ORIGINAL RESEARCH

published: 24 September 2019

doi: 10.3389/fonc.2019.00946

Surgical Treatment Options for Giant

Cell Tumors of Bone Around the
Knee Joint: Extended Curettage or
Segmental Resection?
Hongbo He 1 , Hao Zeng 1 , Wei Luo 1 , Yupeng Liu 1 , Can Zhang 1 and Qing Liu 1,2*
1
Department of Orthopaedics, Xiangya Hospital, Central South University, Changsha, China, 2 Department of Spine Surgery,
The Second Xiangya Hospital, Central South University, Changsha, China

Aims: This study aimed to compare and evaluate the oncological and functional
prognosis of two surgical approaches for giant cell tumor of the bone (GCTB) around
the knee joint and provide worthy suggestion for clinical treatment.
Edited by:
Zongbing You, Patients and Methods: This study included 93 patients, who were divided into
Tulane University, United States
the extended curettage (EC) group and segmental resection (SR) group. Relevant
Reviewed by:
Qingsong Zhang,
preoperative and postoperative data were collected, oncological and functional
Wuhan Pu’ai Hospital, China prognosis were evaluated, and postoperative complications of the two groups
Jiacan Su,
were analyzed. Local recurrence was assessed via clinical and radiological tests.
Second Military Medical
University, China Functional prognosis was evaluated using the Musculoskeletal tumor Society (MSTS)
*Correspondence: scoring system.
Qing Liu
[email protected]
Results: The EC group had 69 patients; it included 57 primary cases and 12 recurrent
cases. The SR group had 24 patients (12 men and 12 women; mean age, 34.9 years),
Specialty section: including 15 primary cases and 9 recurrent cases. In this study, six cases (6.5%; EC
This article was submitted to
group, 5 cases; SR group 1 case) recurred within 18 months postoperatively. There was
Surgical Oncology,
a section of the journal a significant difference in the mean MSTS score between the two groups (p < 0.001).
Frontiers in Oncology Nononcologic complications occurred frequently in the EC group than in the SR group
Received: 10 July 2019 (28.0 vs. 16.7%), but no complications had serious consequences, and the functional
Accepted: 09 September 2019
Published: 24 September 2019
prognosis was not affected.
Citation: Conclusion: EC and SR for GCTB around the knee joint can achieve satisfactory
He H, Zeng H, Luo W, Liu Y, Zhang C
oncological prognosis, but we should individually select the most suitable surgical
and Liu Q (2019) Surgical Treatment
Options for Giant Cell Tumors of Bone method according to Campanacci grade, age, and long-term complications of patients
Around the Knee Joint: Extended and consider the functional prognosis to ensure excellent oncological prognosis.
Curettage or Segmental Resection?
Front. Oncol. 9:946. Keywords: giant cell tumors of bone, extended curettage, segmental resection, oncological prognosis, functional
doi: 10.3389/fonc.2019.00946 prognosis

Frontiers in Oncology | www.frontiersin.org 1 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

