Surgery of Pelvic Bone Tumors PDF

Surgery
of Pelvic Bone
Tumors
Pietro Ruggieri
Andrea Angelini
Editors
123
Surgery of Pelvic Bone Tumors
Pietro Ruggieri • Andrea Angelini
Editors
Surgery of Pelvic Bone

Tumors
Editors
Pietro Ruggieri Andrea Angelini
Department of Orthopedics and Department of Orthopedics and
Orthopedic Oncology Orthopedic Oncology
University of Padova University of Padova
Padova Padova
Italy Italy
ISBN 978-3-030-77006-8 ISBN 978-3-030-77007-5 (eBook)

https://doi.org/10.1007/978-3-030-77007-5
© Springer Nature Switzerland AG 2021

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
1 Surgical Approaches in Pelvic Bone Tumors �� 1

Andrea Angelini, Alberto Crimì, Elisa Pala,
and Pietro Ruggieri
2 The History of Pelvic Tumor Surgery�� 15
Peter S. Rose and Franklin H. Sim
3 Imaging of Pelvic Bone Tumors�� 23
Andrea Angelini, Joele Canapeti, Giulia Trovarelli,
Joseph Benevenia, and Pietro Ruggieri
4 Most Common Histological Type of Pelvic Bone Tumors�� 33
Marilyn M. Bui and Andrew E. Rosenberg
5 Benign Pelvic Bone Tumors�� 45
John E. Mullinax and G. Douglas Letson
6 Chondrosarcoma of the Pelvis�� 55
Gerhard M. Hobusch and Reinhard Windhager
7 Osteosarcoma of the Pelvis�� 63
Andre Spiguel, D. Ian English, Cory Couch,
and Mark Scarborough
8 Ewing’s Sarcoma of the Pelvis�� 73
Benjamin V. Kelley, Danielle E. Greig, and Francis J. Hornicek
9 Pelvic Metastases: Diagnosis and Treatment�� 87
Eduardo J. Ortiz-Cruz, Manuel Peleteiro-Pensado,
Irene Barrientos-Ruiz, and Raquel Ortiz-Hernando
10 Demolitive Surgery for Pelvic Bone Tumors�� 97
Joseph Benevenia, Luis A. Guinand, Kathleen S. Beebe,
Joseph Ippolito, Andrea Angelini, and Pietro Ruggieri
11 Principles of Pelvic Surgery�� 113
Ashley Aratani, Chung Ming Chan, C. Parker Gibbs,
12 Innovative Techniques in Pelvic Reconstructions �� 123
Howard Park, Akash A. Shah, and Francis J. Hornicek
v
vi Contents
13 Navigation in Pelvic Surgery�� 135

David M. Joyce
14 Spinopelvic Fixation After Sacrectomy�� 155
Vasilios G. Igoumenou, Andreas F. Mavrogenis, Andrea
Angelini, Farzam Vazifehdan, Konstantinos Soultanis,
Pavlos Patapis, Pietro Ruggieri,
and Panayiotis J. Papagelopoulos
15 Reconstruction vs. No Reconstruction for Pelvic
Resections�� 165
Andreas F. Mavrogenis, Dimitrios Papadopoulos,
Andreas G. Tsantes, Andrea Angelini,
Panayiotis J. Papagelopoulos, and Pietro Ruggieri
16 The Hip Transposition as a Reconstructive Technique
After Pelvic Resection�� 185
Timo Lübben and Georg Gosheger
17 3D Printed Reconstructions�� 191
Andrea Angelini, Daniel Kotrych, Andrzej Szafrański,
and Pietro Ruggieri
18 Particle Radiotherapy�� 203
Petra Georg and Eugen Boris Hug
19 Chemotherapy for Pelvic Bone Tumors �� 215
Antonella Brunello and Vittorina Zagonel
20 Role of Plastic Surgery in the Treatment of Pelvic Tumors �� 225
Cesare Tiengo, Andrea Monticelli, Regina Sonda,
Andrea Angelini, Pietro Ruggieri, and Franco Bassetto
21 Anesthesia for Pelvic Bone Cancer Surgery: From
Risk Evaluation to Postoperative Course�� 233
Gian Mario Parise, Bianca Ferrarese, Alessandro Graziano,
Manuela Funes, Francesco Ambrosio, and Paolo Navalesi
22 The Importance of a Multidisciplinary Approach to Pelvic
Tumours�� 241
Andreas Leithner, Marko Bergovec,
and Dimosthenis Andreou
23 Functional Rehabilitation of Pelvic Resection
and Reconstruction�� 247
Stefano Masiero, Giacomo Magro, Mariarosa Avenia,
and Francesca Caneva
Surgical Approaches in Pelvic
Bone Tumors 1
Andrea Angelini, Alberto Crimì, Elisa Pala,
and Pietro Ruggieri
1.1 Introduction other authors [7–11]. The main ones are the
T-incision, the question mark incision, the verti-
Surgical approaches to the pelvis in musculoskel- cal posterior extension to the vertebral midline,
etal oncology are employed primarily for tumor and the ilioinguinal approach extended to the
removal and, in recent years for pelvic recon- contralateral pubic ramus. These approaches
structions. Because of the constraints posed by require an appropriate preoperative planning and
pelvic anatomy and tumor volume, standard surgeon’ familiarity with the anatomic relation-
“traumatological” exposures are often inade- ships of pelvic region [12]. In some cases, a mul-
quate. Moreover, preoperative biopsy is fre- tidisciplinary approach with two different team
quently performed to reach the definitive for resection and reconstructive procedures could
diagnosis before surgical treatment and biopsy be useful under oncologic point of view.
tract must be included with the specimen to avoid
local tumor cell seeding [1]. This aspect under-
lines that also the trocar-needle biopsy should be 1.2 Preoperative Evaluation
performed by a surgeon with experience in pelvic
resection [2]. The surgical approach for pelvic 1.2.1 Relative Indications
resections was first described by Enneking in
1978 [3] to achieve the desired surgical objective: Several preoperative considerations must be con-
the utilitarian pelvic incision. This extended ilio- sidered before proceeding with internal/external
inguinal approach has been described and used hemipelvectomy. There are some precautions
for all the primary (benign and malignant) and that should be taken into account to avoid intra/
secondary tumors of the pelvic girdle. It can be postoperative complications.
exploited partially or completely depending on
the tumor malignancy and site as well as it can be 1. As is true for all areas of medicine, a complete
extended for wider pelvic resections [4–6]. Since history is crucial to estimate patient’s suitabil-
the initial description, various modifications have ity for surgery, estimation of comorbidities,
been proposed by Campanacci, Karakousis, and and definition of surgical-related risks. In par-
ticular, in oncologic patients, aspects resulting
A. Angelini · A. Crimì · E. Pala · P. Ruggieri (*) from prior surgery, biopsy tract, radiation
Department of Orthopedics and Orthopedic therapy, history of infection may significantly
Oncology, University of Padova, Padova, Italy influence the choice of surgical procedure and
e-mail: [email protected]; approach. Moreover, depending on the size
[email protected]
© Springer Nature Switzerland AG 2021 1

P. Ruggieri, A. Angelini (eds.), Surgery of Pelvic Bone Tumors,
https://doi.org/10.1007/978-3-030-77007-5_1
2 A. Angelini et al.
and site of the tumor, all pertinent imaging 6. In addiction to optimization of the patient’s
and pathologic studies should be completed cardiopulmonary and general medical health
before the final decision to proceed is made. before such massive surgical procedures are
2. We strongly suggest the use of rectal probe undertaken, the strict collaboration with anes-
placed into the rectum and fixed to the peri- thesiologist to alleviate the burden of local dis-
neal region. Not only does it allow the surgeon ease certainly plays a significant role. As a
ability to demarcate the rectum during sur- large amount of blood loss is sometimes
gery, but also it reduces the risk of wound con- encountered in limb salvage procedures for
tamination by fecal material in the immediate pelvic tumors, it is essential to identify risk
postoperative management. factors predicting the possibility of extensive
3. Localizing the ureters during a surgical proce- hemorrhage. The differences in patients’ gen-
dure can be a challenging task in patients eral condition, blood clotting ability, surgical
undergoing pelvic resection. The ureter lies in team experience as well as speed and volume
the interval between the peritoneum and the of blood transfusion may influence brisk hem-
psoas fascia and may be displaced by large orrhage. Tang et al. focused a study on this
tumors extending medially into the pelvis. A topic, finding that acetabulum or sacrum
prophylactic Double-J ureteral stent placement involvement, a tumor volume greater than
few days before surgery may reduce the chance 400 cm‑ and an anticipated operation time of
of injury to the ureter or increase the chance more than 200 min are likely to have a large
that an injury will be recognized immediately amount of blood loss [19]. We usually suggest
[4, 9, 13, 14]. Moreover, a Foley catheter large amounts of transfused blood and plate-
should always be inserted into the bladder. lets should be prepared in such cases.
4. A general anesthetic is usually administered. Moreover, anesthesiologist should be intraop-
An arterial catheter is inserted for continuous eratively updated on the current blood loss to
hemodynamic monitoring, and a central avoid chasing progressively low hemoglobin
venous catheter is advisable. One or more levels. We usually avoid the use of the Esmarch
large-bore peripheral venous catheters are bandage in patients with oncologic disease,
secured in place. even in case of external hemipelvectomy.
5. Infectious complications following major sur-
gical procedures are a significant source of
morbidity and potential mortality [15–18]. 1.2.2 Patient Positioning
Antibiotic prophylaxis is intended for elective
procedures in which the incision will be Patient positioning and surgical incision depend
closed in the operating room. Numerous pro- on the portion of the pelvis and soft tissue to be
tocols have been designed for pelvic surgery, resected, surgeon taste, and experience. All these
but usually must be adapted to specific resis- positions have pros and cons. The patient can be
tance patterns of each hospital environment positioned in supine position (with a bumper on
[18]. Prophylactic antibiotics should be the contralateral side), so when required the table
administered shortly before or at bacterial can be tilted. A sandbag beneath the lower tho-
inoculation. This should be done 15–60 min racic spine of the affected site is useful to roll the
before skin incision. The majority of studies patient approximately 45° anteriorly during pos-
suggest that a single dose is effective but that terior dissection. In this anterior “floppy-lateral”
for lengthy procedures (>3 h) the dose should position, the skin should be prepared from the
be repeated at intervals one or two times the great toe on the involved site to the level of the
half-life of the drug. It has also been suggested xiphoid proximally (including the entire abdo-
that with large blood loss (>1500 mL), a sec- men above the pubic symphysis), and beyond the
ond dose should be given. midline posteriorly.
1 Surgical Approaches in Pelvic Bone Tumors 3
Lateral decubitus position allows simultane- femur. The starting point is the posterior inferior
ous unilateral ventral and dorsal exposure of the iliac spine, the incision then follows the iliac
hemipelvis, with the abdominal organs shifting crest reaching the anterior superior iliac spine
downward far from the deep surgical plan. The (Fig. 1.1a). At this point, it divides in two
patient is placed with the affected side up and the branches (Fig. 1.1b): the first branch of the inci-
contralateral iliac crest centered over the point of sion extends along the inguinal ligament ending
flexion of the operating room table. Obviously, at the symphysis pubis, the second branch extends
all bony prominences should be protected as well caudally with a gentle curve on the anterior
as the contralateral axilla and upper extremities. aspect of the thigh for 5–7.6 cm and then bends
Combined approaches may be performed laterally crossing the femoral shaft just below the
simultaneously or staged as separate proce- greater trochanter following the posterior aspect
dures depending on tumor site, type of recon- of the femur and the insertion of the gluteus max-
struction, and patient’s comorbidities. When a imus muscle. In the modified T-shaped approach,
custom-made prosthesis with spinal fixation is the surgical incision is much more laterally in the
considered for pelvic reconstruction, a consec- turning point compared to the utilitarian incision;
utive procedure which would allow a change in the distal branch runs straight on the lateral aspect
the patient’s position under the same anesthesia of the thigh and does not turn posteriorly like in
is possible. In this case, we prefer a first surgi- the Enneking approach. It was described for the
cal time in prone position before the second first time by Karakousis in 1989 [20].
surgical time in supine (with the possibility of Some adjustments are necessary based on the
tilting the patient 45°) or lateral position. size and position of the tumor: in periacetabular
Regardless of the variation chosen, a third sur- resections the incision is extended on the lateral
gical time for complete spinopelvic fixation thigh, in posterior resections the posterior part of
could be required. the incision can be extended to the spine (with an
added vertical incision), in anterior resection
(pubic rami resections), the ilioinguinal incision
1.3 Utilitarian Pelvic Incision can be extended to the contralateral side or down-
ward facilitating the femoral vessels identifica-
The utilitarian incision provides access to the tion [9, 13, 14, 20, 21].
inner and outer aspects of the innominate bone, The preoperative planning of the resection and
the lower part of the abdomen, and the proximal a correct biopsy technique are pivotal in order to
a b
Fig. 1.1 Utilitarian pelvic incision. (a) The landmarks rior superior iliac spine. (b) Then it divides in two
are the great trochanter (arrow head), anterior superior branches: the first branch extends along the inguinal liga-
iliac spine (white arrow), and symphysis pubic. The start- ment ending at the symphysis pubis (n. 1) and the second
ing point is the posterior inferior iliac spine (white star) branch extends caudally on the anterior aspect of the thigh
and the incision follows the iliac crest reaching the ante- and then laterally just below the greater trochanter (n. 2)
avoid the jeopardizing effect on the soft tissue neurosis, internal oblique, and transversus
survival and reconstruction. The biopsy has to be abdominis muscles. The anterior osteotomy is
performed along the utilitarian pelvic incision performed through the greater sciatic notch or
because the excision of the biopsy tract to avoid just over the acetabulum (preserving the hip
seeding of the tumor cells can bring to an exten- joint), under direct visualization to prevent inju-
sive soft tissue damage [14]. ries to the superior gluteal nerve and vessels. The
posterior osteotomy is through or near the sacro-
iliac joint using an osteotome directed from pos-
1.4 I ndications Related to Types terior to anterior, with a protection of lumbosacral
of Pelvic Resections trunk and sacral roots. The exposure can be
implemented by the release of the iliolumbar
Pelvic resections are classified according to the ligament at the posterior part of the iliac crest.
Musculoskeletal Tumor Society into four types: The L5 nerve root should be visualized and pre-
(1) Type I—iliac resection, (2) Type II—periace- served because it runs inferior and medial to the
tabular resection, (3) Type III—obturator resec- ligament [9, 13, 14, 20, 21].
tion, (4) Type IV—resections involving sacrum
[13]. Resections combining different portions
can be classified and represented with the rela- 1.4.2 Type II Resection
tive roman numbers, such a resection involving
iliac and acetabular areas is called Type I/II If the tumor involves the acetabulum (a tumor
resection. When all the three parts of the innomi- arising from the acetabulum itself or from the
nate bone are resected preserving the limb, the proximal femur and involving the hip joint in the
procedure is called internal hemipelvectomy acetabular component), a periacetabular Type II
(Type I/II/III) [22]. resection is indicated. In contrast to the iliac
If the resection includes the proximal femur it resection, the internal hemipelvectomy could be
is defined a Type H resection, divided in: Type performed if an adequate wide resection proce-
H1—resection involving the femoral head, Type dure could be performed sparing the major nerves
H2—resection involving the trochanteric area, and preserving a functional limb [3, 21]. The lat-
Type H3—resection involving the subtrochan- eral arm of the incision to the thigh is developed
teric area [13, 21]. through the skin and the subcutaneous tissue,
When resection includes the sacrum, the releasing the tensor fascia lata, sartorius muscle,
subclassification is categorized in four types: and the straight head of the rectus femoris from
Type 1—resection involving a total sacrectomy, their insertions on the iliac crest and anterosupe-
Type 2—resection involving a emisacrectomy, rior iliac spine, respectively. The anterior osteot-
Type 3—resection involving a partial sacrec- omy is performed through the anterior column of
tomy associated with an external hemipelvec- the acetabulum, the base of the superior pubic
tomy, Type 4—resection involving a total ramus. The posterior osteotomy is in the poste-
sacrectomy associated with an external hemi- rior acetabular column or in the ischium. The
pelvectomy [4, 20]. superior osteotomy is through the greater sciatic
notch. If the posterior column is involved, some
authors suggest en-bloc removal of the acetabu-
1.4.1 Type I Resection lum and ischium [9, 20].
In order to obtain a Type I resection, only the first

portion (most posterior part) of the utilitarian pel- 1.4.3 Type III Resection
vic incision is usually needed. Anteriorly, the lat-
eral attachment of the inguinal ligament is Type III pelvic resection requires a medial oste-
resected together with the external oblique apo- otomy (through the pubic symphysis) and it is the
case where the utilitarian incision should be technique, patients’ survival and tumor control
extended to the contralateral pubic ramus. can be achieved with a total sacrectomy [24–27].
Another osteotomy should be performed just If the tumor invades S1, lumbar spine, and pelvis,
medial to the acetabulum, avoiding the hip dislo- the proposed surgical approach is a combined
cation [21]. In this kind of resection, due to their staged posterior and anterior approach. The com-
proximity to the pelvic sidewall, obturator artery, bined approach finds indication in tumors with
vein, and nerve are usually sacrificed with part of high vascularization, primary sacral tumors
the obturator internus muscle [13, 20, 21]. In involving S1 or invading the lumbosacral junc-
order to avoid hernias of the peritoneum, a care- tion [9, 38].
ful reconstruction of the inguinal floor is required
all along the excised part of the pubic ramus.
After the excision of the bony part, femoral ves- 1.5 Deep Surgical Dissection
sels and spermatic cord should be repositioned
deep to the abdominal wall reconstruction [23]. A large flap of the gluteus maximus is reflected
posteriorly in order to give exposure to the greater
and lesser sciatic notches, the ischium, and the
1.4.4 T
ype IV Resections proximal third of the femur. The flap is based on
and Sacrectomies a line that extends from the most medial portion
of the posterior part of the iliac wing to the pos-
Type IV resections involve the sacrum. terolateral aspect of the thigh [3]. The sciatic
Sacrectomy can be partial or total, combined usu- nerve is close to the pelvis at the sciatic notch; it
ally with iliac resections and lower lumbar spine is usually not infiltrated and can be isolated and
resections [4, 21, 24, 25]. The S2 level is pivotal easily separated from the tumor. Iliac muscle,
to define the outcome and surgical approaches to gluteus medius, and gluteus minimus muscles are
obtain a resection with wide margins. A tumor usually excised in order to obtain wide margins
extending below the S2 level can be treated with and good coverage of the pelvic tumor (more glu-
a partial sacrectomy (transverse, sagittal, com- teus medius is not excised more abductor func-
bined) without spino-pelvic reconstruction, with tion will be preserved). The superior gluteal
good expected neurologic results related to blad- artery and vein are sacrificed because the gluteus
der and bowel function [24–27]. Moreover, a medius and gluteus minimus are resected with
posterior-only approach could be used in selected the tumor.
cases [28]. If the sacroiliac joint is not involved In the anterior branch of the approach
by a sacral tumor (lateral sacral tumor), a sagittal (Fig. 1.2a), the inguinal ligament has to be
partial sacrectomy is indicated, whereas in case detached from the anterior superior iliac spine
of sacroiliac joint involvement a partial sacrec- and, as well as in ilioinguinal approach, the apo-
tomy and resection of the posterior part of the neurosis of the external oblique muscle has to be
ileum (type I, IV resection) should be considered incised from the superficial inguinal ring to the
[29]. Sacral midline tumors not involving the sac- anterior superior iliac spine (Fig. 1.2b). Spermatic
roiliac joint are treated with a transverse sacrec- cord in male or the round ligament in female
tomy [30, 31]. A total sacrectomy is indicated patients should be protected and retracted medi-
when an aggressive lesion involves the proximal ally, then the section of the posterior wall of the
sacrum with anterior extension (rarely tumor can ilioinguinal canal (fibers of internal oblique and
penetrate the anterior pelvic fascia extending to transverse abdominis muscles) is performed
the rectum and other pelvic organs) [32–36]. In under tension. The femoral bundle should be
this case, sacral roots are necessarily sacrificed to identified between the pubic tubercle and iliac
obtain wide surgical margins and local tumor crest, just anterior to the superior pubic ramus
control [24, 37]. Despite major complications (Fig. 1.2c). Inferior epigastric artery and vein
and implicit neurological deficits of this resection should be ligated. The multidisciplinary team
should include the plastic surgeon considering surgery in the ureter facilitates its identification as
that, if ipsilateral vertical abdominis musculocu- it crosses the common iliac artery, it must be iden-
taneous flap should be used, the deep inferior epi- tified and should be retracted medially. The poste-
gastric artery should be preserved and protected. rior part of the sacroiliac joint should be visualized;
The important structures in the area should be the L5 nerve roots come out just below the L5 ver-
identified and protected: the spermatic cord tebra’s transverse process, where the iliolumbar
(while round ligament in women can be sacri- ligament attaches to the posterior ilium. In pelvic
ficed), the femoral vessels (section of the iliopec- resection type 4, a posterior approach is usually
tineal fascia in order to mobilize the vascular needed for vertebral instrumentation.
bundle), and iliopsoas muscle with the femoral Once the bone cuts have been completed, the
nerve that lies deep inside the muscle [13, 14, 20, pelvis will open, but the sacrospinal and sacropu-
39]. A large vessel loop is placed around the bic ligaments must be resected to release the hemi-
common iliac vessels to assist with their mobili- pelvis and make it loose still [13, 14, 20, 39]. The
zation (Fig. 1.2d). Arising from the medial and specimen should be compared with preoperative
lateral aspects of the common femoral artery are resection planes and margins macroscopically
the external pudendal and superficial circumflex evaluated (Fig. 1.2f). After that, the reconstructive
iliac arteries that could be ligated to allow mobi- phase can be carried out (Fig. 1.2g–j).
lization of the femoral vessels. Protection of the The “reverse question mark” approach is char-
bladder is required, if pubis ramus osteotomy has acterized by the absence of the anterior branch of
to be performed. The pubic symphysis is exposed the utilitarian pelvic incision (Fig. 1.3a) and could
by detaching the anterior rectus abdominis and be used when pubic osteotomy is planned close to
pyramidalis muscles from their insertion onto the the acetabulum. The deep surgical dissection
ipsilateral pubic crest (Fig. 1.2e). The urethra that includes the same previously described steps start-
lies just inferior to the pubic symphysis and sepa- ing from neurovascular identification (Fig. 1.3b).
rated only by the arcuate ligament, should be pro- One of the advantages of the supine position is the
tected during osteotomy. These structures are intra- and inter-observer reproducibility of radio-
better identified with a Foley catheter inserted. In graphic measurements, especially when a custom-
the following step, the help of a general surgeon made resection (Fig. 1.3c) and prosthetic
is needed at this time to gentle separate the reconstruction are planned (Fig. 1.3d, e).
abdominal organs from the pelvic tumor assess-
ing that wide margins are granted.
In the posterior pelvis after anterior part of the 1.6 Pelvic and Soft Tissue
sacroiliac joint is identified and going further Reconstruction
medially, common iliac vessels should be identi-
fied and followed into the pelvis (the same for the At today, there are few instances in which a staged
inferior vena cava in a right internal hemipelvec- approach may be preferable. In most of the cases,
tomy). A Double-J ureteral stent inserted before a consecutive procedure allows the possibility of
Fig. 1.2 Young patient (12 years-old) with Ewing’s sar- been positioned on the exposed iliac bone and pubic sym-
coma of the left hemipelvis. (a) The skin incision was first physis to perform correct osteotomies. (f) The model of the
drawn follows the utilitarian approach in supine position. tumor and the specimen is shown to emphasize the similar-
(b) The inguinal ligament is incised from the anterior ity between resection plan and actual margins. (g) After
superior iliac spine (white arrow), as well as the aponeuro- tumor removal, it is possible to evaluate bone defect, oste-
sis of the external oblique muscle. (c) The femoral bundle otomy surface of the iliac bone (dashed line), and neuro-
has been identified (asterisk) and (d) protected with a large vascular bundle (asterisk). (h) In this case, reconstruction
vessel loop. The tensor fascia lata, sartorius muscle, and has been performed with an iliofemoral coarctation stabi-
the straight head of the rectus femoris have been released lized with a mesh tube (Trevira; Implantcast, Buxtehude,
from their insertions on the iliac crest and anterosuperior Germany). (i) Soft tissue reconstruction and (j) reinforce-
iliac spine, respectively (black arrow). (e) A cutting jig has ment of the abdominal wall with fascia lata graft
a b
c d
e f
g gh b
i j
Fig. 1.2 (continued)
an immediate reconstruction, with a better reat- Different flap techniques are in use and are
tachment of the soft tissue. In wide pelvic resec- available considering the extension and the soft
tions (mainly in combined type including a type tissue damage during tumor excision [51, 52].
II), the reconstruction of soft tissue defect and Preoperative CT with contrast study is always
adequate implant coverage is crucial [40, 41]. mandatory to properly plan the flap.
Modular prostheses, custom-made 3D-printing
prostheses, massive allografts, and other tech-
niques are used for these challenging reconstruc- 1.6.1 Rectus Abdominis
tions [42–50] and infection remains the main Musculocutaneous Flap
complication [15–18]. In literature, it is widely
reported that good soft tissue coverage of the pros- Local flaps (advancement, rotation, propeller,
thesis is considered one of the most relevant fac- and transposition flaps) are mainly based on a
tors associated with implant survival [16, 18, 46]. perforator as a pedicle. The rectus abdominis
The intersection point of the cutaneous inci- musculocutaneous (RAM) flap could be used as a
sions is at risk of delayed healing and wound muscular or a musculocutaneous flap, and could
necrosis (with finally high risk of periprosthetic), be realized with a transverse RAM (TRAM) or a
significantly higher if the soft tissue reconstruc- vertical RAM (VRAM) based on the orientation
tion of the deep tissues is not adequate. Enneking of the skin paddle, to fill small defects with
suggested, if there was not enough tissue to close exposed vital structures. The VRAM flap is a
the wound primarily, to cover important struc- solution in periacetabular and sacral reconstruc-
tures with flaps of omentum, it dressed with pig- tions [41, 53–57]. In some cases, if there is a
skin and then by skin grafts [3]. large fascial defect, it can be associated with a
a b
c d
e f
Fig. 1.3 Adult patient (52 years-old male) with an osteo- shows the classic use of C-arm fluoroscopy in intraopera-
sarcoma of the left hemipelvis. (a) The “reverse question tive orthopedic procedures. (e) Image intensification is
mark” approach has been drawn on the skin. The classic very useful in the evaluation of bone resection and recon-
anterior branch of the utilitarian pelvic incision is dashed structive aspects, and it allows greater flexibility with
medially to the pubic symphysis. (b) Identification of the standard radiographic projections. (f) Definitive custom-
femoral bundle. (c) The iliac wing should be accurately made 3D printed prosthesis implanted before soft tissue
prepared to fit with the cutting guide jig. (d) The picture reconstruction
synthetic mesh or an acellular dermal matrix. extending from the pubic symphysis to just above
These solutions can be used to repair anterior the umbilicus is performed. The rectus abdominis
defects of the donor site, posterior defects or muscle is then dissected maintaining intact the
both. The patient is positioned supine, the rectus anterior portion of the sheath to avoid damaging
abdominis muscle is palpated and outlined with a the vascular perforators (Fig. 1.4b). The har-
marker, and the flap is designed around the vested rectus flap could be rotated on its pedicle
needed skin island (Fig. 1.4a). A midline incision (Fig. 1.4c, d) and tunneled via an intraperitoneal
a b
c d
e f
Fig. 1.4 Adult patient (54 years-old) with sacral chor- cular perforators. (c, d) The harvested rectus flap can be
doma. (a) Patient in supine position. The skin island is rotated on its pedicle and temporarily placed intraperito-
drawn based on the shape of the rectus abdominis muscle neally. (e, f) During the anterior approach for proximal
and the planned plastic reconstruction. (b) The rectus sacral resection, an omental-pedicled flap based on the
abdominis muscle is then dissected maintaining intact the right gastroepiploic artery is fashioned and used to fill the
anterior portion of the sheath to avoid damaging the vas- dead space
route into the pelvis or via an extrapelvic subcu- are fascio-cutaneous flaps usually considered for
taneous route to support wound closure [56, 57]. partial sacral or total sacrectomies [58]. They
may eventually include the underneath muscle,
even if this myocutaneous technique should be
1.6.2 Other Flaps generally avoided because it can lead to severe
walking impairment. The use of a pedicled
Superior gluteal artery perforator (SGAP) or omental flap has been described as a tool of
inferior gluteal artery perforator (IGAP) flaps decreasing wound complications reducing the
dead space with a vascularized tissue (Fig. 1.4e, f) 11. Steel HH. Partial or complete resection of the hemi-
pelvis: an alternative to hindquarter amputation for
[53, 59]. Anterolateral thigh (ALT) flap is a reli- periacetabular chondrosarcoma of the pelvis. J Bone
able flap that can be used in periacetabular and Joint Surg Am. 1978;60:719–30.
sacral soft tissue reconstruction in some rare sit- 12. Mavrogenis AF, Angelini A, Drago G, Merlino B,
uations, usually to cover perineal or groin soft Ruggieri P. Survival analysis of patients with chondro-
sarcomas of the pelvis. J Surg Oncol. 2013;108(1):19–
tissue defects. Tensor fascia latae (TFL) flap is a 27. https://doi.org/10.1002/jso.23351.
good flap for the coverage of the trochanteric, 13. Enneking WF. The anatomic considerations in

periacetabular, perineum, and abdominal wall tumor surgery: pelvis. In: Enneking WF, editor.
soft tissue defects. It can be both a muscular or Musculoskeletal tumor surgery, vol. 2. New York,
NY: Churchill Livingstone; 1983. p. 483–529.
musculocutaneous flap. 14. Windhager R, Karner J, Kutschera HP, Polterauer P,
Salzer-Kuntschik M, Kotz R. Limb salvage in peri-
acetabular sarcomas: review of 21 consecutive cases.
Clin Orthop Relat Res. 1996;(331):265–76.
References 15. Trovarelli G, Angelini A, Pala E, Cappellari A, Breda
A, Ruggieri P. Infection in orthopaedic oncology: cru-
1. Mavrogenis AF, Angelini A, Errani C, Rimondi cial problem in modern reconstructive techniques. Eur
E. How should musculoskeletal biopsies be per- Rev Med Pharmacol Sci. 2019;23(2 Suppl):271–8.
formed? Orthopedics. 2014;37(9):585–8. https://doi. https://doi.org/10.26355/eurrev_201904_17501.
org/10.3928/01477447-20140825-03. 16. Miwa S, Shirai T, Yamamoto N, Hayashi K, Takeuchi
2. Rimondi E, Rossi G, Bartalena T, Ciminari R, A, Tada K, Kajino Y, Higuchi T, Abe K, Aiba H,
Alberghini M, Ruggieri P, Errani C, Angelini A, Taniguchi Y, Tsuchiya H. Risk factors for surgical site
Calabrò T, Abati CN, Balladelli A, Tranfaglia C, infection after malignant bone tumor resection and
Mavrogenis AF, Vanel D, Mercuri M. Percutaneous reconstruction. BMC Cancer. 2019;19(1):33. https://
CT-guided biopsy of the musculoskeletal system: doi.org/10.1186/s12885-019-5270-8.
results of 2027 cases. Eur J Radiol. 2011;77(1):34– 17. Severyns M, Briand S, Waast D, Touchais S, Hamel A,
42. https://doi.org/10.1016/j.ejrad.2010.06.055. Gouin F. Postoperative infections after limb-sparing
3. Enneking WF, Dunham WK. Resection and recon- surgery for primary bone tumors of the pelvis: inci-
struction for primary neoplasms involving the innomi- dence, characterization and functional impact. Surg
nate bone. J Bone Joint Surg Am. 1978;60(6):731–46. Oncol. 2017;26(2):171–7. https://doi.org/10.1016/j.
4. Mavrogenis AF, Soultanis K, Patapis P, Guerra G, suronc.2017.03.005.
Fabbri N, Ruggieri P, Papagelopoulos PJ. Pelvic 18. Angelini A, Drago G, Trovarelli G, Calabrò T,

resections. Orthopedics. 2012;35(2):e232–43. https:// Ruggieri P. Infection after surgical resection for pel-
doi.org/10.3928/01477447-20120123-40. vic bone tumors: an analysis of 270 patients from one
5. Angelini A, Calabrò T, Pala E, Trovarelli G, Maraldi institution. Clin Orthop Relat Res. 2014;472(1):349–
M, Ruggieri P. Resection and reconstruction of pelvic 59. https://doi.org/10.1007/s11999-013-3250-x.
bone tumors. Orthopedics. 2015;38(2):87–93. https:// 19. Tang X, Guo W, Yang R, Tang S, Ji T. Evaluation
doi.org/10.3928/01477447-20150204-51. of blood loss during limb salvage surgery for pelvic
6. Puri A, Pruthi M, Gulia A. Outcomes after limb spar- tumours. Int Orthop. 2009;33(3):751–6.
ing resection in primary malignant pelvic tumors. 20. Karakousis CP, Emrich LJ, Driscoll DL. Variants of
Eur J Surg Oncol. 2014;40(1):27–33. https://doi. hemipelvectomy and their complications. Am J Surg.
org/10.1016/j.ejso.2013.10.017. 1989;158:404–8.
7. Bickels J, Malawer M. Pelvic resections (internal 21. Conrad EU III, Springfield D, Peabody TD. Pelvis. In:
hemipelvectomies). In: Malawer M, Sugarbaker P, Simon MA, Springfield D, editors. Surgery for bone
editors. Musculoskeletal cancer surgery: treatment of and soft-tissue tumors. Philadelphia, PA: Lipincott-
sarcomas and allied diseases. Boston, MA: Kluwer Raven; 1998. p. 323–41.
Academic Publishers; 2001. p. 405–14. 22.
Masterson EL, Davis AM, Wunder JS, Bell
8. Karakousis CP. Internal hemipelvectomy. In: Sugarbaker RS. Hindquarter amputation for pelvic tumors. Clin
P, Malawer M, editors. Musculoskeletal surgery for can- Orthop Relat Res. 1998;350:187–94.
cer: principles and techniques. New York, NY: Theime 23.
Die Trill J, Madrid JM, Ferrero E, et al.
Medical Publishers, Inc; 1992. p. 150–63. Posthemipelvectomy hernia. Hernia. 2005;9(4):375–7.
9. Campanacci M, Capanna R. Pelvic resections: the 24. Angelini A, Pala E, Calabrò T, Maraldi M, Ruggieri
Rizzoli Institute experience. Orthop Clin North Am. P. Prognostic factors in surgical resection of sacral
1991;22(1):65–86. chordoma. J Surg Oncol. 2015;112(4):344–51. https://
10. Mankin HJ, Hornicek FJ. Internal hemipelvectomy doi.org/10.1002/jso.23987.
for the management of pelvic sarcomas. Surg Oncol 25. Ruggieri P, Angelini A, Ussia G, Montalti M,

Clin N Am. 2005;14:381–96. Mercuri M. Surgical margins and local control in
r esection of sacral chordomas. Clin Orthop Relat Res. 42. Angelini A, Trovarelli G, Berizzi A, Pala E, Breda
2010;468(11):2939–47. A, Ruggieri P. Three-dimension-printed custom-
26. Sar C, Eralp L. Surgical treatment of primary
made prosthetic reconstructions: from revision
tumors of the sacrum. Arch Orthop Trauma Surg. surgery to oncologic reconstructions. Int Orthop.
2002;122:148–55. 2019;43(1):123–32. https://doi.org/10.1007/
27. Wuisman P, Lieshout O, Sugihara S, van Dijk
s00264-018-4232-0.
M. Total sacrectomy and reconstruction: oncologic 43. Delloye C, Banse X, Brichard B, Docquier PL, Cornu
and functional outcome. Clin Orthop Relat Res. O. Pelvic reconstruction with a structural pelvic
2000;381:192–203. allograft after resection of a malignant bone tumor. J
28. Angelini A, Ruggieri P. A new surgical technique Bone Joint Surg Am. 2007;89(3):579–87. https://doi.
(modified Osaka technique) of sacral resection by org/10.2106/JBJS.E.00943.
posterior-only approach: description and preliminary 44. Sun W, Li J, Li Q, Li G, Cai Z. Clinical effectiveness
results. Spine (Phila Pa 1976). 2013;38(3):E185–92. of hemipelvic reconstruction using computer-aided
29. Lin P, Shao Y, Lu H, et al. Pelvic reconstruction with custom-made prostheses after resection of malignant
different rod-screw systems following Enneking pelvic tumors. J Arthroplast. 2011;26:1508–13.
type I/I + IV resection: a clinical study. Oncotarget. 45. Wang J, Min L, Lu M, et al. Three-dimensional-

2017;8(24):38978–89. https://doi.org/10.18632/ printed custom-made hemipelvic endoprosthesis
oncotarget.17164. for primary malignancies involving acetabulum: the
30. Dickey ID, Hugate RR Jr, Fuchs B, Yaszemski MJ, design solution and surgical techniques. J Orthop
Sim FH. Reconstruction after total sacrectomy: early Surg Res. 2019;14:389. https://doi.org/10.1186/
experience with a new surgical technique. Clin Orthop s13018-019-1455-8.
Relat Res. 2005;438:42–50. 46. Angelini A, Kotrych D, Trovarelli G, Szafrański A,
31. Jackson RJ, Gokaslan ZL. Spinal-pelvic fixation in Bohatyrewicz A, Ruggieri P. Analysis of principles
patients with lumbosacral neoplasms. J Neurosurg. inspiring design of three-dimensional-printed custom-
2000;92:61–70. made prostheses in two referral centres. Int Orthop.
32. Gokaslan ZL, Romsdahl MM, Kroll SS, et al. Total 2020;44(5):829–37. https://doi.org/10.1007/s00264-
sacrectomy and Galveston L-rod reconstruction for 020-04523-y. Epub 2020 Mar 13.
malignant neoplasms. Technical note. J Neurosurg. 47. Barrientos-Ruiz I, Ortiz-Cruz EJ, Peleteiro-Pensado
1997;87:781–7. M. Reconstruction after hemipelvectomy with the
33. Capanna R, Briccoli A, Campanacci L, et al. Benign ice-cream cone prosthesis: what are the short-term
and malignant tumors of the sacrum. In: Frymoyer clinical results? [Published correction appears in
JW, editor. The adult spine: principles and practice. Clin Orthop Relat Res. 2017 Mar;475(3):924]. Clin
2nd ed. Philadelphia, PA: Lippincott-Raven; 1997. Orthop Relat Res. 2017;475(3):735–41. https://doi.
p. 2367–405. org/10.1007/s11999-016-4747-x.
34. Guo Y, Yadav R. Improving function after total sacrec- 48. Brown TS, Salib CG, Rose PS, Sim FH, Lewallen DG,
tomy by using a lumbar-sacral corset. Am J Phys Med Abdel MP. Reconstruction of the hip after resection
Rehabil. 2002;81:72–6. of periacetabular oncological lesions: a systematic
35. Angelini A, Ruggieri P. Chordoma of the sacrum. review. Bone Joint J. 2018;100-B(1 Supple A):22–
In: Ruggieri P, Angelini A, Vanel D, Picci P, editors. 30. https://doi.org/10.1302/0301-620X.100B1.BJJ-
Tumors of the sacrum. Cham: Springer; 2017. 2017-0548.R1.
36.
Angelini A, Mavrogenis AF, Ruggieri 49. Liang H, Ji T, Zhang Y, Wang Y, Guo W. Reconstruction
P. Chondrosarcoma of the sacrum. In: Ruggieri P, with 3D-printed pelvic endoprostheses after resection
Angelini A, Vanel D, Picci P, editors. Tumors of the of a pelvic tumour. Bone Joint J. 2017;99-B(2):267–
sacrum. Cham: Springer; 2017. 75. https://doi.org/10.1302/0301-620X.99B2.BJJ-
37.
Cheng EY, Ozerdemoglu RA, Transfeldt EE, 2016-0654.R1.
Thompson RC Jr. Lumbosacral chordoma. Prognostic 50. Wang B, Xie X, Yin J, et al. Reconstruction with
factors and treatment. Spine. 1999;24(16):1639–45. modular hemipelvic endoprosthesis after pelvic tumor
38.
Mindea SA, Salehi SA, Ganju A, et al. resection: a report of 50 consecutive cases. PLoS One.
Lumbosacropelvic junction reconstruction result- 2015;10:0127263.
ing in early ambulation for patients with lumbosa- 51. Vancabeke M, Harper L, Penders W, Putz P. Anterior
cral neoplasms or osteomyelitis. Neurosurg Focus. flap for coverage following hip disarticulation for
2003;15(2):E6. osteomyelitis. Acta Orthop Belg. 1999;65(2):223–5.
39. Simon MA. Surgery for bone and soft-tissue tumors. 52. Siegel GW, Kuzon WM, Hasen JM, Biermann

Philadelphia, PA: Lippincott-Raven; 1998. JS. Staged soft tissue reconstruction following
40. Ruggieri P, Angelini A, Vanel D, Picci P. Tumors of sarcoma excision with anticipated large cutane-
the sacrum: diagnosis and treatment of benign and ous defects: an oncologically safe alternative. Iowa
malignant tumors. Cham: Springer; 2017. Orthop. 2016;36:104–8.
41. Buchel EW, Finical S, Johnson C. Pelvic reconstruc- 53. Asaad M, Rajesh A, Wahood W, Vyas KS, Houdek MT,
tion using vertical rectus abdominis musculocutane- Rose PS, Moran SL. Flap reconstruction for sacrec-
ous flaps. Ann Plast Surg. 2004;52:22–6. tomy defects: a systematic review and meta-analysis.
J Plast Reconstr Aesthet Surg. 2020;73(2):255–68. following wide resection of malignant bone tumours
https://doi.org/10.1016/j.bjps.2019.09.049. of the pelvis. Bone Joint J. 2014;96-B(2):270–3.
54. Kim JE, Pang J, Christensen JM, Coon D, Zadnik https://doi.org/10.1302/0301-620X.96B2.32514.
PL, Wolinsky JP, Gokaslan ZL, Bydon A, Sciubba 57. Zheng L, Dong ZG, Zheng J. Deep inferior epigas-
DM, Witham T, Redett RJ, Sacks JM. Soft-tissue tric vessel-pedicled, muscle-sparing rectus abdomi-
reconstruction after total en bloc sacrectomy. J nis myocutaneous (RAM) flap for reconstruction
Neurosurg Spine. 2015;22(6):571–81. https://doi. of soft tissue defects in pelvic area. Eur J Orthop
org/10.3171/2014.10.SPINE14114. Surg Traumatol. 2015;25(5):859–63. https://doi.
55. Glatt BS, Disa JJ, Mehrara BJ, Pusic AL, Boland P, org/10.1007/s00590-015-1599-0.
Cordeiro PG. Reconstruction of extensive partial or 58. Oswald TM, Lineaweaver WC, Hui K. Recipient ves-
total sacrectomy defects with a transabdominal ver- sels for microsurgical flaps to the groin and pelvis:
tical rectus abdominis myocutaneous flap. Ann Plast a review. J Reconstr Microsurg. 2006;22(1):5–14.
Surg. 2006;56(5):526–30; discussion 530-1. https:// https://doi.org/10.1055/s-2006-931900.
doi.org/10.1097/01.sap.0000205772.15061.39. 59. Garvey PB, Rhines LD, Feng L, Gu X, Butler

56. Ogura K, Miyamoto S, Sakuraba M, Chuman H,
CE. Reconstructive strategies for partial sacrectomy
Fujiwara T, Kawai A. Immediate soft-tissue recon- defects based on surgical outcomes. Plast Reconstr
struction using a rectus abdominis myocutaneous flap Surg. 2011;127(1):190–9.
The History of Pelvic Tumor
Surgery 2
Peter S. Rose and Franklin H. Sim
2.1 Introduction in the selection and management of patients for

these aggressive surgeries.
The field of pelvic tumor surgery has advanced
over the last 125+ years; this progress has been
based on advances in several related areas of 2.2 Early History
medicine:
The first known attempted hemipelvectomy was
• Improved anesthesia and perioperative care by Bilroth in 1891 with a fatal outcome from
capabilities. hemorrhagic shock [1]. A subsequent successful
• Greater understanding of sarcomatous disease operation (for advanced tuberculosis of the hip)
processes and margins. was performed in 1900 by Hogarth-Pringle and is
• Improved imaging capabilities, particularly the first reported in the English literature [2].
the use of computed tomography and mag- Kocher described the first limb sparing pelvic
netic resonance imaging. excision in the late nineteenth century [3], but
• The development of adjuvant chemo- and Putti provides the first well-documented case of
radio-therapy. internal hemipelvectomy in 1914 with successful
• Expansion of orthopedic resections to locally outcome [4].
advanced visceral disease processes. Speed popularized the term “hemipelvec-
• Critical examination of patient results and tomy” to describe radical amputation through the
outcomes. pelvis and replace the cumbersome “inter-ilio-
abdominal amputation,” while Gordon-Taylor
At present, most patients with localized pelvic referenced the procedure as a “hindquarter ampu-
sarcomas are candidates for curative resection, tation” [5, 6]. The modern term “internal hemi-
although high immediate and long-term morbid- pelvectomy” to describe limb sparing approaches
ity remains inherent to these procedures. As well, was first reported by Eilber in 1979 [7], and by
the majority of patients are candidates for limb analogy amputative resections are often referred
salvage operations. Several controversies remain to as “external hemipelvectomies” in contempo-
rary practice.
The early twentieth century publications were
primarily case reports or small case series which
P. S. Rose (*) · F. H. Sim emphasized the surgical anatomy of approaches
Mayo Clinic, Rochester, MN, USA with relatively little data on patient outcome
e-mail: [email protected]

16 P. S. Rose and F. H. Sim
beyond mortality [6, 8]. Initially, operative mor- free, a decade or more after surgery. This work
tality remained prohibitively high during the first remains one of the first and clearest tabulation of
half of the twentieth century. Gordon-Taylor the principles of bone sarcoma resection and the
reported operative mortality in 31 of 55 patients greater than tenfold increase in survival seen with
(56%) treated with hemipelvectomy for sarcoma proper treatment.
or tuberculosis in 1934 and described the proce- Similar results accrued in other bone sarco-
dure as “one of the most colossal mutilations mas and in soft tissue sarcomas to define the
practiced on the human frame” [5, 9]. This opera- strong importance of proper biopsy and en bloc
tive mortality decreased to 22% in a later report resection techniques in the treatment of sarcomas
as their team gained experience [10]. The [12, 13]. Enneking, a pioneering pelvic sarcoma
decrease is likely due to a combination of team surgeon who helped usher in the modern era of
experience and improved perioperative care. treatment, tabulated and popularized these prin-
ciples to guide sarcoma surgery in general [14].
The accumulating experience which helped to
2.3 Advances in Disease define disease processes and these principles and
Understanding the dissemination of them to surgeons helped
propel the role of surgery as a part of curative
The mid-twentieth century brought significant treatment protocols for pelvic neoplasms.
advances in the scientific understanding of sarco-
matous disease processes and the treatment of
tuberculosis (an early indication for major pelvic 2.4 Imaging Advances
resection). This included the establishment of
sarcoma diagnostic categories and a tabulation of The imaging of pelvic sarcomas remains com-
the natural history of conditions. For example, plex today, even with the variety of advanced
Dahlin and Henderson enumerated the basic imaging modalities available. The first pelvic
treatment principles of chondrosarcoma in 1956 tumor surgeries were based on plain film radio-
which remain true for the treatment of pelvic graphs, physical examination, and surgical explo-
chondrosarcomas to this day [11]: ration. Later surgeons used plain film tomograms
to better image the bone in combination with
1. An adequate biopsy specimen for diagnosis catheter angiograms and barium enemas to infer
should be obtained. soft tissue extension [15]. Bone scans were incor-
2. The definitive operation that is carried out is porated as well but lacked spatial resolution.
performed in such a manner that the biopsy The lack of imaging frequently lead to poorly
wound will be excluded from the incision and placed biopsies, inadequate margins, and poor
will be removed with the specimen or limb, or outcomes. Enneking’s large series published in
both, without being opened again. 1978 (patients operated between 1957 and 1977)
3. The tumor itself should be completely excised revealed that one-third of patients treated with
with a zone of surrounding tissue so that the pelvic resections had oncologically inadequate
surgeon does not break into or see the tumor at surgeries for these reasons [16]. Tumor recur-
any time. rence was seen in 100% of patients with inade-
quate surgeries. The certain morbidity of these
Dahlin and Henderson documented the diffi- procedures and far from certain surgical out-
cult and morbid course of tumor recurrence as comes naturally tempered the enthusiasm of phy-
justification for aggressive initial treatment. They sicians and patients alike in selecting aggressive
noted that only 3.4% of patients with inadequate management of pelvic sarcomas.
surgical treatment survived or remained disease- The advent of computed tomography in the
free at 10 years, while 41% of patients treated 1970s significantly improved the ability to image
according to these principles remained disease- patients with pelvic tumors [17, 18]. CT imaging
2 The History of Pelvic Tumor Surgery 17
provided surgeons with two primary benefits. term disease-free survival was rarely achieved in
First, it allowed much improved anatomic defini- these patients.
tion of the extent of pelvic sarcomas to define The advent of doxorubicin-based chemother-
their osseous and soft tissue extension as well as apy immediately and dramatically improved the
visceral relationships. Second, CT scans of the survival of patients with osteosarcoma [22].
chest provided improved sensitivity to detect pul- These advances provided a meaningful potential
monary metastases compared to chest radio- for survival for patients with high-grade axial
graphs or lung tomograms. This second benefit sarcomas and opened the door to consideration of
allowed teams to more reliably exclude from sur- aggressive surgical treatment for what had gener-
gery patients with established metastatic disease. ally been considered a fatal disease. Parallel
CT became widely available at regional tumor advances were made in the treatment of Ewing’s
centers in the early 1980s. In a similar fashion, sarcoma during this era as well [23].
magnetic resonance imaging provided additional Simultaneous advances were made in the
anatomic discrimination of tumor extent and understanding of the use of radiotherapy for pel-
became widely available by 1990 [19]. In the vic Ewing’s sarcoma [24]. Because of the uncer-
recent two decades, positron emission tomogra- tainties of imaging, margin, and prognosis, the
phy has similarly increased the ability of physi- majority of patients with Ewing’s sarcoma of the
cians to properly stage sarcoma patients [20]. pelvis were treated with radiation therapy for
These imaging advances improved the ability local control. Greater enthusiasm grew for surgi-
of surgeons to assess patients for resectability, cal management of pelvic Ewing’s tumors (with
decrease inadvertent positive margins, and to or without radiation) in the 1980s and 1990s with
avoid morbid surgery on patients with metastatic improved imaging, although this remains a con-
disease. The current imaging of pelvic sarcomas troversial aspect of pelvic sarcoma treatment [25,
is center-specific but typically combines CT and 26].
MR imaging of the local disease with CT of the
chest and bone scan (or potentially PET) for
staging. 2.6 Application to Visceral
Diseases
2.5 Adjuvant Treatments While initially associated with high morbidity,

the same conditions which led to advances in pel-
The three most common bone sarcomas encoun- vic sarcoma surgery provided parallel advances
tered in the pelvic region include chondrosar- in surgery for pelvic visceral diseases [27]. This
coma, osteosarcoma, and Ewing’s sarcoma. allowed for the identification of a subset of
Chondrosarcoma remains stubbornly resistant to patients with locally advanced visceral malignan-
any known adjuvant treatment, with prognosis cies and musculoskeletal involvement and no dis-
heavily influenced by grade and surgical margin tant metastases; typical examples would be
for patients with localized pelvic tumors [21]. locally advanced primary or recurrent colorectal
However, dramatic advancements in chemother- cancer invading the sacrum or gynecologic
apy have improved the prognosis for patients malignancy invading the pelvic sidewall or ilium
with osteosarcoma and Ewing’s sarcoma. with no distant tumor spread.
Prior to adjuvant chemotherapy, the survival Musculoskeletal involvement of visceral
of clinically localized osteosarcoma was <15% malignancies had traditionally been considered a
[12]. While specific survival rates for pelvic marker of unresectability. However, by combin-
osteosarcoma in the prechemotherapy era are not ing the advancing understanding of tumor biol-
reliably recorded, these tumors are known to ogy and pelvic resection techniques, extended en
carry an even worse prognosis than extremity bloc resections of visceral disease and involved
tumors, and it is reasonable to infer that long- musculoskeletal structures began in the
id-
m 1980s to provide curative treatment for 2.8 urrent Practice in Pelvic
C
select patients [28, 29]. While initially limited to Tumor Surgery
patients with modest osseous involvement,
expanding experience showed that reasonable Modern imaging now allows reliable determina-
oncologic results and survival could be obtained tion of tumor extent and the overt metastatic sta-
even with extensive resections [30]. At present, tus of patients presenting with pelvic
extended pelvic exenterations (en bloc resection malignancies. As well, current practice provides
of the visceral malignancy and associated muscu- for limb sparing resections in the majority of
loskeletal structures) are now offered at select patients. The common nomenclature for amputa-
cancer centers with reasonable patient morbidity tive resections is either “external hemipelvec-
and oncologic outcome. As is seen in virtually all tomy” or “hindquarter amputation.” Limb sparing
pelvic tumors, margin status is a key determinant resections are termed “internal hemipelvecto-
of outcome, highlighting the role of aggressive mies” and classified as outlined by Enneking and
resections in curative intent procedures. Dunham as to involvement of the iliac bone, ace-
tabulum, or pubic region [16]. Clinical outcome
assessment is still most commonly performed
2.7 Collaboration using the Musculoskeletal Tumor Society rating
and Examination of Results scale [31], although more generalizable patient
reported outcomes are becoming more common.
A key aspect of surgical and scientific advance- Time and institutional practice patterns have
ment is the collaborative sharing and criti- seen different approaches and shifts in the man-
cal examination of results. In parallel with the agement of pelvic sarcoma patients. The initial
development of the field of pelvic sarcoma management of these patients focused on tumor
surgery, several professional organizations removal alone; reconstruction was rarely used
formed to improve progress and better evaluate and difficult with the techniques available [7].
the outcomes of patients with musculoskeletal Recent reports have demonstrated the enduring
malignancies. Notable organizations in this field value of this technique, and it remains a viable
include: surgical option in contemporary practice [32].
However, other centers have demonstrated
• The Musculoskeletal Tumor Society (MSTS, improved functional results with restoration of
est. 1977) femorosacral continuity (anatomic reconstruc-
• The International Society of Limb Salvage tion or substitution) following limb sparing
(ISOLS, est. 1981) resection in the pelvis [33]; this is most difficult
• The European Musculoskeletal Oncology in resections which remove the acetabulum.
Society (EMSOS, est. 1987) A number of different approaches have (and
• The Connective Tissue Oncology Society continue) to be used in these patients. While
(CTOS, est. 1995) cemented and reinforced conventional arthro-
plasty constructs have been reported (commonly
These multidisciplinary professional organi- referred to as the Harrington technique), they are
zations have (and continue) to actively advance most commonly used after surgery for periace-
the practice and understanding of pelvic tumor tabular metastases which typically remove less
surgery. A prime example of this is the evaluation bone than a primary tumor excision with onco-
system for the systematic evaluation of patient logic margins [34]. Iliofemoral arthrodesis was
outcomes initiated at the inception of ISOLS in initially performed for these patients but remained
the 1981 meeting. This culminated in the stan- technically challenging with pseudarthroses and
dard MSTS outcome instrument for evaluating modest functional outcomes [35].
the results of musculoskeletal tumor surgery [31] Early anatomic reconstruction experience uti-
which remains in use to this day. lized massive pelvic allografts or processed
autograft for reconstruction [36]. These recon- closed without the benefit of an amputation
structions were technically demanding and suf- flap. With increasing experience with free flap
fered high complication rates. The saddle coverage and the use of omentum for closure,
prosthesis, an adaptation of an implant for mas- this scenario is becoming less common.
sive bone loss after failed or infected hip arthro- 3. For salvage of patients who experience tumor
plasty, was utilized in tumor resections in an recurrence following internal hemipelvectomy.
attempt to provide a reconstructive option utiliz-
ing a modular endoprosthesis [37]. However, While many teams have been pessimistic
greater experience and follow-up have high- about patient function following external hemi-
lighted the limitations of this method, and its use pelvectomy, modern prosthetic management can
in current practice is rare [38]. allow single hand-free ambulation for many indi-
Modern techniques of periacetabular recon- viduals [45].
struction include modular endoprostheses, cus-
tom prostheses, and porous tantalum implants
[39, 40]. Each of these techniques has relative 2.9 Contemporary Issues
advantages and disadvantages based on resection in Pelvic Tumor Surgery
and remaining bone stock as well as center expe-
rience and preferences. The use of intraoperative Despite the large number of advances made since
navigation or preprinted custom cutting guides the first reported attempt at hemipelvectomy in
allows precise resections to be made to match 1890, a number of areas of pelvic tumor surgery
prefabricated implants. remain unresolved. The need for (and if per-
Not all pelvic resections are commonly consid- formed method of) bony reconstruction after
ered for reconstruction. While reconstructions have major pelvic bone resection remains unclear.
been reported following pubic resections [41], most Reconstruction appears to offer better function at
centers provide soft tissue reconstructions only for the price of higher complications, but selection
these resections. Controversy exists as to whether and center treatment bias clearly influence these
resections of the supra-acetabular ilium require results. True long-term follow-up studies of
reconstruction or not. Some centers advocate for no patients are rare and show an expected decline in
reconstruction to minimize complications and allow function in long-term survivors of their malig-
medialization of the hip center to decrease nancies [46].
Trendelenberg gait (at the expense of leg length dis- The uncertainties of the role and method of
crepancy) [42]. Other centers have shown good reconstruction are magnified in pediatric patients
results with reconstruction of these defects [43]. in whom little published literature exists to guide
Despite advances on many fronts, there surgeons [47]. Most children undergoing major
remains a role for external hemipelvectomy/ pelvic surgery have consideration of reconstruc-
hindquarter amputation in current clinical prac- tion for iliac defects. If the acetabulum is resected,
tice [44]. Patients are currently considered for consideration for reconstruction is given in older
hemipelvectomy in three primary scenarios: adolescents; young patients are generally treated
with resection arthroplasty. While not strictly
1. En bloc resection of a tumor would leave a tabulated, the authors’ clinical experience of this
limb with such little function as to make in young patients is generally favorable.
amputation preferable. This primarily occurs The role of amputation or limb salvage
when tumor extent would require removal of remains controversial. The criteria outlined above
two or three of the critical elements of limb represent the classic criteria for hindquarter
function (the sciatic nerve, the femoral neuro- amputation, but some centers strive to avoid the
vascular bundle, and the acetabulum). morbidity of this by offering limb salvage to
2. Patients in whom resection will result in a soft “borderline” cases. It is not clear which path pro-
tissue defect so large that the wound cannot be vides better functional and oncologic outcomes.
Because of the morbidity of surgical resec- anatomic location in the staging of bone sarco-
tion, many centers try to employ radiotherapy mas (with specific criteria for pelvic tumors) in
when possible. This is most common in patients an attempt to better predict the clinical outcome
with Ewing’s sarcoma in whom local control of these difficult cases [54]. Accumulating data
may be achieved with surgery, radiotherapy, or will hopefully demonstrate whether this change
both. The combination of surgery and radiother- has value in clinical care.
apy appears to decrease the risk of local failure
[48]. Some studies have suggested improved sur-
vival with surgical treatment [25, 26, 49]. 2.10 Conclusions
However, others have not shown a clear benefit
[50]. No studies randomize patients between The field of pelvic tumor has undergone a series
treatment arms. In addition to Ewing’s sarcoma, of advances since the first major pelvic resections
some groups have attempted to employ high-dose were undertaken over a century ago. Modern
proton-based radiotherapy to achieve local con- imaging, improved disease understanding, and
trol of otherwise adversely presenting pelvic sar- adjuvant therapies are the pillars of these
comas with some success [51]. To date, the advancements. However, the morbidity of these
authors’ personal experience with this for non- treatments remains formidable and the prognosis
Ewing’s tumors has been uniformly poor. guarded. Unfortunately, the words of Gordon-
The timing of chemotherapy around major pel- Taylor, a pioneering pelvic tumor surgeon,
vic resections is an area of concern. It is estab- remain true in this field over a half century after
lished in extremity osteosarcoma, for example, they were written [9]:
that delays in resumption of chemotherapy after I still cherish the hope of a golden era of cancer
surgery negatively impact survival [52]. therapy when gross mechanical destruction of dis-
Additionally, a prospective randomized trial ease and cruel mutilation of tissue shall be no
showed no difference in oncologic outcomes in more. Unfortunately, these times are not yet.
osteosarcoma treated with immediate surgery fol-
lowed by chemotherapy compared to a standard
regimen of preoperative chemotherapy, surgery, References
and postoperative chemotherapy [53]. The magni-
tude (and complication profile) of large pelvic 1. Savariaud M. Un cas de disarticulation inter-ilio-
tumor surgeries is such that patients are at high abdominale. Rev Chir (Paris). 1902;26:345–50.
2. Pringle JH. The interpelvi-abdominal amputation. Br
risk to experience significant postoperative delays J Surg. 1916;4:283–96.
in chemotherapy resumption. This has led some 3. Kocher T. Text-book of operative surgery. Stiles H,
centers (including the authors’) to complete most Paul C, trans-editors. London: Adam and Charles
or all chemotherapy prior to surgical resection in Black; 1903. p. 363.
4. Biagini R, Ruggieri P. [Resection of the pelvis due
select pelvic sarcoma patients judged to be at high to bone tumor. The first case treated in the Rizzoli
risk for perioperative complications. It must be Orthopedic Institute in 1914]. Chir Organi Mov.
stated that data regarding this practice are still 1986;71:69–73.
being gathered, and patients undergoing “front- 5. Gordon-Taylor G, Wiles P. Interinnomino-abdominal
(hindquarter) amputation. Br J Surg. 1934;22:671–95.
loading” of chemotherapy are carefully monitored 6. Speed K. Hemipelvectomy. Ann Surg.
with serial imaging studies for disease response. 1932;95:167–73.
Finally, readers should know that the onco- 7. Eilber F, Grant T, Sakai D, Morton D. Internal hemi-
logic staging of pelvic sarcomas has recently pelvectomy – excision of the hemipelvis with limb
preservation. An alternative to hemipelvectomy.
changed. An analysis by the American Joint Cancer. 1979;43:806–9.
Commission on Cancer (AJCC) highlighted the 8. Milch H. Partial resection of the ischium. J Bone Joint
adverse prognosis of axial location on sarcomas. Surg. 1935;17:166–71.
In light of this, the recently released eighth 9. Gordon-Taylor G. On malignant disease in the region
of the hip-joint. J R Coll Surg Edinb. 1959;5:1–21.
Edition AJCC Staging Manual has incorporated
10. Gordon-Taylor G, Wiles P, Patey D, et al. The

27. Brunschwig A. Complete excision of pelvic viscera
Interinnomino-abdominal operation: observations for abdominal carcinoma. Cancer. 1948;1:177–88.
on a series of fifty cases. J Bone Joint Surg (Br). 28. Dowdy S, Mariani A, Cliby W, et al. Radical pelvic
1952;34-B:14–21. resection and intraoperative radiation therapy for
11. Dahlin D, Henderson E. Chondrosarcoma, a surgical recurrent endometrial cancer: technique and analysis
and pathologic problem; review of 212 cases. J Bone of outcomes. Gynecol Oncol. 2006;101:280–6.
Joint Surg Am. 1956;38-A:1025–38. 29.
Magrini S, Nelson H, Gunderson LL, Sim
12. Dahlin D, Coventry M. Osteogenic sarcoma: a study of FH. Sacropelvic resection and intraoperative electron
600 cases. J Bone Joint Surg Am. 1967;49-A:101–10. irradiation in the management of recurrent anorectal
13. Bowden L, Booher R. The principles and technique cancer. Dis Colon Rectum. 1996;39:1–9.
for resection of soft parts for sarcoma. Surgery. 30. Colibaseanu D, Dozois E, Mathis K, Rose P, et al.
1958;44:963–77. Extended sacropelvic resection for locally recur-
14. Enneking W, Spanier S, Goodman M. A system for rent rectal cancer: can it be done safely and with
the surgical staging of musculoskeletal sarcoma. Clin good oncologic outcomes? Dis Colon Rectum.
Orthop Relat Res. 1980;153:106–20. 2014;57:47–55.
15. Hawkins I, Hudson T, Enneking W. Improved mus- 31. Enneking W, Dunham W, Gebhart M, et al. A sys-
culoskeletal angiography with large volume con- tem for the functional evaluation of reconstructive
trast injection and priscoline. Rev Interam Radiol. procedures after surgical treatment of tumors of the
1977;2:83–7. musculoskeletal system. Clin Orthop Relat Res.
16. Enneking W, Dunham W. Resection and reconstruc- 1993;286:241–6.
tion for primary neoplasms involving the innominate 32. Schwartz A, Kiatisevi P, Eilber F, et al. The Friedman-
bone. J Bone Joint Surg Am. 1978;60-A:731–46. Eilber resection arthroplasty of the pelvis. Clin Orthop
17. Carter B, Kahn P, Wolpert S, et al. Unusual pel-
Relat Res. 2009;467:2825–30.
vic masses: a comparison of computed tomo- 33. O’Connor M, Sim F. Salvage of the limb in the

graphic scanning and ultrasonography. Radiology. treatment of pelvic tumors. J Bone Joint Surg Am.
1976;121:383–90. 1989;71:481–94.
18. Hudson T, Schiebler M, Springfield D, et al.
34. Marco R, Sheth D, Boland P, et al. J Bone Joint Surg
Radiologic imaging of osteosarcoma: role in planning Am. 2000;82:642–51.
surgical treatment. Skelet Radiol. 1983;10:137–46. 35. Fuchs B, O’Connor M, Kaufman K, et al. Iliofemoral
19. Sundaram M, McGuire M, Herbold D, et al. Magnetic arthrodesis and pseudarthrosis: a long-term func-
resonance imaging in planning limb-salvage surgery tional outcome evaluation. Clin Orthop Relat Res.
for primary malignant tumors of bone. J Bone Joint 2002;397:29–35.
Surg Am. 1986;68:809–19. 36. Harrington K. The use of hemipelvic allografts or
20. Volker T, Denecke T, Steffen I, et al. Positron emis- autoclaved grafts for reconstruction after wide resec-
sion tomography for staging of pediatric sarcoma tions of malignant tumors of the pelvis. J Bone Joint
patients: results of a prospective multicenter trial. J Surg Am. 1992;74:331–41.
Clin Oncol. 2007;25:5435–41. 37. Aboulafia A, Buch R, Mathews J, et al. Reconstruction
21. Pring M, Weber K, Unni K, Sim F. Chondrosarcoma using the saddle prosthesis following excision of
of the pelvis. A review of sixty-four cases. J Bone primary and metastatic periacetabular tumors. Clin
Joint Surg Am. 2001;83-A:1630–42. Orthop Relat Res. 1995;314:203–13.
22. Cortes E, Holland J, Wang J, et al. Amputation and 38. Donati D, D’Apote G, Boschi M, et al. Clinical and
Adriamycin in primary osteosarcoma. N Engl J Med. functional outcomes of the saddle prosthesis. J Orthop
1974;291:998–1000. Traumatol. 2012;13:79–88.
23. Rosen G, Caparros B, Nirenberg A, et al. Ewing’s 39. Angelini A, Calabro T, Pala E, et al. Resection and
sarcoma: ten-year experience with adjuvant chemo- reconstruction of pelvic bone tumors. Orthopedics.
therapy. Cancer. 1981;47:2204–13. 2015;38:87–93.
24. Tefft M, Razek A, Perez C, et al. Local control and 40. Brown T, Salib C, Rose P, et al. Reconstruction of
survival related to radiation dose and volume and the hip after resection of periacetabular oncologic
to chemotherapy in non-metastatic Ewing’s sar- lesions: a systematic review. Bone Joint J. 2018;100-
coma of pelvic bones. Int J Radiat Oncol Biol Phys. B(1 Supple A):22–30.
1978;4:367–72. 41. Karim S, Colman M, Lozano-Calderon S, et al.

25. Bacci G, Picci P, Gherlinzoni F, et al. Localized
What are the functional results and complications
Ewing’s sarcoma of bone: ten years’ experience at from allograft reconstruction after partial hemi-
the Instituto Ortopedico Rizzoli in 124 cases treated pelvectomy of the pubis? Clin Orthop Relat Res.
with multimodal therapy. Eur J Cancer Clin Oncol. 2015;473:1442–8.
1985;21:163–73. 42. Beadel G, McLaughlin C, Aljassir F, et al. Iliosacral
26. Wilkins R, Pritchard D, Burgert E, Unni K. Ewing’s resection for primary bone tumors: is pelvic
sarcoma of bone. Experience with 140 patients. reconstruction necessary? Clin Orthop Relat Res.
Cancer. 1986;58:2551–5. 2005;438:22–9.
43. Houdek M, Rose P, Bakri K, et al. Outcomes and tumors of the pelvis and sacral bones: the University
complications of reconstruction with use of free vas- of Florida experience. Int J Radiat Oncol Biol Phys.
cularized fibular graft for spinal and pelvic defects 2008;72:41–8.
following resection of a malignant tumor. J Bone Joint 50. Yock T, Krailo M, Fryer C, et al. Local control in
Surg Am. 2017;99:e69. pelvic Ewing sarcoma: analysis from INT-0091 – a
44. Grimer R, Chandrasekar C, Carter S, et al. Hindquarter report from the Children’s Oncology Group. J Clin
amputation: is it still needed and what are the out- Oncol. 2006;24:3838–43.
comes? Bone Joint J. 2013;95-B:127–31. 51. Ciernik I, Niemierko A, Harmon D, et al. Proton-
45. Sherman C, O’Connor M, Sim F. Survival, local
based radiotherapy for unresectable or incompletely
recurrence, and function after pelvic limb salvage resected osteosarcoma. Cancer. 2011;117:4522–30.
at 23 to 38 years followup. Clin Orthop Relat Res. 52. Imran H, Enders F, Krailo M, et al. Effect of time to
2012;470:712–27. resumption of chemotherapy after definitive surgery
46. Kralovek M, Houdek M, Andrews K, et al. Prosthetic on prognosis for non-metastatic osteosarcoma. J Bone
rehabilitation after hip disarticulation or hemipelvec- Joint Surg Am. 2009;91:604–12.
tomy. Am J Phys Med Rehabil. 2015;94:1035–40. 53. Goorin A, Schwartzentruber D, Devidas M, et al.

47. Meazza C, Luksch R, Daolio P, et al. Axial osteosar- Presurgical chemotherapy compared with immedi-
coma: a 25-year monoinstitutional experience in chil- ate surgery and adjuvant chemotherapy for nonmeta-
dren and adolescents. Med Oncol. 2014;31:875. static osteosarcoma: Pediatric Oncology Group Study
48. Ahmed S, Robinson S, Arndt C, et al. Pelvis Ewing POG-8651. J Clin Oncol. 2003;21:1574–80.
sarcoma: local control and survival in the modern era. 54. Kniesl JS, Rosenberg AE, Anderson PM, et al. Bone.
Pediatr Blood Cancer. 2017;64:e26504. In: Amin MB, Edge S, Greene D, et al., editors. AJCC
49. Indelicato D, Keole S, Shahlaee A, et al. Impact of cancer staging manual. 8th ed. Chicago: Springer;
local management on long-term outcomes in Ewing 2018.
Imaging of Pelvic Bone Tumors
3
Andrea Angelini, Joele Canapeti, Giulia Trovarelli,
Joseph Benevenia, and Pietro Ruggieri
3.1 Introduction the imaging characteristics of the most common

pelvic tumors will be discussed.
Imaging of the pelvis can be a challenging task,
especially in the evaluation of tumors and tumor-
like lesion. Numerous primary and secondary 3.2 Hematologic Malignancies
musculoskeletal tumors may affect pelvic bones and Secondary Malignant
and usually many features appear different with Bone Tumors
the same tumors arising in other parts of the body.
Conventional radiographs with multiple views 3.2.1 P
elvic Involvement of Multiple
(inlet, outlet, judet, etc.) represent the first screen- Myeloma and Plasmacytoma
ing approach in the evaluation of osseous lesions
for most symptomatic patients, with the limit of Multiple myeloma is a malignancy of monoclo-
the low sensitivity in detection and diagnosis. A nal plasma cells that represent the second most
correlation with age, history, onset, and duration prevalent blood malignancy (10%) after non-
of symptoms is necessary to raise the clinical sus- Hodgkin’s lymphoma [1]. Bony involvement is
picion. CT and MRI are the most powerful tools very common and pelvic bones are affected in
used for diagnosis, staging, monitoring therapy, 6% of the patients [2]. Lytic bone disease is a
and follow-up. major feature of multiple myeloma, with multi-
Despite the varied appearance and overlap- ple “punched-out” lesions with the absence of
ping radiological features of pelvic tumors, a cor- reactive sclerosis, but sometimes not easily
rect diagnosis should embrace the radiologic detectable at conventional radiographs (Fig. 3.1a)
evaluation with histopathology. In this chapter, [3]. Whole-body low-dose CT (LDCT), PET/CT,
and MRI have a relevant role in the novel diag-
nostic criteria for symptomatic multiple myeloma
A. Angelini · J. Canapeti · G. Trovarelli (Fig. 3.1b, c) [4]. PET/CT demonstrates a signifi-
P. Ruggieri (*) cant higher sensitivity compared with whole-
Department of Orthopedics and Orthopedic
Oncology, University of Padova, Padova, Italy body X-ray for the detection of osteolytic lesions
e-mail: [email protected]; in multiple myeloma [5] and has an independent
[email protected] prognostic value both at diagnosis and after treat-
J. Benevenia ment [6]. Plasmacytoma is a focal, solitary prolif-
Department of Orthopaedics, Rutgers New Jersey eration of plasma cells that seems to be the early
Medical School, Newark, NJ, USA stage of a multiple myeloma.

a b c
Fig. 3.1 Multiple myeloma in a 54-year old woman. (a) gadolinium demonstrate multiple areas of diffuse
Pelvic plain radiograph is not adequate to show the lytic enhancement in the bone marrow of the sacrum and pelvic
lesion in the right iliac bone. No fracture detected and the bones and a soft-tissue mass (white arrow) with symmet-
hip joint demonstrates mild degenerative changes. (b) ric growth outside the right ilium. Note the absence of
Coronal and (c) Axial T1-weighted fat-saturated MRI of cortical disruption
the pelvis obtained after intravenous administration of
3.2.2 Metastases myeloma and osteosarcoma [13], and pelvis is

the most common location [14]. Even if rare,
Metastases are the most common malignant chondrosarcomas may also affect sacral bone
tumors and can derive mainly from breast, lung, [15]. It is a malignant cartilaginous matrix-
prostate, kidney, and thyroid cancers. The inci- producing lesion with a typical progression from
dence of symptomatic bone metastasis affecting a low to high-grade tumor [16, 17]. They may
the pelvis is increasing because of advances in arise from an enchondroma (central chondrosar-
diagnostic work-up, improvement of overall comas) or from an osteochondroma (peripheral
prognosis under chemo-, immune-, and radio- chondrosarcomas), but fast growth may suggest a
therapy [7–9]. These secondary tumors should dedifferentiated chondrosarcoma [18]. Previous
always be considered in the differential diagnosis studies reported several histological parameters
when aggressive lesions are observed in pelvic including grade, tumor necrosis, mitotic count,
bones. The discovery of solitary lesion requires and myxoid tumor matrix for predicting the
further analysis to exclude other primary tumors behavior of the tumor and the prognosis for the
[10]. Osseous metastases may be lytic (more patients, even if also low-grade tumors should be
common), sclerotic, or mixed mainly based on treated with an aggressive surgical approach [17].
histologic subtype (Fig. 3.2). Isotope scan, PET/ On plane radiographs, primary cartilage
CT, and MRI are very sensitive in detecting bone tumors in the pelvis should be approached with
metastases and are mandatory for a complete more caution than those of the extremities,
staging at time of diagnosis [11, 12]. because it is not possible to use the same diag-
nostic criteria to differentiate enchondromas
from low-grade chondrosarcomas. Low-grade
3.3 rimary Malignant Bone
P chondrosarcoma appears as a destructive lytic
Tumors lesion with a lobulated contour, well-defined
margins, endosteal scalloping, and may have cor-
3.3.1 Chondrosarcoma tical expansion. High-grade tumors are usually
diagnosed in stage IIB (based on Enneking clas-
Chondrosarcoma accounts for approximately sification) with cortical destruction, periosteal
20% of malignant bone tumors. It is the third reaction, and an associated soft tissue mass
most common primary malignancy of bone after (Fig. 3.3a, b). The chondroid matrix can show
3 Imaging of Pelvic Bone Tumors 25
a b c
Fig. 3.2 Metastatic renal carcinoma in a 76-year old posing it to a pathological fracture. (b) Coronal and (c)
man. (a) Pelvic plain radiograph demonstrating a predom- Axial pelvic CT Scan showing involvement of the right
inant lytic metastatic lesion (white arrow) in the right acetabular region by the large lytic metastatic lesion. Note
ischium. Note the extensive cortical involvement, predis- the extensive extraosseous involvement (asterisk)
typical “ring and arc” calcifications that are more 3.3.3 Osteosarcoma
evident on CT scans, whereas the nonmineralized
cartilaginous portion of the tumor has a low den- Conventional osteosarcoma rarely affects the pel-
sity. On MRI, the lobules are iso-hypointense on vis, with approximately 8% of all the sites, even if
T1-w images (Fig. 3.3c, d) and have high inten- it accounts for 22% of all primary pelvic bone
sity on T2-w images (Fig. 3.3e, f). The presence malignant tumors [26, 27]. Most of the tumors are
of a large lytic aggressive area adjacent to a carti- secondary, occurring after radiation therapy or in
lage tumor may suggest the diagnosis of dedif- Paget’s disease [28–30]. The plain radiographs are
ferentiated chondrosarcoma, especially if matrix usually diagnostic, with an aggressive permeative
mineralization with a bimorphic pattern on CT pattern, combination of radiolucency and
scan is observed [18, 19]. radiodensity, cortical disruption, and soft tissue
involvement (Fig. 3.5a). On CT, most pelvic osteo-
sarcomas contain “cloud-like” osteoid matrix for-
3.3.2 Ewing Sarcoma mation and show the sunray image (stripes of
density perpendicular to the cortex) (Fig. 3.5b, c)
The imaging appearance of Ewing sarcoma of [27–31]. The telangiectatic osteosarcoma appears
the sacrum and the pelvis is similar to that of the as predominantly lytic bone mass with minimal
extremities, even if it is variable [20, 21]. The sclerosis on radiographs [27, 32]. MRI shows no
tumor usually fills the bone marrow cavity and specific features with the usual pattern of low
destroys the cortex with a moth-eaten and perme- T1-w and high T2-w signal, with heterogeneous
ative pattern on plain radiographs (Fig. 3.4a) and contrast enhancement (Fig. 3.5d, e) [27].
CT. There is often a soft tissue mass associated to Sometimes fluid-fluid levels may be present, espe-
the tumor and a classic sclerotic reaction with a cially in predominantly lytic lesions [32, 33].
concentric expansion called “onion-skin” appear-
ance may be present. CT scans and MRI play a
predominant role in evaluating the soft tissue and 3.4 Benign Tumors
bony extension of the lesion, and in evaluation of
response to adjuvant therapies and surgical plan 3.4.1 Giant Cell Tumor
[22]. MRI features are nonspecific: the tumor is
iso-hypointense on T1-w, with increased signal Giant cell tumors (GCTs) are benign but locally
intensity on T2-w, with variable contrast enhance- aggressive tumors that rarely affect the pelvic
ment (Fig. 3.4b) [23, 24]. PET/CT and dynamic bones (1.5–6.1% of bone GCTs) [34, 35]. On the
MRI are under evaluation as imaging tools for other hand, GCTs are the second most frequent
restaging and tumor response to primary chemo- primary tumors of the sacrum after chordoma
therapy [11, 25]. [36, 37]. On radiographs and CT scans, GCTs are
a b
c d
e f
Fig. 3.3 Chondrosarcoma of the pelvis in a 60-year old a tumor involving the entire acetabulum and lobular soft
woman. (a) Antero-posterior and (b) axial radiographs of tissue extension with iso/hypointense signal. (e) Coronal
the right hip reveal a periacetabular area of lucency with and (f) Axial T2-weighted fat-saturated MR images dem-
surrounding sclerosis and not well-defined calcifications. onstrate a lobulated T2 hyperintense mass compatible
(c) Coronal and (d) Axial T1-weighted MR images show with a cartilage tumor
a b
Fig. 3.4 Ewing’s sarcoma in a 13-year old female. (a) ciation with an overt soft tissue component. (b)
Pelvic plain radiograph shows an expansile bony lesion Gadolinium-enhanced fat-saturated MRI coronal image
centered within the left pubic bone (white arrow) up to the reveals an extensive soft tissue mass with necrotic areas
acetabular area. The lesion did not demonstrate gross cor- and involvement of the adjacent anatomic structures
tical disruption, aggressive periosteal reaction, or an asso-
lytic lesions that appear usually more destructive eccentric, well-defined lytic lesions with a thin
than typically seen in long bones, often with a peripheral rim of sclerosis. Cystic space with
soft tissue mass. Cortex is usually destroyed and fluid-fluid levels and contrast enhancement of the
tumor mass do not present other typical features septa are hallmark features of ABCs on CT and
of extremities GCTs such as sclerotic rim, perios- MRI images. MRI shows a high signal intensity
titis, and mineralization [38]. On MRI, GCTs of the fluid-fluid levels on T1-w sequences with a
usually demonstrate low-signal on T1, but may strong contrast enhancement due to the intense
have a significant heterogeneity on T2-w and vascularization. Telangiectatic osteosarcomas
fluid-sensitive sequences due to hemorrhage or represent the main differential diagnosis, espe-
necrosis [38]. The presence of fibrous compo- cially in lesions with aggressive radiographic
nents and hemosiderin gives the predominantly appearance, cortical destruction, and a soft tissue
low to intermediate signal on T2, whereas fluid extension.
and cystic changes determine the increased T2
signal. Secondary aneurysmal bone cyst may
result in prominent areas with extensive fluid- 3.4.3 Osteochondroma
fluid levels [39].
Osteochondromas are frequently observed in
growing skeleton, especially in metaphysis of the
3.4.2 Aneurysmal Bone Cyst long bones, but every bone can be affected. Ilium
represents the most common site in the pelvis
Aneurysmal bone cyst (ABC) is relative rare [42]. The imaging features are characteristic:
benign expansile osteolytic bone lesion with bony excrescence with well-defined limits
blood-filled cystic spaces. Recent studies demon- (Fig. 3.6a), sessile or pedunculated growth with
strated a neoplastic origin in primary ABC in the continuity of the cortex, and medullary canal
rearrangement of the TRE17/USP6 locus occurs (Fig. 3.6b, c). In the pelvis, these lesions are
resulting in TRE17 overexpression [40]. Flat found incidentally [43]. Differential diagnosis
bones are frequently involved and the pelvis is a with peripheral chondrosarcomas is mandatory
common site, accounting for up to 50% of cases and mainly based on imaging features. In favor of
[41]. On radiographs and CT, ABCs appear as malignancy are size (>5 cm), thick cartilaginous
a b
c d
Fig. 3.5 Osteosarcoma in a 15-year old male. (a) Pelvic Coronal and (e) Axial fat suppressed Gd-chelate enhanced
plain radiograph demonstrates an irregularly calcified MR image show asymmetric soft tissue extension. Tumor
lesion that involves the left hemipelvis. (b) Coronal and is seen to cross the greater sciatic foramen compressing
(c) Axial CT scan show a large associated soft tissue mass and dislocating the anatomic structures inside the pelvis
with cloudy-like areas of matrix mineralization and stripes (asterisk)
of density perpendicular to the cortex (arrowhead). (d)
a b
Fig. 3.6 Solitary osteochondroma in a 25-year old man. bone (arrowhead). (c) Axial T1-w MRI image of the same
(a) Pelvic plain radiograph demonstrates an irregularly lesion demonstrates the continuity of the cortex and med-
calcified, pedunculated lesion (white arrow) arising from ullary portion with the parent bone (arrowhead) and iden-
the right iliac crest. (b) Axial CT scan shows the periph- tifies a thin cartilage cap (asterisk)
eral outgrowth with its cortex in continuity with the iliac
cap (>2 cm), poorly defined cap, irregular calcifi- ostotic, in the latter case associated with multiple
cations, and rapid enhancement on dynamic endocrine abnormalities (McCune-Albright’s syn-
Gd-enhanced MR images (less than 10 s after drome) or intramuscular mixomas (Mazabraud’s
arterial enhancement) [44]. syndrome). Usually asymptomatic diagnosed as
incidental finding, sometimes may cause patho-
logic fracture, discontinuous pain, deformity or
3.4.4 Fibrous Dysplasia lower limb discrepancy. The radiologic presenta-
tion is the same as in extremities, with well-defined
Fibrous dysplasia is an intramedullary hamartoma defect rounded by a rind of bone sclerosis, radiolu-
commonly observed in proximal femur and in pel- cency with “ground glass” appearance, thin and
vic bones [45]. It may be either monostotic or poly- expanded cortex without periosteal reaction
a b
Fig. 3.7 Monostotic fibrous dysplasia in a 30-year old heterogeneous lesion that is largely hyperintense with
woman. (a) Lesions in the right ilium (white arrow) with hypointense foci within the matrix. All images show the
mixed lysis and sclerosis with peripheral bone sclerosis. well-defined character and the absence of cystic
(b) Axial and (c) coronal T2-weighted fat sat image show components
criteria for the diagnosis of multiple myeloma. Lancet

(Fig. 3.7). Cystic cavities and cartilaginous area Oncol. 2014;15:e538–48.
may be present and easily detectable on MRI, 5. Regelink JC, Minnema MC, Terpos E, Kamphuis
MH, Raijmakers PG, Pieters-van den Bos IC,
whereas the lesion appears with a fairly homoge- Heggelman BG, Nievelstein RJ, Otten RH, van
neous low-signal in T1-w images. Lammeren-Venema D, Zijlstra JM, Arens AI, de Rooy
JW, Hoekstra OS, Raymakers R, Sonneveld P, Ostelo
RW, Zweegman S. Comparison of modern and con-
ventional imaging techniques in establishing multiple
References myeloma-related bone disease: a systematic review.
Br J Haematol. 2013;162:50–61.
1. Mavrogenis AF, Angelini A, Pala E, Zinzani P, Ruggieri 6. Nanni C, Versari A, Chauvie S, Bertone E, Bianchi
P. The role of surgery for haematologic neoplasms of A, Rensi M, Bellò M, Gallamini A, Patriarca F, Gay
bone. Acta Orthop Belg. 2012;78(3):382–92. F, Gamberi B, Ghedini P, Cavo M, Fanti S, Zamagni
2. Sakellariou VI, Mavrogenis AF, Savvidou O, Sim E. Image interpretation criteria for FDG PET/CT in
FH, Papagelopoulos PJ. Reconstruction of multiple multiple myeloma: a new proposal from an Italian
myeloma lesions around the pelvis and acetabulum. expert panel. IMPeTUs (Italian Myeloma crite-
Eur J Orthop Surg Traumatol. 2015;25(4):643–53. ria for PET USe). Eur J Nucl Med Mol Imaging.
https://doi.org/10.1007/s00590-014-1555-4. 2015;43:414–21.
3. Mihailovic J, Goldsmith SJ. Multiple myeloma: 7. Ruggieri P, Mavrogenis AF, Angelini A, Mercuri
18F-FDG-PET/CT and diagnostic imaging. M. Metastases of the pelvis: does resection improve
Semin Nucl Med. 2015;45(1):16–31. https://doi. survival? Orthopedics. 2011;34(7):e236–44. https://
org/10.1053/j.semnuclmed.2014.08.002. doi.org/10.3928/01477447-20110526-07.
4. Rajkumar SV, Dimopoulos MA, Palumbo A, Blade 8. Guzik G. Treatment of metastatic lesions localized in
J, Merlini G, Mateos MV, Kumar S, Hillengass J, the acetabulum. J Orthop Surg Res. 2016;11(1):54.
Kastritis E, Richardson P, Landgren O, Paiva B, https://doi.org/10.1186/s13018-016-0384-z.
Dispenzieri A, Weiss B, LeLeu X, Zweegman S, 9. Mavrogenis AF, Angelini A, Vottis C, Pala E, Calabrò
Lonial S, Rosinol L, Zamagni E, Jagannath S, Sezer T, Papagelopoulos PJ, Ruggieri P. Modern palliative
O, Kristinsson SY, Caers J, Usmani SZ, Lahuerta JJ, treatments for metastatic bone disease: awareness
Johnsen HE, Beksac M, Cavo M, Goldschmidt H, of advantages, disadvantages, and guidance. Clin J
Terpos E, Kyle RA, Anderson KC, Durie BG, Miguel Pain. 2016;32(4):337–50. https://doi.org/10.1097/
JF. International Myeloma Working Group updated AJP.0000000000000255.
10. Angelini A, Calabrò T, Pala E, Trovarelli G, Maraldi 23.