INTRODUCTION surgical treatment with limb salvage, and postoperative follow-

up of more than 24 months with integrated data. According
A giant cell tumor of bone (GCTB) is a primary bone tumor to the above criteria, a total of 131 patients with GCTB
with potential invasion, local recurrence, and low probability of located around the knee joint were retrieved. Among them, 35
distant metastasis (1). Studies have shown that GCTB accounts patients lost follow-up, two had amputation due to malignant
for 5–7% of all primary bone tumors and 20% of all benign bone changes, and one received knee arthrodesis. Finally, 93 patients
tumors (2). Its incidence in China was about 14–20%, which was were enrolled in the present study (Figure 1). Radiography,
higher than 5–8% in other eastern countries (3). GCTB tends to computed tomography (CT), and magnetic resonance imaging
occur in people aged 20–40 years, accounting for 60–75% of all (MRI) were performed preoperatively to determine the extent
patients (3), and GCTB occurs in the meta-epiphyseal area of of tumor invasion and pulmonary metastasis. Lesions were
the limbs and in the around knee joint at around 50–65% of the graded according to Campanacci classification of imaging (15).
whole body, especially in the distal femur and proximal tibia. Meanwhile, preoperative puncture biopsy was performed to
GCTBs grow in an expansive manner and easily penetrate confirm the diagnosis of GCTB. Patients were divided into the
the cortex of the bone or even cause pathological fracture. EC group and SR group.
Although they rarely expand into the articular cavity, they This study was approved by Xiangya Hospital Ethics
invade the subchondral bone, which seriously affects knee joint Committee, and written informed consents were obtained from
function (4). These factors lead to an embarrassing situation the patients or their legal guardians.
during treatment because the knee joint is the main load-
bearing joint of the lower limbs and has high functional
requirements. The therapeutic purpose of GCTB around the Procedure
knee joint is to reduce its recurrence rate and maximize the EC was performed as follows. The fenestration from the eccentric
recovery of joint function, while reconstructing the integrity of cortex of the lesion was sufficiently large to avoid opening the
bone structure and articular surface, as well as obtaining normal joint capsule (Figure 2A). To protect peripheral tissues, wet
biomechanics and preventing the occurrence of long-term saline gauze was used before fenestration to reduce tumor cell
osteoarthritis (5–7). implantation. Different types of curettes were used to scrape
There is still controversy about the surgical treatment of the tumors thoroughly, and the cavity was rinsed with sterilize
GCTB in the around knee joint. How to achieve a balance water (Figure 2B). Then, the residual bone ridges in the cavity
between completely removal of tumors to reduce recurrence and were grounded with high-speed burr (Figure 2C); the range of
preservation of knee joint function as much as possible was grinding was 10 mm for the normal cancellous bone and 1 mm
the linchpin for clinicians to balance. The surgical treatment for the cortical bone (only residual bone ridges adjacent to
of GCTB in the around knee joint mainly includes curettage the articular surface were removed). Afterwards, an electrotome
and bone grafting (3, 8), extended curettage (EC) and cement was used to cauterize the cavity wall (Figure 2D), and the
filling (1, 9), segmental resection (SR) and artificial prosthesis blackened bone was scraped again (Figure 2E). Iodine tincture
reconstruction (6, 10). Although these methods have achieved (10% concentration) was applied meticulously using a surgical
certain results in the treatment of GCTB, some problems cotton ball and left for 3 min (Figure 2F). After which, the cavity
occur, such as local recurrence (11, 12), secondary osteoarthritis was irrigated with sterile water again. After thorough curettage,
(1, 13), cartilage surface collapse (14), artificial prosthesis the cavity was filled with cement and allograft. Allografts about
loosening and infection (6, 10), which require deep focus 1 cm thick were transplanted to the side around the articular
and improvement. surface, and the patellar cavity was filled with allograft only
Therefore, this study aimed to analyse the correlation between (Figure 2G). The remaining cavity was filled with cement, and
the choice of surgical treatment for GCTB around the knee internal fixation was performed finally (Figure 2H).
joint and the prognosis of oncology and limb function through SR was performed as follows: The surgical resection margin
a single-center retrospective cohort study to provide a valuable was determined by preoperative T1-weighted-enhanced images.
reference for surgical treatment of GCTB around the knee joint. The tumors were completely resected from normal peripheral
tissues, while the common peroneal nerve and important vessels
of the lower limbs (femoral and posterior tibial arteries) were
PATIENTS AND METHODS protected during the surgery. After resection, bone defects were
Patients repaired with artificial prosthesis followed by soft tissue repair,
Data of 277 GCTB patients who were treated at a single center but the fibula was an exception, where only the lateral ligament
from March 2007 to March 2017 were retrospectively collected. of knee joint was repaired, and the bone defect was not repaired.
The inclusion criteria were as follows: GCTB located in the Patients of the EC group were exempted from weight-
around knee joint, histopathological diagnosis of benign GCTB, bearing for 2 weeks, and they gradually began to perform non-
weight, semi-weight, and full-weight bearing functional exercises
alternately using crutches. Limbs of the patients in the SR
Abbreviations: GCTB, giant cell tumor of bone; CT, computed tomography;
MRI, magnetic resonance imaging; EC, extended curettage; SR, segmental
group were fixed with plaster or braces for 4–6 weeks, and they
resection; MSTS, Musculoskeletal tumor Society; VAS, Visual Analog Scale; gradually began to perform functional exercises, from half-load
K-L, Kellgren-Lawrence. to full load with crutches.

Frontiers in Oncology | www.frontiersin.org 2 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

FIGURE 1 | Flow charts of patients included in this study.

Follow-Up and Evaluation RESULTS

Patients were followed radiographically every 3 months for the
first 2 years after the surgery, every 6 months until the 5th Patients
year, and annually until the 10th year. The radiographs of According to the statistical results of the data (Table 1), the EC
the involved area and CT images of the chest were obtained group included 69 patients (37 men and 32 women), with the
to evaluate cancer prognosis. The prognosis of limb function mean age of 36.3 (range, 17–65) years. The number of involved
was evaluated based on the last follow-up record and those of femur, tibia, and patella were 36, 32, and 1, respectively, and the
recurrent patients were based on the best functional records average length of the lesion was 5.6 cm (range, 2.6–8.8 cm). The
before recurrence. Evaluation tools used were as follows: the preoperative Campanacci grades were I, II, and III in 7, 33, and 29
Musculoskeletal Tumor Society (MSTS) (16) Score was used to cases, respectively. There were 57 primary cases and 12 recurrent
assess function. The visual analog scale (VAS) (17) was used cases (recurrence after initial treatment in other centers). There
for pain evaluation. Osteoarthritic change was evaluated by were 18 preoperative pathological fractures.
the Kellgren-Lawrence (K-L) grading system (18). Rejection, The SR group included 24 patients (12 men and 12 women),
prosthesis loosening, periprosthetic fracture, and infection were with the mean age of 34.9 (range, 17–52) years. In this group,
also recorded. 11 femurs, 7 tibias, and 6 fibulas were examined. The average
length of the lesion was 7.2 cm (4.3–10.2 cm). All cases were
of Campanacci grade III. There were 15 primary cases and 9
Statistical Analysis recurrent cases. There were 14 cases of preoperative pathological
Data were analyzed using SPSS software version 20.0 (IBM fracture and two cases of pulmonary metastasis.
Corp., Armonk, NY), and measurement data were expressed
as mean ± standard deviation. Multivariate and univariate Oncology Prognosis
Cox regressions were used to analyse risk factors of local In this study, six cases (6.5%) of recurrence occurred within 18
tumor recurrence. Continuous variables were compared by months after surgery. There were five recurrence cases in the
one-way analysis of variance, and categorical variables were EC group, including three cases in the femur and two cases
compared by chi-square test. P ≤ 0.05 was considered in the tibia, of which one was far from the articular surface
statistically significant. and the four were around to the articular surface. Preoperative

Frontiers in Oncology | www.frontiersin.org 3 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

FIGURE 2 | Specific implementation steps of extended curettage: (A) Fenestration from the eccentric cortex of the lesion and protecting peripheral tissues with wet
saline gauze. (B) Scraping the lesion thoroughly with different types of curette. (C) Grinding residual bone ridges in cavity with high-speed burr. (D) Cauterization of the
cavity wall with electrotome. (E) Scraping blackened bones and rinsing thoroughly again. (F) Cotton balls with iodine tincture were used to smear the cavity wall for
3 min before rinsing again. (G) Transplantation of allogenic cancellous bone into the subchondral bone at least 1 cm thick. (H) The remaining cavity was filled with
cement, and finally, internal fixation was performed.

TABLE 1 | Patient demographics. TABLE 2 | Prognostic comparative statistics.