Gronemeyer SA, Kauffman WM, Rocha MS,
M, Ruggieri P. Resection and reconstruction of pelvic Steen RG, Fletcher BD. Fat-saturated contrast-
bone tumors. Orthopedics. 2015;38(2):87–93. https:// enhanced T1-weighted MRI in evaluation of
doi.org/10.3928/01477447-20150204-51. osteosarcoma and Ewing sarcoma. J Magn Reson
11. Angelini A, Castellucci P, Ceci F. Future per-
Imaging. 1997;7(3):585–9. https://doi.org/10.1002/
spective of the application of positron emission jmri.1880070321.
tomography- computed tomography-MR imaging in 24. Frassica FJ, Frassica DA, Pritchard DJ, Schomberg
musculoskeletal disorders. PET Clin. 2019;14(1):183– PJ, Wold LE, Sim FH. Ewing sarcoma of the pelvis.
91. https://doi.org/10.1016/j.cpet.2018.08.012. Epub Clinicopathological features and treatment. J Bone
2018 Oct 24. Joint Surg Am. 1993;75(10):1457–65.
12. Yoon MA, Hong SJ, Lee KC, Lee CH. Contrast-
25. Raciborska A, Bilska K, Drabko K, et al. Response
enhanced magnetic resonance imaging of pelvic to chemotherapy estimates by FDG PET is an impor-
bone metastases at 3.0 T: comparison between tant prognostic factor in patients with Ewing sarcoma.
3-dimensional T1-weighted CAIPIRINHA-VIBE Clin Transl Oncol. 2016;18(2):189–95. https://doi.
Sequence and 2-dimensional T1-weighted Turbo org/10.1007/s12094-015-1351-6.
Spin-Echo Sequence. J Comput Assist Tomogr. 26. Rosenberg ZS, Lev S, Schmahmann S, Steiner GC,
2019;43(1):46–50. https://doi.org/10.1097/ Beltran J, Present D. Osteosarcoma: subtle, rare, and
RCT.0000000000000766. misleading plain film features. AJR Am J Roentgenol.
13. Giuffrida AY, Burgueno JE, Koniaris LG, Gutierrez 1995;165(5):1209–14. https://doi.org/10.2214/
JC, Duncan R, Scully SP. Chondrosarcoma in the ajr.165.5.7572505.
United States (1973 to 2003): an analysis of 2890 27. Park SK, Lee IS, Cho KH, Lee YH, Yi JH, Choi
cases from the SEER database. J Bone Joint Surg KU. Osteosarcoma of pelvic bones: imaging fea-
Am. 2009;91(5):1063–72. https://doi.org/10.2106/ tures. Clin Imaging. 2017;41:59–64. https://doi.
JBJS.H.00416. org/10.1016/j.clinimag.2016.10.013. Epub 2016 Oct
14. Murphey MD, Walker EA, Wilson AJ, Kransdorf
17.
MJ, Temple HT, Gannon FH. From the archives 28. Parry MC, Laitinen M, Albergo J, Jeys L, Carter S,
of the AFIP: imaging of primary chondrosarcoma: Gaston CL, Sumathi V, Grimer RJ. Osteosarcoma
radiologic-pathologic correlation. Radiographics. of the pelvis. Bone Joint J. 2016;98-B(4):555–63.
2003;23:1245–78. https://doi.org/10.1302/0301-620X.98B4.36583.
15.
Angelini A, Mavrogenis AF, Ruggieri 29. Calabrò T, Mavrogenis AF, Ruggieri P. Osteoblastic
P. Chondrosarcoma of the sacrum. In: Ruggieri P, osteosarcoma in monostotic Paget’s disease.
Angelini A, Vanel D, Picci P, editors. Tumors of the Musculoskelet Surg. 2011;95(1):37–40. https://doi.
sacrum. Cham: Springer; 2017. org/10.1007/s12306-011-0100-4.
16.
Pring ME, Weber KL, Unni KK, Sim 30. Mavrogenis AF, Angelini A, Pala E, Calabro T,

FH. Chondrosarcoma of the pelvis. A review Bianchi G, Casadei R, Ruggieri P. Radiation-
of sixty-four cases. J Bone Joint Surg Am. induced sarcomas. J Long-Term Eff Med Implants.
2001;83-A(11):1630–42. 2011;21(3):233–40. https://doi.org/10.1615/jlongter-
17. Angelini A, Guerra G, Mavrogenis AF, Pala E, Picci P, meffmedimplants.v21.i3.70.
Ruggieri P. Clinical outcome of central conventional 31. Murphey MD, Robbin MR, McRae GA, Flemming
chondrosarcoma. J Surg Oncol. 2012;106(8):929–37. DJ, Temple HT, Kransdorf MJ. The many faces of
18. Henderson ER, Pala E, Angelini A, Rimondi E,
osteosarcoma. Radiographics. 1997;17(5):1205–31.
Ruggieri P. Dedifferentiated peripheral chon- https://doi.org/10.1148/radiographics.17.5.9308111.
drosarcoma: a review of radiologic characteris- 32. Angelini A, Mavrogenis AF, Trovarelli G, Ferrari

tics. Sarcoma. 2013;2013:505321. https://doi. S, Picci P, Ruggieri P. Telangiectatic osteosarcoma:
org/10.1155/2013/505321. a review of 87 cases. J Cancer Res Clin Oncol.
19. Littrell LA, Wenger DE, Wold LE, Bertoni F, Unni KK, 2016;142(10):2197–207. https://doi.org/10.1007/
White LM, Kandel R, Sundaram M. Radiographic, s00432-016-2210-8.
CT, and MR imaging features of dedifferentiated 33. O’Donnell P, Saifuddin A. The prevalence and diag-
chondrosarcomas: a retrospective review of 174 de nostic significance of fluid-fluid levels in focal lesions
novo cases. Radiographics. 2004;24(5):1397–409. of bone. Skelet Radiol. 2004;33(6):330–6.
20. Bloem JL, Reidsma II. Bone and soft tissue tumors of 34. Sanjay BK, Frassica FJ, Frassica DA, Unni KK,

hip and pelvis. Eur J Radiol. 2012;81(12):3793–801. McLeod RA, Sim FH. Treatment of giant-cell tumor
21. Angelini A, Letson DG, Ruggieri P. Ewing’s sarcoma of the pelvis. J Bone Joint Surg Am. 1993;75:1466–75.
of the sacrum. In: Ruggieri P, Angelini A, Vanel D, 35. Zheng K, Wang Z, Wu SJ, Ye ZM, Xu SF, Xu M, Hu
Picci P, editors. Tumors of the sacrum. Cham: Springer; YC, Yu XC. Giant cell tumor of the pelvis: a system-
2017. https://doi.org/10.1007/978-3-319-51202-0_17. atic review. Orthop Surg. 2015;7(2):102–7. https://
22. Huang WY, Tan WL, Geng DY, Zhang J, Wu G,
doi.org/10.1111/os.12174.
Zhang BY, Li YX, Yin B. Imaging findings of the spi- 36. Ruggieri P, Mavrogenis AF, Ussia G, Angelini A,

nal peripheral Ewing’s sarcoma family of tumours. Papagelopoulos PJ, Mercuri M. Recurrence after
Clin Radiol. 2014;69(2):179–85. and complications associated with adjuvant treat-
ments for sacral giant cell tumor. Clin Orthop Relat 41.
Vergel De Dios AM, Bond JR, Shives TC,
Res. 2010;468(11):2954–61. https://doi.org/10.1007/ McLeod RA, Unni KK. Aneurysmal bone cyst.
s11999-010-1448-8. A clinicopathologic study of 238 cases. Cancer.
37. Mavrogenis AF, Panagopoulos GN, Angelini A,
1992;69(12):2921–31. https://doi.org/10.1002/1097-
Ruggieri P. Giant cell tumor of the sacrum. In: 0142(19920615)69:12<2921::aid-cncr2820691210>3
Ruggieri P, Angelini A, Vanel D, Picci P, editors. .0.co;2-e.
Tumors of the sacrum. Cham: Springer; 2017. https:// 42. Picci P, Vanel D, Gambarotti M, Fabbri N, Manfrini
doi.org/10.1007/978-3-319-51202-0_8. M, editors. Atlas of musculoskeletal tumors and
38. Murphey MD, Nomikos GC, Flemming DJ, Gannon tumor-like lesions: the Rizzoli case archive. Cham:
FH, Temple HT, Kransdorf MJ. From the archives of Springer; 2014.
AFIP. Imaging of giant cell tumor and giant cell repar- 43. Girish G, Finlay K, Morag Y, Brandon C, Jacobson
ative granuloma of bone: radiologic-pathologic corre- J, Jamadar D. Imaging review of skeletal tumors
lation. Radiographics. 2001;21(5):1283–309. https:// of the pelvis—Part I: Benign tumors of the pelvis.
doi.org/10.1148/radiographics.21.5.g01se251283. ScientificWorldJournal. 2012;2012:290930. https://
39. Gutierrez LB, Link TM, Horvai AE, Joseph GB,
doi.org/10.1100/2012/290930.
O’Donnell RJ, Motamedi D. Secondary aneurysmal 44. Brien EW, Mirra JM, Luck JV Jr. Benign and malig-
bone cysts and associated primary lesions: imaging nant cartilage tumors of bone and joint: their anatomic
features of 49 cases. Clin Imaging. 2020;62:23–32. and theoretical basis with an emphasis on radiology,
https://doi.org/10.1016/j.clinimag.2020.01.022. pathology and clinical biology. II. Juxtacortical carti-
40. Lau AW, Pringle LM, Quick L, Riquelme DN, Ye Y, lage tumors. Skelet Radiol. 1999;28(1):1–20.
Oliveira AM, Chou MM. TRE17/ubiquitin- specific 45. Ruggieri P. Fibrous dysplasia. In: Picci P, Manfrini
protease 6 (USP6) oncogene translocated in aneu- M, Fabbri N, Gambarotti M, Vanel D, editors.
rysmal bone cyst blocks osteoblastic maturation Atlas of musculoskeletal tumors and tumor-
via an autocrine mechanism involving bone mor- like lesions. Cham: Springer; 2014. https://doi.
phogenetic protein dysregulation. J Biol Chem. org/10.1007/978-3-319-01748-8_7.
2010;285(47):37111–20. https://doi.org/10.1074/jbc.
M110.175133.
Most Common Histological Type
of Pelvic Bone Tumors 4
Marilyn M. Bui and Andrew E. Rosenberg
4.1 Introduction with the literature review and authors’ institutional

experience, the pathology of 14 most common his-
Previous study of 6000 patients had shown that tological type of pelvic bone tumors will be dis-
patients with pelvic tumors are usually older, and cussed in the order of illustrating the gross and
their tumors are larger relative to patients with histological features of these tumors, highlighting
tumors in extremities. The majority of tumors in the ancillary testing of diagnostic, prognostic, and
the pelvis are malignant (metastases, myeloma, predictive markers, and addressing the collabora-
chondrosarcoma, Ewing sarcoma, osteosarcoma, tive opportunities between pathologists and ortho-
and malignant fibrous histiocytoma (MFH)/fibro- pedic surgeons to improve the quality, safety, and
sarcoma) [1]. The most frequent primary bone value of patient care.
tumors of pelvis include chondrosarcoma (24%), Before getting into the details of each tumor
Ewing sarcoma (16%), osteosarcoma (9%), malig- type, few updates are worth mentioning. For
nant fibrous histiocytoma (MFH)/fibrosarcoma malignant bone tumor staging, the American
(5%), Langerhans cell histocytosis (4%), aneurys- Joint Committee on Cancer (AJCC) eighth edi-
mal bone cyst (4%), fibrous dysplasia (4%), benign tion published in 2017 is generally used [2]. The
miscellaneous bone tumors (25%), and miscella- updates relevant to pelvic bone tumor include:
neous malignant bone tumor (8%) [1]. Combine (1) Pelvis has a separate and distinct TNM clas-
sification but not a separate stage grouping. See
Table 4.1 [2]. (2) Multiple myeloma and primary
malignant lymphoma not staged using this stag-
ing system but rather the plasma cell disorders
M. M. Bui (*) and lymphoma staging system. (3) Stage III is
Department of Pathology, H. Lee Moffitt Cancer reserved for grade 2 (G2) and grade 3 (G3). (4)
Center and Research Institute, Tampa, FL, USA Grade 4 (G4) has been eliminated. Grade 1 (G1)
Department of Oncological Sciences, Morsani is for low grade while G2 and G3 are for high
College of Medicine at the University of South grade. For the definition of the histological tumor
Florida, Tampa, FL, USA types, currently we are using the WHO
Classification of Tumors of Soft Tissue and Bone
A. E. Rosenberg fourth edition where malignant fibrous histocy-
Department of Pathology,
University of Miami Health System, toma (MFH) is replaced by undifferentiated
Miami, FL, USA high-grade pleomorphic sarcoma and separated
e-mail: [email protected] from fibrosarcoma which is a distinct entity.

34 M. M. Bui and A. E. Rosenberg
Table 4.1 AJCC Staging Version 8 for pelvis bone Table 4.2 Conventional Chondrosarcoma grading
tumors
Grade Histology Prognosis
Primary tumor Criteria I Moderately cellular, mildly Locally
TX Primary tumor cannot be assessed atypical nuclei without visible aggressive,
T0 No evidence of primary tumor nucleoli under low rare
T1 Tumor confined to one pelvis segment magnification, no mitosis metastasis,
with no extraosseous extension good
prognosis
T1a Tumor ≤8 cm in greatest dimension
II More cellular and greater Poor
T1b Tumor >8 cm in greatest dimension
degree of nuclear atypia than prognosis
T2 Tumor confined to one pelvis segment grade I, nucleoli can be found
with extraosseous extension or two under high magnification,
segments without extraosseous mitoses are seen
extension
III Hypercellular, markedly Worst
T2a Tumor ≤8 cm in greatest dimension atypical nuclei with prognosis
T2b Tumor >8 cm in greatest dimension hyperchromasia, irregular
T3 Tumor spanning two pelvic segments nuclear contour, and
with extraosseous extension enlargement, prominent
T3a Tumor ≤8 cm in greatest dimension nucleoli easily visible under
T3b Tumor >8 cm in greatest dimension low magnification, frequent
Regional mitosis including atypical
lymph nodes figures
NX Lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Regional lymph node metastasis
rized in Table 4.2. Myxoid change is seen in
Distant grade II and III tumors.
metastasis Conventional chondrosarcoma can undergo
M0 No distant metastasis dedifferentiation and give rise to dedifferentiated
M1 Distant metastasis chondrosarcoma variant. Grossly the demarca-
M1a Lung tion of the cartilaginous and noncartilaginous
M1b Other distant sites components is readily recognizable. The dedif-
ferentiate component is typically a high-grade
undifferentiated sarcoma. Histology of a dedif-
4.2 Chondrosarcoma ferentiated chondrosarcoma is illustrated in
Fig. 4.1a, b. The prognosis of this tumor is dis-
Chondrosarcomas are a heterogeneous group, mal. Heterozygous mutations of the isocitrate
which includes the primary central, secondary dehydrogenase 1 and 2 genes (IDH1 and IDH2)
central, periosteal, dedifferentiated, mesenchy- are found in conventional chondrosarcoma as
mal, and clear cell variants. The distinguishing well as dedifferentiated chondrosarcoma (includ-
hallmark of chondrosarcoma is the tumor cells ing the noncartilaginous component) [4]. This
producing cartilaginous matrix. The most com- biomarker is potentially useful to distinguish a
mon primary central chondrosarcoma is conven- chondrosarcoma from a chondroblastic osteosar-
tional type. Grossly the tumor has the appearance coma. PB1 pathway (TP53) mutation is also a
of hyaline cartilage. The histological criteria used common event in both tumor types [4]. The fea-
for diagnosis include high cellularity, permeation tures of rare variants of chondrosarcoma are sum-
of cortical and/or medullary bone, dysplastic marized in Table 4.3.
chondrocytes, myxoid matrix or chodroid matrix
liquefaction, necrosis, and increased mitotic
activity [3]. The histological grade is the single 4.3 Ewing Sarcoma
most important prognostic factor of local recur-
rence and metastasis. The grading criteria for Ewing sarcoma is a family of tumor with neu-
grade I to III based on the cellularity, nuclear fea- roectodermal origin. It is a high-grade malig-
tures, and mitosis and the prognosis are summa- nancy with small, blue, and round tumors cells
4 Most Common Histological Type of Pelvic Bone Tumors 35
a b
Fig. 4.1 Dedifferentiated chondrosarcoma. (a) Conventional composed of high-grade malignant spindle cells without car-
low-grade chondrosarcoma component with cartilaginous tilaginous matrix
matrix. (b) Dedifferentiated chondrosarcoma component
Table 4.3 Summary of histological variants of Table 4.4 Summary of Ewing sarcoma gene fusions
chondrosarcoma Translocation Genes involved
Tumor type Component Prognosis t(11;22)(q24;q12) EWSR1-FLI1
Conventional Chondrosarcoma Depends t(21;22)(q22;q12) EWSR1-ERG
Grade I on grade t(2;22)(q33;q12) EWSR1-FEV
Grade II t(7;22)(p22;q12) EWSR1-ETV1
Grade III t(17;22)(q12;q12) EWSR1-E1AF
Dedifferentiated Low-grade conventional Poor inv(22)(q12q;12) EWSR1-ZSG
chondrosarcoma plus
t(16;21)(p11;q22) FUS-ERG
high-grade
dedifferentiated
sarcoma or
osteosarcoma Recently, a rare group of Ewing-like sarcoma
Mesenchymal Low-grade conventional Poor is recognized and classified as undifferentiated
chondrosarcoma plus round cell sarcoma by its genomic distinction.
poorly differentiated
malignant small round These tumors exhibit t(4;19)(q35;q13) or t(10;19)
cells (q26;q13) with CIC-DUX4 gene fusion or inv(X)
Clear cell Clear cells or Depends (p11.4p11.22) with BCOR-CCNB3 fusion. Their
chondroblastoma-like on grade clinical behavior is also different from classic
cells
Ewing sarcoma [7, 8].
Grossly, the tumor has tan-gray cut surface
showing pathognomonic molecular signatures without osteoid or cartilaginous matrix. A classic
which are pathogenesis drivers, including Ewing sarcoma is composed of hypercellular
approximately 85% harboring a somatic chro- small round cells with scant cytoplasm and round
mosomal translocation t(11;22)(q24;q12) nuclei arranged in a dyscohesive pattern.
resulted in EWSR1-FLI1 gene fusion [5]. This Neuroectodermal differentiation can be seen with
and other genes involved in Ewing sarcoma are tumor cells forming rosette-like structures. This
summarized in Table 4.4. Molecular testing for small blue round cell pattern resembles lym-
the signature gene and products are useful in phoma, osteosarcoma, rhabdomyosarcoma, and
confirming the diagnosis. Molecular markers, poorly differentiated carcinoma or neuroendo-
such as TP53, telomerase expression, or crine carcinoma. However, the cytoplasm of
CDKN2A loss, have shown prognostic signifi- Ewing sarcoma appears clear and contains glyco-
cance [6]. gen, which is stained positive by periodic acid-
Schiff (PAS). Immunohistochemical stain pattern more pleomorphic nuclei and prominent nucle-
of Ewing sarcoma includes positive vimentin, oli. A radiological, gross, and histological illus-
CD99 (membranous), Keratin (aberrantly tration of Ewing sarcoma is in Fig. 4.2a–e.
expressed in 30% cases), neuroendocrine mark- Intraoperative evaluation of pathologic frac-
ers (aberrant expression sometimes), FLI-1, ture of pelvis caused by Ewing sarcoma can be
ERG, and NKX2.2 [9]. Comparing to lymphoma, challenging, especially when there is no prior
Ewing sarcoma lacks the lymphoglandular bod- biopsy was performed. In our hands, touch prep
ies which are cytoplasmic debris of lymphoma. cytology has been routinely used intraoperatively
Osteosarcoma, especially the small cell variant, in conjunction to frozen section to facilitate a
produces osteoid and lacks the Ewing sarcoma preliminary diagnosis of Ewing sarcoma and tri-
translocation. Rhabdomyosarcomas are immuno- age the tissue for molecular testing using air-
reactive to desmin, myogenin, and/or myoD1. dried slides [10]. In addition, we validated an
Metastatic carcinomas are immunoreactive to antibody PRKCB which is a member of protein
cytokeratin. However, the differential diagnosis kinase C multigene family encoding serine/threo-
between a Ewing sarcoma with aberrant cytoker- nine kinases in our laboratory. This biomarker
atin and/or neuroendocrine marker expression has 98% sensitivity and 96% specificity in detect-
relies on the other immunostain markers listed ing EWSR1-FLI1 rearrangement, hence serves as
above or molecular confirmation. A nonclassic a rapid and economic surrogate diagnostic marker
Ewing sarcoma, the tumor cells are larger with for Ewing sarcoma. These quality improvement
a b
c d e
Fig. 4.2 Ewings Sarcoma. (a) Frontal radiograph of pel- low mass. The iliac crest is composed of white appearing
vis shows poorly defined lucent mass involving the left cartilage as the patient is a child. (d) The tumor is com-
ilium. (b) Coronal stir-weighted MR image shows a large posed of uniform, primitive appearing round cells that
tumor arising in the medullary cavity and transgressing the have fine chromatin and small nucleoli, and clear to eosin-
cortices and forming large extra and intra pelvic soft tissue ophilic cytoplasm. (e) Tumor cells show strong membra-
masses. (c) Coronal section through the ilium post chemo- nous staining for CD99 by immunohistochemistry
therapy shows that the bone is distorted by a necrotic yel-
projects initiated by pathologists greatly improve Table 4.5 Characteristics of histological variants of pri-
mary central osteosarcoma
the delivery of the care of Ewing sarcoma patients
in our institution. Tumor type Component Prognosis
Conventional High-grade High-grade
sarcoma with tumor. Subtype
osteoid formation does not differ
4.4 Osteosarcoma Osteoblastic in prognosis and
(76–80%) therapy
Osteosarcomas are a heterogeneous group, which Chondroblastic
(10–13%)
includes the primary central, secondary central, Fibroblastic (10%)
and surface of the bone, conventional, telangiec- Telangiectatic High-grade Similar to
tatic, and small cell variants. The distinguishing osteosarcoma with conventional
hallmark of osteosarcoma is the tumor cells pro- characteristic type
blood lakes and
ducing osteoid matrix. The most common pri-
spaces
mary central osteosarcoma is conventional type Giant High-grade Similar to
which includes osteoblastic, chondroblastic, and cell-rich osteosarcoma with conventional
fibroblastic variants. Osteoblastic osteosarcomas abundant type
have a predominantly osteoid matrix, which can osteoclast-like
giant cells
be thick or thin and branching. Chondroblastic
Small cell High-grade Slightly worse
osteosarcomas have a predominant chondroid osteosarcoma with prognosis than
matrix. Fibroblastic osteosarcomas produce only characteristic conventional
minimal amounts of osteoid and have high-grade small tumor cells type
spindled cell architecture. Telangiectatic osteo- Low-grade Low-grade Excellent
central osteosarcoma prognosis
sarcoma is characterized by having large blood-
filled spaces, which are usually separated by thin
septa. Although prognosis is thought to be similar
to conventional osteosarcomas, they are much
more sensitive to chemotherapy. Small cell osteo- overall and disease-free survival [11]. The sam-
sarcoma produces variable amounts of osteoid, pling of osteosarcoma includes cross-sectioning
and morphologically resembles Ewing sarcoma, the central and largest slice of the tumor. The
but lacks the t(11;22) translocation. The charac- tumor slice is further divided into 1 cm × 1 cm
teristics of primary central osteosarcoma are slices and prepared for histologic examination.
summarized in Table 4.5. Surface osteosarcoma As a part of the therapy changes, tumor necrosis
very rarely affects the pelvic bone. Grossly the is documented in the pathology report, which is
conventional osteosarcoma shows hard tab-white reversely related to the percentage of viable
cut surface, typically extends into the soft tissue. tumor cells as an independent prognostic of
Histology of osteosarcoma is illustrated in osteosarcoma.
Fig. 4.3a–d. Osteosarcomas are typically immunoreactive
The pathologists’ primary role is to make a to CD99 which is a sensitive but not a specific
definitive diagnosis of osteosarcoma and accu- marker. Osteocalcin is useful for highlighting
rately classify and grade the tumor on preopera- osteoid. Recurrent amplifications at 1q21-23 and
tive biopsy samples. High-grade osteosarcoma is 17p are commonly seen, and comparative
typically treated with neoadjuvant chemotherapy. genomic hybridization analysis has revealed fre-
The pathologists’ second role is to evaluate the quent chromosomal gains, such as the gain of
therapy response which is critically important for 8q23, seen in about half of osteosarcomas [12].
prognosis. Osteosarcomas with greater than 90% CDK4 with or without MDM2 is commonly
tumor necrosis (less than 10% viable tumor cells) amplified in aggressive osteosarcomas. Patients
are considered good responders and have better with hereditary retinoblastoma (RB) and Li
a b
c d
Fig. 4.3 Osteosarcoma. (a) Osteoblastic osteosarcoma tilage. (c) Giant cell-rich osteosarcoma showing multinu-
showing malignant tumor cells producing osteoid matrix. cleated giant cells. (d) Small cell osteosarcoma showing
(b) Chondroblastic osteosarcoma showing malignant car- blue round tumor cells producing osteoid matrix
Fraumeni syndrome have an increased risk of cells have eccentrically located nuclei with abun-
developing osteosarcomas. RB1 alterations have dant cytoplasm. With the increased tumor grade
also been seen in up to 40% of sporadic osteosar- from well-differentiated, to moderately differen-
comas, while TP53 alterations have been seen in tiated and to poorly differentiated myeloma, the
up to 35% of osteosarcomas. Many genetic aber- tumor exhibits increased nuclear size, nuclear
rations have been found in high frequency, some pleomorphism, prominent nucleoli, mitotic activ-
of which may offer prognostic value [13]. ity, and necrosis. The high-grade features resem-
Osteosarcoma of pelvis, in the setting of Paget ble diffuse large B-cell lymphoma. However,
disease and radiation associated are of particu- myeloma cells are immunoreactive to CD38,
larly unfavorable outcomes [14]. CD138 (syndecan-1), and MUM1 with monoclo-
nality of either kappa or lambda chain (kappa or
lambda chain restriction). Flow cytometry is
4.5 Multiple Myeloma/Plasma ideal for fresh tissue to confirm myeloma
Cell Myeloma diagnosis.
On the other hand, intraoperative evaluation of
Plasma cell myeloma commonly occurs in pelvis pathologic fracture of pelvis caused by undiag-
either as a primary tumor or as a part of multiple nosed myeloma can be challenging. In our insti-
myeloma. For the patients with prior history of tution, touch prep cytology has been routinely
plasma cell myeloma, the diagnosis of this tumor used intraoperatively in conjunction to frozen
in pelvic specimen is straight forward. The clas- section to facilitate a rapid diagnosis of myeloma
sic histology of myeloma includes round tumor [10]. The touch prep smears prepared from fresh
a b
Fig. 4.4 Myeloma. (a) Neoplastic plasma cells display eccentrically located nuclei and abundant cytoplasm. (b) Tumor
cells showing lambda chain-restriction by in situ hybridization
tissue are ideal to show the characteristic clock- primary site unknown or without a primary, his-
face like nuclei and perinuclear hof of neoplastic tomorphology in conjunction with pertinent
plasma cells which are diagnostic of this tumor. ancillary testing including immunohistochemis-
Fresh tissue can then be triaged for flow cytometry are used to render a definitive diagnosis.
try study for confirmation. This practice has Intraoperative evaluation of pathologic frac-
proven to be most accurate and efficient in man- ture of pelvis caused by undiagnosed metastatic
aging this type of patients. The histology of carcinoma can be challenging. “Epithelioid
myeloma is illustrated in Fig. 4.4a, b. malignancy” diagnosis is not adequate to guide
Genetically plasma cell myeloma has two dis- the optimal patient care in this type of clinical
tinct groups. One group (40%) harbors a bal- situation. In our institution, touch prep cytology
anced reciprocal translocation of the has been routinely used intraoperatively in con-
immunoglobin heavy-chain locus (IGH) with dif- junction to frozen section to facilitate a rapid
ferent partner genes including FGFR3/MMSET diagnosis [10]. The touch prep smears prepared
on 4p16.3, CCND3 on 6p21, CCND1 on 11q13, from fresh tissue avoid of crush artifact, which is
MAF on 16223, and MAFB on 20q11. Other commonly seen in frozen section slides. The epi-
group (60%) is hyperdiploidy with polysomes 3, thelial nuclei, glandular formation, intracytoplas-
5, 7, 9, 11, 15, 19, and 21. MYC on 8q24 is asso- tic mucin, and squamous cytoplasm provide
ciated with this group. The tumor prognosis is unequivocal evidence for the diagnosis of
associated with multiple genetic markers [15]. metastatic carcinoma intraoperatively. Accurate
and immediate diagnosis of metastatic carcinoma
will guide the surgeon to fix the pathologic frac-
4.6 Metastatic Carcinoma ture in the manner which is totally different from
the fixation of a sarcoma caused pathological
Metastatic disease from carcinoma is common in fracture.
pelvis. In the 4431 metastatic lesions registered
in the archive of the Rizzoli institute, 833 (18.8%)
were found to occur in the pelvic region includ- 4.7 Undifferentiated High-Grade
ing 559 (12.6%) are located in the ilium, 80 Pleomorphic Sarcoma
(1.8%) in the ischium, and 53 (1.2%) in the pubis
[16]. The primary sites of the carcinomas include Undifferentiated pleomorphic sarcoma, is a con-
lung, breast, prostate, kidney, head and neck, and temporary concept to include a group of high-
gastrointestinal tract. When a primary tumor is grade sarcomas, has no identifiable line of
present, the diagnosis of metastatic disease is differentiation when analyzed by current tech-
achieved by comparing the histology of pelvic nologies [13]. Its histology is variable and may
lesion with the primary disease. However, when a show different morphologic patterns composed
of spindle cells, pleomorphic cells, epithelioid 4.10 Langerhans Cell

cells, round cells, and multinucleated giant cells. Histiocytosis/Eosinophilic
Mitotic activity is typically prominent with atypi- Granuloma
cal mitotic figures. Tumor necrosis can be seen.
Due to the lack of consistent and identifiable Although this tumor most frequently involves the
diagnostic biomarkers, this group of tumor ribs, it also involves the pelvis as the primary site.
remains a diagnosis of exclusion. This is an This is a clonal neoplastic-like disease composed
aggressive malignancy with frequent metastases. of Langerhans cells, which are specialized histio-
Tumor necrosis in response to neoadjuvant che- cytes with nuclear grooves (reniform nuclei), and
motherapy provides important prognostic infor- admixed with inflammatory cells including
mation. Further studies are warranted to reveal prominent eosinophilia. The hallmark Langerhans
the prognostic and predictive biomarkers of this cells are immunoreactive to CD1a, CD207/
tumor. Langerin, and S-100, while negative for CD45.
The identification of Langerhans cells are the key
for diagnosis [3]. The prognosis for patient with
4.8 Fibrosarcoma monostotic or limited polystotic diseases is good.
Fibrosarcoma is a very specific diagnosis to

include the tumor composed of intermediate- to 4.11 Desmoplastic Fibroma
high-grade fibroblastic spindle cells, which have
littler genetic or molecular information [13]. This is very rare, benign, and locally aggressive
Grossly, the tumor is firm and tan-white. The spindle cell tumor of fibroblastic origin. The histo-
classic histology exhibits spindle cells devoid of logical features include bland spindle cells with
significant pleomorphism arranged in “herring- abundant collagenous stroma, resemble desmoid
bone” pattern. There is no bone, cartilage or other tumor of the soft tissue. One main differential diag-
line of differentiation other than fibroblastic. It is nosis is low-grade central osteosarcoma. The latter
challenging to make a definite diagnosis of this is typically positive for MDM2 amplification.
tumor on limited biopsy specimen. Because this
is a diagnosis of exclusion, the examination of
the resected specimen is warranted. When a 4.12 Aneurysmal Bone Cyst
tumor has marked cytological atypia and stori-
form growth pattern, it should be classified as Aneurysmal bone cyst is a benign tumor. Grossly
undifferentiated high-grade pleomorphic sar- it is well-defined and composed of blood-filled
coma. The prognosis depends on the patient’s cysts, which lack specific cell-lining and consist of
age, tumor grade, and stage. a wall of spindle cells with scattered osteoclast-
type multinucleated giant cells. The neoplastic
cells are spindle, which can be indistinguishable
4.9 Chondroma/Enchondroma from reactive fibroblasts and myofibroblasts.
However, the tumor cells show USP6 rearrange-
This is a benign hyaline cartilaginous tumor. ment in 70% of the primary aneurysmal bone cyst,
Histologically it is hypocellular, avascular, with not the secondary ones [17]. The spindle cells are
prominent hyaline cartilage matrix, and arranged bland and lack of mitotic activity. Reactive woven
in a multinodular architectural pattern. The chon- bone may be seen with osteoblasts rimming. The
drocytes are bland with no mitotic activity or main differential diagnosis is telangiectatic osteo-
necrosis. Heterozygous somatic mutations of sarcoma, which is characterized by having large
IDH1 and IDH2 have been frequently identified blood-filled spaces, but malignant tumor cells with
in enchondromas as well as chondrosarcomas osteoid formation. Solid variant of aneurysmal
showing the genetic linkage of these two entities. bone cyst may be diagnostically challenging.
However, USP6 rearrangement testing can be used Osteoprotegerin (OPG) is a soluble decoy
to confirm the diagnosis. receptor that is produced by osteoblasts to inhibit
osteoclast differentiation through its binding to
RANKL, which prevents RANK binding. OPG
4.13 Giant Cell Tumor expression reflects a protective mechanism of the
skeleton to compensate increased bone resorp-
Giant cell tumor of bone is a benign but locally tion. Bone remodeling is mainly controlled by
aggressive and recurrent tumor. The tumor is the balance of RANKL/OPG. Osteoprotegerin
composed of numerous characteristic giant cells ligand (OPGL), also named receptor activator of
which are large and osteoclast-like. These cells RANKL is also expressed in the stroma-like
are the background cells reactive to the true neo- tumor cells of GCTB. The ratio of OPGL/OPG
plastic cells. The neoplastic cells are mononu- by tumor cells may contribute to the degree of
clear primitive mesenchymal stromal cells osteoclastogenesis and bone resorption [18].
expressing receptor activator for NF-κB ligand Grossly the tumor is red-brown with hemor-
(RANKL), the master regulator of osteoclast dif- rhage and yellow areas reflect lipid laden
ferentiation. Macrophages and osteoclasts macrophage-rich areas. Histologically, the tumor
express RANK. The interaction between the neo- is composed of numerous multinucleated giant
plastic mononuclear stromal cells and marcoph- cells and scattered mononuclear cells that are
ages/osteoclasts by a RANKL-dependent round or spindle. Because H3-3A (H3F3A) gene
mechanism via the stimulation of macrophage- mutation is common in giant cell tumor of bone
colony stimulation factor (MCSF) results in neo- (95%) and immunohistochemical study of H3.3
plastic proliferation and induces osteoclast G34W is a reliable surrogate marker for this
formation. During this process, tumor-associated mutation. Immunostain of H3.3 G34W is useful
macrophage-like osteoclast precursors, which are in confirming the diagnosis of giant cell tumor of
also mononuclear cells, are recruited by tumoral bone when other morphological differential diag-
stromal cells to participate in osteoclast differen- noses are considered [19]. Histological illustra-
tiation and activation. Because osteoclast formation of giant cell tumor of bone is in Fig. 4.5a, b.
tion is the major consequence of giant cell tumor, Lipid laden or hemosiderin laden macrophages
inhibition of osteoclast formation and activity is are also present. The tumor is mainly solid and
the key for therapeutic approach. For example, may contain cystic areas. Secondary aneurysmal
bisphosphonate inhibits osteoclast-mediated bone cyst component is seen in 10% of giant cell
resorption of bone/osetolysis and anti-RANKL tumor. The tumor may be mitotically active;
antibody targets the RANKL-dependent mecha- however, a benign giant cell tumor typically does
nism of giant cell formation. not have atypical mitosis or significant nuclear
a b
Fig. 4.5 Giant cell tumor of bone. (a) Tumor cells are H3.3 G34W immunohistochemistry which is a reliable
mononuclear. The multinucleated giant cells are nonneo- surrogate marker for underling molecular pathology
plastic. (b) The monocular tumor cells are highlighted by
atypia. The latter is associated with a malignant immediate management of pathologist fracture
transformation. caused by these tumors.
4.14 Fibrous Dysplasia References
This is a benign fibro-osseous lesion that is 1. Bloem JL, Reidsma II. Bone and soft tissue tumors of
hip and pelvis. Eur J Radiol. 2012;81:3793–801.
caused by postzygotic activating missense muta- 2. Jeffrey S, et al. Chapter 38. Bone. In: Amin MB, Edge
tions in the GNAS gene on 20q13 [20]. This S, Greene F, Byrd DR, Brookland RK, Washington
mutation and fibrous dysplasia are also associ- MK, et al., editors. AJCC cancer staging manual.
ated with McCune-Albright syndrome. 8th ed. New York: Sprinter International Publishing;
2017.
Histologically, the tumor is composed of bland 3. Dodd L, Bui MM. Chapter 16.6. Chondrosarcoma.
spindle fibroblastic cells admixed with irregular In: Dodd L, Bui MM, editors. Atlas of soft tissue
bony spicules without osteoblastic rimming. The and bone pathology: with histologic, cytologic and
irregular bony spicules have often been described radiologic correlations. New York: Demos Medical
Publishing; 2015. ISBN 9781620700372.
as “alphabet soup” or “Chinese characters” in 4. Rozeman LB, Hogendoorn PC, Bovee JV. Diagnosis
configuration [3]. The prognosis of monostotic and prognosis of chondrosarcoma of bone. Expert
fibrous dysplasia is excellent. Malignant trans- Rev Mol Diagn. 2002;2(5):461–72.
formation is exceptionally rare. 5. Henderson-Jackson EB, Bui MM. Molecular pathol-
ogy of soft-tissue neoplasms and its role in clinical
practice. Cancer Control. 2015;22(2):186–92.
6. Taylor BS, et al. Advances in sarcoma genom-
4.15 Osteoid Osteoma ics and new therapeutic targets. Nat Rev Cancer.
2011;11(8):541–57.
7. Kao YC, Owosho AA, Sung YS, Zhang L, Fujisawa
This is benign bone-forming tumor. The nidus of Y, Lee JC, Wexler L, Argani P, Swanson D, Dickson
the tumor consists of a combination of osteoid BC, Fletcher CDM, Antonescul CR. BCOR-
and woven bone surrounded by osteoblasts. The CCNB3- fusion positive sarcomas: a clinicopatho-
nidus is vascular-rich given the appearance of logic and molecular analysis of 36 cases with
comparison to morphologic spectrum and clinical
granulation tissue surrounded by sclerotic bone. behavior of other round cell sarcomas. Am J Surg
The diagnosis of osteoid osteoma is typically Pathol. 2018;42(5):604–15. https://doi.org/10.1097/
straightforward with clinical and radiological PAS.0000000000000965.
correlation. 8. Yoshimoto T, Tanaka M, Homme M, Yamazaki Y,
Takazawa Y, Antonescu CR, Nakamura T. CIC-DUX4
induces small round cell sarcomas distinct from
Ewing sarcoma. Cancer Res. 2017;77(11):2927–37.
4.16 Conclusion https://doi.org/10.1158/0008-5472.CAN-16-3351.
9. Wei S, Henderson-Jackson E, Qian X, Bui MM. Soft
tissue tumor immunohistochemistry update:
Pelvic bone tumors are a diverse group of benign, illustrative examples of diagnostic pearls to avoid pit-
intermediate, and malignant neoplasm. Their his- falls. Arch Pathol Lab Med. 2017;141(8):1072–91.
tological type is designated based on their closest 10. Bui MM, Smith P, Agresta S, Cheong D, Letson

histological resemble of the normal mesenchy- D. Practical issues of intraoperative frozen section
diagnosis of bone and soft tissue lesions. Cancer
mal tissue. Ancillary testing such as immunohis- Control. 2008;15(1):7–12.
tochemistry, cytogenetics, and molecular 11. Anninga JK, Gelderblom H, Fiocco M, Kroep JR,
techniques have greatly improved the pathologic Taminiau AH, Hogendoorn PC, et al. Chemotherapeutic
diagnosis of these tumors. Biomarkers that pro- adjuvant treatment for osteosarcoma: where do we
stand? Eur J Cancer. 2011;47(16):2431–45.
vide prognostic and predictive information are 12. Egas-Bejar D, et al. Theranostic profiling for action-
limited. Intraoperative touch prep cytology can able aberrations in advanced high risk osteosarcoma
be used to distinguish metastatic carcinoma and with aggressive biology reveals high molecular diver-
plasmacytoma in conjunction with frozen sec- sity: the human fingerprint hypothesis. Onco Targets
Ther. 2014;1(2):167–79.
tion, which provides valuable information for
13. Gibbs J, Henderson-Jackson E, Bui MM. Bone and aneurysmal bone cysts and are absent in so-called
soft tissue pathology: diagnostic and prognostic secondary aneurysmal bone cysts. Am J Pathol.
implications. Surg Clin N Am. 2016;96(5):915–62. 2004;165(5):1773–80.
14. Rosenberg AE, et al. Conventional osteosarcoma. In: 18. Huang L, Xu J, Wood DJ, Zheng MH. Gene expres-
Fletcher CDM, et al., editors. WHO classification of sion of osteoprotegerin ligand, osteoprotegerin,
tumours of soft tissue and bone. 4th ed. Lyon: IARC and receptor activator of NF-kappaB in giant cell
Press; 2013. 288p. tumor of bone: possible involvement in tumor cell-
15. Lorsbach R, Kluin PM. Plasma cell myeloma. In: induced osteoclast-like cell formation. Am J Pathol.
Fletcher CDM, et al., editors. WHO classification of 2000;156(3):761–7.
tumours of soft tissue and bone. 4th ed. Lyon: IARC 19. Yamamoto H, Iwasaki T, Yamada Y, et al. Diagnostic
Press; 2013. 314p. utility of histone H3.3 G34W, G34R, and G34V
16. Picci P, Manfrini M, Fabbri N, Gambarotti M, Vanel mutant-specific antibodies for giant cell tumors of
D. Atlas of musculoskeletal tumors and tumor-like bone. Hum Pathol. 2018;73:41–50.
lesions. Berlin: Springer; 2014. 20. Bianco P, Riminucci M, Majolagbe A, et al. Mutations
17.
Oliveira AM, Perez-Atayde AR, Inwards CY, of the GNAS1 gene, stromal cell dysfunction, and
Medeiros F, Derr V, Hsi BL, et al. USP6 and CDH11 osteomalacic change in non-McCune-Albright fibrous
oncogenes identify the neoplastic cell in primary dysplasia of bone. J Bone Miner Res. 2000;15:120128.
Benign Pelvic Bone Tumors
5
John E. Mullinax and G. Douglas Letson
5.1 Introduction opposed to the malignant tumors which were

indicated by Roman numerals (I, II, and III). The
The primary diagnostic dichotomy for patients utility of this classification system is largely
with bone tumors is separating a benign tumor based on the easily ascertained variables for each
from one that is malignant. Determination of a lesion that quickly guide the intervention
benign or indolent process from one that is malig- required.
nant with potential for local or distant spread is The radiographic interpretation of bone lesions
crucial to the care of the patient. For those lesions is fundamental and attempts to describe the rate of
deemed to be benign, the treatment decision then growth for each lesion. Benign or indolent lesions
becomes a balance of intervention with observa- are considered those with either no growth or very
tion. This balance is particularly important for slow growth pattern within the bone. The rate of
the benign bone lesions that arise in the pelvis growth in balance with the rate of osteoblastic
due to the anatomical constraints which often activity yields hallmark findings on radiographs.
lead to increased morbidity with intervention The malignant potential of bone lesions was first
relative to other sites of disease. described by Lodwick in 1980 using radiographic
Classification of bone tumors at all sites is findings, followed by Madewell in 1981 using
based on the Enneking classification. First pub- pathologic correlates [2, 3]. Later, Carraciolo
lished in 1980, this construct serves to delineate et al. combined the findings of both classification
benign and malignant bone lesions based on clin- systems with a separate dataset [4]. Termed the
ical, radiographic, and pathologic characteristics Modified Lodwick-Madewell classification sys-
[1]. Using a combination of the radiographic tem, the authors describe six separate categories
appearance, anatomic extent of the tumor, and of radiographic findings that correlate with, to an
the histologic grade, this system was the first to escalating degree, the malignant potential.
classify bone tumors based on their malignant Apart from the imaging findings associated
potential, or lack thereof. In this system, those with the bone lesion of interest, the histologic
tumors considered to be benign or indolent were diagnosis is paramount to understanding the
indicated by Arabic numerals (1, 2, and 3) as behavior of each tumor. The spectrum of specific
diagnoses is wide but well described by the World
J. E. Mullinax · G. D. Letson (*) Health Organization Classification system. Using
Sarcoma Department, H. Lee Moffitt Cancer Center, this pathologic categorical description, bone
Tampa, FL, USA tumors are placed in groups based on the
e-mail: [email protected]; presumed tissue of origin followed by the degree
[email protected]

46 J. E. Mullinax and G. D. Letson
of malignancy. Diagnoses are considered benign, viscera and lower extremities courses through the
low grade, or high grade depending on their rate pelvis and therefore a comprehensive examination
of growth and probability of metastasis. should evaluate any lower extremity symptoms
The purpose of this chapter is to focus on the (i.e., claudication or rest pain) that might be sec-
evaluation and management of benign bone ondary to arterial compromise. Additionally, in the
tumors that arise with the pelvis. Using the clas- lower extremities, unilateral edema is an important
sification systems above, the scope is defined as finding as it may suggest pelvic venous compro-
tumors that are considered benign by the clinical mise with or without an element of thrombosis.
Enneking classification, the radiographic The diagnostic imaging modalities chosen for
Modified Lodwick-Madewell classification, and evaluation should be driven by the clinical exam
the pathologic WHO classification. Attention will but there are some general caveats. In those
be focused on the degree of intervention neces- patients suspected to have a bone tumor within
sary for these tumors with the treatment philoso- the pelvis, a plain radiograph combined with a
phy as a balance between morbidity and need for contrast-enhanced MRI is preferred. The findings
elimination of the lesion. on plain radiograph will allow for discrimination
of malignant potential and the contrast-enhanced
MRI will allow for assessment of the peritumoral
5.2 Initial Evaluation soft tissue. Utilization of axial imaging without
of the Patient plain radiograph will not allow for comprehen-
sive evaluation of a pelvic bone lesion. In those
The primary presentation of the patient with a cases with visceral compromise suspected by
pelvic bone tumor is generally considered either clinical exam, contrast-enhanced CT scan is also
symptomatic or asymptomatic. For those with recommended to assess for external compres-
symptoms, a diagnostic strategy is focused by the sion. A detailed gastroenterological or urologic
presentation whether it is weakness, pain, exam should also be considered for those with
restricted range of motion, or visceral compres- bowel and bladder dysfunction, respectively, as
sion (i.e., bladder or bowel dysfunction). In these symptoms may be coincidental rather than caused
cases, the bone tumor may be a coincidental find- by the pelvic bone tumor.
ing or one that is directly contributing to the con- In addition to radiographs and axial imaging,
stellation of symptoms. Patients with functional imaging is important for many bone
asymptomatic bone tumors of the pelvis are often lesions. Positron-emission tomography (PET)
discovered during an imaging modality per- combined with CT imaging is helpful to describe
formed for an unrelated reason. The two scenar- the metabolic activity of a bone lesion. PET/CT
ios are important to distinguish based on the imaging can be most useful in the patient with a
threshold degree for intervention as the asymp- history of malignancy and new pelvic bone
tomatic patient has a much higher likelihood of lesion. Low metabolic activity would suggest a
morbidity relative to their preoperative state. separate, sporadic process while increased meta-
The clinical exam of the patient with a pelvic bolic activity would suggest possible metastatic
bone tumor is of fundamental importance to delin- disease from the historical primary malignancy.
eate an effective and cost-effective diagnostic Another functional imaging modality is bone
strategy. Primary focus should be on focal weak- scintigraphy, or “bone scan.” Using radiolabeled
ness of muscle groups, range of motion, and neu- methylene diphosphonates, the degree of bone
rovascular deficits. The pelvic anatomy is such turnover and perfusion is assessed which can aid
that assessment of visceral function is also an in the interpretation of the complementary imag-
important factor for clinical exam. Impairment of ing modalities [5].
bowel, bladder, or sexual function may suggest At the conclusion of the diagnostic evaluation,
compression of the viscera directly or neurologic including the clinical exam and imaging modali-
compression. The vascular supply of the pelvic ties, patients found to have a benign bone lesion
5 Benign Pelvic Bone Tumors 47
should be categorized as either symptomatic or the tissue planes. If percutaneous biopsy must be
asymptomatic. Intervention should be considered undertaken, the procedure should be done in con-
more strongly for the former and the focus of cert with the treating surgeon so that the surgical
intervention should be on improvement of symp- approach is considered relative to the planned
toms. Compressive symptoms are more easily biopsy tract.
addressed with surgical intervention while pain
symptoms are less likely to have a durable
response. In the asymptomatic patient, surgical 5.3.2 Enchondroma
intervention should be contemplated relative to
the Ennenking classification with an approach A common benign cartilaginous tumor arising
that escalates with higher grade. Observation of within the medullary canal of long bones is an
grade 1 lesions is an optimal approach while enchondroma. These tumors can be radiographi-
curettage is appropriate for Grade 2 lesions. The cally well-demarcated in the small bones of the
addition of a surgical adjuvant to curettage such hand but can also present with ill-defined borders
as methyl methacrylate or liquid nitrogen (cryo- in long bones. These are most often discovered
therapy) should be considered for Grade 3 lesions. incidentally on plain radiographs. Follow-up
Within these two groups of patients (Grade 2 imaging with MRI characteristically demon-
and 3), the specific histologic diagnosis of the strates a hyperintense mass on T2-weighted
benign bone tumor is the most important factor imaging sequences.
that impacts the decision to operate. The remain- As most are discovered incidentally, most do
der of this chapter focuses on the surgical not need any therapy. Indications for intervention
approach relative to the different subtype include symptoms that can be attributed directly
diagnoses. to the lesion or any features of malignancy.
Radiographic features of malignancy would
include permeative appearance in the surround-
5.3 Chondrogenic Tumors ing bone, periosteal reaction, or adjacent soft tis-
sue mass. Biopsy is not generally required given
5.3.1 General Considerations the characteristic imaging findings but if tissue is
for Cartilage Tumors obtained, pathologic findings concerning for
of the Pelvis malignancy would include nuclear atypia or
increased myxoid component among a hypercel-
Benign cartilaginous tumors of the pelvis are lular hyaline cartilage stroma.
rare. Clinical suspicion of chondrosarcoma
should be high with any bone lesion of the pelvis
that suggests cartilaginous origin on imaging. At 5.3.3 Chondroblastoma
a minimum, a cartilaginous lesion of the pelvis
with benign radiographic appearance warrants Typically involving the epiphysis, this benign
interval imaging follow-up. Any change in the cartilaginous tumor is a rare entity arising from
characteristics of the lesion should warrant resec- within the long bones. They typically present
tion rather than curettage to avoid an intralesional with benign radiographic appearance (sclerotic
procedure on a malignant cartilaginous lesion borders) but may also have a periosteal reaction.
which can have disastrous oncologic effects. In a The gross appearance is consistent with the infil-
similar manner, percutaneous biopsy of bone trate of mononuclear cells characteristic to this
lesion in the pelvis with features consistent of tumor. The degree of atypia is variable on patho-
cartilage origin should generally be avoided. The logic analysis of these lesions and interpretation
gross consistency of chondrosarcoma is often of the histologic findings relative to the
semisolid and thus perforation of the tumor with radiographic findings is helpful in confirming the
biopsy needle can lead to local contamination of diagnosis.
Chondroblastoma is often a symptomatic lesion atic, intervention is preferred for these lesions.
and thus intervention is preferred. One approach to Resection is indicated for these lesions as there is
treatment is an intraoperative biopsy to confirm the significant overlap with chondrosarcoma based
diagnosis followed by curettage and bone graft. on the imaging characteristics and even patho-
Limiting the dissection of tissue planes on approach logic assessment of small tissue quantity obtained
to the lesion should be a focus of operative plan- with biopsy (Fig. 5.2). Understanding this risk of
ning and thus, in the event the lesion is found to be an underlying malignancy, resection of an intact
malignant, proper resection can be undertaken tumor is preferred to curettage.
without a wide field of contamination. This can be
accomplished using a minimally invasive, percuta-
neous technique under fluoroscopic guidance 5.3.5 Osteochondroma
(Fig. 5.1). In addition to curettage, others have
described intraoperative adjuvant therapy such as Arising as outgrowth from the bone, osteochon-
cryosurgery with good functional outcomes and dromas are the most common benign bone lesion.
low recurrence rate [6]. These lesions are considered a benign cartilagi-
nous tumor with continuity to the cortex and
medullary canal. These lesions characteristically
5.3.4 Chondromyxoid Fibroma arise pointing away from a joint and have a
pathognomonic cartilaginous cap. Also termed
Unlike the prior chondrogenic tumors discussed, exostoses, they may arise as a single lesion or in
this benign cartilaginous neoplasm most com- multiple lesions. Approximately 15% of patients
monly presents with pain and can be found in the with multiple lesions have an underlying germ-
iliac wing. The radiographic appearance demon- line mutation in the tumor suppressor genes EXT
strates a scalloped lesion, often with sclerotic 1 or 2 which results in the syndrome known as
noncontiguous borders. As most are symptom- hereditary multiple exostosis [7].
a b c d
e f g h
Fig. 5.1 Percutaneous curettage of pelvic chondroblas- planes. Biopsy is undertaken in a coaxial fashion under
toma. CT demonstrates a pelvic lesion with benign fluoroscopic guidance to prevent contamination of sur-
appearance in the axial (a) and coronal (b) plane. rounding soft tissue (e, f). Once diagnosis of benign lesion
Intraoperative placement of a canula under fluoroscopic is confirmed, curettage is performed through the percuta-
guidance (c, d) allows for dissection through the soft tis- neous access canula (g, h). (Red arrow = bone lesion)
sue with minimal displacement of normal anatomic
a b
c d
Fig. 5.2 Surgical management of pelvic chondromyxoid resection with placement of bone cement, the postopera-
fibroma. Preoperative images demonstrate a lesion within tive MRI (c) and radiograph (d) demonstrate resolution of
the left iliac wing on MRI (a) and CT scan (b). Following the lesion
Intervention is reserved for symptomatic tral, lucent nidus surrounded by a sclerotic

lesions and those that harbor features concerning peripheral zone is readily identified on plain
for malignancy on imaging. The cartilaginous radiographs and confirmed with axial imaging in
cap, a hallmark feature of this lesion, should be the form of a CT scan. These lesions are typically
<2 cm. Features that would raise concern for a found in males between 10 and 20 years old and
malignancy rather than benign lesion include, are, by definition, <2 cm in size. The clinical sce-
rapid growth of the lesion, a cartilage cap >2 cm, nario most often involves pain at night which is
or loss of corticomedullary continuity. These often relieved by aspirin. Intervention is not gen-
findings warrant biopsy before planned excision. erally necessary for these patients but, when
Once the diagnosis is confirmed as an osteochon- required, can be in the form of curettage or sim-
droma rather than a malignant bone lesion, the ple excision. More recently, a percutaneous tech-
removal of these lesions does not require an nique of radiofrequency ablation (RFA) has been
extensive resection but rather excision of the advocated [8, 9]. Technical constraints of this
pedunculated lesion from the bone involved. approach limit RFA to lesions <2 cm and thus
Care should be taken to ensure the lesion is osteoblastomas (below) are not approached in
removed entirely, all the way to the base. this manner. When the latter is performed,
pathologic confirmation of the diagnosis involves
histologic identification of the nidus.
5.4 Osteogenic Tumors
5.4.1 Osteoid Osteoma 5.4.2 Osteoblastoma
Arising within the diaphysis or metaphysis of the Compared to an osteoid osteoma, an osteoblas-
long bones, osteoid osteomas demonstrate a char- toma has the same underlying histologic findings
acteristic finding on plain radiographs. The cen- with the primary clinical difference being the size
of the lesion. While an osteoid osteoma must be size of the tumor and therefore the extent of
<2 cm in size, an osteoblastoma will be larger in resection. Denosumab is a monoclonal antibody
maximal diameter. Additional differences include which binds RANKL, thereby interrupting the
the predilection of osteoblastomas to occur positive stimulation signal between the neoplas-
within the spine and the lack of a clear nocturnal tic stromal and osteoclasts within the tumor [14,
pain cycle as is classic for osteoid osteomas. The 15]. Preoperative treatment with denosumab is of
indications for treatment are not different, though, primary importance for lesions with a large soft
due to the size of the lesion, morbidity can be tissue component as bone formation at the mar-
higher. Planning the surgical resection with gins can assist with resection and limit intraop-
respect to the preoperative symptoms is impor- erative morbidity. Conversely, lesions with a
tant caveat to treating these lesions. planned intralesional (curettage) approach should
not receive preoperative denosumab as this treat-
ment is associated with bone formation at the
5.5 Giant Cell Tumor reactive zone, thereby precluding complete curet-
tage. Finally, radiation has been described for the
With the alternative osteogenic tumors arising treatment of large, locally advanced lesions but
from an over proliferation of osteoblasts, the concern with subsequent development of
giant cell tumor (GCT) of bone is one character- radiation-associated sarcoma limits the broad use
ized by an overproliferation of osteoclasts. The of this approach.
underlying biology is centered on the disruption
of normal bone remodeling physiology where the
osteoclasts proliferate in response to Receptor 5.6 Cystic Disease of the Bone
activator of nuclear factor kappa-Β ligand
(RANKL) which is physiologically secreted by 5.6.1 Unicameral Bone Cyst (UBC)
osteoblasts. In GCT, the neoplastic stromal cells
secrete RANKL without negative feedback, Simple cystic lesion within the bone most com-
thereby driving the tumor growth. monly arises in the adolescent and young adult
The majority of GCT have an indolent, even population with 80% arising within the second
benign clinical course but up to 5% may develop decade of life. Termed unicameral bone cysts
distant metastatic disease, primarily in the lung (UBC), these lesions most often present in the
[10]. For those with localized disease, there have long bones with predilection for the proximal
been several interventions described. Extended humerus and femur. Clinical presentation can be
curettage is the preferred approach and can pro- either incidental or within the context of long
duce excellent long-term results (Fig. 5.3). Several bone fracture. In those with the latter, spontane-
adjuvant treatments are described to reduce local ous resolution can result in 15% of cases owing
recurrence rates such as methyl methacrylate to the bone remodeling associated with fracture
[11]. Radiation therapy has also been described in repair. In those with incidental findings, the deci-
the adjuvant setting but long-term follow-up has sion to intervene is primarily one of impending
not demonstrated local control benefit [12]. fracture risk.
Others advocate the use of liquid nitrogen as a Distinguishing these lesions from more com-
surgical adjuvant [13]. Complete resection is plicated cystic disease such as aneurysmal bone
associated with a recurrence rate <20% but this cyst (ABC) is important prior to treatment.
approach, especially in the pelvis, must be con- Following identification on plain radiograph, CT
sidered relative to the proposed deficit associated scan is helpful to determine the thickness of the
with the resection. Owing to the unique patho- cyst wall and consequent fracture risk.
physiology of the RANKL-axis in the develop- Intervention on UBC should address the impend-
ment of these tumors, neoadjuvant treatment with ing fracture risk without imparting undue mor-
denosumab has been advocated to decrease the bidity. For this reason, excision or en bloc
a b
Fig. 5.3 Giant cell tumor long-term follow-up. Treatment stability and remodeling evident on radiograph 24 years
of a pelvic giant cell tumor (red arrow) with curettage and after the index procedure (b)
adjuvant radiation therapy (a). Long-term follow-up with
resection is not indicated but rather intervention pelvis, this soft tissue component can generate
should address the cystic cavity to allow physio- significant symptoms related to compression of
logic bone remodeling. Following aspiration to neurovascular structures, bladder, or bowel.
confirm the diagnosis, injection of methylpred- The diagnosis of ABC must be confirmed,
nisolone acetate has been suggested as a low risk, apart from a malignant lesion, prior to interven-
even percutaneous, intervention. This procedure tion. Approximately 30% of ABC are described
may also be repeated for those lesions that do not as “secondary” and related to prior trauma or
respond to a single injection [16]. More recently, associated adjacent malignancy. Identification of
this approach has been supplanted by curettage a possible underlying malignant process associ-
with bone graft for pelvic UBC. Given that the ated with the ABC is important to avoid an intra-
UBC will resolve once growth plates close, treat- lesional procedure with resulting tumor
ment should only be considered in the young dissemination. For the primary ABC, the primary
(<10 years) patient while observation is preferred pathologic diagnostic dilemma is distinguishing
in the adolescent patient. these lesions from telangiectatic osteosarcoma
(TO) with the latter requiring multimodal therapy
and surgical intervention with oncologic intent.
5.6.2 Aneurysmal Bone Cyst (ABC) Histologic differences associated with TO
include atypical cells with increased mitoses.
A more complex cystic lesion within the bone is Molecular pathology is helpful to discriminate
the aneurysmal bone cyst (ABC). These lesions, the two diagnoses as well with ABC containing a
rather than simple cystic lesions, are comprised characteristic gene rearrangement in the USP6
of multiloculated regions of abnormal angiogen- gene [17].
esis within a larger area of bone destruction. Like Unlike the focus on intracystic treatment as
UBC, these lesions most commonly arise within with UBC, the only treatment with guaranteed
the long bones but a larger percent (9%) arise complete resolution is wide resection. Often, the
within the pelvis. In addition, while UBCs are size of these lesions precludes complete resection
more commonly restricted to the bone, ABC can due to the associated postoperative deficits in the
have a large soft tissue component emanating setting of a benign process. Intralesional therapy,
from the primary lesion within the bone. In the if undertaken should not be with methylpredniso-
lone as in the case of UBC but rather with ethanol 3. Lodwick GS, Wilson AJ, Farrell C, Virtama P, Dittrich
F. Determining growth rates of focal lesions of bone
or doxycycline [18, 19]. If surgical intervention from radiographs. Radiology. 1980;134(3):577–83.
is required or contemplated, a strong consider- https://doi.org/10.1148/radiology.134.3.6928321.
ation should be given to preoperative emboliza- 4. Caracciolo JT, Temple HT, Letson GD, Kransdorf
tion, especially for pelvic lesions to mitigate the MJ. A modified Lodwick-Madewell grading sys-
tem for the evaluation of lytic bone lesions. AJR
intraoperative blood loss which has frequently Am J Roentgenol. 2016;207(1):150–6. https://doi.
encountered with these lesions. org/10.2214/AJR.15.14368.
5. Brenner AI, Koshy J, Morey J, Lin C, DiPoce J. The
bone scan. Semin Nucl Med. 2012;42(1):11–26.
https://doi.org/10.1053/j.semnuclmed.2011.07.005.
5.7 Hemangioma 6. van der Geest IC, van Noort MP, Schreuder HW,
Pruszczynski M, de Rooy JW, Veth RP. The cryosur-
Benign vascular tumors arising within the bone gical treatment of chondroblastoma of bone: long-
are a common finding in the adult population term oncologic and functional results. J Surg Oncol.
2007;96(3):230–4. https://doi.org/10.1002/jso.20804.
with approximately 10% of patients found to 7. Jennes I, Pedrini E, Zuntini M, Mordenti M, Balkassmi
have a vertebral body hemangioma on imaging S, Asteggiano CG, et al. Multiple osteochondromas:
performed for another reason [20]. Imaging is mutation update and description of the multiple
characteristic with plain radiograph demonstrat- osteochondromas mutation database (MOdb). Hum
Mutat. 2009;30(12):1620–7. https://doi.org/10.1002/
ing pathognomonic mineralization in the form of humu.21123.
phleboliths within the larger lesion. On cross- 8. Faddoul J, Faddoul Y, Kobaiter-Maarrawi S, Moussa
sectional imaging, this mineralization is again R, Rizk T, Nohra G, et al. Radiofrequency ablation
evident in a “polka dot” pattern. Often, the imag- of spinal osteoid osteoma: a prospective study. J
Neurosurg Spine. 2017;26(3):313–8. https://doi.org/1
ing findings are diagnostic, and no biopsy is 0.3171/2016.8.SPINE16462.
needed for these lesions which have universally 9. Beyer T, van Rijswijk CSP, Villagran JM, Rehnitz
benign radiographic appearance. C, Muto M, von Falck C, et al. European multicen-
Bone hemangiomata are most often incidental tre study on technical success and long-term clinical
outcome of radiofrequency ablation for the treat-
findings without symptoms and treatment is not ment of spinal osteoid osteomas and osteoblastomas.
required for these lesions. Serial imaging over a Neuroradiology. 2019;61(8):935–42. https://doi.
period of 2 years is warranted to ensure stability. org/10.1007/s00234-019-02226-9.
Large vertebral lesions may require intervention 10. Campanacci M, Baldini N, Boriani S, Sudanese
A. Giant-cell tumor of bone. J Bone Joint Surg Am.
due to compression at spinal nerve roots or when 1987;69(1):106–14.
impending vertebral body fracture is of concern. In 11. Arbeitsgemeinschaft K, Becker WT, Dohle J, Bernd
these cases, curettage with stabilization is preferred L, Braun A, Cserhati M, et al. Local recurrence of
to resection. Symptomatic lesions in the long bones giant cell tumor of bone after intralesional treatment
with and without adjuvant therapy. J Bone Joint Surg
can be treated with curettage or embolization, the Am. 2008;90(5):1060–7. https://doi.org/10.2106/
latter most easily repeated for nonresolving lesions. JBJS.D.02771.
12. Ruggieri P, Mavrogenis AF, Ussia G, Angelini A,

Papagelopoulos PJ, Mercuri M. Recurrence after
and complications associated with adjuvant treat-
References ments for sacral giant cell tumor. Clin Orthop Relat
Res. 2010;468(11):2954–61. https://doi.org/10.1007/
1. Enneking WF, Spanier SS, Goodman MA. The s11999-010-1448-8.
classic: a system for the surgical staging of muscu- 13. Wu PK, Chen CF, Wang JY, Chen PC, Chang MC,
loskeletal sarcoma (Reprinted from Clin Orthop, Hung SC, et al. Freezing nitrogen ethanol com-
vol 153, pg 106–120 1980). Clin Orthop Relat Res. posite may be a viable approach for cryotherapy of
1980;(153):106–20. https://doi.org/10.1097/01. human giant cell tumor of bone. Clin Orthop Relat
blo.0000093891.12372.0f. Res. 2017;475(6):1650–63. https://doi.org/10.1007/
2. Madewell JE, Ragsdale BD, Sweet DE. Radiologic s11999-017-5239-3.
and pathologic analysis of solitary bone lesions. 14. Thomas D, Henshaw R, Skubitz K, Chawla S,
Part I: Internal margins. Radiol Clin N Am. Staddon A, Blay J-Y, et al. Denosumab in patients
1981;19(4):715–48. with giant-cell tumour of bone: an open-label, phase 2
study. Lancet Oncol. 2010;11(3):275–80. https://doi. 18. Shiels WE, Mayerson JL. Percutaneous doxycycline
org/10.1016/s1470-2045(10)70010-3. treatment of aneurysmal bone cysts with low recur-
15. Rutkowski P, Gaston L, Borkowska A, Stacchiotti S, rence rate: a preliminary report. Clin Orthop Relat
Gelderblom H, Baldi GG, et al. Denosumab treatment Res. 2013;471(8):2675–83. https://doi.org/10.1007/
of inoperable or locally advanced giant cell tumor s11999-013-3043-2.
of bone - multicenter analysis outside clinical trial. 19. Lambot-Juhan K, Pannier S, Grevent D, Pejin Z,

Eur J Surg Oncol. 2018;44(9):1384–90. https://doi. Breton S, Berteloot L, et al. Primary aneurysmal bone
org/10.1016/j.ejso.2018.03.020. cysts in children: percutaneous sclerotherapy with
16. Zhao JG, Wang J, Huang WJ, Zhang P, Ding N,
absolute alcohol and proposal of a vascular classifica-
Shang J. Interventions for treating simple bone tion. Pediatr Radiol. 2012;42(5):599–605. https://doi.
cysts in the long bones of children. Cochrane org/10.1007/s00247-011-2312-z.
Database Syst Rev. 2017;2:CD010847. https://doi. 20. Errani C, Vanel D, Gambarotti M, Alberghini M,

org/10.1002/14651858.CD010847.pub3. Picci P, Faldini C. Vascular bone tumors: a pro-
17. Mascard E, Gomez-Brouchet A, Lambot K. Bone
posal of a classification based on clinicopathologi-
cysts: unicameral and aneurysmal bone cyst. Orthop cal, radiographic and genetic features. Skelet Radiol.
Traumatol Surg Res. 2015;101(1 Suppl):S119–27. 2012;41(12):1495–507. https://doi.org/10.1007/
https://doi.org/10.1016/j.otsr.2014.06.031. s00256-012-1510-6.
Chondrosarcoma of the Pelvis
6
Gerhard M. Hobusch and Reinhard Windhager
6.1 Epidemiology, Presentation, tion by the tumor [8]. Depending on localization,

Diagnosis the regular digestive function is impaired and
voiding disorders and sexual dysfunction can
CS is a rare disease with an estimated incidence occur (see Fig. 6.4). The huge size and anatomic
of 3.4–5/1,000,000/year according to Netherlands involvement of neurovascular abdominal as well
cancer registry from 1989 to 2013 including 2186 as urological structures at the time of diagnosis
patients, the American SEER database from 1973 can create big challenges to both limb-salvage as
to 2003 including 2890 patients or the Vienna well as to ablative surgery [6, 9, 10]. A diagnostic
Bone, and soft tissue tumor registry from 1965 to preoperative CT- or ultrasound-guided biopsy is
2019 including 395 patients with CS [1, 2]. mandatory, even though preoperative histological
According to musculoskeletal-oncologic high- grading should be interpreted with great caution
volume centers, one quarter of all CSs occur in in the pelvis, although it is reliable in the CS of
the pelvis, thereby around 33–80% in iliac wing, the long bones. The concordance between the
13–50% in the acetabular region, and each preoperative biopsy and the final pathological
5–17% in the pubis as well as 5% in the adjacent analysis in terms of histological grade in pelvic
sacrum [3, 4]. There is a male dominance of 55% CS with 36% is much lower than in long-bone CS
and the tumor appears mainly in adults and only with 83% [11]. Independent of bioptic results, a
seldom in children and young adults [5]. wide resection is recommended in all cases of
CS in the pelvis grow deep inside and can pelvic masses because of unreliable prediction
therefore stay undetected and symptomless for a from biopsy and preoperative imaging [12]. All
long time, thus becoming huge masses [6]. surgical steps as the resection as well as the
Especially in low-grade tumors, gradually reconstruction should be planned as well as exe-
increasing pain is the most common presenting cuted by a multidisciplinary team.
symptom [7]. CS can be frequently linked to
compression of pelvic organs as bladder, pros-
tate, or bowel. Olivieri et al. presented a single 6.2 Imaging
case of advanced CS with even urinary obstruc-
Both plane radiographs as well as layered imag-
ing in terms of MRI and CT with facultative 3-D
G. M. Hobusch · R. Windhager (*) reconstructions are necessary in the management
Medical University of Vienna, Vienna, Austria of diagnosis and increasingly in the treatment of
e-mail: [email protected]; pelvic CS. Apart from imaging-differences
[email protected]

56 G. M. Hobusch and R. Windhager
a b c
d f g
Fig. 6.1 Pre- and postoperative radiologic series of a (d) preoperative transversal MRI (T1 TSE Dotarem), (f)
50-year-old female patient with acetabular chondrosar- FDG PET-CT whole body, (g) postoperative ap pelvic
coma G3 and after II + III resection of the left hemipelvis X-ray with custom-made pelvis prosthesis. X-rays ap of a
and reconstruction with a custom-made pelvis prosthesis. 51-year-old patient with a 17-year follow-up after II + III
(a, c) Preoperative coronar pelvic MRI (STIR and T1 TSE resection of the right hemipelvis and reconstruction with a
Dotarem), (b, e) preoperative coronar and transversal CT, custom-made pelvis prosthesis after pelvic CS G2
Fig. 6.2 X-rays ap of a 58-year-old female patient with a 6-months follow-up after G3 dedifferentiated CS, type 2–3
resection left hemipelvis and reconstruction with a stemmed acetabular cup prosthesis
6 Chondrosarcoma of the Pelvis 57
between subtypes of chondrogenic lesions, there CS. On the other hand, different proportions of
are common radiological diagnostic features like low grade (18–26%), grade 2 (37–61%), and
the typical chondroid matrix-calcification (ring- grade 3 (8–22%) [4, 10, 18] tumors among differ-
and-arc-sign) and lobular-type architecture and ent published case series may indicate a large
endosteal scalloping [13]. Along with these variance in various regions in the world or diffi-
mixed lytic and sclerotic patterns, more aggres- culties in tumor grading.
sive patterns of bone destruction like moth-eaten
and permeative lysis can be seen [13]. Brenner
et al. argued a higher standardizes uptake value 6.4 Surgical Treatment
(SUV) in CS patients developing recurrent or
metastatic disease suggesting that as a preopera- According to the literature, hind quarter amputa-
tive staging tool, FDG-PET may provide tion, denoting the unilateral pelvic leg amputa-
improved outcome prediction in combination tion, is done in 13–52% of the CS patients in
with histopathologic findings [14]. Imaging is different tumor centers [4, 6, 17, 19, 20]. Limb-
becoming increasingly important in the surgical salvage in pelvic CS patients is considered the
planning as well as perioperatively as CT or gold standard nowadays and is supported by the
fused preoperative MRI and CT images provide fact that limb-savage surgery is not a risk factor
computer-assisted navigation in surgical resec- for low survival [21]. Furthermore, functional
tions to reduce the risk of intralesional resection outcome shows better results after inner hemipel-
and to avoid too-wide margins to achieve better vectomy/limb salvage with MSTS93 between
functional outcomes. Gerbers et al. demonstrated 61.4% and 86.6% than after external hemipelvec-
a successful hip-joint saving surgery of a patient tomy with MSTS93 score of 20–37.6% in several
suffering from a periacetabular CS by optimiz- case series [10, 20]. Due to the predilection sites
ing surgical margins with computer navigation of pelvic CS, inner hemipelvectomies can be per-
[15, 16]. formed often instead of forequarter amputation.
The Enneking–Dunn classification for internal
hemipelvectomies divides the pelvic bone into
6.3 Pathology three sections and the portion of the sacrum adja-
cent to the ilium as the fourth section and thus
In the pelvis, the conventional central CS is the describes the anatomic regions of resection [22].
most dominating subtype in about 47–80% and Type I (iliosacral) resections depend on
arises in normal bone. It can be distinguished whether the continuity of the pelvic ring can be
from peripheral (secondary) subtypes in 30% that preserved (Type IA versus Type IB). In type IB
occur as a malignant transformation of bone resections, there is agreement not to reconstruct
surface-based osteochondroma or enchondroma. the defect; however, even in Type IA resections,
The risk of malignant transformation in a solitary Beadel et al. promote a less mandatory restora-
lesion is estimated at up to 1% in comparison to tion of the pelvic ring after iliosacral resections
multiple lesions at up to 5%. A further distinction due to good clinical results in a case control study
can be made between rare subtypes like the dedif- with mixed sarcoma. The authors report similar
ferenciated central and peripheral CS in 1.4–15% functional scores like TESS 72% (55–100) in the
and considerably rarer the periosteal, clear cell, non-reconstruction compared with TESS 68%
and mesenchymal CS, in all together 2% of pel- (59–86) in the reconstruction group. Furthermore,
vic CS [2, 3, 6, 17]. Grading of CS is basically the chronic pain medication in the reconstruction
determined by cellularity, size and shape, and group was higher, the mean surgery time was
number of nuclei and mitotic figures and although 25% shorter (463:613 min), and the mean blood
grades 2 und 3 can be well distinguished, it is loss was 30% less (4325:6250 ml) in the non-
sometimes hard in differentiating a benign carti- reconstruction group. Authors conclude that
laginous lesion from a well-differentiated grade 1 complications were more common after recon-
a b
Fig. 6.3 Preoperative (a) and 4-years follow-up (b) X-rays hip-transposition with proximal femur GMRS® tumoren-
(ap) of a 68-year-old patient after dedifferentiated G3 CS doprosthesis, LARS® Ligament and Fixation with
of the right hemipelvis after I–IV resection and type IIb Mitek®-ancor
struction surgery [23]. On the other hand, several essary to preserve hip mobility [27]. Pring et al.
authors consider iliosacral arthrodesis by biologi- and Puri et al. described iliofemoral and ischiofe-
cal (fibula autograft/allograft) or composite moral arthrodesis after CS resection, a way to
osteosynthesis (using polyaxial screws and rods achieve a strong bone union between pelvic bone
in the sacrum and ischium/pubis) a good option and femur of limb-salvage without restoration of
when the acetabulum can be maintained and in the pelvic ring. Both of them describe good final
case a stabilization of the pelvic ring is desired MSTS93 results of 73 points with this technique
[19, 24]. Type II periacetabular resections leave a [19, 26, 27].
bone defect that can be bridged by many different Ever since limb-salvage is the primary surgi-
reconstruction methods or left without recon- cal objective, an anatomic reconstruction is the
struction or being left in a situation with a ilio- overriding wish. Large pelvic allografts, sole
femoral or iliosacral pseudoarthrosis or a hip allografts, or composite allografts in conjunction
transposition/flail hip [19, 25, 26]. The later, with a metal joint replacement as well as mega-
described by Gebert et al., is considered to be an prostheses for pelvic reconstruction are used to
adequate surgical intervention without recon- reach this goal, despite a background of high
struction of pelvic continuity. It can be achieved infectious and mechanical failure rates.
by a turnaround of the inferior part of the acetab- Competing risk analysis by Puchner et al. identi-
ulum and fixation with screws to the sacrum or fied endoprosthetic reconstruction after pelvic
iliac bone (type I), or by an artificial ligament sarcoma resection as the most significant factor
fixation of either the femoral head (type IIa) or a for the first major complication in terms of the
proximal endoprosthetic femoral replacement to ISOLS failure mode classification with a HR 4.9
the sacrum (type IIb) (see Fig. 6.3) [25]. Hip (2.2–9.8) [28]. Future studies will tell us, if the
transposition, according to Puri et al., provides use of personalized pelvic implants, improve-
good MSTS93 functional results of 73 points ments in implant surface-bone congruency as
apart from limp shortening and impaired gait. well as the use of patient-specific instruments
Therefore, this technique emphasizes that the and navigation for a better planning and to
reconstruction of the pelvic continuity is not nec- accomplish shorter operation times will be able
a b
c d e
Fig. 6.4 MRI of a permagna CS G2 of the right hemipel- fusion with Titanium rods and screws of the contralateral
vis crossing the middleline (a, b). X-rays of a postopera- pelvis. X-rays ap and lateral (c, d) and after partial
tive situs after external hemipelvectomy (extended removal of hardware after fusion due to infection (e)
type1–4 resection) and ventral stabilization. Iliospinal
to reduce infection rates and implant survival. sured by Toronto-extremity salvage score (TESS)
Jaiswal et al. sums up data about 98 patients of 70.3% (see Fig. 6.1), functional results change
(>50% CS patients) after resection and recon- for the worse when complications occurred to a
struction with custom-made pelvic endoprosthe- TESS of 37.1%. High complication rates con-
ses. After endoprosthetic reconstruction of cerning infection in 30% and dislocation in 15%
periacetabular tumors, infection rates are high, were also reported after reconstruction after Type
ranging between 10% and 30% [29]. Despite this II and II/III resection by stemmed acetabular ped-
fact, the infection rates in CS-only literature seem estal cups (see Fig. 6.2) [30, 31]. Competing risk
to be lower compared with mixed-sarcoma analysis revealed a surgical revision rate of 48%
cohorts of pelvic reconstruction [6, 20]. High dis- after 10 years [31]. High rates of aseptic loosen-
location rates of 40% in the beginning of endo- ing could suggest impaired bone quality after CS
prosthetic reconstructions were reduced down to for still unknown reasons for this mechanical
a rate of 20% due to efforts in introduction of problem [32]. Still, young patients with the
larger femoral head [29]. Although endopros- explicit demand for anatomic reconstruction
thetic reconstruction provides good functional might give a good indication for stemmed acetab-
results in patients without complications mea- ular cups.
Apart from small case series, saddle prosthe- the reason for radio-resistance. Due to limited
ses were no longer applied in musculoskeletal efficiency of currently available drugs, research
oncology for their considerable morbidity and focuses on the understanding of molecular
complications such as dislocation as well as het- pathomechanisms to develop molecularly tar-
erotopic ossification, vertical migration, and geted therapies [39].
infection leading to MSTS functional outcome of
50.8. Infection occurred in 8 of 18 CS patients
(44%) [33] (Figs. 6.3 and 6.4). 6.6 Oncologic Outcome
An alternative to metal implants in anatomic
reconstructions are allografts. Campanacci et al. Patients with CS are treated primarily by aggres-
concludes in a case series of pelvic massive sive surgical resection. Five-years and 10-years
allograft reconstruction that patients with CS overall survival in pelvic CS range between
might benefit most in comparison to other pri- 55–92% [4, 6, 10, 17, 35, 40] and 45–88% [4, 10,
mary malignancies considering the potential risk 35, 40] with significant differences between
of infection and necrosis and a high general mor- GI-III of 100/50/26% of survival with the lowest
bidity rate of massive allograft, due to less inter-survival of 48% in patients suffering from mesen-
ference with neoadjuvant treatment side effects. chymal and 10–24% in patients suffering from
In fact, CS patients in this cohort had a better dedifferentiating CS [2, 41]. Recurrent tumors
allograft survival and low allograft-necrosis than occur in 19–45% of pelvic CS patients [4, 6, 19,
patients with other malignancies [34]. In line 20, 40, 42, 43], the type of surgery may not be a
with this statement, Langlais et al. specify that reason for it [40]. Furthermore, several authors
massive allograft reconstructions should be done report a higher survival to death in Grade I in
in patients younger than 60 years, in good gen- comparison to Grade II or III CS [4, 10, 17, 18].
eral health condition and physically active Deloin et al. reported a significant higher propor-
because it is a demanding procedure [35]. tion of metastasis between low- and high-grade
Functional outcomes in periacetabular recon- CS patients and a consecutively lower survival in
structions with allografts are good and sometimes high-grade CS. According to Mavrogenis et al.,
excellent with MSTS93 of 61–67% [34, 35]. patients with metastasis are sixfold more likely to
In type III (ischiopubic) resections, recon- be high-grade than low-grade CS. Many different
structions are usually not mandatory due to per- factors like periacetabular location, pathologic
sistent pelvic continuity. fracture and tumor size are associated with low
To conclude, the surgeon’s choice ought to be survival [12, 17, 18, 44]. Surgical margins, con-
made under consideration of individual patients’ ferring to Stevenson et al., determine local recur-
attributes, functional requirements, and personal rence in all tumor-grades, but local recurrence
wishes [27]. itself affects disease-specific survival only in
grade II and III CS [12]. Competing risk analysis
revealed recurrent tumor as the most important
6.5 Radiotherapy risk factors for survival of CS patients with haz-
and Chemotherapy ard ratios of 5.5 and 6.07, respectively, which
emphasizes the importance of wide resections as
CS are considered as chemo- and radiation- a mainstay in CS surgery [4, 12]. Important to
resistant tumors, therefore being deployed only know is that even in recurrent cases of CS,
very sporadically in cases after recurrent tumor aggressive surgical intervention can lead to long-
or marginal resection borders [36]. The low vas- term survival in about 50% of the patients [45].
cularity of the tumors as well as the extracellular
matrix of hyaline and restricted drug penetrance Conflict of Interest Statement No benefits have been or
might be factors that count for chemoresistance will be received from a commercial party related directed
or indirectly to the subject matter of this article.
[37, 38]. The low mitotic fraction might also be
References drosarcoma patients. Eur J Nucl Med Mol Imaging.

2004;31(2):189–95.
15. Jeys L, Matharu GS, Nandra RS, Grimer RJ. Can
1. van Praag Veroniek VM, Rueten-Budde AJ, Ho
computer navigation-assisted surgery reduce the risk
V, Dijkstra PDS, Study group Bone and Soft tis-
of an intralesional margin and reduce the rate of local
sue tumours (WeBot), Fiocco M, et al. Incidence,
recurrence in patients with a tumour of the pelvis or
outcomes and prognostic factors during 25 years
sacrum? Bone Joint J. 2013;95-B(10):1417–24.
of treatment of chondrosarcomas. Surg Oncol.
16. Gerbers JG, Jutte PC. Hip-sparing approach using
2018;27(3):402–8.
computer navigation in periacetabular chondrosar-
2. Giuffrida AY, Burgueno JE, Koniaris LG, Gutierrez
coma. Comput Aided Surg. 2013;18(1–2):27–32.
JC, Duncan R, Scully SP. Chondrosarcoma in the
17. Bus MPA, Campanacci DA, Albergo JI, Leithner A,
United States (1973 to 2003): an analysis of 2890
van de Sande MAJ, Gaston CL, et al. Conventional
cases from the SEER database. J Bone Joint Surg Am.
primary central chondrosarcoma of the pelvis: prog-
2009;91(5):1063–72.
nostic factors and outcome of surgical treatment in 162
3. Picci P, Manfrini M, Fabbri N, Gambarotti M, Vanel
patients. J Bone Joint Surg Am. 2018;100(4):316–25.
D, editors. Atlas of musculoskeletal tumors and
18. Mavrogenis AF, Angelini A, Drago G, Merlino

tumorlike lesions: the Rizzoli case archive [Internet].
B, Ruggieri P. Survival analysis of patients with
Cham: Springer International Publishing; 2014. [cited
chondrosarcomas of the pelvis. J Surg Oncol.
2019 Aug 10]. Available from: https://www.springer.
2013;108(1):19–27.
com/de/book/9783319017471
19.
Pring ME, Weber KL, Unni KK, Sim
4. Stihsen C, Panotopoulos J, Puchner SE, Sevelda F,
FH. Chondrosarcoma of the pelvis. A review
Kaider A, Windhager R, et al. The outcome of the sur-
of sixty-four cases. J Bone Joint Surg Am.
gical treatment of pelvic chondrosarcomas: a compet-
2001;83(11):1630–42.
ing risk analysis of 58 tumours from a single centre.
20. Wirbel RJ, Schulte M, Maier B, Koschnik M,

Bone Joint J. 2017;99-B(5):686–96.
Mutschler WE. Chondrosarcoma of the pel-
5. Aprin H, Riseborough EJ, Hall JE. Chondrosarcoma
vis: oncologic and functional outcome. Sarcoma.
in children and adolescents. Clin Orthop Relat Res.
2000;4(4):161–8.
1982;166:226–32.
21. Kim C-Y, Collier CD, Liu RW, Getty PJ. Are limb-
6. Guo W, Li D, Tang X, Ji T. Surgical treatment of pel-
sparing surgical resections comparable to amputation
vic chondrosarcoma involving periacetabulum. J Surg
for patients with pelvic chondrosarcoma? A case-
Oncol. 2010;101(2):160–5.
control, propensity score-matched analysis of the
7. Karpik M, Reszeć J. Low grade chondrosarcoma -
National Cancer Database. Clin Orthop Relat Res.
epidemiology, diagnosis, treatment. Ortop Traumatol
2019;477(3):596–605.
Rehabil. 2018;20(1):65–70.
22. Enneking WF, Dunham WK. Resection and recon-
8. Olivieri V, Fortunati V, Annoscia S, Massarelli M,
struction for primary neoplasms involving the innomi-
Bellei L, Ollino M, et al. Advanced chondrosarcoma
nate bone. J Bone Joint Surg Am. 1978;60(6):731–46.
of the pelvis: a rare case of urinary obstruction. Arch
23. Beadel GP, McLaughlin CE, Aljassir F, Turcotte RE,
Ital Urol Androl. 2019;90(4):293–4.
Isler MH, Ferguson P, et al. Iliosacral resection for
9. Frassica FJ, Chao EY, Sim FH. Special prob-
primary bone tumors: is pelvic reconstruction neces-
lems in limb-salvage surgery. Semin Surg Oncol.
sary? Clin Orthop Relat Res. 2005;438:22–9.
1997;13(1):55–63.
24. Ozaki T, Hillmann A, Bettin D, Wuisman P,

10. Deloin X, Dumaine V, Biau D, Karoubi M, Babinet
Winkelmann W. High complication rates with pelvic
A, Tomeno B, et al. Pelvic chondrosarcomas: surgi-
allografts. Experience of 22 sarcoma resections. Acta
cal treatment options. Orthop Traumatol Surg Res.
Orthop Scand. 1996;67(4):333–8.
2009;95(6):393–401.
25. Gebert C, Wessling M, Hoffmann C, Roedl R,

11. Roitman PD, Farfalli GL, Ayerza MA, Múscolo DL,
Winkelmann W, Gosheger G, et al. Hip transpo-
Milano FE, Aponte-Tinao LA. Is needle biopsy clini-
sition as a limb salvage procedure following the
cally useful in preoperative grading of central chon-
resection of periacetabular tumors. J Surg Oncol.
drosarcoma of the pelvis and long bones? Clin Orthop
2011;103(3):269–75.
Relat Res. 2017;475(3):808–14.
26. Schwartz AJ, Kiatisevi P, Eilber FC, Eilber FR,

12. Stevenson JD, Laitinen MK, Parry MC, Sumathi V,
Eckardt JJ. The Friedman-Eilber resection arthro-
Grimer RJ, Jeys LM. The role of surgical margins in
plasty of the pelvis. Clin Orthop Relat Res.
chondrosarcoma. Eur J Surg Oncol. 2018;44(9):1412–8.
2009;467(11):2825–30.
13. Murphey MD, Walker EA, Wilson AJ, Kransdorf

27. Puri A, Pruthi M, Gulia A. Outcomes after limb spar-
MJ, Temple HT, Gannon FH. From the archives
ing resection in primary malignant pelvic tumors. Eur
of the AFIP: imaging of primary chondrosarcoma:
J Surg Oncol. 2014;40(1):27–33.
radiologic-pathologic correlation. Radiographics.
28. Puchner SE, Funovics PT, Böhler C, Kaider A,

2003;23(5):1245–78.
Stihsen C, Hobusch GM, et al. Oncological and surgi-
14. Brenner W, Conrad EU, Eary JF. FDG PET imag-
cal outcome after treatment of pelvic sarcomas. PLoS
ing for grading and prediction of outcome in chon-
One. 2017;12(2):e0172203.
29. Jaiswal PK, Aston WJS, Grimer RJ, Abudu A, Carter the development of chondrosarcoma of bone and
S, Blunn G, et al. Peri-acetabular resection and endo- implications for targeted treatment. Lancet Oncol.
prosthetic reconstruction for tumours of the acetabu- 2005;6(8):599–607.
lum. J Bone Joint Surg Br. 2008;90(9):1222–7. 38. van Oosterwijk JG, Herpers B, Meijer D, Briaire-de
30. Bus MPA, Boerhout EJ, Bramer JAM, Dijkstra
Bruijn IH, Cleton-Jansen AM, Gelderblom H, et al.
PDS. Clinical outcome of pedestal cup endoprosthetic Restoration of chemosensitivity for doxorubicin and
reconstruction after resection of a peri-acetabular cisplatin in chondrosarcoma in vitro: BCL-2 fam-
tumour. Bone Joint J. 2014;96-B(12):1706–12. ily members cause chemoresistance. Ann Oncol.
31. Hipfl C, Stihsen C, Puchner SE, Kaider A, Dominkus 2012;23(6):1617–26.
M, Funovics PT, et al. Pelvic reconstruction fol- 39. Polychronidou G, Karavasilis V, Pollack SM, Huang
lowing resection of malignant bone tumours using PH, Lee A, Jones RL. Novel therapeutic approaches in
a stemmed acetabular pedestal cup. Bone Joint J. chondrosarcoma. Future Oncol. 2017;13(7):637–48.
2017;99-B(6):841–8. 40. Donati D, El Ghoneimy A, Bertoni F, Di Bella C,
32. Hobusch GM, Tiefenboeck TM, Patsch J, Krall C, Mercuri M. Surgical treatment and outcome of con-
Holzer G. Do patients after chondrosarcoma treatment ventional pelvic chondrosarcoma. J Bone Joint Surg
have age-appropriate bone mineral density in the long Br. 2005;87(11):1527–30.
term? Clin Orthop Relat Res. 2016;474(6):1508–15. 41. Lex JR, Evans S, Stevenson JD, Parry M, Jeys LM,
33. Aljassir F, Beadel GP, Turcotte RE, Griffin AM, Bell Grimer RJ. Dedifferentiated chondrosarcoma of the
RS, Wunder JS, et al. Outcome after pelvic sarcoma pelvis: clinical outcomes and current treatment. Clin
resection reconstructed with saddle prosthesis. Clin Sarcoma Res. 2018;8:23.
Orthop Relat Res. 2005;438:36–41. 42. Ozaki T, Hillmann A, Lindner N, Blasius S,

34. Campanacci D, Chacon S, Mondanelli N, Beltrami G, Winkelmann W. Chondrosarcoma of the pelvis. Clin
Scoccianti G, Caff G, et al. Pelvic massive allograft Orthop Relat Res. 1997;337:226–39.
reconstruction after bone tumour resection. Int 43. Sheth DS, Yasko AW, Johnson ME, Ayala AG, Murray
Orthop. 2012;36(12):2529–36. JA, Romsdahl MM. Chondrosarcoma of the pelvis.
35. Langlais F, Lambotte JC, Thomazeau H. Long-term Prognostic factors for 67 patients treated with defini-
results of hemipelvis reconstruction with allografts. tive surgery. Cancer. 1996;78(4):745–50.
Clin Orthop Relat Res. 2001;388:178–86. 44. Chen L, Long C, Liu J, Duan X, Xiang Z. Prognostic
36. Riedel RF, Larrier N, Dodd L, Kirsch D, Martinez nomograms to predict overall survival and cancer-
S, Brigman BE. The clinical management of specific survival in patients with pelvic chondrosar-
chondrosarcoma. Curr Treat Options Oncol. coma. Cancer Med. 2019;8(12):5438–49.
2009;10(1–2):94–106. 45. Weber KL, Pring ME, Sim FH. Treatment and out-
37.
Bovée JVMG, Cleton-Jansen A-M, Taminiau come of recurrent pelvic chondrosarcoma. Clin
AHM, Hogendoorn PCW. Emerging pathways in Orthop Relat Res. 2002;397:19–28.
Osteosarcoma of the Pelvis
7
Andre Spiguel, D. Ian English, Cory Couch,
7.1 Introduction Lastly, the histologic subtypes found in pelvic

osteosarcoma also seem to contribute to overall
Primary malignancies of bone are rare and poor survival. A higher percentage of these osteo-
account for less than 1% of all malignancies. sarcomas are chondroblastic when compared
Osteosarcoma is the most common primary bone with appendicular osteosarcomas, a histologic
tumor and it can occur anywhere in the axial and subtype that responds poorly to chemotherapy.
appendicular skeleton [1]. Survival today, with Secondary osteosarcomas also make up one-third
the use of multimodal chemotherapy and surgery, of pelvic osteosarcomas, another histologic sub-
is 60–70% at 5 years. Osteosarcoma of the pelvis, type with an inherently poor prognosis [4].
however, accounts for less than 10% of all osteo-
sarcomas and survival is far worse, ranging from
20% to 40% at 5 years [2]. 7.2 Histology/Pathophysiology
This stark contrast in survival can be attributed
to many factors. Patients with pelvic osteosar- Osteosarcoma is an osteoid-producing neoplastic
coma often present with larger tumors and up to tumor of mesenchymal origin. In the pelvis,
50% of these patients present with metastasis at osteosarcoma most commonly presents as intra-
diagnosis [2]. They often have a delay in diagno- medullary or conventional high-grade osteosar-
sis or misdiagnosis and present with longer dura- coma between ages 15 and 30. Intramedullary
tion of symptoms. Symptoms are frequently osteosarcomas are further classified into subtypes
associated with vague pain often mimicking other differentiated by histologic and radiographic
benign conditions [3]. Surgical management of appearance including conventional (osteoblastic,
these tumors is quite challenging. There is poor chondroblastic, fibroblastic), telangiectatic, small
tumor compartmentalization within the pelvis and cell, giant cell rich, and osteoblastoma-like. Less
achieving a margin can be difficult, given tumor common are surface-based pelvic osteosarcoma
size and proximity to vital structures. subtypes including periosteal, parosteal, and
high-grade surface osteosarcomas.
Studies show that the most common pelvic
subtype is osteogenic osteosarcoma, making up
A. Spiguel · D. I. English · C. Couch about 70% of cases [5]. Parry et al. reported on
M. Scarborough (*) the largest series of primary pelvic osteosarcoma
Department of Orthopaedics and Rehabilitation,
University of Florida, Gainesville, FL, USA
and found that the chondroblastic variant was
e-mail: [email protected]; [email protected]. most common, with worse 5-year survival, as
edu; [email protected]; [email protected] previously mentioned [2].
64 A. Spiguel et al.
A significant proportion of pelvic osteosarco- from the bone, and osteoid production is often
mas are secondary and affect patients who are seen throughout the soft tissue mass on radio-
60–80 years of age. These secondary osteosarco- graphs. Plain X-rays are also important in the
mas arise in the setting of prior pelvic radiation postoperative setting for purposes of
or from Paget’s Disease. The reasons for prior surveillance.
pelvic radiation in these patients can include gen- Axial imaging for pelvic osteosarcoma is
itourinary or lower intestinal carcinomas, most essential and allows the surgeon the ability to
commonly prostate cancer in men and cervical/ define the local extent of the tumor. CT scan of
anorectal cancer in women. Paget’s Osteosarcoma the tumor provides information in regard to the
has significantly poorer prognosis with some bony anatomy and ossification/mineralization of
series showing 0% 5-year survival [2]. This type the soft tissues. It helps delineate the osteoid for-
of secondary osteosarcoma is typically both mation and can show the cortical destruction
radiotherapy and chemotherapy resistant and caused by the malignancy (Fig. 7.4). MRI has
poor survival is seen given patients advanced age been shown to be the most sensitive study to
and comorbidities [6]. determine tumor boundaries when preparing for
surgical resection (Fig. 7.5) [7]. MRI with and
without gadolinium contrast has supplanted CT
7.3 Imaging Osteosarcoma as the imaging modality of choice for many sur-
of the Pelvis geons. MRI allows visualization of the intramed-
ullary extent of the disease in addition to defining
Appropriate imaging is critical for the diagnosis, the soft tissue expansion of the tumor. Osteoid
staging, and subsequent surgical management of matrix demonstrates hypointensity on both T1
osteosarcoma of the pelvis. Multimodal imaging and T2 weighted images. MRI can also clarify
is obtained, and each facet of the workup pro- the relationship of the tumor to critical neurovas-
vides unique and important data. cular and visceral structures in order to determine
While advanced imaging has enhanced the the feasibility of surgical resection. The entire
surgeon’s ability to visualize a tumor and its rela- pelvis is imaged to assess for skip lesions, which
tionship to vital structures and surrounding soft is important in both surgical planning and is an
tissues, plain radiographs remain the first line. important negative prognostic indicator. MRI of
Workup typically consists of an AP Pelvis radio- the pelvis will show the hip joint and femoral
graph, Judet views and inlet/outlet X-rays can heads to check for intra-articular involvement
also prove useful when trying to understanding (Fig. 7.6). The absence of a hip joint effusion has
the anatomy of the tumor (Figs. 7.1–7.3). been shown to have a high negative predictive
Radiographs of conventional osteoblastic osteo- value, but the presence of an effusion carries a
sarcoma reveal a mixed radiolucent and sclerotic low sensitivity [8]. The sacroiliac joint should
lesion with a distorted trabecular pattern. The also be closely examined, trans-articular exten-
boundaries of these aggressive malignancies are sion can occur and happens more frequently in
usually ill-defined. Rapid growth may lead to osteosarcoma and chondrosarcoma than Ewing’s
elevated periosteum, creating a Codman’s trian- sarcoma [9]. When osteosarcoma does contami-
gle, and new bone growth under the elevated nate the SI joint, it has been shown to preferen-
periosteum can lead to a sunburst pattern. A key tially cross at the intraosseous ligamentous
finding of osteosarcoma is osteoid production, portion in the posterior joint as opposed to the
which is commonly described as amorphous and articular cartilage anteriorly [10], so this should
ill-defined with an inability to determine where be carefully scrutinized for disease.
the lesion starts and stops. Patients commonly In addition to delineating the local extent of
present with large soft tissue masses expanding the tumor, advances in MRI technology may
7 Osteosarcoma of the Pelvis 65
Figs. 7.1–7.3 AP Pelvis and Judet radiographs of a imaging reveals the ossified soft tissue component is
58-year-old male with left sided hip pain. Obtaining along the posterior ilium. The iliac oblique radiograph
radiographs is an important first step in the diagnosis of shows the mixed sclerotic and lytic lesion involving of the
pelvic osteosarcoma. These radiographs depict a poorly caudal aspect of the posterior ilium with expansion
defined, destructive lesion of the posterior ilium with towards the sciatic notch
patchy extraosseous mineralization. Obturator oblique
allow evaluation of the response to neoadjuvant [11]. The roles of these MRI sequences in the
therapy. Dynamic contrast-enhanced (DCE) MRI treatment of osteosarcoma are still under investi-
depicts the microvascularity to the tumor, while gation. Care must be taken when applying these
apparent diffusion coefficient (ADC) quantifies advanced modalities because the higher water
the changing tumoral water content in response content of chondroblastic osteosarcoma can be
to treatment. Magnetic resonance spectroscopy misinterpreted for necrosis [12].
(MRS) detects cell turnover activity and may act Patients must be assessed for metastatic dis-
as a stand-in for biologic aggressiveness and also ease once the diagnosis of pelvic osteosarcoma is
reveal the cellular response to chemotherapy suspected. The most frequent site of metastasis is
the lungs, and a non-contrast CT scan has been tigraphy and 18-Fluorodeoxy-glucose Positron
shown to be more sensitive than chest X-rays in emission tomography (18FDG-PET) combined
detecting pulmonary nodules [13]. with whole body CT. Tc-99 bone scan remains
Other imaging modalities to evaluate for the standard of care for staging and is a low-cost
metastases include Tc-99 whole-body bone scin- means of examining the entire skeleton (Fig. 7.7).
The Children’s Oncology Group imaging guide-
lines for children with osteosarcoma states that
bone scintigraphy is required for staging, while
performing a single photon emission computer
tomography (SPECT) in conjunction with the
bone scan is a recommendation. The group rec-
ommends the use of FDG-PET for osteosarcoma
but clearly states this is not a requirement [14].
FDG is a radiotracer that is metabolized in highly
active tissue (Fig. 7.8). A PET-CT can therefore
quantify metabolic activity of a tumor and possi-
bly show responses to chemotherapy in addition
to revealing distant metastasis. The ability of
PET scans to distinguish neoplastic pulmonary
nodules from benign masses has also proven use-
Fig. 7.4 CT scan of the pelvis shows the destructive ful, both in initial staging and during surveillance
nature of the osteosarcoma with obliterated cortices along [15]. Another utility of PET lies in detecting local
the posterior ilium and loss of the trabecular architecture. recurrence and separating return of malignancy
Osteoid production in the soft tissue mass is also well
visualized on CT imaging
versus benign postoperative changes [16].
Figs. 7.5 and 7.6 On axial T1 MRI with gadolinium structures are free from involvement. There is intraosse-
contrast, this osteosarcoma is hypointense and demon- ous expansion to the posterior SI joint and posterior S2
strates contrast enhancement. The tumor involves the glu- and S3 neuroforamen. The coronal MRI shows the tumor
teus medius posteriorly, but the anterior neurovascular is extraarticular and did not invade the hip joint
7.4 Medical Treatment for Pelvic