General information EC group SR group Variable EC group SR group P-value

Mean age (sd) 36.3 (12.5) 34.9 (9.9) Duration of follow-up (month) 67.8 ± 38.7 70.9 ± 27.7 0.720
Gender, n (%) Mean pre-op VAS (sd) 4.5 (1.9) 5.5 (1.8) 0.033
M 37 (53.6%) 12 (50.0%) Mean post-op VAS (sd) 0.3 (0.5) 1.0 (0.8) 0.000
F 32 (46.4%) 12 (50.0%) Mean MSTS score (sd) 28.2 (1.8) 26.5 (1.4) 0.000
Lesion length (mm, mean ± SD) 5.6 ± 1.2 7.2 ± 1.3 Local recurrence, n (%) 5 (7.2%) 1 (4.2%) 0.597
Lesion location, n (%) Complication, n (%)
Femur 36 (52.2%) 11 (45.8%) Osteoarthritis 6 (8.6%) 0
Tibia 32 (46.4%) 7 (29.2%) Rejection reaction 17 (24.6%) 0
Fibula 0 6 (25.0%) Joint stiffness 5 (7.2%) 4 (16.7%) 0.006
Patella 1 (1.4%) 0 Fracture 1 (1.4%) 0
Campanacci grade, n (%) Reoperation, n (%) 5 (7.2%) 1 (4.2%) 0.597
I 7 (10.1%) 0
II 33 (47.8%) 0
III 29 (42.0%) 24 (100%) situ 11 months after curettage at other centers. We performed
Prior surgery, n (%) 12 (17.4%) 9 (37.5%) SR and artificial prosthesis reconstruction for the first time.
Pathological fracture, n (%) 18 (26.1%) 14 (58.3%) Unfortunately, the patient had a recurrence at the tibia 15 months
after the first operation in our hospital. We re-performed SR
and artificial prosthesis reconstruction for patients after puncture
pathological fracture occurred in two cases, and two cases were biopsy with confirmed GCTB, and the latest follow-up was
transferred from another hospital as local recurrence. All five satisfactory (Figure 3). One of the two patients with preoperative
patients, including three cases with EC and two cases with SR, pulmonary metastases underwent laparoscopic resection, and
were reoperated. No recurrence or metastasis was found at the the other refused to undergo surgery and continued follow-
latest follow-up (Table 2). There was only one recurrence in the up. No patients had secondary pulmonary metastases after
SR group, of which the patient had distal femoral recurrence in operation. We found that the recurrence rate of the EC group

Frontiers in Oncology | www.frontiersin.org 4 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

FIGURE 3 | Typical imaging manifestations of patients with recurrence treated with segmental resection. (a) Anteroposterior radiographs of recurrence after curettage
in other hospitals. (b) Computed tomography (CT) image showing that the lesion had penetrated the bone cortex and involved the intercondylar fossa and the posterior
part of the joint. (c,d) Anteroposterior, lateral, and total length of lower limb radiographs after segmental resection and artificial prosthesis reconstruction. (e) Osteolytic
lesions were found in the proximal tibia 15 months after the operation. (f) CT scan confirmed that the lesion invaded the peripheral tissues. (g,h) Anteroposterior and
total length on the lower limb radiographs after segmental resection was performed again and artificial prosthesis reconstruction (white arrow points to the lesion).

was higher than that of the SR group, but the difference was not follow-up. Pain symptoms improved significantly. Moreover, 84
statistically significant (p = 0.597). Among the disease-related patients (90.3%) returned to their previous jobs (Table 2).
and demographic factors analyzed for their effects on recurrence,
age, sex, lesion location, lesion length, pathological fracture,
recurrence or not, Campanacci grade, etc., appear to have no
Complication
Nononcologic complications occurred frequently in the EC
significant effects on local recurrence (Table 2).
group than in the SR group (28.0% [29/69] vs. 16.7% [4/24]).
In the EC group, six patients had secondary osteoarthritis
(five cases with K-L grade 2 and one case with K-L grade 3).
Limb Function Prognosis Symptoms of osteoarthritis occurred at a mean of 33 months
A significant difference was found in the mean MSTS score after surgery, but fortunately, these patients did not need surgical
between the two groups (EC group, 28.2 points; range, 24–30 treatment for the time being. Seventeen patients developed
points, 95% CI 27.8–28.5; SR group, 26.5 points; range, 27.7– mild rejection within 1 week after operation, and symptoms
28.5 points, 95% CI 0.58–0.94; p < 0.001). All patients in this disappeared after oral administration of low-dose hormones. Five
study resumed normal activity after operation. Of the 93 patients, patients developed joint stiffness, and the patient with patellar
82 patients (88.2%) returned to their full level of preoperative lesion developed fracture after complete healing of the lesion.
function and had excellent functional recovery at the latest In the SR group, joint stiffness developed in four patients, while