Osteosarcoma
Until the application of chemotherapy, the sur-

vival rate of osteosarcoma was below 20%.
Patients were commonly treated with ablative
surgeries such as limb amputation or hemipelvec-
tomy, but despite local control via tumor removal,
survival rates failed to improve. In the 1970s,
methotrexate was used to treat osteosarcoma fol-
lowing its success with leukemia patients. Various
trials of systemic chemotherapeutic agents
through this period demonstrated improved sur-
vival and reduced metastatic burden. Each indi-
vidual drug demonstrates a unique response rate
in treating osteosarcoma, most commonly around
30–40%. The scientific community quickly rec-
ognized the benefit of combining chemothera-
peutic agents. Randomized, prospective trials in
the 1980s clearly demonstrated superiority of
adjuvant multimodal chemotherapy over surgery
alone [17, 18].
The development of neoadjuvant chemother-
apy prior to surgery developed as a means of pre-
venting metastasis during the months required to
fabricate custom implants for joint reconstruction
[19]. Concerns regarding the delayed delivery of
chemotherapy with this approach were put to rest
Fig. 7.7 Nuclear medicine bone scan is used to evaluate with a prospective trial showing no difference
for local and distant metastases. This patient has an iso- between neoadjuvant chemotherapy followed by
lated tumor of the left posterior ilium
surgery versus immediate surgery and adjuvant
chemotherapy [20]. A benefit of neoadjuvant
treatment is that biologic response can be
assessed by evaluation of the extent of tumor
necrosis at the time of surgical resection.
Most treatment protocols call for 10 weeks of
neoadjuvant chemotherapy followed by surgical
tumor resection, followed by 20 weeks of adju-
vant chemotherapy. The most common used
agents today are methotrexate, doxorubicin, and
cisplatin, owing the acronym MAP to usage of
drug trade names (Adriamycin® [Bedford
Fig. 7.8 PET-CT shows uptake of 2-deoxy-fluoro-d-glu-
Laboratories; Bedford, OH] Platinol® [Bristol-
cose in the metabolically active malignancy. The red col-
oring depicts the areas highest avidity, demonstrated here Myers Squibb; Princeton, NJ]). Several other
along the posterior periphery of the tumor drugs have also proven effective against osteosar-
coma including Ifosfamide, etoposide, cyclo-
phosphamide, vincristine, and bleomycin.
Ifosfamide and etoposide have been shown to be
especially useful in treating recurrent and apy, in which T cells are engineered to detect and
metastatic disease. Unfortunately, the addition of destroy tumor cells that have downregulated cer-
Ifosfamide and etoposide upfront to MAP neoad- tain proteins in order to avoid recognition by the
juvant protocols has not been shown to reduce the immune system. Chimeric antigen receptor T
rates of local recurrence or metastasis [21]. cells (CAR-Ts) and Tumor-infiltrating lympho-
With current multimodal treatment, the sur- cytes (TILs) are two such examples of adoptive
vival rate most frequently quoted is 70% for cell immunity.
osteosarcoma. This often-cited rate applies to
localized disease of the extremities in which wide
surgical margins are achieved [22]. However, 7.5 Surgical Treatment
osteosarcoma of the pelvis conveys a much
poorer prognosis. Regardless of disease location, 7.5.1 General Principles
the statistics for survival have not changed sig-
nificantly in the past three decades. Stagnation in En bloc resection with a wide margin is the rec-
curative rates and dismal outcomes for pelvic and ommended surgical treatment for osteosarcoma.
metastatic disease have inspired efforts to find Patients with pelvic osteosarcomas that present
new means of treating osteosarcoma. with smaller tumors are more likely to have a
There are several exciting therapies emerging limb sparing resection or an internal hemipelvec-
in the treatment of osteosarcoma, most of which tomy, whereas larger tumors are more likely to
do not involve archetypal cytotoxic drugs. Most require limb sacrificing resection or an external
strategies of these novel therapeutics rely on hemipelvectomy [26, 27].
advances in knowledge regarding how osteosar- Due to location, proximity to viscera and neu-
coma avoids detection and destruction by the rovascular structures and large size at presenta-
patient’s immune system. One such example is tion, osteosarcoma of the pelvis is a challenge to
muramyl tripeptide-phosphatidyl ethanolamine treat surgically. This challenge leads to prolonged
(MTP-PE, or mifamurtide), which is an analog surgical time, high rate of complications, and
of a bacterial cell wall and causes the activation high blood loss. The goals of surgical treatment
of monocytes and macrophages to become of osteosarcoma of the pelvis is adequate local
tumoricidal. Mifamurtide is currently approved oncologic control and to optimize function for
for clinical use in Europe but not in the United quality of life.
States [23].
The development of monoclonal antibodies to
osteosarcoma cell surface markers may also lead 7.5.2 Effect of Margins
to tumor destruction. There have been several
antibody designs to attack osteosarcoma cell Parry et al. showed that poorer outcomes resulted
lines, including bispecific T-cell engagers (BiTE) in patients with pelvic osteosarcoma that had an
antibodies, which bring tumor cells in close prox- intralesional margin [2]. They found a local
imity to T-cells that can then be activated and recurrence rate of 17% in patients with wide mar-
directly cause cellular destruction. Antibodies gins, 32% for marginal margins, and 44% for
have also been designed to selectively deliver intralesional margins.
cytotoxic chemicals to tumor cells, such as vedo- Other authors have found that despite achiev-
tin [24]. Vaccines consisting of tumor cells have ing wide margins in pelvic osteosarcoma sur-
also been developed. Dendritic cells detect the gery, the local recurrence rate is still high.
tumor antigens in the vaccines and prime T cells Donati et al. showed in a study of 60 patients
to create cytotoxic T lymphocytes. Oncolytic with primary high-grade pelvic osteosarcoma
viruses that will only reproduce within tumor that although they obtained wide margins in 18
cells are also being developed [25]. Another divi- of the cases, 8 patients had local recurrence
sion of emerging therapies is adoptive cell ther- [28]. Similar outcomes have also been shown in
a series of primary pelvic sarcomas by Shin correlation of higher tumor necrosis response to
et al. [29] chemotherapy linked with a better prognosis.
Internal hemipelvectomy should not be con-
sidered in cases where sciatic nerve roots must be
7.5.3 L
imb Sacrificing Resection/ sacrificed.
External Hemipelvectomy
External hemipelvectomy or hindquarter amputa- 7.5.5 Computer Navigation

tion is surgical resection of the osteosarcoma-
involved pelvis with sacrifice of the lower limb. Limb sparing pelvic tumor resection can be a dif-
External hemipelvectomy should be considered ficult operation due to the complexity of pelvic
in cases where a functional limb salvage cannot anatomy. Computer navigation has recently
be performed. This procedure is not recom- become available for improving the precision of
mended in cases of widespread metastatic dis- pelvic tumor resection (Fig. 7.9). Computer-
ease that is unresponsive to chemotherapy, except navigated pelvic resection is a novel technology,
for palliation in patients with intractable pain. and therefore it has not been widely adopted, but
This procedure is classically performed with a reports have been published on its effectiveness.
posterior fasciocutaneous or myofascial-Laitinen et al. concluded that navigation-assisted
cutaneous flap with the gluteus maximus for surgery for posterior ilium and sacral tumor
wound coverage. External hemipelvectomy can resection improves oncological outcome and
also be performed with an anterior myofascial- safety for the patient since navigation reduced the
cutaneous flap with the quadriceps musculature rate of intralesional margins [31]. Abraham et al.
or a filet flap of the lower leg musculature on a showed that in a series of 23 patients undergoing
vascular pedicle maintaining the external iliac navigated resection of pelvic and sacral tumors
vessels in the pelvis, superficial femoral vessels, that they were able to perform resection with
and popliteal vessels as it courses through the negative margins in 21 of the patients [32]. Wong
lower extremity to supply the flap. et al. showed that in simulated pelvic tumor
resections, patient-specific instrumentation is as
accurate as navigation, but with shorter bone
7.5.4 Limb Sparing Resection/ resection time [33].
Internal Hemipelvectomy
Internal hemipelvectomy is the surgical resection

of pelvic osteosarcoma and can include the
sacrum, ilium, ischium, and/or pubis, while spar-
ing a functional lower extremity. Advances in the
diagnostic and therapeutic regimens as well as
implementation of a multidisciplinary team
approach to treating pelvic osteosarcoma have
paved the path to limb salvage.
Appropriate patient selection and planning of
surgical resection with wide margins in an inter-
nal hemipelvectomy can provide a good func-
tional outcome. Salunke et al. showed in their
series of 23 patients with pelvic bone sarcomas
treated with internal hemipelvectomy that there Fig. 7.9 Post-operative radiograph reveals a modified
was an 83% 5-year survival rate and a good mean hemipelvectomy with preservation of the hip joint and
functional MSTS score [30]. They did show a resection of the sacroiliac joint and sacral ala
7.6 Summary 12. Wang J, Sun M, Liu D, Hu X, Pui MH, Meng Q,

Gao Z. Correlation between apparent diffusion coef-
ficient and histopathology subtypes of osteosar-
Osteosarcoma of the pelvis continues to be one of coma after neoadjuvant chemotherapy. Acta Radiol.
the most challenging diseases treated by orthope- 2017;58(8):971–6.
dic oncologists. Treatment goals are the same in 13. Marina N, Gebhardt M, Teot L, Gorlick R. Biology
and therapeutic advance for pediatric osteosarcoma.
all osteosarcomas. However, there are significant Oncologist. 2004;9:422–41.
negative factors associated with this entity includ- 14. Dehury PP. Imaging guidelines for children with

ing increased local recurrence rates, decreased Ewing sarcoma and osteosarcoma: a report from the
survival, and overall worse function and morbid- Children’s Oncology Group Bone Tumor Committee.
Pediatr Blood Cancer. 2008;51:163–70.
ity after surgical management. 15. Cistaro A, Lopci E, Gastaldo L, Fania P, del Prever B,
Adalberto, Fagioli F. The role of 18F-FDG PET/CT
in the metabolic characterization of lung nodules in
pediatric patients with bone sarcoma. Pediatr Blood
Cancer. 2012;59(7):1206–10.
References 16. Rajiah P, Ilaslan H, Sundaram M. Imaging of sar-
comas of pelvic bones. Semin Ultrasound CT MRI.
1. Fletcher CDM, Bridge JA, Hogendoorn PCW, 2011;32(5):433–41.
Mertens F, editors. World Health Organization, classi- 17. Eilber BF, Giuliano A, Eckardt J, Patterson K,

fication of tumours: pathology and genetics of tumors Moseley S, Goodnight J. Adjuvant chemotherapy for
of soft tissue and bone. Lyon: IARC Press; 2013. osteosarcoma: a randomized prospective trial. J Clin
2. Parry MC, Jeys L, Carter S, Gaston CL, Sumathi V, Oncol. 1987;5(1):21–6.
Grimer RJ, Albergo J. Osteosarcoma of the pelvis. 18. Link MP, Goorin AM, Miser AW, Green AA, Pratt
Bone Joint J. 2016;98B(4):555–63. CB, Belasco JB, Pritchard J, Malpas JS, Baker AR,
3. Kadhim M, Oyoun NA, Womer RB, Dormans Kirkpatrick JA, Ayala AG, Shuster JJ, Abelson HT,
JP. Clinical and radiographic presentation of pelvic Simone J, T. J. The effect of adjuvant chemotherapy
sarcoma in children. SICOT J. 2018;4:44. on relapse-free survival in patients with osteosarcoma
4. Fuchs B, Hoekzema N, Larson DR, Inwards CY, of the extremity. N Engl J Med. 1986;314(25):1600–6.
Sim FH. Osteosarcoma of the pelvis: outcome anal- 19. Rosen G, Marcove RC, Caparros B, et al. Primary
ysis of surgical treatment. Clin Orthop Relat Res. osteogenic sarcoma: the rationale for preopera-
2009;467:510–8. tive chemotherapy and delayed surgery. Cancer.
5. Tang N, et al. Osteosarcoma development and 1979;43:2163–77.
stem cell differentiation. Clin Orthop Relat Res. 20. Goorin AM, Schwartzentruber DJ, Devidas M,

2008;466(9):2114–30. Gebhardt MC, Ayala AG, Harris MB, et al. Presurgical
6. Grimer RJ, Cannon SR, Taminiau AM, Bielack S, chemotherapy compared with immediate surgery and
Kempf-Bielack B, Windhager R, Van der Eijken adjuvant chemotherapy for nonmetastatic osteosar-
J. Osteosarcoma over the age of forty. Eur J Cancer. coma: Pediatric Oncology Group Study POG-8651. J
2003;39(2):157–63. Clin Oncol. 2003;21(8):1574–80.
7. Gillespy T, Manfrini M, Ruggieri P, Spanier SS, 21. Marina NM, Smeland S, Bielack SS, Bernstein M,
Pettersson H, Springfield DS. Staging of intraosseous Jovic G, Krailo MD, et al. Comparison of MAPIE
extent of osteosarcoma: correlation of preoperative versus MAP in patients with a poor response to preop-
CT and MR imaging with pathologic macroslides. erative chemotherapy for newly diagnosed high-grade
Radiology. 1988;167(3):765–7. osteosarcoma (EURAMOS-1): an open-label, inter-
8. Ozaki T, Putzke M, Bürger H, Gosheger G, national, randomised controlled trial. Lancet Oncol.
Winkelmann W, Lindner N. Infiltration of sar- 2016;17(10):1396–408.
comas into the hip joint. Acta Orthop Scand. 22. Chou AJ, Malek F. Osteosarcoma of bone. In:

2002;73(2):220–6. Biermann JS, editor. Orthopaedic knowledge update
9. Ozaki T, Lindner N, Hillmann A, Link T, Winkelmann 3: musculoskeletal tumors. Chicago, IL: American
W. Transarticular invasion of iliopelvic sarcomas into Academy of Orthopaedic Surgeons; 2014. p. 159–70.
the sacrum. Radiological analysis of 47 cases. Acta 23. Wedekind MF, Wagner LM, Cripe TP. Immunotherapy
Orthop Scand. 1997;68(4):381–3. for osteosarcoma: where do we go from here? Pediatr
10. Chhaya S, White LM, Kandel R, Wunder JS,
Blood Cancer. 2018;65(9):1–9.
Ferguson P, Agur A. Transarticular invasion of bone 24. Roth M, Barris D, Piperdi S, Kuo V, Everts S, Geller
tumours across the sacroiliac joint. Skelet Radiol. D, Houghton P, Kolb EA, Hawthorne T, Gill J, Gorlick
2005;34(12):771–7. R. Targeting glycoprotein NMB with antibody-
11. Fukuda T, Wengler K, de Carvalho R, Boonsri P, drug conjugate, Glembatumumab Vedotin, for the
Schweitzer ME. MRI biomarkers in osseous tumors. J treatment of osteosarcoma. Pediatr Blood Cancer.
Magn Reson Imaging. 2019;50(3):702–18. 2016;63:32–8.
25. Martinez-Velez N, Xipell E, Jauregui P, Zalacain M, 30. Salunke AA, Shah J, Warikoo V, Chakraborty A,

Marrodan L, Zandueta C, et al. The oncolytic adeno- Sahijwani H, Sharma M, Jain A. Surgical manage-
virus Δ24-RGD in combination with cisplatin exerts a ment of pelvic bone sarcoma with internal hemipel-
potent anti-osteosarcoma activity. J Bone Miner Res. vectomy: oncologic and functional outcomes. J Clin
2014;29(10):2287–96. Orthop Trauma. 2017;8(3):249–53.
26. Ham SJ, Koops HS, Veth RPH, Van Horn JR, Eisma 31. Laitinen MK, Parry MC, Albergo JI, Grimer RJ, Jeys
WH, Hoekstra HJ. External and internal hemipel- LM. Is computer navigation when used in the surgery
vectomy for sarcomas of the pelvic girdle: conse- of iliosacral pelvic bone tumours safer for the patient?
quences of limb-salvage treatment. Eur J Surg Oncol. Bone Joint J. 2017;99-B(2):261–6.
1997;23(6):540–6. 32. Abraham JA, Kenneally B, Amer K, Geller DS. Can
27. Huth JF, Eckardt JJ, Pignatti G, Eilber FR. Resection navigation-assisted surgery help achieve negative
of malignant bone tumors of the pelvic gir- margins in resection of pelvic and sacral tumors? Clin
dle without extremity amputation. Arch Surg. Orthop Relat Res. 2018;476(3):499–508.
1988;123(9):1121–4. 33. Wong KC, Sze KY, Wong IO, Wong CM, Kumta
28. Donati D, Giacomini S, Gozzi E, et al. Osteosarcoma SM. Patient-specific instrument can achieve same
of the pelvis. Eur J Surg Oncol. 2004;30(3):332–40. accuracy with less resection time than navigation
29. Shin KH, Rougraff BT, Simon MA. Oncologic out- assistance in periacetabular pelvic tumor surgery:
comes of primary bone sarcomas of the pelvis. Clin a cadaveric study. Int J Comput Assist Radiol Surg.
Orthop Relat Res. 1994;(304):207–17. 2016;11(2):307–16.
Ewing’s Sarcoma of the Pelvis
8
Benjamin V. Kelley, Danielle E. Greig,
and Francis J. Hornicek
8.1 Introduction 8.2 Histology
Although rare, Ewing’s sarcoma (ES) is the sec- EFT are high-grade malignancies characterized
ond most frequent primary bone malignancy after histologically by sheets of small, round blue
osteosarcoma [1, 2]. ES is part of a spectrum of cells with scant cytoplasm (Fig. 8.1a–c). These
neoplastic diseases known as the Ewing’s Family primitive, undifferentiated cells do not resemble
of Tumors (EFT) [3]. In addition to ES, this group mesodermal tissue in contrast to those of the
includes peripheral neuroectodermal tumor other primary bone malignancies osteosarcoma
(PNET), extraosseous ES (EES), atypical ES, and and chondrosarcoma, which demonstrate histo-
malignant small-cell tumors of the thoraco- logic features of bone and cartilage, respectively
pulmonary region (Askin tumor) [4–8]. Common [18–20]. The histologic features of ES may be
histologic and immunohistochemical features in distinguished from PNET, as the latter is charac-
addition to non-random chromosomal features terized by neural differentiation identified on
and similar responses to radiation and chemother- light microscopy and by a neural phenotype on
apy suggest that this group shares a common cell immunohistochemical staining [21]. Sensitive
of origin [9]. Increasing evidence suggests that immunohistochemical markers for EFT include
this is specifically a mesenchymal progenitor cell positive vimentin, FLI-1, NKX2.2, and CD99
[10], though this remains debated. Overall out- (Fig. 8.2), also known as MIC2 [22].
comes of patients with EFT have dramatically
improved with optimization of multimodal thera-
pies [11]. However, pelvic EFT has been associ- 8.3 Molecular Pathogenesis
ated with inferior outcomes compared with other
skeletal sites as a result of a higher rate of meta- The pathologic molecular alteration of EFT is the
static disease at presentation in addition to nearby reciprocal translocation between chromosomes
radiosensitive organs limiting the ability to 11 and 22, t(11;22)(q24;q12), which results in
achieve local control via surgical resection or the fusion gene EWSR1-FLI1 [23–28]. This
radiation without excessive morbidity [12–17]. fusion gene is present in 90–95% of cases, while
similar variant translocations and the respective
B. V. Kelley · D. E. Greig · F. J. Hornicek (*) fusion genes they have produced, such as EWSR-
University of California, Los Angeles, ERG, make up the remainder [23, 29]. These
Los Angeles, CA, USA fusion genes can be detected using fluorescence
e-mail: [email protected]; dgreig@mednet. in situ hybridization (FISH) or reverse transcrip-
ucla.edu; [email protected]

74 B. V. Kelley et al.
a b
Fig. 8.1 (a) H&E staining at 100× shows small, round tions, with sheets of cells divided into lobules by thin
blue cells with geographic necrosis (pseudo-rosettes). (b) fibrous strands. (c) H&E staining at 400× demonstrates a
H&E staining at 200× demonstrates typical rosette forma- uniform sheet of small, round blue cells
tase polymerase chain reaction (RT-PCR) for

molecular diagnosis of EFT with high levels of
sensitivity and specificity [30].
There are two likely mechanisms for the con-
tribution of these chimeric proteins to neoplastic
transformation [24, 27, 31, 32]: by influencing
the transcription of (1) genes normally regulated
by native FLI or EWS protein; and (2) genes dif-
ferent from those normally regulated by native
FLI or EWS proteins. Acting as transcriptional
activators, the chimeric fusion proteins drive
oncogenesis by deregulation of cell signaling,
apoptosis, telomerase activity, proliferation, and
Fig. 8.2 CD99 immunohistochemistry at 400× demon- metastasis [33]. The gene affected by the fusion
strates a typical CD99 positive membranous staining proteins may serve as a target for future
pattern
molecular-based therapy [34].
8 Ewing’s Sarcoma of the Pelvis 75
8.4 Epidemiology
Primary EFT may present throughout the axial

and appendicular skeleton [35]. Based on data
from 975 patients from the European Intergroup
Cooperative Ewing Sarcoma Studies (EI-CESS)
trials [36, 37], the pelvis is the primary site in
25% of ES cases [12, 38, 39]. According to the
National Cancer Institute’s Surveillance,
Epidemiology, and End Results Program (SEER),
there is a dramatic variation in race across
patients with ES, with white children having an
approximate sixfold higher incidence rate than
black children [40].
The peak incidence of EFT is in the second Fig. 8.3 Antero-posterior (AP) X-ray of the pelvis dem-
decade, while 30% of patients presenting before onstrating a large destructive lesion in the right iliac wing
with internal calcifications consistent with Ewing’s
the age of 10 and an additional 30% of patients sarcoma
after the age of 20 [41]. EFTs are rare after the
third decade [42]. There is significant racial and
gender variation in rates of EFT [40]. Caucasian
and male patients are disproportionately affected
and EFT are extremely rare in black or Asian
patients [40].
8.5 Imaging
The radiographic appearance of ES is variable,

but is classically described as “moth-eaten,”
demonstrating an aggressive poorly marginated
destructive lesion (Fig. 8.3) [43]. These tumors
fill the bone marrow cavity and destroy the cor- Fig. 8.4 Axial CT scan taken during image-guided
tex. Common features include permeative lesions biopsy of a patient with Ewing’s sarcoma demonstrating a
with associated periosteal reaction and sclerosis. lytic lesion in the right iliac wing with cortical disruption
This radiographic sclerosis is correlated with and internal calcifications
histologic sclerosis. While 80% of cases have an
associated soft tissue mass, soft tissue calcifica- involvement of adjacent structures. Computed
tion is rare and is found in only 10% of cases [44]. tomography (CT) allows for accurate delinea-
The tumor may expand the cortex and displace tion of cortical destruction (Fig. 8.4). Magnetic
the overlying periosteum, forming the character- resonance imaging (MRI) is preferred for defini-
istic clinical sign of Codman’s triangle—though tion of tumor size and tumor relation to fascial
this is more readily appreciated on long bone planes, blood vessels, nerves, and pelvic organs
lesions rather than pelvic tumors. In addition, (Fig. 8.5a–c). EFT may present with either a hot
most tumors demonstrate a laminated “onion or cold bone scan. Positron emission tomogra-
skin” periosteal reaction [43]. Cross-sectional phy (PET) is useful for staging but has not been
imaging is essential for preoperative planning shown to be predictive of chemotherapy respon-
in order to define tumor extent and evaluate for siveness [45–47].
a b c
Fig. 8.5 Coronal STIR (a), axial T2 (b), and sagittal T1 right iliac bone with a large, bi-cortical soft tissue compo-
(c) MRI of the pelvis in a patient with Ewing’s sarcoma nent and surrounding inflammatory zone
demonstrating a heterogeneous mass arising from the
marrow (most commonly spine), while lymph

8.6 Staging node, liver, and brain metastases are rare. Among
patients with clinical metastases, there is a trend
There is no staging system specific to EFT, but the for improved survival among those with lung
commonly utilized tumor, node, metastases metastases compared with those with bone
(TNM) system may be applied. Chest radio- metastases [58–60]. Patients with pelvic EFT are
graphs, chest/abdomen/pelvis CT scans, whole- significantly more likely to present with meta-
body technetium bone scans, and PET scans are static disease compared with those with a differ-
standard imaging modalities to detect metastases ent primary site (25% vs. 16%) [38]. Other risk
[45–50]. In general, EFT patients are staged as factors associated with clinically evident meta-
having either localized or metastatic disease, static disease at the time of diagnosis include
which is the strongest predictor of prognosis [51]. elevated lactic dehydrogenase (LDH), fever, and
Minimal metastatic disease in the peripheral an interval between symptom onset and diagno-
blood or bone marrow detected with PCR but not sis of less than 3 months [58, 61–63]. Other
detected on traditional imaging has been reported strong prognostic factors for poor overall survival
in as high as 30% of patients with otherwise local- include tumor volume over 200 ml, older age,
ized disease [52]. The prognostic significance of low socioeconomic status, and a histologic
molecularly detectable minimal disseminated dis- response of less than 100% to chemotherapy [64–
ease (MRD) remains controversial but may play a 67]. There has been moderate but conflicting evi-
role in explaining why some patients with local- dence regarding the prognostic significance of
ized disease have poor outcomes [53, 54]. achieving negative, disease-free tumor margins
during surgical resection on event-free survival
[65, 68].
8.7 Prognostic Factors
The presence of metastatic disease at presenta- 8.8 Biopsy

tion is the most important prognostic factor for
EFT [55, 56]. Five-year survival for patients with As with other bone and soft tissue malignancies,
localized disease is approximately 70%, com- pathologic biopsy is central to diagnosis. To avoid
pared with 33% in patients with metastatic dis- compromising potential future limb salvage sur-
ease at presentation [57]. The most common sites gery, surgeons must be consulted to plan the site
of metastatic disease are the lung and bone/bone of biopsy [69]. CT-guided core needle biopsy is
the most common mode of tissue sampling; how- The timing of chemotherapy in relation to
ever, open biopsy may be required if there is a local control has been identified as an important
high rate of tissue necrosis to reduce the effect of factor in the management of EFT [92–95].
sampling error and to obtain sufficient tissue for Results from the National Cancer Database dem-
multiple immunohistochemical tests [69]. onstrated improved survival when patients initi-
ated local therapy by week 15 of induction
chemotherapy. As a result, most current protocols
8.9 Treatment advocate for 12 weeks of induction chemother-
apy followed by immediate local control with
A multidisciplinary approach in treating patients surgery or radiation therapy [92].
with EFT is essential [11, 70–72]. EFT is a sys-
temic disease, and patients treated with local
therapy alone have high relapse rates [73]. 8.11 Surgery
Chemotherapy can eradicate metastatic deposits
if initiated when tumor burden is low and is Surgical resection is regarded as superior to radi-
therefore critical in the treatment of EFT [71]. ation therapy for local control and performed
Current standard treatment regimens for local- whenever a marginal or wide resection is feasi-
ized EFT consist of multiagent neoadjuvant che- ble [96–98]. While there has been no random-
motherapy followed by local treatment with ized prospective study to determine which is
surgical resection and/or radiation therapy, fol- superior, retrospective series including a 956
lowed by additional postoperative adjuvant che- patient cohort from the Children’s Oncology
motherapy [11, 74–77]. The treatment strategy Group (COG) have suggested higher failure
for patients with metastatic disease is similar to rates in patients treated with radiation therapy
that of localized disease as multimodal therapy alone as compared with surgery alone [97, 99].
with chemotherapy, radiation therapy, and surgi- Anatomic location within the pelvis plays a key
cal resection has the potential for cure, and can role in determining the approach for local con-
prolong progression-free survival and relieve trol [12–16]. Lesions of the iliac wing, ischium,
pain [75, 78]. In particular, the decision to pursue or pubis are more amenable to resection than
surgical resection in patients with metastatic dis- lesions involving the acetabulum, sacroiliac
ease must be carefully weighed against the joint, or sacrum [100, 101]. Surgical planning
impact of a long period of time off chemotherapy should be based on the MRI scan after induction
[55]. chemotherapy, which can be dramatically differ-
ent from the pre-chemotherapy MRI scan [102,
103]. Internal hemipelvectomy with en bloc
8.10 Chemotherapy excision of the affected hemipelvis is the most
common approach for surgical treatment of
The chemotherapy treatment strategy for EFT patients with pelvic EFT [13, 104, 105].
has been optimized from the efforts of multiple Preserving the extremity improves functional
collaborative national and international trials and psychological outcomes while producing
[79–81]. Multi-neoadjuvant chemotherapy equivalent oncologic results to external hemipel-
increases the likelihood of local control by reduc- vectomy (hindquarter amputation) [15, 102].
ing tumor burden prior to surgery [82–84]. In Depending on the clinical situation, surgeon, and
many cases, the soft tissue component of the patient preference, resection may or may not be
tumor may disappear completely [44, 85]. The accompanied by reconstruction with allograft or
traditional agents used in therapy include vincris- endoprosthesis. For tumors deemed unresectable
tine, doxorubicin, cyclophosphamide, dactino- following induction therapy, patients should be
mycin, and ifosfamide, with some regimens also referred for definitive radiation therapy rather
including etoposide [86–91]. than attempting debulking surgery [14].
8.12 Radiation Therapy approach similar to that of a primary lesion but

must be tailored to the patient on an individual-
The radiosensitivity of ES was noted in James ized basis, as the lower chance of disease eradica-
Ewing’s original description of the tumor in tion may guide treatment toward palliation [130,
1921. EFTs are radiosensitive but radiotherapy 132, 133]. Salvage chemotherapeutic treatment
alone is insufficient to prevent local recurrence protocols may include irinotecan, gemcitabine/
[106–108]. Radiation therapy is indicated for pri- docetaxel, temozolomide, or aerosol rubitecan
mary local control if the tumor is deemed unre- [84, 134]. Anti-insulin-like growth factor-1
sectable, or can be used postoperatively in the receptor (IGF-1R) antibodies have also been used
setting of marginal resection [14, 100, 109–113]. to treat patients with relapsed EFT and have dem-
Recommended doses range from 55 Gray (Gy) onstrated short-term, but not durable, benefit
for most primary sites of disease to 45 Gy for [135–137]. Late recurrence up to 10 years after
microscopic disease [76, 107, 114]. Complications treatment has been observed for EFT, making
of radiation therapy include radiation-induced surveillance crucial [138, 139]. Surveillance
sarcomas, skin fibrosis, edema, wound healing guidelines are offered by the National
problems, pathologic fracture, exacerbation of Comprehensive Cancer Network (NCCN) and
chemotherapy-induced myelosuppression, and include 3-month interval visits for the first 2 years
growth deformity or arrest [110, 115–119]. The followed by 4–6-month interval visits for the next
use of proton beam therapy rather than conven- 3 years.
tional photon therapy has recently been advo-
cated as a way to reduce these side effects,
especially in the pelvis where toxicity to the spi- 8.14 Case Presentations
nal canal, intestine, rectum, bladder, and femoral
head must be mitigated [74, 100, 120–122]. 8.14.1 Case 1
There is concern that proton beam therapy may
be associated with neutron scatter radiation and a A 19-year-old male initially sought care for left
risk for secondary malignancies, warranting hip and leg pain for 2 months. Pain was atrau-
future investigation [119]. Intensity-modulated matic in onset and had been gradually worsening
radiation therapy (IMRT) [123–125] and three- since it began. Pain was worse with weight-
dimensional conformal radiation therapy bearing, had failed to respond to anti-
(3CD-CRT) [124] are newer strategies that are inflammatory medications, and was accompanied
utilized to improve accuracy, spare normal tissue, by subjective weakness and loss of function. The
and reduce complications [126]. Data guiding the patient also reported waking up multiple times a
role of adjuvant radiation therapy in the setting of night with pain. His past medical and surgical
surgery for patients with inadequate margins is history were unremarkable. His family history
conflicting [127]. The Euro-E.W.I.N.G. group was positive for breast cancer in his maternal
reported on 599 patients and found a reduction grandmother. He presented to the emergency
in local recurrence in patients treated with adju- department, where exam was notable only for
vant radiation therapy [128], while the pain with left hip range of motion and mild swell-
Cooperative Ewing Sarcoma Study (CESS) trials ing about the left gluteal region.
[37, 114, 115, 129] demonstrated no difference. Initial X-ray imaging demonstrated a subtle
lytic lesion within the left iliac wing (Fig. 8.6)
that prompted further imaging and laboratory
8.13 Surveillance and Recurrence workup. Labs were notable for a moderately ele-
vated c-reactive protein and pro-calcitonin but
Local recurrence is a poor prognostic sign and were otherwise normal. CT (Fig. 8.7a, b) and
can occur with or without metastases [12, 130, MRI (Fig. 8.8a, b) were notable for a heteroge-
131]. Local recurrences can be treated with an neous mass arising from the left iliac bone with a
large associated soft tissue component. He under-

went CT-guided biopsy of the lesion, with histol-
ogy demonstrating sheets of small, round blue
cells consistent with Ewing’s sarcoma (Fig. 8.9a).
These cells stained positive for both CD99 (mem-
branous staining) and NKX2.2 (nuclear staining)
(Fig. 8.9b). Fluorescence in situ hybridization
(FISH) analysis was positive for the characteris-
tic EWSR1 rearrangement.
The patient underwent staging with a positron
emission tomography (PET) scan, on which the
large iliac wing mass demonstrated intense meta-
bolic activity (Fig. 8.10). In addition, the PET
scan revealed diffuse osseous metastatic disease
to the right humerus, spine, pelvis, and left proxi-
mal femur, as well as possible metastatic pelvic
lymphadenopathy. Given the findings of widely
Fig. 8.6 Initial antero-posterior (AP) X-ray of the pelvis
in a patient presenting with left hip and leg pain demon- metastatic Ewing’s sarcoma, he was started on
strates a subtle lytic lesion within the left iliac wing chemotherapy consisting of vincristine, doxoru-
a b
Fig. 8.7 Coronal (a) and axial (b) CT scan demonstrates irregularity of the left iliac wing with associated periosteal
reaction (“onion skin” appearance) and a large soft tissue component
a b
Fig. 8.8 Coronal (a) and axial (b) T2-weighted MRI shows a large mass arising from the left iliac bone, with a bi-
cortical soft tissue component extending into the iliacus and gluteal muscles
a b
Fig. 8.9 Histology of a patient with Ewing’s sarcoma that demonstrates (a) sheets of small, round blue cells on H&E
stain and (b) positive nuclear staining for NKX2.2
bicin, and cyclophosphamide. Surgical planning

was postponed until response to chemotherapy
was able to be assessed.
8.14.2 Case 2
A 2-year-old previously healthy female presented

with 5 days of limping. Her parents denied any
history of witnessed trauma. They noted she has
been progressively inverting her right foot and
elevating her right hip while walking. The limp
had not improved with ibuprofen. She had no
complaints of pain, and no history of recent ill-
ness, fevers, or chills. Upon presentation, she
exhibited a right-sided antalgic gait, but had full
painless right hip range of motion, no focal ten-
derness, and an otherwise normal exam.
Laboratory analysis including complete blood
count, electrolytes, and inflammatory markers
(erythrocyte sedimentation rate and c-reactive
protein) was normal.
Initial X-ray imaging demonstrated a poorly
defined lesion within the right ilium, superior
to the acetabular dome, with a wide zone of
transition and no periosteal reaction (Fig. 8.11).
Given these findings, she underwent CT scan
(Fig. 8.12a, b) and MRI (Fig. 8.13a, b), which
Fig. 8.10 Positron emission tomography (PET) scan of a
revealed a large, permeative lesion in the right
patient with metastatic Ewing’s sarcoma demonstrates a
large left pelvic lesion with high metabolic activity, as well iliac wing with a surrounding superficial and
as metastases to the right proximal humerus and spine deep soft tissue component. CT-guided biopsy
of the mass was performed. Histology demon-

strated clusters and sheets of small, round blue
cells with high nuclear-to-cytoplasmic ratios,
round nuclei, and irregular borders. These cells
stained positive for CD99 in a diffuse, mem-
branous pattern. FISH analysis was positive for
the ESWR1 rearrangement, solidifying a sus-
pected diagnosis of Ewing’s sarcoma of the
pelvis.
She underwent staging, including a CT scan
of her chest, abdomen, and pelvis, which was
negative for metastasis, and PET scan, which
revealed diffusely increased uptake in the right
iliac wing, but no evidence of hypermetabolic
Fig. 8.11 Antero-posterior X-ray of the pelvis in a
lymphadenopathy or metastatic disease.
2-year-old female presenting with a limp demonstrates a
poorly defined, permeative lesion within the right ilium, The treatment strategy for this patient con-
superior to the acetabular dome, with a wide zone of tran- sisted of neoadjuvant chemotherapy, surgical
sition and internal calcifications resection, and consolidation chemotherapy. She
a b
Fig. 8.12 Axial (a) and coronal (b) CT scan shows a permeative lytic lesion of the right iliac wing with internal dys-
trophic calcifications and faint periosteal reaction
a b
Fig. 8.13 Axial T2-weighted (a) and coronal STIR (b) MRI reveals a mass of the right iliac wing extending to the
anterior sacroiliac joint and through the acetabular dome, with a large superficial and deep soft tissue component
was treated with 6 cycles of vincristine, doxoru- and sacral bones: the University of Florida experi-
ence. Int J Radiat Oncol Biol Phys. 2008;72(1):41–8.
bicin, and cyclophosphamide with excellent 15. Mankin HJ, et al. Malignant tumors of the pel-
response and reduction of her primary tumor vis: an outcome study. Clin Orthop Relat Res.
size. She then underwent right internal hemipel- 2004;425:212–7.
vectomy of zones 1 and 2, followed by an addi- 16. Rodl RW, et al. Ewing’s sarcoma of the pelvis: com-
bined surgery and radiotherapy treatment. J Surg
tional 11 rounds of chemotherapy. Surgical Oncol. 2003;83(3):154–60.
pathology demonstrated Ewing’s sarcoma with 17. Sucato DJ, et al. Ewing’s sarcoma of the pelvis.
>99% tumor necrosis and negative margins. Long-term survival and functional outcome. Clin
Orthop Relat Res. 2000;373:193–201.
18. Taylor C, et al. Diagnosis of Ewing’s sarcoma and
peripheral neuroectodermal tumour based on the
References detection of t(11;22) using fluorescence in situ
hybridisation. Br J Cancer. 1993;67(1):128–33.
1. Gatta G, et al. Childhood cancer survival trends in 19. Terrier P, Llombart-Bosch A, Contesso G. Small
Europe: a EUROCARE Working Group study. J Clin round blue cell tumors in bone: prognostic factors
Oncol. 2005;23(16):3742–51. correlated to Ewing’s sarcoma and neuroectodermal
2. Wilkins RM, et al. Ewing’s sarcoma of tumors. Semin Diagn Pathol. 1996;13(3):250–7.
bone. Experience with 140 patients. Cancer. 20. Tirode F, et al. Mesenchymal stem cell features of
1986;58(11):2551–5. Ewing tumors. Cancer Cell. 2007;11(5):421–9.
3. Jaffe R, et al. The neuroectodermal tumor of bone. 21. Fellinger EJ, et al. Immunohistochemical analysis of
Am J Surg Pathol. 1984;8(12):885–98. Ewing’s sarcoma cell surface antigen p30/32MIC2.
4. Askin FB, et al. Malignant small cell tumor of the Am J Pathol. 1991;139(2):317–25.
thoracopulmonary region in childhood: a distinctive 22. Ambros IM, et al. MIC2 is a specific marker for
clinicopathologic entity of uncertain histogenesis. Ewing’s sarcoma and peripheral primitive neuroecto-
Cancer. 1979;43(6):2438–51. dermal tumors. Evidence for a common histogenesis
5. Lee CS, et al. Primary cutaneous Ewing’s sar- of Ewing’s sarcoma and peripheral primitive neuroec-
coma/peripheral primitive neuroectodermal tumors todermal tumors from MIC2 expression and specific
in childhood. A molecular, cytogenetic, and chromosome aberration. Cancer. 1991;67(7):1886–93.
immunohistochemical study. Diagn Mol Pathol. 23. Crompton BD, et al. The genomic landscape
1995;4(3):174–81. of pediatric Ewing sarcoma. Cancer Discov.
6. Pradhan A, et al. Oncological outcomes of patients 2014;4(11):1326–41.
with Ewing’s sarcoma: is there a difference between 24. Dagher R, et al. Molecular confirmation of Ewing sar-
skeletal and extra-skeletal Ewing’s sarcoma? J Bone coma. J Pediatr Hematol Oncol. 2001;23(4):221–4.
Joint Surg Br. 2011;93(4):531–6. 25. Diffin F, et al. Rapid and specific diagnosis of
7. Raney RB, et al. Ewing’s sarcoma of soft tis- t(11;22) translocation in paediatric Ewing’s sarcoma
sues in childhood: a report from the Intergroup and primitive neuroectodermal tumours using RNA-
Rhabdomyosarcoma Study, 1972 to 1991. J Clin PCR. J Clin Pathol. 1994;47(6):562–4.
Oncol. 1997;15(2):574–82. 26. Downing JR, et al. Detection of the (11;22)
8. Rud NP, et al. Extraosseous Ewing’s sarcoma. A (q24;q12) translocation of Ewing’s sarcoma and
study of 42 cases. Cancer. 1989;64(7):1548–53. peripheral neuroectodermal tumor by reverse tran-
9. Grier HE. The Ewing family of tumors. Ewing’s sar- scription polymerase chain reaction. Am J Pathol.
coma and primitive neuroectodermal tumors. Pediatr 1993;143(5):1294–300.
Clin N Am. 1997;44(4):991–1004. 27. Lin PP, et al. Differential transactivation by alterna-
10. Jedlicka P. Ewing sarcoma, an enigmatic malignancy tive EWS-FLI1 fusion proteins correlates with clini-
of likely progenitor cell origin, driven by transcrip- cal heterogeneity in Ewing’s sarcoma. Cancer Res.
tion factor oncogenic fusions. Int J Clin Exp Pathol. 1999;59(7):1428–32.
2010;3(4):338–47. 28. Ohno T, Rao VN, Reddy ES. EWS/Fli-1 chimeric
11. Gaspar N, et al. Ewing sarcoma: current manage- protein is a transcriptional activator. Cancer Res.
ment and future approaches through collaboration. J 1993;53(24):5859–63.
Clin Oncol. 2015;33(27):3036–46. 29. Zoubek A, et al. Does expression of different EWS
12. Ahmed SK, et al. Pelvis Ewing sarcoma: local con- chimeric transcripts define clinically distinct risk
trol and survival in the modern era. Pediatr Blood groups of Ewing tumor patients? J Clin Oncol.
Cancer. 2017;64(9). 1996;14(4):1245–51.
13. Burgers JM, et al. Ewing’s sarcoma of the pelvis: 30. Denny CT. Gene rearrangements in Ewing’s sar-
changes over 25 years in treatment and results. Eur J coma. Cancer Investig. 1996;14(1):83–8.
Cancer. 1997;33(14):2360–7. 31. de Alava E, Gerald WL. Molecular biology of the
14. Indelicato DJ, et al. Impact of local management on Ewing’s sarcoma/primitive neuroectodermal tumor
long-term outcomes in Ewing tumors of the pelvis family. J Clin Oncol. 2000;18(1):204–13.
32. Delattre O, et al. The Ewing family of tumors— Ewing sarcoma family of tumors. J Clin Oncol.
a subgroup of small-round-cell tumors defined 2005;23(34):8828–34.
by specific chimeric transcripts. N Engl J Med. 49. Panicek DM, et al. CT and MR imaging in the
1994;331(5):294–9. local staging of primary malignant musculoskel-
33. Yoshida A, et al. NKX2.2 is a useful immunohis- etal neoplasms: report of the Radiology Diagnostic
tochemical marker for Ewing sarcoma. Am J Surg Oncology Group. Radiology. 1997;202(1):237–46.
Pathol. 2012;36(7):993–9. 50. Volker T, et al. Positron emission tomography
34. Hattinger CM, et al. Prognostic impact of deletions for staging of pediatric sarcoma patients: results
at 1p36 and numerical aberrations in Ewing tumors. of a prospective multicenter trial. J Clin Oncol.
Genes Chromosomes Cancer. 1999;24(3):243–54. 2007;25(34):5435–41.
35. Sneppen O, Hansen LM. Presenting symptoms and 51. Glaubiger DL, et al. Determination of prognos-
treatment delay in osteosarcoma and Ewing’s sar- tic factors and their influence on therapeutic
coma. Acta Radiol Oncol. 1984;23(2–3):159–62. results in patients with Ewing’s sarcoma. Cancer.
36. Paulussen M, et al. Primary metastatic (stage IV) 1980;45(8):2213–9.
Ewing tumor: survival analysis of 171 patients 52. Fagnou C, et al. Presence of tumor cells in bone
from the EICESS studies. European Intergroup marrow but not in blood is associated with adverse
Cooperative Ewing Sarcoma Studies. Ann Oncol. prognosis in patients with Ewing’s tumor. Societe
1998;9(3):275–81. Francaise d’Oncologie Pediatrique. J Clin Oncol.
37. Schuck A, et al. Local therapy in localized Ewing 1998;16(5):1707–11.
tumors: results of 1058 patients treated in the CESS 53. Bacci G, et al. Prognostic significance of serum lac-
81, CESS 86, and EICESS 92 trials. Int J Radiat tic acid dehydrogenase in Ewing’s tumor of bone.
Oncol Biol Phys. 2003;55(1):168–77. Ric Clin Lab. 1985;15(1):89–96.
38. Cotterill SJ, et al. Prognostic factors in Ewing’s 54. West DC, et al. Detection of circulating tumor cells
tumor of bone: analysis of 975 patients from the in patients with Ewing’s sarcoma and peripheral
European Intergroup Cooperative Ewing’s Sarcoma primitive neuroectodermal tumor. J Clin Oncol.
Study Group. J Clin Oncol. 2000;18(17):3108–14. 1997;15(2):583–8.
39. Natarajan MV, et al. Surgical management 55. Cangir A, et al. Ewing’s sarcoma metastatic at diag-
of pelvic Ewing’s sarcoma. Indian J Orthop. nosis. Results and comparisons of two intergroup
2010;44(4):397–401. Ewing’s sarcoma studies. Cancer. 1990;66(5):887–93.
40. Jawad MU, et al. Ewing sarcoma demonstrates 56. Rodriguez-Galindo C, et al. Analysis of prognostic
racial disparities in incidence-related and sex- factors in Ewing sarcoma family of tumors: review
related differences in outcome: an analysis of 1631 of St. Jude Children’s Research Hospital studies.
cases from the SEER database, 1973–2005. Cancer. Cancer. 2007;110(2):375–84.
2009;115(15):3526–36. 57. Hayes FA, et al. Metastatic Ewing’s sarcoma:
41. Widhe B, Widhe T. Initial symptoms and clinical remission induction and survival. J Clin Oncol.
features in osteosarcoma and Ewing sarcoma. J 1987;5(8):1199–204.
Bone Joint Surg Am. 2000;82(5):667–74. 58. Parasuraman S, et al. Brain metastases in pediatric
42. Baldini EH, et al. Adults with Ewing’s sarcoma/ Ewing sarcoma and rhabdomyosarcoma: the St. Jude
primitive neuroectodermal tumor: adverse effect of Children’s Research Hospital experience. J Pediatr
older age and primary extraosseous disease on out- Hematol Oncol. 1999;21(5):370–7.
come. Ann Surg. 1999;230(1):79–86. 59. Rosito P, et al. Italian Cooperative Study for the
43. Meyer JS, et al. Imaging guidelines for children with treatment of children and young adults with local-
Ewing sarcoma and osteosarcoma: a report from the ized Ewing sarcoma of bone: a preliminary report of
Children’s Oncology Group Bone Tumor Committee. 6 years of experience. Cancer. 1999;86(3):421–8.
Pediatr Blood Cancer. 2008;51(2):163–70. 60. Sandoval C, et al. Outcome in 43 children present-
44. Mendenhall CM, et al. The prognostic significance ing with metastatic Ewing sarcoma: the St. Jude
of soft tissue extension in Ewing’s sarcoma. Cancer. Children’s Research Hospital experience, 1962 to
1983;51(5):913–7. 1992. Med Pediatr Oncol. 1996;26(3):180–5.
45. Franzius C, et al. FDG-PET for detection of osseous 61. Ferrari S, et al. Ewing’s sarcoma of bone: relation
metastases from malignant primary bone tumours: between clinical characteristics and staging. Oncol
comparison with bone scintigraphy. Eur J Nucl Med. Rep. 2001;8(3):553–6.
2000;27(9):1305–11. 62. Ferrari S, et al. Ifosfamide and actinomycin-D, added
46. Gyorke T, et al. Impact of FDG PET for staging in the induction phase to vincristine, cyclophospha-
of Ewing sarcomas and primitive neuroectodermal mide and doxorubicin, improve histologic response
tumours. Nucl Med Commun. 2006;27(1):17–24. and prognosis in patients with non metastatic
47. Kleis M, et al. Diagnostic value of PET/CT for the Ewing’s sarcoma of the extremity. J Chemother.
staging and restaging of pediatric tumors. Eur J Nucl 1998;10(6):484–91.
Med Mol Imaging. 2009;36(1):23–36. 63. Whelan JS, et al. A systematic review of the role of
48. Hawkins DS, et al. [18F]Fluorodeoxyglucose posi- pulmonary irradiation in the management of primary
tron emission tomography predicts outcome for bone tumours. Ann Oncol. 2002;13(1):23–30.
64. Biermann JS, et al. NCCN guidelines insights: bone 81. Pilepich MV, et al. Radiotherapy and combina-
Cancer, Version 2.2017. J Natl Compr Cancer Netw. tion chemotherapy in advanced Ewing’s Sarcoma-
2017;15(2):155–67. Intergroup study. Cancer. 1981;47(8):1930–6.
65. Bosma SE, et al. Prognostic factors for survival in 82. Bacci G, et al. Neoadjuvant chemotherapy for
Ewing sarcoma: a systematic review. Surg Oncol. peripheral malignant neuroectodermal tumor of
2018;27(4):603–10. bone: recent experience at the Istituto Rizzoli. J Clin
66. Parham DM, et al. Neuroectodermal differentiation Oncol. 2000;18(4):885–92.
in Ewing’s sarcoma family of tumors does not pre- 83. Bacci G, et al. Multimodal therapy for the treatment
dict tumor behavior. Hum Pathol. 1999;30(8):911–8. of nonmetastatic Ewing sarcoma of pelvis. J Pediatr
67. Wunder JS, et al. The histological response to che- Hematol Oncol. 2003;25(2):118–24.
motherapy as a predictor of the oncological outcome 84. Kolb EA, et al. Long-term event-free survival after
of operative treatment of Ewing sarcoma. J Bone intensive chemotherapy for Ewing’s family of
Joint Surg Am. 1998;80(7):1020–33. tumors in children and young adults. J Clin Oncol.
68. Ozaki T, et al. Significance of surgical margin on 2003;21(18):3423–30.
the prognosis of patients with Ewing’s sarcoma. 85. Nesbit ME Jr, et al. Multimodal therapy for the
A report from the Cooperative Ewing’s Sarcoma management of primary, nonmetastatic Ewing’s
Study. Cancer. 1996;78(4):892–900. sarcoma of bone: a long-term follow-up of the First
69. Hau A, et al. Accuracy of CT-guided biopsies in 359 Intergroup study. J Clin Oncol. 1990;8(10):1664–74.
patients with musculoskeletal lesions. Skelet Radiol. 86. Grier HE, et al. Addition of ifosfamide and etoposide
2002;31(6):349–53. to standard chemotherapy for Ewing’s sarcoma and
70. Marcove RC, Rosen G. Radical en bloc exci- primitive neuroectodermal tumor of bone. N Engl J
sion of Ewing’s sarcoma. Clin Orthop Relat Res. Med. 2003;348(8):694–701.
1980;153:86–91. 87. Juergens C, et al. Safety assessment of intensive
71. Nilbert M, et al. Ewing’s sarcoma treatment in induction with vincristine, ifosfamide, doxorubicin,
Scandinavia 1984–1990—ten-year results of the and etoposide (VIDE) in the treatment of Ewing
Scandinavian Sarcoma Group Protocol SSGIV. Acta tumors in the EURO-E.W.I.N.G. 99 clinical trial.
Oncol. 1998;37(4):375–8. Pediatr Blood Cancer. 2006;47(1):22–9.
72. Pinkerton CR, et al. Treatment strategies for 88. Miser JS, et al. Treatment of metastatic Ewing’s
metastatic Ewing’s sarcoma. Eur J Cancer. sarcoma or primitive neuroectodermal tumor
2001;37(11):1338–44. of bone: evaluation of combination ifosfamide
73. Yock TI, et al. Local control in pelvic Ewing sar- and etoposide—a Children’s Cancer Group and
coma: analysis from INT-0091—a report from Pediatric Oncology Group study. J Clin Oncol.
the Children’s Oncology Group. J Clin Oncol. 2004;22(14):2873–6.
2006;24(24):3838–43. 89. Pretz JL, et al. Localized adult Ewing sarcoma:
74. Krasin MJ, et al. Definitive irradiation in multidis- favorable outcomes with alternating vincristine,
ciplinary management of localized Ewing sarcoma doxorubicin, cyclophosphamide, and ifosfamide,
family of tumors in pediatric patients: outcome and etoposide (VDC/IE)-based multimodality therapy.
prognostic factors. Int J Radiat Oncol Biol Phys. Oncologist. 2017;22(10):1265–70.
2004;60(3):830–8. 90. Wagner MJ, et al. Vincristine, Ifosfamide, and doxo-
75. Krasin MJ, et al. Efficacy of combined surgery and rubicin for initial treatment of Ewing sarcoma in
irradiation for localized Ewings sarcoma family of adults. Oncologist. 2017;22(10):1271–7.
tumors. Pediatr Blood Cancer. 2004;43(3):229–36. 91. Wexler LH, et al. Ifosfamide and etoposide plus
76. La TH, et al. Radiation therapy for Ewing’s sar- vincristine, doxorubicin, and cyclophosphamide for
coma: results from Memorial Sloan-Kettering newly diagnosed Ewing’s sarcoma family of tumors.
in the modern era. Int J Radiat Oncol Biol Phys. Cancer. 1996;78(4):901–11.
2006;64(2):544–50. 92. Granowetter L, et al. Dose-intensified compared with
77. Obata H, et al. Clinical outcome of patients with standard chemotherapy for nonmetastatic Ewing
Ewing sarcoma family of tumors of bone in Japan: sarcoma family of tumors: a Children’s Oncology
the Japanese Musculoskeletal Oncology Group Group Study. J Clin Oncol. 2009;27(15):2536–41.
cooperative study. Cancer. 2007;109(4):767–75. 93. Gupta AA, et al. Clinical outcome of children and
78. Werier J, et al. A systematic review of optimal treat- adults with localized Ewing sarcoma: impact of che-
ment strategies for localized Ewing’s sarcoma of motherapy dose and timing of local therapy. Cancer.
bone after neo-adjuvant chemotherapy. Surg Oncol. 2010;116(13):3189–94.
2016;25(1):16–23. 94. Verrill MW, et al. The use of paediatric chemother-
79. Bernstein M, et al. Ewing’s sarcoma fam- apy protocols at full dose is both a rational and feasi-
ily of tumors: current management. Oncologist. ble treatment strategy in adults with Ewing’s family
2006;11(5):503–19. tumours. Ann Oncol. 1997;8(11):1099–105.
80. Picci P, et al. Chemotherapy-induced tumor necrosis 95. Womer RB, et al. Randomized controlled trial of
as a prognostic factor in localized Ewing’s sarcoma interval-compressed chemotherapy for the treat-
of the extremities. J Clin Oncol. 1997;15(4):1553–9. ment of localized Ewing sarcoma: a report from
the Children’s Oncology Group. J Clin Oncol. disease control, limb function, and treatment toxic-
2012;30(33):4148–54. ity. Int J Radiat Oncol Biol Phys. 2008;72(3):871–7.
96. Bacci G, et al. The role of surgical margins in treat- 111. Koontz BF, Clough RW, Halperin EC. Palliative
ment of Ewing’s sarcoma family tumors: experience radiation therapy for metastatic Ewing sarcoma.
of a single institution with 512 patients treated with Cancer. 2006;106(8):1790–3.
adjuvant and neoadjuvant chemotherapy. Int J Radiat 112. Ng VY, et al. The effect of surgery with radiation
Oncol Biol Phys. 2006;65(3):766–72. on pelvic Ewing sarcoma survival. J Surg Oncol.
97. DuBois SG, et al. Comparative evaluation of local 2015;112(8):861–5.
control strategies in localized Ewing sarcoma of 113. Talleur AC, et al. Limited margin radiation therapy
bone: a report from the Children’s Oncology Group. for children and young adults with Ewing sarcoma
Cancer. 2015;121(3):467–75. achieves high rates of local tumor control. Int J
98. Givens SS, et al. Non-metastatic Ewing’s sarcoma: Radiat Oncol Biol Phys. 2016;96(1):119–26.
twenty years of experience suggests that surgery is 114. Dunst J, et al. Radiation therapy as local treatment
a prime factor for successful multimodality therapy. in Ewing’s sarcoma. Results of the Cooperative
Int J Oncol. 1999;14(6):1039–43. Ewing’s Sarcoma Studies CESS 81 and CESS 86.
99. Bhatia S, et al. Therapy-related myelodysplasia Cancer. 1991;67(11):2818–25.
and acute myeloid leukemia after Ewing sarcoma 115. Dunst J, et al. Second malignancies after treatment
and primitive neuroectodermal tumor of bone: a for Ewing’s sarcoma: a report of the CESS-studies.
report from the Children’s Oncology Group. Blood. Int J Radiat Oncol Biol Phys. 1998;42(2):379–84.
2007;109(1):46–51. 116. Fuchs B, et al. Complications in long-term survivors
100. Evans R, et al. Local recurrence, rate and sites of of Ewing sarcoma. Cancer. 2003;98(12):2687–92.
metastases, and time to relapse as a function of treat- 117. McLean TW, et al. Late events in pediatric patients
ment regimen, size of primary and surgical history in with Ewing sarcoma/primitive neuroectodermal
62 patients presenting with non-metastatic Ewing’s tumor of bone: the Dana-Farber Cancer Institute/
sarcoma of the pelvic bones. Int J Radiat Oncol Biol Children’s Hospital experience. J Pediatr Hematol
Phys. 1985;11(1):129–36. Oncol. 1999;21(6):486–93.
101. Hesla AC, et al. Improved prognosis for patients 118. Navid F, et al. Second cancers in patients with the
with Ewing sarcoma in the sacrum compared Ewing sarcoma family of tumours. Eur J Cancer.
with the innominate bones: the Scandinavian 2008;44(7):983–91.
Sarcoma Group Experience. J Bone Joint Surg Am. 119. Paulussen M, et al. Second malignancies after
2016;98(3):199–210. Ewing tumor treatment in 690 patients from a coop-
102. Hoffmann C, et al. Pelvic Ewing sarcoma: a erative German/Austrian/Dutch study. Ann Oncol.
retrospective analysis of 241 cases. Cancer. 2001;12(11):1619–30.
1999;85(4):869–77. 120. Evans RG, et al. Multimodal therapy for the man-
103. Puri A, et al. Results of surgical resection in pelvic agement of localized Ewing’s sarcoma of pelvic and
Ewing’s sarcoma. J Surg Oncol. 2012;106(4):417–22. sacral bones: a report from the second intergroup
104. Mankin HJ, Hornicek FJ. Internal hemipelvectomy study. J Clin Oncol. 1991;9(7):1173–80.
for the management of pelvic sarcomas. Surg Oncol 121. Ning MS, et al. Role of radiation in the treatment
Clin N Am. 2005;14(2):381–96. of non-metastatic osseous Ewing sarcoma. J Med
105. Raciborska A, et al. Internal hemipelvectomy in the Imaging Radiat Oncol. 2016;60(1):119–28.
management of pelvic Ewing sarcoma - are out- 122. Rombi B, et al. Proton radiotherapy for pediatric
comes better than with radiation therapy? J Pediatr Ewing’s sarcoma: initial clinical outcomes. Int J
Surg. 2014;49(10):1500–4. Radiat Oncol Biol Phys. 2012;82(3):1142–8.
106. Barbieri E, et al. Combined therapy of local- 123. Hall EJ. Intensity-modulated radiation therapy, pro-
ized Ewing’s sarcoma of bone: analysis of results tons, and the risk of second cancers. Int J Radiat
in 100 patients. Int J Radiat Oncol Biol Phys. Oncol Biol Phys. 2006;65(1):1–7.
1990;19(5):1165–70. 124. Mounessi FS, et al. Pelvic Ewing sarcomas.
107. Donaldson SS, et al. A multidisciplinary study Three-dimensional conformal vs. intensity-
investigating radiotherapy in Ewing’s sarcoma: end modulated radiotherapy. Strahlenther Onkol.
results of POG #8346. Pediatric Oncology Group. 2013;189(4):308–14.
Int J Radiat Oncol Biol Phys. 1998;42(1):125–35. 125. Stea B, et al. Treatment of pelvic sarcomas in ado-
108. Paulino AC, et al. Dose response and local control lescents and young adults with intensive combined
using radiotherapy in non-metastatic Ewing sar- modality therapy. Int J Radiat Oncol Biol Phys.
coma. Pediatr Blood Cancer. 2007;49(2):145–8. 1987;13(12):1797–805.
109. Arai Y, et al. Ewing’s sarcoma: local tumor control 126. Polf JC, Newhauser WD, Titt U. Patient neutron
and patterns of failure following limited-volume dose equivalent exposures outside of the pro-
radiation therapy. Int J Radiat Oncol Biol Phys. ton therapy treatment field. Radiat Prot Dosim.
1991;21(6):1501–8. 2005;115(1–4):154–8.
110. Indelicato DJ, et al. Definitive radiotherapy for 127. Merchant TE, et al. Effect of low-dose radia-
Ewing tumors of extremities and pelvis: long-term tion therapy when combined with surgical resec-
tion for Ewing sarcoma. Med Pediatr Oncol. tor antibody, in patients with metastatic Ewing fam-
1999;33(2):65–70. ily tumors or desmoplastic small round cell tumors.
128. Foulon S, et al. Can postoperative radiotherapy be J Clin Oncol. 2012;30(15):1849–56.
omitted in localised standard-risk Ewing sarcoma? 135. Juergens H, et al. Preliminary efficacy of the anti-
An observational study of the Euro-E.W.I.N.G insulin-like growth factor type 1 receptor antibody
group. Eur J Cancer. 2016;61:128–36. figitumumab in patients with refractory Ewing sar-
129. Paulussen M, et al. Localized Ewing tumor of bone: coma. J Clin Oncol. 2011;29(34):4534–40.
final results of the cooperative Ewing’s Sarcoma 136. Naing A, et al. Insulin growth factor-receptor
Study CESS 86. J Clin Oncol. 2001;19(6):1818–29. (IGF-1R) antibody cixutumumab combined with
130. Rodriguez-Galindo C, et al. Survival after recur- the mTOR inhibitor temsirolimus in patients with
rence of Ewing tumors: the St Jude Children’s refractory Ewing’s sarcoma family tumors. Clin
Research Hospital experience, 1979–1999. Cancer. Cancer Res. 2012;18(9):2625–31.
2002;94(2):561–9. 137. Pappo AS, et al. R1507, a monoclonal antibody to
131. Stahl M, et al. Risk of recurrence and survival after the insulin-like growth factor 1 receptor, in patients
relapse in patients with Ewing sarcoma. Pediatr with recurrent or refractory Ewing sarcoma family
Blood Cancer. 2011;57(4):549–53. of tumors: results of a phase II Sarcoma Alliance for
132. Jurgens H, et al. Multidisciplinary treatment of Research through Collaboration study. J Clin Oncol.
primary Ewing’s sarcoma of bone. A 6-year expe- 2011;29(34):4541–7.
rience of a European Cooperative Trial. Cancer. 138. van Doorninck JA, et al. Current treatment pro-
1988;61(1):23–32. tocols have eliminated the prognostic advantage
133. Oberlin O, et al. Prognostic factors in localized of type 1 fusions in Ewing sarcoma: a report from
Ewing’s tumours and peripheral neuroectodermal the Children’s Oncology Group. J Clin Oncol.
tumours: the third study of the French Society of 2010;28(12):1989–94.
Paediatric Oncology (EW88 study). Br J Cancer. 139. Wasilewski-Masker K, et al. Late recurrence in
2001;85(11):1646–54. pediatric cancer: a report from the Childhood
134. Tap WD, et al. Phase II study of ganitumab, a fully Cancer Survivor Study. J Natl Cancer Inst.
human anti-type-1 insulin-like growth factor recep- 2009;101(24):1709–20.
Pelvic Metastases: Diagnosis
and Treatment 9
Eduardo J. Ortiz-Cruz, Manuel Peleteiro-Pensado,
Irene Barrientos-Ruiz, and Raquel Ortiz-Hernando
9.1 Introduction that more of the pelvic surgery indications are

made [1].
Metastatic bone disease (MBD) to the pelvis is a Metastatic involvement of the pelvis is com-
challenging problem that affects the patient’s life mon, second only to axial involvement. Certainly,
and Quality of Life (QoL) and the orthopedic sur- 833 (18.8%) of all 4431 metastatic lesions docu-
geons increasingly face it. Due to the relatively mented in the archive of the Rizzoli Institute
large dimension of the pelvic cavity, tumors at were found to occur in the pelvic region: 12.6%
that location could reach considerable size before are located in the ilium, 1.8% in the ischium, and
symptoms appear, like a pain, pathologic frac- 1.2% in the pubis [2].
tures, and limit the ability to ambulate indepen-
dently [1].
The optimal treatment of bone metastasis may 9.2 Diagnosis
be complex and demand multimodality treatment
strategies to achieve optimal outcomes. Some Detection of bone metastases is essential for
locations of metastases within the pelvis have no accurate staging and optimal treatment. The
significant impact on pelvic stability and function objectives of imaging are to recognize sites of
(e.g., ilium and pubis), but tumors located on the metastasis and to consider involved locations in
posterior ilium may carry a risk to lumbosacral which complications are likely, such as patho-
integrity; tumors of the acetabulum may impair logic fracture. Imaging is also used to monitor
the hip function and the weight-bearing; due to the biopsy, if believed indispensable.
the high mechanical loads and this is the location There are four situations at diagnosis:
E. J. Ortiz-Cruz (*) 1. Patient with no previous history of cancer.

Orthopaedic Oncology Surgery Unit, Hospital The assessment of a patient with no previ-
Universitario La Paz, Madrid, Spain ous history of cancer has to be similar that the
Orthopaedic Oncology Surgery Unit, Hospital MD evaluation for solitary symptomatic bone
Anderson Cancer Center, Madrid, Spain lesion. National Comprehensive Cancer
M. Peleteiro-Pensado · I. Barrientos-Ruiz Network (NCCN) guidelines [3] suggest com-
Orthopaedic Oncology Surgery Unit, Hospital plete a workup for potential bone metastasis
Universitario La Paz, Madrid, Spain for symptomatic bone lesion associated with
R. Ortiz-Hernando abnormal radiograph in patients over 40 years
Facultad de Medicina, Universidad CEU San Pablo, of age.
Madrid, Spain

88 E. J. Ortiz-Cruz et al.
2. Patients with a past history of remission and prognoses. Surgery on the pelvic bone is a
cancer, it does not matter how long ago the challenging procedure both for the patient and
cancer happened. surgeon [7], and that is why a clear understanding
Most of these patients require biopsy con- of all potential varieties of treatment, beginning
firmation. If this event is the first evidence of from nonsurgical therapies to surgeries options
recurrence or the disease progression, the [1, 4, 8].
staging assessment must include a contrast-
enhanced CT of the chest, abdomen, and pel-
vis for visceral organ evaluation, a bone scan 9.3.1 Nonoperative Treatment
or FDG-PET/CT may be also indicated.
3. Patients with known history of cancer at The radiation therapy is effective in providing
the present time, with known metastases. palliation and relief of painful bone metastasis,
These types of patients who were found to with a global pain response rate as 60%.
have other visceral metastasis on restaging Therefore, the external irradiation is the standard
evaluation, presumptive diagnosis with imag- nonoperative care for patients with localized
ing modalities may be sufficient for a reason- bone pain and palliation [9].
able diagnosis of bone metastasis and the Lesions of the hemipelvis not directly involv-
biopsy might be not indicated. ing the hip joint as avulsion fractures of the ante-
4. Patients with a history of cancer at the pres- rior superior/inferior iliac spines, iliac crest, and
ent time and without known metastases. superior/inferior pubic rami can generally be
Generally, the diagnostic staging evalua- treated with external beam radiation and modifi-
tion should precede a biopsy of suspicious cation of weight-bearing.
bone lesions, particularly if a pathologic or
impending pathologic fracture is present or
suspected. 9.3.2 M
inimally Invasive Palliative
Procedures
9.3 Treatment Planning Patients who have intractable pain at a site of pre-
viously irradiated and may not be eligible for
There are mostly three types of treatment: nonop- supplementary radiation and neither for surgery,
erative, minimally invasive procedures, and sur- the interventional radiology adds to the arma-
gery. The selected procedure should offer an mentarium an effective palliative treatment of
adequate treatment to the patient to achieve the their symptoms [10]. The most frequent tech-
best possible QoL, while avoiding under or over- niques are radiofrequency-ablation, microwave
treatment. The factors associated with poor QoL tumors ablation, cryoablation, and cemento-
include loss of limb function, bedridden, and plasty, which could be used in combination with
pathologic fracture. the previous techniques [11–14].
The treatment depends on the patient’s symp-
toms, prognosis, patient class [4], histologic type,
site of the metastasis, amount of bone loss, and 9.3.3 Surgical Treatment
functional performance such as the Eastern
Cooperative Oncology Group (ECOG) The decision to proceed with surgery could be
Performance Status scales [5] and family aims. difficult for the surgeon and patient too, because
The main indications for surgery include the risks of surgery may outweigh the proposed
patients with compromised skeletal stability, pain benefits of improved pain and function. It is
that not responding to other modalities of treat- indispensable to identify the next six key
ment or some solitary bone metastases [1, 4, 6], subjects.
9 Pelvic Metastases: Diagnosis and Treatment 89
9.3.3.1 Types of Pelvis Resection treated conservatively by chemotherapy, hor-

According to Enneking classification that is monotherapy, and/or radiation therapy.
based on the resected region of the pelvis. Type I
(ilium); type II (periacetabular region); zone III, 9.3.3.3 Zones At Risk for Mechanical
pubis and ischium. When resection of the poste- Failure
rior ilium with the sacral wing is classified as According to Muller and Capanna [16], type 2
either an extended type I or type IV resection parallels to the nonarticular part of long bones
[15]. When the MBD is located at periacetabular (humerus, femur, and tibia) and those are the
region (type II), most of the nonoperative treat- lesions with a high risk for mechanical failure
ments such as radiotherapy, chemotherapy, hor- due to progressive destruction of the hip joint
monal therapy, and bisphosphonates are and they have a surgery indication (Fig. 9.2a–d).
insufficient to reduce pain and restore ambulation ZONES 1 and 3 are comparable to non-weight-
and this is the zone that is more frequent to indi- bearing and expendable bones of the extremity
cate surgery (Fig. 9.1a, b). (clavicle, sternum, and fibula) and they do not
compromise the mechanical stability of the pel-
9.3.3.2 Patients’ Classes vic ring (Fig. 9.3a–d) and they do not need bone
It is critical to individualize the type of patient reconstruction after the resection.
who requires a surgical treatment. Capanna and
Campanacci [4] introduced a protocol in long 9.3.3.4 Metastatic Acetabular
bone metastases that provide an aim to look for a Classification
proper treatment and it is adapted for pelvic It is important to keep in mind the classification
metastases too. The patients are divided into four described by Harrington, in order to select the
classes: (1) solitary lesion with good prognosis; best type of surgical management of acetabu-
(2) pathologic fracture; (3) impending fracture; lum [17].
and (4) other lesions. Periacetabular reconstruction for class II and
All patients included in Classes 1, 2, and 3 class III lesions require restoration of structural
should have been referral to oncology orthopedic integrity of the medial wall, superior dome, and
surgeon for surgical treatment and have to be lateral wall to adjust the acetabular components.
evaluated by the medical oncologist and medical Harrington [17] described good results with no
radiotherapist for the assessment of neo or adju- loosening of implants by using a protrusio ring
vant treatments. Most of the class 4 patients are with or without mesh in class II and adding
a b
Fig. 9.1 (a) Anteroposterior (AP) radiograph of the pel- patient and destructive progression of the lytic lesion at
vis that shows an osteolytic lesion at left zone 2, in a zone 2, after nonoperative treatments, at this point is
70-year-old male patient with a lung cancer (class 3 changed a class 2 patient
patient according to Muller and Cappana [16]). (b) Same
a b
c d
Fig. 9.2 (a) AP radiograph of the pelvis that shows an graph of metastatic osteolytic lesion in femoral diaphysis,
osteolytic lesion zone 2, in 75-year-old female patient, with a high risk of pathologic fracture, that parallels to the
with metastatic thyroid cancer (class 3 patient). (b) Same lesion at zone 2 of pelvis, as displayed in a. (d) AP view
patient, with tumor progression and demonstrates a pre- of the same lesion of the c, currently with pathologic frac-
dictable femoral head protrusio (class 2) that probably had ture, that parallels to the lesion at zone 2 of pelvis with
to be restored it before happened. (c) Conventional radio- femoral head protrusion as b
threaded-Steinmann pins in class III lesions. In Modifications of the technique have been pro-
spite of the limited life expectancy of patients, posed. Using retrograde screws placed through
reconstruction with anti-protrusio cages, screws, the protrusio cage to transmit weight loads to
and cemented hip replacement is beneficial for structurally intact bone, which follows the prin-
improving their QoL. ciples outlined by Harrington [18, 19].
a b
c d
Fig. 9.3 (a) Osteolytic infiltrative lesion at zone 1 (supra- imal end of the fibula (expandable bone) in a 65-year-old
acetabular) in a 63-year-old female patient with solitary male with lung cancer, which parallels with a zone 1 or 3
bone metastasis of follicular thyroid carcinoma (class 3). of the pelvis, which are possible to resect them without
(b) Identical patient, after wide tumor resection and not reconstruction. (d) Radiograph after wide tumor resection
pelvic reconstruction, due that there is no compromise of without reconstruction of the proximal fibula; that paral-
the mechanical stability of the pelvic ring. (c) AP view of lels with wide tumoral resection of the zones 1 and 3 of
the tibia and fibula that shows an osteolytic lesion at prox- the pelvis, as shown in b
9.3.3.5 Patient Prognosis ing the clinical studies which reported pain relief,
The estimation of survival will assist to select the function outcomes, perioperative complications,
appropriate treatment indication. A systematic and mortality, they found an advantage for surgi-
review of the literature made by Wood and cal management; however, there is significant
coworkers [20] in patients with MBD to long risk of morbidity and mortality that had to be
bones and/or pelvis, treated surgically and revis- considered.
There are tools which estimate the likelihood Indications for Surgical Reconstruction of the
of survival after surgery for patients with skeletal Acetabulum
metastases, and some of them are located in the 1. Lesions that compromise the load transfer

web and easy to check (www.pathfx.org) [21], from the lower limb to spine are as follows:
and they help for surgical decision-making. (a) Periacetabular weight-bearing dome com-
promised by tumor.
9.3.3.6 Type of Surgery (b) Fracture of the medial or weight-bearing
Pelvic metastases are treated either with tumor dome of the acetabulum.
intralesional resection and not reconstruction, (c) Posterior Ilium lesions not involving the
intralesional resection, and reconstruction or acetabulum can be treated by intralesional
wide resection and reconstruction or not. resection and cement augmentation.
Respect which type of resection has to be indi- (d) Acetabular lesions that are contained

cated, there are few data available to compare the (with an intact medial wall) can be recon-
outcome of wide resection and intralesional structed by a cemented arthroplasty.
resection for pelvic metastases. The study of Protrusio acetabular cups compensate for
Pietro Ruggieri and coworkers [22] evaluated the deficiencies of the medial wall, while
role of intralesional or marginal resection and cement and pin fixation [18, 19] with
compared with wide resection; they did not find modified Harrington methods can be used
difference in survival to death between wide effectively to reconstruct large defects in
resection and intralesional resection even in the acetabular column and dome.
patients with solitary metastases. Nevertheless,
the wide margin and reconstruction could be Reconstruction of the Acetabulum: Options
appropriate for patient with solitary metastases, Metal cage like device, fixed in place with a com-
in order to attempt to increase the survival [16]. bination of screws and cement. Into this cage is
The Interdisciplinary Consensus on the placed a constrained hip cup or preferable double
Management of Bone Metastases from Renal mobility cup to prevent dislocation of the total hip
Cell Carcinoma [23] supports the wide resection replacement [24]. A long-stem femoral compo-
essentially if the metastases are from thyroid or nent is often used, not just to complete the total hip
renal carcinomas. Preoperative embolization of reconstruction, but also to prevent against patho-
these tumors is strongly suggested to reduce logic fracture of the femur in the case of disease
intraoperative blood loss. progression. Nevertheless, long cemented femoral
stems may lead to adverse events, for instance,
Periacetabular Defects hypotension or desaturation that are supposed to
Larger lesions can be treated by tumor intrale- be secondary to embolic phenomena, but this
sional resection (curettage) and cement pack- event is still in study [25, 26]. Cemented compo-
ing and if a solitary metastases lesion, as nents are generally chosen in the context of meta-
mentioned before, it could be indicated to get static bone disease since the associated use of
wide margin. radiation therapy will limit the degree of bone
Large lesions with impending or completed ingrowth with uncemented prostheses.
acetabular fractures may require surgery, with
the goal of constructing a durable hip joint to
offer pain relief and allow immediate weight- (a) Minor lytic lesions when the cortical bone is
bearing. This kind of reconstruction requires a undamaged, it is a good solution to perform
construct that effectively bypasses the acetabu- PMMA-augmentation with the use of verte-
lum, mainly with total hip arthroplasty with broplasty kits. Bone cement increases the
some combination of pelvic or acetabular hard- resistance of the acetabulum and allows full
ware and cement. loading of the affected limb (Fig. 9.4).

(b) In case the acetabulum cannot be recon- and reconstructions are massive and involve
structed, but significant iliac crest is available, significant blood loss and risk of complica-
stemmed acetabular implants (inverted ice tions, it is important to have determined pre-
cream cone prosthesis or pedestal cup) let operatively that the benefits outweigh the
anchorage of the acetabular shell into the pos- risks.
terior ilium with the stem [27] (Fig. 9.5a, b). (c) In cases where there are no further recon-
Alternatively, a customs acetabular or pelvic struction options available, a flail limb recon-
prosthesis may be used. As these resections struction by removing the femoral head and
debulking of the tumor can be performed;
resection arthroplasty may be considered in
patients thought to be too great a risk for
reconstructive surgery although this proce-
dure generally will preclude return to ambu-
lation in this fragile population and pain
relief is variable (Fig. 9.6a, b) [28].
(d) Hindquarter amputation is a last measure

reserved for cases of tumor fungating through
skin, non-suppressive deep infections, or
uncontrollable pain. Although, these proce-
dures have a high rate of complications, and
a death rate of about 50% within 12 months
[29], the patient’s life expectancy and the
Fig. 9.4 AP view of the pelvis with an osteolytic lesion improvement of the quality of life produced
of supra-acetabular region with percutaneous acetabulo-
plasty with PMMA
could justify the surgical risks.
a b
Fig. 9.5 (a) Bone metastases from breast carcinoma in a tumor board committee. (b) Intralesional resection zone II
female patient of 45 years old, with severe pain and capa- and wide resection of proximal femur. The reconstruction
ble of all self-care, but unable to carry out any work activi- was doing by inverted ice cream cone prostheses (Ice-
ties. The radiograph demonstrated destruction of the Cream Cone-style implant (Coned; Stanmore Worldwide
superior, lateral, and medial walls of the right acetabulum Ltd, Elmstree, UK)) with PMMA filling the gap of the
(zone 2) and impending femoral of the femoral neck. She acetabulum, dual mobility, no constrained cup, and
is ECOG 2 and class 3 patient; surgery was indicated after cemented proximal femur tumoral prostheses
a b
Fig. 9.6 (a, b) The X-rays shows an acetabular protru- Nonsurgical pretreatments have failed to relieve severe
sion of the 59-year-old female patient affected by bladder pain. Consequently, the surgical indication to dismiss the
Carcinoma and multiples bone and visceral metastases. pain was a resection arthroplasty after tumor board and
She has severe left hip pain. She has a pathologic fracture consent by the patient and relatives
involving a posterior column and pubic osteolisis.
3. Biermann JS, Chow W, Reed DR, Lucas D, Adkins

9.4 Conclusions DR, Agulnik M, Benjamin RS, Brigman B, Budd
GT, Curry WT, Didwania A, Fabbri N, Hornicek
FJ, Kuechle JB, Lindskog D, Mayerson J, McGarry
Pelvis metastases is a challenging problem that SV, Million L, Morris CD, Movva S, O’Donnell
affects the quality of life. Individual treatment RJ, Randall RL, Rose P, Santana VM, Satcher RL,
should be based on rational guidelines; however, Schwartz H, Siegel HJ, Thornton K, Villalobos
V, Bergman MA, Scavone JL. NCCN guidelines
the outcomes of these procedures depend on an insights: bone cancer, Version 2.2017. J Natl Compr
interdisciplinary decision on treatment by an Canc Netw. 2017;15(2):155–67.
entire multidisciplinary team. You need to know 4. Capanna R, Campanacci DA. The treatment of metas-
the prognoses, class of patient, and types of “sur- tases in the appendicular skeleton. J Bone Joint Surg
Br. 2001;83(4):471–81.
gical” local treatment. 5. Kelly CM, Shahrokni A. Moving beyond Karnofsky
The future of the treatment of metastatic bone and ECOG performance status assessments with
disease is stimulating. We hope with time, these new technologies. J Oncol. 2016; https://doi.
complex reconstructions will be addressed more org/10.1155/2016/6186543.
6. Hedge V, Bernthal NM. Metastatic bone disease: pel-
easily with more standard and predictable implant vis. In: Randall RL, editor. Metastatic bone disease:
constructs. an integrated approach to patient care. New York:
Springer; 2016. p. 267–77.
7. Krishnan CK, Han I, Kim HS. Outcome after surgery
for metastases to the pelvic bone: a single institutional
References experience. Clin Orthop Surg. 2017;9(1):116–25.
https://doi.org/10.4055/cios.2017.9.1.116.
1. Ortiz-Cruz EJ, Peleteiro-Pensado M, Barrientos- 8. Capana R, Rugieri P, Plana R, Campanacci D, Di Martino
Ruiz I, Carbonell-Escobar R. Metastases to pelvis. A, Daolio PA, Angelini A, Maccauro G. Linne Guida
In: Denaro V, Di Martino A, Piccioli A, editors. Siot Ed Algoritmo Di Trattamento Nelle Localizzacioni
Management of bone metastases. A multidisciplinary Metastachiche Del Bacino. Giornale Italiano Di
guide. Cham: Springer International Publishing AG, Ortopedia E Traumatologia. 2016;42:366–75.
part of Springer Nature; 2019. p. 159–66. ISBN 9. Ellsworth SG, Alcorn SR, Hales RK, McNutt TR,
978-3-319-73485-9. DeWeese TL, Smith TJ. Patterns of care among
2. Casadei R, Gambarroti M. Metastatic carcinoma. patients receiving radiation therapy for bone metas-
In: Picci P, Manfrini M, Fabbri N, Gambarotti M, tases at a large academic institution. Int J Radiat
Vanel D, editors. Atlas of musculoskeletal tumors and Oncol Biol Phys. 2014;89(5):1100–5. https://doi.
tumorlike lesions. Berlin: Springer; 2014. p. 251–5. org/10.1016/j.ijrobp.2014.04.028.
10. Rosenthal D, Callstrom MR. Critical review and

Gasbarrini A, Biagini R, Piana R, Fazioli F, Luzzati
state of the art in interventional oncology: benign A, Di Martino A, Nicolosi F, Camnasio F, Rosa MA,
and metastatic disease involving bone. Radiology. Campanacci DA, Denaro Capanna R. How do we
2012;262(3):765–80. estimate survival? External validation of a tool for
11. Lane MD, Le HB, Lee S, Young C, Heran MK, Badii survival estimation in patients with metastatic bone
M, Clarkson PW, Munk PL. Combination radiofre- disease-decision analysis and comparison of three
quency ablation and cementoplasty for palliative international patient populations. BMC Cancer.
treatment of painful neoplastic bone metastasis: expe- 2015;15:424.
rience with 53 treated lesions in 36 patients. Skelet 22. Ruggieri P, Mavrogenis AF, Angelini A, Mercuri

Radiol. 2011;40(1):25–32. M. Metastases of the pelvis: does resection improve
12. Pusceddu C, Sotgia B, Fele RM, Ballicu N, Melis survival? Orthopedics. 2011;34(7):e236–44.
L. Combined microwave ablation and cemento- 23. Grünwald V, Eberhardt B, Bex A, Flörcken A, Gauler
plasty in patients with painful bone metastases at T, Derlin T, Panzica M, Dürr HR, Grötz KA, Giles RH,
high risk of fracture. Cardiovasc Intervent Radiol. von Falck C, Graser A, Muacevic A, Staehler M. An
2016;39(1):74–80. interdisciplinary consensus on the management of
13. Durfee RA, Sabo SA, Letson GD, Binitie O, Cheong bone metastases from renal cell carcinoma. Nat Rev
D. Percutaneous acetabuloplasty for metastatic Urol. 2018;15(8):511–21. https://doi.org/10.1038/
lesions to the pelvis. Orthopedics. 2017;40(1):e170– s41585-018-0034-9. Review.
5. https://doi.org/10.3928/01477447-20161017-05. 24. Plummer D, Passen E, Alexander J, Vajapey S, Frantz
14. Castaneda Rodriguez WR, Callstrom MR. Effective T, Niedermeier S, Pettit R, Scharschmidt T. Rapid
pain palliation and prevention of fracture for axial- return to function and stability with dual mobility
loading skeletal metastases using combined cryoab- components cemented into an acetabular reconstruc-
lation and cementoplasty. Tech Vasc Interv Radiol. tive cage for large osseous defects in the setting of
2011;14(3):160–9. periacetabular metastatic disease. J Surg Oncol.
15. Enneking WF. The anatomic considerations in
2019;119(8):1155–60. https://doi.org/10.1002/
tumor surgery: pelvis. In: Enneking WF, editor. jso.25463.
Musculoskeletal tumor surgery, vol. 2. New York: 25. Xing Z, Moon BS, Satcher RL, Lin PP, Lewis VO. A
Churchill Livingstone; 1983. p. 483–529. long femoral stem is not always required in hip arthro-
16. Muller D, Capanna R. The surgical treatment of pel- plasty for patients with proximal femur metastases.
vic bone metastases. Adv Orthop. 2015;2015:525363. Clin Orthop Relat Res. 2013;471(5):1622–7. https://
17. Harrington KD. The management of acetabular insuf- doi.org/10.1007/s11999-013-2790-4.
ficiency secondary to metastatic malignant disease. J 26. Peterson JR, Decilveo AP, O’Connor IT, Golub

Bone Joint Surg Am. 1981;4:653–64. I, Wittig JC. What are the functional results and
18. Vielgut I, Sadoghi P, Gregori M, et al. The modified complications with long stem hemiarthroplasty in
Harrington procedure for metastatic peri-acetabular patients with metastases to the proximal femur? Clin
bone destruction. Int Orthop. 2013;37(10):1981–5. Orthop Relat Res. 2017;475(3):745–56. https://doi.
https://doi.org/10.1007/s00264-013-1940-3. org/10.1007/s11999-016-4810-7.
19. Lozano-Calderon SA, Kaiser CL, Osler PM, Raskin 27. Barrientos-Ruiz I, Ortiz-Cruz EJ, Peleteiro-Pensado
KA. Cemented total hip arthroplasty with retrograde M. Reconstruction after hemipelvectomy with the ice-
ischioacetabular Steinmann pin reconstruction for cream cone prosthesis: what are the short-term clinical
periacetabular metastatic carcinoma. J Arthroplast. results? Clin Orthop Relat Res. 2017;475(3):735–41.
2016;31(7):1555–60. https://doi.org/10.1016/j. https://doi.org/10.1007/s11999-016-4747-x.
arth.2016.01.014. 28. Quinn RH. Metastatic disease to the hip and pelvis: sur-
20. Wood TJ, Racano A, Yeung H, Farrokhyar F, Ghert M, gical management. Tech Orthop. 2007;22(2):78–87.
Deheshi BM. Surgical management of bone metasta- 29. Wunder JS, Ferguson PC, Griffin AM, Pressman A,
ses: quality of evidence and systematic review. Ann Bell RS. Acetabular metastases: planning for recon-
Surg Oncol. 2014;21(13):4081–9. struction and review of results. Clin Orthop Relat Res.
21. Piccioli A, Spinelli MS, Forsberg JA, Wedin R, Healey 2003;415S:187–97.
JH, Ippolito V, Daolio PA, Ruggieri P, Maccauro G,
Demolitive Surgery for Pelvic Bone
Tumors 10
Joseph Benevenia, Luis A. Guinand,
Kathleen S. Beebe, Joseph Ippolito,
Andrea Angelini, and Pietro Ruggieri
10.1 Introduction (History) external hemipelvectomies make up 0.5% of

lower extremity amputations.
The need for external hemipelvectomy, like other The first report of this procedure in the litera-
amputations for musculoskeletal tumors, has ture was in 1889, which was performed by
decreased in frequency since the advent of limb Theodor Billroth, who was unsuccessful in his
salvage in the 1970s [1–3]. The amputative hemi- attempt after the patient died from shock
pelvectomy is a procedure that involves the (Fig. 10.1) [5]. In 1895, Caciopoli, Jaboulay, and
removal of all or part of the hemipelvis along Girard [6, 7], all considered contemporaries of
with ipsilateral lower extremity. This procedure Billroth, were credited for the first hemipelvec-
results in dramatic functional impairments with tomy performed with the patient surviving. By
increased energy and oxygen requirements for 1909, Ransohoff [8] was recognized as being the
locomotion [4]. Medical comorbidities and psy- first surgeon in the United States to complete an
chosocial issues regarding body image add to the external hemipelvectomy [5, 9].
challenge faced by these patients. While the true In 1932, Kellogg Speed [10] is reported in the
incidence of external hemipelvectomy in the literature as the first to have coined the term
United States is not known, it is estimated to be “hemipelvectomy.” Gordon-Taylor performed
about one per one million each year [5]. Overall, the procedure on a cohort of 41 patients, where
the perioperative mortality was 61%. In 1934, he
observed a perioperative mortality rate of 59.5%,
with the most common causes of death being
J. Benevenia · L. A. Guinand · K. S. Beebe shock and hemorrhage [11, 12]. In 1948, Robert
J. Ippolito Wise reported an innovative operative technique
Department of Orthopaedics, University Hospital,
Newark, NJ, USA used to control hemorrhage by safely ligating the
external iliac vessels during a hemipelvectomy
Department of Orthopaedics, Divison of
Musculoskeletal Oncology, Rutgers—New Jersey (Fig. 10.2) [13, 14]. During the subsequent two
Medical School, Newark, NJ, USA decades, progress was made in these procedures,
e-mail: [email protected]; with postoperative mortality rates reduced to
[email protected]; [email protected]; 22% and a 21% survival rate at 5 years [11, 12,
[email protected]
15, 16]. Hemipelvic amputations were largely
A. Angelini · P. Ruggieri (*) replaced by limb-salvage procedures using
Oncology, University of Padova, Padova, Italy allografts and prosthetic composites pioneered
e-mail: [email protected]; by William Enneking and Henry Mankin during
[email protected] the 1960–1970s [1, 2, 17].
98 J. Benevenia et al.
10.2 I ndication (Bone, Soft Tissue,

Nerve, or Vascular)
Anatomical considerations for limb-salvage ver-

sus amputative are dependent on the three major
structures: the periacetabular bone, femoral ves-
sels, and sciatic nerve. A relative indication for
amputative (external) hemipelvectomy has been
in cases where two of the three anatomical struc-
tures must be resected to achieve an adequate
margin [5, 18]. Other indications are for aggres-
sive and malignant lower limb tumors of bone
and soft-tissue that cannot be managed with
lower level ablative procedures such as hip disar-
ticulation or high above-knee amputation. Other
non-oncologic conditions may necessitate hemi-
pelvectomy including extensive infection,
trauma-related problems, and complications of
pelvic and hip surgery. Decision-making is pri-
marily based upon pathologic analysis and asso-
ciated margins. If function after a limb-salvage is
no better than amputation, a hemipelvectomy
Fig. 10.1 Theodor Billroth. Prussian-born Austrian sur-
geon (and amateur musician). Reported to be the first to
should be considered. From this point forward,
attempt a hemipelvic amputation. (Courtesy of Bert & the use of the term “hemipelvectomy” in this
Judith van der Waal van Dijk) chapter will describe an amputative procedure.
10.3 Preoperative
Staging is necessary prior to consideration of surgi-

cal options. Computed tomography (CT) and mag-
netic resonance imaging (MRI) imaging are
essential to determine the anatomic extent of tumor
resection. The goal of surgery is an adequate (R0 or
R1) margin [19]. Although angiography was uti-
lized in the past, it has been supplanted by CT angi-
ography (CTA) and MR angiography (MRA). With
CT data from the bony pelvis, an individualized
three-dimensional pelvic model can be used for
preoperative planning. The biopsy should be posi-
tioned in line with the incision location for the
hemipelvectomy. In cases where needle biopsy is
not diagnostic, an open biopsy is neccessary. It is
Fig. 10.2 Early hemostatic technique. Drawing of tem- important to avoid contamination of the retroperi-
porary control of the common iliac artery during a hemi- toneum during the biopsy. When neoadjuvant treat-
pelvectomy procedure, using controlling tape placed ment is completed, the tumor is restaged to assess
around the common iliac artery. (Courtesy of Wolters the plan for surgical margins. Imaging with MRI is
Kluwer Health, Inc. Robert A. Wise, Control of the
Common Iliac Artery during Sacro-Iliiac Disarticulation necessary in three planes (axial, coronal, and sagit-
(Hemipelvectomy), Annals of Surgery 1948) tal), and at least T1 (fat sequences) and T2 (fat sup-
10 Demolitive Surgery for Pelvic Bone Tumors 99
Arm rest
Axillary roll
Radiolucent table
Radiolucent foam support
Fig. 10.3 Lateral decubitus position. The positioning of table, for imaging purposes, with a bump under the
the patient in the later decubitus position is used for the patient’s lateral side and axilla. In the upper extremity is
standard hemipelvectomy and subsequent hemipelvec- secured in an arm rest
tomy subtypes. The patient is placed on a radiolucent
pressed) is necessary. Contrast with gadolinium is Insert of a Foley catheter to initiate. The patient is
usually helpful to define tumor extent, especially in then placed in a lateral decubitus position with the
recurrence and after prior surgeries. involved side up on a radiolucent table, for use of
fluoroscopy (Fig. 10.3). The patient is supported
on the table to best facilitate anterior and posterior
10.4 Perioperative Considerations dissection by modifying table tilt. These initial
instructions are applicable to all hemipelvectomy
A multidisciplinary approach may include vascular subtypes.
surgery, general surgery, musculoskeletal oncology, Standard hemipelvectomy involves the disar-
urology, and plastic surgery. In cases of visceral ticulation of the ipsilateral pelvis from the sacro-
involvement, a patient may need a staged diverting iliac joint to the pubic symphysis and removal of
colostomy and ureteral stents. Preoperative medi- the ipsilateral lower extremity (Fig. 10.4). This
cal, oncologic, cardiac, and anesthetic evaluations requires the division of the pelvic vasculature
are necessary. Team composition includes surgical (Fig. 10.5). A posterior myocutaneous or fascio-
technologists, radiological technologist, and proper cutaneous flap from the gluteal region is utilized
instrumentation. Use of a Foley catheter before to cover the resulting defect. This procedure is
beginning the procedure. Scheduling a postopera- reserved for tumors originating from within the
tive bed in the step-down or intensive care unit. pelvis, as well as high-grade pelvic tumors
located in the anterior and lateral aspect of the
pelvis and thigh (Fig. 10.6). This amputative pro-
10.5 Amputative cedure is performed in patients where limb sal-
Hemipelvectomy Types vage cannot be done safely.
10.5.1 Standard Hemipelvectomy 10.5.1.1 Surgical Technique

• Step 1: An anterior dissection is performed
The following sections will be dedicated to the first; the incision (Fig. 10.7) extends from
surgical technique for a standard hemipelvectomy: approximately 5 cm above the anterior supe-
rior iliac spine to the pubic tubercle. Deepen parietal peritoneum off the iliac vessels, per-
the incision through the tensor fascia, external mitting it to fall inferiorly with the viscera.
oblique aponeurosis, and internal oblique and Ligation of the inferior epigastric vessels fol-
transversalis musculature. Followed by retrac- lows. Release the rectus muscle and sheath
tion of the spermatic cord medially. Expose from the pubis. Once the iliac vessels are iden-
the iliac fossa by blunt dissection. Elevate the tified, retract the ureter medially, and ligate
and divide the common iliac artery and vein.
s
Putting lateral traction on the iliac artery and
e vein and ligate and divide their branches to the
sacrum, bladder, and rectum separating the
bladder and rectum from the pelvic sidewall
and exposing the sacral nerve roots. In cases
requiring further exposure, divide the sym-
physis pubis and sacroiliac joint before this
m dissection. The anterior wound is then packed
with warm, moist gauze.
• Step 2: Make a posterior skin incision
(Fig. 10.7), extending approximately from
5 cm above the anterior superior iliac spine,
going over the anterior aspect of the greater
trochanter, parallel to the gluteal fold around
the thigh, connecting with the inferior end of
Fig. 10.4 Hemipelvectomy anatomy. Illustration of the
the anterior incision. The posterior flap is
differing osteotomies for standard hemipelvectomy and raised by dissecting the gluteal fascia directly
subtypes: Standard (s), Modified (m), and Extended (e) off the gluteus maximus (Fig. 10.8). Include
Fig. 10.5 Hemipelvectomy Common iliac Internal iliac

pelvic vascular anatomy artery artery
Iliolumbar artery
(Lumbar branch)
External iliac Lateral sacral

artery artery
Superior gluteal
artery
Umbilical Internal pudendal

artery artery
Superior vesical
artery
Obturator artery
a b
* *
Fig. 10.6 Posterior flap hemipelvectomy case. A with preservation of the gluteus maximus. The tumor was
51-year-old male patient with chondrosarcoma (high- found near the right greater sciatic notch with notable
grade) of the right pelvis. Presented with 4–5 months of peritumoral edema. A decision was made to perform an
significant right lower limb pain, following the sciatic external hemipelvectomy. (a) Magnetic resonance imag-
nerve distribution, and an enlarging right pelvic mass. ing (MRI) T1 (left) showing preservation of the gluteus
Imaging showed a large tumor that begins from the most maximus (asterisks) and skin (arrows) for posterior flap
proximal portion of the right ilium involving the posterior wound closure. (b) MRI T2 (right) showing chondrosar-
soft tissue of the gluteus medius and gluteus minimus, coma on the anterolateral side of the right pelvis
the fascia with the flap. When possible, include

the medial portion of the gluteus maximus
with the flap. Superiorly elevate the flap off
the iliac crest. Divide the external oblique,
sacrospinalis, latissimus dorsi, and quadratus
lumborum from the crest of the ilium. Reflect
the gluteus maximus from the sacrotuberous
ligament, coccyx, and sacrum. Divide the ilio-
psoas muscle; genitofemoral, obturator, and
femoral nerves; and lumbosacral nerve trunk
at the level of the iliac crest.
Abduct the hip, placing tension on the soft tis-

sues around the symphysis pubis. Pass a long right-
angle clamp around the symphysis and divide it
with a scalpel. The sacral nerve roots are divided
preserving the nervi erigentes. Reflect the iliacus
muscle laterally, exposing the anterior aspect of the
sacroiliac joint. Divide the joint anteriorly with a
scalpel or osteotome and divide the iliolumbar liga-
ment. Place considerable traction on the extremity,
separating the pelvic sidewall from the viscera.
Proceeding from anterior to posterior, divide the
following from the pelvic sidewall: urogenital dia-
Fig. 10.7 Posterior flap incision. The incision begins at
the iliac crest, extending across the anterior superior iliac
phragm, pubococcygeus, ischiococcygeus, iliococ-
spine, posteriorly across the gluteal crease, and anteriorly cygeus, piriformis, sacrotuberous ligament, and
along the pubis in the groin to complete the incision sacrospinous ligaments. These structures are all
Gluteus medius muscle

Piriformis
muscle
Deep superior gluteal artery
Sciatic nerve
Superficial gluteal artery
Inferior gluteal
artery
Superficial superior gluteal
artery
Fig. 10.8 Posterior flap dissection
section of the posterior thigh [20]. The resulting

operative defect involving the gluteal region,

hemipelvis, and lower extremity is then enveloped
using an anterior flap made from the quadriceps
femoris muscles, subcutaneous layers, and skin.
The superficial femoral artery is spared to supply
the anterior myocutaneous flap (Fig. 10.12, 10.13,
10.14, 10.15, and 10.16).
Fig. 10.9 Closure of posterior flap. Gluteal fascia is
sutured to the fascia of the abdominal wall, with suction 10.5.2.1 Surgical Technique
drains (not shown) placed prior to skin closure The patient is placed in the lateral decubitus posi-
tion with the operated side up. Secure the patient
divided under tension. Move the extremity anteri- to the table so that it can be tilted to facilitate the
orly and divide the posterior aspect of the sacroiliac anterior and posterior dissections (Fig. 10.6). The
joint to complete the dissection. skin from toes to rib cages is prepared and drape
Closure: Place suction drains in the wound the extremity free. A mark is then made for the
and suture the gluteal fascia to the fascia of the skin incision that should include the length and
abdominal wall. Close the skin (Fig. 10.9). width of the anterior flap that will adequately
Regarding postoperative care, drains, urinary cover the posterior defect. An incision is then
catheters, and vacuum sponge dressings are used made superiorly across the iliac crest to the mid-
as per surgeon preference. Drains are discontin- lateral point, around the buttock just lateral to the
ued when output volume is <50 cc per 12-h shift. anus, and the mid-medial point of the thigh. The
incision is taken down the thigh a distance ade-
quate to cover the posterior defect, across the front
10.5.2 Anterior Flap of the thigh to the mid-lateral point, and superiorly
Hemipelvectomy to join the superior incision (Fig. 10.10).
Anterior flap hemipelvectomy is reserved for • Step 1: Posterior dissection is done first. Skin
tumors located in the gluteal region and proximal margins are preserved, at least 3 cm from the
rectus femoris from the anterior inferior spine,

and the rectus abdominis from the pubis.
Retract the flap medially and dissect along the
femoral nerve into the pelvis to expose the
iliac vessels. Separate the symphysis pubis
while protecting the bladder and urethra.
Ligate and divide the internal iliac vessels at
Fig. 10.10 Anterior Flap incision (right). The incision is
their origin from the common iliac. While
made going superiorly across the iliac crest to the mid-
lateral point, around the buttock just lateral to the anus, applying medial traction on the bladder and
and to the mid-medial point of the thigh. The incision is rectus, divide the visceral branches of the
then brought down the thigh a distance that is adequate to internal iliac vessels. Divide the psoas muscle
cover the posterior defect, across the front of the thigh to
as it joins the iliacus muscle and divide the
the mid-lateral point, and superiorly to join the superior
incision underlying obturator nerve, but protect the
femoral nerve going into the flap. Separate the
lumbosacral nerve and the sacral nerve roots.
anus. The gluteus maximus and sacrospinalis Apply traction on the pelvic diaphragm by
are detached from the sacrum; followed by elevating the extremity and divide the urogen-
the external oblique, sacrospinalis, latissimus ital diaphragm, levator ani, and piriformis
dorsi, and quadratus lumborum muscles from near the pelvis. Separate the sacroiliac joint
the iliac crest. Flex the hip and place the tis- and the iliolumbar ligament and remove the
sues in the region of the gluteal fold under tumor.
tension. Detach the origins of the gluteus
maximus from the coccyx and sacrotuberous Closure: Turn the quadriceps flap onto the
ligament. Bluntly dissect laterally to the rec- posterior defect and suture to the abdominal wall,
tum into the ischiorectal fossa. Maneuver to sacrospinalis, sacrum, and pelvic diaphragm,
the front of the patient and deepen the ante- with a suction drain placed prior to skin closure
rior incision at the linking point of the middle (Fig. 10.17). For postoperative care, the patient
and distal thirds of the thigh through the may ambulate when comfort and stability
quadriceps to the femur. Continue the dissec- permits.
tion laterally in a cephalad direction to the
anterior superior spine severing the vastus
lateralis from the femur; separate the tensor 10.5.3 Extended Hemipelvectomy
fascia femoris from its fascia including the
tumor. Under certain circumstances, a standard external
• Step 2: Medially dissect, starting at Hunter’s hemipelvectomy is not sufficient to prevent a
canal, and ligate and divide the superficial tumor from spreading. In instances where there is
femoral vessels. Track the vessels superiorly tumor involvement of the ilium or ischium that
to the inguinal ligament, dividing and ligating extends up to the sacroiliac joint (SI) and invades
multiple small branches to the adductor mus- through the cartilage, performing a standard
cles. Apply upward traction on the myocuta- hemipelvectomy, in this case, would lead to pos-
neous flap and detach the vastus medialis sible tumor spillage and dissemination. Therefore,
muscle and intermedius from the femur. a more proximal transection of bone through the
Ligate and divide the deep femoral vessels at sacral foramina, known as an extended hemipel-
their origin from the common femoral artery vectomy, can prevent these potentially fatal com-
and vein (Fig. 10.11). Separate the myocuta- plications from occurring. Buttock tumors
neous flap from the pelvis by releasing the located within the gluteus muscles and some soft
abdominal muscles from the iliac crest, the tissue sarcomas that can extend along nerve
sartorius from the anterior superior spine, the sheaths can also be managed with an extended
Common femoral artery
Common iliac artery Proximal profunda

femoris artery
Lateral circumflex artery
External iliac artery Distal profunda

femoris artery
Vastus lateralis muscle

Internal iliac artery
Vastus intermedialis muscle
Produnda femoris artery Rectus femoris muscle
Superficial femoral artery
Vastus medialis muscle

Perforating artery
Sartorius muscle
Superficial femoral artery
Fig. 10.11 Myocutaneous anterior flap anatomy with preservation of the superficial femoral vessels, deep femoral
vessels, and quadricep musculature
Fig. 10.12 Anterior flap hemipelvectomy. Case of the left hemipelvis, extending posteriorly to the gluteal
1—42-year-old female patient with a history of osteosar- region. CT Pelvis Axial (top) shows destruction of the left
coma in the left femur with previous surgical attempts at iliac wing. MRI T1 axial (bottom) shows thinning of the
limb salvage, who presented with persistent left lower gluteus maximus (yellow arrows). Based upon the ana-
extremity pain and enlarging mass. Imaging demonstrated tomic extent of the recurrence, anterior flap hemipelvec-
recurrent left-sided pelvic osteosarcoma with destruction tomy was performed
a b
Fig. 10.13 Anterior flap hemipelvectomy Case 2. Thirty- appears as soft tissue mass with a slightly higher density
five-year-old female patient with synovial sarcoma of the than muscle and with the easily detectable calcifications.
left hemipelvis. (a) Axial CT scan shows extensive (b) Sagittal and coronal CT images of the pelvis better
involvement of the soft tissue in the posterior region of the demonstrate the tumor extension from the proximal ilium
left hemipelvis. The synovial sarcoma (white arrow) to the subtrochanteric area, including the sciatic nerve
a b c
Fig. 10.14 Anterior flap hemipelvectomy Case 2 cont. strate the marked inhomogeneity, enhancement, and sep-
Thirty-five-year-old female patient with synovial sarcoma tation. Pathology confirmed a synovial sarcoma with
of the left hemipelvis. A subsequent MRI demonstrates typical SYT-SSX chromosomal translocation
with coronal (a) sagittal (b), and axial (c) images demon-
hemipelvectomy. This procedure includes the ligation is often necessary. If an anterior flap
removal of the hemipelvis along with additional is being utilized, the medial skin incision
structures not limited to the ipsilateral sacrum, should be made over the mid-sacral spines,
lumbar spine, and/or contralateral pelvis allowing for visualization of the dorsal sacral
(Fig. 10.18). foramina. If a posterior flap is being utilized,
this flap must be dissected back to the mid-
10.5.3.1 Surgical Technique sacral spines to allow for accurate localization
• Step 1: In most circumstances, bone transec- of the dorsal foramina. In the anterior aspect,
tion is done through the sacral foramina; a all of the branches of the internal iliac artery
lumbo-sacral laminectomy with nerve root overlying the sacral nerve roots must be very
a b
c d
Fig. 10.15 Anterior flap hemipelvectomy Case 2 cont. Then, it extends caudally on the anterior aspect of the
Thirty-five-year-old female patient with synovial sarcoma thigh and then laterally just below the greater trochanter
of the left hemipelvis. Surgery consists of an external (white arrow). (a) Vascularization of the flap is based on
hemipelvectomy with partial resection of the ilium and the major vascular pedicle of the pelvis and extremity ves-
atypical anterior thigh flap considering the extensive sels (dot black line). (b) Lateral incision of the myocuta-
involvement of the gluteus. The landmarks are the great neous flap. (c) Release of the quadriceps femoris muscle
trochanter, anterior superior iliac spine. The starting point from the femur (asterisk). (d) During this phase, care must
is the posterior inferior iliac spine and the incision follows be taken not to separate muscle bundles of the myocutane-
the iliac crest reaching the anterior superior iliac spine. ous flap from the overlying skin
carefully dissected. Failure to secure these the osteotome through the bone with the mal-
vessels may lead to substantial blood loss let. The surgeon’s left hand will have located
when the nerve roots and sacrum are being the next highest foramina and will guide the
transected. direction of bony transection. The bone at the
• Step 2: Once the vessels are secured, move to upper portion of the sacrum is fragile, and
the posterior aspect of the patient. Beginning therefore the surgeon must be careful to not
at the tip of the coccyx, an osteotome is now accidentally fracture through the bone during
utilized to divide the coccyx and sacrum in a osteotomy. Finally, the lumbosacral ligament
plane that divides the sacral foramina. The ini- is divided, and the specimen is released.
tial course of the osteotome should be parallel
to the mid-sacral spines. Then, the surgeon
should reach around the coccyx with their left 10.5.4 Modified Hemipelvectomy
hand to find the S-5 neural foramina within
the sacrum, and then work superiorly, pro- The modified hemipelvectomy involves resection
gressing from one foramina to the next. The of the hemipelvis through the ilium for tumors of
surgeon should be holding the osteotome in the lower pelvis sparing a portion of the upper
their right hand and allow the assistant to drive bony pelvis (Fig. 10.19). The wing of the ilium is
a b
c d
Fig. 10.16 Anterior flap hemipelvectomy Case 2 cont. operative defect. (c) Closure, the myocutaneous flap is
Thirty-five-year-old female patient with synovial sarcoma folded posteriorly and bleeding points are secured. (d)
of the left hemipelvis. (a) Surgical field after tumor Postoperative X-ray shows the partial resection of the
removal. (b) The myocutaneous flap is well vascularized ilium and the pubis (modified hemipelvectomy)
(dot black line) and the skin is adequate for coverage of
the most commonly spared structure; however,

the spared portion depends on the location of the
bony pelvic tumor. The most common indication
for a modified hemipelvectomy is large soft tis-
sue sarcomas located high in the medial thigh.
The modified hemipelvectomy may add a cos-
metic and functional aspect. Sparing of the wing
of the ilium allows the patient to maintain a nor-
Fig. 10.17 Closure of anterior flap (right). The quadri- mal waistline, leading to clothes feeling and fit-
ceps flap is turned onto the posterior defect for the closure ting better for the patient. The remaining ilium
of the wound, finished by suturing the quadriceps to the
abdominal wall, sacrospinalis, sacrum, and pelvic dia- may serve as a “post” for a prosthesis. Technical
phragm, with the use of drains, urinary catheters, and advantages are also present, as this procedure
vacuum sponges similar to posterior flap spares the superior and inferior gluteal vessels
Fig. 10.18 Extended hemipelvectomy case. A 28-year- MRI T1 (top) and CT (bottom) axials shows tumor
old female with left pelvic osteosarcoma (right). involvement of the sacroiliac joint (SI) (arrow) with pen-
Presented to the ICU with a fungating large soft tissue etration of gluteal structures. Due to extensive tumor
mass at the left pelvis. Imaging showed tumor replace- involvement at the hemipelvis and sacrum, as well as
ment of the posterior aspect of the ilium, extension tumor crossing the midline, a decision was made to per-
through the SI joint, and a large soft tissue mass expand- form an extended hemipelvectomy
ing through the gluteal musculature and lateral thigh.
of the anterior superior iliac spine and a longer

posterior skin flap is created for closure.
Dissection of the posterior skin flap is consid-
ered complete when an imaginary line can be
drawn between the anterior superior iliac
spine and the tip of the coccyx. After this step,
shift focus to the anterior aspect of the patient
and perform blunt dissection to expose the
common iliac artery and vein. The ipsilateral
rectus muscle should be released from the
superior pubic ramus, then the pubic symphy-
sis is divided. The greater sciatic foramen
should be identified from both the anterior and
posterior aspects. All of the gluteus muscles
should then be divided with electrocautery
Fig. 10.19 Modified hemipelvectomy. Postoperative AP between the anterior superior iliac spine and
radiography of a patient with a right pelvic sarcoma that the posterior aspect of the greater sciatic fora-
underwent a modified hemipelvectomy, with preservation men. The iliacus muscle and psoas tendon
of a portion of ilium (yellow arrow)
should be divided on the anterior to expose the
pelvis. Next, the sartorius muscle is identified
along with the posterior portion of the gluteus arising from the anterior superior iliac spine
maximus muscle, improving the vascularity of and is dissected through its tendon.
the long posterior skin flap used to cover the • Step 2: Now with the pelvis exposed, a Gigli
resulting wound. saw is used to divide the pelvis from the sci-
atic foramen to the section of bone between
10.5.4.1 Surgical Technique the origin of the rectus femoris and sartorius
• Step 1: The incision and early dissection for a muscles. The superior vesical, obturator, and
modified hemipelvectomy are identical to the inferior vesical vessels are severed. Since the
steps described above for a standard hemipel- superior and inferior gluteal vessels are pre-
vectomy, except less skin is incised at the level served, the gluteus muscles maintain perfu-
sion. The sciatic nerve is now divided at the of the hemipelvis may be indicated in such
same level that the muscles were divided. circumstances.
Strong upward traction is placed on the Patients best suited for this procedure are
extremity to transect the urogenital diaphragm those with locally aggressive pelvic tumors that
and levator ani muscle to create a free plane are of low-grade malignancy. Compound hemi-
above the urethra, bladder, and rectum. pelvectomy tends to carry an increased risk of
Drainage and closure of the wound are similar postoperative infection due to the resection of the
to the steps described above for a standard abdominal or pelvic viscera.
hemipelvectomy. It is important to conserve A multidisciplinary approach and consults of
the skin over the anterior superior iliac spine a variety of services may be warranted given the
as increased tension over the bony prominence nature of this subtype.
may lead to delayed wound healing.
10.6 Soft-Tissue Coverage

10.5.5 Compound Hemipelvectomy
Flap sectional should be considered preopera-
In certain cases, pelvic tumors can extend to the tively for proper planning. Surgical site closure
rectum, female adnexal region, bladder, and can be accomplished with an anterior or posterior
other viscera (Fig. 10.20). Resection of abdomi- myocutaneous flap depending on the location of
nal and/or pelvic viscera along with amputation the pelvic tumor. While myocutaneous hemipel-
Fig. 10.20 Compound hemipelvectomy case. Fifty- T2 Coronal (left) and Sagittal (right) imaging show tumor
three-
year-old male patient with undifferentiated pleo- involvement of spermatic cord (yellow arrows). Given the
morphic sarcoma (UPS). Presented with 5 months of right extent of tumor involvement to the spermatic cord, a com-
knee and hip pain. Imaging demonstrated tumor involve- pound hemipelvectomy was performed
ment in the right hemipelvis, right lower extremity. MRI
vectomy flaps are adequate for closure, cases do testinal injuries (3%) [23, 25, 27, 28]. Regarding
have a high probability of wound morbidity after quality of life measures, a study by Beck et al.
soft-tissue reconstruction [21]. The involvement observed that patient’s overall quality of life
of plastic surgeons in cases that would include parameters was comparable between groups
the use of a free flap closure is prudent and some- undergoing both external and internal hemipel-
times required. The availability and consultation vectomy. Beck would also observe those who
of a plastic surgery service are dependent on underwent an external hemipelvectomy had an
institution and surgeon preference. In a study by increase in experienced pain severity and were
Apfelstaedt et al., there was no statistical differ- less independent in bladder function [29].
ence between flap failure and ligation of the com-
mon iliac artery compared with ligation of the
external iliac artery only. Increased operative
References
time and perioperative complications can also
lead to an increase in fap necrosis and infection. 1. Enneking WF. Local resection of malignant lesions of
The best option for reconstructive faps is the use the hip and pelvis. 1966. Clin Orthop. 2002;397:3–11.
of the amputated tissue (free fillet faps) [21–23] https://doi.org/10.1097/00003086-200204000-00002.
struction for primary neoplasms involving the innomi-
nate bone. J Bone Joint Surg Am. 1978;60(6):731–46.
10.7 Complications 3. Mavrogenis AF, Soultanis K, Patapis P, Guerra
G, Fabbri N, Ruggieri P, et al. Pelvic resections.
The rate of associated morbidities regarding exter- Orthopedics. 2012;35(2):e232–43. https://doi.
org/10.3928/01477447-20120123-40.
nal hemipelvectomy is high. While literature 4. Nowroozi F, Salvanelli ML, Gerber LH. Energy
related to the complications following hemipel- expenditure in hip disarticulation and hemipelvec-
vectomy is scarce, rates are estimated to be from tomy amputees. Arch Phys Med Rehabil. 1983;
20% to 50% [5, 24]. Given the high incidence of 64(7):300–3.
5. Mayerson JL, Wooldridge AN, Scharschmidt
postoperative complications, in a review of 136 TJ. Pelvic resection: current concepts. J Am Acad
patients who underwent external hemipelvectomy, Orthop Surg. 2014;22(4):214–22. https://doi.
Kiiski et al. [25] observed postoperative 30-day org/10.5435/jaaos-22-04-214.
mortality of less than 1%. Senchenkov et al. [26] 6. Ariel IM, Hark FW. Disarticulation of an innominate
bone (hemipelvectomy) for primary and metastatic
discovered that the complexity of the case and cancer. Ann Surg. 1949;130(1):76–99. https://doi.
increased time in the operating room were factors org/10.1097/00000658-194907000-00009.
associated with increased rates of postoperative 7. Lee CM Jr, Alt LP. Hemipelvectomy and hip disartic-
complications. Flap necrosis and wound infection ulation for malignant tumors of the pelvis and lower
extremity. Ann Surg. 1953;137(5):704–17. https://doi.
are considered to be the most common complica- org/10.1097/00000658-195305000-00013.
tions in both internal and external hemipelvec- 8. Ransohoff JXIII. Interilio-abdominal amputation:
tomy. Flap necrosis typically requires operative with report of a case. Ann Surg. 1909;50(5):925–
debridement and tissue coverage, often delaying 35. https://doi.org/10.1097/00000658-190911000-
00013.
the patient’s rehabilitation process. 9. Grimer RJ, Chandrasekar CR, Carter SR, Abudu
Many patients have significant phantom pain A, Tillman RM, Jeys L. Hindquarter amputa-
in the early postoperative course. Residual limb tion: is it still needed and what are the outcomes?
spasm has been reported to occur more commonly Bone Joint J. 2013;95-b(1):127–31. https://doi.
org/10.1302/0301-620x.95b1.29131.
than phantom pain and may present weeks or even 10. SpeedK. Hemipelvectomy.AnnSurg.1932;95(2):167–73.
months after the procedure; it is most common https://doi.org/10.1097/00000658-193202000-00002.
after traumatic hemipelvectomy. In their review 11. Gordon-Taylor G. On malignant disease in the region
of 160 external hemipelvectomies, Senchenkov of the hip-joint. J R Coll Surg Edinb. 1959;5:1–21.
12. Gordon-Taylor G, Wiles P. Interinnomino-abdominal
et al. reported a morbidity rate of 54%, including (hind-quarter) amputation. Clin Orthop Relat Res.
intraoperative genitourinary (18%) and gastroin- 1967;53:3–11.
13. Wise RA. Hemipelvectomy for malignant tumors

23. Senchenkov A, Moran SL, Petty PM, Knoetgen J III,
of the bony pelvis and upper part of the thigh. Arch Tran NV, Clay RP, et al. Soft-tissue reconstruction
Surg. 1949;58(6):867–74. https://doi.org/10.1001/ of external hemipelvectomy defects. Plast Reconstr
archsurg.1949.01240030878013. Surg. 2009;124(1):144–55. https://doi.org/10.1097/
14. Wise RA. Control of the common iliac artery
PRS.0b013e3181a80557.
during sacroiliac disarticulation (hemipelvec- 24. Baliski CR, Schachar NS, McKinnon JG, Stuart GC,
tomy). Ann Surg. 1948;128(5):993–8. https://doi. Temple WJ. Hemipelvectomy: a changing perspective
org/10.1097/00000658-194811000-00016. for a rare procedure. Can J Surg. 2004;47(2):99–103.
15. Pack GT, Miller TR. Exarticulation of the innominate 25. Kiiski J, Parry MC, Le Nail LR, Sumathi V,

bone and corresponding lower extremity (hemipel- Stevenson JD, Kaartinen IS, et al. Surgical and onco-
vectomy) for primary and metastatic cancer. A report logical outcomes after hindquarter amputation for
of one hundred and one cases with analysis of the end pelvic sarcoma. Bone Joint J. 2020;102-b(6):788–
results. J Bone Joint Surg Am. 1964;46:91–5. 94. https://doi.org/10.1302/0301-620x.102b6.Bjj-
16. Miller TR. Interilio-abdominal amputation. A report 2019-1317.R1.
of 32 cases. Acta Radiol Suppl. 1959;188:173–89. 26. Senchenkov A, Moran SL, Petty PM, Knoetgen J

https://doi.org/10.1177/0284185159052s18816. III, Clay RP, et al. Predictors of complications and
17. Mankin HJ, Doppelt S, Tomford W. Clinical experi- outcomes of external hemipelvectomy wounds:
ence with allograft implantation. The first ten years. account of 160 consecutive cases. Ann Surg Oncol.
Clin Orthop Relat Res. 1983;174:69–86. 2008;15(1):355–63. https://doi.org/10.1245/s10434-
18. Freitas RR, Crivellaro AL, Mello GJ, Neto MA,
007-9672-5.
Filho Gde F, Silva LV. Hemipelvectomy: Erasto 27. Guder WK, Hardes J, Gosheger G, Henrichs MP,

Gaertner Hospital’s experiences with 32 cases in 10 Nottrott M, Streitbürger A. Analysis of surgical and
years. Rev Bras Ortop. 2010;45(4):413–9. https://doi. oncological outcome in internal and external hemipel-
org/10.1016/s2255-4971(15)30390-6. vectomy in 34 patients above the age of 65 years at a
19. AJCC manual for staging of cancer. 3rd ed.
mean follow-up of 56 months. BMC Musculoskelet
Philadelphia: JB Lippincott; 1988. Disord. 2015;16:33. https://doi.org/10.1186/
20.
Mavrogenis AF, Soultanis K, Patapis P, s12891-015-0494-5.
Papagelopoulos PJ. Anterior thigh flap extended 28. Ham SJ, Schraffordt Koops H, Veth RP, van Horn
hemipelvectomy and spinoiliac arthrodesis. JR, Eisma WH, Hoekstra HJ. External and internal
Surg Oncol. 2011;20(4):e215–21. https://doi. hemipelvectomy for sarcomas of the pelvic girdle:
org/10.1016/j.suronc.2011.07.001. consequences of limb-salvage treatment. Eur J Surg
21. Mat Saad AZ, Halim AS, Faisham WI, Azman WS, Oncol. 1997;23(6):540–6. https://doi.org/10.1016/
Zulmi W. Soft tissue reconstruction following hemi- s0748-7983(97)93173-5.
pelvectomy: eight-year experience and literature 29. Beck LA, Einertson MJ, Winemiller MH, DePompolo
review. ScientificWorldJournal. 2012;2012:702904. RW, Hoppe KM, Sim FF. Functional outcomes
https://doi.org/10.1100/2012/702904. and quality of life after tumor-related hemipelvec-
22. Apffelstaedt JP, Driscoll DL, Spellman JE, Velez AF, tomy. Phys Ther. 2008;88(8):916–27. https://doi.
Gibbs JF, Karakousis CP. Complications and outcome org/10.2522/ptj.20070184.
of external hemipelvectomy in the management of
pelvic tumors. Ann Surg Oncol. 1996;3(3):304–9.
https://doi.org/10.1007/bf02306287.
Principles of Pelvic Surgery
11
Ashley Aratani, Chung Ming Chan,
C. Parker Gibbs, and Mark Scarborough
11.1 Introduction 11.2 Epidemiology
Hemipelvectomies and pelvic resections are infre- Primary bone sarcomas make up about 0.2% of
quently performed procedures, mainly performed all cancers with 15–20% of those being located in
for primary malignancies of bone and soft tissue. the pelvis [2–4]. The most common malignant
These surgeries may also be performed for meta- primary bone sarcomas encountered are osteosar-
static disease, infection, severe trauma, and pallia- coma (35%), chondrosarcoma (30%), and
tion. A hemipelvectomy, or hindquarter Ewing’s sarcoma (16%) [2]. Similarly, they are
amputation, involves resecting the entire hemipel- the most common primary bone sarcomas found
vis and ipsilateral extremity, and prior to the 1970s, in the pelvis. However, 45% of Ewing’s sarcoma
pelvic tumors were often treated with hindquarter cases are located in the pelvis compared to only
amputations. In 1978, Enneking and Dunham [1] 5% of osteosarcomas found in the pelvis. Benign
described a large series of pelvic resections clas- bone tumors that occur in the pelvis that often
sifying them by the segment of the innominate require surgical treatment include giant cell
bone resected. This widely accepted classification, tumor of bone, osteoblastoma, and aneurysmal
and its basis forms the foundation of the subse- bone cyst. Primary soft tissue sarcomas as a
quent decades of research on pelvic resections or group are more common than primary bone sar-
modified internal hemipelvectomies. It is the find- comas and account for 1.0% of all cancers with
ing of multiple centers from around the globe, that 5% being located in the pelvis [3–5]. The same
with appropriate patient selection, advances in surgical principles used in the management of
imaging, adjuvant treatments, and contemporary pelvic primary bone sarcoma can be applied to
resection and reconstructive techniques, limb sal- soft tissue sarcomas involving adjacent pelvic
vage surgery for pelvic sarcoma can be performed bony structures.
safely with reasonable outcomes.
11.3 Surgical Indications

A. Aratani
Stanford University, Redwood City, CA, USA
Surgical extirpation is the cornerstone of sarcoma
management and when the sarcoma involves the
C. M. Chan · C. P. Gibbs · M. Scarborough (*)
innominate bone, surgical resection is often indi-
University of Florida, Gainesville, FL, USA
e-mail: [email protected]; cated if it is technically feasible. Pelvic sarcomas
[email protected] often present when they are already advanced

114 A. Aratani et al.
owing to the significant capacity of the pelvis to mas. Similarly to lymphoma and myeloma,
accommodate a growing tumor prior to it becom- pelvic metastatic disease is usually first treated
ing symptomatic to the patient. This delay in with nonoperative management. However, stud-
detection translates into tumors that are large and ies have reported improved survival and lower
in close proximity to viscera and neurovascular morbidity with pelvic resection compared to
structures. Therefore, surgical treatment becomes curettage in certain patient populations [12].
difficult. These included patients with a solitary metastatic
The goal of surgical management is to remove lesion, primary carcinoma with good prognosis,
the tumor en bloc and with negative surgical mar- and a long time interval between diagnosis of pri-
gins. Where preservation of the limb is possible mary and metastasis [12, 14]. Management of
with the least effect on the patient’s function, metastatic disease has been discussed in greater
limb salvage surgery is preferred [6]. If the extent detail in an earlier chapter.
of tumor involvement in the pelvis makes resec-
tion with negative margins not feasible, this
would be an indication for an external hemipel- 11.4 Surgical Planning
vectomy or hindquarter amputation.
Some patients’ pelvic bone tumors may be Management of pelvic tumors requires a multi-
managed with radiotherapy or other modes of disciplinary approach from both medical and sur-
surgical intervention. An example is Ewing’s sar- gical subspecialties. These include radiology,
coma, which is generally sensitive to chemother- radiation oncology, pathology, medical oncology,
apy and radiation therapy. In such patients where orthopedic oncology, urology, vascular surgery,
resecting the pelvic tumor would cause signifi- surgical oncology, colorectal surgery, neurosur-
cant morbidity, definitive radiation therapy is an gery, gynecology, and plastic surgery. Medical
option that has been reported to provide similar oncologists and radiation oncologists manage
local control rates and acceptable complications adjuvant therapy regimens and treatment timing.
[7, 8]. However, this is debatable as some studies Radiologists are able to assess tumor extent and
have documented a benefit to surgical manage- help plan/perform the biopsy. Depending on
ment of pelvic Ewing’s sarcoma [9–11]. Other structures involved, other surgical subspecialties
minimally invasive options have also been may be required intraoperatively for successful
described for metastasis, including CT-guided and safe tumor resection.
ethanol and thermal ablation, cryoablation, radio- Preoperative laboratory and imaging workup
frequency ablation, and percutaneous acetabulo- are essential to guiding treatment in pelvic malig-
plasty with cement [12]. nancies. Laboratory studies including complete
Other malignancies that may affect the pelvis blood count with differential, electrolyte panel,
include lymphoma and myeloma, where the and serum and urine protein electrophoresis are
mainstay treatments for those patients are che- helpful for diagnosing pancytopenia, life-
motherapy, radiation therapy, and bisphospho- threatening hypercalcemia, and multiple
nates. However, if patients have sustained or are myeloma. Imaging studies should start with plain
at risk for a pathologic fracture, surgical manage- radiographs of the pelvis to get a complete evalu-
ment such as curettage, cementation, and joint ation of pelvic involvement, to identify the pat-
reconstruction may be performed [3, 12]. tern of bone formation in the tumors, and serve as
There has been an overall decrease in mortal- a baseline for future surveillance. Advanced
ity from cancer with the advances in systemic imaging studies including a contrasted MRI are
treatment, with mortality rates in the USA crucial for determining the extent of local dis-
decreasing for men and women by 1.8% and ease, the involvement of major neurovascular
1.4%, respectively [13]. Metastatic disease structures and viscera, and the feasibility of limb
affecting the pelvis is significantly more preva- salvage surgery. A CT of the pelvis is useful to
lent than pelvic primary bone or soft tissue sarco- evaluate the amount of bone destruction and is
11 Principles of Pelvic Surgery 115
often complementary to findings seen on MRI. If include local imaging (plain X-rays, MRI, and/or
the extent of vascular involvement is unclear CT) and metastatic imaging studies (CT chest,
from a contrasted MRI, formal MR or CT angi- bone scan). Where the extent of the tumor raises
ography may be useful. When a primary bone the concern for visceral involvement, the neces-
sarcoma is suspected, staging studies include a sary consultations with urologists, gynecologic
CT chest and bone scan. If a soft tissue sarcoma oncologists, and/or colorectal surgeons to coordi-
is suspected, CT chest for staging is required. nate concurrent visceral resection is
Should the pelvic lesion be suspected to be meta- recommended.
static in nature, a CT chest, abdomen, and pelvis Timing of surgery is usually about 4 weeks
is also performed to identify a visceral primary after finishing the last cycle of chemotherapy and
lesion. In multiple myeloma patients, a skeletal similar timing after neoadjuvant radiation ther-
survey should be performed (CT or plain X-ray) apy to allow cell counts and overlying skin to
to evaluate for other sites of disease. recover. In patients who received neoadjuvant
Once all appropriate imaging has been per- chemotherapy, absolute neutrophil count (ANC)
formed, the biopsy can be planned. The goals of and platelet count need to recover appropriately
the biopsy are to obtain an adequate amount of prior to surgery. Five hundred to 1000 cells/μL
tissue that results in a definitive diagnosis with- and 50,000 cells/μL are considered appropriate
out compromising the planned resection. The ANC and platelet levels, respectively [16]. If cell
biopsy is a fundamental part of the workup and counts have not recovered in time for surgery, the
can cause significant morbidity if done inappro- patient may benefit from another cycle of chemo-
priately. Inappropriate or inadequate biopsy can therapy prior to resection.
lead to a change in treatment plan, making wide
resection and limb salvage surgery no longer fea-
sible [6, 15]. Percutaneous core needle biopsies 11.5 Types of Resection
(with or without CT guidance) and open biopsy
techniques are performed. If a CT-guided biopsy Where limb salvage and a curative resection are
is planned, there must be a discussion between the indicated, resection of the tumor with nega-
the treating surgeon and the radiologist perform- tive margins is the goal. Anatomic considerations
ing the biopsy about the planned approach for determine the technical feasibility of performing
resection in order to prevent contamination of this with preservation of vital structures.
uninvolved compartments. Care must also be In 1978, Enneking and Dunham [1] described
taken not to contaminate the retroperitoneum. If pelvic resections based on the part of the innomi-
an incisional biopsy is performed, the biopsy nate bone excised and surrounding soft tissue.
should be made in line with the planned surgical This widely accepted classification of pelvic
incision. This is usually along the iliac crest. resections and its subsequent modifications are
Intraoperative fluoroscopy may be necessary for based on the bony anatomy of the innominate
tumors without a soft tissue component and con- bone. This classification provides a framework
tained within the bone to ensure lesional tissue is for appreciating how the level of bony resection
collected. Once enough tissue has been obtained, is accompanied by the requirement for dissection
meticulous hemostasis must be achieved to pre- of certain soft tissue structures to permit suffi-
vent hematoma formation as this may extend the cient mobilization of the bony segment for resec-
area of tumor contamination. Lastly, careful tion (Fig. 11.1).
wound closure is important to minimize wound A type I excision involves resection of the
complications, especially if neoadjuvant chemo- ilium and type IA included the buttock muscles
therapy or radiation therapy is anticipated [15]. and sciatic nerve, if necessary (Fig. 11.2). A type
Following neoadjuvant treatment (chemother- II excision is a periacetabular resection involving
apy or radiation therapy), restaging studies resection of the acetabulum while preserving the
should be performed. These studies typically femoral head, whereas a type IIA resection
involves en bloc extra-articular resection of the type IIIA resection includes resection of the fem-
hip joint, i.e., the acetabulum and femoral head. oral neurovascular bundle and surrounding mus-
A type III excision is a pubic resection, and a cles, sparing the hip joint. Depending on the
location and size of the tumor, these resections
can be combined; e.g., a type II/III resection
involves removing the acetabulum and ischium
(Fig. 11.3). A hemipelvectomy would be classi-
fied as type I/II/III. Additions to this classifica-
tion system include sacral and femoral head
resections. Sacral resections are classified as a
type IV resection. Pelvic resections that include
the femoral head are designated H and subdi-
vided into three types: (1) femoral head, (2)
peritrochanteric, and (3) subtrochanteric [17].
Spinopelvic resections have also been classified
into four types based on the how much of the
sacrum is resected and if the sacral resections are
Fig. 11.1 Classification of pelvic resections combined with an external hemipelvectomy [18].
Fig. 11.2 Type I resection for Ewing’s sarcoma with allograft reconstruction
Fig. 11.3 Type II/III H1 resection without reconstruction performed
ment to become the femoral artery. The ureters

11.6 Anatomic Considerations cross over at the bifurcation of the internal and
external iliac vessels, traveling from lateral to
The anatomy of the pelvis is complex, and there medial to enter the bladder.
are important visceral, neurovascular, and bone The gluteus medius and minimus originate
and soft tissue structures that are prone to tumor from the outer table of the ilium and insert onto
involvement or surgical injury and must be con- the greater trochanter. These muscles usually
sidered when performing surgical resections. The provide a margin for tumors arising from the
pelvic ring is stabilized by the pubic symphysis ilium with soft tissue extension laterally. The ilia-
anteriorly and the sacroiliac joint, sacrospinous, cus originates from the iliac fossa and joins the
and sacrotuberous ligaments posteriorly. The psoas major to form the iliopsoas tendon that
abdominal aorta bifurcates into the common iliac inserts on the lesser trochanter. Similarly, the ilia-
arteries usually at the level of the fourth lumbar cus usually provides a medial margin for tumors
vertebra. The internal and external iliac vessels arising from the ilium with soft tissue extension
bifurcate anterior to the sacroiliac joint. The medially. The adductor muscles, gracilis, pelvic
internal iliac vessels exit the pelvis through the floor muscles, and hamstrings originate from dif-
greater sciatic notch, terminating as the superior ferent areas of the pubis and ischium and may
and inferior gluteal vessels. These vessels per- need to be released depending on the tumor’s
fuse the gluteal muscles and preservation of the location. The femoral nerve can be found running
internal iliac artery, and its branches are impor- between the iliacus and psoas major before
tant for maintaining their vascular supply. The exiting the pelvis with the femoral vessels. The
external iliac artery exits the pelvis medial to the sciatic nerve exits the greater sciatic notch ante-
iliopsoas tendon and deep to the inguinal liga- rior to the piriformis.
In male patients, the spermatic cord running in taken not to violate the periosteum when dissect-
the inguinal canal is important to identify and ing and releasing muscles from the pelvis near
protect. Other structures including the prostate, the tumor. A more extensive discussion of
bladder, and corpus of the penis may either be approaches for pelvic resection is included in an
involved or in close proximity to the tumor and earlier chapter.
should be protected. Pelvic resection can affect
male sexual function by damage to the corpus
origin or through neurologic damage. In female 11.8 External Hemipelvectomy/
patients, the ovaries, uterus, and vagina may be Hindquarter Amputation
involved or in close proximity to the tumor. The
urethra runs just inferior to the pubic symphysis Limb salvage surgery for sarcoma may be pre-
and must be retracted if a pubic symphyseal cut is cluded by extensive invasion of critical structures
required. The rectum may also be involved by tumor. The indications for an external hemipel-
depending on the tumor’s location. Any prior vectomy or a hindquarter amputation are mostly
radiation therapy to the pelvic area increases related to how advanced a disease is locally and
fibrosis and adhesions and increases the risk to the extent to which adjacent neurovascular struc-
visceral damage. tures are invaded. Should obtaining clear surgical
margins be deemed unfeasible with limb salvage
surgery in a patient without systemic disease,
11.7 Surgical Approaches hindquarter amputation should be considered.
Neural structures at the level of the pelvis in par-
A utilitarian extended ilioinguinal approach is ticular do not lend themselves well to reconstruc-
commonly used for pelvic resection surgery. It tion, owing to the distance from the level of injury
permits access to the pubic symphysis anteriorly, at the time of transection to the sensory and motor
the sacroiliac joint and sacrum posteriorly, as target end organs in the lower extremity. Loss of
well as the hip and proximal femur distally. muscular control or loss of nerve function can be
Depending on the location of the tumor, modifi- compensated for with the use of external orthoses
cations to this incision may be necessary. For and is not an absolute contraindication to limb
type I resections, only the posterior limb of the salvage. Major artery reconstruction when indi-
extended ilioinguinal approach may be required. cated adds to the complexity of the surgery but
In type II resections, a lateral limb down the lat- with venous or endoprosthetic grafting can be
eral thigh is required and can be achieved by associated with acceptable limb salvage rate [25].
curving the distal and medial end of the incision When resection of major nerves is paired with an
back toward the lateral aspect of the femur. unstable skeleton, external hemipelvectomy
Alternatives to this include a T-incision [19] and should be considered as the resulting lower
the modified Ollier transtrochanteric approach extremity may be insensate, weak, or flail and
[20, 21]. If the resection is to remain extra- unstable. Living with such a limb may be more
articular, the joint capsule is kept intact and the burdensome and disabling to a patient than living
proximal femur is osteotomized at the appropri- without it. The three structures that should be con-
ate level. If the resection is intra-articular, the sidering in the decision making are the periace-
capsule is circumferentially incised to allow fem- tabular portion of the pelvis, the sciatic nerve, and
oral head dislocation. Type III resections may the femoral nerve. Where two of these three struc-
require extending the anterior limb over to the tures require resection, external hemipelvectomy
contralateral side of the pubic symphysis. should be seriously considered [3, 16].
Anterior, posterior, or combined approaches may When an external hemipelvectomy is indi-
be needed for type IV resections depending on cated, the two most common conventional
location, size, and presacral soft tissue extension approaches are the posterior flap hemipelvec-
[16, 22–24]. With all resections, care must be tomy or the anterior flap hemipelvectomy.
A posterior flap hemipelvectomy involves the and lateral incisions along the borders of the
following components: (1) Ilioinguinal approach quadriceps muscles and raising the anterior com-
to explore the retroperitoneal space and for partments of the thigh as a flap off the femur. The
release of the anterior abdominal wall muscles, superficial femoral artery is ligated distally as it
mobilization, ligation, and transection of struc- courses from the anterior compartment through
tures including the iliac vessels. (2) Perineal dis- the adductor hiatus to the popliteal fossa. The
section involving a posterior extension of the adductor compartment may also be included if a
medial end of the ilioinguinal approach toward larger flap is desired or if part of the anterior
the posterior aspect of the thigh and the region of compartment of the thigh has to be removed.
the horizontal gluteal crease. This permits access
to the pubic symphysis for their disarticulation or
the medial ends of the pubic rami for their tran- 11.9 Complications
section. (3) Raising the posterior flap which typi-
cally extends from the iliotibial band laterally, Pelvic tumor resection surgery is complex and
connecting to the ilioinguinal incision superiorly often prolonged and is associated with a signifi-
and coursing posteriorly along the horizontal glu- cant risk of complications that patients should be
teal fold. This flap may be a fasciocutaneous flap duly counseled on. The mortality in the periop-
in the classical hemipelvectomy/hindquarter erative period is an important risk, and studies
amputation where its perfusion is based of fascio- document rates of perioperative mortality from
cutaneous perforators from its base. However, 0% to 10% [16, 26–30]. This illustrates that with
where the gluteus maximus is not involved by good perioperative care, such surgeries can be
disease, preservation of this muscle and the supe- performed safely. However, apart from the risk of
rior gluteal artery with it results in a more robust mortality, the complication rate for pelvic tumor
flap for closure. (4) Mobilization and division of resection surgery is quite significant. Studies
the pelvic floor muscles to complete the detach- vary in their reported rates which range from
ment of the pelvis inferiorly. (5) Final mobiliza- 15% to 68% [26–29, 31–37]. Since complica-
tion and amputation of the hemipelvis. A classical tions occur at such a high rate, they should be
hemipelvectomy involves disarticulation of the anticipated and prevented when possible and
sacroiliac joint. In a modified hemipelvectomy, dealt with as they occur.
the innominate bone is osteotomized through the These complications may be divided into the
sciatic notch, while in an extended hemipelvec- intraoperative complications, complications
tomy, the osteotomy is through the sacrum. diagnosed in the early postoperative period, those
An anterior flap hemipelvectomy is indicated diagnosed in the intermediate postoperative
for the management of sarcoma when the disease period, and late complications.
extends posteriorly and precludes the preserva- Intraoperative complications are related to the
tion of the posterior flap. Prior surgery with con- frequently prolonged nature of these surgeries,
tamination of the posterior flap by sarcoma is the risk of sudden blood loss during certain por-
also an indication for this approach. Nononcologic tions of the procedure, and inadvertent injury to
indication for this procedure includes extensive pelvic viscera. The prolonged surgeries put
sacral or trochanteric decubitus ulcers and osteo- patients at risk of pressure sores, deep vein
myelitis such as in paraplegic patients. The ante- thrombosis, and respiratory compromise. Patient
rior flap is a myocutaneous flap that is perfused with significant cardiovascular comorbidities
by the femoral artery and includes muscles of the should be closely monitored owing to the poten-
anterior compartment of the thigh and the overly- tial for significant blood loss throughout the
ing skin. The good vascularity of this flap is course of the surgery as well as the risk of sudden
regarded as an advantage, and the muscle bulk is brisk blood loss. The extensive nature of some
useful in occluding dead space on wound closure. pelvic tumors requires mobilization of pelvic vis-
Dissection of the anterior flap involves medial cera, and this puts patients at risk of injury to
these pelvic viscera. Identification of injury to be a superficial infection. Studies have also found
pelvic viscera is crucial if these injuries are to be an association of reconstruction with a higher
managed appropriately and to permit appropriate rate of complications [33, 40, 41]. While antibi-
consultation with other surgical subspecialties if otic therapy may be adequate for treating some
indicated to address these complications. The cases of soft tissue infection, severe superficial
pelvic viscera or vessels are at particular risk of infections and deep infection frequently require
injury in type III resections [29]. Due to these repeat surgery. Deep infection in the setting of
risks, high-risk anesthesia teams are needed, and endoprosthetic or allograft reconstruction may
patients should be adequately resuscitated and also involve removal of the components and
arrangements should be made for adequate allograft used in the reconstruction. The develop-
amounts of blood products to be available. ment of wound complications such as infection
Typically, patients are taken care of in monitored or wound dehiscence may preclude the initiation
intensive care units postoperatively. or restarting of systemic cytotoxic chemotherapy.
In the early postoperative period, wound com- It is thus imperative that the indication for recon-
plications are of significant concern and neuro- struction be weighed against the increased risk of
logic deficits resulting from surgery are often complications and the impact that this may have
apparent. Flap necrosis, wound dehiscence, and on a patient’s subsequent systemic treatment.
soft tissue infections are common especially con- Complications that are diagnosed late include
sidering the large size of these wounds, the risk of abdominal hernia and complications related to
bacterial translocation from pelvic viscera, and the failure of reconstructive procedures. Abdominal
challenges in maintaining the hygiene of portions hernia is also a known complication owing to the
of the wounds that are adjacent to the perineum. need to detach the insertions of the abdominal
Increased rate of wound complications and flap wall musculature, and abdominal wall reconstruc-
necrosis have been associated with longer surgical tion may be indicated where there is a significant
times; the extent of resection and ligation of the defect [31, 42, 43]. Prosthetic reconstructions
common iliac artery is also associated with higher may also fail after the early postoperative period,
rates of flap necrosis [31]. Neurologic deficits are and these include periprosthetic fractures, implant
also a common complication as manipulation and fracture, dislocation, and loosening.
mobilization of nerves are required to perform pel- Lastly, there is the risk of local recurrence.
vic resection, while many of these deficits may Local recurrence rates vary significantly in pub-
improve with time others do not improve owing to lished studies from 9% to 45%, while it is clear that
the severity of the injury [29, 38]. Resections intralesional margins result in poorer outcomes,
involving the ilium involve mobilizing the psoas several large studies have not shown differences in
muscle and femoral nerve, resection around the recurrence rates in marginal and wide resections
sciatic notch may involve significant manipulation [19, 28, 33, 34, 36, 44–48]. Where local recurrence
of the sciatic nerve, and resection involving the or deep infection arises and limb preservation is not
sacrum frequently involves significant nerve root feasible, a secondary hindquarter amputation may
manipulation or sacrifice [39]. Reconstructive probe indicated, and studies report a rate of 8–12% of
cedures may also result in an increase in the ten- patients requiring this [34, 37, 49].
sion across the nerves and result in injury. Nerve
injuries may also involve the contralateral side
which may be related to surgical manipulation, 11.10 Conclusion
prolonged surgery, or ischemia.
While superficial infections and wound dehis- Pelvic resection for oncologic disease requires
cence are often diagnosed in the early postopera- meticulous planning, multidisciplinary teams,
tive period, deep infections may only be and careful execution owing to the complex anat-
diagnosed at a later stage or after failure of anti- omy, challenging exposure, and often advanced
biotic therapy to manage what appears at first to nature of these tumors on presentation. The
understanding gained from decades of research 15. Mankin HJ, Lange TA, Spanier SS. The hazards of
biopsy in patients with malignant primary bone
on the topic has improved the safety and out- and soft-tissue tumors. J Bone Joint Surg Am.
comes of this type of surgery. 1982;64:1121–7.
16. Mavrogenis AF, Soultanis K, Patapis P, Guerra G,
Fabbri N, Ruggieri P, Papagelopoulos PJ. Pelvic
resections. Orthopedics. 2012;35:e232–43.
17. Simon MA, Springfield D, editors. Surgery for bone
References and soft-tissue tumors. 1st ed. Philadelphia, PA:
LWW; 1997.
1. Enneking WF, Dunham WK. Resection and recon- 18. Rose PS, Yaszemski MJ, Dekutoski MB, Huddleston
struction for primary neoplasms involving the innom- PM, Nassr A, Shives TC, Sim FH. 7.p.12 classifica-
inate bone. J Bone Joint Surg Am. 1978;60:731–46. tion of spinopelvic resections: oncologic and recon-
2. Biermann JS, Chow W, Reed DR, et al. NCCN guide- structive implications. Orthop Proc. 2010;92-B:468.
lines insights: bone cancer, Version 2.2017. J Natl 19. Lackman RD, Crawford EA, Hosalkar HS, King JJ,
Compr Cancer Netw. 2017;15:155–67. Ogilvie CM. Internal hemipelvectomy for pelvic
3. Mayerson JL, Wooldridge AN, Scharschmidt sarcomas using a T-incision surgical approach. Clin
TJ. Pelvic resection: current concepts. J Am Acad Orthop. 2009;467:2677–84.
Orthop Surg. 2014;22:214–22. 20.
McDowell S, Mullis B, Knight BS, Dahners
4. Beckingsale TB, Shaw C. Epidemiology of LE. Modified Ollier transtrochanteric approach for
bone and soft-tissue sarcomas. Orthop Trauma. the treatment of acetabular fractures. Orthopedics.
2017;31:144–50. 2012;35:e132–6.
5. von Mehren M, Randall RL, Benjamin RS, et al. 21. Ko J-Y, Wang C-J, Lin C-FJ, Shih C-H. Periacetabular
Soft tissue sarcoma, Version 2.2018, NCCN clinical osteotomy through a modified Ollier transtrochanteric
practice guidelines in oncology. J Natl Compr Cancer approach for treatment of painful dysplastic hips. J
Netw. 2018;16:536–63. Bone Joint Surg Am. 2002;84:1594–604.
6. O’Connor MI, Sim FH. Salvage of the limb in the 22. Accardo S, Gonzalez R. Anatomy and surgical

treatment of malignant pelvic tumors. J Bone Joint approaches to the sacrum. In: Ruggieri P, Angelini
Surg Am. 1989;71:481–94. A, Vanel D, Picci P, editors. Tumors of the sacrum.
7. Indelicato DJ, Keole SR, Shahlaee AH, Shi W, Morris Diagnosis and treatment of benign and malignant
CG, Gibbs CP, Scarborough MT, Marcus RB. Impact tumors. Cham: Springer International Publishing;
of local management on long-term outcomes in Ewing 2017. p. 247–54.
tumors of the pelvis and sacral bones: the University 23. Zhang Y, Guo W, Tang X, Yang R, Ji T, Yang Y, Wang
of Florida experience. Int J Radiat Oncol Biol Phys. Y, Wei R. En bloc resection of pelvic sarcomas with
2008;72:41–8. sacral invasion: a classification of surgical approaches
8. Ahmed SK, Robinson SI, Arndt CAS, Petersen IA, and outcomes. Bone Jt J. 2018;100-B:798–805.
Haddock MG, Rose PS, Issa Laack NN. Pelvis Ewing 24. Dubory A, Missenard G, Lambert B, Court C. “En
sarcoma: local control and survival in the modern era. bloc” resection of sacral chordomas by combined
Pediatr Blood Cancer. 2017; https://doi.org/10.1002/ anterior and posterior surgical approach: a monocen-
pbc.26504. tric retrospective review about 29 cases. Eur Spine J.
9. Rödl RW, Hoffmann C, Gosheger G, Leidinger B, 2014;23:1940–8.
Jürgens H, Winkelmann W. Ewing’s sarcoma of the 25. McGoldrick NP, Butler JS, Lavelle M, Sheehan S,
pelvis: combined surgery and radiotherapy treatment. Dudeney S, O’Toole GC. Resection and reconstruc-
J Surg Oncol. 2003;83:154–60. tion of pelvic and extremity soft tissue sarcomas with
10. Yang RS, Eckardt JJ, Eilber FR, Rosen G, Forscher major vascular involvement: current concepts. World
CA, Dorey FJ, Kelly CM, al-Shaikh R. Surgical indi- J Orthop. 2016;7:293–300.
cations for Ewing’s sarcoma of the pelvis. Cancer. 26. Apffelstaedt JP, Driscoll DL, Spellman JE, Velez AF,
1995;76:1388–97. Gibbs JF, Karakousis CP. Complications and outcome
11. Thorpe SW, Weiss KR, Goodman MA, Heyl AE,
of external hemipelvectomy in the management of
McGough RL. Should aggressive surgical local con- pelvic tumors. Ann Surg Oncol. 1996;3:304–9.
trol be attempted in all patients with metastatic or pel- 27.
Apffelstaedt JP, Driscoll DL, Karakousis
vic Ewing’s sarcoma? Sarcoma. 2012;2012:953602. CP. Partial and complete internal hemipelvectomy:
12. Müller DA, Capanna R. The surgical treatment of pel- complications and long-term follow-up. J Am Coll
vic bone metastases. Adv Orthop. 2015;2015:525363. Surg. 1995;181:43–8.
13. www.seer.cancer.gov. 28. Ham SJ, Schraffordt Koops H, Veth RP, van Horn JR,
14. Guzik G. Oncological, surgical and functional results Eisma WH, Hoekstra HJ. External and internal hemi-
of the treatment of patients after hemipelvectomy pelvectomy for sarcomas of the pelvic girdle: conse-
due to metastases. BMC Musculoskelet Disord. quences of limb-salvage treatment. Eur J Surg Oncol.
2018;19:63. 1997;23:540–6.
29. Capanna R, van Horn JR, Guernelli N, Briccoli A, 39. Sabourin M, Biau D, Babinet A, Dumaine V, Tomeno
Ruggieri P, Biagini R, Bettelli G, Campanacci M, B, Anract P. Surgical management of pelvic primary
Bettellini G. Complications of pelvic resections. Arch bone tumors involving the sacroiliac joint. Orthop
Orthop Trauma Surg. 1987;106:71–7. Traumatol Surg Res. 2009;95:284–92.
30. Yuen A, Ek ET, Choong PF. Research: Is resection of 40. Jansen JA, van de Sande MAJ, Dijkstra PDS. Poor
tumours involving the pelvic ring justified?: a review long-term clinical results of saddle prosthesis after
of 49 consecutive cases. Int Semin Surg Oncol. resection of periacetabular tumors. Clin Orthop.
2005;2:9. 2013;471:324–31.
31. Senchenkov A, Moran SL, Petty PM, Knoetgen J, 41. Benatto MT, Hussein AM, Gava NF, Maranho DA,
Clay RP, Bite U, Barnes SA, Sim FH. Predictors of Engel EE. Complications and cost analysis of hemi-
complications and outcomes of external hemipelvec- pelvectomy for the treatment of pelvic tumors. Acta
tomy wounds: account of 160 consecutive cases. Ann Ortop Bras. 2019;27:104–7.
Surg Oncol. 2008;15:355–63. 42.
Die Trill J, Madrid JMF, Ferrero E, et al.
32. Baliski CR, Schachar NS, McKinnon JG, Stuart GC, Posthemipelvectomy hernia. Hernia. 2005;9:375–7.
Temple WJ. Hemipelvectomy: a changing perspective 43.
Kraybill WG, Standiford SB, Johnson
for a rare procedure. Can J Surg. 2004;47:99–103. FE. Posthemipelvectomy hernia. J Surg Oncol.
33. Farfalli GL, Albergo JI, Ritacco LE, Ayerza MA,
1992;51:38–41.
Muscolo DL, Aponte-Tinao LA. Oncologic and clini- 44. Shin KH, Rougraff BT, Simon MA. Oncologic out-
cal outcomes in pelvic primary bone sarcomas treated comes of primary bone sarcomas of the pelvis. Clin
with limb salvage surgery. Musculoskelet Surg. Orthop. 1994;(304):207–17.
2015;99:237–42. 45. Mankin HJ, Hornicek FJ. Internal hemipelvectomy
34. Angelini A, Calabrò T, Pala E, Trovarelli G, Maraldi for the management of pelvic sarcomas. Surg Oncol
M, Ruggieri P. Resection and reconstruction of pelvic Clin N Am. 2005;14:381–96.
bone tumors. Orthopedics. 2015;38:87–93. 46. Scully SP, Temple HT, O’Keefe RJ, Scarborough MT,
35. Angelini A, Drago G, Trovarelli G, Calabrò T,
Mankin HJ, Gebhardt MC. Role of surgical resec-
Ruggieri P. Infection after surgical resection for pel- tion in pelvic Ewing’s sarcoma. J Clin Oncol. 2016;
vic bone tumors: an analysis of 270 patients from one https://doi.org/10.1200/JCO.1995.13.9.2336.
institution. Clin Orthop. 2014;472:349–59. 47. Han I, Lee YM, Cho HS, Oh JH, Lee SH, Kim

36. Abudu A, Grimer RJ, Cannon SR, Carter SR, Sneath H-S. Outcome after surgical treatment of pelvic sar-
RS. Reconstruction of the hemipelvis after the exci- comas. Clin Orthop Surg. 2010;2:160–6.
sion of malignant tumours. Complications and func- 48. Shao QD, Yan X, Sun JY, Xu TM. Internal hemi-
tional outcome of prostheses. J Bone Joint Surg Br. pelvectomy with reconstruction for primary pel-
1997;79:773–9. vic neoplasm: a systematic review. ANZ J Surg.
37. Hillmann A, Hoffmann C, Gosheger G, Rödl R,
2015;85:553–60.
Winkelmann W, Ozaki T. Tumors of the pelvis: com- 49. Guder WK, Hardes J, Gosheger G, Henrichs M-P,
plications after reconstruction. Arch Orthop Trauma Nottrott M, Streitbürger A. Analysis of surgical and
Surg. 2003;123:340–4. oncological outcome in internal and external hemipel-
38. Campanacci M, Capanna R. Pelvic resections: the vectomy in 34 patients above the age of 65 years at a
Rizzoli Institute experience. Orthop Clin North Am. mean follow-up of 56 months. BMC Musculoskelet
1991;22:65–86. Disord. 2015;16:33.
Innovative Techniques in Pelvic
Reconstructions 12
Howard Park, Akash A. Shah,
and Francis J. Hornicek
12.1 Introduction to spinopelvic, sacral, bulk allograft, and proxi-

mal femoral reconstructions.
Significant advancements have been made in pelvic
reconstruction since the first documented pelvic
resections were attempted at the turn of the twenti- 12.2 Indications
eth century. The first documented nonfatal pelvic
resection was for sarcoma performed in 1895 by Pelvic resections including hemipelvectomies
Girard of Berne, and in 1935, Sir Gordon Gordon- occur relatively rarely with rough estimates
Taylor of Britain called such resections as “one of approximating 1 per one million persons annu-
the most colossal mutilations practiced on the ally [3]. Enneking and Dunham classified pelvic
human frame” [1, 2]. Since that time, the knowl- resections in relation to sarcoma of the innomi-
edge base of pelvic anatomy, advancements in tech- nate bone that failed treatment by medical means
nique and cross-sectional imaging, and the rapid [4]. A variety of oncologic processes can indicate
sophistication of metallurgy and implant develop- a pelvic resection and reconstruction. Pelvic pri-
ment have made this once morbid procedure now mary bone tumors compose 15–20% of all pri-
safe and effective. mary bone tumors. Furthermore, chondrosarcoma,
Pelvic resections can be left flail or be recon- osteosarcoma, and Ewing’s sarcoma compose
structed to maximize function. In order to achieve 50–80% of all pelvic bone tumors [5, 6]. Ewing’s
an adequate reconstruction, innovative tech- sarcoma and osteosarcoma are most highly prev-
niques have been proposed with various use of alent in the adolescent or young adult population,
autograft, allograft, and custom implants that whereas chondrosarcoma most often presents in
span from the spine to the femur. Primary muscu- the fourth to seventh decades of life [5].
loskeletal tumors, metastatic lesions, trauma, and Chordoma, fibrosarcoma, Langerhans cells his-
infection of the pelvis are among the indications tiocytosis, aneurysmal bone cyst, giant cell
for this relatively uncommon procedure. The tumor, and fibrous dysplasia can also necessitate
scope of this discussion will focus on the indica- pelvic resection although with less frequency
tions, relevant anatomy, and innovations related than those oncologic process aforementioned.
Metastatic disease to the pelvis can originate
from the breast, lung, prostate, kidney, and thy-
H. Park · A. A. Shah · F. J. Hornicek (*)
roid; however, many of these lesions can be man-
UCLA, Los Angeles, CA, USA
e-mail: [email protected]; aged with radiation or chemotherapy with a
[email protected]; [email protected] minority of metastatic lesions indicating pelvic

124 H. Park et al.
resection [7, 8]. Infection, complications from resection significantly increases the morbidity to
arthroplasty, and trauma can also require pelvic the patient. Therefore, in principle, the indication
resection, albeit at a relatively lower frequency. for pelvic reconstruction is any patient who has
undergone destabilizing pelvic resection for the
aforementioned pathologies that is willing and
12.3 Classification of Pelvic able to withstand the extensive surgery and reha-
Resection bilitation intended to maximize function.
Pelvic resections vary widely in location and

size, and therefore, pelvic reconstructions can 12.4 Spinopelvic Reconstruction
vary. To organize these pelvic resections and the
necessity and types of reconstructions, it is help- Pelvic resections that extend to the sacrum or
ful to understand the Enneking and Dunham clas- lumbar spine often require spinopelvic recon-
sification of pelvic resections. The Enneking and struction to reestablish the relationship between
Dunham classification is based on the specific the pelvic and axial skeleton. Depending on the
anatomic locations of resection. In brief, resec- size and extent of resection, the reconstruction
tions of the ilium are Type 1, resections of the construct can vary widely. Instrumentation to aid
periacetabular region are Type 2, resections of in these reconstructions also varies with the
the pubic rami or obturator rings are Type 3, and employment of pedicle screws, plates and screws,
resections of the sacrum are Type 4, which have rods, allograft, autograft, etc., and there exists
subtypes depending on resection of adjacent very limited literature and no standard of care
anatomy [4]. Reconstruction can be considered in with regard to these constructs. The following are
resection types that disrupt the pelvic ring, illustrative cases and insights into spinopelvic
although leaving a patient flail is often a viable reconstructions by the authors of this chapter.
alternative. For example, within Type 1 resec-
tions of the ilium, reconstruction may be favored
if the pelvic ring is disrupted as in the case of 12.4.1 Case 1
complete ilium resection but not necessary in
cases of partial resection with an intact pelvic The patient is a 33-year-old male who initially pre-
ring. Extensive resections can disrupt the conti- sented with back and hip pain with subsequent
nuity of the pelvic ring and the ability of the pel- biopsy-proven grade 3 chondroblastoma of the pel-
vis to support the continuity between the lower vis without metastatic disease. The patient under-
extremity and axial skeleton. went neoadjuvant chemotherapy with 90% tumor
The overarching goal for pelvic reconstruction necrosis prior to referral to a tertiary orthopedic
is to provide pelvic girdle support for maximal oncology service. Advanced cross-sectional imag-
function. The function of the pelvis is to provide ing of the tumor revealed an expansive lesion about
continuity between the lower extremity and the the left lower lumbar spine, sacrum, and ilium with
axial skeleton to allow for locomotion in addition posterior soft tissue extension (Fig. 12.1). The
to protecting the pelvic contents and providing operation to resect and reconstruct the spinopelvic
muscular attachments for the torso and lower defect was undertaken in two stages.
extremities. Although pelvic resection without
reconstruction should be considered, the func- 12.4.1.1 Stage 1
tional outcomes of resection without reconstruc- First, the patient was positioned prone and a pos-
tion are often unacceptable to patient and surgeon terior longitudinal midline incision with exten-
in light of modern techniques and implant design. sion over the left gluteus was utilized to access the
However, reconstruction must be considered lumbar spine and sacrum. Pedicle screws were
carefully with a full assessment of the complica- placed at the L2/L3 levels bilaterally and L4/L5
tion profile as reconstruction following a pelvic levels on the right; additionally, rods were placed
12 Innovative Techniques in Pelvic Reconstructions 125
Fig. 12.1 Grade 3 chondroblastoma of the pelvis in a 33-year-old male. Magnetic resonance cross-sectional imaging
show expansive lesion involving the lumbar spine, sacrum, and ilium
to enable spinopelvic fixation in the second stage

of the surgery. Laminectomy and ligation of the
nerve roots on the left lower lumbar and sacral
levels were completed as well as posterior oste-
otomy cuts through margin negative pelvis with a
diamond tip burred under navigation guidance.
Dissection of the tumor to achieve negative mar-
gins were carried out from the posterior incision.
12.4.1.2 Stage 2
In stage 2 of the operation performed 3 days later, the
patient was positioned supine. Vascular surgery per-
formed an anterior approach to the lower lumbar
spine and sacrum with mobilization of the great ves-
sels which required ligation of the left internal iliac
artery and left common iliac vein. The osteotomy cuts
were completed from the anterior approach under
navigation guidance and the tumor, hemisacrum, and
hemi-ilium were resected in an en bloc fashion while Fig. 12.2 Same patient. Postoperative radiograph shows
the spinopelvic reconstruction with multiple pedicled
preserving acetabulum. Plastic surgery then per- screws and rods associated with vascularized fibular auto-
formed a vascularized fibular autograft to span the graft and a humeral shaft allograft for bone support
lumbar spine to acetabulum adjacent to a humeral
shaft allograft for structural support. Multiple pedicle
screws into the remaining pelvis were placed and con- vastus lateralis rotational muscle flap was mobilized
nected to the rods placed during the first stage to fur- anteriorly to provide coverage over the hardware
ther reconstruct the resected hemipelvis (Fig. 12.2). A within the pelvis. Toe-touch weight-bearing restric-
126 H. Park et al.
tions were in place for 8 weeks. At last follow-up, the 12.5.1 Case 2
patient was ambulating independently with a walker
within his house. At 6 months follow-up, the patient The patient is a 32-year-old male with right leg
has remained recurrence-free. sciatica with calf atrophy for several months who
initially attempted nonoperative measures. MRI
revealed a large tumoral lesion centered about the
12.5 Sacral Reconstruction R sacral ala extending into the S1 and S2 neural
foramen and into the pelvis displacing the blad-
The indications for sacral reconstruction most der (Fig. 12.3). Subsequent biopsy demonstrated
frequently include chordoma, chondrosarcoma, grade 1 chondrosarcoma, and staging revealed no
giant cell tumor, and osteosarcoma [9]. This rela- metastatic disease.
tively rare reconstruction can be addressed with
various techniques, but the principle of maintain- 12.5.1.1 Stage 1
ing pelvic ring continuity remains paramount. The patient was positioned supine for an anterior
The use of autograft, allograft, titanium bars, approach to the sacrum by vascular surgery. The
plates, and custom 3-D printed implants have all great vessels were mobilized, and the right inter-
been documented in the literature [9–12]. nal iliac artery and vein were ligated to gain
Sacrectomy and reconstruction carry a significant access to the sacrum. The tumor was dissected
morbidity with a neurologic deficit expected in away from the bowel and peritoneum. Utilizing
most reconstructions in addition to a high infec- navigation, osteotomies with a diamond-tipped
tion and wound-healing complication rate. burr were performed in a longitudinal fashion
Furthermore, nerve ligation at the sacral level can along the sacrum and L5–S1 disk on the right
affect bladder, bowel, and sexual function, and it side with preservation of the left-sided hemisa-
is our preference to preserve at least one side of crum. The left hemisacrum was preserved in
the sacrum in an attempt to maximize function. order to maximize the patient’s postoperative
Based on the current literature, there is no con- bowel, bladder, and sexual function.
sensus on the best reconstruction method, but
anterior spinal column fixation in conjunction 12.5.1.2 Stage 2
with posterior instrumentation may be required Two days later, the patient was positioned prone
to minimize the risk of hardware failure. and a midline lumbar incision with lateral exten-
Fig. 12.3 Grade 1 chondrosarcoma of the sacrum in a 32-year-old male. Sagittal and axial MRI show the tumor exten-
sion into the proximal sacrum (S1 and S2 neural foramen)
tumor lesion emanating from the left lumbosacral

junction which extended both posterior and ante-
rior to the ilium (Fig. 12.5). Subsequent biopsy
revealed a grade 1 chondrosarcoma.
12.5.2.1 Operation
A midline longitudinal incision through the lum-
bar spine which extended over the gluteus to the
left lateral thigh was utilized to access the lumbar
spine and tumor. The tumor was carefully dis-
sected away from the lamina of the low lumbar
spine, and the nerve roots were meticulously dis-
sected away from the tumor. Laminectomy was
completed in the lower lumbar spine and sacrum.
Utilizing navigation, a diamond tip burr was uti-
Fig. 12.4 Same patient. Postoperative radiograph shows lized for the near total sacrectomy. Posteriorly
the resection performed with navigation (longitudinal based iliac osteotomies were completed, and the
split of the sacrum and acetabular sparing) and recon- tumor was dissected from the pelvic contents and
struction with pedicled screws, rods, and allograft
resected en bloc. A humeral allograft was utilized
to span the defect from right sacral remnant to
sion over the gluteus maximus toward the mid- left iliac wing to reconstruct the pelvic ring.
portion of the right thigh was undertaken. The Pedicle screws, iliac bolts, and multiple rods
sciatic nerve was dissected proximally to the sci- were utilized to reconstruct and support the spi-
atic notch. Laminectomy of the lumbosacral nopelvic junction (Fig. 12.6). The patient has
junction was completed, and utilizing navigation, partial paresis in the sciatic distribution of the left
posterior osteotomy to join the anterior osteot- lower extremity and continues his rehabilitation
omy made in stage 1 was completed. The sacrum 9-month follow-up.
was split longitudinally and then to the right at
the L5–S1 disk space. The ilium was freed of
muscular attachments on the left, and a supra- 12.6 Pelvic Allograft
acetabular osteotomy was completed. The left Reconstruction
hemisacrum, left iliac wing, and tumor were then
resected en bloc. A humeral shaft allograft was Wide resection of primary malignant tumors of
placed from the low lumbar spine to the supra- the pelvis coupled with limb-sparing surgery
acetabular pelvis. Pedicle screws were placed when possible is the preferred treatment of our
into the lumbar spine, left ilium, and supra- group. Although there is no unanimous opinion
acetabular pelvis on the right side. Multiple rods regarding the ideal reconstruction method, pelvic
and connecters were utilized to reconstruct the allograft is a viable option for reconstruction. An
pelvic ring (Fig. 12.4). The patient is ambulatory advantage of allograft reconstruction is that it can
with preserved bowel and bowel function at be shaped and sized in order to match the postre-
20-month follow-up. section defect. Allograft use also provides the
advantage of patient bone stock preservation
[13]. Furthermore, use of an allograft may allow
12.5.2 Case 3 the patient to avoid a flail hip or arthrodesis.
Pelvic allograft accommodates anatomic recon-
The patient is a 50-year-old male who presented struction of pelvic bony architecture as well as
with low back pain and difficulty with bowel move- allows for multiple hip preservation options [14].
ments. Advanced imaging revealed large expansile Patients report a high rate of functionality with
128 H. Park et al.
Fig. 12.5 Grade 1 chondrosarcoma of the sacrum with posterior soft tissue extension in a 50-year-old male. Sagittal
and axial MRI show the typical characteristics and growth pattern of the tumor
Yet, as is the case with most pelvis recon-

structions options, pelvic allograft reconstruc-
tions are associated with significant rates of
infection and mechanical failure. Infection rates
who underwent allograft reconstruction after
pelvic resection or internal hemipelvectomy
range from 12.5% to 20%, respectively [14, 16].
Given the high baseline risk of infection, history
of persistent infection or immunosuppressed
state should preclude use of allograft [17].
Sciatic and/or femoral nerve palsies can occur
in as many as 25% of patients; however, this is
most common in the setting of periacetabular
resections. Most of these palsies are complete
and partially resolve [14]. Hip instability is a
known complication of allograft reconstruction.
Patients must be followed for fracture and non-
union in the setting of allograft use or irradiated
bone. In a series of 24 patients who underwent
pelvic allograft reconstruction after tumor
resection, 12% of the cohort developed non-
Fig. 12.6 Same patient. En bloc tumor resection required
union. Two of the three nonunions in the series
total sacrectomy extended to the ilium. Postoperative occurred at the site of fixation of the allograft to
radiograph shows pedicle screws, iliac bolts, and multiple the ilium. Rates of allograft fracture have been
rods that were utilized to reconstruct and support the spi- reported to range between 0% and 21% [14, 18,
nopelvic junction
19]. Due to the many possible complications
that the patient must be monitored for long-
regard to pain and ambulation [14, 15]. Children term, the social situation of the patient must be
and adolescents achieve substantially increased considered before opting for allograft
functionality compared to adults [14]. reconstruction.
12.6.1 Case 4
The patient is a 57-year-old male with history of

right pelvic liposarcoma status post chemoradia-
tion, resection, and right total hip arthroplasty
presenting with worsening right hip pain and dif-
ficulty ambulating. Physical exam was notable
for mild tenderness to palpation over the right
iliac wing.
MRI of the right hip demonstrated marrow
replacement destruction of the iliac crest without
associated soft tissue mass and pathologic frac-
ture of the right ilium, suggestive of osteoradio-
necrosis. Radiation-induced sarcoma was
suspected given the patient’s history of radiation
in the area. CT of the abdomen and pelvis dem-
onstrated diffuse sclerotic appearance of the right
iliac wing concerning for neoplastic infiltration
(Fig. 12.7). Furthermore, PET scan revealed Fig. 12.7 Radiation-induced sarcoma of the right hemi-
pelvis in a 57-year-old male, with a total hip arthroplasty
avidity in the right iliac wing. Subsequent
CT-guided biopsy was diagnostic for radiation-
associated sarcoma. The patient was scheduled to
undergo resection of the pelvic sarcoma in a
staged manner.
12.6.1.1 Stage 1
In the first stage of this staged resection, the
patient was positioned prone and a posterior inci-
sion was made longitudinally in line with the
lumbar spine; then, dissection was carried out to
the right toward the right buttock and sciatic
nerve. With guidance from intraoperative naviga-
tion, an osteotomy was made just medial to the
right sacroiliac joint. Pedicle screws were placed
at L3 and L4 as well as two large iliac bolts in
anticipation of the spinopelvic reconstruction. An
L5 osteotomy was performed. Further bone wax
was placed in the defect, and rods were used to
span the defect, positioned strategically for sub-
sequent anterior reconstruction (Fig. 12.8).
12.6.1.2 Stage 2 Fig. 12.8 Same patient. First surgical stage with poste-
rior approach. Pedicle screws were placed at L3 and L4 as
Five days following stage 1, the patient under-
well as two large iliac bolts in anticipation of the spinopel-
went the second stage. The patient was positioned vic reconstruction
in lateral decubitus, and a curvilinear incision
was made extending from the posterior superior
iliac spine toward the anterior iliac spine and tables of the pelvis. Tissue along the iliac crest
toward the knee. The tumor was situated within and the hip abductors was detached with the
the ilium extending out from the inner and outer mass.
130 H. Park et al.
Using intraoperative navigation, pelvic oste- 12.7 Proximal Femoral

otomies were performed. The osteotomy was Reconstruction
created from the superior ramus and then
through the inferior ramus toward the ischium, Wide resection of a tumor in the periacetabular
distal to the prior acetabular cup from the prior region requires en bloc resection of proximal femur
hip reconstruction. Further dissection was per- with subsequent complex reconstruction.
formed proximally along the anterior sacrum, Periacetabular resection without reconstruction will
and an osteotomy was completed in line with likely result in instability. Reconstruction can com-
the prior posterior osteotomy in the first stage. prise of allograft arthrodesis, intercalary allografts,
The tumor was resected en bloc and sent to endoprostheses, and allograft–endoprosthesis com-
pathology for analysis. Negative margins were posites. Although allograft techniques have an
obtained. advantage of possible bony host-to-allograft incor-
A pelvic allograft was fashioned to fit the poration, they also come with the significant risks of
defect from the right sacroiliac joint to the dis- nonunion, fracture, and infection. Endoprostheses,
tal right ischium. The plastic surgery service on the other hand, are technically simpler recon-
had concurrently harvested a free vascularized structions and provided the added benefit of shorter
fibular graft with its associated vascular pedi- time to weight-bearing [20–23]. While endopros-
cle. The graft was fit and placed on the inner theses share some of disadvantages of allografts,
aspect of the allograft, held in place with they also possess the risk of instability [13].
screws and abutting the pelvic graft. Finally, In a series of 137 patients who underwent
the hip capsule was entered, and the proximal proximal femoral reconstruction, a difference in
femoral component was removed. Proximal outcomes was found to exist between osteoarticu-
femoral replacement was performed with a lar allografts and all other allograft reconstruc-
bipolar prosthesis into the allograft. The recon- tions. Osteoarticular allografts were significantly
struction defects were covered with a vastus less likely to have a positive result than allograft–
lateralis flap. Final reconstruction is shown in endoprosthesis composites. Development of
Fig. 12.9. osteoarthritis in patients with osteoarticular
allografts led to a 39% rate of total joint replace-
ment [13]. Other studies in other anatomic sites
confirm that patients with allograft-endoprosthetic
reconstruction have superior functional results
compared to osteoarticular allografts [24–26].
Nonetheless, infection remains a serious compli-
cation regardless of reconstruction type. Wide
resection, multiple surgeries, avascular allograft,
and neoadjuvant chemoradiation all lead to a high
baseline likelihood of infection in these oncologic
surgeries [13]. Massive alloprosthetic reconstruc-
tion using allograft bone combined with arthro-
plasty may preserve limb length and maximize
function, particularly in young patients [18, 27].
12.7.1 Case 5
The patient is a 37-year-old male with prior diag-

Fig. 12.9 Same patient. Final reconstruction using a pel- nosis of right pelvis undifferentiated pleomorphic
vic allograft sarcoma status post-neoadjuvant chemoradiation
tions obtained were negative for tumor, including

paraspinal musculature margins. The mass was
kept intact without violating the capsule. The
right-sided sacral nerve roots were ligated with
silk ligatures and sacrificed. A construct was
placed connecting the left L4 and L5 pedicle
screws to the left iliac bolts. This was connected
to the right L4 and L5 pedicle screws via cross
connectors. Closure was performed, and the
patient was transferred to the intensive care unit
postoperatively.
Fig. 12.10 Huge sarcoma of the hemipelvis in a 37-year-
old male. Axial CT scan shows the aggressiveness of the 12.7.1.2 Stage 2
tumor with soft tissue extension within the pelvis and in Two days after the first stage, the patient returned
the posterior area to the operating room for the second stage. A cur-
vilinear incision was made from the mid-back
with worsening right hip pain and right lower extending along the iliac crest to the distal thigh.
extremity weakness with symptoms consistent A second limb of the incision provided anterior
with sciatica. The patient was found to have a exposure of the distal vastus lateralis tendon,
large fungating soft tissue mass consistent with which was cut distally. The dissection was car-
known sarcoma. ried from distally to proximally. This was diffi-
CT of the pelvis demonstrated a large soft tis- cult to perform due to scar tissue from prior
sue mass centered about and infiltrating the R radiation therapy. The tumor was adherent to the
iliac crest, with extension across the right sacro- sciatic nerve, and it was thus difficult to obtain a
iliac joint and into the sacrum. The lesion involves wide margin in this area. Nonetheless, margins
the right S2, S3, and S4 neural foramina. Within were negative for tumor in frozen sections.
the pelvis, it abuts the posterior sciatic nerve After completion of soft tissue dissection
before it enters the sciatic foramen (Fig. 12.10). around the tumor, osteotomies were performed
The patient was scheduled for staged resection of with the aid of intraoperative navigation. The
the pelvic mass. femoral neck was cut with an oscillating saw in
order to better expose the acetabulum prior to its
12.7.1.1 Stage 1 osteotomy. An osteotomy then was made through
The patient was positioned prone, and a midline the acetabulum to partially reflect the tumor mass
incision was made from the mid-lumbar spine to in order to better expose the sciatic nerve and
the coccyx. The incision was ellipsed over the tumor pseudocapsule interface. Further dissection
right soft tissue mass. Pedicle screws were placed was carried out releasing further in the proximal
bilaterally at L4 and L5. Decompression was per- sacrum as well as distal L5. The prior osteotomy
formed from L4/5 to the distal sacrum and was connected through where the sacral osteot-
coccyx. omy was performed in stage 1. The tumor was
Using intraoperative navigation, an osteotomy removed en bloc.
was carried through the distal sacrum toward the After the wound was irrigated thoroughly,
left side. The piriformis was identified, and blunt attention was turned to reconstruction. An
dissection was performed over the anterior aspect allograft pelvis was sized to fit into the large
of the sacrum. Further dissection was carried out osseous defect. The pelvic graft was anchored in
around the mass on the right side through the glu- place with pedicle screws and wedged into the
teus maximus muscle primarily. One dural defect remaining acetabulum. With the aid of intraoper-
was noted in an area of scar tissue adherent to the ative navigation, the graft was reamed sequen-
dura and concerning for tumor. All frozen sec- tially in order to reconstruct the acetabulum. A
132 H. Park et al.
defining soft tissue planes even more challenging

due to distortion of normal anatomy by scar.
Local recurrence rates are alarmingly common in
cases of marginal resection and nearly 100%
after intralesional resection; it is thus of the
utmost importance to achieve negative margins.
Standard resection techniques have been found to
result in intralesional resection rate of 29% [28].
Even the most experienced surgeons have diffi-
culty replicating a proposed osteotomy more than
50% of the time in a sawbones model [29].
Navigation-assisted resection may be useful in
this situation, although it may add time and cost to
an already complex and expensive procedure.
Computer-aided navigation has become increas-
ingly used in musculoskeletal tumor surgery [30,
31]. Navigation technology has been used with
success in other surgical disciplines including
Fig. 12.11 Same patient. Final reconstruction after en neurosurgery, urology, spinal surgery, otolaryn-
bloc resection gology, orthopedic trauma, and arthroplasty [32].
Navigation requires preoperative imaging that is
hemispherical shell cup was inserted and then integrated by software to help develop a pre-
anchored with screws into both native and operative plan. CT provides bony detail, MRI pro-
allograft pelvis. The proximal femur was vides soft tissue detail, and PET-CT provides
broached and reamed. A modular mobile bearing excellent discrimination between tumor and non-
construct was employed. A constraining liner tumor tissue. Intraoperative navigation is based
was not chosen since anchorage into the pelvis on overlying the preoperative imaging onto fixed
would be more than 50% allograft bone. Further, anatomic landmarks (e.g., anterior superior iliac
screws were placed into the allograft and host spine) to provide proposed bone cuts.
bone. The sacral hardware was joined to the con- In a series of 24 patients with primary tumors
struct. An A-like frame was constructed to allow of the pelvis or sacrum, computer-aided naviga-
for good stability. Final structural reconstruction tion resulted in negative bony margins in all
is shown in Fig. 12.11. patients and negative soft tissue margins in 91%
With the plastic surgery service, soft tissue of cases [33]. Similarly, in a series of 31 patients
reconstruction was performed using a pedicled with pelvic tumors, there was a reduction in intra-
vastus lateralis flap. A lateral hamstring flap was lesional resection rates from 29% to 8.7% using
also reflected into the defect, and the defect was intraoperative navigation [34].
closed completely. The patient was transferred
postoperatively to the intensive care unit.
12.9 Conclusion
12.8 Intraoperative Navigation With advances in our understanding of anatomy,

imaging, surgical technique, and implant devel-
One of the greatest challenges when resecting opment, pelvic reconstruction in the setting of
malignant sacropelvic tumors is achieving nega- oncologic resection has evolved from a morbid
tive tumor margins. The complex anatomy of this procedure to one that is relatively safe and effec-
region makes adequate resection difficult. tive. The Enneking and Dunham classification
Furthermore, neoadjuvant radiation may make organizes the breadth of pelvic resections into
specific anatomic locations of resection; recon- 9. Bederman S, Shah KN, Hasan JM, Hoang BH, Kiester
PD, Bhatia NN. Surgical techniques for spinopelvic
struction may be favored in the setting of pelvic reconstruction following total sacrectomy: a systemic
ring disruption. The goal of pelvic reconstruction review. Eur Spine J. 2014;23(2):305–19.
is to provide continuity between the lower 10. Wei R, Guo W, Ji T, Zhang Y, Liang H. One-step
extremity and the axial skeleton in order to maxi- reconstruction with a 3D-printed, custom-made pros-
thesis after total en bloc sacrectomy: a technical note.
mize function. Eur Spine J. 2017;26(7):1902–9.
Spinopelvic reconstruction is often required in 11. McLoughlin GS, Sciubba DM, Suk I, Witham T,

the setting of pelvic resections that extend to the Bydon A, Gokaslan ZL, et al. En bloc total sacrec-
sacrum or lumbar spine. Sacral reconstruction is tomy performed in a single stage through a pos-
terior approach. Neurosurgery. 2008;63(1 Suppl
relatively rare and generally performed after 1):ONS115–20; discussion ONS120.
sacrectomy for chordoma, chondrosarcoma, 12.
Zileli M, Hoscoskun C, Brastianos P, Sabah
giant cell tumor, or osteosarcoma. Wide resection D. Surgical treatment of primary sacral tumors: com-
of a tumor in the periacetabular region requires plications associated with sacrectomy. Neurosurg
Focus. 2003;15:E9.
en bloc resection of proximal femur with subse- 13. Fox EJ, Hau MA, Gebhardt MC, Hornicek FJ,

quent complex proximal femoral resection. Bony Tomford WW, Mankin HJ. Long-term followup of
reconstruction can be performed with pelvic proximal femoral allografts. Clin Orthop Relat Res.
allograft or autograft with employment of pedicle 2002;397:106–13.
14. Delloye C, Banse X, Brichard B, Docquier P, Cornu
screws, plates, screws, and/or rods. Intraoperative O. Pelvic reconstruction with a structural pelvic
navigation technology can be used to provide allograft after resection of malignant bone tumor. J
real-time intraoperative feedback with increased Bone Joint Surg Am. 2017;89-A(3):579–87.
likelihood of negative bony margins. Finally, soft 15. Karim SM, Colman MW, Lozano-Calderon SA,

Raskin KA, Schwab JH, Hornicek FJ. What are the
tissue coverage of the resulting spinopelvic functional results and complications from allograft
defect is of the utmost importance to avoid reconstruction after partial hemipelvectomy of the
wound-related complications. Rotational flaps pubis? Clin Orthop Relat Res. 2015;473:1442–8.
such as the pedicled vastus lateralis flap may pro- 16. Mankin HJ, Hornicek FJ. Internal hemipelvectomy
for the management of pelvic sarcomas. Surg Oncol
vide adequate coverage. Clin N Am. 2005;14:381–96.
17. Hugate R, Sim FH. Pelvic reconstruction techniques.
Orthop Clin N Am. 2006;37(1):85–97.
18. Harrington KD. The use of hemipelvic allografts or
References autoclaved grafts for reconstruction after wide resec-
tion of malignant bone tumors of the pelvis. J Bone
1. Gordon-Taylor G, Wiles P. Interinnomino-abdominal Joint Surg Am. 1992;74:331–41.
(hind-quarter) amputation. Clin Orthop Relat Res. 19. Langlais F, Lambotte JC, Thomazeau H. Long-term
1967;53:3–11. results of hemipelvis reconstruction with allografts.
2. Ch G. Désarticulation de l’os iliaque pour sarcoma. Clin Orthop Relat Res. 2001;388:178–86.
Rev Chir. 1895;15:952. 20.
Bickels J, Meller I, Henshaw RM, Malawer
3. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, MM. Reconstruction of hip stability after proxi-
2010. CA Cancer J Clin. 2010;60(5):277–300. mal and total femur resections. Clin Orthop.
4. Enneking WF, Dunham WK. Resection and recon- 2000;375:218–30.
struction for primary neoplasms involving the innomi- 21. Freedman EL, Eckhardt JJ. A modular endoprosthetic
nate bone. J Bone Joint Surg Am. 1978;60(6):731–46. system for tumor and non-tumor reconstructions: pre-
5. Biermann JS, Adkins DR, Benjamin RS, Brigman liminary experience. Orthopedics. 1997;20:27–35.
B, Chow W, Conrad EU, et al. Bone cancer. J Natl 22. Kabukcuoglu Y, Grimer RJ, Tillman RM, Carter

Compr Cancer Netw. 2010;8(6):688–712. SR. Endoprosthetic replacement for primary malig-
6. Bloem JL, Reidsma II. Bone and soft tissue tumors of nant tumors of the proximal femur. Clin Orthop.
hip and pelvis. Eur J Radiol. 2012;81(12):3793–801. 1999;358:8–14.
7. Benevenia J, Cyran FP, Biermann JS, Patterson FR, 23. Zehr RJ, Enneking WF, Scarborough MT. Allograft-
Leeson MC. Treatment of advanced metastatic lesions prosthetic composite versus megaprosthesis in
of the acetabulum using the saddle prosthesis. Clin proximal femoral reconstruction. Clin Orthop.
Orthop Relat Res. 2004;426:23–31. 1996;322:207–23.
8. Damron TA, Sim FH. Surgical treatment for meta- 24. Henja MF, Gitelis S. Allograft prosthetic composite
static disease of the pelvis and the proximal end of the reconstruction for bone tumors. Semin Surg Oncol.
femur. Instr Course Lect. 2000;49:461–70. 1997;13:18–24.
134 H. Park et al.
25. Mankin HJ, Gebhardt MC, Jennings CJ, Springfield struction within the pelvis: an experimental study.
DS, Tomford WW. Long-term results of allograft Acta Orthop. 2008;79:695–702.
replacement in the management of bone tumors. Clin 30. Hufner T, Kfuri M, Galanski M, Bastian L, Loss

Orthop. 1996;324:86–97. M, Pohlemann T, Krettek C. New indications for
26.
McGoveran B, Davis AM, Gross AE, Bell computer-assisted surgery: tumor resection in the pel-
RS. Evaluation of the allograft-prosthesis compos- vis. Clin Orthop Relat Res. 2004;426:219–25.
ite technique for proximal femoral reconstruction 31. Krettek C, Geerling J, Bastian L, Citak M, Rucker F,
after resection of a primary bone tumor. Can J Surg. Kendoff D, Hufner T. Computer aided tumor resection
1999;42:37–45. in the pelvis. Injury. 2004;35(Suppl 1):S-A79–83.
27. Beadel GP, McLaughlin CE, Wunder JS, Griffin
32. Thadani VN, Riaz MJ, Singh G. The evolution of
AM, Ferguson PC, Bell RS. Outcome in two groups three-dimensional technology in musculoskeletal
of patients with allograft-prosthetic reconstruc- oncology. J Clin Orthop Trauma. 2018;9(3):269–74.
tion of pelvic tumor defects. Clin Orthop Relat Res. 33. Abraham JA, Kenneally B, Amer K, Geller DS. Can
2005;438:30–5. navigation-assisted surgery help achieve negative
28. Jeys L, Grimer R, Carter S, Tillman R, Abudu
margins in resection of pelvic and sacral tumors? Clin
S. Outcomes of primary bone tumours of the pel- Orthop Relat Res. 2018;476:499–508.
vis: the ROH experience. J Bone Joint Surg Br. 34. Jeys L, et al. Can computer navigation-assisted

2012;94(Suppl):39. surgery reduce the risk of an intralesional margin
29. Cartiaux O, Docquier PL, Paul L, Francq BG,
and reduce the rate of local recurrence in patients
Cornu OH, Delloye C, Rlucent B, Dehez B, Banse with a tumor of the pelvis or sacrum? Bone Joint J.
X. Surgical inaccuracy of tumor resection and recon- 2013;95B(10):1417–24.
Navigation in Pelvic Surgery
13
David M. Joyce
13.1 Introduction 13.2 W

hy Pelvic Computer
Navigation
Computer navigation for bony pelvic oncology
cases was introduced as a guidance tool in the Prior to computer navigation the thought of limb sal-
early 2000s [1, 2]. Prior to this, the only option for vage, maintained function and decreased morbidity
intraoperative guidance in pelvic bone resection of were almost impossible without sacrificing one of
tumors was 2-dimensional (2D) fluoroscopy typi- the former to obtain adequate surgical margins [14].
cally used in fracture fixation cases. This limited The complex 3-dimensional nature of the pelvis
options for resection types and affected recon- often prevents direct line of sight, with no way to
structive options. Navigation began its incorpora- place a retractor to prevent inadvertent tumor viola-
tion into oncology because those with experience tion. The difficulty is only magnified by structural
in pelvic oncologic surgery felt a frustration with alterations caused by tumor, neoadjuvant treatment,
the results of the standard resection used for sur- and rarity contributing to lack of experience on part
gery [3–10]. Malignant tumors in the pelvis that of surgeon in conventional pelvic surgery which led
involve pelvic bone often have large soft tissue to larger bone resections with surgeons opting for
components that compromise visualization [11]. the more morbid procedure out of the need to obtain
Computer navigation was specifically introduced the best oncologic margins and hence the best onco-
in the pelvis to help in identifying the complete logic outcome with little concern for function [14].
complex 3-dimensional extent of the malignant Looking back at pelvic oncology experience involv-
neoplasm involving both soft tissue and bone and ing sarcoma resections in the pelvis, there are higher
safely assist in executing a local wide excision positive margin rates and higher local recurrence
[12]. Navigation has improved the visualization of rates compared to sarcomas in the extremities con-
the surgical field through a virtual 3D reconstruc- firming the difficulty in obtaining negative margins
tion allowing for precise osteotomy resection lev- in pelvic sarcoma surgery [2, 15] which is partly due
els [11, 13]. This section is dedicated to advanced to inherent complexity of the pelvis. The pelvic
imaging (computed tomography (CT) and mag- complexity and added inaccuracy with free-hand
netic resonance imaging (MRI)) computer naviga- cuts have led to positive surgical margins in the
tion of the bony pelvis in pelvic oncology. majority of cases, with local recurrence approaching
rates of 70–80% when using conventional methods
involving fluoroscopy, visual, and tactile references
D. M. Joyce (*)
[3–6, 11, 16–18]. In practice, surgeons performing
Sarcoma Department Moffitt Cancer Center,
Tampa, FL, USA conventional pelvic surgery have to plan at least a
e-mail: [email protected] 2 cm margins in order to be assured no tumor viola-
136 D. M. Joyce
tion [8]. The purpose of computer navigation in the interpretation of preoperative imaging on MRI and
musculoskeletal oncology setting is to plan and CT and then properly correlates to bony landmarks
assist in executing a local wide excision, to reduce intraoperatively on the patient’s pelvis during sur-
and eliminate the rate of intralesional resection with gery. This relationship between what is seen on pre-
preservation of vital structures as well as mechanical operative imaging and what is identified in the
structures. Navigation allows for accurate identifica- patient can lead to errors that translate into positive
tion of the local bony anatomy and define the extent margins in the noncomputer-assisted surgery.
of the tumor which helps the surgeon better under- Navigation allows a surgeon to appreciate the extent
stand the relationship between the tumor and native of the tumor infiltration in the bone on the computer
structures in a distorted environment [19–23]. This navigation monitor and know that their tool is cor-
technology aids a surgeon by identifying critical rectly located on the planned resection plane avoid-
structures, improve margins, and preserve structure ing tumor. One of the main goals of computer
and theoretically function [24]. It allows for real- navigation is to help eliminate positive margins
time visual feedback that can be seen on a monitor hopefully leading to more disease-free patients. The
intraoperatively and augments and enhances the sur- other benefit the navigation is the accuracy of the
geon’s tactile perception of the surgical field. cuts because with accurate 3D navigation you now
Primary pelvic tumors often have an intraosseous have a very precise cut that allows the surgeon to
component making it nearly impossible to appreci- take advantage of precision 3D printing (Fig. 13.1)
ate extent of tumor infiltration intraoperatively [23]. now being used in both models and custom 3D
In conventional pelvic resection, infiltration of the printed anatomic reconstructions for use in the pel-
tumor within the bone marrow cannot be seen intra- vis (Fig. 13.2). Improved accuracy through this
operatively and requires a reliance of the surgeon’s modality is one of the primary benefits and the
Fig. 13.1 3D virtual printing used for planning after resection lines are placed on 3D printed model
13 Navigation in Pelvic Surgery 137
Fig. 13.2 Custom 3D printed anatomic reconstruction
ability to get closer to the tumor if needed to save

critical structures without compromising margins.
While pelvic oncology surgery first principle is to
remove all the tumor with a negative margin,
using computer navigation one can remove less
normal tissue which aids in reconstruction with-
out increasing the risk of inadvertent tumor viola-
tion [21].
13.3 History of Navigation
While computer navigational surgery in pelvic

Fig. 13.3 Optical Tracking System that can be moved
oncology is relatively new, computer-assisted
around to improve “line of sight”
surgery has been used in other surgical fields
such as neurosurgery for cranial tumor resec-
tions and biopsies for at least the last 25 years
[25, 26]. Several different computer navigation 13.4 Computer Navigation
systems exist in the commercial world consist- Process Overview
ing of imageless navigation, fluoroscopic navi-
gation, and advanced imaging navigation [27]. There are required components to computer naviga-
Adoption of computer navigation in recon- tion in the setting of the pelvis [22, 34] including an
structive surgery was born out of a need for intraoperative computer platform loaded with com-
more accurate component placement and most puter navigation software and tracking system. A
navigation was developed for spine, trauma or patient’s tracker or dynamic reference base (DRB),
hip and knee surgery for accuracy of hardware which can either consist of optical trackers (consist-
placement within 1 mm of the desired location ing of three OCD cameras (Figs. 13.3 and 13.4)) or
[1, 26, 28–33]. In 1997, it was described in use electromagnetic trackers can track the instruments
for periacetabular osteotomies and thus may relative to the patient in space [35–39]. An intraop-
have served as the foundation for using it in erative registration process in which the patient’s
pelvic oncology [26, 33]. Possibly, the first use preoperative imaging and patient’s intraoperative
of computer- navigated chisels for a pelvic position is matched and creates a referencing sys-
tumor was reported in 2004 to perform an oste- tem that allows the patient and tools to be tracked in
otomy for a sacral tumor [2]. space with respect to each other. The registration
138 D. M. Joyce
quately identify appropriate margins on imaging

will dictate which imaging is needed preopera-
tively. CT is ideal for cortical bone resection
planning, but MRI is essential to determine mar-
row infiltration and true bony involvement [46,
47] and provides the best method for defining
marrow involvement for bone malignancy and
soft tissue malignancy when planning for resec-
tion margins [29]. MRI and CT (if cortical
destruction) to a point have a high sensitivity and
specificity for bony tumor infiltration, but mar-
gins are difficult to identify intraoperatively, and
intralesional resection is very much possible in
the pelvis [7, 19] (Fig. 13.5). A surgeon can mea-
sure on MRI the marrow infiltration and then rec-
reate this measurement on the CT used in
computer navigation as long as a similar slice
thickness is used or it can be augmented or
“fused” with a MRI for the benefit of identifying
marrow or periosteal edema then uploaded to the
navigation system [23, 29, 46]. In general, when
performing surgery in the pelvis using computer
Fig. 13.4 Patient Tracker or Dynamic Reference Base
(DRB) attached to iliac wing and covered with sterile bag navigation, the surgeon may only use the CT
imaging, although some will use a CT-MRI
fusion [12, 18, 29, 46]. Some users feel the
process can involve intraoperative CT, 3D fluoro- CT-MRI fusion is beneficial to the surgical pro-
scopic, imageless, kinematic, bone morphing, or cedure by being able to identify marrow margins
“pair matching” with bony landmarks or fiducials, or margins outside peritumoral edema [29, 34,
surface mapping, and ultrasound referencing [27, 46]. While one does not have to take advantage of
40, 41]. Navigation of the pelvis cannot be per- CT-MRI fusion, this modality is probably the
formed with imageless referencing because this best way to look at soft tissue when using com-
mode uses the pelvic plane and bony landmarks for puter navigation. CT-MRI fusion allows for
reference, and soft tissue on the pelvis and tumor determining the extent of tumor resection planes
render this method fairly inaccurate [42–44] requir- based off the bony involvement seen on the navi-
ing all pelvic oncology navigation be based on gated software [48]. Currently, surgeons not
advanced imaging such as MRI and CT [34]. using navigation rely on their interpretation of
tumor location on CT and MRI obtained preop-
eratively and then attempt to translate that inter-
13.5 Preoperative Planning pretation into pelvic landmarks during the
intraoperative resection leading to significant
The use of computer navigation starts with inaccuracies in tumor resection [20, 23]. Planning
obtaining advanced 3D imaging of the pelvis and for resection can be done on the computer navi-
tumor in question. At a minimum, a CT scan of gation system prior to surgery or can be used as a
the affected bone must be obtained with way to augment the visual representation of the
0.5–2 mm cuts, continuous with no overlap that surgeon. If the surgeon elects to plan resection
covers the entire compartment, for the pelvis that planes prior to surgery, this virtual planning
includes both innominate bones and sacrum [2, comes at the time when a CT and/or MRI scan is
20, 22, 23, 45]. Understanding how best to ade- imported into the navigational software prior to
Fig. 13.5 MRI vs. CT scan for identifying marrow infiltration, CT when compared to MRI under appreciates the extent
of tumor in the intraosseous bone
surgery and the surgeon elects planes to deter- obtained. The choice based on the technology
mine the starting point and vector of the intended available to surgeons with the options includes
plane for future osteotomy or resection plane surface matching, paired-point matching with
(Fig. 13.6). One can create planes that represent or without fiducial marker, or creating a intra-
resection levels and osteotomy sites. One can operative CT scan of a segment of bone to
also paint the tumor and provide the user with match the preoperative imaging. During com-
collision warnings to minimize that chance of puter-assisted surgery, the surgeon must appre-
getting into tumor (Fig. 13.7). One can create ciate the importance of DRB placement. Certain
multiple planes to allow for cuts than are multi- types of DRBs can have issues with line of
planar (Fig. 13.8). sight such as optical trackers, and it is impor-
tant to think about surgeon’s approach and
patient’s positioning when using navigated
13.6 Intraoperative Computer tools [49]. Some navigation systems use elec-
Navigation tromagnetic trackers for the DRB and the
instrument and will not have the same line of
The intraoperative process consists of tracker sight problem that optical trackers with infrared
or dynamic reference base (DRB) placement on sensors do; however, they can be prone to elec-
the patient, registration, and then resection. tromagnetic interference [36–39, 50]. No track-
Registration of the patient to the system occurs ing system is perfect, and each has their inherent
after the patient is asleep and draped into a ster- problems [37, 50–52]. However, one study was
ile field and can occur either at the beginning able to minimize registration error, in the set-
prior to major surgical incision or at the time ting of fiducial marker use, by placing the DRB
prior to bone resection after bone exposure is as far away from the fiducial markers and resec-
140 D. M. Joyce
Fig. 13.6 Three planes were created in the virtual planning area to identify the pubic symphysis (yellow), intended
resection plane (green) and tumor margin (pink)
tion zone as possible [53]. The DRB can be pelvis away from the surgical site to minimize
placed far enough away from the surgical site interference and compromise to the registration
so to minimize possible disturbances during [53]. One can place pins for the tracker in the
surgery to the tracker and system; i.e., leaning same side iliac wing as long as the pins our out-
on the tracker with a retractor or hand, the sur- side the peritumoral edema zone and outside
geon should place it in a relatively safe but the resection zone [55]. Only solid pins with
visualized area. Some authors have recom- good fixation should be accepted because any
mended the tracker be placed on the contralat- change in tracker position will affect your accu-
eral side of the pelvis and when surface racy. The DRB must be connected to the bone
matching to create a large enough exposure to of an intact pelvic ring for the planned resection
allow for all the working area of bone to be or osteotomy because ligament resection can
included in the sampling surface area [53]. allow bones to move independently and will
While other authors have expressed concern affect the accuracy of the bone interface.
that the further the tracker is placed away from Placing three pins better stabilizes the tracker
the resection plane the more room there is for on the subjects body [56]. Tracker pins that
play in the system through stretching and loosen corrupt the registration accuracy and the
stressing soft tissue connections in the pelvis registration process will be required to be per-
[51, 54]. Pins are placed into a stable part of the formed again [2].
Fig. 13.7 Tumor painted in yellow
13.7 Registration 27, 34, 48]. This requires the surgeon to a ccurately
identify both on the CT or MRI image and on the
All advanced imaging-based navigation systems patient’s pelvis the corresponding points. Often,
require a registration process prior to navigated the pubic symphysis, ASIS, AIIS, pubic tubercle,
surgical resection [57]. Registration is the most and PSIS are used for patient to image matching
significant and error-prone step in navigation due and confirmation [20, 53]. However, there are
to the fact the surgeon variation can be intro- specific bony landmarks often that cannot be
duced into the system [58]. This occurs after a identified or palpated due to patient positioning.
DRB has been placed in a position that will be Surface mapping has also been used where one
stable throughout surgery [2, 24]. This can be can select 50–65 points on the bony pelvis to reg-
done several different ways and is mostly based ister the patient [24]. The difficulty with surface
on the type of computer navigation system avail- mapping is soft tissue components such as carti-
able. If no advanced intraoperative imaging is lage, ligament, and soft tissue which will inter-
available registration, then paired-point matching fere due to pliability; thus, bone cortex is used
is primarily used and requires a minimum of four due to its consistency and ease of ability to be
points or “paired points” that are chosen on the defined on a CT scan [48]. MRI is not typically
patient’s preoperative MRI or CT image and then used for surface registration due to the system
identified on patient’s bony anatomy [22, 24, 25, having difficulty identifying cortex on MRI, and
142 D. M. Joyce
Fig. 13.8 Multiplanar cut with all planes identified at once
thus, CT is the preferred modality when using tion process as a way for quicker and more reli-
surface mapping [12, 29]. Both of these processes ably accurate paired-point registration with
have inaccuracies of the registration, thus trans- minimization of the registration error, i.e., the
lating into errors in resection location; thus, there difference between the preoperative imaging and
was a demand for decreasing registration error, the actual patient bony anatomy. Fiducials allow
i.e., the difference between the picked point on for better paired-point registration than would be
the preoperative CT and the correct point in the obtained with bony landmarks alone. These
patient. Patients with significant body habitus or implants serve as fiducials for pair-point match-
BMI [22] will limit exposure and make it difficult ing during patient registration at the time of sur-
to accurately identify bony landmarks for paired- gical resection [2]. Fiducial markers can be
point matching. Planned registration landmarks placed in the pelvis, but an additional surgery to
whether surface features or fiducials should not place these markers must occur prior to the pre-
involve osseous features that may be included in operative CT [45, 57]. Fiducial positioning is
the resection [29], and when using either paired- chosen based on ease of access during surgery
point or surface matching, the bone and/or the such as the iliac crests, ASIS, AIIS, and posterior
fiducials must be adequately exposed to perform iliac spines [2, 18]. No more than four fiducial
the registration. markers are needed for an accuracy of 1.5 mm
Fiducial markers or small implants such as a [59], but they need to be accessible in both the
screw or pin were added to the computer naviga- prone and supine position if the patient needs two
approaches. With CT, small titanium fiducial to registration errors because merging is still done
markers such as a Kirschner wire [12] or screw visually by the surgeon leading to potential error
are the preferred implant as stainless steel will even with the best processes [29, 64].
have too much metal artifact to identify a 1 mm
point of registration. Titanium screws or K-wires
chosen as fiducial markers do not allow for the 13.8 Registration Accuracy
MRI to be used in preoperative imaging for navi-
gation. Bioabsorbable fiducial markers are Registration error is a way to determine accuracy
required when MRI is used in paired-point regis- of the “match” can be reported in some systems
tration in order to obtain a registration error of based on the method used to match the patient to
<1 mm [18] because of metal artifact. the preoperative CT scan or CT/MRI fusion scan
Bioabsorbable nonmetallic 1.5-mm pins placed [29, 46]. Registration error can be calculated by
beyond the tumor resection can act as fiducials so the navigation software and gives an indication of
that CT-MRI fusion and MRI images alone can the mismatch and provides a value of difference
be used for patient registration [29]. When using between the point picked on the preoperative
fiducial markers, the slice thickness of either the advanced image on the intraoperative monitor and
preoperative MRI or CT scan must be less than the corresponding location on patient’s bony. It
the width of the fiducial; otherwise, the possibil- can help the surgeon decide if they need to per-
ity exists that the fiducial marker will be missed form registration again depending on the differ-
on the scan [29]. Due to the possibility of needing ence seen. A registration error of <1 mm is the
to flip the patient in the case of some surgeries goal, but some will accept below 2 mm [22, 24,
involving parts of the sacrum, fiducial markers 46, 48, 65, 66]. Most would advocate for a CT or
(1.8-mm titanium K-wires) can limit the need for MRI scan slice thickness of 0.5–1 mm in order to
a second intraoperative image acquisition that obtain registration errors less than 2 mm. For
allow them to quickly reregister the patient after paired-point and surface matching, a registration
replacing the patient tracker [2]. When placing error of greater than 2 mm will not infrequently be
fiducials, they should be placed in solid bone obtained which can be due to soft tissue such as
because placing them erroneously in soft tissue thick cartilage, tendinous, or ligamentous inser-
would lead to significant registration errors and tion being in the way of a bony landmark leading
an inaccurate navigated tool. to an inaccurate registration [22]. Even with fidu-
Registration with intraoperative advanced cial markers, placing the navigational tool inside a
imaging through either a 3 D fluoroscopy C-arm screw head (variability of 1–2 mm) or at the tip of
(Fig. 13.9a, b) CT scan machine or O-arm intra- a K-wire can introduce error by not selecting the
operatively (Fig. 13.9c–f) to create a CT image correct point on the preoperative imaging and
that can be matched with the preoperative imag- then by not matching that location well enough
ing [45, 60–63] to make the registration process intraoperatively. Depending on the software used,
more successfully consistent with minimal regis- the surgeon has options to improve the accuracy
tration error. Image to image registration using of the registration through several different pro-
intraoperative advanced imaging such as 3D fluo- cesses. Surface matching originally intended as a
roscopy or intraoperative CT [45] offers the ben- way to avoid a second surgery (preresection fidu-
efit of not requiring fiducial markers along with a cial surgery placement) is a way to improve the
preresection surgery and can have improved reg- accuracy of the registration [21, 41, 57]. The navi-
istration accuracy over pair-point matching using gational probe can be used to select in continuous
bony landmarks. These systems can allow the sur- succession a minimum of 30 points but often up to
geon to manually match and then use automatic 100 points on the patient’s exposed pelvis boney
matching done by the navigational software to surface [22, 24]. If the registration error is unac-
fine-tune the process (Fig. 13.10). The image ceptable even with attempts at refinement, the sur-
fusion process is not without its own contribution geon will have to start over again by locating
144 D. M. Joyce
a b
c d e
Fig. 13.9 (a) 3D fluoroscopy unit used to create intraop- (c) Intraoperative CT scanner. (d, e) Intraoperative image
erative CT like image. (b) CT like image created by 3D acquisition. (f) A particular of a CT scan machine
fluoroscopy unit for import into navigation system.
Fig. 13.10 Process of “fusing” or matching the preoperative CT scan with the intraoperative pseudo CT image created
by the 3D fluoroscopy machine
points after confirming in the navigation system

he has picked his correct landmarks on the preop-
erative scan assuming that metal scatter is not
interfering with identifying the point. One may
not be able to obtain a less than 1 mm registration
error, and one will have to accept a 2 mm error.
Rarely does one have to abandon navigation.
13.9 Computer-Navigated
Resection
Once registration is complete, and most of the

surgical dissection is performed navigated
resection can be attempted. Custom tracker
connectors allow about any type of tool to be
navigated as long as it can be calibrated regis-
tered to the patient with a tracker, so they can
be tracked in space (Fig. 13.11). Several differ-
ent tools can be used for navigation and can be
navigated and include diathermy device [22,
24], ostetomes [67], chisels [1, 2, 33], drills Fig. 13.11 Navigated 1/4 in. osteotome
[68, 69], burrs [18, 46], screwdriver [67], and
oscillating saws [20, 24, 34, 70]. One thing to
be aware is tool localization may differ between The navigated saw and navigated osteotome
the display of the instrument on the monitor are some of the most common tools used. The
and what is seen in the operative field because navigated saw with a thinner, smaller blade
tool registration can degrade with striking a allows one to use it in smaller exposures with
tool or using a vibrating instrument can cause limited excursion, but the instability due to
loosening of the instrument tracker causing vibrations with increased flexibility creates
inaccuracies over time [71]. instability and produces potential error in the
146 D. M. Joyce
a b c
Fig. 13.12 (a) Navigated ½ in. osteotome for completion of cut. (b) Navigated osteotome for completion of cut. (c)
Planned cuts
planned resections [72] as well as injury to tissue or

disruption of tumor due to oscillations [24]. A slight
push by the surgeon, the smaller saw blade allows
significant flexion of several millimeters that can
cause an inaccurate positional reading of the saw
blade [23, 70]. Even with these drawbacks with the
saw, there have been significant improvements for
angle of cut and location of the cut plane (2.8 mm)
when using a navigated saw compared to the free-
hand process (5.7 mm) [20]. The other navigational
tool at hand is the osteotome that allows the user to
have nonflexible tool that gives reliable depth and
trajectory feedback via the computer navigational
screen to the surgeon (Fig. 13.12a, b). The draw-
back to the osteotome is that using it can cause
unwanted fractures in bone. Both the saw and the
osteotome are ideal for uniplanar cuts, but when a
multiplanar three- dimensional cut is needed,
another method is used. The surgeon can use the
navigational pointer to identify these “way-ward
points” [12] in a multiplanar cut and mark them
with cautery or a sterile marker [11, 23, 48, 68]. At
this point a drill or a burr can be used to create sev- Fig. 13.13 Navigated burr 1.5 mm head
eral holes along the planned cut can then be com-
pleted with either an osteotome or burr at the tion by introducing a certain amount of uncertainty
discretion of the surgeon. A burr has benefit over a into the system for the next cut [18]. This is particu-
drill, in that it can be used to thin cortex on the far larly important because cutting sacroiliac and sym-
side before coming into contact with tumor or a physeal ligaments in the pelvis can create subtle
critical structure that may exist on the opposite side mobility that can affect the correct location of the
(Fig. 13.13). Disruption of the ring by ligament or bone cut. An inappropriately timed osteotomy
bone sectioning can potentially disturb the accuracy introduces error into the navigation system by lead-
of the spatial relationship and corrupt the registra- ing to inaccurate second osteotomy [22].
13.10 T
opics Unique to Computer 13.11 Visualization, Accuracy,
Navigation and Margin Benefits
One of the drawbacks to navigation is the added The overall benefit to navigated surgery is the
time to surgery, and with added surgical time, reproducibility of the surgical resection. In a
complications rates can increase. Tracker inser- study of 28 patients with 61 osteotomies using
tion and registration increases surgical time paired-point registration, the quantitative differ-
between 15 and 47 min but, after the learning ence between the planned osteotomies and per-
curve, can often decrease from an average of formed osteotomies was 2.52 ± 2.32 mm for all
30 min to 20 min after the surgeon has performed patients and 2.82 ± 2.01 for the pelvis [23, 75].
more navigated surgeries [22]. Relatively speak- An ex vivo experimental study showed the accu-
ing, this added time in pelvic surgery is probably racy of performed osteotomy planes with respect
insignificant given 30 min of delayed surgical to the planned planes in the pelvis was signifi-
time may only represent 6% of surgical time in an cantly improved by almost 9 mm using a navi-
8-h case. Although additional operating time is gated saw, averaging 2.8 mm compared to
needed for navigation set up; planning and defin- 11.2 mm for the freehand saw (p < 0.001), and no
ing the resection plane on preoperatively obtained intralesional tumor resections were executed
images can reduce the overall surgical time since compared to 22% (N = 5) intralesional violations
the osteotomy is already defined on the naviga- in the freehand group (N = 23) [70]. This showed
tion screen, thus negating the need for multiple by choosing a desired safe margin of 10 mm that
fluoroscopic images [48]. Time is then saved dur- the maximum difference achieved between the
ing surgery by not having to bring in C-arm or cut and desired margin was 6.5 mm for the navi-
X-ray for multiple orthogonal images to deter- gated saw cut compared to the 13 mm conven-
mine the correct resection level if that can even tional cut [70]. In a separate laboratory study
be appreciated on fluoroscopic images. Using involving intraoperative CT registration and nav-
navigation, one can eliminate the need for a igated saw, the navigated sawbones’ entry cuts
C-arm to come in and out of the surgical field, were within 1.4 ± 1 mm and exit cuts 1.9 ± 1.2 mm
thereby decreasing the risk of inadvertent con- from the intended osteotomy plane and were sig-
tamination as well as limiting the radiation expo- nificantly different (p < =0.01) to nonnavigated
sure to the surgical staff [40, 73, 74]. Navigation 2.8 ± 4.9 mm entry cut and 3.5 ± 4.6 mm exit cut
often can allow surgery to be performed through in a pelvic bone model [72]. The navigated saw
one approach which could negate the need for accuracy was evaluated on a cadaver and pro-
flipping the patient. In the situation or having to duced similar differences in intended and exe-
flip back and forth, bony landmarks can be identi- cuted osteotomies which showed a navigated
fied in other areas of the pelvis that can then be entry cut of 1.5 mm ± 0.9 mm and navigated exit
marked on the computer navigation software that cut of 2.1 ± 1.5 mm [72]. Navigation theoreti-
can be used to reestablish the registration [24]. cally allows a 95% certainty of avoiding a posi-
Fiducial markers can make things easier or tive margin as long as the osteotomy is planned
needed if the patient is needed to be flipped from no closer than 5 mm of the tumor [72]. Using
supine to prone; not all the time can the surgery conventional techniques, there is only a 52%
be performed only through a posterior approach probability of achieving a 1 cm margin in a tri-
[8, 49]. Fiducial markers can improve and may plane-simulated tumor model of the pelvis due to
negate the need for another image acquisition the complex 3-D geometry [15]. Another experi-
spin need be performed, but rather the fiducial mental test comparing navigated vs. nonnavi-
markers can be marked as rescue points in order gated saw cuts with a nonnavigated saw had a
to alleviate the need for another spin. 22% intralesional violation rate compared to 0%
148 D. M. Joyce
in the navigated modality [60]. One study look- be saved without compromising margins [76] as
ing at navigation in the pelvis reduced the intral- a free-hand navigated saw improves cutting accu-
esional rate to 8.7% (n = 2) with clear bone racy [72, 77]. Even a navigated tool’s bone loss
margins in all cases (n = 31) with a 13% local related to the saw blade thickness or other resec-
recurrence rate compared to the traditional tion tool can be adjusted for on computer naviga-
method where intralesional rate and local recur- tion planning by shifting the planned resection
rence rate were found to be 29% and 27%, planes by 1.5 mm [20]. In fact, most resection
respectively [24]. planes can be adjusted preoperatively based on
known resection width tools such as the osteo-
tome (0.6 mm) and oscillating saw blade
13.12 Tissue Sparing (1.25 mm) which produced a loss of bone width
and Reconstruction of 2 mm due to oscillation [72]. The precision
and reproducibility of navigation can allow sur-
Surgeons using computer navigation believe it geons to resect the tumor and then use the naviga-
allows for more complex resections and recon- tion software to plan for an allograft piece of
structions than are possible with conventional pelvis to reconstruct allowing for functional limb
surgery allowing preservation of sacral nerve root salvage with a joint reconstruction [78–81].
controlling bladder and bowel (42% of time), Computer navigation osteotomies help with
resect unresectable tumors (13% of the time), and matching to surface contact area of the allograft
avoidance of hindquarter amputation (10% of the to the host bone that can minimize nonunion rates
time) [24]. Navigation can allow for preservation [79]. Navigation, with it’s better accuracy and
of the complete or partial joint making recon- precision is now used to save bone and produce
structions less complex because more bone can joint-sparing bone cuts [60, 76] (Fig. 13.14). The
Fig. 13.14 Examples of joint and pelvic ring preserving cut of the hip joint for a chondrosarcoma
system can also be used to produce pelvic ring not only have to be used for resection and recon-
sparing cuts that will prevent the ring from being struction of malignant tumors; its use in benign
disrupted. In the era of 3-D printing, custom tumors has been described [13] as there is benefit
prosthesis can now be created that eliminate the to using navigation for a en bloc resection over
need to modify an allograft in order maximize the choice of using curettage in an effort to mini-
contact area (Fig. 13.15). In order for the custom mize recurrence and not affect structural integrity
prosthesis to fit precisely, only computer naviga- of the pelvis [13, 82]. Computer navigation using
tion osteotomies can consistently provide enough an O-arm has been used to treat benign tumors
accuracy. The precise cuts and virtual model can and hematologic tumors tumor ablation and
be loaded to the system to allow for prosthesis kyphoplasty in benign and malignant nonprimary
matching osteotomies (Fig. 13.16). Theoretically, bone tumors [83].
this can minimize reconstruction and wound While navigation may be useful in achieving
exposure time as well as having better recon- negative margins, there is some that feel that with
struction to host contact that can reduce compli- some of the difficulty in learning to use navigation
cations and failures. Computer navigation does that navigation may not routinely be needed for all
Fig. 13.15 3D planning and printing for navigated resection and reconstruction of the pelvic sarcoma
150 D. M. Joyce
Fig. 13.16 Two 3D virtual models with one showing the ability to plan for the new hip center and the other model
(yellow) being loaded to the intraoperative navigational software to assist with planning osteotomies
oncologic resections in bone [55, 84]. Pelvic sur- their patients more accurate surgical resection and
gery is difficult, and adding navigation only helps reconstructions. Computer navigation has
if one is familiar with the system. One must under- increased the precision and accuracy of pelvic
stand the limitation of the navigation system one is resections leading to preservation pelvic structures
using as well as having knowledge of accuracy of with the intention of not compromising oncologic
the system [27, 50–52, 85]. Doing 1 or 2 pelvic outcomes in terms of margins and recurrence.
cases a year does not allow the surgeon to become
familiar enough with using pelvic navigation in
this setting. Surgeons are often frustrated with 13.13 Computer Navigation
need for multiple steps needed to make one com- Summary
plete resection with computer navigation.
Computer navigation is a learned process and skill The overall benefit to navigated surgery is the accu-
that takes relatively large numbers of cases to mas- racy and reproducibility of the surgical osteotomy.
ter. This is not a once a year tool. As with any tech- It is difficult to achieve negative margins in pelvic
nology unless you routinely use it, you would not surgery [8] as tumors involved in the pelvis have a
see the benefit it provides. Centers that do high higher prevalence of positive margins [9] leading to
volume of computer navigation cases can offer higher recurrence rates and poorer outcomes.
Pelvic three-dimensional anatomy is difficult to References

understand and master conceptually when trying to
resect tumors [14, 72] making it difficult to get 1. Krettek C, et al. Computer aided tumor resection in
the pelvis. Injury. 2004;35(Suppl 1):S-A79–83.
tumor-free margin leading to higher local recur- 2. Hufner T, et al. New indications for computer-assisted
rence [2, 21, 86, 87]. Without navigation, perform- surgery: tumor resection in the pelvis. Clin Orthop
ing the planned resection with precision and Relat Res. 2004;426:219–25.
accuracy is very difficult to achieve with conven- 3. Court C, et al. Surgical excision of bone sarcomas
involving the sacroiliac joint. Clin Orthop Relat Res.
tional resections in the pelvis [15, 70]. Computer- 2006;451:189–94.
navigated surgery can increase precision of the 4. Pring ME, et al. Chondrosarcoma of the pelvis. A
osteotomies for tumor resection [1, 2, 12, 18, 48]. review of sixty-four cases. J Bone Joint Surg Am.
Clinical studies have shown that navigated tools 2001;83(11):1630–42.
5. Sucato DJ, et al. Ewing’s sarcoma of the pelvis. Long-
have assisted in attaining negative margins in term survival and functional outcome. Clin Orthop
malignant pelvic tumors [1, 18, 20, 34, 68, 86]. The Relat Res. 2000;373:193–201.
surgeon can plan the desired margin with respect to 6. Wirbel RJ, Schulte M, Mutschler WE. Surgical treat-
the tumor on the navigational software and then use ment of pelvic sarcomas: oncologic and functional
outcome. Clin Orthop Relat Res. 2001;390:190–205.
it to assist during resections, maximizing accuracy 7. Ozaki T, et al. Osteosarcoma of the pelvis: experience
and precision and minimizing sacrifice of critical of the Cooperative Osteosarcoma Study Group. J Clin
structure [88]. Where computer navigation excels Oncol. 2003;21(2):334–41.
is the ability to “see beyond walls” because it is 8. Hulen CA, et al. Oncologic and functional outcome
following sacrectomy for sacral chordoma. J Bone
hard to know what lies on the other side of the pel- Joint Surg Am. 2006;88(7):1532–9.
vic cut. Navigation literally shows the surgeon 9. Kawai A, et al. Prognostic factors for patients with sar-
where the instrument is heading in real time and comas of the pelvic bones. Cancer. 1998;82(5):851–9.
how close the tool is coming to tumor on the other 10. Angelini A, et al. Infection after surgical resec-

tion for pelvic bone tumors: an analysis of 270
side. It can help the surgeon minimize the chances patients from one institution. Clin Orthop Relat Res.
of positive bone margins. It can also help the sur- 2014;472(1):349–59.
geon perform resections that make reconstruction 11. Fehlberg S, et al. Computer-assisted pelvic tumor

easier and help maintain function. The future of resection: fields of application, limits, and perspec-
tives. Recent Results Cancer Res. 2009;179:169–82.
computer navigation surgery will likely include the 12. Cho HS, et al. Joint-preserving limb salvage sur-

benefit of robotic assistive devices and patient-spe- gery under navigation guidance. J Surg Oncol.
cific instrumentation [86]. Computer navigation 2009;100(3):227–32.
currently is a passive system which only provides 13. Zhang Y, et al. Three-dimensional printing and com-
puter navigation assisted hemipelvectomy for en bloc
information or feedback, while the future com- resection of osteochondroma: a case report. Medicine
puter-navigated robotic-assisted surgeries will be (Baltimore). 2017;96(12):e6414.
performed with a more active role that physically 14. Grimer RJ, et al. Hindquarter amputation: is it still
guides and limits the surgeon from straying outside needed and what are the outcomes? Bone Joint J.
2013;95-B(1):127–31.
of predetermined resection planes [89]. The benefit 15. Cartiaux O, et al. Surgical inaccuracy of tumor resec-
to robotic-assisted surgery involves still maintain- tion and reconstruction within the pelvis: an experi-
ing control of saws and osteotomes while minimiz- mental study. Acta Orthop. 2008;79(5):695–702.
ing the effect of tool vibration and fatigue on part of 16. Kawai A, et al. Osteosarcoma of the pelvis. Oncologic
results of 40 patients. Clin Orthop Relat Res.
the surgeons hand that navigation alone cannot cur- 1998;348:196–207.
rently provide [90]. During surgery, computer navi- 17. Fuchs B, et al. Osteosarcoma of the pelvis: outcome
gation offers a lot of benefits; however, the system analysis of surgical treatment. Clin Orthop Relat Res.
is not full proof. Navigation does not guarantee the 2009;467(2):510–8.
18. Cho HS, et al. Computer-assisted sacral tumor

margin will be clear because the incorrect margin resection. A case report. J Bone Joint Surg Am.
or resection level is surgeon chosen on the preop- 2008;90(7):1561–6.
erative imaging, but the system does provide accu- 19. Kojima T, et al. The usefulness and limits of magnetic
racy, consistency, and reproducibility with the bony resonance imaging in the differential diagnosis of pel-
vic tumors. Oncol Rep. 2001;8(4):867–9.
resection plane [24].
152 D. M. Joyce
20. Docquier PL, et al. Computer-assisted resection and 39. Stevens F, et al. Minimizing electromagnetic inter-
reconstruction of pelvic tumor sarcoma. Sarcoma. ference from surgical instruments on electromag-
2010;2010:125162. netic surgical navigation. Clin Orthop Relat Res.
21. Reijnders K, et al. Image guided surgery: new tech- 2010;468(8):2244–50.
nology for surgery of soft tissue and bone sarcomas. 40. Stockle U, Schaser K, Konig B. Image guidance in
Eur J Surg Oncol. 2007;33(3):390–8. pelvic and acetabular surgery—expectations, success
22. Young PS, Bell SW, Mahendra A. The evolving role and limitations. Injury. 2007;38(4):450–62.
of computer-assisted navigation in musculoskeletal 41. Bachler R, Bunke H, Nolte LP. Restricted surface
oncology. Bone Joint J. 2015;97-B(2):258–64. matching—numerical optimization and technical
23. Ritacco LE, et al. Accuracy of 3-D planning and
evaluation. Comput Aided Surg. 2001;6(3):143–52.
navigation in bone tumor resection. Orthopedics. 42. Lembeck B, et al. Pelvic tilt makes acetabular cup nav-
2013;36(7):e942–50. igation inaccurate. Acta Orthop. 2005;76(4):517–23.
24. Jeys L, et al. Can computer navigation-assisted sur- 43. Thaler M, et al. Accuracy of an image-guided naviga-
gery reduce the risk of an intralesional margin and tion system for pelvic surgery based on a multimodal-
reduce the rate of local recurrence in patients with ity registration object: a cadaver study. Am J Orthop
a tumour of the pelvis or sacrum? Bone Joint J. (Belle Mead NJ). 2010;39(8):382–5.
2013;95-B(10):1417–24. 44. Van Hellemondt G, et al. Computer-assisted pelvic
25. Schlondorff G. Computer-assisted surgery: historical surgery: an in vitro study of two registration proto-
remarks. Comput Aided Surg. 1998;3(4):150–2. cols. Clin Orthop Relat Res. 2002;405:287–93.
26. Berlemann U, et al. [Computer-assisted orthopedic 45. Takao M, et al. Application of a CT-3D fluoroscopy
surgery. From pedicle screw insertion to further appli- matching navigation system to the pelvic and femoral
cations]. Orthopade. 1997;26(5):463–9. regions. Comput Aided Surg. 2012;17(2):69–76.
27. Stiehl JB, Heck DA. Computer-assisted surgery: basic 46. Wong KC, et al. Image fusion for computer-

concepts. Instr Course Lect. 2008;57:689–97. assisted bone tumor surgery. Clin Orthop Relat Res.
28. Hufner T, et al. Computer-assisted fracture reduction 2008;466(10):2533–41.
of pelvic ring fractures: an in vitro study. Clin Orthop 47. Jones BC, Sundaram M, Kransdorf MJ. Synovial sar-
Relat Res. 2002;399:231–9. coma: MR imaging findings in 34 patients. AJR Am J
29. Cho HS, et al. Direct application of MR images to Roentgenol. 1993;161(4):827–30.
computer-assisted bone tumor surgery. J Orthop Sci. 48. Wong KC, et al. Precision tumour resection and

2011;16(2):190–5. reconstruction using image-guided computer naviga-
30. Dessenne V, et al. Computer-assisted knee anterior tion. J Bone Joint Surg Br. 2007;89(7):943–7.
cruciate ligament reconstruction: first clinical tests. J 49. Amiot LP, Poulin F. Computed tomography-based

Image Guid Surg. 1995;1(1):59–64. navigation for hip, knee, and spine surgery. Clin
31. Jaramaz B, et al. Computer assisted measurement of Orthop Relat Res. 2004;421:77–86.
cup placement in total hip replacement. Clin Orthop 50. Frantz DD, et al. Accuracy assessment protocols for
Relat Res. 1998;354:70–81. electromagnetic tracking systems. Phys Med Biol.
32. Hube R, et al. CT-based and fluoroscopy-based navi- 2003;48(14):2241–51.
gation for cup implantation in total hip arthroplasty 51. Khadem R, et al. Comparative tracking error analy-
(THA). Surg Technol Int. 2003;11:275–80. sis of five different optical tracking systems. Comput
33. Langlotz F, et al. The first twelve cases of computer Aided Surg. 2000;5(2):98–107.
assisted periacetabular osteotomy. Comput Aided 52. Milne AD, et al. Accuracy of an electromagnetic

Surg. 1997;2(6):317–26. tracking device: a study of the optimal range and
34. So TY, Lam YL, Mak KL. Computer-assisted naviga- metal interference. J Biomech. 1996;29(6):791–3.
tion in bone tumor surgery: seamless workflow model 53.
Docquier PL, et al. Registration accuracy in
and evolution of technique. Clin Orthop Relat Res. computer- assisted pelvic surgery. Comput Aided
2010;468(11):2985–91. Surg. 2009;14(1–3):37–44.
35. Lionberger DR. The attraction of electromagnetic
54. Wiles AD, Thompson DG, Frantz DD. Accuracy

computer-assisted navigation in orthopaedic sur- assessment and interpretation for optical tracking sys-
gery. In: Stiehl JB, Konermann WH, Hacker R, edi- tems. Med Imaging. 2004;5367:1–12.
tors. Navigation and MIS in orthopaedic surgery. 55. Farfalli GL, et al. What is the expected learning curve
Heidelberg: Springer; 2006. p. 44–53. in computer-assisted navigation for bone tumor resec-
36. Song EK, et al. Accuracy of navigation: a comparative tion? Clin Orthop Relat Res. 2017;475(3):668–75.
study of infrared optical and electromagnetic naviga- 56. Mayr E, et al. The effect of fixation and location on
tion. Orthopedics. 2008;31(10 Suppl 1). the stability of the markers in navigated total hip
37.
Hernandez-Vaquero D, Suarez-Vazquez arthroplasty: a cadaver study. J Bone Joint Surg Br.
A. Complications of fixed infrared emitters in 2006;88(2):168–72.
computer-assisted total knee arthroplasties. BMC 57. Oszwald M, et al. Accuracy of navigated surgery of
Musculoskelet Disord. 2007;8:71. the pelvis after surface matching with an a-mode
38. Lionberger DR, et al. How does electromagnetic navi- ultrasound probe. J Orthop Res. 2008;26(6):860–4.
gation stack up against infrared navigation in mini- 58. Arand M, Kinzl L, Gebhard F. [Sources of error

mally invasive total knee arthroplasties? J Arthroplast. and risks in CT based navigation]. Orthopade.
2008;23(4):573–80. 2002;31(4):378–84.
59. Salehi SA, Ondra SL. Use of internal fiducial mark- 75. Ritacco LE, et al. Bone tumor resection: analy-

ers in frameless stereotactic navigational systems sis about 3D preoperative planning and navigation
during spinal surgery: technical note. Neurosurgery. method using a virtual specimen. Stud Health Technol
2000;47(6):1460–2. Inform. 2013;192:1162.
60. Sternheim A, et al. Cone-beam computed tomography- 76. Lam YL, et al. Is it possible and safe to perform
guided navigation in complex osteotomies improves acetabular-preserving resections for malignant neo-
accuracy at all competence levels: a study assessing plasms of the periacetabular region? Clin Orthop
accuracy and reproducibility of joint-sparing bone cuts. Relat Res. 2017;475(3):656–65.
J Bone Joint Surg Am. 2018;100(10):e67. 77. Haider H, Barrera OA, Garvin KL. Minimally inva-
61. Kobayashi H, et al. Three-dimensional fluoroscopic sive total knee arthroplasty surgery through navigated
navigation-assisted surgery for tumors in patients with freehand bone cutting: winner of the 2005 “HAP”
tumor-induced osteomalacia in the bones. Comput PAUL AWARD. J Arthroplast. 2007;22(4):535–42.
Assist Surg (Abingdon). 2017;22(1):14–9. 78. Abe K, et al. The usefulness of wide excision assisted
62. Fujiwara T, et al. Intraoperative O-arm-navigated
by a computer navigation system and reconstruction
resection in musculoskeletal tumors. J Orthop Sci. using a frozen bone autograft for malignant acetabu-
2018;23(6):1045–50. lar bone tumors: a report of two cases. BMC Cancer.
63. Fujiwara T, et al. Mini-open excision of osteoid oste- 2018;18(1):1036.
oma using intraoperative O-arm/Stealth navigation. J 79. Aponte-Tinao L, et al. Does intraoperative naviga-
Orthop Sci. 2019;24(2):337–41. tion assistance improve bone tumor resection and
64. Pappas IP, et al. New method to assess the registration allograft reconstruction results? Clin Orthop Relat
of CT-MR images of the head. Injury. 2004;35(Suppl Res. 2015;473(3):796–804.
1):S-A105–12. 80. Paul L, et al. Selection of massive bone allografts
65. Wong KC, Kumta SM. Computer-assisted tumor sur- using shape-matching 3-dimensional registration.
gery in malignant bone tumors. Clin Orthop Relat Acta Orthop. 2010;81(2):250–5.
Res. 2013;471(3):750–61. 81. Paul L, et al. Inaccuracy in selection of massive bone
66. Wong KC, Kumta SM. Joint-preserving tumor
allograft using template comparison method. Cell
resection and reconstruction using image-guided Tissue Bank. 2008;9(2):83–90.
computer navigation. Clin Orthop Relat Res. 82. Farfalli GL, et al. Is navigation-guided en bloc resec-
2013;471(3):762–73. tion advantageous compared with intralesional curet-
67. Cheong D, Letson GD. Computer-assisted naviga-
tage for locally aggressive bone tumors? Clin Orthop
tion and musculoskeletal sarcoma surgery. Cancer Relat Res. 2018;476(3):511–7.
Control. 2011;18(3):171–6. 83. Wu K, et al. Intraoperative navigation for minimally
68. Wong KC, et al. Computer assisted pelvic tumor
invasive resection of periarticular and pelvic tumors.
resection and reconstruction with a custom-made Orthopedics. 2011;34(5):372.
prosthesis using an innovative adaptation and its vali- 84. Staats K, et al. Computer navigation-assisted sur-

dation. Comput Aided Surg. 2007;12(4):225–32. gery for musculoskeletal tumors: a closer look into
69. Kendoff D, et al. Navigated Iso-C3D-based percuta- the learning curve. Eur J Orthop Surg Traumatol.
neous osteoid osteoma resection: a preliminary clini- 2017;27(6):851–8.
cal report. Comput Aided Surg. 2005;10(3):157–63. 85. Stiehl JB, Heck DA, Lazzeri M. Accuracy of ace-
70. Cartiaux O, et al. Computer-assisted planning and tabular component positioning with a fluoroscopi-
navigation improves cutting accuracy during simu- cally referenced CAOS system. Comput Aided Surg.
lated bone tumor surgery of the pelvis. Comput Aided 2005;10(5–6):321–7.
Surg. 2013;18(1–2):19–26. 86. Cartiaux O, et al. Improved accuracy with 3D plan-
71. Zura RD, Kahler DM. A transverse acetabular non- ning and patient-specific instruments during simu-
union treated with computer-assisted percutaneous lated pelvic bone tumor surgery. Ann Biomed Eng.
internal fixation. A case report. J Bone Joint Surg Am. 2014;42(1):205–13.
2000;82(2):219–24. 87. Delloye C, et al. Pelvic reconstruction with a structural
72. Sternheim A, et al. Navigated pelvic osteotomy and pelvic allograft after resection of a malignant bone
tumor resection: a study assessing the accuracy and tumor. J Bone Joint Surg Am. 2007;89(3):579–87.
reproducibility of resection planes in sawbones and 88. Cho HS, et al. The outcomes of navigation-assisted
cadavers. J Bone Joint Surg Am. 2015;97(1):40–6. bone tumour surgery: minimum three-year follow-up.
73. Zwingmann J, et al. Computer-navigated iliosacral J Bone Joint Surg Br. 2012;94(10):1414–20.
screw insertion reduces malposition rate and radia- 89. Sugano N. Computer-assisted orthopaedic surgery

tion exposure. Clin Orthop Relat Res. 2009;467(7): and robotic surgery in total hip arthroplasty. Clin
1833–8. Orthop Surg. 2013;5(1):1–9.
74. Arand M, Kinzl L, Gebhard F. Computer-guidance 90. Khan F, et al. Haptic robot-assisted surgery improves
in percutaneous screw stabilization of the iliosacral accuracy of wide resection of bone tumors: a pilot
joint. Clin Orthop Relat Res. 2004;422:201–7. study. Clin Orthop Relat Res. 2013;471(3):851–9.
Spinopelvic Fixation After
Sacrectomy 14
Vasilios G. Igoumenou, Andreas F. Mavrogenis,
Andrea Angelini, Farzam Vazifehdan,
Konstantinos Soultanis, Pavlos Patapis,
Pietro Ruggieri, and Panayiotis J. Papagelopoulos
14.1 Introduction for a spinopelvic fixation, due to the special ana-

tomical (large vessels, bladder, bowel) and biome-
Obtaining solid arthrodesis of the lumbosacral chanical characteristics of the lumbosacral region.
region of the spine continues to be a challenge for From a biomechanical point, the sacrum–sacroil-
spine surgeons. Various spinal pathologies iac joint–ilium complex cannot be discussed sepa-
require a spinopelvic fixation including adult rately, as the vertical load from the axial skeleton
deformities with coronal or sagittal malalign- is transferred to the lower limbs via this area;
ment, neuromuscular scoliosis with pelvic obliq- thereby, an intact sacropelvic region is one of the
uity, high-grade spondylolisthesis, and lumbosacral key points of the human upright posture and walk-
tumors (primary or secondary) [1–3]. The latter ing ability. Tumors in this region alone as well as
represents actually the most challenging indication their surgical resection can significantly influence
the biomechanics of the sacropelvic complex
resulting in failure of axial load transmission.
Therefore, to perform an oncologically and func-
V. G. Igoumenou · F. Vazifehdan tionally optimal surgery, both the anatomical and
Spine Center Stuttgart, Diakonie-Klinikum Stuttgart,
biomechanical consequences must be carefully
Stuttgart, Germany
e-mail: [email protected] considered during preoperative planning [4].
En bloc resection of tumors in the lumbosacral
A. F. Mavrogenis · K. Soultanis
P. J. Papagelopoulos (*) region with procedures such as total sacrectomy
First Department of Orthopaedics, National and or L5 spondylectomy is typically indicated for
Kapodistrian University of Athens, School of patients with locally invasive primary sacral
Medicine, Athens, Greece
tumors such as chordomas, sarcomas, chondro-
e-mail: [email protected];
[email protected]; [email protected] sarcomas, or giant cell tumors. In contrast, this
strategy has been applied in limited cases to met-
A. Angelini · P. Ruggieri
Department of Orthopedics and Orthopedic astatic diseases, mainly due to recent advances in
Oncology, University of Padova, Padova, Italy adjuvant treatment in surgical oncology [5].
e-mail: [email protected]; Additionally, treatment of metastatic tumors in
[email protected]
the lumbar spine near the lumbosacral junction
P. Patapis often requires pelvic fixation even with separa-
Third Department of Surgery,
tion surgery procedures to allow for adequate sta-
National and Kapodistrian University of Athens,
School of Medicine, Athens, Greece bilization of the lumbosacral junction, as these
e-mail: [email protected] patients often have poor bone quality [5].

156 V. G. Igoumenou et al.
Therefore, standardizing universal surgical inability to obtain a tumor-free margin of resec-

procedures and techniques is a copious attempt tion is similarly a relative contraindication. The
for spinal and orthopedic surgeons. medical status of the patients also is important;
patients receiving chemotherapy frequently
require alterations in their chemotherapy sched-
14.2 Indications ules to allow for surgery of this magnitude. All
patients are subject to an intense preoperative
Total sacrectomy is indicated for primary sacral medical evaluation including a dobutamine stress
malignancies. More rarely, sacrectomy is indi- echocardiogram for (a) anyone with known car-
cated for patients with primary or recurrent pel- diovascular disease; (b) men above age of
vic visceral tumors (most commonly colorectal 40 years; or (c) women above age of 50 years [6].
carcinoma with sacral involvement by direct Special consideration has been made regard-
extension) and no evidence of metastatic or nodal ing the use of spinopelvic fixation with dual iliac
disease. The techniques hereby described may screws. These constructs are necessary mostly if
also be adapted to intralesional treatment of (a) total sacrectomy is performed where the
benign tumors such as osteoblastoma and aneu- whole sacroiliac joint is removed on both sides
rysmal bone cyst. Last, some carefully selected [7], (b) partial sacrectomy is performed involving
benign aggressive sacral tumors may also be con- more than 50% of sacroiliac joint on each side, or
sidered for en bloc resection, particularly if small partial sacrectomy involving less than one half of
or recurrent [6]. the sacrum but with one-side sacroiliac joint
A surgical procedure of a much larger scale is resection [8], and (c) in palliative fixation cases
the amputative sacral resection that extends into for unstable destructive lumbosacral metastatic
the pelvis and may be necessary in the following lesions where pedicle screw anchorage in the
instances: tumor epicenter within the ilium but sacrum is extremely poor [1, 9].
disease extends across the sacroiliac joint, or
tumor epicenter in the sacrum but removal
requires resection of the femoral nerve as well as 14.3 Classification
the lumbosacral trunk or the lumbosacral trunk
and the hip joint articulation. In these cases, the There is no standard classification of these proce-
ultimate function of the limb is so poor that exter- dures. Mayo Clinic has proposed the following
nal hemipelvectomy in conjunction with sacral classification, where resections could be divided
resection allows the maximal oncologic margin into five types based on the extent of the lumbo-
to be obtained and provides for healthy and sacral resection and the need for an associated
robust flap coverage from the limb [6]. external hemipelvectomy [10, 11]. They are as
On the other side, the presence of dissemi- follows: Type IA resection—total sacrectomy;
nated malignancy is a strong relative contraindi- Type IB resection—subtotal sacrectomy above
cation for sacrectomy. The procedure is of such the S1 foramen; Type IC resection—subtotal
magnitude and generally entails deliberate neuro- sacrectomy below the S1 foramen (the SI joints
logical defects with frequent loss of bowel, blad- are not disrupted here, and a reconstruction is not
der, sexual, and potentially lower extremity typically performed); Type II resection—hemisa-
function that it is usually inappropriate to pursue crectomy with or without partial lumbar excision,
without curative intent. Patients with tumor and iliac wing resection; Type III resection—
thrombus in the iliac veins or vena cava by sar- external hemipelvectomy with hemisacrectomy
coma predictably have a rapid development of with or without partial lumbar excision; Type IV
metastatic disease and demise [6]; evidence of resection—external hemipelvectomy with total
the above on preoperative imaging can prompt sacrectomy with or without lumbar excision; and
catheter-directed biopsy, while its finding at time Type V resection—hemicorporectomy-type pro-
of surgery prompts abortion of resection [6]. The cedures (Fig. 14.1).
14 Spinopelvic Fixation After Sacrectomy 157
a b c d
Fig. 14.1 Mayo classification of spinopelvic resections: extended external hemipelvectomy), and (d) total sacral
(a) total sacral resection (total sacrectomy), (b) hemisa- and lower lumbar spine resection (total sacrectomy and
cral resection (partial sacrectomy), (c) hemisacral and [extended] external hemipelvectomy)
lower lumbar spine resection (partial sacrectomy and
(direct implantation of L rods into the iliac

14.4 Techniques and Implants crests), intrasacral rods, transiliac bars, iliac
screws, and S2-alar-iliac screws [5]. Some of the
Generally, there are three components in spino- earliest reported techniques used Harrington rods
pelvic surgery, spinopelvic fixation, posterior [13, 14], or hooks and CD rods connected to
pelvic ring fixation, and anterior spinal column transiliac bars [15]. The Galveston technique for
support [5]. The infrequency of these procedures spinopelvic fixation was initially described by
does not allow for an established gold-standard Allen and Ferguson for use in scoliosis surgery in
technique, as the various instrumentation tech- 1982 [16]. It was later modified for use in spino-
niques are scattered across the literature in case pelvic reconstruction after sacrectomy [17, 18].
reports and small case series; however, it was Since then, various modifications have been pro-
implied that incorporation of anterior spinal col- posed and the direct implantation of rods into
umn fixation in reconstructing the spinopelvic iliac crests has been practically abandoned. In the
junction may lead to improved outcomes with modified Galveston technique, the rods are fixed
lower rates of hardware failure and other surgical to the pelvis with iliac screws in the caudal part
complications including blood loss [12]. of the system. Further modifications include the
multiple rod–screw techniques such as the four-
rod reconstruction or double-rod double iliac
14.4.1 Spinopelvic Fixation screw reconstruction [19]. The use of one single
U-shaped rod anchored with iliac screws—the
The goal of the surgical stabilization after an closed-loop technique—was published in 2009
extended oncological procedure such as sacrec- providing a more harmonic stress distribution
tomy is to restore the load-bearing structures along the whole construct [20].
from the lumbar spine to the remaining pelvis. Iliac screws represent a modified version of
There are several techniques for spinopelvic the Galveston technique having three times more
instrumentation that were originally described biomechanical strength than the Galveston tech-
for spinal deformity or trauma surgery. These nique, and at the same time, they are applicable in
include sacral sublaminar wires and hooks, S1 every case, unless a hemi- or partial pelvectomy
tricortical screws, Galveston rod technique is required [5]. In fact, one of the main indications
for the use of dual iliac screws is the need to per- fixation to bilateral iliac, transiliac bars, and
form partial or total sacrectomy in order to have a cages [5]. With the triangular frame reconstruc-
solid basis at the bottom of the spinopelvic con- tion, the pulled down L5 vertebral body is affixed
struct. In addition, the dual iliac screw techniques to the bilateral ilium with sacral rods. The pelvis
may be used in palliative fixation of metastatic is also connected to the spinal rods with a second
lumbosacral lesions with extremely poor sacral sacral rod [4]. Murakami et al. [24] showed in
bone quality [1]. Several biomechanical cadav- their in vitro and in silico biomechanical analyses
eric studies have evaluated theses fixation tech- that there was less stress concentration on the
niques. In the setting of total sacrectomy, Mindea implants with this technique; however, excessive
et al. [7] showed that the double-rod double iliac stress occurred in the iliac bones that could be
screw technique provided the most rigid fixation, associated with loosening of the sacral rods.
followed by the single-rod double iliac screw Gallia et al. [25] published a challenging tech-
fixation, in comparison with single-rod single nique known as the Johns Hopkins University
iliac screw or modified Galveston technique. Yu (JHU) technique; a modified Galveston technique
et al. [21] showed that dual iliac screws, when all was used, where a transiliac bar was inserted
inserted in the lower iliac column, exhibited through the iliac crests, and single iliac screws
higher compressive and torsional stiffness not were implanted and linked with a horizontal rod.
only when compared to single iliac screws (short The spinal rods were attached to the transiliac bar
and long) but also to dual iliac screws where two with L connectors, and the transiliac bar, the hori-
screws are inserted in the lower iliac column and zontal rod, and a horizontal connector between
two screws in the upper iliac column. the vertical spinal rods were connected with one
In terms of selecting iliac screw length, accord- other using vertical connectors. Last, a femoral
ing to the biomechanical study of Zheng et al. allograft was placed horizontally, between the
[22], short iliac screws (7 mm in diameter and two iliac crests bridging the defect.
70 mm in length) are susceptible to loosening
after cyclic loading. Bone cement augmentation
of short screws has shown a significant increase in 14.4.3 Anterior Spinal Column
the fixation strength of short screws to an extent Support
similar to that of long iliac screws (7 mm in diam-
eter and 120 mm in length). Therefore, given the The importance of anterior spinal column sup-
potential complications of long screw breach, port in lumbopelvic reconstruction after total
short iliac screw fixation with augmentation with sacrectomy has been discussed extensively. In
bone cement may be a viable option for lumbo- 2005, Dickey et al. [26] published the use of
pelvic reconstruction, although much larger screw bilateral fibular grafts. The fibular grafts are
diameters are currently available and more com- placed between the L5 vertebra and the bilateral
monly used [5]. According to biomechanical anal- iliopectineal area, and this technique can be com-
yses, it cannot be argued that a stronger construct bined with the posterior stabilization techniques.
with multiple rods and screws increases the rigid- In vitro and in silico biomechanical study showed
ity of the construct. However, it should be also that with the help of these combined systems,
kept in mind that more metal implants increase greater rigidity can be achieved; however, the
the risk of wound healing problems; therefore, increase of the morbidity from the extension of
increased caution is required [23]. the surgery has to be also considered [4]. A
cadaveric biomechanical study by Cheng et al.
[27] evaluated the following four constructs:
14.4.2 Posterior Pelvic Ring Fixation sacral rod reconstruction; bilateral fibular flap
reconstruction; four-rod reconstruction; and
Techniques for posterior pelvic ring reconstruc- improved compound reconstruction (a combina-
tion include allografts (femur or tibia) with screw tion of the previous methods). Among these,
improved compound reconstruction that utilized ally to posterior spinal instrumentation.

the sacral rod and the fibular triangular construct Pedicle screw instrumentation is performed in
in the anterior approach produced optimal struc- usually the remaining three to four vertebral
tural stability after total sacrectomy. Similarly, body sites (Fig. 14.2a, b). Prior instrumenta-
Clark et al. [28] examined the biomechanical tion, appropriate changes to the surgical table
strength of three constructs: femoral strut should be made to restore lumbar lordosis.
allograft reconstruction, where a femoral allograft Pedicle screws are placed aggressively to
was placed between iliac and secured with bone extend to the anterior cortex or even bicorti-
screws; L5–iliac cage strut reconstruction, where cally. Usually, after the sacrum is removed, a
two titanium cages were placed obliquely, each hand can be placed ventral to the spine to feel
wedged between the inferior L5 endplate and the the pedicle screws as they come through to
iliac bone; and S1 body replacement with expand- allow for safe bicortical placement. Screws
able cage reconstruction, in which a rod was are placed in the remaining ilium, ideally with
placed from the inferior L5 endplate and fixed to the double iliac screw techniques. “Docking
a transiliac bar and a 22-mm expandable cage sites” are placed for fibula strut grafts in the
was placed between the L5 endplate and the supra-acetabular region. A burr is used to
transiliac bar. They concluded that the latter tech- place these from behind. If the level of iliac
nique provided the most biomechanically stable resection prohibits this, the ischium is usually
structure. an appropriate site for docking stations as
well. Once this is done, fibula strut grafts are
• Type 1 and 2 resections placed as described by Dickey et al. [26], in a
• Resections at or below the level of the S2 neu- “cathedral fashion”; struts are placed in the
roforamen are generally resected through a supra-acetabular region and then end in the
posterior approach unless there is involvement last remaining vertebral segment. Appropriate
of pelvic visceral or vascular structures. Given rods are placed after the strut grafts are posi-
the need to obtain an oncologic margin, this tioned, and compression is achieved across
generally implies lesions at or below the S2/3 these to lock the fibula grafts in. If the patient
vestigial disk [6]. has undergone prior pelvic radiation, consid-
• Lesions cephalad to this level or involving pel- eration is given to using vascularized fibular
vic structures are treated first with anterior grafts. This significantly extends the operative
mobilization of pelvic structures, vessel liga- time and may require staging to a further day
tion, and unicortical anterior sacral osteotomy. [6, 11].
The use of pedicle flaps is encouraged for • Type 3 and 4 resections
facilitating wound healing. A pedicled myocu- • Partial and total sacrectomies in conjunction
taneous rectus abdominis flap can be har- with external hemipelvectomy represent the
vested in this stage and tucked into the amputative sacrectomies. These procedures
abdomen with the anterior procedure. Tumor are performed in one stage. Patients undergo-
resection is then completed through a poste- ing Type 3 resections are considered for an
rior approach, and the rectus flap is pulled instrumented spinopelvic arthrodesis to the
through the abdomen and rotated to assist in remaining limb if more than 50% of the lum-
wound closure and reconstruction of the pos- bosacral articulation is resected. The instru-
terior abdominal wall. The posterior approach mentation can be performed 48 h after the
can be performed 48 h later, unless the rectum amputation to allow time for final margins to
is devascularized and requires resection with be ascertained and to minimize the physio-
the tumor specimen [6]. logic impact on the patient. In Type 4 resec-
• Resections cephalad to the S1 neuroforamen tions, the resection is performed in a single
require spinopelvic reconstruction. Fibula stage, and the spinopelvic instrumentation
autografts or allografts can be used addition- between the remaining lumbar spine and
Fig. 14.2 (a) A 35-year-old man with a sacropelvic chondrosarcoma. (b) Type II resection and spinopelvic reconstruc-
tion were done without evidence of local recurrence at 8-year follow-up
remaining limb is in a second stage. Therefore, patients likely benefit from instrumented
in Type 4 resections, the tumor-free portion of fusion across the spinopelvic junction. This is
the amputated femur is ideally stored sterilely generally performed in a second stage
in a liquid nitrogen freezer until this second approximately 48 h after the index surgical
stage of the surgery [29]. procedure. It is usually simple to reopen the
• The need for an instrumented spinopelvic wound (and probably advantageous to wash
reconstruction after a Type 3 resection is con- out the inevitable degree of hematoma which
troversial. In our experience, if the majority of develops). Reconstruction is performed using
the lumbosacral articulation is resected, spinopelvic instrumentation from L4 through
the ilium on the retained side. There is usually lumbar spine. Rods and screws allow for fixa-
excellent exposure to perform a discectomy of tion and compression across this graft. An
the remaining disk at the L5–S1 segment and alloderm or similar membrane can be prophy-
provide an anterior interbody graft at this lactically used to sequester the abdominal
junction. Depending on the vascular mobiliza- contents away from the instrumentation.
tion achieved in the index procedure or desired Similar to the Type 3 resection, the anterior
in the secondary procedure, similar anterior thigh flap is inserted to close the soft tissue
lumbar interbody fusion can be performed at defect. As Type 4 resections commonly
the L4–5 level as well [29]. involve resection of the anus and genital
• In the case of a Type 4 resection, it is neces- structures, the amount of skin defect may
sary to provide reconstruction between the require the full aspect of skin from the quad-
remaining lumbar spine and remaining hemi- riceps flap [29].
pelvis and limb. Because of the very large
magnitude of the oncologic resection, these
procedures are staged at least 48 h and often- 14.5 Results
times longer after the index procedure, once
the patient has physiologically recovered Oncologic results are most favorable when com-
appropriately. Key aspects of the reconstruc- plete resection of the tumor is obtained. It is best
tion of a Type 4 procedure include centraliz- illustrated by the data of Fuchs et al. [30], report-
ing the remaining hemipelvis and limb under ing the operative management of sacral chor-
the lumbar spine as well as providing a robust doma. In a series of 52 patients undergoing
autograft strut between the lowest remaining surgery, complete survival was seen in all patients
vertebral body and the hemipelvis. At the in whom a wide margin was achieved at the time
time of the index resection, a portion of the of surgery. In contrast, the majority of patients
femur of the amputated limb that is largely with less than a wide margin resection succumb
free of tumor is saved sterilely in a liquid to disease. Results of more aggressive tumors
nitrogen freezer. This provides a strut graft to depended heavily upon the response to chemo-
bridge the gap between the remaining lumbar therapy. Regarding neurologic function after
spine and pelvis on the retained side. Pedicle major sacrectomy, preservation of bilateral S2
screw instrumentation is performed into at nerve roots and a unilateral S3 nerve root or uni-
least the lowest three segments of the lumbar lateral S2, S3, and S4 nerve roots is required for
spine on the remaining side. Screw fixation is predictable maintenance of bowel and bladder
obtained in the bone stock of the remaining function [31–33]. In those patients undergoing
ilium avoiding the hip joint. In performing the major spinopelvic reconstruction, a study of 45
reconstruction after a Type 4 resection, two patients [chondrosarcoma (n = 11); other sarco-
key factors are involved. First, the pelvis mas (n = 11); osteosarcoma (n = 9); chordoma
should be externally rotated centralizing the (n = 6); locally invasive carcinoma (n = 5); and
remaining lumbar spine over the remaining others (n = 3)] at mean 38-month follow-up has
pelvis such that the patient’s center of gravity shown that 28 were living and 17 were deceased;
is relatively uniform. Second, a foraminot- 22 of 28 surviving patients were disease-free and
omy of the lowest one or two lumbar seg- 19 of surviving patients were independent in their
ments remaining should be performed, in activities of daily living; 20 patients required
order to avoid too much traction on the lum- early operation for wound healing; and 16 of
bar nerve roots to the remaining leg from the these 20 patients had a deep infection; in the
previous maneuver. Once instrumentation is patients requiring reoperation, a mean of three
in place, the femoral autograft from the reoperations was necessary; 4 patients in this
resected limb is used as a strut graft between cohort have been revised for instrumentation fail-
the supra-acetabular pelvis and the remaining ure [10]. These results pertain to very large resec-
tions, which disrupt spinopelvic continuity; much anterior spinal column fixation (14 patients), and
fewer complications and more favorable results spinopelvic fixation alone (8 patients). Compared
are seen with lesser sacral resections provided to the other two groups, the endoprosthesis group
appropriate margins are obtained [34]. had significantly better spinopelvic stability and
The current literature has not focused on implant survival with no greater intraoperative
pseudarthrosis in the setting of lumbopelvic hemorrhage or perioperative complications.
reconstruction, and the nonunion rate cannot be Authors found also radiological evidence of
assessed. Likewise, there is limited data on the implant osseointegration at a mean of 7.2 months.
mechanical failure rate. In a systematic review by However, the study’s retrospective design, the
Bederman et al. [12], it was shown that instru- small sample size and short follow-up period
mentation failure was evident in 16.1% of (mean 22.1 months), and the fact that some
patients (5 of 31 patients). Although there was no patients from the 3D implant group underwent
statistically significant difference, patients with- supplemental reconstruction at the time of sur-
out anterior column support tended to have high gery led to inevitable selection bias that cannot
mechanical failure rates (17.4% vs. 12.5%). This be ignored. Nevertheless, further research and
was also shown recently by Tang et al. [35]; in development of novel materials could be the
their study, 63 patients who underwent spinopel- future answer to the treatment of these extremely
vic reconstruction following total sacrectomy complex and challenging cases.
were studied. Postoperative mechanical failure of
the fixation occurred in 25% of patients, and the
factors associated with this failure were: single- References
rod instrumentation with single or double iliac
screws; posterior fixation without anterior aug- 1. Bourghli A, Boissiere L, Obeid I. Dual iliac screws
mentation; and female gender. in spinopelvic fixation: a systematic review. Eur
This evidence is suggestive of the potential Spine J. 2019;28(9):2053–9. https://doi.org/10.1007/
s00586-019-06065-3.
benefit of adding anterior column support to spi- 2. Esmende SM, Shah KN, Daniels AH. Spinopelvic fix-
nopelvic reconstruction after total sacrectomy, ation. J Am Acad Orthop Surg. 2018;26(11):396–401.
but since extensive instrumentation both anteri- https://doi.org/10.5435/JAAOS-D-15-00738.
orly and posteriorly requires more operative time 3. Kuklo TR, Bridwell KH, Lewis SJ, Baldus C, Blanke
K, Iffrig TM, Lenke LG. Minimum 2-year analysis of
and more sophisticated techniques, the potential sacropelvic fixation and L5-S1 fusion using S1 and
complications must be thoroughly discussed with iliac screws. Spine. 2001;26(18):1976–83. https://doi.
the patients [5]. org/10.1097/00007632-200109150-00007.
4. Varga PP, Szoverfi Z, Lazary A. Surgical resection and
reconstruction after resection of tumors involving the
sacropelvic region. Neurol Res. 2014;36(6):588–96.
14.6 Future Perspectives https://doi.org/10.1179/1743132814Y.0000000370.
5. Ishida W, Elder BD, Lo SL, Witham TF. Spinopelvic
So far, there have not been any methods/implants reconstruction following lumbosacral tumor resec-
tion. World Spinal Column J. 2016;7(1):25–30.
available for total or partial SI joint replacement; 6. Rose PS, Sim FH, Yaszemski MJ. Sacrectomy. In: Sim
thus, all kinds of stabilization are far from the FH, Choong PFM, Weber KL, editors. Orthopaedic
natural biomechanics. Current stabilization tech- oncology and complex reconstruction. Master tech-
niques try to ensure a stable fixation between the niques in orthopaedic surgery. 1st ed. Philadelphia,
PA: Lippincott Williams & Wilkins, a Wolters Kluwer
lumbar spine and the pelvis with metal or com- business; 2011.
bined systems [4]. Recently, investigators from 7. Mindea SA, Chinthakunta S, Moldavsky M, Gudipally
China reported the use of a 3D-printed sacral M, Khalil S. Biomechanical comparison of spinopel-
endoprosthesis after total en bloc sacrectomy [36, vic reconstruction techniques in the setting of total
sacrectomy. Spine. 2012;37(26):E1622–7. https://doi.
37]. In their series, the authors compared the org/10.1097/BRS.0b013e31827619d3.
reconstruction with 3D-printed prosthesis (10 8. Yu BS, Zhuang XM, Li ZM, Zheng ZM, Zhou ZY,
patients) to combined reconstruction, including Zou XN, Lu WW. Biomechanical effects of the extent
of sacrectomy on the stability of lumbo-iliac recon- 22. Zheng ZM, Zhang KB, Zhang JF, Yu BS, Liu H,
struction using iliac screw techniques: what level of Zhuang XM. The effect of screw length and bone
sacrectomy requires the bilateral dual iliac screw tech- cement augmentation on the fixation strength of
nique? Clin Biomech. 2010;25(9):867–72. https://doi. iliac screws: a biomechanical study. J Spinal Disord
org/10.1016/j.clinbiomech.2010.06.012. Tech. 2009;22(8):545–50. https://doi.org/10.1097/
9. Fujibayashi S, Neo M, Nakamura T. Palliative dual BSD.0b013e3181929e0c.
iliac screw fixation for lumbosacral metastasis. 23. Chang DW, Friel MT, Youssef AA. Reconstructive
Technical note. J Neurosurg Spine. 2007;7(1):99– strategies in soft tissue reconstruction after resection
102. https://doi.org/10.3171/SPI-07/07/099. of spinal neoplasms. Spine. 2007;32(10):1101–6.
10. Rose P, Yaszemski M, Delrutoski M, Wenger D, Sim https://doi.org/10.1097/01.brs.0000261555.72265.3f.
F. Classification of spinopelvic resections – implica- 24. Murakami H, Kawahara N, Tomita K, Sakamoto J, Oda J.
tions for surgical approach, procedural staging, recon- Biomechanical evaluation of reconstructed lumbosacral
struction, and patient outcomes. Orthop Proc Suppl spine after total sacrectomy. J Orthop Sci. 2002;7(6):658–
Bone Joint J. 94-B(SUPP_XXXVIII). 64. https://doi.org/10.1007/s007760200117.
11. Rose PS, Yaszemski MJ, Sim FH. A biomechani- 25. Gallia GL, Haque R, Garonzik I, Witham TF, Khavkin
cal approach to advances in sacropelvic reconstruc- YA, Wolinsky JP, Suk I, Gokaslan ZL. Spinal pelvic
tion. Hamdan Med J. 2018;11:193–7. https://doi. reconstruction after total sacrectomy for en bloc
org/10.4103/HMJ.HMJ_88_18. resection of a giant sacral chordoma. Technical note.
12. Bederman SS, Shah KN, Hassan JM, Hoang BH,
J Neurosurg Spine. 2005;3(6):501–6. https://doi.
Kiester PD, Bhatia NN. Surgical techniques for spi- org/10.3171/spi.2005.3.6.0501.
nopelvic reconstruction following total sacrectomy: a 26. Dickey ID, Hugate RR Jr, Fuchs B, Yaszemski MJ,
systematic review. Eur Spine J. 2014;23(2):305–19. Sim FH. Reconstruction after total sacrectomy:
https://doi.org/10.1007/s00586-013-3075-z. early experience with a new surgical technique.
13. Shikata J, Yamamuro T, Kotoura Y, Mikawa Y, Iida Clin Orthop Relat Res. 2005;438:42–50. https://doi.
H, Maetani S. Total sacrectomy and reconstruction org/10.1097/01.blo.0000180054.76969.41.
for primary tumors. Report of two cases. J Bone Joint 27. Cheng L, Yu Y, Zhu R, Lv H, Jia Y, Zeng Z, Chen B,
Surg Am. 1988;70(1):122–5. Ding Z. Structural stability of different reconstruction
14.
Tomita K, Tsuchiya H. Total sacrectomy techniques following total sacrectomy: a biomechani-
and reconstruction for huge sacral tumors. cal study. Clin Biomech. 2011;26(10):977–81. https://
Spine. 1990;15(11):1223–7. https://doi. doi.org/10.1016/j.clinbiomech.2011.06.003.
org/10.1097/00007632-199011010-00024. 28. Clark AJ, Tang JA, Leasure JM, Ivan ME, Kondrashov
15. Santi MD, Mitsunaga MM, Lockett JL. Total sacrec- D, Buckley JM, Deviren V, Ames CP. Gait-simulating
tomy for a giant sacral schwannoma. A case report. fatigue loading analysis and sagittal alignment fail-
Clin Orthop Relat Res. 1993;(294):285–9. ure of spinal pelvic reconstruction after total sacrec-
16. Allen BL Jr, Ferguson RL. The Galveston tech-
tomy: comparison of 3 techniques. J Neurosurg Spine.
nique for L rod instrumentation of the scoli- 2014;20(4):364–70. https://doi.org/10.3171/2013.12.
otic spine. Spine. 1982;7(3):276–84. https://doi. SPINE13386.
org/10.1097/00007632-198205000-00014. 29.
Papagelopoulos PJ, Mavrogenis AF, Rose PS,
17. Gokaslan ZL, Romsdahl MM, Kroll SS, Walsh
Yaszemski MJ. Amputative sacral resections. In: Sim
GL, Gillis TA, Wildrick DM, Leavens ME. Total FH, Choong PFM, Weber KL, editors. Orthopaedic
sacrectomy and Galveston L-rod reconstruction for oncology and complex reconstruction. Master tech-
malignant neoplasms. Technical note. J Neurosurg. niques in orthopaedic surgery. 1st ed. Philadelphia,
1997;87(5):781–7. https://doi.org/10.3171/ PA: Lippincott Williams & Wilkins, a Wolters Kluwer
jns.1997.87.5.0781. business; 2011.
18. Jackson RJ, Gokaslan ZL. Spinal-pelvic fixation in 30. Fuchs B, Dickey ID, Yaszemski MJ, Inwards CY,
patients with lumbosacral neoplasms. J Neurosurg. Sim FH. Operative management of sacral chordoma.
2000;92(1 Suppl):61–70. https://doi.org/10.3171/ J Bone Joint Surg Am. 2005;87(10):2211–6. https://
spi.2000.92.1.0061. doi.org/10.2106/JBJS.D.02693.
19. Eck JS, Yaszemski MJ, Sim FH. Sacrectomy and
31. Gunterberg B, Kewenter J, Petersen I, Stener

spinopelvic reconstruction. Semin Spine Surg. B. Anorectal function after major resections of the
2009;21(2):99–105. https://doi.org/10.1053/j. sacrum with bilateral or unilateral sacrifice of sacral
semss.2009.03.009. nerves. Br J Surg. 1976;63(7):546–54. https://doi.
20. Varga PP, Bors I, Lazary A. Sacral tumors and man- org/10.1002/bjs.1800630713.
agement. Orthop Clin North Am. 2009;40(1):105– 32. Todd LT Jr, Yaszemski MJ, Currier BL, Fuchs B, Kim
123, vii. https://doi.org/10.1016/j.ocl.2008.09.010. CW, Sim FH. Bowel and bladder function after major
21. Yu BS, Zhuang XM, Zheng ZM, Li ZM, Wang TP, Lu sacral resection. Clin Orthop Relat Res. 2002;397:36–9.
WW. Biomechanical advantages of dual over single https://doi.org/10.1097/00003086-200204000-00006.
iliac screws in lumbo-iliac fixation construct. Eur 33.
Gunterberg B, Norlen L, Stener B, Sundin
Spine J. 2010;19(7):1121–8. https://doi.org/10.1007/ T. Neurologic evaluation after resection of the sacrum.
s00586-010-1343-8. Investig Urol. 1975;13:183–8.
34.
Zileli M, Hoscoskun C, Brastianos P, Sabah prosthesis after total en bloc sacrectomy: a techni-
D. Surgical treatment of primary sacral tumors: com- cal note. Eur Spine J. 2017;26(7):1902–9. https://doi.
plications associated with sacrectomy. Neurosurg org/10.1007/s00586-016-4871-z.
Focus. 2003;15(5):E9. 37. Wei R, Guo W, Yang R, Tang X, Yang Y, Ji T, Liang
35. Tang X, Yang R, Qu H, Cai Z, Guo W. Factors asso- H. Reconstruction of the pelvic ring after total en
ciated with spinopelvic fixation mechanical failure bloc sacrectomy using a 3D-printed sacral endopros-
after total sacrectomy. Spine. 2018;43(18):1268–74. thesis with re-establishment of spinopelvic stabil-
https://doi.org/10.1097/BRS.0000000000002635. ity: a retrospective comparative study. Bone Joint J.
36. Wei R, Guo W, Ji T, Zhang Y, Liang H. One-step 2019;101-B(7):880–8. https://doi.org/10.1302/0301-
reconstruction with a 3D-printed, custom-made 620X.101B7.BJJ-2018-1010.R2.
Reconstruction vs. No
Reconstruction for Pelvic 15
Resections
Andreas F. Mavrogenis, Dimitrios Papadopoulos,
Andreas G. Tsantes, Andrea Angelini,
Panayiotis J. Papagelopoulos, and Pietro Ruggieri
15.1 Introduction radiographic findings in early stages are not diag-

nostic, so these tumors are usually detectable on
Approximately 5–10% of all primary bone radiographs when they have already invaded an
tumors are located in the pelvis and pose a chal- extensive part of the pelvis and have resulted in
lenging problem for tumor orthopedic surgeons significant bone destruction or sclerosis.
[1]. The most common types of pelvic sarcomas Subsequently, initial detection of pelvic tumors is
are chondrosarcoma, Ewing’s sarcoma, and usually made with advanced imaging including
osteosarcoma, while chordoma is the most com- computer tomography (CT), magnetic resonance
mon primary bone tumor in the sacrum [2]. (MR) imaging, and bone scintigraphy.
Although pelvic bone metastases are not rare, As a result of the late diagnosis, the operation
most metastatic tumors in the pelvis are managed for adequate oncologic resection of pelvic tumors
with radiation and pain palliation, and they do not with safe margins may be exceptionally difficult.
require surgical treatment. Due to significant dif- The late diagnosis and wide spread of these tumors
ficulties in terms of diagnosis and surgical treat- at the time of initial detection, the highly demand-
ment of pelvic tumors, the survival rate for these ing procedures for complete resection of pelvic
patients is much lower compared to malignancies tumors, and the high rate of complications follow-
in the extremities [2]. Diagnosis of these lesions ing these procedures are some of the causative fac-
on a clinical base is extremely difficult since due tors for the poor prognosis for these patients.
to their deep location most pelvic tumors are
impalpable, while they can substantially extend
without inducing local symptoms. Moreover, the 15.2 Examination and Imaging
The main symptom associated with pelvic tumors

A. F. Mavrogenis · D. Papadopoulos is a vague, often poorly localized pain around the
A. G. Tsantes · P. J. Papagelopoulos
First Department of Orthopaedics,
pelvis that may also radiate downward to the
National and Kapodistrian University of Athens, limbs due to compression, irritation, and/or
School of Medicine, Athens, Greece encasement of the peripheral nerve roots and
e-mail: [email protected]; [email protected] major vessels. A palpable mass that is the second
A. Angelini · P. Ruggieri (*) most common symptom of all tumors is absent in
Department of Orthopedics and Orthopedic most cases, especially during the early stages.
Oncology, University of Padova, Padova, Italy
e-mail: [email protected]; These tumors may become palpable only when
[email protected] they have increased in size substantially.

166 A. F. Mavrogenis et al.
Radiographs are not diagnostic in most early around the ischium, an open biopsy is usually
cases since radiographs have a low sensitivity for performed through a Ludloff approach, while for
pelvic tumors in early stages, especially for the tumors in the pubic rami, an open biopsy can be
inexperienced physician. In patients with indica- performed through an anterior approach medially
tive symptoms for a pelvic tumor or when radio- to the neurovascular bundle. Regardless of the
graphic signs of malignancy are suspected on biopsy method, the route for accessing the tumor
simple radiographs, further imaging with MR during biopsy must be the same with the route of
imaging and CT should proceed. Although a the surgical approach that will be used for the
complete imaging evaluation of a lesion can con- final resection because the biopsy tract (either
firm or exclude malignancy with enough cer- closed or open) should be excised with the final
tainty, bone biopsy is always mandatory. tumor specimen.
15.3 Biopsy 15.4 Preoperative Management
Biopsy must always follow the imaging evalua- A multidisciplinary approach is required, as nec-
tion of a suspicious pelvic lesion in order to con- essary for any malignancy. The multidisciplinary
firm diagnosis and identify the histological type team may include urologists, vascular surgeons,
of the lesion. Due to high risk for contamination colorectal surgeons, orthopedic surgeons, neuro-
of the surrounding tissue with cancer cells during surgeons, plastic surgeons, and spine surgeons
biopsy, the procedure must be performed only by [3–8]. After histological confirmation of the
experienced oncologic surgeons. Unfortunately, tumor, a surgical staging protocol is recom-
several incidents have been described in which an mended using comprehensive imaging evalua-
improperly performed biopsy by unexperienced tion; lung CT, abdomen CT, and a whole-body
surgeons resulted in inoperability of a tumor that bone scintigraphy are mandatory in order to
was initially resectable. Contamination of the ret- investigate whether bone, lung, or intra-abdominal
roperitoneum with cancer cells during biopsy is a metastases are present (Table 15.1). Moreover, a
detrimental event that can alter and/or worsen sonography of the inguinal and para-aortal lymph
patients’ survival [3]. The success of a biopsy nodes is helpful for detection of cancerous infil-
depends mostly on the accurate targeting of the tration as this adverse event can significantly alter
involved area and the proper placement of the the surgical plan and the overall management of
biopsy tract at this area; therefore, an imaging- the patients. Preoperative evaluation may also
guided closed biopsy technique is strongly
recommended.
Table 15.1 Imaging studies included in staging protocol
Biopsy can be performed under CT guidance
for pelvic malignancies
using a 3–6-mm gauge needle (core biopsy tech-
Imaging study Investigation
nique with a tru-cut® needle). Tumors in the iliac
Lung CT Lung metastases
wing can be easily accessed through the iliac Abdomen CT Abdominal metastases
crest, whereby the tip of the biopsy needle should Whole-body bone Remote bone metastases
follow a subperiosteal route. The biopsy speci- scintigraphy
men should additionally include a thin layer of Sonography of regional Inguinal and para-aortal
the soft tissue that covers the medial aspect of the lymph nodes lymph nodes infiltration
iliac wing [4]. When a needle biopsy is not diag- MR angiography or Infiltration of iliac
intravenous pyelography vessels
nostic, we recommend to repeat the closed biopsy (IVP)
technique with imaging guidance and coopera- Pyeloureterogram or Ureteral or cyst
tion with the interventional radiologist; if the cystoscopy cancerous invasion
repeat closed biopsy technique is nondiagnostic, Rectoscopy Rectal cancerous
an open biopsy is indicated. For lesions located invasion
15 Reconstruction vs. No Reconstruction for Pelvic Resections 167
include MR angiography or intravenous pyelogra- 15.5 Surgery and Resection Types

phy (IVP) to assess possible invasion of the iliac
vessels, to evaluate the anatomic relationship of Wide resection (resection margins outside the
the tumor with these vessels, and to check the vas- reactive zone of the tumor in healthy tissue) is the
cular supply of soft tissue flaps that may be used recommended type of oncologic resection for
for wound closure. Based on the site and the pelvic sarcomas. There is a general consent that
extent of the lesion, further studies may also be for a wide resection a healthy surrounding osse-
required such as pyeloureterogram, cystoscopy, ous area of approximately 2–3 cm must be
and rectoscopy. In case of ureteral involvement, included in the resected specimen, while when
consultation by an urologist is necessary, and an the tumor invades the surrounding soft tissue, a
ureter stent or a pigtail catheter may be required. more extensive area of normal soft tissue of up to
Regarding the preoperative laboratory evaluation 5 cm is required. In general, small pelvic tumors
of patients undergoing tumor resections, 500– are addressed with partial (limited) pelvic resec-
1000 cells/ml for absolute neutrophil count and tions, while larger tumors are managed with (1)
50,000 cells/ml for platelet count are considered external hemipelvectomy with hindquarter ampu-
relatively safe cutoff values for an adequate post- tation, (2) internal hemipelvectomy with preser-
operative immunologic response and coagulation vation of the limb, (3) sacrectomy for sacral
of the patients [1]. Due to the significantly tumors, and (4) extended hemipelvectomy (exter-
increased risk for venous thromboembolism in nal hemipelvectomy with additional resection of
these patients, several tertiary tumor centers now the lower lumbar spine). Depending on the
routinely insert a vena cava filter in all patients resected section of the pelvis and whether this
undergoing major pelvic resections [9]. section involves a weight-bearing or moving ele-
Based on the histology of the tumor, neoadju- ment of the pelvis, the resulted bone defect may
vant chemotherapy or radiation therapy may be be reconstructed or not.
recommended since most pelvic sarcomas such as The techniques for reconstruction of pelvic
Ewing sarcomas are sensitive to neoadjuvant bone defects are broadly classified into biological
treatments for tumor response, facilitation of procedures such as bone grafting, and nonbio-
tumor resection, and improvement of patients’ logical methods such as endoprosthetic recon-
survival. In certain cases, some inoperable tumors structions. Before the 1970s, almost all tumors
may even become operable after neoadjuvant che- involving critical parts of the pelvis were
motherapy. However, there are types of sarcoma addressed with external hemipelvectomy and
such as chondrosarcomas that are not sensitive to amputation. Nowadays, due to the significant
(neo-)adjuvant chemotherapy or radiotherapy. A advances in the fields of bioengineering, imaging
safety hold-off period of 2–5 weeks after the last techniques and adjuvant treatments internal
chemotherapy session and of 4–5 weeks after the hemipelvectomy with some forms of pelvic
last radiotherapy session is usually recommended reconstruction are considered a favorable option
for the surgical procedure to be implemented [6]. without compromising the survival of the patients
Restaging after the neoadjuvant therapy should be [1]. Regardless of the method, it is of great sig-
performed. This new evaluation will set the final nificance to ensure a sufficient musculature for
margins in three dimensions (sagittal, coronal, wound closure and reconstruction coverage.
and frontal) and six planes (proximal, distal, ante- Although the advances in microsurgical flaps
rior, posterior, lateral, and medial) for the proper over the past decades have significantly decreased
resection of the tumor [3, 4, 6–8, 10]. At this the rate of wound healing complications, these
point, using novel technologies, a 3D pelvic complications with their devastating repercus-
model can be designed based on the restaging sions are still very common [12–17]. The treat-
imaging. This approach allows reevaluation of the ment plan regarding the preoperative
planning of resection and the possible reconstruc- management, type of resection, and subsequent
tion of the resulting bone defect [11]. reconstruction of the bone defect should be based
Table 15.2 Enneking and Dunham classification for pel- Type H3 when the proximal femur is resected [4,
vis resections
8]. Last, when the tumor not only involves the
Resection type Pelvic region posterior part of the pelvis and sacrum but also
Type I Iliac wing extends to the lower lumbar spine, a more exten-
Type II Periacetabular area
sive procedure called extended hemipelvectomy
Type III Pubic rami
is required. This procedure includes external
Type IV Sacrum
hemipelvectomy and resection of ilium, sacrum,
Femoral resections are designated as follows: H1 for fem-
oral head resections, H2 for peritrochanteric area resec- and part of the lower lumbar spine. Certain details
tions, H3 for proximal femoral resections about the surgical approach, the extent of resec-
tion, and the methods for reconstruction of the
bone defects must be set out based on the
Enneking and Dunham classification.
15.6 External Hemipelvectomy

Versus Limb-Salvage
The decision about lower limb amputation versus

limb-salvage surgery is complex, and surgeons
are called to decide whether an adequate wide
resection is feasible without ending up with a
severely compromised lower limb. When sur-
geons decide to proceed with limb-salvage sur-
Fig. 15.1 Enneking and Dunham classification of pelvic gery, the postoperative functional results should
resections
always be aimed to be superior to those of an
external hemipelvectomy [21, 24–27]. Although
on consideration of the anatomic location and there are several well-defined prerequisites that
extent of the tumor, the age and comorbidities of must be fulfilled for a limb-salvage procedure, in
the patients, and the overall prognosis of the many cases, fulfillment of all these criteria is
malignancy [7, 18, 19]. debatable. The main contraindications for a limb-
The most widespread and efficient surgical salvage procedure include (1) recurrence of the
classification system for pelvic resections is the tumor after a previous limb-salvage procedure
Enneking and Dunham classification by which unless a wide resection can be definitely achieved
pelvic resections are classified into four types and the overall status of the patients allows for a
(Table 15.2) [20]; Type I resection involves the new limb-sparing procedure, (2) extension of the
ilium, Type II the periacetabular area, Type III the tumor in the sacral foramens and involvement of
pubic rami, and Type IV the sacrum (Fig. 15.1). the nerve roots, (3) extension of the tumor in sci-
Several combinations of these types can be made, atic notch and involvement of the sciatic nerve,
and a pelvic resection that includes more than and (4) infiltration of iliac vessels resulting in
one type is defined by the combined number of nonrepairable vascular lesions (Table 15.3).
the individual types (e.g., Type II–III resection). Although femoral involvement or extensive infil-
The resection that includes the largest part of the tration of the psoas muscle is not considered con-
hemipelvis (Type I–II–III) is called hemipelvec- traindications for internal hemipelvectomy, the
tomy [21–23]. When part of the femur is also pros and cons of such a major procedure with a
resected, the resection is designated as Type H significantly compromised lower limb afterward
and is further subclassified into three types: Type must be considered with caution. According to
H1 when the femoral head is resected, Type H2 O’Connor and Sim the acetabulum, the sciatic
when the pertrochanteric area is resected, and nerve and the femoral neurovascular bundle are
Table 15.3 Indications for external hemipelvectomy Table 15.4 Reconstruction of bone defects after Type I
resections
Indications
Extension of the tumor in the sacral foramens and Type I resection Reconstruction
involvement of the nerve roots Partial iliac resection Resection without
Extension of the tumor in sciatic notch and without pelvic ring reconstruction
involvement of the sciatic nerve disruption
Infiltration of iliac vessels resulting in nonrepairable 1. Iliac crest tumors
vascular lesions 2. Small tumors of the
iliac wing
Recurrence of the tumor after a previous limb-salvage
Complete iliac resection
procedure (unless a wide resection can be definitely
with pelvic ring disruption
achieved and the overall status of the patients allows for a
new limb-sparing procedure) 1. Tumors involving the Osseous reconstruction
largest part of the with autograft or
ilium allograft
three major anatomic elements of the pelvis, and 2. Tumors extending to Sacroiliac arthrodesis
sacroiliac joint with grafts
when two of these three must be resected, exter-
nal hemipelvectomy should be considered [28].
required. Similar to these tumors, more distally
located tumors that are small in size and can be
15.6.1 Type I Resections: Os Ilium adequately resected without disrupting the conti-
nuity of the iliac wing can be also solely resected.
Tumors in the iliac wing can be accessed through
the posterior part of the ilioinguinal approach; a 15.6.1.2 Complete Resection
simpler approach through an incision along the of the Ilium with Pelvic Ring
iliac crest is usually adequate for smaller tumors Disruption
in the iliac crest. The resection of these tumors is Bone defects after pelvic resections that involve
achieved through two osteotomies, one proximal the largest part of the supra-acetabular ilium or
and one distal. The distal iliac osteotomy is at the extend to the sacroiliac joint destabilize the pel-
supra-acetabular level and can be carried out vic ring and usually lead to certain functional
through the sciatic notch, while the proximal iliac deficits. In most of these cases, bone defects are
osteotomy is usually adjacent to the iliosacral managed with some form of bone reconstruction
joint. On the proximal osteotomy site, the ilio- or arthrodesis (sacroiliac arthrodesis) to support
lumbar ligament that marks the position of the L5 the two sides of the resection (ilium–ilium or
root (the root is located inferior and medially to sacrum–ilium) and to provide adequate pelvic
the ligament) must be identified and usually ring stability. This can be achieved with grafts,
needs to be cut for adequate exposure. In some metallic implants, or a combination of both.
cases, the proximal extent of the tumor is too Usually, a tibial or a fibular autograft is harvested
close to the sacroiliac joint, and a wide resection and placed at the resection site in order to bridge
can only be achieved through disarticulation of the defect. The one side of the graft is fixed at the
the sacroiliac joint [3, 4, 6–8]. The subsequent supra-acetabular osteotomy site, while the other
reconstruction of the bone defect following Type at the proximal iliac osteotomy or at the sacral
I resections depends on the integrity of pelvic osteotomy. In case of a very large bone defect,
girdle (Table 15.4). massive allografts can also be used.
The functional outcomes after these recon-
15.6.1.1 P artial Resection of the Ilium structive procedures are reported to be excellent,
Without Pelvic Ring while complication rates are low. Although
Disruption reconstruction for these defects is recommended
When iliac resection is only partial and does not by most authors, Beadel et al. compared the
disrupt the pelvic continuity as with tumors results with and without reconstruction of bone
located in the iliac crest, reconstruction is not defects after iliosacral resections (4 and 12
Table 15.5 Reconstruction Reconstruction Techniques

of bone defects after Type II Nonreconstruction 1. External hemipelvectomy
and Type II–III resections 2. Iliofemoral arthrodesis
3. Resection arthroplasty
4. Hip transposition technique
Reconstruction 1. Biological reconstruction
(a) Bulk pelvic allograft
(b) Autoclaved or irradiated pelvic autograft
2. Endoprosthetic reconstruction
(a) Saddle prosthesis
(b) Stemmed cup
Reverse ice-cream cone implant
Pedestal cup
(c) Custom-made implant
(d) Allograft/prosthetic composite
patients, respectively) and suggested that recon- around the base of the upper pubic rami, and the
struction to restore pelvic stability after sacroiliac posterior osteotomy is at the proximal part of the
resections is not always necessary [29]. posterior acetabular column or more distally at
Interestingly, in this case–control study, although the ischium. Both reconstruction and nonrecon-
the functional scores (Toronto Extremity Salvage struction options, and biological methods, and
Score and Musculoskeletal Tumor Society), the endoprosthetic options are available for the man-
rates of local recurrence, and survival were simi- agement of the periacetabular bone defects
lar between the groups, patients without recon- (Table 15.5).
struction had a lesser need for walking supports,
required less pain medications, and were more 15.6.2.1 Nonreconstruction Methods:
likely to return to work [29]. Therefore, any Iliofemoral Arthrodesis
reconstruction should be considered with Iliofemoral arthrodesis was a very popular
caution. method for Type II or Type I–II resections during
the previous decades. By this type of arthrodesis,
the stability of the pelvic ring is maintained,
15.6.2 Type II Resections: while the patients preserve a certain degree of a
Periacetabular painless limb function [30]. The disadvantages of
iliofemoral arthrodesis include loss of joint func-
For tumors located around the acetabulum with tion, constant use of gait support, limb length dis-
or without extension to the proximal femur, a crepancy, and long consolidation times [31, 32].
Type II pelvic resection including the acetabulum The functional results of this operation are sig-
is warranted. This type of resection is a highly nificantly better when the proximal iliac osteot-
demanding procedure, and good functional out- omy is close to the native acetabulum so that limb
comes necessitate restoration of the native hip shortening is minimized and successful fusion is
biomechanics. The hip center of rotation and the more likely. However, the attempt to maintain a
force transmission along the pelvic ring must be large part of the ilium may compromise safe
restored so that the weight-bearing pattern resections margins and result in an increased risk
throughout the gait cycle is not altered [20]. For for local recurrence. For the fusion between the
this type of resection, three osteotomies are per- proximal femur and the remaining pelvis, a com-
formed. The superior osteotomy at the supra- bination of hardware can be used including
acetabular area is usually performed through the plates, cerclage wires, and cables [33–35].
greater sciatic notch, the anterior osteotomy is Moreover, the fusion can be augmented with
performed at the anterior acetabular column bone grafts that are fixed along the fusion site.
Commonly used autografts include the iliac crest reported satisfactory results regarding LLD
graft and the free fibula graft, either avascular or (average, 2 cm) and postoperative functional sta-
vascularized. Although the use of the free vascu- tus [39]. In another study, the total cost of three
larized fibula graft increases the complexity of different methods (external hemipelvectomy
the procedure and requires microsurgical skills, with use orthotic device, internal hemipelvec-
the reported union time with this technique is sig- tomy with endoprosthetic reconstruction, and
nificantly decreased and stability of the pelvic iliofemoral arthrodesis) for the management of
ring is enhanced [36–39]. In cases that part of the bone defects after oncologic pelvic resections
pubis or the ischium needs to be resected as well was compared; iliofemoral arthrodesis was
(e.g., Type II–III resections), additional fusion of clearly associated with a significantly lower cost
the femur to the ischium or to the pubic bone is compared to the other two methods [43].
recommended [8, 40]. Last, another fusion tech-
nique that can be used in selected cases is the 15.6.2.2 Nonreconstruction Methods:
tibia–hindfoot rotationplasty with calcaneopelvic Resection Arthroplasty
arthrodesis. This technique involves partial resec- Similar to the Girdlestone procedure, the method
tion of the pelvis, resection of femur, and preser- of resection arthroplasty (or flail hip technique)
vation of the lower limb below the knee. The after oncologic pelvic resections is an alternative
remaining lower limb is rotated 180°, the midfoot option that was also very popular in the early
and forefoot are resected, and the calcaneus is days of pelvic tumor surgery and is gaining sup-
fixed to the pelvic osteotomy site [35, 41]. port again [44]. As with any limb-salvage proce-
There is a large body of evidence regarding dure, this method is indicated only when wide
the results of iliofemoral arthrodesis. One of the resection is possible without significant compro-
first large studies for this technique included 60 mise of the lower limb function. According to
patients with iliofemoral fusions. The results in this technique, the involved acetabular area is
this study were disappointing since pseudarthro- solely resected resulting in a flail hip without a
sis occurred in 50% of the patients [28]. In supporting pelvic articular surface. The aim of
another study by Fuchs et al. enrolling 32 patients this surgery is to establish a fibrous union between
who treated by this method, solid fusion was the remaining ilium and the femur as opposed to
done in 21 patients, while primary pseudarthrosis arthrodesis in which an osseous union is aimed.
with adaption of the femur to the ilium was done As expected, this surgery is associated with sig-
in 5 patients [42]. In the group of patients with nificantly limb shortening which can be addressed
primary iliofemoral arthrodesis, failed fusion and with either shoe blocks in mild cases or distrac-
pseudarthrosis were observed in 14% of the tion osteogenesis in more severe cases. Even
cases. In terms of functional results, patients with though unassisted ambulation has been reported
primary pseudarthroses reported significantly for patients treated in this fashion, supporting
lower values in the Musculoskeletal Tumor walking aids are usually required [45]. The func-
Society Score (MSTS) than those with solid tional results of resection arthroplasty are greatly
fusions. All patients in this study regardless of depended on the extent of the resected ilium [46].
the success of the fusion reported mild low back Although this method is not very common today,
pain, probably due to altered forces on their lum- the low rates of complications in terms of wound
bar spine. The average leg length discrepancy healing problems and mechanical failures have
(LLD) in this series was 4.8 cm that was compen- led to a renaissance of this method during the
sated with shoe lifts. The authors concluded that past years, especially in severely compromised
iliofemoral arthrodesis should be preferred in patients.
young patients with more demanding functional Schwartz et al. evaluated the capacity for
activities, while primary pseudarthrosis must be independent walking and the functional out-
performed in older patients. In a smaller study comes after resection arthroplasty in eight
evaluating the results of this method, the authors patients, with a 2.9 years minimum follow-up
[45]. The authors reported excellent results; at geometry of the corresponding pelvic defect, in
the last follow-up, seven of the eight patients theory they can be used to reconstruct the normal
were free of pain and able to walk without sup- pelvic anatomy after almost any type of resec-
ports, while the overall mean MSTS score was tion. In contrast, a prerequisite for the use of the
73.3% (range, 53.3–80.0%). resected bone as an autograft is tumor denatur-
ation that can be achieved through several tech-
15.6.2.3 Nonreconstruction Methods: niques including autoclave, pasteurization,
Hip Transposition Technique freezing, and extracorporeal irradiation.
In the hip transposition technique, the hip joint is Following the process of tumor denaturation with
transferred cranially to the level of the proximal these techniques, the graft is reimplanted to fit
osteotomy. The inferior part of the acetabulum into the defect. This method has certain advan-
(in cases that it can be preserved) is rotated 90° tages compared to allografts such as the easy
and fixed with soft tissue or textile implants at the accessibility of the graft since bone bank is not
proximal osteotomy site in order to form a required, and the optimum fit of the graft into the
pseudo-joint. If the acetabulum is completely defect [59–61]. Moreover, there is no need for
resected, the femoral head can be wrapped into immunosuppression (which decreases the poten-
an artificial capsule that is attached with bone tial for bone healing), while the risk of infection
anchors to the remaining ilium [47]. Even though is also lower [58, 62, 63]. Compared to nonrecon-
this technique has fairly good functional results, struction methods, biological reconstruction has
limb shortening as with any nonreconstruction the obvious advantage of preservation of the hip
technique is a significant drawback [48]. joint function, while compared to endoprosthe-
Distraction osteogenesis for the management of ses, reattachment of the surrounding musculature
the resulting LLD has been recommended by is easier [64]. However, grafts have been linked
many surgeons [49]. Revision surgery after the with high rates of serious complications such as
hip transposition technique is rarely reported, and infections or mechanical complications such as
the rates of postoperative complications are sig- failure of consolidation [2, 10, 56].
nificantly lower compared to more advanced In a large study of 945 patients investigating
methods for reconstruction of bone defects such the factors influencing graft consolidation, the
as the endoprosthetic reconstruction and the use nonunion rate was reported 17.3% [65]. A signifi-
of structural allografts [50, 51]. Hillman et al. cant risk factor for failure of consolidation was
reported the outcomes of hip transposition in 17 the suboptimal fit of the graft into the defect and
patients after Type II, Type I–II, Type II–III, and the subsequent large interfragmentary gaps. The
Type I–II–III resections [50]. LLD was evident in overall infection rate of allografts ranged from
four patients. Although three of these patients 10% to 33% [28, 66]. Some authors recommend
(18%) had postoperative skin problems, the over- impregnation of the grafts with antibiotics such
all complication rate was very low. The authors as rifampicin in order to reduce infectious com-
highlighted that no incidence of deep infection or plications [67]. Dellove et al. in their study using
local recurrence occurred in this series. structural bone allografts after periacetabular
oncologic resections in 24 patients reported a
15.6.2.4 Reconstruction Methods: high rate of complications (11 patients, 46%),
Graft Implantation requiring a revision surgery in all cases [68].
Biological reconstruction of the bone defect Regarding the functional outcomes in this study,
refers to application of a structural bone graft for authors reported an average MSTS score of 73%,
substitution of the resected bone segment. This but only half of the patients were capable for
graft may be either the resected pelvic segment independent walking without crutches. The
reimplanted as an autograft or a pelvic massive authors also highlighted that the functional out-
allograft [10, 52–58]. Since allografts can be comes were clearly better for the younger patients
shaped intraoperatively to copy the complex (average MSTS score 82% in patients <20 years
old compared to average MSTS score 65% in the simulates a saddle hinges over this notch [70].
elderly) [68]. In an another study of 18 patients Initial results of this endoprosthesis in terms of
with irradiated autografts after periacetabular mechanical complications such as loosening, hip
pelvic resections, the authors reported good func- dislocation, and intraoperative or postoperative
tional results with an average MSTS score of fractures were very discouraging. A major cause
73% and an average TESS of 71% [69]. The for these poor results was the final eccentric posi-
authors also reported that three revision surgeries tion of the artificial joint. The high rates of com-
were performed in three patients with deep infec- plications led to many modifications of these
tions; although graft union was difficult to identi- implants; newer implants had the advantage of
fied, there was no evidence of nonunion that modularity, offering more options in terms of res-
required a revision surgery [69]. toration of the native hip biomechanics [80].
Aboulafia et al. studied the outcomes of sad-
15.6.2.5 Reconstruction Methods: dle prostheses in 17 patients with pelvic tumors
Endoprostheses [70]. Functional results in their study were
Endoprostheses for periacetabular reconstruc- reported to be excellent or good in 12 patients
tions are the first option in most practices with and fair or poor in 5 patients. Dislocation rate
many different types of endoprostheses devel- was high (53%), whereas other complications
oped over the last decades [57, 70–78]. A solid included wound healing problems and infections.
and stable fixation of the prosthesis into the Implant loosening and cranial migration of the
remaining pelvis is a prerequisite for successful prosthesis were also frequently reported in the
outcomes. Also, in order to increase hip joint sta- long term. Some of the critical points for
bility and decrease dislocation rates, large femo- improved functional outcomes for these implants
ral head sizes have been recommended, while the were preservation of a sufficient iliac notch and a
peripelvic musculature including the gluteal fas- proper selection of the implant length. In this
cia must be preserved and reattached to femur way, the muscle tension of the iliopsoas and the
when possible [8]. Until the development of a hip abductors is restored and the stability between
pseudocapsule around the prosthetic joint, artifi- the pelvis and femur is increased. In line with
cial ligaments can also be used to augment stabil- this, poor iliac bone stock and suboptimal status
ity and musculature reattachment, although their of the psoas and the hip abductor muscles should
efficacy is questioned. Last, for the restoration of be considered contraindications for the use of
the native hip biomechanics the final position of these implants.
the acetabular cup must be symmetrical with the
opposite site in terms of height, lateral distance, Custom-Made Pelvic Prosthesis
and orientation [79]. Due to the high rates of infection following the
use of allograft reconstructions and the need for
Saddle Prosthesis improved functional results, modular custom-
Saddle prosthesis is one of the first implants that made endoprotheses were widely distributed in
was developed for reconstruction after Type II the middle 1990s for reconstruction of periace-
and Type II–III pelvic resections. These implants tabular bone defects [2, 50, 78]. For the g eneration
were initially developed and used for reconstruc- of such implants, a preoperative CT of the pelvis
tion of large pelvic defects after failed total hip is required, and a 3D pelvic model is formed onto
arthroplasties or after resections in infected which the resection margins are simulated and
reconstructions [73]. By the late 1990s though, marked and a modular custom-made pelvic endo-
there was already a sufficient body of evidence prosthesis is subsequently manufactured. The
regarding the use of these implants after onco- theoretical advantage of this method is that since
logic pelvic resections [72]. For the insertion of the endoprosthesis matches exactly to the defect
this implant, a notch is created in the remaining and replaces with high precision the resected
ilium and the proximal part of the prosthesis that bone segment, the native hip function is suppos-
edly completely restored. To achieve that though, similar to that of a long bone with a thick outer
the intraoperative resection must be copied as cortex and a dense trabecular medulla, the ratio-
much as possible to the predesigned resection on nale for the use of these implants is that their
the pelvic model. Despite the initial enthusiasm stem that is inserted into the ilium will provide
and the excellent functional results, the long-term the required rigid fixation of the implant to the
outcomes were not as expected as these bulk remaining pelvis. The two main types of stemmed
implants were associated with certain complica- cups are the reversed “ice-cream cone-style”
tions including loosening, migration, and implant prostheses and the pedestal cups [84, 85].
failure. Moreover, the large dead space that was Although the main principles involved in the
created after insertion of these implants resulted design of these two implants are similar, the
in high rates of hematoma formation and postop- newer types of pedestal cups such as the LUMiC®
erative infection. The high rates of these compli- endoprosthesis are more versatile regarding their
cations led in the late 2000s to a significant stem length, orientation, and diameter of the cup.
decline in the use of these implants [2]. However, These endoprostheses require only a small cra-
due to the precise preoperative planning of the nial part of the ilium, while the proper orientation
resection margins and the extensive pelvic resec- of the stem restores load transmission along the
tion that is required for these implants, these iliosacral joint; therefore, the anatomical and bio-
modular endoprostheses were associated with mechanical continuity between the spine and the
decreased rates of local tumor recurrence [59]. lower limb is undisrupted. There are several other
There are many series about custom-made pros- advantages that are associated with these implants
theses after oncologic pelvic resections [71, 74]. such as the fact that restoration of the continuity
Hillman et al. reported the outcomes and complica- of the pelvic ring is not necessary since the
tions of several reconstructive techniques after pel- weight-bearing load is transferred to the socket
vic resections [50]. In 16 patients, a custom-made and the high versatility regarding the stem length
prosthesis was inserted and the complication rate and orientation of the cup. This versatility pro-
was significantly high (62.5%). The infection rate vides many options for restoration of the native
was 38%, while local recurrence occurred in 3 hip center of rotation and limb length. Moreover,
patients, and 1 patient presented with wound heal- due to the small size of these implants, the
ing problems. The authors noted that 30 additional resulted dead space is substantially decreased,
procedures were required in these patients (average and soft tissue coverage is more easily achieved,
1.9 revision operations per patient). if necessary. Last, these implants are commer-
cially available in a wide variety of stem length
Stemmed Acetabular Prostheses (Cups) and cup orientation; thus, there is no need for pre-
Although saddle prostheses and custom-made operative time for planning the construction and
implants have been used for many years in pelvic manufacturing as with custom-made implants.
tumor surgery, due to the high complication rates Secondary to these features, there is also a sig-
and the high cost of these implants, there was a nificant decrease in complications such as hema-
growing skepticism about their use [81, 82]. To toma formation and infection [83]. However,
overcome these concerns, a different type of despite the precise restoration of the native hip
endoprostheses that was initially developed and biomechanics, the functional results of these
used in revision surgery after failed THA gained implants widely vary.
ground over the past years [75, 83]. These acetab- Pedestal cups were initially developed for
ular prostheses were called stemmed cups and revision surgery in patients with failed
are composed of an inferior shell and a large THA. There are only few studied in the literature
superior stem that is fixed proximally into the regarding the use of these cups after oncologic
intramedullary space of the ilium aiming to the periacetabular resections [80, 86–88]. In the larg-
posterior superior iliac spine or the sacroiliac est study of 48 patients with pelvic tumors and a
joint. Since the structure of the iliac isthmus is 6-year follow-up, Hipfl et al. [86] reported the
outcomes of a pedestal cup (Schoellner cup; Interestingly, the authors reported that none of the
Zimmer Biomet Inc., Warsaw, Indiana) for onco- ten patients required revision surgery for implant
logic pelvic reconstruction. The complication removal; thus, the failure incidence at 2 years was
rate in this study was high, including deep infec- 0% [85]. In another study by Fisher et al. ice-
tion in 17% of patients, hip dislocation in 15% of cream cones-style implants were inserted in 27
patients, and aseptic loosening in 6% of patients. patients after periacetabular oncologic resections
Also, the survival rate of these cups at 1- and [92]. Compared to the previous study, while dislo-
5-year follow-up was 72% and 51%, respectively. cation rate was almost similar (14.8%), the infec-
Bus et al. also evaluated the same pedestal cup tion rate was significantly lower (11.1%). During
(Schoellner cup; Zimmer Biomet Inc., Warsaw, the follow-up (mean, 39 months, range,
Indiana) after oncologic pelvic resections in 19 18–80 months), only 1 patient required implant
patients with a mean follow-up of 7.9 years [84]. removal due to loosening. In another recent study
Although the 5-year failure rate was similar to enrolling 24 patients with periacetabular onco-
that of Hipfl et al. (50%), there were no revisions logic resections who underwent reconstruction
for mechanical reasons and all revision cases with an ice-cream cone-style implant, at least one
were due to periprosthetic infections. complication occurred in 58% of the patients
Complications rate was also high in another [93]. These complications included deep infection
study by Bus et al. that evaluated the LUMiC® (17%), dislocation (18%), and mechanical fail-
pedestal cup for periacetabular reconstruction ures (8%). At 5-year follow-up, implants survival
after oncologic resections [89]. In a total of 47 in this study was 75%.
patients, 30% experienced at least one complica-
tion. Particularly, infections occurred in 13 Allograft/Prosthetic Composites
patients (28%), while 10 patients (22%) had at Another option for reconstruction of periacetabu-
least one dislocation. The authors highlighted lar defects is the use of a composite structure
that when dual mobility cups were used the dislo- including a conventional acetabular prosthesis
cation rate dropped to 4%. The cumulative failure surrounded by a bone graft. The theoretical
incidences for the LUMiC® endoprosthesis in advantage of this technique is that the additional
that study at 2 and 5 years were 19.4% and use of the graft will allow for reconstruction of
26.5%, respectively [89], much lower than those larger bone defects, while the conventional pros-
reported in the study of Hipfl et al. [86]. thesis will restore hip range of motion resulting
There are also limited data regarding the use of in good functional results. The bone graft that is
reversed ice-cream cones-style implants for peri- attached to the prosthesis can be either the
acetabular reconstruction after oncologic pelvic resected bone segment used as an autoclaved or
resections [90, 91]. The newer types of ice-cream irradiated autograft or a bank bone allograft [52,
cone implants are basically modifications of the 94–100]. Autografts have the advantage of the
McMinn implant (LINK, Hamburg, Germany) optimum fit to the defect, while they are easily
that as the pedestal cups was initially developed accessible if the oncologic center is properly
for reconstruction of bone defects after failed equipped and a bone bank is available. Although
THA. Barriento-Ruiz et al. in a recent study this method yields good functional results, the
reported the outcomes of two ice-cream cones- use of an autograft or an allograft has in long
style implants (Coned®, Stanmore Worldwide term all these complications that are associated
Ltd, Elstree, UK; and Socincer®, Gijon, Spain) in with bone grafts such as infection, nonunion,
ten patients after Type II pelvic resections with a loosening, and fracture. The process for tumoral
median follow-up of 3 years [85]. Infections necrosis such as irradiation and heating for the
occurred in four patients (40%), two of which autografts compromises the bone quality of the
were superficial wound infections and the two grafts that are major risk factors for graft fracture
were deep infections, and dislocation occurred in and consolidation failure [97]. The reported
one patient (10%) during the follow- up. infection rates after these operations range from
15% to 50% [50, 52, 54]. The wide range in this theses, while structural bone grafts (either a
rate is probably due to the varying extent of soft structural allograft or an autograft) are an alterna-
tissue and bone resection and the subsequently tive option [10, 12]. Allografts can be used for
varying extent of the resulting dead space. reconstruction following almost any type of pel-
vic resections, and although they have the advan-
tage of good functional results and restoration of
15.6.3 Type I–II Resections native hip biomechanics (at least during the initial
postoperative period), the high rates of complica-
Tumors that involve not only the acetabular/supra- tions such as loosening, infection, and fractures
acetabular area but also extend to the largest part remain an unsolved problem. Therefore, due to
of ilium (resection Type I–II) are common, and their high complication rates and their limited
their management requires cautious preoperative availability, structural pelvic allografts for such
planning and a more ablative procedure. A critical extensive defects are not very popular.
factor in terms of managing the bone defects is
whether an adequate cranial part of the ilium is
preserved (Table 15.6). The nonreconstruction 15.6.4 Type III Resections:
options for Type I–II resections are similar to Ischiopubic Rami
those for Type II resections including iliofemoral
arthrodesis (or sacrofemoral arthrodesis), resec- Type III resection is required for tumors around
tion arthroplasty, and hip transposition. When an the pubic and ischial rami. For this type of resec-
adequate cranial part of the ilium is preserved, the tion, two pelvic osteotomies are performed. The
reconstruction options are also similar to those medial level of resection is through the pubic sym-
described for Type II resections including biologi- physis or in some cases through the opposite pubic
cal reconstructions and endoprosthetic recon- rami. The lateral osteotomy is performed medially
structions with saddle prostheses, stemmed cups, to the acetabulum. In some cases, the obturator
and custom-made prostheses. If the cranial part of nerve with the neighboring obturator vessels must
the ilium is not adequate, endoprosthetic recon- be sacrificed due to their close proximity to the
struction is possible only with custom-made pros- tumor. Tumors of the upper or lower pubic ramus
can be solely resected without further manage-
ment of the bone defects since this type of resec-
Table 15.6 Reconstruction of bone defects after Type tion does not disrupt the native pelvic ring stability.
I–II and Type I–II–III resections
An important part of this p rocedure is the cautious
Reconstruction Techniques reconstruction of the soft tissue envelope in order
Nonreconstruction 1. External hemipelvectomy to prevent herniation of the bladder or intestines
2. Iliofemoral or sacrofemoral
arthrodesis through the defect. Reconstruction of the inguinal
3. Resection arthroplasty floor is mandatory to prevent peritoneal herniation
4. Hip transposition technique [101, 102]. Augmentation of the soft tissue recon-
Reconstruction struction with a synthetic mesh or fascia lata
1. Biological 1. Bulk pelvic allograft allograft is commonly performed.
reconstruction 2. Autoclaved or irradiated
pelvic autograft
2. Endoprosthetic 1. Saddle prosthesis
reconstruction 2. Stemmed cup 15.6.5 Type II–III Resections
(a) Adequate Reverse ice-cream cone
preservation of implant
When tumors involve the upper or lower pubic
cranial iliac part Pedestal cup
3. Custom-made implant rami with additional extension to the acetabulum,
(b) Inadequate Custom-made implant a more extended approach compared to a Type II
preservation of resection is required. While the medial level of
cranial iliac part the resection is similar to that of a Type II resec-
tion, the lateral resection must be extended proxi- Table 15.7 Reconstruction of bone defects after Type IV
resections
mally to the supra-acetabular level (Type II–III).
The resulting bone defect after this type of resec- Type IV resection Reconstruction
tion can be addressed with any of the methods Partial sacrectomy Sole resection
<50% of the sacroiliac
that have been described for Type II resections. joint on each side (e.g.,
tumors below S1)
Total sacrectomy
15.6.6 Type I–II–III–IV Resections 1. Tumors involving the Resection and spinopelvic
largest part of sacrum stabilization
2. Proximal sacral tumors Resection without
In some cases, the tumor extends to include not stabilization (“spine on a
only the largest part of hemipelvis, but part of the biological sling”)
sacrum as well. In these cases, complete resection
of the hemipelvis must be performed either as an
external or an internal hemipelvectomy. The indi- junction or for tumors with extensive presacral
cations for external hemipelvectomy and lower soft tissue invasion, a combined approach may
limb amputation versus internal hemipelvectomy be required. From the anterior approach, the
have already been discussed previously. tumors can be accessed either through a transab-
The main difference compared to Type I–II or dominal route or a retroperitoneal route.
Type I–II–III resections is that there is no remain- Depending on the tumor’s extension into the
ing proximal iliac wing; therefore, the reconstruc- sacrum, sacrectomies can be either partial or
tion and nonreconstruction methods that are total (Table 15.7) [8, 22, 23, 103]. Partial resec-
available for the management of the resulted bone tions may be transverse, sagittal, or a combina-
defect are similar to those of Type I–II resections tion. Lateral sacral tumors are addressed with a
when no adequate cranial part of the ilium is pre- sagittal partial sacrectomy, whereas midline
served. Nonreconstruction options include resec- tumors are addressed with a transverse partial
tion arthroplasty, sacrofemoral arthrodesis, or a sacrectomy. For tumors located below the S2
hip transposition, while reconstructive options level, a partial sacrectomy is usually adequate
include custom-made implants or allografts. for wide resection of the tumor. In such cases,
Allografts have the advantage of reattachment of the lumbosacral joint is preserved, while blad-
the pelvic musculature to the graft, which in the- der and bowel function are usually not affected
ory leads to improved postoperative functional [104, 105]. Conversely, for tumors that develop
results. Custom-made prostheses are mega- around the proximal part of the sacrum with
implants that replace the patients’ hemipelvis and anterior expansion, total sacrectomy is required.
are associated with high rates of loosening and In total sacrectomies, the adjacent nerve roots
migration. Spontaneous periprosthetic hetero- are sacrificed in order to achieve a wide-margin
topic ossification around these implants can resection. These operations are associated with
develop that is associated with lower rates of loos- severe neurological dysfunction and a high rate
ening and dislocation. In most cases, an external of wound complications, although the use of the
hemipelvectomy is performed at the index sur- transpelvic vertical rectus abdominis myocuta-
gery or after local tumor recurrence or compli- neous flap has significantly decreased the latter
cated reconstructions for these extensive tumors. [106–112].
Resections for tumors below the S1 level do
not result in pelvic ring and lumbosacral insta-
15.6.7 Type IV Resections: Sacrum bility since the sacroiliac joints and the lumbosa-
cral junction are left intact. On the other hand,
The surgical approach for a sacrectomy can be sacrectomies that require S1 resection disrupt
either anterior or posterior, while for large prox- the lumbosacral and iliosacral junctions, lead to
imal tumors that extend to the lumbopelvic instability, and necessitate some form of spino-
pelvic stabilization [113–122]. Without stabili- 15.7 Extended Hemipelvectomy

zation, after a total sacrectomy, the lumbar spine
usually migrates downward and remains between When pelvic tumors expand to the lower lumbar
the iliac bones. The inferiorly migrated lumbar vertebrae, an extended hemipelvectomy with
spine is maintained at this new position by a additional resection of the lower lumbar spine is
“biological sling” that is formed by the sur- required. Since this surgery leads to spinopelvic
rounding musculature and the developing scar dissociation, the bone defect after an internal
tissue. Although patients commonly describe extended hemipelvectomy should be recon-
back or leg pain, walking is possible with the use structed [124, 129]. Many authors recommend
of a brace [119, 120, 123]. Therefore, some that internal extended hemipelvectomy must be
authors do not recommend bone reconstruction preserved only for patients with localized dis-
after total sacrectomies since the postoperative ease, while it should not be performed in cases
ambulatory status is acceptable and the rate of when prior surgery in the same area has been per-
complications such as surgical site infections is formed due to the increased risk for local recur-
significantly lower compared to more complex rence and complications [8, 129]. Additionally,
skeletal reconstructions [120–122]. However, given the high rate of failure and the increased
according to most authors, the indications for morbidity that is associated with extended inter-
lumbopelvic stabilization include a total sacrec- nal hemipelvectomy, this surgery is not indicated
tomy and a partial sacrectomy that involves at for compromised patients or for patients with
least 50% of the sacroiliac joint on each side metastatic bone disease.
[124–126]. Although the oncologic outcomes after inter-
Several different techniques using variable nal or external (extended) hemipelvectomy do
types of implants such as plates, screws, wires, not significantly differ, the postoperative function
and bars have been described for spinopelvic of the lower limbs after internal extended hemi-
stabilization after major spinopelvic resec- pelvectomy is very poor due to resection of the
tions. In cases of partial sacrectomies, spino- lumbosacral plexus [129]. In reality, amputation
pelvic fusion can be achieved with a of the lower limb (extended external hemipelvec-
combination of bars or screws into the sacrum tomy) is preferred in almost all cases and internal
and rods (Zimmer, Inc., Warsaw, Indiana) with extended hemipelvectomy is very rarely per-
pedicle screws into the lumbar vertebrae. formed [22, 23]. After an extended external
Otherwise, in cases of total sacrectomies, a hemipelvectomy, a spinopelvic fusion for preser-
commonly used technique for spinopelvic vation of the function of the contralateral limb is
fusion is the Luque–Galveston technique, in required. Part of the resected femur of the ampu-
which the pelvis is engaged into the fusion tated limb such as the femoral condyles can be
with bars or screws into the iliac bones bilater- used as a strut autograft for augmentation of this
ally. Many modifications of spinopelvic fusion fusion [22].
have been developed over the past years [113,
127, 128]. Although the newer implants may
achieve a more rigid fixation to the pelvis, the 15.8 Conclusions
issue of a robust proximal lumbar fixation has
still not been solved. Currently, the pedicle Whether an external hemipelvectomy or a partial
screw–rod construct is the most common pelvic resection should be performed is a deci-
instrumentation system for spinopelvic fixa- sion that should be made by the surgeon based on
tion since they are easily inserted and provide a several factors such as the underlying diagnosis,
more rigid fixation compared to other implants the patients’ medical status, the tumor’s charac-
such as hooks or wires. As a general consent, at teristics, and the correlated expected survival. It
least the three lower spinal segments must be has been proven that with proper patient selection
included in the fusion. and proper surgical technique internal hemipel-
vectomies do not carry a higher risk for recur- 10. Harrington KD. The use of hemipelvic allografts or
autoclaved grafts for reconstruction after wide resec-
rence compared to external hemipelvectomy, tions of malignant tumors of the pelvis. J Bone Joint
with similar survival rates [19, 130]. On the other Surg Am. 1992;74(3):331–41.
hand, external hemipelvectomy and amputation 11. Bruns J, Habermann CR, Rüther W, Delling D. The
have a lower incidence of complications and a use of CT derived solid modelling of the pelvis
in planning cancer resections. Eur J Surg Oncol.
faster recovery time compared to partial pelvic 2010;36:594–8.
resections. Also, even though amputees experi- 12. Yasko AW, Johnson ME. Surgical management of
ence certain limitations in daily activities such as primary bone sarcomas. Hematol Oncol Clin North
walking, rising and sitting down or climbing Am. 1995;9:719–31.
13. Gradinger R, Rechl H, Ascherl R, et al.
chairs, they can have a relatively high level of Endoprothetischer Teilersatz des Beckens bei
activities [131]. Reconstruction of the bone malignen Tumoren. Orthopade. 1993;22:167–73.
defect must be considered on a case-by-case 14. Gordley K, Marco R, Klebuc M. Supercharged
basis as it is a complex issue with no strong internal hemipelvectomy that enhances flap perfu-
sion and decreases wound-healing complications.
guidelines. The available reconstruction tech- Orthopedics. 2008;31(12).
niques provide better functional results over non- 15. Dick HM, Strauch RJ. Infection of massive bone
reconstruction since hip and lower limb functions allografts. Clin Orthop Relat Res. 1994:46–53.
are preserved, but at the expense of a higher com- 16. Wuisman P, Sugihara S. Allograft-Rekonstruktion
nach Resektion muskuloskeletaler Tumoren. In:
plication rate [3, 5–7, 10, 70, 132]. Maligne Knochentumoren, Aktueller Stand der
Diagnostik & Therapie. Berlin: Logos Verlag; 1997.
p. 93–100.
17. Knox K, Bitzos I, Granick M, Datiashvili R,
References Benevenia J, Patterson F. Immediate reconstruction
of oncologic hemipelvectomy defects. Ann Plast
1. Mavrogenis A, Soultanis K, Patapis P, Guerra Surg. 2006;57(2):184–9.
G, Fabbri N, Ruggieri P, et al. Pelvic resections. 18. Shih LY, Chen TS, Lo WH. Limb salvage surgery
Orthopedics. 2012;35(2):e232–43. for locally aggressive and malignant bone tumors. J
2. Schwameis E, Dominkus M, Krepler P, Dorotka R, Surg Oncol. 1993;53:154–60.
Lang S, Windhager R, Kotz R. Reconstruction of the 19. O’Connor MI. Malignant pelvic tumors: limb-
pelvis after tumor resection in children and adoles- sparing resection and reconstruction. Semin Surg
cents. Clin Orthop Relat Res. 2002;402:220–35. Oncol. 1997;13:49–54.
3. Karakousis CP, Emrich LJ, Driscoll DL. Variants 20. Enneking WF, Dunham WK. Resection and
of hemipelvectomy and their complications. Am J reconstruction for primary neoplasms involv-
Surg. 1989;158(5):404–8. ing the innominate bone. J Bone Joint Surg Am.
4. Enneking WF. The anatomic considerations in 1978;60(6):731–46.
tumor surgery: pelvis. In: Enneking WF, editor. 21. Masterson EL, Davis AM, Wunder JS, Bell
Musculoskeletal tumor surgery, vol. 2. New York: RS. Hindquarter amputation for pelvic tumors. Clin
Churchill Livingstone; 1983. p. 483–529. Orthop Relat Res. 1998;350:187–94.
5. Frassica FJ, Frassica DA, Pritchard DJ, et al. 22. Papagelopoulos PJ, Mavrogenis AF, Rose PS,
Ewing sarcoma of the pelvis. Clinicopathological Yaszemski MJ. Amputative sacral resections. In:
features and treatment. J Bone Joint Surg Am. Sim FH, Choong PFM, Weber KL, editors. Master
1993;75(10):1457–65. techniques in orthopaedic surgery: orthopaedic
6. Campanacci M, Capanna R. Pelvic resections: the oncology and complex reconstruction. New York:
Rizzoli Institute experience. Orthop Clin North Am. Lippincott Williams & Wilkins; 2011. p. 93–100.
1991;22(1):65–86. 23. Rose PS, Yaszemski MJ, Dekutoski MB, et al.
7. Windhager R, Karner J, Kutschera HP, Polterauer P, Classification of spinopelvic resections: oncologic
Salzer-Kuntschik M, Kotz R. Limb salvage in peri- and reconstructive implications. Paper presented at
acetabular sarcomas: review of 21 consecutive cases. the International Society of Limb Salvage Meeting,
Clin Orthop Relat Res. 1996;331:265–76. Boston, MA, September 24–27; 2009.
8. Conrad EU III, Springfield D, Peabody TD. Pelvis. 24. Shin KH, Rougraff BT, Simon MA. Oncologic out-
In: Simon MA, Springfield D, editors. Surgery for comes of primary bone sarcomas of the pelvis. Clin
bone and soft-tissue tumors. Philadelphia, PA: Orthop Relat Res. 1994;304:207–17.
Lipincott-Raven; 1998. p. 323–41. 25. Kawai A, Healey JH, Boland PJ, et al. Prognostic
9. Helmberger T. Vena cava filter [in German]. factors for patients with sarcomas of the pelvic
Radiologe. 2007;47:439–42. bones. Cancer. 1998;82:851–9.
26. Court C, Bosca L, Le Cesne A, Nordin JY, calcaneo-sacral fixation. Int Semin Surg Oncol.
Missenard G. Surgical excision of bone sarcomas 2008;5:1.
involving the sacroiliac joint. Clin Orthop Relat Res. 42. Fuchs B, O’Connor MI, Kaufman KR, Padgett DJ,
2006;451:189–94. Sim FH. Iliofemoral arthrodesis and pseudarthro-
27. Pollock RC, Skinner JA, Blunn GW, Pringle JA, sis: a long-term functional outcome evaluation. Clin
Briggs TW, Cannon SR. The swing procedure for Orthop Relat Res. 2002;397:29–35.
pelvic ring reconstruction following tumour exci- 43. Bruns J, Luessenhop S, Behrens P. Cost analy-
sion. Eur J Surg Oncol. 2003;29:59–63. sis of three different surgical procedures for treat-
28. O’Connor MI, Sim FH. Salvage of the limb in the ment of a pelvic tumour. Langenbecks Arch Surg.
treatment of malignant pelvic tumors. J Bone Joint 1998;383:359–63.
Surg Am. 1989;71(4):481–94. 44. Salzer M. Hemipelvectomy (amputation interileo-
29. Beadel GP, McLaughlin CE, Aljassir F, et al. abdominalis, hindquarter amputation). Wien Klin
Iliosacral resection for primary bone tumors: is pel- Wochenschr. 1965;77:330–1.
vic reconstruction necessary? Clin Orthop Relat Res. 45. Schwartz AJ, Kiatisevi P, Eilber FC, Eilber FR,
2005;438:22–9. Eckardt JJ. The Friedman-Eilber resection arthro-
30. Capanna R, Guernelli N, Ruggieri P. Periacetabular plasty of the pelvis. Clin Orthop Relat Res.
pelvic resections. In: Enneking WF, editor. Limb 2009;467(11):2825–30.
salvage in musculoskeletal oncology. New York: 46. Takami M, Ieguchi M, Takamatsu K, Kitano T, Aono
Churchill Livingstone; 1987. p. 141–6. M, Ishida T, Yamano Y. Functional evaluation of flail
31. Menendez LR, Nichter LS. Reconstructive consid- hip joint after periacetabular resection of the pelvis.
erations for limb salvage surgery. Orthop Clin North Osaka City Med J. 1997;43:173–83.
Am. 1993;24(3):511–21. 47. Gebert C, Gosheger G, Winkelmann W. Hip trans-
32. Yuen A, Ek ET, Choong PF. Research: Is resection of position as a universal surgical procedure for peri-
tumours involving the pelvic ring justified? A review acetabular tumors of the pelvis. J Surg Oncol.
of 49 consecutive cases. Int Semin Surg Oncol. 2009;99(3):169–72.
2005;2(1):9. 48. Winkelmann W. A new surgical method in malig-
33. Guo Z, Li J, Pei GX, Li XD, Wang Z. Pelvic recon- nant tumors of the ilium. Z Orthop Ihre Grenzgeb.
struction with a combined hemipelvic prostheses 1988;126:671–4.
after resection of primary malignant tumor. Surg 49. Rödl R, Gosheger G, Leidinger B, Lindner N,
Oncol. 2010;19(2):95–105. Winkelmann W, Ozaki T. Correction of leg-length
34. Guo W, Li D, Tang X, Yang Y, Ji T. Reconstruction discrepancy after hip transposition. Clin Orthop
with modular hemipelvic prostheses for periacetabu- Relat Res. 2003;416:271–7.
lar tumor. Clin Orthop Relat Res. 2007;461:180–8. 50. Hillmann A, Hoffmann C, Gosheger G, Rödl R,
35. Peterson CA II, Koch LD, Wood MB. Tibia-hindfoot Winkelmann W, Ozaki T. Tumors of the pelvis: com-
osteomusculocutaneous rotationplasty with calcane- plications after reconstruction. Arch Orthop Trauma
opelvic arthrodesis for extensive loss of bone from Surg. 2003;123:340–4.
the proximal part of the femur. A report of two cases. 51. Hoffmann C, Gosheger G, Gebert C, Jürgens H,
J Bone Joint Surg Am. 1997;79:1504–9. Winkelmann W. Functional results and quality of
36. Nagoya S, Usui M, Wada T, Yamashita T, Ishii life after treatment of pelvic sarcomas involving the
S. Reconstruction and limb salvage using a free vas- acetabulum. J Bone Joint Surg Am. 2006;88:575–82.
cularised fibular graft for periacetabular malignant 52. Bell RS, Davis AM, Wunder JS, Buconjic T,
bone tumours. J Bone Joint Surg Br. 2000;82:1121–4. McGoveran B, Gross AE. Allograft reconstruction
37. Sakuraba M, Kimata Y, Iida H, Beppu Y, Chuman H, of the acetabulum after resection of stage-IIB sar-
Kawai A. Pelvic ring reconstruction with the double- coma. Intermediate-term results. J Bone Joint Surg
barreled vascularized fibular free flap. Plast Reconstr Am. 1997;79:1663–74.
Surg. 2005;116:1340–5. 53. Delloye C, de Nayer P, Allington N, Munting E,
38. Yajima H, Tamai S. Twin-barrelled vascularized fib- Coutelier L, Vincent A. Massive bone allografts in
ular grafting to the pelvis and lower extremity. Clin large skeletal defects after tumor surgery: a clini-
Orthop Relat Res. 1994;303:178–84. cal and microradiographic evaluation. Arch Orthop
39. Yu G, Zhang F, Zhou J, Chang S, Cheng L, Jia Y, Li Trauma Surg. 1988;107:31–41.
H, Lineaweaver WC. Microsurgical fibular flap for 54. Langlais F, Lambotte JC, Thomazeau H. Long-term
pelvic ring reconstruction after periacetabular tumor results of hemipelvis reconstruction with allografts.
resection. J Reconstr Microsurg. 2007;23:137–42. Clin Orthop Relat Res. 2001;388:178–86.
40. Kollender Y, Shabat S, Bickels J, et al. Internal hemi- 55. Mankin HJ, Doppelt S, Tomford W. Clinical experi-
pelvectomy for bone sarcomas in children and young ence with allograft implantation. The first ten years.
adults: surgical considerations. Eur J Surg Oncol. Clin Orthop Relat Res. 1983;174:69–86.
2000;26(4):398–404. 56. Ozaki T, Hillmann A, Bettin D, Wuisman P,
41. Kong GY, Rudiger HA, Ek ET, Morrison WA, Winkelmann W. High complication rates with pel-
Choong PF. Reconstruction after external hemipelvic allografts. Experience of 22 sarcoma resections.
vectomy using tibia-hindfoot rotationplasty with Acta Orthop Scand. 1996;67:333–8.
57. Satcher RL Jr, O’Donnell RJ, Johnston 71. Angelini A, Kotrych D, Trovarelli G, Szafrański
JO. Reconstruction of the pelvis after resection of A, Bohatyrewicz A, Ruggieri P. Analysis of prin-
tumors about the acetabulum. Clin Orthop Relat Res. ciples inspiring design of three-dimensional-printed
2003;409:209–17. custom-made prostheses in two referral centres. Int
58. Von Sys G, Uyttendaele D, Poffyn B, Verdonk R, Orthop. 2020;44(5):829–37. https://doi.org/10.1007/
Verstraete L. Extracorporeally irradiated autografts s00264-020-04523-y. Epub 2020 Mar 13.
in pelvic reconstruction after malignant tumour 72. Van der Lei B, Hoekstra HJ, Veth RP, Ham SJ,
resection. Int Orthop. 2002;26:174–8. Oldhoff J, Schraffordt Koops H. The use of the
59. Gradinger R, Rechl H, Hipp E. Pelvic osteosarcoma. saddle prosthesis for reconstruction of the hip joint
Resection, reconstruction, local control, and survival after tumor resection of the pelvis. J Surg Oncol.
statistics. Clin Orthop Relat Res. 1991;270:149–58. 1992;50:216–9.
60. Harrington KD, Johnston JO, Kaufer HN, Luck JV 73. Nieder E, Elson RA, Engelbrecht E, Kasselt MR,
Jr, Moore TM. Limb salvage and prosthetic joint Keller A, Steinbrink K. The saddle prosthesis for
reconstruction for low-grade and selected high- salvage of the destroyed acetabulum. J Bone Joint
grade sarcomas of bone after wide resection and Surg Br. 1990;72:1014–22.
replacement by autoclaved [corrected] autogenetic 74. Angelini A, Trovarelli G, Berizzi A, Pala E, Breda
grafts. Clin Orthop Relat Res. 1986;211:180–214. A, Ruggieri P. Three-dimension-printed custom-
Erratum in: Clin Orthop 216:312. made prosthetic reconstructions: from revision
61. Mnaymneh W, Malinin T, Mnaymneh LG, Robinson surgery to oncologic reconstructions. Int Orthop.
D. Pelvic allograft. A case report with a follow- 2019;43(1):123–32. https://doi.org/10.1007/
up evaluation of 5.5 years. Clin Orthop Relat Res. s00264-018-4232-0.
1990;255:128–32. 75. Schoellner C, Schoellner D. Pedestal cup opera-
62. Friedlaender GE. Bone allografts: the biological tion in acetabular defects after hip cup loosen-
consequences of immunological events. J Bone Joint ing. A progress report. Z Orthop Ihre Grenzgeb.
Surg Am. 1991;73:1119–22. 2000;138:215–21.
63. Tomford WW, Thongphasuk J, Mankin HJ, Ferraro 76. Uchida A, Myoui A, Araki N, Yoshikawa H, Ueda
MJ. Frozen musculoskeletal allografts. A study T, Aoki Y. Prosthetic reconstruction for periace-
of the clinical incidence and causes of infection tabular malignant tumors. Clin Orthop Relat Res.
associated with their use. J Bone Joint Surg Am. 1996;326:238–45.
1990;72:1137–43. 77. Walker RH. Pelvic reconstruction/total hip arthro-
64. Zhang HY, Thongtrangan I, Balabhadra RS, et al. plasty for metastatic acetabular insufficiency. Clin
Surgical techniques for total sacrectomy and Orthop Relat Res. 1993;294:170–5.
spinopelvic reconstruction. Neurosurg Focus. 78. Windhager R, Welkerling H, Kastner N, Krepler
2003;15(2):E5. P. Surgical therapy of pelvis and spine in primary
65. Hornicek FJ, Gebhard MC, Tomford WW, et al. malignant bone tumors. Orthopade. 2003;32:
Factors affecting nonunion of the allograft- 971–82.
host junction. Clin Orthop Relat Res. 2001;382: 79. Dai KR, Yan MN, Zhu ZA, Sun YH. Computer-
87–98. aided custom-made hemipelvic prosthesis used in
66. Yoshida Y, Osaka S, Tokuhashi Y. Analysis of extensive pelvic lesions. J Arthroplast. 2007;22(7):
limb function after various reconstruction methods 981–6.
according to tumor location following resection 80. Wirbel RJ, Schulte M, Mutschler WE. Surgical treat-
of pediatric malignant bone tumors. World J Surg ment of pelvic sarcomas: oncologic and functional
Oncol. 2010;19:39. outcome. Clin Orthop Relat Res. 2001;390:190–205.
67. Delloye C. Bone banking in orthopaedic surgery. In: 81. Cottias P, Jeanrot C, Vinh TS, Tomeno B, Anract
Gallinaro P, Lemaire R, editors. Surgical techniques P. Complications and functional evaluation of 17
in orthopaedics and traumatology. Paris: Elsevier; saddle prostheses for resection of periacetabular
2000. p. 55–61. tumors. J Surg Oncol. 2001;78:90–100.
68. Delloye C, Banse X, Brichard B, Docquier PL, 82. Jansen JA, van de Sande MA, Dijkstra PD. Poor
Cornu O. Pelvic reconstruction with a structural long-term clinical results of saddle prosthesis after
pelvic allograft after resection of a malignant bone resection of periacetabular tumors. Clin Orthop
tumor. J Bone Joint Surg Am. 2007;89:579–87. Relat Res. 2013;471:324–31.
69. Wafa H, Grimer R, Jeys L, Abudu A, Carter S, 83. Toma C, Kubista B, Funovics P, Abdolvahab F,
Tillman R. The use of extracorporeally irradiated Dominkus M, Kotz R. Pedestal cups for acetabular
autografts in pelvic reconstruction following tumour and pelvic defects following tumor resection. A new
resection. Bone Joint J. 2014;96-B(10):1404–10. approach to an old problem. Abstract book, 13th
70. Aboulafia AJ, Buch R, Mathews J, Li W, Malawer ISOLS Korea; 2005. p. 148.
MM. Reconstruction using the saddle prosthe- 84. Bus MP, Boerhout EJ, Bramer JA, Dijkstra
sis following excision of primary and metastatic PD. Clinical outcome of pedestal cup endoprosthetic
periacetabular tumors. Clin Orthop Relat Res. reconstruction after resection of a peri-acetabular
1995;314:203–13. tumour. Bone Joint J. 2014;96:1706–12.
85. Barrientos-Ruiz I, Ortiz-Cruz E, Peleteiro-Pensado 99. Tsuchiya H, Wan SL, Sakayama K, Yamamoto N,
M. Reconstruction after hemipelvectomy with Nishida H, Tomita K. Reconstruction using an auto-
the ice-cream cone prosthesis: what are the short- graft containing tumour treated by liquid nitrogen. J
term clinical results? Clin Orthop Relat Res. Bone Joint Surg Br. 2005;87:218–25.
2016;475:735–41. 100. Yamamoto N, Tsuchiya H, Tomita K. Effects of liq-
86. Hipfl C, Stihsen C, Puchner S, Kaider A, Dominkus uid nitrogen treatment on the proliferation of osteo-
M, Funovics P, et al. Pelvic reconstruction follow- sarcoma and the biomechanical properties of normal
ing resection of malignant bone tumours using a bone. J Orthop Sci. 2003;8:374–80.
stemmed acetabular pedestal cup. Bone Joint J. 101. Bhattacharya S, Biswas S. Post hemipelvectomy
2017;99-B:841–8. incisional hernia. Hernia. 2008;13(4):435–7.
87. Rudert M, Holzapfel BM, Pilge H, Rechl H, 102. Die Trill J, Madrid JM, Ferrero E, et al.
Gradinger R. Partial pelvic resection (internal Posthemipelvectomy hernia. Hernia.
hemipelvectomy) and endoprosthetic replacement 2005;9(4):375–7.
in periacetabular tumors. Oper Orthop Traumatol. 103. Mavrogenis AF, Patapis P, Kostopanagiotou
2012;24:196–214. G, Papagelopoulos PJ. Tumors of the sacrum.
88. Shao Q, Yan X, Sun J, Xu T. Internal hemipelvectomy Orthopedics. 2009;32(5):342.
with reconstruction for primary pelvic neoplasm: a 104. Baratti D, Gronchi A, Pennacchioli E, et al.
systematic review. ANZ J Surg. 2014;85:553–60. Chordoma: natural history and results in 28 patients
89. Bus MP, Szafranski A, Sellevold S, Goryn T, Jutte P, treated at a single institution. Ann Surg Oncol.
Bramer J, et al. LUMiC® endoprosthetic reconstruc- 2003;10:291–6.
tion after periacetabular tumor resection: short-term 105. Wuisman P, Lieshout O, Sugihara S, van Dijk
results. Clin Orthop Relat Res. 2016;475:686–95. M. Total sacrectomy and reconstruction: oncologic
90. McMinn DJ, Grigoris P, Roberts P. A stemmed and functional outcome. Clin Orthop Relat Res.
acetabular cup for complex hip arthroplasty. J Bone 2000;381:192–203.
Joint Surg Br. 1993;75-B:123. 106. Miles WK, Chang DW, Kroll SS, et al. Reconstruction
91. Stihsen C, Hipfl C, Kubista B, et al. Review of the of large sacral defects following total sacrectomy.
outcomes of complex acetabular reconstructions Plast Reconstr Surg. 2000;105:2387–94.
using a stemmed acetabular pedestal component. 107. Fourney DR, Gokaslan ZL. Current management of
Bone Joint J. 2016;98-B:772–9. sacral chordoma. Neurosurg Focus. 2003;15(2):E9.
92. Fisher NE, Patton JT, Grimer RJ, et al. Ice-cream 108. Localio SA, Eng K, Ranson JH. Abdominosacral
cone reconstruction of the pelvis: a new type of pel- approach for retrorectal tumors. Ann Surg.
vic replacement: early results. J Bone Joint Surg Br. 1980;191:555–60.
2011;93-B:684–8. 109. Stener B, Gunterberg B. High amputation of the
93. Issa S, Biau D, Babinet A, Dumaine V, Le Hanneur sacrum for extirpation of tumors. Principles and
M, Anract P. Pelvic reconstructions following peri- technique. Spine. 1978;3:351–66.
acetabular bone tumour resections using a cement- 110. Biagini R, Ruggieri P, Mercuri M, et al. Neurologic
less ice-cream cone prosthesis with dual mobility deficit after resection of the sacrum. Chir Organi
cup. Int Orthop. 2018;42:1987–97. Mov. 1997;82:357–72.
94. Bohm P, Springfeld R, Springer H. Re-implantation 111. Nakai S, Yoshizawa H, Kobayashi S, Maeda K,
of autoclaved bone segments in musculoskeletal Okumura Y. Anorectal and bladder function after
tumor surgery. Clinical experience in 9 patients fol- sacrifice of the sacral nerves. Spine. 2000;25:2234–9.
lowed for 1.1–8.4 years and review of the literature. 112. Todd LT Jr, Yaszemski MJ, Currier BL, Fuchs B,
Arch Orthop Trauma Surg. 1998;118:57–65. Kim CW, Sim FH. Bowel and bladder function
95. Chen TH, Chen WM, Huang CK. Reconstruction after major sacral resection. Clin Orthop Relat Res.
after intercalary resection of malignant bone 2002;397:36–9.
tumours: comparison between segmental allograft 113. Gokaslan ZL, Romsdahl MM, Kroll SS, et al. Total
and extracorporeally irradiated autograft. J Bone sacrectomy and Galveston L-rod reconstruction for
Joint Surg Br. 2005;87:704–9. malignant neoplasms. Technical note. J Neurosurg.
96. Guest CB, Bell RS, Davis A, Langer F, Ling H, 1997;87:781–7.
Gross AE, Czitrom A. Allograft-implant compos- 114. Shikata J, Yamamuro T, Kotoura Y, Mikawa Y, Iida
ite reconstruction following periacetabular sarcoma H, Maetani S. Total sacrectomy and reconstruction
resection. J Arthroplast. 1990;5(Suppl):S25–34. for primary tumors. Report of two cases. J Bone
97. Jeon DG, Kim MS, Cho WH, Song WS, Lee Joint Surg Am. 1988;70:122–5.
SY. Reconstruction with pasteurized autograft total 115. Kuklo TR, Bridwell KH, Lewis SJ, et al. Minimum
hip prosthesis composite for periacetabular tumors. 2-year analysis of sacropelvic fixation and
J Surg Oncol. 2007;96:493–502. L5-S1 fusion using S1 and iliac screws. Spine.
98. Manabe J, Ahmed AR, Kawaguchi N, Matsumoto S, 2001;26:1976–83.
Kuroda H. Pasteurized autologous bone graft in sur- 116. Gunterberg B, Romanus B, Stener B. Pelvic strength
gery for bone and soft tissue sarcoma. Clin Orthop after major amputation of the sacrum: an experimen-
Relat Res. 2004;419:258–66. tal study. Acta Orthop Scand Suppl. 1976;47:635–42.
117. Hüfner T, Kfuri M Jr, Galanski M, et al. New early experience with a new surgical technique. Clin
indications for computer-assisted surgery: tumor Orthop Relat Res. 2005;438:42–50.
resection in the pelvis. Clin Orthop Relat Res. 125. Doita M, Harada T, Iguchi T, et al. Total, sacrec-
2004;426:219–25. tomy and reconstruction for sacral tumors. Spine.
118. Randall R, Bruckner J, Lloyd C, Pohlman TH, 2003;28:E296–301.
Conrad EU III. Sacral resection and reconstruction 126. Jackson RJ, Gokaslan ZL. Spinal-pelvic fixation in
for tumors and tumor-like conditions. Orthopedics. patients with lumbosacral neoplasms. J Neurosurg.
2005;28:307–13. 2000;92:61–70.
119. Shen FH, Harper M, Foster WC, Marks I, Arlet 127. Tomita K, Tsuchiya H. Total sacrectomy and
V. A novel “four-rod technique” for lumbopelvic reconstruction for huge sacral tumors. Spine.
reconstruction: theory and technical considerations. 1990;15:1223–7.
Spine. 2006;31:1395–401. 128. Blatter G, Halter Ward EG, Ruflin G, Jeanneret
120. Ruggieri P, Angelini A, Ussia G, Montalti M, B. The problem of stabilization after sacrectomy.
Mercuri M. Surgical margins and local control in Arch Orthop Trauma Surg. 1994;114:40–2.
resection of sacral chordomas. Clin Orthop Relat 129. Fuchs B, Yaszemski MJ, Sim FH. Combined poste-
Res. 2010;468(11):2939–47. rior pelvis and lumbar spine resection for sarcoma.
121. Michel A. Total sacrectomy and lower spine resec- Clin Orthop Relat Res. 2002;397:12–8.
tion for giant cell tumor: one case report. Chir 130. Guo W, Li D, Tang X, Ji T. Surgical treatment of
Organi Mov. 1990;75:117–8. pelvic chondrosarcoma involving periacetabulum. J
122. Simpson AH, Porter A, Davis A, Griffi NA, McLeod Surg Oncol. 2010;101:160–5.
RS, Bell RS. Cephalad sacral resection with a com- 131. Yari P, Dijkstra P, Geertzen J. Functional out-
bined extended ilioinguinal and posterior approach. come of hip disarticulation and hemipelvectomy:
J Bone Joint Surg Am. 1995;77:405–11. a cross-sectional national descriptive study in the
123. Guo Y, Yadav R. Improving function after total Netherlands. Clin Rehabil. 2008;22(12):1127–33.
sacrectomy by using a lumbar-sacral corset. Am J 132. Ruggieri P, Mavrogenis AF, Angelini A, Mercuri
Phys Med Rehabil. 2002;81:72–6. M. Metastases of the pelvis: does resection improve
124. Dickey ID, Hugate RR Jr, Fuchs B, Yaszemski MJ, survival? Orthopedics. 2011;34(7):e236–44.
Sim FH. Reconstruction after total sacrectomy:
The Hip Transposition
as a Reconstructive Technique 16
After Pelvic Resection
Timo Lübben and Georg Gosheger
16.1 Introduction
Resection of pelvic tumors is one of the most

challenging surgical procedures in tumor surgery.
Wide resection following the definition of
Enneking [1] is to be considered standard of care
for most malignancies. Limb salvage can be
achieved in most of the cases of pelvic bone
tumors without narrowing the margins of the
resection [2]. While endoaprosthetic replacement
and allograft/autograft reconstruction suffer
severe postoperative problems due to infection
and poor soft tissue coverage [3–9], hip transpo- ap x-ray; arrow mark → parosteal osteosarcoma of the
sition, although far from perfect, can be consid- iliac bone
ered a functionally good and long-lasting surgical
alternative [10].
16.3 Contraindications
16.2 Indication The contraindications for hip transposition are

the same as for internal hemipelvectomy.
Hip transposition can be the procedure of choice Absolute
for Enneking type P2-, P1-2, P1-3, and P1-4 Tumors crossing the middle of the sacral bone
resection of the pelvis. Alternative operative pro- (when wide resection is not possible) and meta-
cedures, i.e., endoprosthetic replacement, in the static disease (specialties (i.e., single late metas-
case of P2-/P2-3-resection should be discussed. tasis of renal cell carcinoma) excluded)
Relative
Bad general condition, bad soft tissue cover-
age (especially involvement of the gluteus maxi-
T. Lübben (*) · G. Gosheger
Department for orthopedic surgery, University Clinics mus muscle), involvement of the sciatic nerve,
of Münster, Münster University Hospital (UKM), and involvement of the femoral nerve/the femoral
Munster, Germany artery [11]. Age above 65 years is associated with
e-mail: [email protected];
higher rate of complications [12].
[email protected]

186 T. Lübben and G. Gosheger
16.4 Preparation HINT: No false ambitions! Good soft tissue

coverage doesn’t help in the case of tumor con-
Performing an internal hemipelvectomy is chal- tamination!!!! WIDE RESECTION FIRST!
lenging and needs careful preoperative planning
and preparation!
We recommend:
1. MRI-based planning of the tumor resection

and soft tissue closure (MRI not older than
4 weeks/matched navigation if possible).
2. Use Enneking’s approach! For extraarticular
resection of the proximal femur, a second cut
from the SIAS following the tensor fascia lata
to the lateral thigh (lateral part of Judet’s
approach) can be helpful.
3. Two skilled surgeons/surgical teams perform-
ing the operation (simultaneous ventral and
dorsal approach) help to shorten intraopera-
tive time.
4. If possible use preoperative epidural catheter situs after internal extraarticular hemipelvectomy
for intra- and postoperative pain therapy. P1-4: red loops, A. iliaca externa and femoralis;
5. Urinary catheter placement (green colored)
blue loop, V. iliaca externa; white loop, N. femo-
and ureteral stenting (ipsi- or bilateral) for ralis; and strap sciatic nerve.
easy palpation of ureter, urethra, and bladder.
6. Positioning of the patient in lateral position
for easy unfolding and through maximizing 16.5 Performing Hip
space between ribs and pelvis. Transposition
7. Intraoperatively preserve the vessels feeding
the gluteus maximus flap whenever possible; Three types of hip transpositions are described:
alternatively think about performing a rectus While Type 1 is in need of osteosynthesis
abdominis flap/free flaps or if not possible (with additional complications like pseudarthro-
external hemipelvectomy. sis or screw displacements),
types 2a and 2b are easy to perform.

16 The Hip Transposition as a Reconstructive Technique After Pelvic Resection 187
Relevant in clinical terms are Type 2a: hip

transposition after intraarticular resection of the
hip joint.
and 2b: hip transposition after extraarticular

resection of the hip joint.
Type 2a: 2. place 3–5 bone anchors in the remaining bone

of the ileum or sacrum (hint: not all kinds of
1. try to narrow the space between the femoral anchors do the job [13]) and fix the attach-
head and the remaining bone, when this is ment tube at the site of optimal fixation,
possible proceed, otherwise perform soft tis-
sue release (necessary mostly of the adductor
muscles),
After the resection of the proximal femur, pro-

ceed implanting a proximal femur replacement.
In standard care, we use Implantcast MUTARS
prox. For femur replacement (silver coated for
reduction of deep infections [15–17]/reconstruc-
tion length 80 mm, 100 mm, and then every
10 mm) and ic-bipolar head (HINT: use small
size, i.e., 44 mm)/CAVE in the case of an allergic
reaction to nickel/chrome, etc., ions switch to a
full ceramic bipolar head (i.e., Mathys Bionit 2).
When finished implanting and testing the endo-
prosthesis, proceed as in 2a reconstructions.
3. attach the tube on the femoral head with non-

resorbable sutures,
Rö einfügen
4. attach remaining muscles to the attachment
tube or to the bone for narrowing dead space;
HINT: The psoas muscle can be transferred to
the trochanter major for better abduction,
5. minimum 2 deep drains,
6. Flip gluteus maximus flap to the front and
close musculature and fascia.
7. Close wound, the use of clamps or sutures can
be discussed [14], and hard evidence for the
use of skin sutures for pelvic wounds has not
yet been established.
Type 2 b hip transposition
16 The Hip Transposition as a Reconstructive Technique After Pelvic Resection 189
HINT: While leg length discrepancy is fixed in 16.6 Short summary

Type 2a reconstructions, you can adapt the length
of the proximal femur replacement in type 2b While the operative procedure of the hip transpo-
reconstructions depending on the soft tissue sition is fairly easy to perform after successful
coverage! internal hemipelvectomy, postoperative compli-
postoperative care cations are to be expected. Pelvic resections and
reconstruction remain to be among the most chal-
–– at days 1–5, prolonged antibiotic prophylaxis/
lenging operative interventions and should be
antibiotic therapy (second generation cephalo-
performed at specialized centers of care.
sporin) is administered.
–– immobilization without splint for 2–6 weeks/
alternative procedures like external stabiliza-
References
tion via fixateur externe are possible [18].
–– partial weight bearing for 3 months after the 1. Enneking WF, Spanier SS, Goodman MA. A system
operation. for the surgical staging of musculoskeletal sarcoma.
–– restrictions in movement depending on the Clin Orthop Relat Res. 1980;153:106–20.
soft tissue coverage and wound healing, 2. Ham SJ, Schraffordt Koops H, Veth RP, van Horn JR,
Eisma WH, Hoekstra HJ. External and internal hemi-
pelvectomy for sarcomas of the pelvic girdle: conse-
complications quences of limb-salvage treatment. Eur J Surg Oncol.
1997;23(6):540–6.
–– prolonged wound healing/wound necro- 3. Abudu A, Grimer RJ, Cannon SR, Carter SR, Sneath
sis → wait for 5–7 days till revision operation RS. Reconstruction of the hemipelvis after the exci-
for minimizing the numbers of revisions; sion of malignant tumours. Complications and func-
close monitoring of CRP and leukocyte count. tional outcome of prostheses. J Bone Joint Surg.
1997;79(5):773–9.
–– infection → don’t hesitate to revise and 4. Windhager R, Karner J, Kutschera HP, Polterauer P,
debride; when used, change endoprosthetic Salzer-Kuntschik M, Kotz R. Limb salvage in peri-
material; remove attachment tube early. acetabular sarcomas: review of 21 consecutive cases.
Clin Orthop Relat Res. 1996;331:265–76.
–– persistent infection → complete removal of all
endoprosthetic material, depending on the struction for primary neoplasms involving the innomi-
entity of the tumor change procedure to flail nate bone. J Bone Joint Surg Am. 1978;60(6):731–46.
hip +− vacuum foam or spacer. 6. Hillmann A, Hoffmann C, Gosheger G, Rodl R,
Winkelmann W, Ozaki T. Tumors of the pelvis: com-
–– HINT: Don’t postpone relevant postoperative
plications after reconstruction. Arch Orthop Trauma
chemotherapy for more than 6 weeks. Surg. 2003;123(7):340–4. https://doi.org/10.1007/
–– complete flap necrosis → shorten leg, change s00402-003-0543-7.
flap, use free flap, or change to external 7. Windhager R, Welkerling H, Kastner N, Krepler
P. Surgical therapy of pelvis and spine in pri-
hemipelvectomy,
mary malignant bone tumors. Der Orthopade.
2003;32(11):971–82. https://doi.org/10.1007/
follow-up s00132-003-0551-x.
8. Bell RS, Davis AM, Wunder JS, Buconjic T,
–– depending on the entity of the tumor, in gen- McGoveran B, Gross AE. Allograft reconstruc-
eral, every 3 months for 2 years, every 6 month tion of the acetabulum after resection of stage-
for 5 years, and then every year, IIB sarcoma. Intermediate-term results. J Bone
Joint Surg Am. 1997;79(11):1663–74. https://doi.
late follow-up org/10.2106/00004623-199711000-00008.
9. Donati D, Capanna R, Campanacci D, Del Ben M,
–– In the case of severe leg length discrepancy Ercolani C, Masetti C, et al. The use of massive bone
allografts for intercalary reconstruction and arthrod-
more than 5 years after the primary operation,
eses after tumor resection. A multicentric European
discuss secondary leg lengthening, i.e., via study. La Chirurgia degli organi di movimento.
intramedullary nail or external fixateur [19]. 1993;78(2):81–94.
10. Gebert C, Wessling M, Hoffmann C, Roedl R,

15. Gosheger G, Hardes J, Ahrens H, Streitburger A,

Winkelmann W, Gosheger G, et al. Hip transpo- Buerger H, Erren M, et al. Silver-coated megaendo-
sition as a limb salvage procedure following the prostheses in a rabbit model—an analysis of the infec-
resection of periacetabular tumors. J Surg Oncol. tion rate and toxicological side effects. Biomaterials.
2011;103(3):269–75. https://doi.org/10.1002/ 2004;25(24):5547–56. https://doi.org/10.1016/j.
jso.21820. biomaterials.2004.01.008.
11. Nottrott M, Gosheger G, Dieckmann R, Gebert C, 16. Schmidt-Braekling T, Streitbuerger A, Gosheger

Henrichs M-P, Hardes J. Hemipelvectomy: what G, Boettner F, Nottrott M, Ahrens H, et al. Silver-
is important? OP-J. 2010;26(3):192–5. https://doi. coated megaprostheses: review of the literature. Eur
org/10.1055/s-0030-1250544. J Orthop Surg Traumatol. 2017;27(4):483–9. https://
12. Guder WK, Hardes J, Gosheger G, Henrichs MP,
doi.org/10.1007/s00590-017-1933-9.
Nottrott M, Streitburger A. Analysis of surgical and 17. Streitbuerger A, Henrichs MP, Hauschild G, Nottrott
oncological outcome in internal and external hemipel- M, Guder W, Hardes J. Silver-coated megaprostheses
vectomy in 34 patients above the age of 65 years at a in the proximal femur in patients with sarcoma. Eur
mean follow-up of 56 months. BMC Musculoskelet J Orthop Surg Traumatol. 2019;29(1):79–85. https://
Disord. 2015;16:33. https://doi.org/10.1186/ doi.org/10.1007/s00590-018-2270-3.
s12891-015-0494-5. 18. Kunisada T, Fujiwara T, Hasei J, Nakata E, Senda
13. Wessling M, Aach M, Herbort M, Gosheger G,
M, Ozaki T. Temporary external fixation can sta-
Hardes J, Gebert C. The primary stability of the bilize hip transposition arthroplasty after resec-
hip transposition type IIb: a biomechanical in vitro tion of malignant periacetabular bone tumors. Clin
study. Clin Biomech. 2009;24(4):361–5. https://doi. Orthop Relat Res. 2019; https://doi.org/10.1097/
org/10.1016/j.clinbiomech.2009.01.008. CORR.0000000000000764.
14. Maurer ERA, Maschuw K, Aminossadati B, Neubert 19. Rodl R, Gosheger G, Leidinger B, Lindner N,

T, Schade-Brittinger C, Bartsch DK. Superficial sur- Winkelmann W, Ozaki T. Correction of leg-length
gical site infection following the use of intracutaneous discrepancy after hip transposition. Clin Orthop Relat
sutures versus staples—a randomized single-center Res. 2003;416:271–7. https://doi.org/10.1097/01.
trial in an elective gastrointestinal surgery setting. blo.0000081933.75404.f8.
Dtsch Arztebl Int. 2019;116:365–71. https://doi.
org/10.3238/arztebl.2019.0365.
3D Printed Reconstructions
17
Andrea Angelini, Daniel Kotrych,
Andrzej Szafrański, and Pietro Ruggieri
17.1 Introduction plex bone fractures [27–31], and treatment of

infective complications [32]. A biocompatible
Limb salvage surgery has been shown to be fea- and tailored 3D-printed titanium implant is the
sible and effective for most of the primary and final result of a multistep process that begins with
secondary pelvic bone tumors [1–4]. The surgical the imaging data acquisition and the fabrication
accuracy improved significantly in recent years, of bone models. Of course, each step needs to be
thanks to the advances in surgical techniques, monitored and carried out in a structured and pre-
biomedical engineering, tumor segmentation on cise manner [33, 34]. In the following paragraphs,
imaging, preoperative planning, and intraopera- we will get into the details of each aspect of
tive tools for surgical margins [5–6]. The increas- 3D-printing technology.
ing interest in the development of 3D printing
technology is based on the possibility to improve
precision surgery and to realize a personalized 17.2 Anatomical Models
custom implant on patient’s anatomy [7–11]. In and Surgical Plan
fact, 3D printing techniques have been widely
used in numerous fields of orthopedic surgery The first step in the use of 3D-printing technol-
and musculoskeletal oncology: bone tumor resec- ogy is the “segmentation” process whereby imag-
tion and functional reconstruction [12–19], pri- ing data (digital imaging and communications in
mary and revision arthroplasty [14, 20–23], medicine—DICOM) are purchases and con-
spinal surgery [13, 24–26], management of com- verted into a digital 3D-model. This process is
particularly relevant in patients with pelvic bone
tumors because of the complex anatomy of the
pelvis and the need of accurate visualization of
A. Angelini · P. Ruggieri (*)
Department of Orthopedics and Orthopedic tumor extension. Usually, a CT scan of the pelvis
Oncology, University of Padova, Padova, Italy with 1 mm cut every 1 mm step is required to
e-mail: [email protected]; have a voxel size of 1 mm square that defines the
[email protected] level of accuracy of the plan. A virtual 3D model
D. Kotrych may be realized and analyzed by a multidisci-
Department of Orthopedics, Traumatology and plinary team that involves surgeons and engi-
Orthopedic Oncology, Pomeranian Medical
University, Szczecin, Poland neers, with higher information compared to a
multiplanar CT image alone (Fig. 17.1) [35]. In
A. Szafrański
Institute of Mother & Child, Warsaw, Poland the virtual 3D planning, it is possible to add

a b
Fig. 17.1 Osteosarcoma of the pelvis in a 52-year-old mentation process; (b) creation of a digital 3D model with
man. (a) Pelvic plain radiographs with the resection lev- tumor highlighted (green color); (c) The expected bone
els annotated directly by the surgeon. The pictures are defect can be easily analyzed on the virtual 3D model
then transferred to engineers for the analysis after seg-
a natomic landmarks and resection planes and to bone tumors [37–42]. The authors believe that
visualize specifically the tumor volume there is a strong difference in preparing the surgi-
(Fig. 17.2a–c). These aspects should be trans- cal plan for oncologic disease compared to other
ferred from the surgeon to the engineers consid- complex fractures or revision surgery. The main
ering the surgical approach (based on tumor site), aspect is the possibility of relevant changes in
the extent of soft tissue involvement, and close tumor volume from the imaging acquisition to
critical anatomic structures. All of these elements the surgical procedure. It is important to take into
should be visualized during surgery for a safe account the worst case scenario with a tumor
resection (Fig. 17.2d, e). These models can be growth not respondent to preoperative chemo-
printed with materials fulfilling biocompatibility therapy or other neoadjuvant procedures, thereby
standards in a 1:1 scale, so they can be sterilized considering a planned bone cut to be safe this
and brought to the surgical field (Fig. 17.3a) [36] occurrence [43].
or in plastic transparent material for academic/
education aims. 3D printed models improve the
surgeon’s understanding of the surgical challenge 17.3 Patient-Specific Tools
and can be used to explain surgical procedure to and Guides
the patient (Fig. 17.3b, c), to test the printed sur-
gical tools, and as simulation to improve safety in Three-dimensional printing is a simple way to
real-life surgery (Fig. 17.3d). These models are obtain patient-specific instrumentation (PSI)
an effective tool for finalizing surgical planning, that enables the surgeon to follow and real-
also considering the crucial role of resection mar- ize the surgical plan based on preoperative
gins as prognostic factor in most of the malignant imaging.
17 3D Printed Reconstructions 193
a b
c d e
Fig. 17.2 Ewing’s sarcoma of the pelvis in a 13-year- Corporate Medical Systems, Villanova San Daniele del
old girl. (a) Axial CT scan with 1 mm cut every 1 mm step Friuli, Italy), it is possible to visualize the tumor volume
and (b) axial T1-weighted MR images were obtained after (red area) and the involvement of the contralateral ischio-
preoperative neoadjuvant chemotherapy for the evaluation pubic branches. (d, e) Resection planes have been added
of tumor volume. The tumor (white arrow) is highlighted considering the surgical approach, the extent of soft tissue
by the surgeon on both examinations. (c) In the virtual 3D involvement, and safe resection margins
planning (using web-based platform Promade, Lima
a b
c d
Fig. 17.3 Same patient: Ewing’s sarcoma of the pelvis with detachable parts. (d) 3D printed cutting jigs. (Implant
in a 13-year-old girl. (a) The 1:1-sized 3D printed model designed with Promade, Lima Corporate Medical
of the entire hemipelvis allows the surgeon to accurately Systems, Villanova San Daniele del Friuli, Italy)
appreciate the anatomy and (b, c) the surgical planning
a b
c d
Fig. 17.4 Same patient: Ewing’s sarcoma of the pelvis Intraoperative photograph showing the use of
in a 13-year-old girl. (a) Preoperative virtual planning 3-dimensional model of the specimen and the correspon-
with custom cutting jigs and (b) 3D-printed patient- dent resected tumor (Implant designed with Promade,
specific instrumentations available for intraoperative use. Lima Corporate Medical Systems, Villanova San Daniele
(c) The jigs are fixed to the bone with K-wires. (d) del Friuli, Italy)
Standard cutting jigs are widely used in pri- specific trajectory and can be mounted on the
mary and revision arthroplasty, whereas their use definitive implant for bone fixation. The intraop-
is precluded in musculoskeletal oncology. The erative availability of sterilized 3D-printed mod-
main objectives of using 3D printed PSI are to els helps the surgeon with correct orientation,
increase the accuracy of tumor resection in the especially when they present detachable parts to
pelvis and the precision of implant placement show the position of cutting PSI (Fig. 17.4d).
[11, 44]. In fact, custom-made prostheses usually Nowaday, we think that the 3D-printed PSIs
need the absolute precision in performing the should be considered a less expensive, easier, and
multiplanar osteotomies [45], and more than one comparable alternative to computer navigation
cut planes are necessary for the tumor resection for challenging tumor resections. Some Authors
(Fig. 17.4a). There are specific tools designed for [46] specifically analyzed this topic, showing a
bone resection (Fig. 17.4b), and other customized clinically acceptable accuracy of 2.62 mm vs
drill guides are studied for improving screw tra- 3.6 mm at the resection planes comparing PSI
jectories maximizing implant stability. The cut- guides with computer navigation. The relative
ting jigs reduce the errors derived from the difficulty compared to computer navigation is to
freehand use of oscillating saw blades, uncorrect make absolutely bare the target bony surface,
directions, and contamination due to tumor mar- which will perfectly align with the PSI footprint.
gin violation. The jigs should be fixed to the bone In the authors’ experience, the application of PSI
with one or more K-wires after correct alignment is valuable in most of the cases after a detailed
with host bone to reduce further errors related to preoperative planning together with dedicated
vibration of the saw blade or guide displacement engineers. A wide surgical exposure is important
during bone cut (Fig. 17.4c). Drill-guided PSI is because correct guide placement and accessibil-
used to guide drill holes that are planned for a ity with different cutting tools (osteotome or
oscillating saw) provide safe resection margins (EBM) technologies, with adequate internal
without compromising neurovascular structures. porous structures that could be considerably valid
The ability to improve accuracy in preopera- alternatives to allografts in terms of mechanical
tive planning and in real-life surgery using 3D scaffold. Based on the preoperative imaging stud-
models and 3D printed PSI may strongly influ- ies, it is possible to determine exactly the bone
ence the surgical, oncological, and functional defect. Many prosthetic models may be design
outcome. based on the unaffected site, but we usually sug-
gest a peripheral downsize of 1 mm in order to
achieve optimal fit of the implant and adequate
17.4 3D-Printed Prostheses soft tissue coverage, with attention for any mis-
in Pelvic Reconstruction calculation (too small or too short implant).
In recent years, the improvement of 3D printing

technology has reached high levels, up to the 17.4.2 Shape and Structure
point of being able to produce prosthetic implants.
These customized prostheses may be used as an The challenge in the use of custom implants in
innovative alternative to different biologic and pelvic reconstructions is achieving a perfect fit
prosthetic reconstructive strategies [1, 9, 47–50]. considering the distribution of forces during
However, few studied with limited case series weight-bearing activities at physiological status.
reported results at early/mid-term follow-up in Looking the literature, a wide spectrum of
pelvic reconstructions [12, 14, 15, 51, 52]. We implant designs has been reported, with different
recently reported our experience in a relatively shapes and concepts [12, 14, 15, 51, 52, 55, 56].
large series (41 cases) treated with custom- Custom-made 3D printed prostheses have spe-
designed 3D-printed prostheses in different cific indications in the reconstruction of periace-
oncologic and nononcologic settings [15]. tabular area, which represents the most
There are some concepts that should be con- demanding site for anatomy and hip joint func-
sidered in the implant design of a 3D-printed pel- tion. Dai et al. reported their experience in ten
vic prosthesis: (1) how to fill the bone defect; (2) patients with different designs and fixation on the
how to assess the correct shape and areas with remaining ilium (or the sacrum), pubic rami
porous surface structure; and (3) how to obtain a (same side or other side), and the ischial rami
stable fixation at long term; (4) how to optimize [56]. Wang et al. specifically evaluated the out-
the soft tissue reattachment, the osseointegration comes in 11 periacetabular 3D printed implants
between prosthesis and host bone, and its cover- with similar reconstructions, searching with dif-
age with vascularized tissue. ferent strategies a complete restoration of the pel-
vic ring to obtain satisfactory functional results
[19]. We do not emphasize a strict anatomical
17.4.1 How to Fill the Bone Defect pelvic ring reconstruction [14, 15]. In our experi-
ence, a 3D-printed custom-made prosthesis can
Historically, allograft prosthetic composite has be safely used when a good and stable interface
been widely used to reconstruct large bone bone/implant could be obtained in residual part
defects in the pelvis, despite the relatively high of the ilium or sacrum (Fig. 17.5a, b). The hip
complication rate [53, 54]. With the use of 3D joint can be therefore replaced with “conven-
printed technology, it is possible to print custom tional” modular cups (Fig. 17.5c), as well as the
implants in titanium metal with increasingly pop- proximal femur (Fig. 17.5d). Usually, the bone-
ularity and low reduction of cost and time. The contacting surfaces should be realized with a
industrial application of this process alloy porous structure to facilitate the bone ingrowth as
implants is that they are manufactured by metal proven by in vitro and in vivo studies [7]. The
laser sintering (DMLS) or electron beam melting metal 3D printing allows the production of
a b
c d
Fig. 17.5 Chondrosarcoma gr. 2 of the right acetabu- 3D-printed prosthesis with “conventional” modular cup.
lar area in a 38-year-old male treated with type II-III (d) Postoperative radiograph shows our philosophy of not
resection. (a) Preoperative CT scan and (b) surgical plan- pursuing the complete restoration of the pelvic ring
ning on virtual 3D model. (c) intraoperative photograph (Implant designed with C-Fit 3D®, Implantcast Ltd.,
showing the definitive implant of a custom-made Buxtehude, Germany)
implants with complex shapes, alternating porous according to a computer-aided design (CAD)
surfaces (implant/bone interface) and smooth model, so there is the possibility to create a
surfaces and minimizing local friction to overly- porous surface with ingrowth bone characteris-
ing important soft tissues. On the other hand, the tics [7, 8, 57]. The perfect fit between host bone
weight of the implant and stress forces concen- and prosthesis is a primary requirement for
tration can be optimized working on the custom- immediate stability and long-term fixation. Wong
izable texture of the internal structure. et al. described the use of printed guide plates for
intraoperative precise resection and implant
installation [46]. In our experience, 3D-printed
17.4.3 Long-Term Fixation PSIs are routinely used and are considered the
most effective tools for guided resection and
The long-term mechanical strength of a 3D reconstruction. During the last few years, we
printed pelvic prostheses is not guaranteed con- used different strategies for primary fixation to
sidering the limits of recent clinical use and mid- host bone, which today should be considered in
term follow-up of the largest published series combination: long cancellous screws, short corti-
[12, 14, 15, 51]. The EBM or SLM technology cal screws, press-fit porous stems, and small
alloys a successive layering of melted titanium hooks for stabilization (Fig. 17.6). In the design
Fig. 17.6 Same patient: Chondrosarcoma gr. Two of with 4.5 mm of diameter (black arrows), press-fit porous
the right acetabular area in a 38-year-old male. The stems (asterisks), and hook for stabilization (white star).
transparent virtual 3D model shows the strategies for pri- (Implant designed with C-Fit 3D®, Implantcast Ltd.,
mary fixation to host bone: long cancellous screws with Buxtehude, Germany)
6.5 mm of diameter (white arrows), short cortical screws
phase of the custom implant, we suggest to plan one of the most adaptable flaps for both periace-
counter screws as safety locking to avoid screw tabular and sacral reconstructions.
mobilization. When resection involves the proxi-
mal part of the ilium, sacroiliac joint, or the
sacrum, there is a need of further stabilization of 17.5 A
dvantages of 3D Printing
the implant for adequate loading transfer and bal- Technology
ance. In these cases, we suggest a posterior pedi- in Reconstruction for Pelvic
cle screw-rod fixation connected to tulip-head Defect
polyaxial screws directly designed on the custom
prosthesis, to add stability to the spinopelvic Advantages of 3D-printed pelvic prostheses are
continuity. listed in the following:
–– The shape and design of the 3D-printed pelvic

17.4.4 Soft Tissue Reattachment prostheses were based on thin-layer CT scans
and Prosthetic Coverage converted into a digital 3D-model. Thanks to
the multidisciplinary collaboration between
The soft tissue coverage of the prosthesis with engineers and surgeons, a more accurate
well-vascularized muscles is one of the most resection with PSIs may allow an improve-
important aspects to avoid wound dehiscence or ment of oncologic outcome. Moreover, preop-
deep infection [53, 58–60]. The porous surface erative 3D printed models of the anatomy and
can be realized not only in bone/implant interface tumor site may help to visualize, predict, and
but also in some areas (i.e., outer part of the better understand the surgical challenge.
ilium) to increase the friction and adherence of –– The optimization of 3D printed custom-jigs
the residual soft tissues to the prosthesis. In some guides for intraoperative resection is safe
cases, we designed specific large holes to guaran- and cheap compared to intraoperative navi-
tee a fixation of muscles/tendons through the gation in our experience, considering the
prosthesis [15], as well as reported by other limits of the latter reported in literature [16,
Authors [61]. Different flaps can be used by the 41]. This aspect influences the operation
reconstructive surgeon to plan a specific solution time and intraoperative implant exposure,
for implant coverage and wound closure [62, 63]. reducing consequently the risk of infectious
The rectus abdominis musculocutaneous flap is complications.
–– The possibility to produce anatomic implants planning and navigation in bone tumor resection.
with a wide freedom in terms of shapes and Orthopedics. 2013;36(7):e942–50.
6. Wong TM, Jin J, Lau TW, Fang C, Yan CH, Yeung
structure allows a stable reconstruction and a K, et al. The use of three-dimensional printing tech-
good restoration of function. The reported nology in orthopaedic surgery. J Orthop Surg (Hong
functional results evaluated with MSTS score Kong). 2017;25(1):2309499016684077.
at a mid-term follow-up ranged from 63% to 7. Shah FA, Snis A, Matic A, Thomsen P, Palmquist
A. 3D printed Ti6Al4V implant surface promotes
85% [12, 14, 15, 51, 52, 56]. bone maturation and retains a higher density of less
–– The bony ingrowth in the porous structure of aged osteocytes at the bone-implant interface. Acta
the prosthesis may reduce mechanical failures Biomater. 2016;30:357–67.
at long-term follow-up that usually occurs due 8. Sing SL, An J, Yeong WY, Wiria FE. Laser and
electron-beam powder-bed additive manufacturing
to osteolysis or nonunion at the junction in the of metallic implants: a review on processes, materials
reconstructions with APC [1, 64, 65]. and designs. J Orthop Res. 2016;34:369–85. https://
–– There are potential developments in 3D-printed doi.org/10.1002/jor.23075.
technology looking at newer materials to 9. Sun W, Li J, Li Q, Li G, Cai Z. Clinical effectiveness
of hemipelvic reconstruction using computer-aided
decrease the risks of infection and improve custom-made prostheses after resection of malig-
bone ingrowth. nant pelvic tumors. J Arthroplast. 2011;26:1508–13.
https://doi.org/10.1016/j.arth.2011.02.018.
10. Mumith A, Thomas M, Shah Z, Coathup M,

Blunn G. Additive manufacturing. Bone Jt J.
17.6 Conclusions 2018;100-B(4):455–60.
11. Mulford JS, Babazadeh S, Mackay N. Three-

The use of custom-made 3D-printed prostheses dimensional printing in orthopaedic surgery: review
represents today a well-established method for of current and future applications. ANZ J Surg.
2016;86(9):648–53.
reconstruction of large pelvic bone defects, which 12. Liang H, Ji T, Zhang Y, Wang Y, Guo W. Reconstruction
will become increasingly accessible in the next with 3D-printed pelvic endoprostheses after resection
decades. These implants have the potential objec- of a pelvic tumour. Bone Joint J. 2017;99-B:267–
tives to improve oncologic and functional out- 75. https://doi.org/10.1302/0301-620X.99B2.BJJ-
2016-0654.R1.
comes in patients with pelvic bone tumors, with 13. Wei R, Guo W, Ji T, Zhang Y, Liang H. One-step
acceptable complication rate and satisfactory reconstruction with a 3D-printed, custom-made pros-
safety. thesis after total en blocsacrectomy: a technical note.
Eur Spine J. 2017;26(7):1902–9.
14. Angelini A, Trovarelli G, Berizzi A, Pala E, Breda
A, Ruggieri P. Three-dimension-printed custom-
References made prosthetic reconstructions: from revision
surgery to oncologic reconstructions. Int Orthop.
1. Angelini A, Calabrò T, Pala E, Trovarelli G, Maraldi 2019;43(1):123–32. https://doi.org/10.1007/
M, Ruggieri P. Resection and reconstruction of pelvic s00264-018-4232-0.
bone tumors. Orthopedics. 2015;38(2):87–93. https:// 15. Angelini A, Kotrych D, Trovarelli G, Szafrański A,
doi.org/10.3928/01477447-20150204-51. Bohatyrewicz A, Ruggieri P. Analysis of principles
2. Ayvaz M, Bekmez S, Mermerkaya MU, Caglar O, inspiring design of three-dimensional-printed custom-
Acaroglu E, Tokgozoglu AM. Long-term results made prostheses in two referral centres. Int Orthop.
of reconstruction with pelvic allografts after 2020;44(5):829–37. https://doi.org/10.1007/s00264-
wide resection of pelvic sarcomas. Sci World J. 020-04523-y. Epub 2020 Mar 13.
2014;2014:605019. 16. Wong KC, Kumta SM, Geel NV, Demol J. One-step
3. Campanacci M, Capanna R. Pelvic resections: the reconstruction with a 3D-printed, biomechanically
Rizzoli Institute experience. Orthop Clin North Am. evaluated custom implant after complex pelvic tumor
1991;22(1):65–86. resection. Comput Aided Surg. 2015;20(1):14–23.
4. Sim FH, Choong PF, Weber KL. Master techniques 17. Imanishi J, Choong PF. Three-dimensional printed
in orthopaedic surgery: orthopaedic oncology and calcaneal prosthesis following total calcanectomy. Int
complex reconstruction. Montgomery, IL: Lippincott J Surg Case Rep. 2015;10:83–7.
Williams & Wilkins; 2011. 18. Fan H, Fu J, Li X, Pei Y, Li X, Pei G, Guo Z. Implantation
5. Ritacco LE, Milano FE, Farfalli GL, Ayerza MA, of customized 3-D printed titanium prosthesis in limb
Muscolo DL, Aponte-Tinao LA. Accuracy of 3-D salvage surgery: a case series and review of the lit-
erature. World J Surg Oncol. 2015;13:308. https://doi. 31. Hung CC, Li YT, Chou YC, Chen JE, Wu CC,

org/10.1186/s12957-015-0723-2. Shen HC, et al. Conventional plate fixation method
19. Wang B, Hao Y, Pu F, Jiang W, Shao Z. Computer- versus pre-operative virtual simulation and three-
aided designed, three dimensional-printed hemipel- dimensional printing-assisted contoured plate fixa-
vic prosthesis for periacetabular malignant bone tion method in the treatment of anterior pelvic ring
tumour. Int Orthop. 2018;42(3):687–94. https://doi. fracture. Int Orthop. 2018;43(2):425–31. https://doi.
org/10.1007/s00264-017-3645-5. org/10.1007/s00264-018-3963-2.
20. Li H, Qu X, Mao Y, Dai K, Zhu Z. Custom acetabular 32. Inzana JA, Trombetta RP, Schwarz EM, Kates SL,
cages offer stable fixation and improved hip scores Awad HA. 3D printed bioceramics for dual antibiotic
for revision THA with severe bone defects. Clin delivery to treat implant-associated bone infection.
Orthop Relat Res. 2016;474(3):731–40. https://doi. Eur Cell Mater. 2015;30:232–47.
org/10.1007/s11999-015-4587-0. 33. Song B, Zhao X, Li S, Han CJ, Wei QS, Wen SF, et al.
21. Berasi CC, Berend KR, Adams JB, Ruh EL, Lombardi Differences in microstructure and properties between
AV Jr. Are custom triflange acetabular components selective laser melting and traditional manufacturing
effective for reconstruction of catastrophic bone loss? for fabrication of metal parts: a review. Front Mech
Clin Orthop Relat Res. 2015;473(2):528–35. https:// Eng. 2015;10(2):111–25.
doi.org/10.1007/s11999-014-3969-z. 34. Palmquist A, Shah FA, Emanuelsson L, Omar O,

22. Wind MA Jr, Swank ML, Sorger JI. Short-term
Suska F. A technique for evaluating bone ingrowth
results of a custom triflange acetabular compo- into 3D printed, porous Ti6Al4V implants accurately
nent for massive acetabular bone loss in revision using X-ray micro-computed tomography and histo-
THA. Orthopedics. 2013;36(3):e260–5. https://doi. morphometry. Micron. 2017;94:1–8.
org/10.3928/01477447-20130222-11. 35. Fadero PE, Shah M. Three dimensional (3D) model-
23. Wyatt MC. Custom 3D-printed acetabular implants ling and surgical planning in trauma and orthopaedics.
in hip surgery: innovative breakthrough or expensive Surgeon. 2014;12(6):328–33.
bespoke upgrade? Hip Int. 2015;25(4):375–9. https:// 36. Perez M, Block M, Espalin D, Winker R, Hoppe

doi.org/10.5301/hipint.5000294. T, Medina F et al. editors. Sterilization of FDM-
24. Xu N, Wei F, Liu X, Jiang L, Cai H, Li Z, Yu M, Wu manufactured parts. Proceedings. Annual International
F, Liu Z. Reconstruction of the upper cervical spine Solid Freeform Fabrication Symposium, Austin, TX,
using a personalized 3D-printed vertebral body in Aug 2012; 2012.
an adolescent with Ewing sarcoma. Spine (Phila, 37. Angelini A, Guerra G, Mavrogenis AF, Pala E, Picci P,
PA 1976). 2016;41:E50–4. https://doi.org/10.1097/ Ruggieri P. Clinical outcome of central conventional
BRS.0000000000001179. chondrosarcoma. J Surg Oncol. 2012;106(8):929–37.
25. Mobbs RJ, Coughlan M, Thompson R, Sutterlin CE, 38. Loh AH, Wu H, Bahrami A, et al. Influence of bony
Phan K. The utility of 3D printing for surgical plan- resection margins and surgico pathological factors on
ning and patient-specific implant design for complex outcomes in limb-sparing surgery for extremity osteo-
spinal pathologies: case report. J Neurosurg Spine. sarcoma. Pediatr Blood Cancer. 2015;62(2):246–51.
2017;26(4):513–8. https://doi.org/10.3171/2016.9.SP 39. He F, Zhang W, Shen Y, et al. Effects of resection mar-
INE16371. gins on local recurrence of osteosarcoma in extremity
26. Yang M, Li C, Li Y, Zhao Y, Wei X, Zhang G, et al. and pelvis: systematic review and meta-analysis. Int J
Application of 3D rapid prototyping technology in Surg. 2016;36(Pt A):283–92.
posterior corrective surgery for Lenke 1 adolescent 40. Albergo JI, Farfalli GL, Ayerza MA, Ritacco LE,

idiopathic scoliosis patients. Medicine (Baltimore). Aponte-Tinao LA. Computer-assisted surgery (CAS)
2015;94(8):e582. in orthopedic oncology. Which were the indications,
27. Mulhern JL, Protzman NM, White AM, Brigido
problems and results in our first consecutive 203
SA. Salvage of failed total ankle replacement patients? [published online ahead of print, 2020 Jun
using a custom titanium truss. J Foot Ankle 24]. Eur J Surg Oncol. 2020:S0748-7983(20)30528-X.
Surg. 2016;55:868–73. https://doi.org/10.1053/j. https://doi.org/10.1016/j.ejso.2020.06.008.
jfas.2015.12.011. 41. Abraham JA, Kenneally B, Amer K, Geller DS. Can
28. Chana-Rodriguez F, Mananes RP, Rojo-Manaute J, navigation-assisted surgery help achieve negative
Gil P, Martinez-Gomiz JM, Vaquero-Martin J. 3D sur- margins in resection of pelvic and sacral tumors? Clin
gical printing and pre contoured plates for acetabular Orthop Relat Res. 2018;476(3):499–508. https://doi.
fractures. Injury. 2016;47(11):2507–11. org/10.1007/s11999.0000000000000064.
29. Boudissa M, Courvoisier A, Chabanas M, Tonetti
42. Ahmad S, Stevenson J, Mangham C, Cribb G,

J. Computer assisted surgery in preoperative planning Cool P. Accuracy of magnetic resonance imaging
of acetabular fracture surgery: state of the art. Expert in planning the osseous resection margins of bony
Rev Med Devices. 2018;15(1):81–9. tumours in the proximal femur: based on coronal
30. Cromeens BP, Ray WC, Hoehne B, Abayneh F, Adler T1-weighted versus STIR images. Skelet Radiol.
B, Besner GE. Facilitating surgeon understanding of 2014;43(12):1679–86.
complex anatomy using a three-dimensional printed 43. Sternheim A, Daly M, Qiu J, et al. Navigated pelvic
model. J Surg Res. 2017;216:18–25. osteotomy and tumor resection: a study assessing
the accuracy and reproducibility of resection planes institution. Clin Orthop Relat Res. 2014;472(1):349–
in sawbones and cadavers. J Bone Joint Surg Am. 59. https://doi.org/10.1007/s11999-013-3250-x.
2015;97(1):40–6. 54. Wilke BK, Houdek MT, Rose PS, Sim FH. Proximal
44. Gouin F, Paul L, Odri GA, Cartiaux O. Computer- femoral allograft-prosthetic composites: do they
assisted planning and patient-specific instruments for really restore bone? A retrospective review of revi-
bone tumor resection within the pelvis: a series of 11 sion allograft-prosthetic composites. J Arthroplast.
patients. Sarcoma. 2014;2014:842709. 2019;34(2):346–51.
45. Cernat E, Docquier PL, Paul L, Banse X, Codorean 55. Jentzsch T, Vlachopoulos L, Furnstahl P, Muller

IB. Patient specific instruments for complex DA, Fuchs B. Tumor resection at the pelvis using
tumor resection-reconstruction surgery within three-dimensional planning and patient-specific
the pelvis: a series of 4 cases. Chirurgia (Bucur). instruments: a case series. World J Surg Oncol.
2016;111(5):439–44. 2016;14(1):249.
46. Wong KC, Sze KY, Wong IO, Wong CM, Kumta 56. Dai KR, Yan MN, Zhu ZA, Sun YH. Computer-aided
SM. Patient-specific instrument can achieve same custom-made hemipelvic prosthesis used in extensive
accuracy with less resection time than naviga- pelvic lesions. J Arthroplast. 2007;22:981–6. https://
tion assistance in periacetabular pelvic tumor sur- doi.org/10.1016/j.arth.2007.05.002.
gery: a cadaveric study. Int J Comput Assist Radiol 57. Xiu P, Jia Z, Lv J, Yin C, Cheng Y, Zhang K, Song
Surg. 2016;11(2):307–16. https://doi.org/10.1007/ C, Leng H, Zheng Y, Cai H, Liu Z. Tailored surface
s11548-015-1250-x. treatment of 3D printed porous Ti6Al4V by micro-
47. Mankin HJ, Gebhardt MC, Jennings LC, Springfield arc oxidation for enhanced osseointegration via
DS, Tomford WW. Long term results of allograft optimized bone in-growth patterns and interlocked
replacement in the management of bone tumors. bone/implant interface. ACS Appl Mater Interfaces.
Clin Orthop Relat Res. 1996;324:86–97. https://doi. 2016;8(28):17,964–75.
org/10.1097/00003086-199603000-00011. 58. Tsantes AG, Papadopoulos DV, Vrioni G, Sioutis

48. Traub F, Andreou D, Niethard M, Tiedke C, Werner S, Sapkas G, Benzakour A, Benzakour T, Angelini
M, Tunn PU. Biological reconstruction follow- A, Ruggieri P, Mavrogenis AF. World association
ing the resection of malignant bone tumors of the against infection in orthopedics and trauma W A I
pelvis. Sarcoma. 2013;2013:745360. https://doi. O T Study Group on bone and joint infection defini-
org/10.1155/2013/745360. tions. spinal infections: an update. Microorganisms.
49. Abudu A, Grimer RJ, Cannon SR, Carter SR,
2020;8(4):476. Published 2020 Mar 27. https://doi.
Sneath RS. Reconstruction of the hemipel- org/10.3390/microorganisms8040476.
vis after the excision of malignant tumours. 59. Trovarelli G, Angelini A, Pala E, Cappellari A, Breda
Complications and functional outcome of prostheses. A, Ruggieri P. Infection in orthopaedic oncology: cru-
J Bone Joint Surg (Br). 1997;79:773–9. https://doi. cial problem in modern reconstructive techniques. Eur
org/10.1302/0301-620x.79b5.6749. Rev Med Pharmacol Sci. 2019;23(2 Suppl):271–8.
50. Ji T, Guo W, Yang RL, Tang XD, Wang YF. Modular https://doi.org/10.26355/eurrev_201904_17501.
hemipelvic endoprosthesis reconstruction–experience 60. Maricevich M, Maricevich R, Chim H, Moran SL,
in 100 patients with mid-term follow-up results. Eur J Rose PS, Mardini S. Reconstruction following partial
Surg Oncol. 2013;39:53–60. https://doi.org/10.1016/j. and total sacrectomy defects: an analysis of outcomes
ejso.2012.10.002. and complications. J Plast Reconstr Aesthetic Surg
51. Wang J, Min L, Lu M, Zhang Y, Wang Y, Luo Y, JPRAS. 2014;67(9):1257–66.
Zhou Y, Duan H, Tu C. Three-dimensional-printed 61.
Guo W. Reconstruction with 3D-printed
custom-made hemipelvic endoprosthesis for pri- Endoprosthesis. In: Guo W, Hornicek F, Sim
mary malignancies involving acetabulum: the design F, editors. Surgery of the pelvic and sacral
solution and surgical techniques. J Orthop Surg Res. tumor. Dordrecht: Springer; 2020. https://doi.
2019;14(1):389. Published 2019 Nov 27. https://doi. org/10.1007/978-94-024-1945-0_15.
org/10.1186/s13018-019-1455-8. 62. Stein MJ, Karir A, Ramji M, Allen M, Bain JR, Avram
52. Wang J, Min L, Lu M, Zhang Y, Wang Y, Luo Y, Zhou R, Boushey R, Auer R, Jarmuske M. Surgical outcomes
Y, Duan H, Tu C. What are the complications of of VRAM versus gracilis flaps for the reconstruction
three-dimensional-printed custom-made integrative of pelvic defects following oncologic resection✰. J
hemipelvic endoprostheses in patients with primary Plast Reconstr Aesthet Surg. 2019;72(4):565–71.
malignancies involving the acetabulum, and what is https://doi.org/10.1016/j.bjps.2018.12.044.
the function of these patients? Clin Orthop Relat Res. 63. Sagebiel TL, Faria SC, Balachandran A, Sacks JM,
2020. [Published online ahead of print, 2020 May 11].; You YN, Bhosale PR. Pelvic reconstruction with
https://doi.org/10.1097/CORR.0000000000001297. omental and VRAM flaps: anatomy, surgical tech-
53. Angelini A, Drago G, Trovarelli G, Calabrò T,
nique, normal postoperative findings, and complica-
Ruggieri P. Infection after surgical resection for pel- tions. Radiographics. 2011;31(7):2005–19. https://
vic bone tumors: an analysis of 270 patients from one doi.org/10.1148/rg.317115112.
64. Ippolito J, Thomson J, Beebe K, Patterson F,

65. Beadel GP, McLaughlin CE, Wunder JS, Griffin

Benevenia J. Outcomes following periacetabular AM, Ferguson PC, Bell RS. Outcome in two groups
tumor resection: A 25-year institutional experi- of patients with allograft-prosthetic reconstruc-
ence. J Surg Oncol. 2020. [Published online ahead tion of pelvic tumor defects. Clin Orthop Relat Res.
of print, 2020 Jun 28]; https://doi.org/10.1002/ 2005;438:30–5. https://doi.org/10.1097/01.blo.00001
jso.26088. 80048.43208.2f.
Particle Radiotherapy
18
Petra Georg and Eugen Boris Hug
18.1 Background Extra dose

Biological effect of carbon ions
Particles used in the radiotherapy are energetic

protons or positive ions (e.g., carbon ions). Photons
Particle therapy has been introduced to increase Protons
cure rates and/or decrease side effects for patients Carbon ions
Dose
who need radiotherapy as part of their cancer Extra dose

Tumor
treatment. The most common type of particle
therapy is proton therapy. The second type rou-
tinely used is treatment with carbon ions. For
protons and carbon ions, the dose increases,
while the particle penetrates the tissue and loses
energy continuously. Hence, the dose increases Depth
with increasing thickness up to the Bragg peak, Fig. 18.1 Dose depth curve for photons, protons and car-
where the maximum energy deposits and it bon ions
occurs near the end of the particle’s range. Both
proton and carbon ion therapies exhibit a defined
Bragg peak in the body, so they deliver their max- carbon ions cause damage of the DNA of tissue
imum lethal dosage at or near the tumor. cells, ultimately causing their death. Because of
Beyond the Bragg peak, the dose drops to zero their reduced ability to repair damaged DNA,
(for protons) or almost zero (for heavier ions) cancerous cells are particularly vulnerable to this
(Fig. 18.1). damage. From a radiation biology standpoint,
The advantage of this energy deposition pro- there is considerable rationale to support the use
file is that less energy is deposited into the healthy of carbon ion beams in treating cancer patients.
tissue surrounding the target tissue. Protons and Carbon ions are heavier than protons and so pro-
vide a higher relative biological effectiveness
(RBE), which increases with depth to reach the
maximum at the end of the beam’s range. Thus,
P. Georg (*) · E. B. Hug the RBE of a carbon ion beam increases as the
MedAustron Ion Therapy Center,
ions advance deeper into the tumor-lying region.
Wiener Neustadt, Austria
e-mail: [email protected]; Carbon ion radiotherapy provides the highest lin-
[email protected] ear energy transfer (LET) of any currently

204 P. Georg and E. B. Hug
Photons Protons Carbon ions
Tumors may repair Protons cause slightly more Carbon ions cause
or resist some damage damage than photons. 2-3 times more damage.
from photon irradiation.
Fig. 18.2 Biological differencies between photons, protons and carbon ions
available form of clinical radiation. This high can also be located primarily in the pelvic ring.
energy delivery to the tumor results in many Both tumor entities are characterized by slow, but
double-strand DNA breaks, which are very diffi- locally aggressive and invasive growth patterns
cult for the tumor to repair (Fig. 18.2). In com- leading to large tumor volumes. The management
parison, conventional radiation produces of these tumors is challenging, because they lie in
principally single strand DNA breaks, which close vicinity to critical structures, like spinal cord,
allow many of the tumor cells to survive. nerve roots, or bowel structures. These anatomic
There are clear advantages to treat otherwise structures often limit surgical access and resect-
intractable hypoxic and radioresistant cancers with ability, as well as the delivery of high radiation
carbon ions, while opening the door for substantially doses. Nonetheless, surgery remains the initial
hypofractionated treatment of normal and radiosen- standard treatment and postoperative high-dose
sitive disease with protons. The higher outright cell radiotherapy is frequently recommended. The
mortality produced by carbon ion radiotherapy may major problem is the insufficient local control, due
also provide a clearer antigen signature to stimulate to limited possibility of high dose application
the patient’s immune system. This might be benefi- because of normal tissue tolerance. Large random-
cial in improving the disease outcome. ized prospective studies comparing photons with
As far as pelvic bone tumors are difficult to protons and/or carbon ions are missing to date.
manage, particle therapy may widen the scope of This might be due to the fact that particle therapy
treatment possibilities with the aim of improving is presently not easily available and that these sub-
disease control in this patient population. groups of bone tumors have a comparatively low
incidence rate, so systematic randomized trials are
difficult to perform [1].
18.1.1 Chordoma
and Chondrosarcoma 18.1.1.2 C linical Evidence of Particle
Therapy
18.1.1.1 Rationale for Particle By using protons or carbon ions, the local control
Therapy rates can be improved to 70–100% in patients
Chordoma is a rare (1–4% of all malignant bone with extracranial chordomas [2–12]. Focusing on
tumors), slowly growing tumor arising from cel- the sacral chordoma, there are reports using pro-
lular remnants of the notochord, anywhere along ton or carbon treatment.
the spine. Sacrum is the predominant location of
this tumor in the pelvis. Chondrosarcomas are 18.1.1.3 Carbon Ion Radiotherapy
relatively rare bone tumors. Their predominant Japanese data are focusing on definitive carbon
location in the pelvis is also in the sacrum, but they ion radiotherapy (CIRT).
18 Particle Radiotherapy 205
At National Institute of Radiological Sciences higher toxicity occurred after radiation treatment.
in Chiba in Japan (NIRS), carbon ion therapy is Five patients showed a decrease in pain after
used since 20 years. Imai et al. reported their treatment [6].
results on local control and side effects. In the At Hyogo Ion Beam Medical Center (HIBMC)
initial report on phase I and II clinical trials in 38 in Japan, protons or carbon ions were used for
patients, the 5-year local control and survival rate treating patients with sacral chordomas, and they
was 89% and 86% in 38 patients [2]. Two patients report 94% local control and 83% overall sur-
(5%) developed severe skin toxicity after treat- vival at 3 years in 23 patients. In 9 patients (39%),
ment, 2 of them required skin grafting, and 15 they observed grade 3 or higher late toxicity. The
patients (16%) had neuropathic pain that most frequent event was dermatitis [grade 4 in
impacted their quality of life. In the next report five patients (22%)], followed by neuropathies,
on 95 patients, the authors showed 5-year local including motor disorder, sensory disorder/pain,
control and survival rates of 88% and 86%, 2 and urinary retention [grade 3 in four patients
patients had late grade 3 skin, and 2 patients had (17%)] [7].
grade 4 late skin and soft tissue complication; Clinical outcome comparison between CIRT
90% of patients remained ambulatory, and 50% and surgery for sacral chordoma was performed
needed no pain medication [3]. by Nishida et al. [8]. Although the patient popula-
In their extended patient population of 188 tion was very limited, 10 patients were treated
patients, their report local control was 77.2% and with surgery and 7 patients with CIRT. The local
overall survival was 81.1% at 5 years. There were recurrence-free survival rate at 5 years was 62.5%
grade 3 toxicity of the peripheral nerves in 6 for the surgery and 100% for the carbon ion RT
patients and grade 4 toxicity of the skin in 2 group. Urinary anorectal function worsened in 6
patients [4]. patients (60%) of the surgical group and remained
The second institution using carbon ions in the unchanged in all patients treated with carbon
treatment of sacral chordoma is the Heidelberg ions. Postoperative wound complications requir-
Ion Therapy center. In 2014, they started a ran- ing surgical treatment occurred in 3 patients
domized phase II trial of hypofractionated proton (30%) of the surgical group and in 1 patient
versus carbon ion radiation therapy in patients (14%) after CIRT. The functional outcome evalu-
with sacrococcygeal chordoma—the ISAC trial ated using the Muskuloskeletal Tumor Society
protocol. This is the only one randomized study scoring system (MSTS) was 55% in the surgery
comparing protons and carbon ions using the group and 75% in the carbon ion group, and car-
same dose and fractionation Scheme (22 frac- bon ion group had significantly higher scores in
tions with 3 Gy RBE per fraction). The aim of the emotional acceptance than did the surgery
this study is to confirm the toxicity results of the group [8].
Japanese data and to compare them with the tox- A direct comparison of clinical and functional
icity analysis of proton therapy given in the same outcomes between surgery and CIRT for pelvic
fractionation [5]. Their preliminary results using chondrosarcoma was published only in one report
carbon ion therapy in a patient cohort of 56 from Osaka Medical Center. They compared 24
patients show 79% local control and 100% over- patients who underwent surgery with 7 patients
all survival at 2 years. The 2- and 3-year local treated by CIRT, which was performed at NIRS
control probability was 76% and 53%, respec- and HIBMC. The surgery was performed with
tively. However, in this cohort, 15 patients were reconstruction in 13 patients, and 11 patients
treated for recurrent tumors. Local control was have resection only. The 5-year local control and
significantly better in the naive patients with 85% overall survival rates were 68% and 72%. Despite
at 2 years vs 47% in recurrent patients. A total of the limited number of patients treated with CIRT,
23 patients were irradiated with carbon ions in there were no significant differences in survival
combination with photon IMRT, while 33 or local control between these patients and surgi-
received carbon ion therapy only. No grade 3 or cal treated patients. The functional outcome was
measured with Musculoskeletal Tumor Society doses of 77.1–77.4 Gy RBE to central sacral chor-
(MSTS) functional score available in 21 patients. domas, where spinal canal/sacral nerve sparing
The mean MSTS score was 59%, and the mean was not possible because of tumor location [10].
score in CIRT patients (n = 7) was 72.6% vs In their updated results with 7.3 years of median
49.6% in patients treated with surgery (n = 14), follow-up, the 5- and 8-year actuarial local con-
showing better outcomes in CIRT patients. trol rates for the whole group were 81% and 74%,
Complication developed in 9 patients after sur- and for primary tumors, the rates were higher with
gery, and seven patients developed deep wound 95% and 85%, because local recurrence was less
infection at a median follow-up of 6 months. All common for primary tumors (11%) than for recur-
seven patients had periacetabular tumors and rent tumors (50%). The overall survival was 84%
were treated with prosthetic pelvic reconstruc- and 65% at 5 and 8 years. The actuarial risk for
tion. To reduce the infection, three patients late grade 3 and 4 toxicity was 13% at 8 years. No
needed removal of the implant, three needed late neurological toxicities were found with radia-
wound debridement, and one needed a flap recon- tion doses 72.0 Gy RBE, but 3 sacral neuropathies
struction. Implant dislocation was seen in 2 developed after doses of 76.6–77.4 Gy RBE [11].
patients, and pulmonary thrombosis in one In the updated analysis of tumor response,
patient. One patient exhibited nerve palsy, but Kabolizadeh et al. evaluated 40 patients with
eventually recovered. Complications developed unresected chordoma, treated with definitive pho-
in 5 patients after CIRT. Five pelvic compression ton/proton radiation therapy. Tumor location in
fractures and four cases of avascular femoral the sacrum was in 27 patients. With a median
head necrosis were seen in 5 patients. One patient follow-up of 50.3 months, the 5-year local control
needed a unilateral hip replacement because of and overall survival rates were 85.1% and 81.9%.
femoral neck fracture [9]. Additionally, the authors evaluated volumetric
response of the total target volume using MRI and
18.1.1.4 Proton Radiotherapy CT imaging. They found that significant volumet-
At Massachusetts General Hospital in Boston, the ric reduction of the total target volume presented
standard treatment for sacral tumors is surgery. at a median follow-up of 18 months, followed by
Within a phase II trial, DeLaney et al. reported further gradual reduction throughout the rest of
their results on 50 patients with tumors in the the follow-up period. Volumetric analysis was
spine and sacrum. There were 29 chordomas, 14 more reliable and reproducible in comparison
chondrosarcomas, and 7 other histologies. The with the modified RECIST [12].
majority of patients had surgery, and 13 patients The impact of tumor volume on overall sur-
had biopsy only (26%). Only 26 patients had pri- vival was also described by the Boston group in a
mary location of the disease in the sacrum. retrospective analysis of 24 patients with newly
Patients with sacral tumor received preoperative diagnosed, previously untreated spinal chordo-
radiotherapy to reduce risk of seeding during sur- mas (only core biopsy, no prior incision or resec-
gery, and after surgery, patients were treated up to tion). At a median follow-up of 56 months,
70.2 Gy RBE, if resections were R0 and up to overall survival was 91% and local control was
77.4 Gy RBE if R0 was not achieved. A mixture 81.5%. Tumor volume more than 500 cm3 was
of photons and protons was used. Local control correlated with worse survival (50% at
and overall survival for the whole group were 56 months). Long-term side effects included 8
78% and 87% at 5 years. Radiation complications sacral insufficiency fractures, 1 secondary malig-
grade 3 or more developed in 6 patients (two neu- nancy, 1 foot drop, 1 erectile dysfunction, 1 peri-
ropathies, one erectile dysfunction, one rectal neal numbness, 2 worsening urinary/fecal
bleeding, and two sacral insufficiency fractures). incontinence, and 4 grade 2 rectal bleedings [13].
No spinal cord injuries were seen. Grade 3 sacral Authors from Paul Scherrer Institute (PSI)
neuropathies and erectile dysfunction occurred at reported their initial experience with spot
scanning-based technology using protons in 40 In a large retrospective cohort, authors from

chordoma patients. Published results are not Hyogo Ion Beam Medical center evaluated pro-
focused on sacrum, but reporting mobile spine ton beam therapy results in 96 patients with bone
and sacrum together, 11 patients had tumor sarcomas. Seventy two patients (75%) had chor-
located in sacrum. Twenty one patients under- doma, 20 patients (20.8%) had chondrosarcoma,
went macroscopically complete surgical resec- and four patients (7.2%) had osteosarcoma.
tion (no GTV left); the remaining 19 were Patients received a median total dose of 70 Gy
treated with macroscopic disease, 31 patients RBE. The most frequent location was skull base
were treated with protons only, 9 with a mix of in 68 patients (70.8%) and sacral spine only in 13
photons and protons. Thirty-seven patients patients (13.5%), 12 patients with sacral tumors
(93%) received >70 Gy RBE (range, 59.4–75.2). had chordoma, and only 2 patients of 13 patients
Local control was 62% and overall survival was with sacral tumors underwent surgical resection.
80% at 5 years. In patients without metal stabi- The authors did not report data of sacral tumors
lization device (part of this may be explained by separately, but they reported the outcome data of
the dosimetric uncertainties due to metal implant patients with tumors of the spine (including 8
artifacts and part may be due to a correlation patients with tumor location in cervical spine,
between need of surgical stabilization and big- 5 in lumbar spine, 2 with lumbosacral spine, and
ger and more infiltrating tumors), the local con- 13 with sacral tumors, total 28 patients). The
trol was 100%.In sacral patients, there was only 5-year local control and overall survival rates
one case of late grade 3 toxicity with subcutane- were 55.6% and 70.7%. Late grade 3 and higher
ous fistula requiring multiple surgical debride- toxicities occurred in 9 patients (9.4%), and the
ment [14]. For the first time here, the negative noncerebral toxicities were musculoskeletal and
effect of metal (e.g., titanium) implant material connective tissue disorders in 3 and necrosis in 2
used for reconstruction/spinal stabilization on patients [16].
the local control was described. The reasons All data on carbon ion or proton radiotherapy
may be multifactorial, the authors describe the suggest good local control and functional out-
dosimetric uncertainties, because they may comes, suggesting a good alternative for surgery
impact the range calculation for particle therapy, or even avoiding surgery. The Italian Sarcoma
but it also includes the uncertainty in precise Group initiated in 2016 a randomized and obser-
delineation of target and organs at risk (espe- vational study on surgery versus definitive radia-
cially in the spinal canal) because of the imag- tion therapy in primary localized disease (sacral
ing artifacts. If a debulking surgery is planned, chordoma, SACRO Study). This study is aimed
the possibility of substituting metal implants at estimating the effectiveness of definitive radio-
with carbon fiber devices should be considered therapy as compared to standard surgical treat-
to enable radiation with curative intent. ment for patients with primary sacral chordoma
In the clinical outcome paper on 26 pediatric who are candidates to a complete en bloc resec-
patients (mean age at time of proton therapy was tion, in terms of relapse-free survival (RFS). The
13.2 years) with chordoma and chondrosarcoma radiotherapy options used in this study are proton
treated at PSI, they reported local control rates of or carbon ion radiotherapy. The secondary objec-
81% for chordoma and 80% for chondrosarcoma tives are to estimate the efficacy, activity, safety,
at 5 years and the corresponding overall survival and quality of life for definitive radiotherapy as
rates were 89% for chordoma and 75% for chon- compared to standard surgery, as well as to iden-
drosarcoma, with no high-grade late toxicities tify radiological and pathological characteristics
observed. However, the majority of patients had that might be used as predictors of relapse-free
skull base location of the tumors (17 patients), 8 survival, progression-free survival, and overall
patients had spinal tumor location, and only one survival (ClinicalTrials.gov identifier (NCT num-
patient had sacral chordoma [15]. ber): NCT02986516).
18.1.2 Ewing Sarcoma age. No severe acute toxicity was observed. One
patient with spinal tumor developed local failure,
18.1.2.1 Rationale for Particle and the tumors located in the pelvis were con-
Therapy trolled. But all pelvic tumor patients developed
Ewing’s sarcomas are highly sensitive to radiation distant metastases. So the authors concluded that
therapy. Their locations often involve bones, local control is favorable, but distant control is
which are not so easily resectable, where radio- unsatisfactory [20].
therapy is well-established treatment. Radiotherapy
is used in the postoperative setting for patients 18.1.2.4 Proton Radiotherapy
with close or positive resection margins and some- Report from Paul Scherrer Institute on 38 pediat-
times in the setting of a poor or slow clinical ric patients (median age 9.9 years) with Ewing
response to neoadjuvant chemotherapy as an addi- sarcoma treated with pencil beam scanning pro-
tive neoadjuvant treatment. Radiation is typically ton therapy shows 81.5% local control and 85%
used instead of surgery for children with unresect- overall survival at 5 years. The majority of these
able tumors or in cases in which surgery would patients had the primary tumor location in the
result in severe, mutilating morbidity. Nevertheless, axial/pelvic site (27 pts., 71.7%). The outcome
surgery plays a substantial role in the management data in this patient subcohort were 75.1% local
of these tumors due to several reports where not control and 80.6% overall survival at 5 years. All
only local but also distant failure occurred more local recurrences developed in the irradiation
frequently in patients treated with radiotherapy field and in nonextremity locations. Two patients
only than in those treated with surgery or surgery– developed late grade 3 toxicity (kyphoscoliosis in
radiotherapy [17–19]. However, definitive radio- 1 pt. and endocrine dysfunction in 1 pt) [21].
therapy causes side effects as a result of the volume At Massachusetts General Hospital, Rombi
of normal tissue also irradiated. Due to the steep et al. evaluated the outcome of 30 pediatric
dose profile, particle therapy offers better sparing Ewing sarcoma patients treated with proton ther-
of normal tissue, therefore reducing the toxicity. apy. The 3-year local control and overall survival
rates were 86% and 89%, respectively. The only
18.1.2.2 C linical Evidence of Particle severe late toxicities were hematological malig-
Therapy nancies associated with the use of topoisomerase
A majority of reviewed reports on particle ther- and anthracycline. However, only 4 patients in
apy for Ewing sarcoma focus on the use of pro- this cohort had pelvic tumor mass and 2 patients
tons. This has two major reasons; first of all, the had lumbosacral spine tumors [22].
accessibility of proton therapy is much broader,
because there are simply more proton only cen-
ters than carbon ion therapy centers. The other 18.1.3 Osteosarcoma
reason is also the radiosensitivity of this histol-
ogy. Carbon ions offer stronger biological effects 18.1.3.1 Rationale for Particle
because of their high LET, but this is the stron- Therapy
gest argument in their use in radioresistant tumors Osteosarcoma is the most common primary bone
such as osteosarcomas. malignancy in children and adolescents and is
very radioresistant [23]. Neoadjuvant chemother-
18.1.2.3 Carbon Ion Radiotherapy apy, followed by surgical resection, and further
The only report about the use of carbon ion radio- adjuvant chemotherapy is the typical treatment
therapy in Ewing sarcomas is coming from the approach for high-grade osteosarcomas [24].
NIRS experience. They reported on five unresect- While this approach yields an acceptable overall
able patients: 3 patients had tumors located in the survival for resectable cases, outcomes remain
pelvis, and two patients in the spine. After carbon poor for unresectable osteosarcomas such as
ion radiotherapy, 2 patients showed tumor shrink- those in the pelvis or trunk [25].
Carbon ion radiotherapy (CIRT) has stronger survival rates were 79% and 52%. Sacral tumors
biological effects and more conformal dose dis- were excluded, the major histology was osteosar-
tribution compared to photon- and proton-based coma in 13 patients and chondrosarcoma in 13
therapies [26, 27]. patients, and other histologies were chordoma
[9], malignant fibrous histiocytoma (MFH) [7],
18.1.3.2 C linical Evidence of Particle Ewing sarcoma, [2] and others [4]. One patient
Therapy had grade 3 and one patient grade 4 late skin tox-
Surgery remains the standard for the treatment of icity with skin ulcer requiring grafts. Vertebral
osteosarcoma; however, similar to Ewing sar- body compression occurred in 7 patients. Twenty-
coma, osteosarcoma also develops in bones, two of the surviving 28 patients who had primary
where a radical surgery without major morbidity tumors remained ambulatory without supportive
is not possible. Osteosarcomas are well known as devices, so the authors concluded that CIRT was
radioresistant tumors, so the rationale of using both effective and safe for the treatment of
carbon ion therapy is obvious. patients with unresectable spinal sarcoma [29].
First experience on pediatric unresectable osteo-
18.1.3.3 Carbon Ion Radiotherapy sarcoma presented results on 26 patients (median
First report on efficacy and safety of carbon ion age 16 years) with inoperable osteosarcoma of the
radiotherapy in bone and soft tissue sarcomas trunk (24 pelvic, 1 mediastinal, and 1 paraverte-
comes from NIRS from a phase I/II dose escala- bral). Median CIRT dose was 70.4 Gy RBE deliv-
tion study. Fifty-seven patients with 64 sites of ered in 16 fractions. Local control was 69.9% and
bone and soft tissue sarcomas not suitable for 62.9%, and overall survival was 50.0% and 41.7%
resection received carbon ion radiotherapy. at 3 and 5 years, respectively. Grade 3–4 adverse
Tumors involved the spine or paraspinal soft tis- events excluding fractures of affected bone were
sues in 19 patients, pelvis in 32 patients, and observed in 4 cases. There was one case of grade 3
extremities in six patients. Bone sarcomas had 41 skin toxicity, one case of grade 4 skin toxicity, and
patients with osteosarcoma [15], chordoma [11], 2 cases of neurologic dysfunction due to nerve
chondrosarcoma [6], primitive neuroectodermal injury. There was 1 case of grade 4 bone toxicity in
tumor (PNET) [5], malignant fibrous histiocy- which the sacrum (S1) was involved with the dis-
toma (MFH) [1], and other histologies [3], and 16 ease and developed a fracture after CIRT [30].
patients had soft tissue sarcomas. The total dose At Heidelberg Ion Therapy Center, a nonran-
ranged from 52.8 to 73.6 Gy (RBE) and was domized therapy trial to determine the safety and
administered in 16 fixed fractions over 4 weeks. efficacy of heavy ion radiotherapy in patients
The median tumor size was 559 cm3 (range: with nonresectable osteosarcoma in children
20–2290 cm3). The local control rates were 88% older than 6 years was started in 2010. Desired
and 73% at 1 year and 3 years of follow-up, target dose is 60–66 Gy RBE with 45 Gy RBE
respectively. The 1- and 3-year overall survival proton therapy and a carbon ion boost of
rates were 82% and 46%, respectively. Looking 15–21 Gy RBE. The primary objectives are the
on the bone and cartilage manifestations, local determination of feasibility and toxicity of proton
control could be reached in 11/15 osteosarcoma therapy and heavy ion therapy using carbon ions
cases, in 12/13 chordoma manifestations, and 5/7 boost. Secondary objectives are tumor response,
chondrosarcoma lesions [28]. disease-free survival and overall survival. The
The next experience with CIRT in the treat- aim is to improve outcome for patients with non-
ment of unresectable sarcoma concentrated on resectable osteosarcoma [31].
spinal sarcomas only. Matsumoto et al. evaluated The experience from HIBMC showed very
the outcome in 47 patients with 48 medically good results in 91 patients with nonmetastatic
unresectable spinal sarcomas included in phase I/ unresectable or incompletely resected pelvic
II and phase II clinical trials for bone and soft tis- bone and soft tissue sarcomas of the pelvis. The
sue sarcomas. Five-year local control and overall particle therapy with protons was performed in
52 patients, and carbon ion therapy was per- tion with surgery. Here, local control rates up to
formed in 39 patients. The histologic type was 70–85% at 3–5 years can be reached. If surgery is
chordoma in 53 patients, chondrosarcoma in 14, not possible, the reports on definitive carbon
osteosarcoma in 10, malignant fibrous histiocy- treatment are increasing with reported local con-
toma/undifferentiated pleomorphic sarcoma in 5, trol rates of 70–94% at 3–5 years (Table 18.1).
and others in 9 patients. The 3-year local control All together, the data presented suggest that the
and overall survival rates were 92% and 83%. use of particle therapy is reasonable and offers a
Late grade 3 or higher toxicity was observed in good therapeutic alternative, which can avoid,
23 patients (25%). Late grade 3 toxicities involved i.e., mutilating surgery in many cases.
peripheral nerves in 6 patients, pain in 5, bone in However, clear recommendation to select
2, genitourinary tract in 2, muscle in 2, skin in 2, between the indication for protons or carbon ion
and the vascular system in 1 patient. Late grade 4 cannot be given based on the reported data.
toxicities involved skin in 9 patients [32]. Randomized trials comparing protons vs carbon
ions in bone tumors are limited to the ISAC pro-
18.1.3.4 Proton Radiotherapy tocol for sacrococcygeal chordoma [5]. However,
At Massachusetts General Hospital, unresectable in cases when surgery is possible without major
or incompletely resected osteosarcoma is treated mutilation, a combination with proton treatment
with proton therapy or with combination of pho- especially if gross tumor resection was performed
ton and proton therapy. The initial report on 55 is indicated. In cases of inoperability or major
patients with a median age of 29 years (2–76) mutilation, definitive carbon treatment offers a
showed local control 82% and 72% at 3 and valuable treatment choice (Table 18.1).
5 years, and the five-year OS was 67%. The extent
of surgical resection did not correlate with out-
come. Grade 3–4 late toxicity was seen in 30.1% 18.3 Conclusion
of patients. One patient died from treatment-
associated acute lymphocytic leukemia, and one The management of pelvic bone tumors remains
from secondary carcinoma of the maxilla [33]. very challenging. The main reason for that is the
limited radical resectability, so additive treatment
modalities should be used. Even in cases where
18.2 Summary resectability can be reached, impaired functional
outcome can be expected. In cases of implant
Bone tumors involving the pelvis are in the reconstruction, deep wound infection is still a
majority of cases chordomas, chondrosarcomas, major problem.
Ewing sarcomas, and osteosarcomas. For other Particle therapy opens the window of new ther-
histologies, the evidence of particle therapy is apeutic options aiming at improving not only the
very limited and it can be recommended in “classical” outcome parameters like local control
selected cases, i.e., in pediatric patients and and survival but also functional outcomes.
young adults when reduction of integral dose In the case of sacral chordoma, the recommen-
delivered to healthy tissue is needed in order to dation for using carbon ion or proton radiotherapy
avoid late complications or secondary radiation- for definitive treatment after biopsy only or in
induced malignancies. Other examples are when patients who refused surgery was postulated by the
dose escalation is aimed for or in previously irra- Chordoma Global Consensus Group [34]. Also, in
diated cases. For the histologies mentioned cases of local-regional recurrent chordoma, parti-
above, there is evidence for the use of proton as cle therapy should be considered [35]. ESMO–
well as carbon ion therapy in combination with PaedCan–EURACAN Clinical Practice Guidelines
surgery or as an exclusive local treatment option. recommend also new techniques (e.g., proton and
According to the literature, proton treatment is carbon ion beam RT), particularly for unresectable
used more in pediatric patients and in combina- primary osteosarcoma and chondrosarcoma [36].
Table 18.1 Treatment outcomes of particle radiotherapy for pelvic bone tumors
Author Diagnosis Patients (n) Particle Local control Overall survival
Imai et al. Chordoma 38 Carbon ion 89% (5y) 86% (5y)
(2010) [2]
Imai et al. Chordoma 95 Carbon ion 88% (5y) 86% (5y)
(2011) [3]
Imai et al. Chordoma 188 Carbon ion 77.2% (5y) 81.1% (5y)
(2016) [4]
Uhl et al. Chordoma 23 Photon/carbon ion 79% (2y) 100% (2y)
(2015) [5] 33 Carbon ion
Mima et al. Chordoma 23 Carbon ion 94% (3y) 83% (3y)
(2014) [6]
DeLaney et al. Chordoma, chondrosarcoma, 50 Photon/proton± 78% (5y) 87% (5y)
(2019) [11] and others surgery
DeLaney et al. Chordoma, chondrosarcoma, 50 Photon/proton± 74% (8y) 65% (8y)
(2014) [12] and others Surgery
Kabolizadeh Chordoma 40 Proton/proton± 85.1% (5y) 81.9% (5y)
et al. (2017) [12] Surgery
Chen et al. Chordoma (spine) 24 Proton 81.5% (5y) 91% 85y)
(2013) [13]
Staab et al. Chordoma 40 Proton ± surgery 62% (5y) 80% (5y)
(2011) [14]
Rombi et al. Chordoma, 26 Proton ± surgery 81% (5y) 89% (5y)
(2013) [15] Chondrosarcoma, and 80% (5y) 75% (5y)
pediatrics (skull base, spine,
sacrum)
Demizu et al. Chordoma, 72 Proton ± surgery 55.6% (5y) 70.7% (5y)
(2017) [16] Chondrosarcoma, and 20
osteosarcoma 4
(skull base, spine, and sacrum)
Weber et al. Ewing sarcoma 38 Proton ± surgery 81.5% (5y) 85% (5y)
(2017) [21] Axial and pelvic 27 75.1% (5y) 80.6% (5y)
Rombi et al. Ewing sarcoma 30 Proton ± surgery 86% (3y) 89% (3y)
(2012) [22]
Kamada et al. Osteosarcoma, 15 Carbon ion 73% (3y) 46% (3y)
(2002) [28] Chordoma, 11
Chondrosarcoma, and PNETa, 6
MFHb, and other, 5
Soft tissue sarcoma 1
3
16
Matsumoto et al. Osteosarcoma, 13 Carbon ion 79% (5y) 52% (5y)
(2013) [29] Chondrosarcoma, 13
Chordoma, 9
MFHb, 7
Ewing sarcoma, and 2
Other 4
Mohamad et al. Osteosarcoma 26 Carbon ion 69.9% (3y) 50% (3y)
(2018) [30] Pelvis [24], trunk 62.9% (5y) 41.7% (5y)
Demizu et al. Chordoma, 53 Proton (52) 92% (3y) 83% (3y)
(2017) [32] Chondrosarcoma 14 Carbon ion (39)
Osteosarcoma 10
MFHb 5
Other 9
Ciernik et al. Osteosarcoma 55 Proton or photon/ 82% (3y) 67% (5y)
(2011) [33] proton + surgery 72% (5y)
PNET Primitive neuroectodermal tumor
a
MFH Malignant fibrous histiocytoma

b
There are no randomized trials directly com- chondrosarcomas, and other sarcomas. J Surg Oncol.
2014;110(2):115–22.
paring protons and carbon ions published so far. 12. Kabolizadeh P, Chen YL, Liebsch N, et al. Updated
One randomized trial is currently recruiting outcome and analysis of tumor response in mobile
patients for sacral chordoma [5].The limited spine and sacral chordoma treated with definitive
reported evidence shows no difference between high-dose photon/proton radiation therapy. Int J
Radiat Oncol Biol Phys. 2017;97(2):254–62.
using protons vs carbons. However, looking into 13. Chen YL, Liebsch N, Kobayashi W, et al. Definitive
biological behavior of these particles, carbons high-dose photon/proton radiotherapy for unresected
should be considered in radioresistant tumor his- mobile spine and sacral chordomas. Spine (Phila PA
tologies and in large nonresectable tumor volumes, 1976). 2013;38(15):E930–6.
14. Staab A, Rutz HP, Ares C, et al. Spot-scanning proton
where a big tumor burden has to be managed. radiation therapy for pediatric chordoma and chon-
drosarcoma. Clinical outcome of 26 patients treated at
Paul Scherrer Institute. Int J Radiat Oncol Biol Phys.
2011;81(4):489–96.
References 15. Rombi B, Ares C, Hug EB, et al. Spot-scanning pro-
ton radiation therapy for pediatric chordoma and
1. Goitein M. Trials and tribulations in charged particle chordrosarcoma:clinical outcome of 26 patients
therapy. Radiother Oncol. 2010;95(1):23–31. treated at Paul Scherrer Institute. Int J Radiat Oncol
2. Imai R, Kamada T, Tsuji H, et al. Effect of carbon ion Biol Phys. 2013;86(3):578–84.
radiotherapy for sacral chordoma: results of phase I-II 16. Demizu Y, Mizumoto M, Onoe T, et al. Proton beam
and phase II clinical trials. Int J Radiat Oncol Biol therapy for bone sarcomas of the skull base and spine.
Phys. 2010;77(5):1470–6. A retrospective nationwide multicenter study in
3. Imai R, Kamada T, Sugahara, et al. Carbon ion Japan. Cancer Sci. 2017;108:972–7.
radiotherapy for sacral chordoma. Br J Radiol. 17. Indelicato DJ, Keole SR, Shahlaee AH, et al. Impact
2011;84(Spec 1):S48–54. of local management on long-term outcomes in Ewing
4. Imai R, Kamada T, Araki N. Carbon ion radia- tumors of the pelvis and sacral bones: the University
tion therapy for unresectable sacral chordoma: an of Florida experience. Int J Radiat Oncol Biol Phys.
analysis of 188 cases. Int J Radiat Oncol Biol Phys. 2008;72:41–8.
2016;95(1):322–7. 18. Paulussen M, Ahrens S, Dunst J, et al. Localized
5. Uhl M, Edler L, Jensen AD, et al. Randomized phase Ewing tumor of bone: final results of the coopera-
II trial of hypofractionated proton versus carbon ion tive Ewing’s sarcoma study CESS 86. J Clin Oncol.
radiation therapy in patients with sacrococcygeal 2001;19:1818–29.
chordoma-the ISAC trial protocol. Radiat Oncol. 19. Rodriguez-Galindo C, Navid F, Liu T, et al. Prognostic
2014;9:1–10. factors for local and distant control in Ewing sarcoma
6. Uhl M, Welzel T, Jensen A, et al. Carbon ion beam family of tumors. Ann Oncol. 2008;19:814–20.
treatment in patients with primary and recurrent 20. Iwata S, Yonemoto T, Ishii T, et al. Efficacy of carbon-
sacrococcygeal chordoma. Strahlenther Onkol. ion radiotherapy and high-dose chemotherapy for
2015;191:597–603. patients with unresectable Ewing’s sarcoma family of
7. Mima M, Demizu Y, Jin D, et al. Particle therapy tumors. Int J Clin Oncol. 2013;18(6):1114–8.
using carbon ions or protons as a definitive therapy for 21. Weber DC, Murray FR, Correia D, et al. Pencil

patients with primary sacral chordoma. Br J Radiol. beam scanned protons for the treatment of patients
2014;87(1033):20130512. with Ewing sarcoma. Pediatr Blood Cancer.
8. Nishida Y, Kamada T, Imai R, et al. Clinical outcome 2017;64:e26644.
of sacral chordoma with carbon ion radiotherapy 22. Rombi B, DeLaney T, McDonald S, et al. Proton
compared with surgery. Int J Radiat Oncol Biol Phys. radiotherapy for pediatric Ewing sarcoma. Initial
2011;79(1):110–6. clinical outcomes. Int J Radiat Oncol Biol Phys.
9. Outani H, Hamada K, Imura Y, et al. Comparison 2012;82(3):1142–8.
of clinical and functional outcome between surgical 23. Longhi A, Errani C, De Paolis M, Mercuri M, Bacci
treatment and carbon ion radiotherapy for pelvic chor- G. Primary bone osteosarcoma in the pediatric age:
dosarcoma. Int J Clin Oncol. 2016;21:186–93. state of the art. Cancer Treat Rev. 2006;32:423–36.
10. DeLaney TF, Liebsch NJ, Pedlow FX, et al. Phase II 24. Bielack SS, Kempf-Bielack B, Delling B, et al.

study of high-dose photon/proton radiotherapy in the Prognostic factors in high-grade osteosarcoma of the
management of spine sarcomas. Int J Radiat Oncol extremities or trunk: an analysis of 1,702 patients
Biol Phys. 2009;74(3):732–9. treated on neoadjuvant cooperative osteosarcoma
11. DeLaney TF, Liebsch NJ, Pedlow FX. Long-term
study group protocols. J Clin Oncol. 2002;20:776–90.
results of phase II study of high dose photon/proton 25.
Ham SJ, Kroon HM, Koops HS, Hoekstra
radiotherapy in the management of spine chordomas, HJ. Osteosarcoma of the pelvis oncological results of
40 patients registered by The Netherlands Committee 32. Demizu Y, Jin D, Sulaiman NS, et al. Particle

on Bone Tumours. Eur J Surg Oncol. 2000;26: therapy using protons or carbon ions for unresect-
53–60. able or incompletely resected bone and soft tissue
26. Schulz-Ertner D, Jakel O, Schlegel W. Radiation
sarcomas of the pelvis. J Radiat Oncol Biol Phys.
therapy with charged particles. Semin Radiat Oncol. 2017;98(2):367–74.
2006;16:249–59. 33. Ciernik F, Niemierko A, Harmon DC, et al.

27. Pompos A, Durante M, Choy H. Heavy ions in cancer Proton-based radiotherapy for unresectable or
therapy. JAMA Oncol. 2016;2:1539–40. incompletely resected osteosarcoma. Cancer.
28. Kamada T, Tsujii H, Tsuji H, et al. Efficacy and safety 2011;117(19):4522–30.
of carbon ion radiotherapy in bone and soft tissue sar- 34. Stacchiotti S, Sommer J, Ares C, et al. Building a
comas. J Clin Oncol. 2002;20:4466–71. global consensus approach to chordoma: a posi-
29. Matsumoto K, Imai R, Kamada T, et al. Working tion paper from the medical and patient community.
Group for Bone and Soft Tissue Sarcomas. Soft tis- Lancet Oncol. 2015;16(2):e71–83.
sue, impact of carbon ion radiotherapy for primary 35. Stacchiotti S, Gronchi A, Fossati P, et al. Best prac-
spinal sarcoma. Cancer. 2013;119:3496–503. tices for the management of local-regional recurrent
30. Mohamad O, Imai R, Kamada T, et al. Carbon ion chordoma: a position paper by the Chordoma Global
radiotherapy for inoperable pediatric osteosarcoma. Consensus Group. Ann Oncol. 2017;28(6):1230–42.
Oncotarget. 2018;9(33):22,976–85. 36. Casali PG, Bielack S, Abecassis N, et al. Bone sarco-
31. Blattmann C, Oertel S, Schulz-Ertner D, et al. Non- mas: ESMO-PaedCan-EURACAN Clinical Practice
randomized therapy trial to determine the safety Guidelines for diagnosis, treatment and follow-up.
and efficacy of heavy ion radiotherapy in patients ESMO guidelines committee, PaedCan and ERN
with non-resectable osteosarcoma. BMC Cancer. EURACAN. Ann Oncol. 2018;29(Supplement_4)
2010;10:96. :iv79–95.
Chemotherapy for Pelvic Bone
Tumors 19
Antonella Brunello and Vittorina Zagonel
19.1 Introduction endocrine- sensitive breast cancer with bone

metastases only) there is evidence supporting
The vast majority of pelvic bone tumors are sec- less-intensive schedules [2, 3].
ondary, due to metastatic spread of cancer of Primary malignant bone tumors are chondro-
other primary sites and hematopoietic neoplasms sarcoma, osteosarcoma, Ewing sarcoma, high
such as multiple myeloma or solitary plasmacy- grade sarcomas, and chordomas. Giant cell
toma of bone. tumors, which are locally aggressive benign
Primary bone tumors are rare, accounting for tumors that may rarely metastasize, are also
less than 0.2% of malignant neoplasms as among primary tumors of bone, which may local-
reported in the EUROCARE (European Cancer ize in the pelvis.
Registry-based study on survival and care of can- Given the rarity of sarcoma, it is of utmost
cer patients) database [1]. importance to have initial diagnosis and treatment
Medical treatment of metastatic bone tumors guided by a specialized multidisciplinary tumor
varies widely according to primary tumor type board, and recent evidence suggests that such an
and may be highly effective especially for approach is associated with improved survival [4].
endocrine- sensitive tumors (i.e., breast cancer
and prostate cancer). Multidisciplinary discus-
sion must be therefore undertaken before making 19.2 Osteosarcoma
definitive decisions.
In many cases, treatment of bone metastases Osteosarcoma is the most common type of bone
from solid tumors or myeloma includes the use of sarcoma in children and young adults, along with
antiresorptive bone agents (i.e., zoledronic acid and Ewing sarcoma, but it can also occur in older
denosumab), beside antitumor-directed therapy. subjects. In children and young adults, osteosar-
Both zoledronic acid and denosumab are gen- coma is most frequently located in the extremity,
erally administered monthly, but for patients with with an increasing proportion of axial localiza-
tumors which have good prognosis (i.e., tion with age.
Chemotherapy has been recognized as essen-
tial for the treatment of high-grade osteosarcoma
since the pivotal study by Link and colleagues
A. Brunello (*) · V. Zagonel
[5]. Indeed, before the introduction of systemic
Istituto Oncologico Veneto—IOV, IRCCS,
Padova, Italy therapy, cure rates for osteosarcoma were less
e-mail: [email protected] than 20%, even among patients who presented

216 A. Brunello and V. Zagonel
with no detectable metastatic disease. Since that Based on these observations, several trials
time, evidence has mounted on the presence of have investigated the possibility of improving
micrometastatic disease at the time of diagnosis outcomes for patients with tumors exhibiting low
in most patients, providing the background for necrosis following initial chemotherapy. In the
the role of systemic therapy in achieving cure. European and American Osteosarcoma Study
The introduction of combination chemotherapy Group (EURAMOS) trial, patients received neo-
improved the probability for cure to rates as high adjuvant therapy with the MAP regimen for
as 70% [5]. Cytotoxic drugs, which have been 10 weeks and were randomized after definitive
shown to be active in osteosarcoma, are surgery to different adjuvant treatments accord-
Doxorubicin, Ifosfamide, Methotrexate, and ing to necrosis at definitive surgery. Patients with
Cisplatin [6], yielding objective response rates, good histologic response were randomized to
when used as single agents, of 43%, 33%, 32%, receive either continuation of MAP or MAP with
and 26%, respectively. the addition of interferon-alfa, whereas patients
Conventionally, surgery of the primary tumor with poor histologic response were randomized
is performed after preoperative chemotherapy, an to receive either continuation of MAP or MAP
approach pioneered in the 1970s by Rosen and with the addition of high-dose ifosfamide and
colleagues [7]. Although no survival advantage etoposide [11]. The addition of interferon to
of neoadjuvant compared to post-operative che- MAP did not improve the risk of events for
motherapy has been demonstrated, and surgery is patients with higher necrosis [10]. The addition
rarely influenced, obtaining symptom improve- of high-dose ifosfamide and etoposide to MAP
ment, time available for surgical planning, and did not improve the risk of events for patients
availability of the primary tumor for assessment with less necrosis [12].
of histologic response to preoperative chemo- Since the first pivotal studies on neoadjuvant
therapy have made this approach a standard of chemotherapy, very little progress has been made
care for the majority of patients. In specific cases, with regard to further improvement of survival
such as when diagnostic uncertainty exists or rates obtained with the MAP regimen.
when demolitive surgery cannot be avoided, then The Intergroup Study 0133 studied the role of
immediate resection could be suggested. liposomal muramyl tripeptide (MTP), a deriva-
When osteosarcoma is treated in the neoadju- tive of the Bacillus Calmette-Guérin cell wall,
vant setting (that is, before surgery), necrosis in which stimulates macrophages’ response against
the primary tumor can be assessed at the time of tumor cells, in an adjuvant randomized trial,
definitive surgical resection. Chemotherapy- which showed that the addition of MTP to stan-
induced necrosis strongly correlates with event- dard chemotherapy provided a trend toward
free survival (EFS) and overall survival (OS) [8], improved EFS and a statistically significant
with high necrosis in the primary tumor correlat- improvement in overall survival [9].
ing with lower probability of recurrence and In light of the results of this trial, MTP has
death. Radiological response likely underesti- been approved for use in combination with che-
mates real antineoplastic activity, as osteosarcoma motherapy in the treatment of localized osteosar-
may present with substantial necrosis following coma for patients age 2–30 years in several
chemotherapy without change in dimensions due Countries.
to the osteoid matrix produced by the tumor. Pelvic primary site is associated with a more
Chemotherapy regimens based on a combina- dismal prognosis compared to patients with pri-
tion of high-dose methotrexate, doxorubicin, and mary osteosarcoma of the extremities, with a
cisplatin (the so-called “MAP” regimen) have 5-year survival of about 30% compared to 5-year
been shown to provide 3-year EFS roughly survival for localized extremity osteosarcoma of
around 70% [9, 10], with differences in outcome approximately 70%, [8, 13]. The poor survival of
strongly correlated with the histological response patients with pelvic osteosarcoma is multifacto-
in terms of necrosis. rial, including large tumor volume at presentation,
19 Chemotherapy for Pelvic Bone Tumors 217
more problematic surgery due to higher morbidity, prognosis, with higher necrosis associated with
difficulty in achieving adequate surgical margins,better outcomes.
higher likelihood of metastatic disease at presenta- The Intergroup trial INT-0091conducted by the
tion, inferior necrosis after preoperative chemo- Pediatric Oncology Group and the Children’s
therapy, and presence of macroscopic tumor Cancer Group demonstrated that the addition of
emboli in the large regional vessels [14–19]. ifosfamide and etoposide to cyclophosphamide,
With recurrent osteosarcoma, chemotherapy doxorubicin, and vincristine significantly improved
options are quite limited. The combination of outcomes for patients with localized Ewing sar-
ifosfamide and etoposide appears to be one of the coma [28]. Among patients with localized disease,
most active regimens for these patients [20, 21]. patients randomized to cyclophosphamide, doxo-
The combination of gemcitabine and docetaxel rubicin, and vincristine alternating with ifosfamide
has also been investigated for patients with recur-
and etoposide had a statistically significant better
rent osteosarcoma [22] with findings that indicate5-year EFS compared to patients randomized to
only modest activity. cyclophosphamide, doxorubicin, and vincristine
Newer agents are being studied. Currently, foralone (69% and 54%, respectively). The addition of
relapsed osteosarcoma, there is evidence of activ-ifosfamide to etoposide did not improve outcomes
ity of antiangiogenetic agents, and data from phase
for patients with metastatic disease at initial pre-
II trials with sorafenib, also in combination withsentation, though [31].
everolimus, and regorafenib have been recently In contrast to other bone and soft tissue
published [23–26], with median progression-free tumors, Ewing sarcoma outcomes seem to
survival times of about 4 months. improve with dose intensification.
A single-center study from the Memorial
Sloan Kettering Cancer Center reported a high
19.3 Ewing Sarcoma rate of EFS with the use of very high-dose
alkylating agent therapy given over a shorter
Before the introduction of systemic chemother- duration of just 21 weeks [32]. In a Children’s
apy, Ewing sarcoma had a cure rate of less than Oncology Group trial, dose intensification was
10%, even among patients who presented with obtained by shortening the interval between
localized disease [27]. The outcomes for patients chemotherapy cycles [30], with the administra-
with Ewing sarcoma have dramatically improved tion of the usual five-drug combination every
since then, with as many as 70% of patients pre- 2 weeks for 28 weeks, achieving better out-
senting with localized disease achieving long- comes compared to the five-drug combination
term event-free survival (EFS) with the use of administered every 3 weeks for 42 weeks.
multiagent chemotherapy [28–30]. Patients randomized to the interval-compressed
Survival rates fall down to 30–50% if patients arm had a significantly greater 5-year EFS
present with lung metastases and to less than (73% versus 65% for patients randomized to
20% when patients present with metastases to the standard arm).
distant bones or bone marrow [28]. In the Euro-E.W.I.N.G.99 trial, patients with
Being generally highly sensitive to chemo- localized Ewing sarcoma were randomized
therapy, Ewing sarcoma treatment involves pri- between high-dose chemotherapy with busulfan
mary chemotherapy before local treatment. and melphalan or standard chemotherapy (vin-
Cytotoxics with documented activity, which cristine, dactinomycin, and ifosfamide, seven
are widely used in treatment protocols for courses) if they were at high risk for relapse
Ewing Sarcoma, are Doxorubicin, Ifosfamide, (either poor histologic response after receiving
Cyclophosphamide, Etoposide, Vincristine, and six courses of chemotherapy with vincristine,
D-Actinomycin. ifosfamide, doxorubicin, and etoposide or with
Similar to osteosarcoma, the percentage of tumor volume at diagnosis ≥200 mL if unre-
necrosis postchemotherapy is associated with sected, or initially resected, or resected after
radiotherapy) [33]. In an intent-to-treat analysis, citabine and docetaxel than for each other arm,
the 3- and 8-year EFS was 69% and 60.7% with all comparisons favored the other arms, and the
high-dose chemotherapy versus 56.7% and gemcitabine and docetaxel arm was dropped,
47.1% with standard chemotherapy, respectively. with recruitment continuing in the remaining
Overall survival (OS) also favored high-dose arms.
chemotherapy, with 3- and 8-year OS compared Several targeted agents have been studied,
to standard chemotherapy of 78% vs 64.5% and such as mammalian Target Of Rapamicin
72.2% vs 55.6%, respectively. (mTOR) inhibitors [45] or Insulin-like Growth
Patients with recurrent Ewing sarcoma have a Factor-1 inhibitors [46], with no one proving
number of systemic therapy options. Historically, enough activity to warrant approval. Furthermore,
patients used to be retreated with chemotherapy despite understanding the critical role of EWSR1
combinations used as part of initial therapy, with fusion oncogenes in the pathogenesis of Ewing
some responses and durable remissions reported sarcoma, strategies to target EWSR1 fusion
[34]. There is evidence suggesting activity of oncogenes and oncoproteins have so far been dif-
high doses of ifosfamide (15 g/m2) in patients ficult to develop.
with recurrent Ewing sarcoma who were previ- One of the strongest prognostic factors associ-
ously treated with lower doses as part of initial ated with a poor outcome of patients with local-
therapy [35]. ized Ewing sarcoma is primary tumor located in
Currently, patients with recurrent Ewing sar- the pelvis. Half of the cases of Ewing sarcoma
coma are candidates for clinical trials of novel arise in the extremity bones, followed by pelvis,
agents or may be treated with a number of sal- ribs, and vertebrae [47]. The most common sites
vage chemotherapy regimens with documented within the pelvis are the iliac bones followed by
activity in this setting. the pubis [48]. The higher incidence of primary
Campothecin-based regimens are currently pelvic sites for Ewing sarcoma in adults is one of
among the most active available chemotherapy the possible reasons for poorer outcomes in older
regimens for patients with relapsed Ewing. The subjects compared to younger ones. Also, since
combination of topotecan with cyclophospha- metastatic disease is the most important prognos-
mide has shown activity in this population [36– tic factor for Ewing sarcoma, the worse progno-
38]. The combination of irinotecan and sis of tumors localized in the pelvis is at least
temozolomide has also shown activity in patients partly related to a higher percentage of metastatic
with relapsed Ewing sarcoma [39–42]. cases at diagnosis [49–52].
Also, the combination of gemcitabine with The optimal strategy of addressing local ther-
docetaxel has shown some activity in patients apy in pelvic disease with surgery, radiation ther-
with recurrent Ewing sarcoma [43]. apy, or both has been long investigated, yet
The first interim results of the currently ongo- remains highly controversial [53].
ing rEECur trial, assessing these regimens pro- Indeed, the time interval between chemother-
spectively in a randomized manner, have recently apy initiation and definitive local therapy impact
been reported [44]. In this trial, patients with event-free survival, with longer intervals being
refractory or recurrent Ewing sarcoma were ran- associated with unfavorable outcomes [54].
domized to receive either topotecan and cyclo- In pelvic primary localization, the timing of
phosphamide, irinotecan and temolozomide, and local therapy commonly relies on multidisci-
gemcitabine and docetaxel or high-dose ifos- plinary assessment, being dependent on tumor
famide. Patients randomized in the gemcitabine size, specific location, and response to neoadju-
and docetaxel arm had 11.5% response rate, vant chemotherapy. As a result, the decision as to
median progression-free survival of 3 months, when to offer local therapy for large pelvic
and median OS 13.7 months. After assessing the tumors often involves a thorough multidisci-
probabilities that overall response and plinary discussion among medical, radiation, and
progression-free survival were better for gem- orthopedic oncologists. In patients with chemo-
responsive disease, whether the maximal benefit high-grade chondrosarcoma of the extremities
of chemotherapy should be pursued to facilitate [61]. Globally, high-grade chondrosarcoma of
the best possible surgical resection or radiation of pelvic bones is associated with the highest rates
the smallest tumor volume is still matter of of local recurrence and worst prognosis after sur-
debate. gical treatment [62].
On the one hand, continuing chemotherapy to The majority of chondrosarcomas display
maximum response could be helpful to reduce total conventional subtype features, but some other
tumor volume and would avoid the risk of an inter- subtypes have been described, namely, a mesen-
ruption with local therapy, given the fact that recov- chymal and clear cell chondrosarcoma. Moreover,
ery from surgery for axial tumors is often prolonged, conventional chondrosarcoma can ‘dedifferenti-
making it difficult to timely resume chemotherapy. ate’ into a highly aggressive form, which is the
Moreover, the delivery of concomitant radiation dedifferentiated chondrosarcoma, typically
therapy and chemotherapy can be sometimes diffi- yielding a very poor prognosis [63, 64].
cult, especially in older patients. On the other hand, In chondrosarcoma, unlike osteosarcoma and
delaying local therapy may not be beneficial. In Ewing’s sarcoma, chemotherapy and radiation
some cases, radiation therapy in the absence of sur- therapy have not been proven to be effective nei-
gery could provide sufficient local control of Ewing ther as adjuvant treatment nor for distant control,
sarcoma, particularly those arising in the sacrum, and therefore, surgery is the primary treatment.
which displays a better prognosis than Ewing sar- Inoperable, locally advanced and metastatic
coma of the hip bones [55], probably at least in part high-grade chondrosarcomas have a poor prog-
due to smaller tumor volumes in the sacrum than in nosis [65], and this may also be related to the
the hip bones. So far, no randomized controlled tri- inefficacy of chemotherapy in conventional
als have directly compared radiation and surgery. chondrosarcoma.
Despite historical data of two to threefold However, there is some evidence for chemo-
increased local failure and decreased survival in sensitivity of mesenchymal chondrosarcoma,
primary Ewing sarcoma of the pelvis compared thus suggesting a role for adjuvant or neoadju-
to other sites, more recent series report local and vant therapy with Ewing-type regimens [66, 67].
distant control rates for pelvic Ewing’s sarcoma Dedifferentiated chondrosarcoma is com-
that approach those of nonpelvic [53, 56–59] and monly treated as a high-grade bone sarcoma, with
this may be due to optimizing chemotherapy systemic and local therapies that need to be
doses and schedules with improved supportive adapted to patient’s age [65, 68]. Among available
therapy, better radiation techniques, and more drugs, beside anthracyclines, ifosfamide, and cis-
aggressive surgery. platin, some activity of gemcitabine in combina-
tion with docetaxel has been reported [69].
19.4 Chondrosarcoma
19.5 Chordoma
Chondrosarcoma is the most frequent bone sar-
coma of adulthood, and it is primary located in Chordoma is a very rare mesenchymal neoplasm
the extremities and the axial skeleton, with the arising from embryonic remnants of the noto-
pelvis representing the most common location. chord in axial skeleton, with a reported yearly
Typically, chondrosarcomas are low-grade, incidence of approximately 0.08/100,000 people
locally aggressive, nonmetastasizing tumors [70, 71], and it affects predominantly the mobile
(grade I), but in some cases, chondrosarcoma can spine and the sacrum in older adults [72].
be high grade (grades II–III) [60]. Conventional chordoma is a low-grade, locally
Chondrosarcoma of the pelvis is often high grade invasive malignancy. Immunohistochemistry
and is almost twice as likely to present with met- nuclear positivity for Brachyury is the diagnostic
astatic disease at presentation compared with hallmark [73].
Although typically slow-growing, the natural For epithelioid hemangioendothelioma, which

history of chordoma is marked by a high ten- is advanced or metastatic and not amenable to
dency toward local recurrence, with reported local treatment, sirolimus has been reported to
local failure in approximately 40–50% of patients display some activity [81].
undergoing surgery [74].
Dedifferentiated chordomas account for less
than 5% of all cases, with brachyury expression 19.7 Giant Cell Tumors of Bone
that can be lost, and behave more aggressively
than the conventional counterpart. Approximately Giant cell tumors are benign, locally aggressive,
30% of patients with chordoma will develop and rarely metastasizing intramedullary bone
metastases, usually late in the natural history of tumors composed of mononuclear cells and
the disease and mostly after local recurrence. osteoclast-like multinucleated giant cells, with a
The mainstay of treatment for chordoma is variable and unpredictable potential for aggres-
surgery and/or radiation therapy. When a local sive growth [82]. Giant cell tumor of bone can
relapse occurs, it carries very poor survival rates localize in the pelvic bone although this is not
and local control is hardly achieved [70]. very common, with the acetabular area being the
Chemotherapy for advanced chordoma is gen- most frequent site [83].
erally inactive, yet there is some evidence of activ- Giant cell tumor cells typically present with a
ity of tyrosin-chinase inhibitors, such as imatinib high expression of Receptor Activator of Nuclear
and sorafenib, which can provide stabilization of factor Kappa-Β (RANK)/RANK-ligand (RANKL),
disease and nondimensional tumor responses, and and Denosumab, a RANKL inhibitor, has been
studies of epidermal growth factor receptor and successfully used to treat unresectable or meta-
mTOR inhibitors are ongoing [75–77]. static disease [84].
Denosumab may provide an option for treat-
ment of initially locally advanced tumors when
19.6 High-Grade Sarcomas used as neoadjuvant therapy to facilitate complete
of Bone surgical resection or avoid mutilating surgery [85].
High-grade spindle/pleomorphic sarcomas con-

stitute a heterogeneous group of primary malig- References
nant bone tumors that do not fulfill the histological
criteria for a diagnosis of osteosarcoma, chondro- 1. Stiller CA, Trama A, Serraino D, Rossi S, Navarro C,
sarcoma, or Ewing sarcoma [78]. Chirlaque MD, et al. Descriptive epidemiology of sar-
comas in Europe: report from the RARECARE proj-
Vascular sarcomas, namely, epithelioid ect. Eur J Cancer. 2013;49:684–95.
hemangioendothelioma and angiosarcoma, may 2. Amadori D, Aglietta M, Alessi B, Gianni L, Ibrahim T,
occasionally arise in the bone. Angiosarcoma is Farina G, et al. Efficacy and safety of 12-weekly ver-
highly aggressive, with a poor prognosis, whereas sus 4-weekly zoledronic acid for prolonged treatment
of patients with bone metastases from breast cancer
epithelioid hemangioendothelioma has an inter- (ZOOM): a phase 3, open-label, randomised, non-
mediate biological behavior. Vascular tumors of inferiority trial. Lancet Oncol. 2013;14(7):663–70.
bone may present as unifocal or multifocal dis- 3. Yee AJ, Raje NS. Denosumab for the treatment of
ease [79]. bone disease in solid tumors and multiple myeloma.
Future Oncol. 2018;14(3):195–203.
Given the rarity of these forms, the evidence is 4. Blay JY, Soibinet P, Penel N, Bompas E, Duffaud F,
limited regarding treatment and outcomes. For Stoeckle E, et al. Improved survival using special-
patients with angiosarcoma of bone, improve- ized multidisciplinary board in sarcoma patients. Ann
ments in survival rates may be obtained with the Oncol. 2017;28:2852–9.
5. Link MP, Goorin AM, Miser AW, Green AA, Pratt CB,
use of adjuvant chemotherapy based on cisplatin, Belasco JB, et al. The effect of adjuvant chemotherapy
doxorubicin, and ifosfamide, but there are reports on relapse-free survival in patients with osteosarcoma
of the efficacy of taxanes as well [80]. of the extremity. N Engl J Med. 1986;314:1600–6.
6. Anninga JK, Gelderblom H, Fiocco M, Kroep JR, cooperative osteosarcoma study group protocols. Ann
Taminiau AH, Hogendoorn PC, et al. Chemotherapeutic Oncol. 2011;22:1228–35.
adjuvant treatment for osteosarcoma: where do we 19. Grimer RJ, Carter SR, Tillman RM, Spooner D,

stand? Eur J Cancer. 2011;47:2431–45. Mangham DC, Kabukcuoglu Y. Osteosarcoma of the
7. Rosen G, Murphy ML, Huvos AG, Gutierrez M, pelvis. J Bone Jt Surg Br. 1999;81:796–802.
Marcove RC. Chemotherapy, en bloc resection, and 20. Miser JS, Kinsella TJ, Triche TJ, Sokos M, Jarosinski
prosthetic bone replacement in the treatment of osteo- P, Forquer R, et al. Ifosfamide with mesna uro-
genic sarcoma. Cancer. 1976;37:1–11. protection and etoposide: an effective regimen
8. Meyers PA, Heller G, Healey J, Huvos A, Lane J, in the treatment of recurrent sarcomas and other
Marcove R, et al. Chemotherapy for nonmetastatic tumors of children and young adults. J Clin Oncol.
osteogenic sarcoma: the memorial Sloan-Kettering 1987;5:1191–8.
experience. J Clin Oncol. 1992;10:5–15. 21. Gentet JC, Brunat-Mentigny M, Demaille MC, Pein
9. Meyers PA, Schwartz CL, Krailo MD, Healey JH, F, Avet-Loiseau H, Berger C, et al. Ifosfamide and
Bernstein ML, Betcher D, et al. Osteosarcoma: the etoposide in childhood osteosarcoma. A phase II
addition of muramyl tripeptide to chemotherapy study of the French Society of Paediatric Oncology.
improves overall survival-a report from the Children's Eur J Cancer. 1997;33:232–7.
oncology group. J Clin Oncol. 2008;26:633–8. 22. Lee HY, Shin SJ, Kim HS, Hong SJ, Han JW, Lim ST,
10. Bielack SS, Smeland S, Whelan JS, Marina N, Jovic et al. Weekly gemcitabine and docetaxel in refractory
G, Hook JM, et al. Methotrexate, doxorubicin, and soft tissue sarcoma: a retrospective analysis. Cancer
cisplatin (MAP) plus maintenance pegylated inter- Treat Res. 2012;44:43–9.
feron alfa-2b versus MAP alone in patients with 23. Grignani G, Palmerini E, Dileo P, Asaftei SD,

resectable high-grade osteosarcoma and good histo- D’Ambrosio L, Pignochino Y, et al. A phase II trial
logic response to preoperative MAP: first results of of sorafenib in relapsed and unresectable high-grade
the EURAMOS-1 good response randomized con- ostesarcoma after failure of standard multimodal
trolled trial. J Clin Oncol. 2015;33:2279–87. therapy: an Italian sarcoma group study. Ann Oncol.
11. Whelan JS, Bielack SS, Marina N, Smeland S,
2012;23:508–16.
Jovic G, Hook JM, et al. EURAMOS-1, an interna- 24. Grignani G, Palmerini E, Ferraresi V, D’Ambrosio L,
tional randomised study for osteosarcoma: results Bertulli R, Asaftei SD, et al. Sorafenib and everoli-
from pre- randomisation treatment. Ann Oncol. mus for patients with unresectable high-grade osteo-
2015;26:407–14. sarcoma progressing after standard treatment: a
12. Marina NM, Smeland S, Bielack SS, Bernstein M, non-randomised phase 2 clinical trial. Lancet Oncol.
Jovic G, Krailo MD, et al. Comparison of MAPIE 2015;16(1):98–107.
versus MAP in patients with a poor response to preop- 25. Davis LE, Bolejack V, Ryan CW, Ganjoo KN, Loggers
erative chemotherapy for newly diagnosed high-grade ET, Chawla S, et al. Randomized double-blind phase
osteosarcoma (EURAMOS-1): an open-label, inter- II study of Regorafenib in patients with metastatic
national, randomised controlled trial. Lancet Oncol. osteosarcoma. J Clin Oncol. 2019;37(16):1424–31.
2016;17(10):1396–408. 26. Duffaud F, Mir O, Boudou-Rouquette P, Piperno-

13. Donati D, Giacomini S, Gozzi E, Ferrari S, Sangiorgi Neumann S, Penel N, Bompas E, et al. Efficacy and
L, Tienghi A, et al. Osteosarcoma of the pelvis. Eur J safety of regorafenib in adult patients with meta-
Surg Oncol. 2004;30:332–40. static osteosarcoma: a non-comparative, randomised,
14. Ozaki T, Flege S, Kevric M, Lindner N, Maas R, double- blind, placebo-controlled, phase 2 study.
Delling G, et al. Osteosarcoma of the pelvis: experi- Lancet Oncol. 2019;20(1):120–33.
ence of the cooperative osteosarcoma study group. J 27. Nesbit M. Ewing's sarcoma. CA Cancer J Clin.

Clin Oncol. 2003;21:334–41. 1976;26(3):174–80.
15. Wurtz LD, Peabody TD, Simon MA. Delay in the 28. Grier HE, Krailo MD, Tarbell NJ, Link MP, Fryer CJ,
diagnosis and treatment of primary bone sarcoma of Pritchard DJ, et al. Addition of ifosfamide and etopo-
the pelvis. J Bone Joint Surg Am. 1999;81:317–25. side to standard chemotherapy for Ewing's sarcoma
16. Kawai A, Healey JH, Boland P, Lin PP, Huvos AG, and primitive neuroectodermal tumor of bone. N Engl
Meyers PA. Prognostic factors for patients with sarco- J Med. 2003;348:694–701.
mas of the pelvic bones. Cancer. 1998;82:851–9. 29. Granowetter L, Womer R, Devidas M, Krailo M,

17. Ferrari S, Bielack SS, Smeland S, Longhi A, Egerer Wang C, Bernstein M, et al. Dose-intensified com-
G, Hall KS, et al. EURO-B.O.S.S.: a European study pared with standard chemotherapy for nonmetastatic
on chemotherapy in bone-sarcoma patients aged over Ewing sarcoma family of tumors: a Children's oncol-
40: outcome in primary high-grade osteosarcoma. ogy group study. J Clin Oncol. 2009;27:2536–41.
Tumori J. 2018;104(1):30–6. 30. Womer RB, West DC, Krailo MD, Dickman PS,

18. Andreou D, Bielack SS, Carrle D, Kevric M, Kotz Pawel BR, Grier HE, et al. Randomized controlled
R, Winkelmann W, et al. The influence of tumor- and trial of interval-compressed chemotherapy for the
treatment-related factors on the development of local treatment of localized Ewing sarcoma: a report
recurrence in osteosarcoma after adequate surgery. from the Children's oncology group. J Clin Oncol.
An analysis of 1355 patients treated on neoadjuvant 2012;30:4148–54.
31. Miser JS, Krailo MD, Tarbell NJ, Link MP, Fryer CJ, can, and temozolomide in patients with relapsed and
Pritchard DJ, et al. Treatment of metastatic Ewing’s refractory Ewing sarcoma. Pediatr Blood Cancer.
sarcoma or primitive neuroectodermal tumor of bone: 2013;60:1621–5.
evaluation of combination ifosfamide and etoposide– 43. Mora J, Castañeda A, Perez-Jaume S, Lopez-Pousa
a Children’s Cancer group and pediatric oncology A, Maradiegue E, Valverde C, et al. GEIS-21: a mul-
group study. J Clin Oncol. 2004;22:2873–6. ticentric phase II study of intensive chemotherapy
32. Kolb EA, Kushner BH, Gorlick R, Laverdiere C,
including gemcitabine and docetaxel for the treatment
Healey JH, LaQuaglia MP, et al. Long-term event- of Ewing sarcoma of children and adults: a report
free survival after intensive chemotherapy for Ewing's from the Spanish sarcoma group (GEIS). Br J Cancer.
family of tumors in children and young adults. J Clin 2017;117(6):767–74.
Oncol. 2003;21:3423–30. 44. McCabe MG, Moroz V, Khan M, Dirksen U, Evans
33. Whelan J, Le Deley MC, Dirksen U, Le Teuff G, A, Fenwick N, et al. Results of the first interim
Brennan B, Gaspar N, et al. High-dose chemotherapy assessment of rEECur, an international randomized
and blood autologous stem-cell rescue compared controlled trial of chemotherapy for the treatment of
with standard chemotherapy in localized high-risk recurrent and primary refractory Ewing sarcoma. J
ewing sarcoma: results of Euro-E.W.I.N.G.99 and Clin Oncol. 2019;37. (suppl; abstr 11007).
Ewing- 2008. J Clin Oncol. 2018:JCO2018782516. 45. Wagner LM, Fouladi M, Ahmed A, Krailo MD,

https://doi.org/10.1200/JCO.2018.78.2516. [Epub Weigel B, DuBois SG, et al. Phase II study of cixutu-
ahead of print]. mumab in combination with temsirolimus in pediatric
34. Hayes FA, Thompson EI, Kumar M, Hustu HO. Long- patients and young adults with recurrent or refractory
term survival in patients with Ewing’s sarcoma relaps- sarcoma: a report from the Children's oncology group.
ing after completing treatment. Med Pediatr Oncol. Pediatr Blood Cancer. 2015;62(3):440–4.
1987;15:254–6. 46. van de Luijtgaarden AC, Versleijen-Jonkers YM,

35. Ferrari S, del Prever AB, Palmerini E, Staals E, Berta Roeffen MH, Schreuder HW, Flucke UE, van der
M, Balladelli A, et al. Response to high-dose ifos- Graaf WT. Prognostic and therapeutic relevance of
famide in patients with advanced/recurrent Ewing the IGF pathway in Ewing's sarcoma patients. Target
sarcoma. Pediatr Blood Cancer. 2009;52:581–4. Oncol. 2013;8(4):253–60.
36. Saylors RL 3rd, Stine KC, Sullivan J, Kepner JL, 47. Valery PC, Laversanne M, Bray F. Bone cancer inci-
Wall DA, Bernstein ML, et al. Cyclophosphamide dence by morphological subtype: a global assessment.
plus topotecan in children with recurrent or refrac- Cancer Causes Control. 2015;26:1127–39.
tory solid tumors: a pediatric oncology group phase II 48. Evans RG, Nesbit ME, Gehan EA, Garnsey LA,

study. J Clin Oncol. 2001;19:3463–9. Burgert O Jr, Vietti TJ, et al. Multimodal therapy
37. Hunold A, Weddeling N, Paulussen M, Ranft A,
for the management of localized Ewing’s sarcoma
Liebscher C, Jürgens H. Topotecan and cyclophos- of pelvic and sacral bones: a report from the second
phamide in patients with refractory or relapsed Ewing intergroup study. J Clin Oncol. 1991;9:1173–80.
tumors. Pediatr Blood Cancer. 2006;47:795–800. 49. Cotterill SJ, Ahrens S, Paulussen M, Jürgens HF,

38. Bernstein ML, Devidas M, Lafreniere D, Souid AK, Voûte PA, Gadner H, Craft AW. Prognostic factors in
Meyers PA, Gebhardt M, et al. Intensive therapy with Ewing’s tumor of bone: analysis of 975 patients from
growth factor support for patients with Ewing tumor the European intergroup cooperative Ewing’s sarcoma
metastatic at diagnosis: Pediatric Oncology Group/ study group. J Clin Oncol. 2000;18(17):3108–14.
Children’s Cancer Group phase II study 9457–A 50. Burgers JM, Oldenburger F, de Kraker J, van

report from the Children’s Oncology Group. J Clin Bunningen BN, van der Eijken JW, Delemarre JF,
Oncol. 2006;24:152–9. et al. Ewing’s sarcoma of the pelvis: changes over
39. Wagner LM, Crews KR, Iacono LC, Houghton PJ, 25 years in treatment and results. Eur J Cancer.
Fuller CE, McCarville MB, et al. Phase I trial of 1997;33(14):2360–7.
temozolomide and protracted irinotecan in pediatric 51. Hoffmann C, Ahrens S, Dunst J, Hillmann A,

patients with refractory solid tumors. Clin Cancer Winkelmann W, Craft A, et al. Pelvic Ewing sar-
Res. 2004;10:840–8. coma: a retrospective analysis of 241 cases. Cancer.
40. Wagner LM, McAllister N, Goldsby RE, , Rausen 1999;85(4):869–77.
AR, McNall-Knapp RY, McCarville MB, Albritton 52. Evans R, Nesbit M, Askin F, Burgert O, Cangir A,
K. Temozolomide and intravenous irinotecan for Foulkes M, et al. Local recurrence, rate and sites of
treatment of advanced Ewing sarcoma. Pediatr Blood metastases, and time to relapse as a function of treat-
Cancer 2007;48:132–139. ment regimen, size of primary and surgical history in
41. Casey DA, Wexler LH, Merchant MS, Chou AJ,
62 patients presenting with non-metastatic Ewing’s
Merola PR, Price AP, Meyers PA. Irinotecan and sarcoma of the pelvic bones. Int J Radiat Oncol Biol
temozolomide for Ewing sarcoma: the memorial Phys. 1985;11:129–36.
Sloan-Kettering experience. Pediatr Blood Cancer. 53. Yock TI, Krailo M, Fryer CJ, Donaldson SS, Miser JS,
2009;53:1029–34. Chen Z, et al. Local control in pelvic Ewing sarcoma:
42. Raciborska A, Bilska K, Drabko K, Chaber R,
analysis from INT-0091—a report from the Children’s
Pogorzala M, Wyrobek E, et al. Vincristine, irinote- oncology group. J Clin Oncol. 2006;24:3838–43.
54. Gupta AA, Pappo A, Saunders N, Hopyan S, Ferguson A European Musculoskeletal Oncology Society study.
P, Wunder J, et al. Clinical outcome of children and Eur J Cancer. 2015;51:374–81.
adults with localized Ewing sarcoma: impact of che- 68. Grimer RJ, Gosheger G, Taminiau A, Biau D,

motherapy dose and timing of local therapy. Cancer. Matejovsky Z, Kollender Y, et al. Dedifferentiated
2010;116(13):3189–94. chondrosarcoma: prognostic factors and outcome from
55. Hesla AC, Tsagozis P, Jebsen N, Zaikova O, Bauer H, a European group. Eur J Cancer. 2007;43:2060–5.
Brosjö O. Improved prognosis for patients with Ewing 69. Fox E, Patel S, Wathen JK, Schuetze S, Chawla S,
sarcoma in the sacrum compared with the innominate Harmon D, et al. Phase II study of sequential gem-
bones: the Scandinavian sarcoma group experience. J citabine followed by docetaxel for recurrent Ewing
Bone Joint Surg Am. 2016;98(3):199–210. sarcoma, osteosarcoma, or unresectable or locally
56. La TH, Meyers PA, Wexler LH, Alektiar KM,
recurrent chondrosarcoma: results of sarcoma alli-
Healey JH, Laquaglia MP, et al. Radiation therapy ance for research through collaboration study 003.
for Ewing’s sarcoma: results from memorial Sloan- Oncologist. 2012;17:321.
Kettering in the modern era. Int J Radiat Oncol Biol 70. Smoll NR, Gautschi OP, Radovanovic I, Schaller

Phys. 2006;64:544–50. K, Weber DC. Incidence and relative survival of
57. Kolb EA, Kushner BH, Gorlick R, Laverdiere C,
chordomas: the standardized mortality ratio and
Healey JH, LaQuaglia MP, et al. Long-term event-free the impact of chordomas on a population. Cancer.
survival after intensive chemotherapy for Ewing’s 2013;119(11):2029–37.
family of tumors in children and young adults. J Clin 71. Gatta G, Capocaccia R, Botta L, Mallone S, De

Oncol. 2003;21:3423–30. Angelis R, Ardanaz E, et al. Burden and central-
58. Rodriguez-Galindo C, Liu T, Krasin MJ, Wu J,
ised treatment in Europe of rare tumours: results of
Billups CA, Daw NC, et al. Analysis of prognostic RARECAREnet-a population-based study. Lancet
factors in Ewing sarcoma family of tumors: review of Oncol. 2017;18(8):1022–39.
St. Jude Children’s Research Hospital studies. Cancer. 72. Lau CS, Mahendraraj K, Ward A, Chamberlain

2007;110:375–84. RS. Pediatric chordomas: a population-based clinical
59. Granowetter L, Womer R, Devidas M, Krailo M,
outcome study involving 86 patients from the surveil-
Wang C, Bernstein M, et al. Dose-intensified com- lance, epidemiology, and end result (SEER) database
pared with standard chemotherapy for nonmetastatic (1973-2011). Pediatr Neurosurg. 2016;51(3):127–36.
Ewing sarcoma family of tumors: a Children’s oncol- 73. Vujovic S, Henderson S, Presneau N, Odell E, Jacques
ogy group study. J Clin Oncol. 2009;27:2536–41. TS, Tirabosco R, et al. Brachyury, a crucial regulator
60.
Fletcher CDM, Bridge JA, Hogendoorn PCW, of notochordal development, is a novel biomarker for
Mertens F. Pathology and genetics of tumours of soft chordomas. J Pathol. 2006;209:157–65.
tissue and bone. World Health Organization. Lyon: 74. Stacchiotti S, Gronchi A, Fossati P, Akiyama T,

IARC Press; 2013. Alapetite C, Baumann M, et al. Best practices for the
61. Lee FY, Mankin HJ, Fondren G, Gebhardt MC,
management of local-regional recurrent chordoma:
Springfield DS, Rosenberg AE, Jennings LC. a position paper by the chordoma Global Consensus
Chondrosarcoma of bone: an assessment of outcome. Group. Ann Oncol. 2017;28(6):1230–42.
J Bone Joint Surg Am. 1999;81:326–38. 75. Stacchiotti S, Longhi A, Ferraresi V, Grignani G,

62.
Pring ME, Weber KL, Unni K, Sim FH. Comandone A, Stupp R, et al. Phase II study of
Chondrosarcoma of the pelvis: a review of sixty-four imatinib in advanced chordoma. J Clin Oncol.
cases. J Bone Jt Surg. 2001;83A:1630–42. 2012;30:914–20.
63. Whelan JS, Davis LE. Osteosarcoma, chondrosar-
76. Bompas E, Le Cesne A, Tresch-Bruneel E, Lebellec
coma, and chordoma. J Clin Oncol. 2018;36:188–93. L, Laurence V, Collard O, et al. Sorafenib in patients
64.
Gelderblom H, Hogendoorn PC, Dijkstra SD, with locally advanced and metastatic chordomas: a
van Rijswijk CS, Krol AD, Taminiau AH, Bovée phase II trial of the French Sarcoma Group (GSF/
JV. The clinical approach towards chondrosarcoma. GETO). Ann Oncol. 2015;26:2168–73.
Oncologist. 2008;13:320–9. 77. Lebellec L, Chauffert B, Blay JY, Le Cesne A,

65. Italiano A, Mir O, Cioffi A, Palmerini E, Piperno- Chevreau C, Bompas E, et al. Advanced chordoma
Neumann S, Perrin C, et al. Advanced chondrosarco- treated by first-line molecular targeted therapies: out-
mas: role of chemotherapy and survival. Ann Oncol. comes and prognostic factors. A retrospective study
2013;24:2916–22. of the French Sarcoma Group (GSF/ GETO) and the
66. Dantonello TM, Int-Veen C, Leuschner I, Schuck
Association des Neuro-Oncologues d’Expression
A, Furtwaengler R, Claviez A, et al. Mesenchymal Francaise (ANOCEF). Eur J Cancer. 2017;79:119–28.
chondrosarcoma of soft tissues and bone in children, 78. Pakos EE, Grimer RJ, Peake D, Spooner D, Carter
adolescents, and young adults: experiences of the SR, Tillman RM, et al. The ‘other’ bone sarcomas:
CWS and COSS study groups. Cancer. 2008;112: prognostic factors and outcomes of spindle cell
2424–31. sarcomas of bone. J Bone Joint Surg Br. 2011;93:
67. Frezza AM, Cesari M, Baumhoer D, Biau D, Bielack 1271–8.
S, Campanacci DA, et al. Mesenchymal chondrosar- 79. Verbeke SL, Bertoni F, Bacchini P, Sciot R, Fletcher
coma: prognostic factors and outcome in 113 patients. CD, Kroon HM, et al. Distinct histological fea-
tures characterize primary angiosarcoma of bone. giant cell tumor of bone. J Bone Joint Surg Am.
Histopathology. 2011;58(2):254–64. 2014;96(23):1999–2007.
80. Palmerini E, Maki RG, Staals EL, Alberghini M,
83. Zheng K, Wang Z, Wu SJ, Ye ZM, Xu SF, Xu M, et al.
Antonescu CR, Ferrari C, et al. Primary angiosarcoma Giant cell tumor of the pelvis: a systematic review.
of bone: a retrospective analysis of 60 patients from 2 Orthop Surg. 2015;7(2):102–7.
institutions. Am J Clin Oncol. 2014;37(6):528–34. 84. Thomas D, Henshaw R, Skubitz K, Chawla S, Staddon
81. Stacchiotti S, Provenzano S, Dagrada G, Negri T, A, Blay JY, et al. Denosumab in patients with giant-
Brich S, Basso U, et al. Sirolimus in advanced epi- cell tumour of bone: an open-label, phase 2 study.
thelioid hemangioendothelioma: a retrospective case- Lancet Oncol. 2010;11(3):275–80.
series analysis from the Italian Rare Cancer Network 85. Rutkowski P, Gaston L, Borkowska A, Stacchiotti S,
Database. Ann Surg Oncol. 2016;23(9):2735–44. Gelderblom H, Baldi GG, et al. Denosumab treatment
82. Liede A, Bach BA, Stryker S, Hernandez RK,
of inoperable or locally advanced giant cell tumor of
Sobocki P, Bennett B, Wong SS. Regional varia- bone—multicenter analysis outside clinical trial. Eur
tion and challenges in estimating the incidence of J Surg Oncol. 2018;44(9):1384–90.
Role of Plastic Surgery
in the Treatment of Pelvic Tumors 20
Cesare Tiengo, Andrea Monticelli, Regina Sonda,
Andrea Angelini, Pietro Ruggieri,
and Franco Bassetto
Primary pelvic tumors are rare, and the principal structive surgery. This scenario, in fact, is charged
goal of the surgical treatment is to obtain a free by a high rate of complications and dissatisfying
margin resection. Although the majority of mechanical and nonmechanical results: infec-
patients can be treated with limb-salvaging inter- tions and wound dehiscences are common, espe-
nal hemipelvectomies [1, 2], hindquarter amputa- cially when associated with radiotherapy.
tions and hip disarticulation are still performed. In Nevertheless the poor cosmetic, physical and
order to extirpate completely the tumor, the recon- psychological outcomes, and soft-tissue recon-
structive surgeon may be faced to large soft-tissue structions can improve the quality of life of the
defects. A concomitant one-step autologous patients. Over years, these techniques allowed
reconstruction, mostly using myocutaneous or even more aggressive resections.
fasciocutaneous flaps, has been proved to lead to The purpose of this chapter is to provide an
better results, thanks to the well-vascularized tis- overview of the most used reconstructive tech-
sues, which are used to cover defects, grafts, and niques following bone tumor resection in the pel-
implants or just to obliterate the dead space. vic district. The hardest part to create a
Moreover, these flaps can be hard to be performed reconstructive algorithm is that each patient
in a second time, especially when radiotherapy or needs a tailored solution, due to the high com-
infections compromise the donor site. plexity and variability of these cases.
The reconstruction of periacetabular and
sacral defects, after pelvic tumor resection, can
be classified among the most challenging proce- 20.1 Flap Planning
dures in orthopedic oncology and plastic recon-
The reconstructive surgeon must be involved in
the preoperative multidisciplinary meetings in
C. Tiengo (*) · A. Monticelli
order to plan the possible solutions. Even though,
R. Sonda · F. Bassetto
Plastic and Reconstructive Surgery Unit, in any case, no risks must be taken not to com-
University of Padova, Padova, Italy pletely extirpate the tumor, the multidisciplinary
e-mail: [email protected]; team approach can better plan collateral proce-
[email protected]
dures, such as colostomy or ileostomy, leaving
A. Angelini · P. Ruggieri open many reconstructive techniques. Imaging
exams must be acquired and collectively dis-
Oncology, University of Padova, Padova, Italy
e-mail: [email protected]; cussed. The presence of vascular pedicles must
[email protected] be detailed, usually in our practice with an

226 C. Tiengo et al.
Angio-CT, and the operative plan should be sheath passing anterior to the arcuate line.
agreed and proposed to the patient. For sacral Normally, it divides into two or three major ves-
resection, we usually prefer to place ureteral sels that run cranially into the muscle belly, origi-
stents that can help their identification during the nating several muscular or musculocutaneous
tumor resection. Patient positioning depends perforators. These vessels, which are usually big-
mostly on the location of the tumor. We advise to ger when originate from the medial raw, supply
mark the skin in the preoperative setting. Usually, the skin paddle, which must be placed over them.
we start with the patient in the supine position for Among our institute, an angio-TC, if not yet per-
periacetabular tumor and in the prone position for formed, is usually requested to investigate the
the tumor of the sacrum, except for those involv- abdominal wall vessels, because several surgical
ing S2 or higher levels where we prefer to per- procedures can have damaged deep inferior epi-
form an anterior abdominal exploration and a gastric vessels, such as colostomy, appendec-
posterior resection. The eventual changing of the tomy, etc. The main venous drainage is granted
position of the patient can be needed for the flap by the deep inferior epigastric vein, which is con-
dissection. nected to the external iliac vein.
In our experience, flaps using tissue from rectus The VRAM is one of the principal solutions
abdominis muscle or from the gluteal area allow for both periacetabular (Fig. 20.1) and sacral
the best coverage for sacral resections, providing a (Fig. 20.2) reconstructions. The orientation of the
moderate quantity of tissue to fill the dead space. skin paddle is decided based mainly on the exten-
Other solutions might be the omental flap or the sion and on the location of the skin defect. In the
combination of the previous techniques. case of periacetabular resections or sacral low
resections, when the patient can be placed in the
supine position for all the time, a simultaneous
20.2 Rectus Abdominis flap elevation can be performed by secondary
Musculocutaneous Flap equipment [3]. If a change of position is manda-
tory, the flap elevation can be performed prior to
The rectus Abdominis Musculocutaneous flap is the abdominal exploration, if needed, or after the
type III according to the Mathes and Nahai clas- posterior resection, if the entire demolitive time
sification. It can be used as a muscular or as a can be completed in the prone position. Usually,
musculocutaneous flap, allowing several types of the rectus sheath is approached medially, far from
reconstructions. It can be named differently the skin paddle in order not to damage skin perfo-
based on the orientation of its skin paddle: trans- rators. The muscle can be easily separated from
verse rectus abdominis musculocutaneous the posterior fascia, but care must be taken to
(TRAM) flap or vertical rectus abdominis mus- handle the pedicle, which usually lies on the pos-
culocutaneous (VRAM) flap. It is supported by terior surface of the muscle belly. If flap elevation
two main vessels: the deep inferior epigastric is performed prior to turning the patient, the
artery (DIEA), which is the principal one, and the authors advise to leave the pubis insertion
superior epigastric artery. The rectus abdominis attached, in order to avoid pedicle twisting and to
muscle measures about 30 x 10 cm and can be put the flap in a plastic bag and leaving it where it
used as pedicled or microsurgical flap. The prin- can be clearly retrieved and managed from the
cipal pedicle, the DIEA, measures 14–18 cm in posterior approach.
length and has a caliber of about 3–4 mm. It In the case of large defects of the fascia, it can
arises from the external femoral artery and trav- be reconstructed placing a synthetic mesh or an
els superomedially in the extraperitoneal tissue acellular dermal matrix. These solutions can be
piercing the transversalis fascia. After leaving used to repair anterior defects of the donor site,
several and constant branches, it enters the rectus posterior defects, or both. These repairs should
20 Role of Plastic Surgery in the Treatment of Pelvic Tumors 227
Fig. 20.1 Delayed wound healing in a patient previously custom-made prosthesis and soft tissue reconstruction
treated for aseptic loosening of acetabular component with a pedicled VRAM flap
after sarcoma resection, undergone a revision with
be tension-free. Even though both alternatives neath muscle. The myocutaneous technique
have been described, many Authors still prefer should be considered only in paraplegic patients;
the synthetic options, especially to avoid long- otherwise, a severe dysfunctional deambulation is
term hernias or bulging, which are costly and caused. In this case, the flap is considered a type
hard to treat. Care must be taken leaving a gap III Mathes–Nahai classification. Usually, they are
posterior to make the pedicle passing through. It proposed for partial or total sacrectomy
is very important that the mesh doesn’t decubi- (Fig. 20.3). Even though it has been described, the
tate on the vessels and its course is regular and microsurgical employments are rare. These flaps
without twistings or kinkings. Finally, the inset- are irrorated by several perforator vessels coming
ting of the flap must be tension-free to avoid from the superior and the inferior gluteal arteries.
necrosis, flap loss, or wound cracking during the The superior gluteal artery is the largest branch of
change of position. the internal iliac artery, which is the posterior
Advancement SGAP (superior gluteal artery division. It originates above from the upper bor-
perforator) or IGAP (inferior gluteal artery perfo- der of the piriformis muscle, dividing soon into a
rator) Flaps. superficial and a deep branch. The deep one runs
The advancement SGAP flaps are fasciocuta- vascularizating the gluteus medius and the iliac
neous flaps that can eventually include the under- bone, while the superficial one the upper portion
Fig. 20.2 Sacrectomy for chordoma resection, undergone colostomy surgery and repaired with a pedicled VRAM flap
of the anatomic area. From this last branch, our than those originated from the superior gluteal
perforators of interest usually originate. artery. The mean length of the completed dis-
The inferior gluteal artery (IGA) represents sected pedicles is 5–7 cm for the superior gluteal
the terminal branch of the anterior division of the flap and 7–10 cm for the inferior gluteal flap, but
internal iliac artery. This artery travels with the normally a minimal subfascial dissection is
greater sciatic nerve through the greater sciatic required, avoiding risks. When longer move-
foramen, and then, it pierces the sacral fascia ments are necessaries, we prefer to use other
toward the surface. Normally, under the inferior techniques, usually the VRAM flap, also because
portion of the gluteus maximus, perforators are they’re most likely associated with the need of
given off to supply the overlying fat and skin filling the dead space. The venous drainage
(Fig. 20.3). These last vessels run more oblique accompanies the arterial system.
Fig. 20.3 Decubitus sacral and ischial ulcer in unresectable chondrosarcoma, repaired with advancement fasciocutane-
ous SGAP flap for sacral region and IGAP flap for the ischial region
20.3 Anterolateral Thigh (ALT) Flap on the defect size. Although a larger width than
8 cm may not be closed primarily, greater dimen-
The anterolateral thigh (ALT) flap is a well- sions have been described. The pedicle length is
known and reliable flap. Although it is a work- around 11 cm with a caliber of 2.1 mm. The
house technique, in sacral and periacetabular venous drainage is granted by a comitantes sys-
reconstructions, it can be performed in very rare tem, which finally drains in the greater saphenous
situations, mainly using it pedicled to cover groin vein. This flap can be raised as fasciocutaneous,
or perineal defects. The pedicle originates from fascial, composite, including a portion of the rec-
the lateral circumflex femoral artery (LCFA). tus femoris muscle, or chimeric with a part of the
This branch of the profunda femoris artery origi- vastus lateralis muscle. Preoperative Angio–CT
nates 8–10 cm inferior to the anterior superior scan is mandatory to properly plan the flap, espe-
iliac spine, and it divides into three branches: cially in these patients where there can have
ascending, transverse, and descending. The much damage of this vascular system. The flap is
descending one runs on the medial edge of the raised medially to laterally, in the subfascial plan,
vastus lateralis giving off several perforator ves- looking for the chosen perforator. Once identi-
sels to the fascia and the overlying fat and skin. fied, the perforator is carefully dissected through
Most of these perforators have an intramuscular the muscle to its origin from the descending
path, while few can present a septal one. The skin branch of the LCFA (Fig. 20.4). Depending on
island is generally taken centrally over the cho- the necessity of the pedicle length, the rising of
sen perforator, with a variable dimension based the LCFA can be continued proximally.
Fig. 20.4 Wound dehiscence after sarcoma and lymphatic resection, repaired with a pedicled ALT flap
20.4 T
ensor Fascia Latae comitantes system. The skin island can be 20 cm
(TFL) Flap long and 10 cm wide, even though only a width of
8 cm consents a primary closure of the donor site.
The tensor fascia latae (TFL) flap is a type I
Mathes Nahai. It can be considered a reliable flap
for the coverage of the trochanteric, periacetabu- 20.5 Disarticulation
lar, perineum, and abdominal wall as pedicled, and Hemipelvectomy
and it has also been employed as microsurgical
flap. It can rise both as muscular or musculocuta- Disarticulation of the hip is the surgical removal
neous flap. This muscle originates from the exter- of the entire lower limb through the hip joint,
nal border of the iliac crest between the sartorius while the hemipelvectomy requires the ablation
and the gluteus medius with a tendon 5 cm wide. of the entire or a portion of the iliac bone [4].
It descends as a band, inserted onto the iliotibial Disarticulations or Hemipelvectomies are
band. The pedicle is represented by the transverse extremely mutilating procedures, but sometimes
branch of the LCFA, while the descending has they are still required to obtain a radical resec-
been already described and the ascending branch tion. Standard techniques are Boyd’s method,
travels to the gluteus minimus muscle. The trans- and the posterior flap is described by Slocum.
verse branch divides into three branches before Alternative options are represented by the ante-
entering the TFL muscle at a point around 8 – rior flap, or “anomalous flap”, or adapted modi-
10 cm inferior to the anterior superior iliac spine. fied techniques. First, inguinal or iliac lymph
It has a length of 4 – 6 cm and a caliber of 2 – nodes can be removed or not depending on the
3 mm. A small fragment of iliac crest bone can be preoperative planning. Boyd described an ana-
included if needed, preserving little branches that tomic strategy, which is still the basic procedure.
reach it. Venous drainage is usually granted by the It is oriented to the transection of the muscles at
their origin to minimize the blood loss and to pro- operative time in order to make the patient sitting
vide a well-padded stump to grant a weight- and deambulating soon.
bearing surface for prosthesis.
Slocum’s technique employs a posteromedial
flap to cover the stump, which can be used for both 20.7 Complications
hip disarticulation and hemipelvectomy. This flap
is about 10 cm long or more, in order to gently Pelvic reconstruction surgery is usually associ-
cover the stump without tension. It is important to ated with a high ratio of complications [1, 7].
ligate the femoral vessels to make them falling Despite the use of the Clinitron© bed, the greatest
above the inguinal ligament and to make the two number of them is related to the skin dehiscence,
branches of the obturator nerve retract, to avoid the due to a problem in the surgical wound healing
pressure areas. The tensor fascia latae is divided at process [1, 8]. It is essential to properly and early
the level of the greater trochanter, where the mus- treat this skin issues, mainly because they can
cle reaches it. The gluteus maximus is divided at contraindicate the start of adjuvant therapies
the distal end to the posterior skin flap. This flap, (such as chemotherapy), which are essential in
containing this muscle, is finally rotated anteriorly most of the cases. This need makes the wound
to cover the stump. Although Boyd’s technique is healing care an absolute priority for the patient.
considered the basic technique, among our To prevent wound dehiscences or infections,
Institute, we prefer Slocum’s technique to better among our institute, we usually place one or mul-
provide a useful a nonpainful stump. tiple drains in order to avoid hematoma or seroma
The standard hemipelvectomy is generally formations. We strongly believe that these are the
performed drawing a posterior or a gluteal flap to first step through a surgical failure, due to the
cover the defect. Standard hemipelvectomy dis- pression that an accumulation of fluids can place
articulates the symphysis pubis and sacroiliac on the skin margins and on the flap itself. This
joint, while in extended hemipelvectomy, the process can lead to a reduction of the blood sup-
resection line passes through the sacrum and in port. According to the oncologic principles, we
conservative hemipelvectomy through the ilium place the drains exit on the skin near to the surgi-
above the acetabulum, leaving the iliac crest. cal wound, allowing an eventually subsequent en
Finally, internal hemipelvectomy is defined as a bloc resection. Speaking of irradiated wounds,
limb-sparing procedure [5]. large flaps, or sacral and posterior resection in
Anterior flap disarticulation or hemipelvec- general, we prefer to keep drains longer than
tomy is used for lesions of the buttock or poste- what literature generally advises (less than 30 cc
rior proximal thigh. It employs a quadriceps per day).
myocutaneous flap based on the superficial femo- The incidence of a flap loss is low and depends
ral artery [6]. mostly on the area and the type of flap used.
Speaking of the sacrum, these problems regard
mostly the V–Y advancement flaps, which are
20.6 Postoperative Management less likely to obliterate the dead space, allowing
the hematoma or seroma formations. In a recent
Normally, these patients require at least one night study, [9] it has been found that the use of acel-
in the ICU. We started using Air Fluidized lular dermal matrix to reconstruct the posterior
Therapy beds, Clinitron©, especially for those abdominal wall reduces the risk of posterior
patients undergone to a sacral resection, for not to bowel herniations, bowel obstructions, and fis-
let the weight bearing area on the flap. Obviously, tula formations. Due to these reasons, for poste-
the fluid resuscitation is mandatory to maintain rior reconstructions, the Authors suggest to use
an adequate perfusion and the choice to use gluteal flaps only for smaller defects, performing
hydrocolloids or blood product depends on the a VRAM flap plus dermal matrix for any greater
postoperative blood tests. Physical and rehabili- demolition, especially for combined anterior and
tative programs are scheduled early in the post- posterior approaches [1].
References pressure ulcers. Cham, Switzerland: Springer; 2015.

p. 221–42.
6. Vancabeke M, Harper L, Penders W, Putz P. Anterior
1. Ruggieri P, Angelini A, Vanel D, Picci P. Tumors of the
flap for coverage following hip disarticulation for
sacrum: diagnosis and treatment of benign and malig-
osteomyelitis. Acta Orthop Belg. 1999;65(2):223–5.
nant tumors. Cham, Switzerland: Springer; 2017, 1 p.
7. Garvey PB, Rhines LD, Feng L, Gu X, Butler
2. Grimer RJ, Chandrasekar CR, Carter SR, Abudu A,
CE. Reconstructive strategies for partial sacrectomy
Tillman RM, Jeys L. Hindquarter amputation: is it
defects based on surgical outcomes. Plast Reconstr
still needed and what are the outcomes? Bone Jt J.
Surg. 2011;127(1):190–9.
2013;95-B(1):127–31.
8. Fuchs B, Dickey ID, Yaszemski MJ, Inwards CY, Sim
3. Angelini A, Kotrych D, Trovarelli G, Szafrański A,
FH. Operative management of sacral chordoma. J
Bohatyrewicz A, Ruggieri P. Analysis of principles
Bone Joint Surg Am. 2005;87(10):2211–6.
inspiring design of three-dimensional-printed custom-
9. Maricevich M, Maricevich R, Chim H, Moran SL,
made prostheses in two referral centres. Int Orthop.
Rose PS, Mardini S. Reconstruction following partial
2020; https://doi.org/10.1007/s00264-020-04523-y.
and total sacrectomy defects: an analysis of outcomes
4. Record EEMD. Atlas of limb prosthetics. Surgical and
and complications. J Plast Reconstr Aesthetic Surg
prosthetic principles. JBJS. 1981;63(6):1038.
JPRAS. 2014;67(9):1257–66.
5. Rubayi S. Disarticulation and total thigh flap. In:
Rubayi S, editor. Reconstructive plastic surgery of
Anesthesia for Pelvic Bone Cancer
Surgery: From Risk Evaluation 21
to Postoperative Course
Gian Mario Parise, Bianca Ferrarese,
Alessandro Graziano, Manuela Funes,
Francesco Ambrosio, and Paolo Navalesi
Orthopedic oncologic surgeries are challenging and/or pulmonary evaluation (spirometry, arterial
for the anesthesiologist. When planning anesthe- blood gas). In this phase, postoperative ICU stay
sia, numerous factors must be considered. Pelvic should be discussed according to the type of sur-
oncologic surgeries vary in length and complex- gery and the patient’s medical history.
ity. Thus, a well-formulated anesthetic plan cre- The management of patients undergoing pel-
ated by effective communication between the vic tumor surgeries is particularly complicated
anesthesiologist and the surgeon is essential to because of the possibility of massive hemorrhage
ensure optimal patient outcomes [1, 2]. [1–6]. According to studies, pelvic tumor surger-
Preoperative assessment of the patient for ies show an intraoperative blood loss ranging
anesthesia begins 2–3 weeks prior to the sched- from 2500 to 5000 mL [2, 4]. Although, in most
uled surgery date. Patients are seen by the anes- circumstances, the administration of blood and/
thesiologist and they must receive a thorough or blood products can be used to effectively cor-
preoperative evaluation to elucidate significant rect hemoglobin concentrations and coagulation
comorbidities or side effects secondary to che- function, there is a growing body of evidence that
motherapy and radiation [1, 2]. The anesthetic shows the potential adverse effects of allogeneic
preoperative consultation should include full blood product administration. These adverse
blood examination including baseline hemoglo- effects include the transmission of infectious dis-
bin level, hematocrit, platelet count, iron and eases, immunosuppression, transfusion-related
coagulation function (prothrombin time (PT) acute lung injury, transfusion reactions, and they
with international normalized ratio (INR), and can also be associated with decreased tumor-free
activated partial thromboplastin time (aPTT)). survival [2, 5–9]. Moreover, patients with comor-
Preoperative evaluation should also include ECG bidities or those who have undergone chemother-
and chest radiograph. Based on the patient’s apy and radiotherapy may often have significant
medical history, physical exam, and test results, anemia and thrombocytopenia. Advances in
the anesthesiologist may also require further car- anesthesia, surgery, and transfusion medicine
diac evaluation (echocardiography, stress testing) over the past decade have led to the development
of “patient blood management,” a multimodal,
G. M. Parise (*) · B. Ferrarese · A. Graziano evidence-based preoperative and perioperative
M. Funes · F. Ambrosio · P. Navalesi strategy, aimed to minimize the need for red
Department of Internal Medicine, Anesthesia and blood cell transfusions, consisting of treating
Intensive Care Unit, Padua University Hospital,
Padua, Italy anemia and reducing perioperative blood loss [8,
e-mail: [email protected] 10]. Depending on the cause and degree of

234 G. M. Parise et al.
anemia, the urgency of the procedure, and the in higher postoperative hemoglobin, quicker
anticipated blood loss, patients are given a spe- postsurgical recovery, shorter hospitalization,
cific therapy in order to improve their hemoglo- and decreased surgery-related costs [10].
bin level. Individuals with iron deficiency or iron For patients requiring urgent or emergency
deficiency anemia should be treated with iron, surgery, it may not be possible to correct all fac-
allowing adequate time for effect of treatment tors adversely affecting the baseline hemoglobin
before surgery (typically 2–4 weeks for correc- level and defects affecting hemostasis, and trans-
tion). Intravenous (IV) iron is an option if less fusions may be needed. In the preoperative
than 4–6 weeks are available, and for patients period, patient preferences and acceptance or
with poor tolerance or an inadequate response to refusal to receive various blood components
oral iron. For individuals with anemia of chronic should be discussed, and related consents and
disease/inflammation, we typically administer advanced directives should be obtained and
preoperative EPO (together with supplemental documented.
iron to avoid functional iron deficiency) [8]. Recommended preoperative evaluation also
The table below shows patient blood manage- includes a pain control plan. Oncologic patients
ment according to our procedure (Table 21.1). often have pain prior to surgery and they often
A study also shows that preoperative sucroso- receive significant amounts of drugs and opioids
mial iron supplementation at least 4 weeks prior to control it. The anesthesiologist needs an accu-
to an elective surgery in non-anemic patients lim- rate idea of the patient’s level of pain, opioid tol-
its the drop in postoperative Hb levels, resulting erance, and requirements [2]. Patient preparation
should include adjustments of preoperative medi-
cations to avoid withdrawal effect, treatment to
Table 21.1 Preoperative patient blood management
reduce preoperative pain/anxiety, and beginning
1 LONG PREOPERATIVE TIME (4 WEEKS) of preoperative treatment as part of a multimodal
• low serum iron and low ferritin (<30 ug/L)
Sucrosomial Iron 4 tabs/day p.o. (2 tabs in the pain control plan, which is vital to successful
morning + 2 tabs in the evening) for 30 days postoperative pain management.
• low serum iron and normal ferritin To improve surgical outcomes, adequate
(>30 ug/L) patient hydration and nutrition should be achieved.
Sucrosomial Iron 4 tabs/day p.o. for
30 days + EPO 40000 IU s.c./week Preoperative fluid management should guarantee
• Normal serum iron and high ferritin a near-zero fluid balance. Studies show that opti-
(>100 ug/L) mization of the preoperative metabolism by oper-
EPO 40000 IU s.c./week ating a patient under the influence of insulin after
2 MEDIUM PREOPERATIVE TIME (2–3 WEEKS)
• low serum iron and low ferritin (<30 ug/L) giving a carbohydrate load preoperatively (such
Ferric Carboxymaltose 1000 mg i.v. (check Hb as 100 g oral carbohydrate drink) results in attenu-
e ferritin after 10–12 days) ation of glucagon release, complete abolition of
possible repetition of ferric Carboxymaltose cortisol release, which is thus far less catabolic in
1000 mg ev
• low serum iron and normal ferritin the postoperative period [11].
(>30 ug/L) About preoperative fasting, the American
Ferric Carboxymaltose 1000 mg i.v. + EPO Society of Anesthesiologists (ASA) recommends
40000 IU s.c./week to minimize starvation: patients should stop sol-
• Normal serum iron and high ferritin
(>100 ug/L) ids 8–6 h and liquids 2 h before a procedure [12].
EPO 40000 IU s.c./week In extensive oncologic surgeries, bowel prepa-
3 SHORT PREOPERATIVE TIME (1 WEEK) ration can be useful. Laxatives should be given
• low serum iron and low ferritin 2 days prior to the surgery and adequate intrave-
Ferric Carboxymaltose 1000 mg i.v. + EPO
40000 IU s.c. × 2 nous hydration and nutrition should be ensured.
• Normal serum iron and high ferritin These preoperative interventions (along with
(>100 ug/L) patient and family education and counseling, pre-
EPO 40000 IU s.c. × 2 operative cessation of smoking, drinking, prophy-
21 Anesthesia for Pelvic Bone Cancer Surgery: From Risk Evaluation to Postoperative Course 235
laxis of infection and thromboembolic events, and and connected to the ventilator. The anesthetics
normothermia) are part of standard care pathways generally used are: Propofol, Midazolam, or
known as Enhanced Recovery After Surgery Ketamine to reach unconsciousness, Fentanyl or
(ERAS) protocols. The aims of the ERAS pro- Ketamine for analgesia, Rocuronium or
grams are to reduce surgery-related morbidity, Cisatracurium for neuromuscular block. The
standardize and optimize perioperative medical maintenance of general anesthesia can be
care, decrease the length of hospital stay, and achieved by inhalation of a volatile anesthetic
facilitate the patient’s return to normal life [13]. agent (Desflurane, Sevoflurane) or by total intra-
On the scheduled surgery date, the patient is venous anesthesia (TIVA) or TIVA target-
taken into the operating room. After his identifica- controlled infusion (TCI). Low-dose Remifentanil
tion, a blood pressure cuff, ECG leads, and a pulse (a short-acting synthetic opioid analgesic drug) is
oximeter are applied to monitor his non-invasive given intraoperatively by continuous intravenous
blood pressure, electrical activity of the heart, and infusion to relieve pain.
oxygen saturation. A large peripheral intravenous Placing an arterial catheter for continuous
(IV) line is placed to administer premedication, blood pressure monitoring and sampling, and
drugs, fluids, and possible blood products. acquiring a central venous line (internal jugular
The aims of premedication are reduction of vein or subclavian vein) are imperative prior to
anxiety and pain, enhancing the hypnotic effects surgical incision [2]. The central line placement
of general anesthesia, promotion of retrograde should always be checked via chest X-ray, and
amnesia, and reduction of vagal reflexes. the catheter tip position should be close to the
A great number of anesthetic techniques can border of the superior vena cava and the right
be used. The epidural or spinal+epidural tech- atrium. A central venous catheter is fundamental
nique in combination with general anesthesia for fluids and blood products infusion, infusion
(GA) is usually the best choice [1, 2]. of vasoactive and inotropic agents, and monitor-
The level of the spine at which the epidural ing the central venous pressure. Vasopressors and
catheter is placed should be discussed with the inotropes are often used to maintain an adequate
orthopedic surgeon according to the location and central venous pressure and venous return and to
size of the tumor, type and extension of the sur- optimize the heart rate, especially when a spinal
gery, and patient positioning. Epidural catheters or epidural anesthetic is used. Hypotension may
are usually inserted at T11 to T12 or T12 to L1 cause hypoperfusion, thus blood pressure must
interspaces, prior to induction of general anesthe- be monitored closely, especially in those patients
sia, and they are secured in place to enable both suffering from cardiovascular, pulmonary, or
additional intraoperative drug infusion and post- renal diseases. Studies have shown that an intra-
operative analgesia. operative mean arterial blood pressure value of
The administration of the anesthetic through 60 mmHg is enough to ensure optimal organ per-
the epidural catheter aims at obtaining analgesia fusion, although this target should be higher in
up to dermatomes T2–T3, generating both sys- hypertensive patients and those with cardiovas-
temic and cardiac blockade of the sympathetic cular comorbidities [7, 14].
nervous system, which causes vasodilatation and Invasive blood pressure monitoring with arte-
blocks the cardiac accelerator fibers, leading to rial line is often associated with hemodynamic
hypotension and superior rate control [14]. monitoring (MostCare®, Picco®) based on the
According to the literature, intraoperative analysis of the peripheral arterial waveform,
hypotensive epidural anesthesia may reduce which provides the measurement of the main
bleeding and transfusion, improve the quality of hemodynamic parameters, thanks to specific
the operative field, and shorten operative times algorithms.
[5, 6, 8, 14]. A major complication of pelvic oncologic
After epidural catheter placement, general surgeries is massive blood loss [1–6]. Large
anesthesia is induced, and the patient is intubated tumors, requiring extensive resection with a
close proximity to vascular structures, will be As with any prolonged surgery, careful atten-
more likely to cause a larger blood loss. The tion needs to be given to prevent peripheral nerve
internal iliac vessels are more likely to be trou- compression. Poor patient positioning may result
blesome when tumors are situated posteriorly in in devastating outcomes [2, 14]. Patients must be
the pelvis. Anterior pelvic tumors that require placed in anatomic positions with pressure points
dissection near the bladder neck are problematic padded to prevent compression or stretch-related
because of the perivesical venous plexus that injuries. When a patient is placed in the supine
often bleeds heavily or continuously. Bone cuts position, if the upper extremities are abducted,
expose the bleeding bone and can provide a sus- they should remain supinated and padded at no
tained source of hemorrhage. Renal cell meta- more than 90 degrees. For a patient in the lateral
static tumors are exceptional in being highly decubitus or prone position, the neck should be
vascular and are associated with massive blood maintained in a midline position, and there
loss during surgeries [1]. Therefore, controlled should be no excessive pressure on the eyes and
hypotensive epidural anesthesia is very helpful ears. The abdomen should not be compressed
in aiming to decrease the intraoperative blood because excessive pressure may compromise
loss and red blood cell transfusion requirements ventilation and decrease venous return from the
[5, 6, 8, 14]. A restrictive red blood cell transfu- lower extremities [2].
sion threshold of 8.0 g/dL is recommended for Pelvic tumor surgeries may be complicated by
patients undergoing oncologic orthopedic sur- intraoperative hypothermia [14]. Patient heat loss
gery or when the blood loss is greater than 20% is primarily due to a significant exposure of the
of the total blood volume. To manage the high skin and internal viscera to the operating room
flow of blood and fluids required, rapid infusion air. Forced air warming devices and fluid warm-
devices are often helpful. Autologous transfu- ers can be adopted to prevent this complication.
sions do not improve long-term outcomes over A low body temperature can cause and worsen
exogenous donor red blood cell transfusions [2]. acidosis and coagulopathy, and this combination
Unfortunately, cell salvage cannot be used as it can be lethal (known also as the trauma triad of
may increase the risk of spreading tumor cells death) [7].
systemically, although there is some promising A life-threatening complication is cardiovas-
research on the effectiveness of filtration and cular collapse, due to acute right heart failure sec-
irradiation to reduce the tumor load of salvaged ondary to massive fat embolism, which can occur
blood [1]. during the reconstruction phase. Reaming and
Acute Normovolemic Hemodilution (ANH) is pressurizing of the femoral canal may release
another technique used to decrease the influence emboli into the circulation, although pulmonary
of perioperative transfusions. It is a blood conser- embolism is less common in comparison to hip
vation technique that involves the removal of and knee arthroplasty.
blood from a patient shortly after the induction of Preemptive management involves avoiding
anesthesia, with maintenance of normovolemia hypovolemia and fluid overload, increasing the
using a crystalloid and/or colloid replacement inspired oxygen concentration, minimizing the
fluid. The blood withdrawn from the patient is use of vasodilators, and starting the administra-
frequently kept at room temperature, to be tion of vasopressors (dopamine, norepinephrine,
returned to the patient later in the procedure, no dobutamine).
longer than 8 h after collection. This action has Transoesophageal echocardiography is not
shown some efficacy in reducing allogeneic recommended, and the anesthesiologist should
blood transfusions in cardiac and miscellaneous pay attention to clinical signs that include desatu-
procedures, though not with orthopedic surger- ration, hypoxemia, hypocapnia, hypotension, and
ies. This technique is also often associated with increased pulmonary shunt [1]. If cardiovascular
significantly higher intraoperative fluid and vaso- collapse occurs, the team should start cardiopul-
pressor requirements [2]. monary resuscitation.
At the end of the procedure, patients, who had Table 21.2 Postoperative complications
severe complications or undergo extensive proce- Anemia
dures or have severe comorbidities, should be Postoperative nausea and vomiting
kept electively intubated and admitted to the Excessive postoperative pain
Intensive Care Unit for postoperative ventilation, Cardiovascular system
Myocardial infarction
fluid and electrolyte management, and correction
Congestive cardiac failure
of coagulopathy. Otherwise, neuromuscular
Atrial fibrillation
relaxation is antagonized (using sugammadex or Angina
neostigmine + atropine), maintenance agents are Wound
discontinued, and the patient is allowed to awake. Superficial wound infection
Extubation is performed when spontaneous Deep wound infection
breathing is re-established and full recovery of Wound dehiscence
neuromuscular activity is achieved. Wound hematoma
Afterward, the patient is typically taken to the Urinary tract
Post Anesthesia Care Unit (PACU) for an imme- Urinary tract infection
Acute urinary retention
diate postoperative follow-up, which includes
Vascular system
airway management and oxygen administration, Deep venous thrombosis
monitoring the vital signs (heart rate, blood pres- Venous thromboembolic disease
sure, temperature, and respiratory rate), manag- Respiratory tract
ing postoperative pain, treating postoperative Pneumonia
nausea and vomiting (PONV), treating postanes- Respiratory failure
thetic shivering, and monitoring the surgical sites Symptomatic electrolyte abnormality
for excessive bleeding. Anesthesiologists and Neurological
nurse anesthetists should also prepare and teach Peripheral nerve injury
Postoperative paralysis
the use of patient-controlled units, and additional
Cerebrovascular accident
intravenous and/or epidural infusions could be
Death
administered. The patient remains in the PACU
for 3–4 h to ensure recognition of possible late-
onset pain or sedation. Then, the patient is trans- volatile anesthetic agents, and/or specific patient
ferred to the Orthopedic Oncology Unit. The factors are involved in this pathogenesis. A vari-
arterial line is usually removed, and the central ety of pharmacologic agents are available to
venous line should be maintained. reduce the incidence of PONV, such as serotonin
Most frequent postoperative complications (5HT3) inhibitors (e.g., ondansetron); metoclo-
are symptomatic anemia, postoperative pain, pramide, droperidol, which act primarily through
wound infections, urinary tract infections, and dopaminergic antagonism; dexamethasone [2].
deep venous thrombosis [1, 2] (Table 21.2). In these patients, postoperative pain manage-
Blood loss, Hb levels, and coagulation func- ment is difficult and challenging, due to the sub-
tion should be monitored daily. Anemia is a very stantial area and site of the surgery [1, 2].
common complication. The decision criteria for Oncologic orthopedic patients characteristically
blood transfusion should not be a static value, suffer varying degrees of pain or discomfort long
and the clinical status of the patient should be before operation, and this might cause worse post-
taken into account; however, one of the most operative pain, as mentioned earlier. Adequate
important factors to be considered is the preop- postoperative pain control is crucial and complex
erative hemoglobin baseline. If necessary, blood because poorly managed pain can contribute to
transfusion and fluid therapy should be adminis- the development of short- and long-term postop-
tered via the central venous line. erative complications including atelectasis, pneu-
Postoperative nausea and vomiting (PONV) is monia, ileus, delayed ambulation, and
often observed. The use of opioids, inhalation of postoperative persistent pain. Pain m anagement
should be targeted to: (a) improve functional out- the patient’s demand. Usually the infusion is
comes and reduced in-hospital length of stay, (b) commenced in the ICU or PACU once the patient
reduce the inflammatory and stress response asso- has become hemodynamically stable. The dura-
ciated with surgery, (c) minimize the risk of per- tion typically ranges from 2 to 6 days, based on
sistent postsurgical pain development, and (d) the patient’s pain. A careful postoperative moni-
facilitate the return of patients to the next planned toring of the epidural analgesia should be per-
oncological therapy [15]. formed to detect an early onset of neurological
There are several approaches to control the complications. When epidural analgesia is no
postoperative pain: epidural analgesia (EA) longer needed, the antithrombotic agent is with-
using either continuous epidural administration held for 12–18 h before the catheter can be
(CEA) and/or patient-controlled epidural (PCEA) removed; and sensory and motor integrity should
techniques, or intravenous analgesia (IVA) be reevaluated 6 h after the removal of the
using either continuous intravenous infusion and/ catheter.
or intravenous patient-controlled analgesia When the epidural catheter could not be
(IV-PCA). The method used is dependent on the placed, patients are given intravenous analgesia:
technique chosen for anesthesia. a continuous infusion of analgesic (usually opi-
Patients who have an epidural catheter placed oids such as morphine) or intravenous patient-
(CEA) can be managed with a continuous infu- controlled techniques (IV-PCA). In this case, the
sion of a local anesthetic (ropivacaine or levobu- patient can self-administer analgesic preset
pivacaine) in combination with an opioid boluses (morphine) via a patient-controlled anal-
(sufentanyl) (Table 21.3). gesia pump. As for PCEA, the device is preset by
A patient-controlled analgesia device may be the anesthesiologist and the analgesic bolus is
attached to the epidural line (PCEA). Patient- always followed by a lockout period.
controlled techniques allow patients to self- When possible, epidural analgesia should be
administer small boluses of analgesics, thus always preferred. Studies demonstrated a clear
providing better titration and enhancing respon- antinociceptive superiority of the epidural over
siveness to their analgesic requirements. The the intravenous methods for pain control [2, 16].
device is programmed by the anesthesiologist to Besides, epidural infusion or PCEA often show a
deliver a preset number of analgesic boluses per lower rate of side effects compared with IVA or
day, whenever the patient activates it. Each anal- IV-PCA. Some of the epidural analgesia benefits
gesic bolus is followed by a preset lockout period include excellent pain control despite the decrease
to avoid analgesic overdose. A patient-controlled in the total amount of opioids, with minimal
epidural analgesia (PCEA) usually also consists respiratory depression, decreased somnolence,
of a continuous background infusion of a combi- and early ambulation, although pruritus, nausea,
nation of a local anesthetic (ropivacaine or urine retention, and rare neuroaxial disturbances
levobupivacaine) and an opioid (sufentanyl). Of are still described.
note, during the patients’ stay in the hospital, Another novel drug delivery system that is
additional epidural boluses could be adminis- approved for use in 33 European countries but
tered manually by the attending anesthetist upon not yet in the United States is a sufentanil (opi-
oid) sublingual tablet system (see Fig. 21.1). This
Table 21.3 Epidural analgesia system does not require an IV line. It is a hand-
Epidural analgesia
held, preprogrammed, noninvasive, patient-
Local anesthetic (from []% + Opioid (from []% activated device that delivers sufentanil 15-mcg
to []%) to []%) microtablets on demand. Numerous studies and
Ropivacaine (0.125% to 0.2%) Sufentanyl randomized, placebo-controlled trials have dem-
Or levobupivacaine (0.0625% 0.02%–0.04% onstrated its superiority compared with morphine
to 0.125%) delivery by IV PCA [17–19].
References
1. Molnar R, Emery G, Choong PFM. Anesthesia
for hemipelvectomy: a series of 49 cases. Anaesth
Intensive Care. 2007;35:536–43.
2. Anderson MR, Jeng CL, Wittig JC, et al. Anesthesia
for patients undergoing orthopedic oncologic surger-
ies. J Clin Anesth. 2010;22(7):565–72.
3. Cata JP, Gottumukkala V. Blood loss and massive
transfusion in patients undergoing major oncological
surgery: what do we know? Int Scholarly Res Netw
ISRN Anesthesiol. 2012;2012:918938, 11 pages.
4. Tang X, Guo W, Yang R, et al. Evaluation of blood
loss during limb salvage surgery for pelvic tumours.
Int Orthop. 2009;33:751–6.
5. Jiang J, Zhou R, Li B, et al. Is deliberate hypotension
a safe technique for orthopedic surgery?: a systematic
review and meta-analysis of parallel randomized con-
trolled trials. J Orthop Surg Res. 2019;14(1):409.
6. Freeman AK, Thorne CJ, Gaston CL, et al.
Hypotensive epidural anesthesia reduces blood loss in
pelvic and sacral bone tumor resections. Clin Orthop
Fig. 21.1 Sufentanyl sublingual tablet system Relat Res. 2017;475(3):634–40.
7. Tobias JD. Strategies for minimizing blood loss in
orthopedic surgery. Semin Hematol. 2004;41(1 Suppl
Accordingly, the management of postopera- 1):145–56. Review.
tive pain proves difficult and often requires a 8. Sambandam B, Batra S, Gupta R, et al. Blood conser-
multimodal approach to achieve success. vation strategies in orthopedic surgeries: a review. J
Clin Orthop Trauma. 2013;4(4):164–70.
Parenteral opioids alone rarely provide adequate 9. Wall T, Sherwin A, Ma D, et al. Influence of perioper-
analgesia; the addition of non-opioid strategies ative anaesthetic and analgesic interventions on onco-
(such as acetaminophen, NSAIDs, NMDA recep- logical outcomes: a narrative review. Br J Anaesth.
tor antagonists (ketamine), gabapentinoids (pre- 2019;123(2):135–50.
10. Scardino M, Di Matteo B, Martorelli F, et al. Improved
gabalin or gabapentin), β-blockers, α2-adrenergic patient blood management and cost saving in hip
agonists) is frequently beneficial and should be replacement surgery through the implementation of
considered as pharmacological modalities that pre-operative Sucrosomial® iron supplementation: a
can be combined to regional or systemic analge- quality improvement assessment study. Int Orthop.
2019;43(1):39–46.
sia techniques. However, non-steroidal anti- 11. Chowdhury AH, Lobo DN. Fluids and gastrointes-
inflammatory agents are often avoided due to the tinal function. Curr Opin Clin Nutr Metab Care.
problem of coagulopathy. 2011;14(5):469–76.
Multimodal analgesia should be considered as 12. American Society of Anesthesiologists Committee.
Practice guidelines for preoperative fasting and the
the standard of care modality with the goal of use of pharmacologic agents to reduce the risk of
reducing the surgical stress and opioid consump- pulmonary aspiration: application to healthy patients
tion. It is a key component of enhanced-recovery undergoing elective procedures: an updated report by
pathways, improving short-term postoperative the american society of anesthesiologists task force
on preoperative fasting and the use of pharmacologic
outcomes not only by shortening the length of agents to reduce the risk of pulmonary aspiration*.
stay in the hospital and accelerating the quality of Anesthesiology. 2017;126(3):376–93.
return to “normal life” but also optimizing the 13. Iqbal U, Green JB, Patel S, et al. Preoperative
conditions in which patients can return to their patient preparation in enhanced recovery pathways.
J Anaesthesiol Clin Pharmacol. 2019;35(Suppl
next intended oncological therapy [15]. 1):S14–23.
The anesthesiologist, surgeon, and physiatrist 14. Shah A, Palmer AJR, Klein AA. Strategies to mini-
should work together to minimize the orthopedic mize intraoperative blood loss during major surgery.
oncologic patient’s postoperative discomfort. Br J Surg. 2020;107(2):e26–38.
15. Cata JP, Lasala J, Bugada D. Best practice in the 18. Ringold FG, Minkowitz HS, Gan TJ, et al. Sufentanil
administration of analgesia in postoncological sur- sublingual tablet system for the management of post-
gery. Pain Manag. 2015;5(4):273–84. Epub 2015 Jun operative pain following open abdominal surgery: a
15. randomized, placebo-controlled study. Reg Anesth
16. Weinbroum AA. Superiority of postoperative epi-
Pain Med. 2015;40(1):22–30.
dural over intravenous patient-controlled anal- 19. Melson TI, Boyer DL, Minkowitz HS, et al. Sufentanil
gesia in orthopedic oncologic patients. Surgery. sublingual tablet system vs. intravenous patient-
2005;138(5):869–76. controlled analgesia with morphine for postoperative
17. Jove M, Griffin DW, Minkowitz HS, et al. Sufentanil pain control: a randomized, active-comparator trial.
sublingual tablet system for the management of post- Pain Pract. 2014;14(8):679–88.
operative pain after knee or hip arthroplasty: a ran-
domized, placebo-controlled study. Anesthesiology.
2015;123(2):434.
The Importance
of a Multidisciplinary Approach 22
to Pelvic Tumours
Andreas Leithner, Marko Bergovec,
and Dimosthenis Andreou
In 1969, more than 50 years ago, the Apollo 11

mission successfully reached the moon, an event
that has been internationally celebrated
(Fig. 22.1). One might ask, “What does the moon
landing have to do with pelvic tumours?”
Multidisciplinarity! Such a success has only been
possible due to a multidisciplinary, highly spe-
cialized team [1]. Another moon-related example
is the NASA-team building game, where you are
a member of a space crew and where you have to
decide which items are the most important ones
to successfully reach your 200-mile away space
station. A team of people discussing options usu-
ally gets results closer to the NASA experts’
solution than each of the individuals alone.
Teamwork in critical situations like space mis-
sions is often the key to success [2].
Orthopaedic surgeons have been known to be Fig. 22.1 Buzz Aldrin poses on the Moon [concerning
licencing—picture taken from wikipedia.org—“This file
highly intelligent [3], but even they cannot suc- is in the public domain in the United States because it was
cessfully treat primary pelvic sarcomas alone. As solely created by NASA. NASA copyright policy states
highlighted in the previous chapters, the multidis- that “NASA material is not protected by copyright unless
ciplinary treatment of these tumours is of utmost noted“. (See Template:PD-USGov, NASA copyright pol-
icy page or JPL Image Use Policy)”]
A. Leithner (*) · M. Bergovec importance: chemotherapy has been shown to be

Department of Orthopaedics and Trauma, a prerequisite for survival in patients with osteo-
Medical University of Graz, Graz, Austria sarcomas and Ewing sarcomas, and radiotherapy
e-mail: [email protected]; can improve the outcome in at least some patients
[email protected]
with Ewing sarcoma, while none of us would ever
D. Andreou perform surgery for pelvic chondrosarcoma with-
Sarcoma Center Berlin Brandenburg,
Berlin, Germany out adequate imaging. The 2018 ESMO-
e-mail: [email protected] EURACAN guidelines for soft tissue and visceral

242 A. Leithner et al.
sarcomas therefore state: “A multidisciplinary ferred to our tumour centre. To exclude the dif-
approach is, therefore, mandatory in all cases, ferential diagnosis of brown tumours due to
involving pathologists, radiologists, surgeons, hyperparathyroidism, we determined the para-
radiation therapists, medical oncologists and pae- thormon level in serum, which was exorbitantly
diatric oncologists, as well as nuclear medicine high at 922 pg/ml. Further investigations con-
specialists and organ-based specialists, as appli- firmed a parathyroid adenoma (Fig. 22.2c). After
cable” [4]. Similarly, the 2018 ESMO–PaedCan– its resection serum levels of parathormon
EURACAN guidelines for bone sarcomas are decreased but the patient developed a hungry
focused on multidisciplinarity, stating for e.g. bone syndrome, despite a prophylactic treatment
“Samples must be interpreted by an experienced with a high-dose calcium substitution. After
bone sarcoma pathologist, in collaboration with 1 year of therapy calcium and CrossLaps values
the radiologist, and discussed in a multidisci- returned to normal levels and the radiologic con-
plinary team” or “All new cases of bone tumours trols showed bone consolidation (Fig. 22.2d).
should be formally discussed in a multidisci- The patient was symptomless after 2 years of
plinary team at a bone sarcoma reference centre follow-up. The take home message of this case is
with the radiologist, the pathologist, the surgeon, that a pathologist has to have radiological and
the radiation oncologist and the medical and/or clinical information (e.g. multiple lesions) to
paediatric oncologist” [5]. reach the correct diagnosis.
We all know the importance of multidisci-
plinary treatment concepts; nevertheless, in some
cases, we tend to forget. These are mostly patients 22.2 S
urgery Is Not Always
with symptomatic metastases and benign or the Best Option for Bone
locally aggressive/rarely metastasizing pelvic Metastases
tumours, who are sometimes not discussed in
interdisciplinary tumour boards, as the indication Whereas it is clear that surgery will not be benefi-
for surgical treatment seems clear to the ortho- cial in some patients i.e. with multiple osteoblas-
paedic oncologist and other treatment options are tic prostate cancer metastases (Fig. 22.3),
not considered. But, even in patients with pri- intralesional curettage and stabilization with
mary sarcomas undergoing multidisciplinary Steinmann pins may be necessary in other
therapy, the optimal sequence of the planned patients with i.e. symptomatic osteolytic pelvic
treatment modalities can be unclear, especially if metastases of a renal cell carcinoma, allowing
new findings render a reassessment of the neces- immediate full weight-bearing and improving
sary initial treatment plan. quality of life (Fig. 22.4). This procedure has
been shown to lead to excellent results in selected
patients. As a result of good experiences with sur-
22.1 P
athologists Need Clinical geries like the Harrington procedure, many sur-
and Radiological geons, when asked at conferences and courses,
Information proposed similar surgical techniques in a case of
a large supraacetabular defect in a 59-year old
A 49-year-old woman was primarily admitted female patient with multiple endometrial carci-
with multiple osteolytic tumours (Fig. 22.2a) for noma metastases (Fig. 22.5a). A multidisci-
further surgical treatment after external curettage plinary case discussion involving gynaecologists,
of one of those lesions and the external histologi- medical and radiation oncologists, orthopaedic
cal diagnosis of giant cell tumour (Fig. 22.2b), a surgeons, radiologists and pathologists, however,
diagnosis confirmed by a second histopathologi- came to the conclusion that surgery is not the
cal investigation accomplished by a bone tumour treatment of choice. At 10 months of follow-up,
specialist, who did not have access to the patient’s after local radiotherapy and antihormonal ther-
radiographs. Consequently, the patient was trans- apy with an aromatase inhibitor, the patient was
22 The Importance of a Multidisciplinary Approach to Pelvic Tumours 243
a b
c d
Fig. 22.2 Female, 49y, (a) a pelvis x-ray showing multi- parathyroid adenoma; (d) complete radiological healing
ple osteolytic lesions; (b) the histology presents multiple of the osteolytic lesions 2 years after parathyroid resec-
multinucleate giant cells; (c) surgical specimen of the tion – without any pelvic surgery necessary
22.3 T
he Optimal Local Treatment
Modality in Patients
with Pelvic Ewing Sarcoma Is
Still a Matter of Debate
No randomized controlled trials comparing the

outcome of different local treatment modalities
in patients with pelvic Ewing sarcomas are avail-
able. As a result, treatment strategies are greatly
influenced by the sometimes conflicting evidence
of smaller, retrospective analyses and the philos-
ophy of the respective study groups. Definitive
Fig. 22.3 Male, 59 years, multiple prostate carcinoma radiotherapy tends to be more often utilized in
metastases in all parts of the skeleton patients treated in the United States, whereas
European Cooperative Study Groups consider
pain-free under full weight-bearing and had a the results of surgical treatment with or without
good quality of life without having to undergo radiotherapy to be superior, at least in patients
any surgery. with localized disease. However, the results of
a b
Fig. 22.4 Male, 75 years, multiple renal cell carcinoma weight-bearing 10 months after a Harrington procedure
metastases with (a) a painful disability due to the left following preoperative embolization
supraacetabular osteolysis; (b) stable situation with full
a b
Fig. 22.5 Female, 59 years, multiple metastases of endometrial cancer, (a) with a large osteolytic lesion on the right
supraacetabular site, (b) 10 months after local radiotherapy and antihormonal therapy
surgical treatment appear to depend on the apy point out that it is associated with a lower
tumour localization in the pelvis, as the toxicity and fewer long-term complications,
Scandinavian Sarcoma Group recently demon- compared to postoperative radiotherapy. On the
strated that definitive radiotherapy appears to be other hand, a recent study demonstrated conclu-
adequate for patients with sacral tumour localiza- sively that the prognosis of patients with a poor
tion [6]. A recent report of the Euro-EWING99 response to neoadjuvant chemotherapy alone
consortium verified this finding and additionally could be significantly improved with adjuvant
showed that patients with localized non-sacral high-dose chemotherapy—a highly toxic treat-
Ewing sarcomas had a significantly better prog- ment that cannot be recommended for all patients
nosis following surgery and additional radiother- [8]. Unfortunately, the histological response to
apy, compared with surgery alone [7], raising the treatment cannot be assessed in patients undergo-
question of when radiotherapy should ideally be ing both preoperative chemotherapy and preop-
performed. Proponents of preoperative radiother- erative radiotherapy – as a result, the optimal
22 The Importance of a Multidisciplinary Approach to Pelvic Tumours 245
adjuvant therapy in these patients remains lenge for treating physicians – surgeons do not
unclear. Therefore, no blanket recommendation want to endanger their reconstructions or risk of
for pre- or postoperative radiotherapy can be sup- patients developing septic complications by clear-
ported at this point, and the optimal treatment can ing them to receive chemotherapy in cases of
only be determined after interdisciplinary discus- delayed wound healing, but they also do not want
sion on a case-to-case basis. to have a negative influence on their patients’
prognosis. A collective interdisciplinary decision
on when to proceed with the adjuvant treatment
22.4 “Real-Life” Multidisciplinary can help address this challenge.
Treatment Looks Different
Than on Paper
22.5 Multidisciplinary Treatment
Multidisciplinary treatment protocols for patients Plans Are Not Set in Stone
with osteosarcoma and Ewing sarcoma typically
stipulate that surgery is performed as soon as hae- One of the most common pitfalls in the multidis-
matological recovery can be expected and that ciplinary treatment of patients with pelvic bone
postoperative chemotherapy should commence sarcomas is the lack of flexibility that is some-
7–14 days after the surgical treatment. In reality, times shown after multidisciplinary treatment
though patients tend to experience delays when plans have been decided on the following initial
moving from one discipline to another, especially interdisciplinary case discussion at a tumour
if medical and surgical treatments take place in board, as the involved physicians may sometimes
different hospitals. Additionally, patients with lose sight of the whole picture and only focus on
pelvic tumours undergoing surgery are at a high their part of the plan. For example, a 42-year-old
risk for postoperative complications, which may man presented with a localized pelvic leiomyo-
further delay the adjuvant treatment. However, sarcoma of bone affecting the cranial part of the
emerging evidence suggests that such delays may ilium bone and the adjacent sacral bone
have a negative impact on patients’ prognosis. An (Fig. 22.6a). The interdisciplinary tumour board
as yet unpublished analysis of data of the Euro- recommended neoadjuvant chemotherapy, fol-
EWING99 trial demonstrated that patients with lowed by surgical treatment involving a type I/IV
localized disease and an interval between surgery resection of the ilium and the affected ipsilateral
and adjuvant chemotherapy longer than 16 days sacrum and adjuvant chemotherapy. The MRI
had significantly poorer overall and event-free report after preoperative chemotherapy showed a
survival probabilities compared with patients with partial tumour response and the patient was
shorter intervals [9]. This finding creates a chal- referred for surgical treatment, which was
a b
Fig. 22.6 Male, 42 years, with a leiomyosarcoma of the right ilium bone (a), and a new skip metastasis (b) in the pubic
bone under treatment
a ccordingly planned. However, the musculoskel- 4. Casali PG, Abecassis N, Aro HT, et al. Soft tissue
and visceral sarcomas: ESMO-EURACAN Clinical
etal radiologist presenting the MRI in the preop- Practice Guidelines for diagnosis, treatment and fol-
erative tumour board also noted a new, previously low-up. Ann Oncol. 2018;29(Supplement_4):iv268–9.
undescribed lesion in the ipsilateral pubic bone, 5. Casali PG, Bielack S, Abecassis N, et al. Bone sarco-
which had been presumably missed by the previ- mas: ESMO-PaedCan-EURACAN Clinical Practice
Guidelines for diagnosis, treatment and follow-up.
ous physicians who focused on the extension of Ann Oncol. 2018;29(Supplement_4):iv79–95.
the local tumour (Fig. 22.6b). Biopsy confirmed 6. Hesla CA, Tsagkozis P, Jebsen N, et al. Improved
the diagnosis of a skip metastasis, and further prognosis for patients with Ewing sarcoma in the
two cycles of chemotherapy were applied to rule sacrum compared with the innominate bones. The
Scandinavian sarcoma group experience. J Bone Joint
out further disease progression prior to surgical Surg Am. 2016;98:199–210.
treatment, which now involved a complete inter- 7. Which Factors Are Associated with Local Control and
val hemipelvectomy. Survival of Patients with Localized Pelvic Ewing’s
In conclusion, a multidisciplinary approach to Sarcoma? A Retrospective Analysis of Data from the
Euro-EWING99 Trial. Andreou D, Ranft A, Gosheger
pelvic tumours is strongly advised. Otherwise, to G, Timmermann B, Ladenstein R, Hartmann W, Bauer
apply the space flight comparison again, you S, Baumhoer D, van den Berg H, Dijkstra PDS, Dürr
might not end on the moon but in the marshland HR, Gelderblom H, Hardes J, Hjorth L, Kreyer J,
beside the space centre. Kruseova J, Leithner A, Scobioala S, Streitbürger A,
Tunn PU, Wardelmann E, Windhager R, Jürgens H,
Dirksen U; GPOH-Euro-EWING99 consortium. Clin
Orthop Relat Res. 2020 Feb;478(2):290–302. https://
References doi.org/10.1097/CORR.0000000000000962.
8. Whelan J, Le Deley MC, Dirksen U, et al. High-
1. Fong K. Moon landing: space medicine and the legacy dose chemotherapy and blood autologous stem-
of project Apollo. Lancet. 2019;394(10194):205–7. cell rescue compared with standard chemotherapy
2. King MJ. Apollo 13 creativity: in-the-box innovation. in localized high-risk Ewing sarcoma: results of
J Creat Behav. 1997;31(4):299–308. Euro-E.W.I.N.G.99 and Ewing-2008. J Clin Oncol.
3. Subramanian P, Kantharuban S, Subramanian V, 2018;36:3110–9.
Willis-Owen SA, Willis-Owen CA. Orthopaedic sur- 9. Dirksen U, Ranft A, Baumhoer D, et al. Do treatment
geons: as strong as an ox and almost twice as clever? delays impact the prognosis in Ewing sarcoma? 2016
Multicentre prospective comparative study. BMJ. Annual meeting of the Connective Tissue Oncology
2011;343:d7506. Society, November 2016, Lisbon, Portugal.
Functional Rehabilitation of Pelvic
Resection and Reconstruction 23
Stefano Masiero, Giacomo Magro,
Mariarosa Avenia, and Francesca Caneva
23.1 Introduction The rehabilitation project is the reference for

every intervention carried out by the rehabilita-
When we talk about rehabilitation, we mean “a set tion team: it defines the objectives in the short,
of measures that assist individuals, who experience medium and long terms, the expected times, the
or are likely to experience disability, to achieve and global and specific outcomes, the expectations
maintain optimum functioning in interaction with and priorities of the patient, takes into account
their environments” (WHO, 2011). the global needs and preferences of the patient, of
The purpose of rehabilitation interventions is his impairments, disabilities, and above all, resid-
to maintain or restore clinical stability, regain ual and recoverable abilities.
independence to perform everyday activities, and Some of the professionals who could be
promote reintegration and social involvement. involved in the rehabilitation project are physiat-
Rehabilitation measures are aimed at achiev- rists, orthopedics, nurses, healthcare assistants,
ing broad outcomes: prevention of the loss of physiotherapists, speech therapists, psycholo-
function, slowing the rate of loss of function, gists, neuropsychologists: they form a multidisci-
improvement or restoration of function, compen- plinary team only focused on the patient.
sation for lost function, and maintenance of cur- This kind of multidisciplinary approach is
rent function. Interventions are performed even more useful when dealing with a complex
according to an “individual rehabilitation proj- patient, like a cancer patient.
ect”, designed on the basis of the patient’s needs When patients receive a cancer diagnosis, they
and recovery potential, the patient’s and his fam- face different challenges, both physical and psy-
ily’s preferences and the resources available. chological. The lives of these people are totally
upset in a short time, the condition of indepen-
dence and autonomy in daily activities is lost,
S. Masiero (*) their social and family position changes, motor
Rehabilitation Unit Department of Neuroscience, disabilities prevent them from performing simple
University General Hospital of Padua, Padova, Italy
actions, such as washing and being tidied up, eat-
Physical Medicine and Rehabilitation School, ing, performing postural steps, transferring from
University of Padua, Padova, Italy
e-mail: [email protected] the bed to the chair and vice versa, climbing
stairs, and walking. All of this leads to a drastic
G. Magro (*) · M. Avenia (*) · F. Caneva (*)
Physical Medicine and Rehabilitation School, reduction in the quality of life, also in relation to
University of Padua, Padova, Italy the patient’s demand of functionality.

248 S. Masiero et al.
The increase in cancer survivors determines tion of mechanical prostheses, often involves the
an increase in the demand for functionality: it partial or complete removal of the stabilizing
implies not only the correction of deformity or muscles of the pelvis, the flexor-extensors and
disability, but also their prevention and some- the abductors of the thigh. The resection of the
times the need for episodic rehabilitation for life. head insertions of these muscles will clearly
During the clinical assessment, the physiatrist result in a loss or a lack of function in the follow-
evaluates organ functions and disabilities, physi- ing phases.
cal, cognitive and behavioral impairments, social The functional complications most often
participation and quality of life, family and social observed after a pelvis resection are diverse and
environment surrounding the patient. include: presence of moderate-strong pain espe-
As a result of the medical examination, the cially in the first post-operative days, edema of
rehabilitation specialist can prescribe motor reha- soft tissue that can affect the ipsilateral lower
bilitation, physical therapies in support of the limb, reduction of the range of joint movement
motor and functional rehabilitation, treatment of (ROM) in flexion-extension and abduction of the
pain syndromes, training in the use of prostheses, ipsilateral limb, a decrease in the muscle strength
orthotics (hip orthosis and reclining wheelchair, (ilio-psoas, quadriceps, gluteus) also correlated
pelvic-thigh brace) and assistive technologies, to the period of time spent in unloading before
rehabilitation with the aid of robotized assis- surgery, poor motor control (alteration of pro-
tance, rehabilitation of organ functions and activ- prioception) with the presence of compensatory
ities aimed at achieving autonomy. mechanisms, a discrepancy in the limb length
Moreover, the educational and informative (heterometry due to prosthesis length) and neuro-
activities carried out by an interprofessional team pathic pain, which in turn can affect daily activi-
are of great importance such as: training of the ties [1].
patient and their family on how to manage dis- The factors that influence rehabilitation in this
abling issues and how to use prostheses, orthotics type of patient are the complexity of surgery (sur-
and assistive technologies, regular informational gical demolition, tumor extension, etc.), the type
meetings with the patient and their family, and of prosthesis, the residual structures to be recov-
involvement of the patient and their family in the ered, the functional demands and the quality of
development and updating of the rehabilitation life expected.
path and programs, giving information and Radiation therapy for bone cancer can cause
advice to social workers, teachers, co-workers stiffness and less ability to move joints, as well as
and anyone who could be involved in the man- loss of muscle strength. A physical therapist can
agement of the disabling issues and the condition teach exercises to help keep joints and muscles
of the patient. healthy and to work properly (also before radia-
tion therapy) and it may be necessary for a long
time after the end of the treatment.
23.2 Rehabilitation Finally, the treatment of the surgical wound
in Musculoskeletal should not be overlooked: surgical wounds in
Oncologic Disease these patients are often more extensive than those
associated with traditional prostheses due to the
Patients with bone cancer can present general highly invasive surgery required, so manual
problems common to all cancers, such as damage debridement may be required [2].Indeed, onco-
from chemotherapy, radiation damage, disuse, logical resection requires large resections, which
asthenia and psychological problems; on the also includes a portion of the soft tissue cover not
other hand, organ-specific problems due to the involved as a surgical margin.
direct involvement of vascular, nerve, bone and Furthermore, it may be necessary to remove or
muscle structures can be seen. A large pelvic sur- repair the neuro-vascular bundle nearby, so a
gery, in addition to bone resection and implanta- complete assessment of the neuro-motor loss
23 Functional Rehabilitation of Pelvic Resection and Reconstruction 249
would be needed to plan dynamic strength train- adjustments of flexion/extension and abduction/
ing and external support requirements. Partial or adduction of the hip, giving greater safety to the
complete loss of the joint capsule and dynamic patient and allowing an early ambulation.
stabilizers of the hip joint during tumor resection To achieve a truly complete recovery and
can leave the hip joint vulnerable to dislocation. effective motor control, the patient needs to re-
This can be enhanced with certain combination of establish an effective proprioceptive sensitivity.
movements, if these joint movements are allowed The computerized proprioceptive platform
beyond a certain limit. This restriction depends (Fig. 23.1) is an absolutely current and techno-
largely on the surgical approach. The posterolat- logically advanced tool for rehabilitation. It is a
eral approach is more common in the limb sal- pivoting platform (the degrees of which can be
vage surgery (LSS) of this site. Hip rotations, decided on the basis of the patient’s condition)
particularly internal rotation, flexion greater than connected to a computer that allows certain
60° and adduction of the hip joint should be pre- movements to be performed in order to fully
vented up to 6 weeks. These movement limita- recover the proprioception, based on a personal-
tions could be achieved using the hip abduction ized training program.
pillow/bracing and the de-rotation splint. The patient can move in an orthostatic posi-
Before patients are discharged from the hospi- tion or sit according to the adopted strategy and
tal, it becomes imperative to train them regarding objectives, breech or mono-breech based on the
the transfer from the bed in the supine position to type of injury and at the current stage of rehabili-
standing (to lie supine and to sit on a chair/com- tation. The computer screen shows a very
fortable in the initial phase of rehabilitation). important visual feedback for the patient’s true
Knee joint mobilization should be initiated early
from the edge of the bed, with the hip joint well
supported or laterally with the cushions between
the legs. Any restriction of the knee joint interval
would adversely affect the overall function since
the function of the hip joint of the ipsilateral leg
has already been impaired. From a biomechani-
cal viewpoint, lateral pelvic stability is provided
by a hip abductor [3]. The reconstruction of
abductor muscles, which provide hip joint stabil-
ity, has been reported as a key to achieve a better
functional outcome after proximal femur resec-
tion [4]. On the other hand, the pelvic site of the
bony attachment of the abductor muscles is typi-
cally resected without abductor reconstruction in
patients undergoing pelvic tumor resection.
Using pelvic-thigh brace, particularly indi-
cated in prosthetic surgery, helps to treat post-
surgical condition, thanks to the immobilization
due to the thigh support. This peculiar brace pres-
ents a greater prolongation on the medial part of
the thigh, which, by resting on the medial con-
dyle of the femur, prevents unfavorable rotations
of the hip joint. This gives great security to the
Fig. 23.1 The computerized proprioceptive platform is a
patient, stabilizes the hip, limits mobility and pivoting platform connected to a computer that allows
allows the bipedal station and early intercourse. certain movements to be performed in order to fully
In addition, it allows different and progressive recover the proprioception
Fig. 23.2 Treadmills
equipped with digital
mirror able to detect
every single angular
movement of the runner
perception. Only in this way does he really real- 23.3 Assessment of Functional
ize what kind of movement, error and speed of Status
correction execution he is performing.
It is now scientifically consolidated that the The variability of the factors involved and the
control of our reflection helps us to improve our need to customize the rehabilitation treatment is
movement through the “continuous feedback the basis of the rehabilitation process in onco-
system” method. The patient experiences a great logic patients.
advantage in reconstructing his motor map, It is, therefore, necessary to try to define a
thanks to the help of the reflected image with pro- common line of therapeutic intervention, use the
gressive stimulations. most accurate possible assessment scales, which
Nowadays, there are particular treadmills include measures of strength, mobility, general
equipped with a 3D camera (Fig. 23.2) that trans- state of psycho-physical health and the quality of
form the classic mirror in to a digital mirror, life of the patients.
which has the ability to detect every single angu- Different tools have been employed to assess
lar movement of the runner, with precision and the functional status of patients, the most vali-
reliability. Patients can immerse themselves in dated in literature are:
virtual environments and get a real-time stimulat-
ing feedback, both of postural and symmetry type • Range of Motion (ROM): measured with a
of the supports on the ground. simple hand-held goniometer to obtain both
The latest studies [3], with the incorporation of passive and active ranges of motion, [3] pro-
objective and validated measures of the function, vides a measure of joint mobility. It may
indicated that the patient undergoing LSS obtained indicate the presence of partial or complete
higher scores than those with amputations [4]. stiffness (Fig. 23.3).
This suggests the need to examine the post- • Strength can be measured under both isometric
surgical functional outcomes and plan a person- and isokinetic conditions by utilizing iso-
alized treatment in order to provide the best kinetic dynamometers [7]; strength measure-
functional results and maximum achievable inde- ments of the unaffected limb may be used as a
pendence [5, 6]. reference.
velocity, cadence, ROM, single-limb sup-

port time, swing and stance times, and dou-
ble limb support time, and if it is associated
with a dynamic electromyography, we can
study patterns of muscle activity during
gait.
–– -Oxygen consumption during gait provides
an objective measure of gait performance.
Individual gait efficiency may be calcu-
lated quantitatively by measuring the oxy-
gen consumed per unit of distance travelled
per kilogram of body mass.
Fig. 23.3 Physical evaluation of ROM and strength
23.4 Rehabilitation Protocols

• Subjective functionality scale such as: the in Pelvic Resection
Musculoskeletal Tumor Society scoring sys-
tem (MSTS) [8]. It is a disease-specific instru- Although LSS for malignant bone tumors is con-
ment to determine the physical and mental sidered the treatment of choice, rehabilitation
health for patients with extremity sarcoma, guidelines for this kind of patients have yet to be
which is used to evaluate six items, including formally established. Detailed guidelines for this
pain, function, emotional acceptance, use of patient population are stratified by the anatomical
any external support, walking ability, and gait position, the type of prosthesis, the width of resec-
alteration tion, in order to make the rehabilitation protocols
• In addition, the Toronto Extremity Salvage applicable and reproducible. With regard to the
Score (TESS) is widely used for the functional pelvic and proximal reconstruction of the femur, it
assessment of patients following surgery for can be differentiated according to the involvement
musculoskeletal tumors. of the acetabulum or its exclusion in the periace-
• Health status survey short form-36 (SF-36v2) tabular areas and in the non-acetabular areas (iliac
is also frequently used as a measure of health bone, ischium, pubis and sacrum).The protocol
status. It consists of eight sections that evalu- also varies according to the type of reconstruction
ate vitality, physical functioning, bodily pain, performed, which can be: with prosthesis, with
general health perceptions, physical role func- bone graft, with prosthesis and graft [9].
tioning, emotional role functioning, social
role functioning, and mental health.
• FMA (Functional Mobility assessment) is a 23.5 Rehabilitation Program
self-report outcomes tool designed to measure
the effectiveness of wheeled mobility and The rehabilitation program should be designed to
seating interventions for patients with address:
disabilities.
• Energy expenditure testing to assess differ- 1. Pain. Since cancer patients generally experi-
ences between groups and provide a global ence multiple concurrent pain syndromes,
measure of functional performance for com- adequate pain control is an absolute requisite
paring a patient’s status before and after an for successful rehabilitation. Pain control
intervention: might require the integrated use of anticancer
–– -Gait analysis is used to assess and treat treatments, agents from multiple analgesic
individuals with conditions affecting their classes, interventional techniques, topical
ability to walk; it can be used to measure agents, manual approaches, and modalities.
The unique disease context, in which cancer It is known to be the most common symp-
pain develops, distinguishes it from many tom experienced by cancer patients. It affects
other pain-associated diagnoses managed by almost 75% of these patients and 60% of them
physiatrists. The majority of cancer pain is due consider it more disabling than pain. CRF is
to tumor effects, and for this reason, disease- especially associated with chemotherapy and
modifying, anticancer therapy plays a critical drug therapy cycles and a quite aggressive
role in pain management. For example, radio- incidence. In fact, a majority of patients in
therapy often offers a definitive and effective active treatment rate their fatigue as “severe”
means of controlling pain associated with or 7 or more on an 11-point numerical rating
symptomatic and uncomplicated bone metas- scale.
tases. Bone metastases occur in 60–84% of 4. Education of the patient and caregiver to pro-
patients with solid tumors. Pain intensity does vide continuity of assistance. Learning of
not correlate with the number, size, or location basic nursing techniques and functional reha-
of bone metastases. Bone pain is particularly bilitation (postural steps, assisted walking,
relevant to physiatrists because recruiting and maintenance exercises).
muscles that act on or loading affected struc-
tures can precipitate severe pain [10].
Pharmacological and instrumental physical 23.6 Rehabilitation Phases
therapy to manage pain can include:

(a) Drugs: paracetamol, NSAIDS, opioids, The rehabilitation program is distinguished by
antidepressants, anticonvulsants. three phases:
(b)
Transcutaneous Electrical Nerve
Stimulation(TENS) with rectangular cur- 1. Pre-operative: absolute proscription of the

rents pulsed from 30–150 μs at a low fre- load from the moment of diagnosis, education
quency (10–150 Hz); in walking with brachial sticks. Rehabilitation
(c) Low level laser therapy (LLLT) with anal- even before starting primary cancer therapy
gesic and decontracting effects; and surgery, such as crutch muscles strength-
(d) Massotherapy for the reduction of muscle ening, could be of great benefit in the post-
contraction and stasis edema. This should treatment functional outcome.
be applied only 1 month after the end of 2. Early post-operative: simple isometric con-

chemotherapy treatment, because there is tractions with the operated limb, contralateral
a risk of promoting the spread in the blood limb mobilization for vasomotor and anti-
or in the lymphatic circulation of meta- thromboembolic purposes, and diaphragmatic
static components. respiratory rehabilitation. Slow and concentric
2. ROM recovery, muscle tone, mobilization and exercises are preferred to stimulate the slow
toning from the immediate post-operative type I fibers, which undergo a greater volu-
phase, first in discharge and then in treatment, metric and functional reduction from rest (up
postural and behavioral hygiene, compensa- to 30% after 5 weeks). Walking with overflow
tory strategies, and use of guardians. load.
3. Contrast of Cancer Related Fatigue syndrome 3. Advanced post-operative: granting of the pro-
(CRF). The National Comprehensive Cancer gressive load and, if applicable, complete; the
Network defines CRF as “an unusual, persistent, use of aids to stimulate walking and the
subjective sense of tiredness related to cancer or patient’s “fear of fracture”; co-contraction
cancer treatment that interferes with usual func- exercises of the antagonist muscles of the
tioning”. This is an integral part of the patient’s lower limb promote stability and load transfer;
symptoms, disproportionate to the degree of proprioceptive exercises; muscle strengthen-
activity and not reducible with sleep or rest. ing exercises. Two daily sessions (Table 23.1).
Table 23.1 Example of rehabilitation protocol in pelvic or proximal femoral resection

1 Step (week 1–4) 2 steps (week 5–8) 3 step (week 9–12)
• Extension braceactive knee and • Release of the brace at 45° from • 60th day, brace unblocked at 90°
ankle joint mobilization and the 30th day, then at 60° on the and progressive load
active limb contralateral 45th–50th day, • From 90th day the brace is removed
• Quadriceps isometric • Mobilization. Passive and and the full load is granted
reinforcement and ipsilateral active assisted hip flexion • Muscle reinforcement with
sural triceps for maintaining the extension under the pain concentric exercises and in isometric
tone and the trophism threshold seal
• Distal mobilization exercises of • Stimulation with electrotherapy, • Aerobic exercises
the treated limb to prevent preparatory to walking. • Functional exercises in preparation
venous and lymphatic pooling. • Gradual replacement of the for return home, including stairs
• Self-mobilization and walking frame by elbow training, re-education in the
strengthening exercises of the crutches: Gait showed slight activities of daily living and the
body districts not involved in the abduction of the hip (which postural movements involved in
surgery gradually regressed) entering and leaving a car.
23.7 Conclusion Vincent S. Paramanandam et al. [2] tried to

establish some fundamental concepts on which
The rehabilitation of orthopedic oncology rehabilitative treatment should be based for this
patients needs a global management by a multi- type of patients, first of all, the presence of a
disciplinary and expert team that defines a project rehabilitation team of cooperating professionals
and a rehabilitation program based on their spe- for the whole period of hospitalization (protocols
cific characteristics. last more or less 60 days).
Despite the fact that early physical rehabilita- Another essential concept is training patients
tion is the key to achieve good functional out- and their caregivers on how to use specific aids
come and quality of life after LSS, rehabilitation (hip orthosis and reclining wheelchair, and
techniques following LSS are largely neither pelvic- thigh brace) to provide support to the
tested nor documented in detail [11]. limb.
Most of the resection and oncological recon- New technologies, such as computerized
struction varies from one individual to another proprioceptive platforms or the treadmill asso-
even in a particular site and requires a personal- ciated with virtual reality, can greatly help
ized rehabilitation protocol to design an individ- patient in motor recovery, making him an active
ual rehabilitation program. part of the rehabilitation process and setting
In a recent paper, Shehadeh et al. [12] reported short-term goals that transmit confidence in his
that, following a standardized rehabilitation pro- own means.
tocol produced an improved functional outcome, The role of the physiatrist, in addition to coor-
even though their conclusion is based on a small dinating the multidisciplinary team, is to guide
observational study with heterogeneous popula- the patient through a long and difficult path and a
tion who received different types of LSS for dif- physical and psychological challenge. This also
ferent anatomical sites. includes the assessment of the new impairments,
Although limb savage surgery for primary the attempt to recover the normal motor function,
malignant tumors has improved because of a sig- and in many cases, the acceptance of the disabil-
nificant progress in surgical techniques and endo- ity with the learning of adaptation strategies.
prosthetic design and manufacture, without It is essential for rehabilitation team to under-
optimal peri and postoperative physical rehabili- stand the real needs of the patient and set realistic
tation, achieving the desired quality of life may goals, in order to achieve the highest possible
not be feasible. quality of life.
References receiving a large-resection hip prosthesis because

of a phosphaturic mesenchymal tumor. Clin Pract.
2015;5:814.
1. Renard AJ, Veth RP, Schreuder HW, van Loon CJ,
8. Enneking WF, et al. A system for the functional evalu-
Koops HS, van Horn JR. Function and complications
ation of reconstructive procedures after surgical treat-
after ablative and limb-salvage therapy in lower extrem-
ment of tumors of the musculoskeletal system. Clin
ity sarcoma of bone. J Surg Oncol. 2000;73(4):198–205.
Orthop. 1993:241–6.
2. Paramanandam VS, Daptardar AA, Gulia
9. Mei J, Zhu X-Z, Wang Z-Y, Cai X-S. Functional out-
A. Rehabilitation following limb-salvage surgery
comes and quality of life in patients with osteosar-
in sarcoma. J Bone Soft Tissue Tumors. 2016;2(2):
coma treated with amputation versus limb salvage
19–21.
surgery: a systematic review and meta-analysis. Arch
3. Marchese VG, et al. Relationship among range of
Orthop Trauma Surg. 2014;134(11):1507–16.
motion, functional mobility, and quality of life in
10. Capanna R, Pasquetti P, Poli R, Nicodemo ML,

children and adolescents after limb-sparing surgery
Zambelan G. Libro bianco della riabilitazione—I
for lower-extremity sarcoma. Pediatr Phys Ther.
bisogni riabilitativi nei sarcomi primitivi dell’osso.
2006;18:238–44.
Project H.O. C.U.R.A.—Health Organization of
4. Ginsberg JP, et al. A comparative analysis of func-
Cancer Units for Rehabilitation Activities, Ministero
tional outcomes in adolescents and young adults with
Italiano del Lavoro, della Salute e delle Politiche
lower-extremity bone sarcoma. Pediatr Blood Cancer.
Sociali; 2008. p. 140–146.
2007;49:964–9.
11. Nagarajan R, Neglia JP, Clohisy DR, Robison

5. Lane JM, Christ GH, Khan SN, Backus
LL. Limb salvage and amputation in survivors
SI. Rehabilitation for limb salvage patients: kine-
of pediatric lower-extremity bone tumors: what
siological parameters and psychological assessment.
are the long- term implications? J Clin Oncol.
Cancer. 2001;92(4 Suppl):1013–9.
2002;20:4493–501.
6. Zati B, Ruggieri M. La Riabilitazione dei pazienti
12. Shehadeh A, El Dahleh M, Salem A, Sarhan Y, Sultan
operati per neoplasie primitive dell’ osso. Bologna,
I, Henshaw RM, Aboulafia AJ. Standardization of
Italy: Timeo; 2017.
rehabilitation after limb salvage surgery for sarcomas
7. Lopresti M, Daolio PA, Rancati JM, Ligabue N,
improves patients’ outcome. Hematol Oncol Stem
Andreolli A, Panella L. Rehabilitation of a patient
Cell Ther. 2013;6(3–4):105–11.

Surgery of Pelvic Bone Tumors PDF

Uploaded by

Copyright:

Available Formats

Surgery of Pelvic Bone Tumors PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Surgery of Pelvic Bone Tumors PDF

Uploaded by

Copyright:

Available Formats

What are some of the surgical approaches discussed in the document?

What are some of the surgical approaches discussed in the document?

What are some of the common tumor types mentioned?

What are some of the common tumor types mentioned?

Surgery

Surgery of Pelvic Bone

ISBN 978-3-030-77006-8 ISBN 978-3-030-77007-5 (eBook)

© Springer Nature Switzerland AG 2021

1 Surgical Approaches in Pelvic Bone Tumors �������������������������������� 1

13 Navigation in Pelvic Surgery���������������������������������������������������������� 135

© Springer Nature Switzerland AG 2021 1

In order to obtain a Type I resection, only the first

2.1 Introduction in the selection and management of patients for

© Springer Nature Switzerland AG 2021 15

2.5 Adjuvant Treatments While initially associated with high morbidity,

10. Gordon-Taylor G, Wiles P, Patey D, et al. The

3.1 Introduction the imaging characteristics of the most common

© Springer Nature Switzerland AG 2021 23

3.2.2 Metastases myeloma and osteosarcoma [13], and pelvis is

criteria for the diagnosis of multiple myeloma. Lancet

10. Angelini A, Calabrò T, Pala E, Trovarelli G, Maraldi 23.

4.1 Introduction with the literature review and authors’ institutional

© Springer Nature Switzerland AG 2021 33

of spindle cells, pleomorphic cells, epithelioid 4.10 Langerhans Cell

Fibrosarcoma is a very specific diagnosis to

4.14 Fibrous Dysplasia References

5.1 Introduction opposed to the malignant tumors which were

© Springer Nature Switzerland AG 2021 45

Intervention is reserved for symptomatic tral, lucent nidus surrounded by a sclerotic

5.4.1 Osteoid Osteoma 5.4.2 Osteoblastoma

6.1 Epidemiology, Presentation, tion by the tumor [8]. Depending on localization,

© Springer Nature Switzerland AG 2021 55

References drosarcoma patients. Eur J Nucl Med Mol Imaging.

7.1 Introduction Lastly, the histologic subtypes found in pelvic

7.4 Medical Treatment for Pelvic

Until the application of chemotherapy, the sur-

External hemipelvectomy or hindquarter amputa- 7.5.5 Computer Navigation

Internal hemipelvectomy is the surgical resection

7.6 Summary 12. Wang J, Sun M, Liu D, Hu X, Pui MH, Meng Q,

8.1 Introduction 8.2 Histology

© Springer Nature Switzerland AG 2021 73

tase polymerase chain reaction (RT-PCR) for

Primary EFT may present throughout the axial

The radiographic appearance of ES is variable,

marrow (most commonly spine), while lymph

The presence of metastatic disease at presenta- 8.8 Biopsy

8.12 Radiation Therapy approach similar to that of a primary lesion but

large associated soft tissue component. He under-

bicin, and cyclophosphamide. Surgical planning

A 2-year-old previously healthy female presented

of the mass was performed. Histology demon-

9.1 Introduction that more of the pelvic surgery indications are

E. J. Ortiz-Cruz (*) 1. Patient with no previous history of cancer.

© Springer Nature Switzerland AG 2021 87

9.3.3.1 Types of Pelvis Resection treated conservatively by chemotherapy, hor-

3. Biermann JS, Chow W, Reed DR, Lucas D, Adkins

10. Rosenthal D, Callstrom MR. Critical review and

10.1 Introduction (History) external hemipelvectomies make up 0.5% of

10.2 I ndication (Bone, Soft Tissue,

1 Surgical Approaches in Pelvic Bone Tumors �� 1

13 Navigation in Pelvic Surgery�� 135

2.1 Introduction in the selection and management of patients for

2.5 Adjuvant Treatments While initially associated with high morbidity,

3.1 Introduction the imaging characteristics of the most common

3.2.2 Metastases myeloma and osteosarcoma [13], and pelvis is

4.1 Introduction with the literature review and authors’ institutional

of spindle cells, pleomorphic cells, epithelioid 4.10 Langerhans Cell

4.14 Fibrous Dysplasia References

5.1 Introduction opposed to the malignant tumors which were

5.4.1 Osteoid Osteoma 5.4.2 Osteoblastoma

6.1 Epidemiology, Presentation, tion by the tumor [8]. Depending on localization,

7.1 Introduction Lastly, the histologic subtypes found in pelvic

7.4 Medical Treatment for Pelvic

External hemipelvectomy or hindquarter amputa- 7.5.5 Computer Navigation

7.6 Summary 12. Wang J, Sun M, Liu D, Hu X, Pui MH, Meng Q,

8.1 Introduction 8.2 Histology

The presence of metastatic disease at presenta- 8.8 Biopsy

8.12 Radiation Therapy approach similar to that of a primary lesion but

9.1 Introduction that more of the pelvic surgery indications are

9.3.3.1 Types of Pelvis Resection treated conservatively by chemotherapy, hor-

10.1 Introduction (History) external hemipelvectomies make up 0.5% of

10.2 I ndication (Bone, Soft Tissue,

10.5.1 Standard Hemipelvectomy 10.5.1.1 Surgical Technique

10.6 Soft-Tissue Coverage

11.1 Introduction 11.2 Epidemiology

11.3 Surgical Indications

12.1 Introduction to spinopelvic, sacral, bulk allograft, and proxi-

Using intraoperative navigation, pelvic oste- 12.7 Proximal Femoral

12.8 Intraoperative Navigation With advances in our understanding of anatomy,

13.1 Introduction 13.2 W

13.3 History of Navigation

14.1 Introduction for a spinopelvic fixation, due to the special ana-

15.1 Introduction radiographic findings in early stages are not diag-

15.3 Biopsy 15.4 Preoperative Management

include MR angiography or intravenous pyelogra- 15.5 Surgery and Resection Types

15.6 External Hemipelvectomy