Frontiers in Oncology | www.frontiersin.org 5 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

other complications were not observed. Postoperative fracture, extensively, and finally, iodine tincture can denature the protein
infection, and failure of internal fixation were not observed in and cause coagulation necrosis of the cells. Consequently,
both groups. successive combined treatment can treat the lesions from an
omni-direction and multi-angle to reduce recurrence.
DISCUSSION Although EC can achieve excellent oncological prognosis, we
had summed up some experience from recurrent cases: ➀ The
GCTBs in the around knee joint were a clinical challenge in window must be large enough for curettage under direct vision,
orthopedics, as the knee joint is the most important weight- and application of sterile oral endoscopy may help in the removal
bearing joint with high functional requirements. Furthermore, of small lesions in blind visual field. ➁ The use of adjuvant
biologically, GCTB showed expansive growth, which can easily therapy should focus on the treatment of articular lateral tumors
break through the bone cortex and even cause pathological to preserve the subchondral bone as much as possible and to
fracture (1, 3, 4, 7). Although it rarely spread into the articular achieve the goal of EC. ➂ For patients with pathological fracture
cavity, subchondral bone involvement was not uncommon, and recurrence, as long as the fracture line or lesion did not
which can have a serious effect on the function of the knee involve the articular cartilage, it can still be treated by EC, and
joint. The treatment of GCTB in the around knee joint should the patient’s oncological prognosis was still satisfactory.
follow the principle of thoroughness and functionality. Thus, it is The repair of bone defect after EC was also a focus of
necessary to thoroughly remove the tumor tissue to recover joint current clinical controversy. Previous studies have confirmed
function. Therefore, how to achieve a balance between radical that cement has many benefits in repairing bone defects after
removal of tumors to reduce recurrence and preserve knee joint EC of GCTB: ➀ The heat released during cement solidification
function is very important for GCTB around the knee joint. can kill the residual tumor cells in the cavity to achieve the
Considering that the main surgical methods were EC (4, 19) effect of extended curettage (25). ➁ It provides strong support to
and SR (5–7), it is of great practical significance to analyse allow early weight-bearing. ➂ Cement filling was suitable for all
and evaluate the efficacy of EC and SR of GCTB around the shapes and sizes of bone defects. ➃ Recurrence can be detected
knee joint. The recurrence rate of GCTB in curettage and bone early on X-ray imaging (7). Although cement provided many
grafting was 40–60% because of insufficient surgical margin (3). satisfactory benefits, due to the difference in elastic modulus
Consequently, in GCTB treatment, some authors used physical between the cement and normal bone, the surrounding bone can
or chemical methods such as high-speed burr (20), ethanol (21), be gradually absorbed after stress and lead to the loosening of
phenol (22), liquid nitrogen (23), and other physical or chemical the cement, resulting in the “ball effect,” so it is often necessary
methods to expand the surgical curettage boundary, followed by to use internal fixation when filling with cement. Considering
cement repair of bone defects to achieved satisfactory results. that cement directly adheres to the subchondral bone or articular
In the present study, we used high-speed burr, electrotome cartilage, its local thermal effect and stress concentration on the
cauterization, and iodine tincture to treat the tumors subchondral bone and articular surface can easily lead to cartilage
successively, and we achieved satisfactory oncological prognosis damage, which increases the risk of intra-articular fracture and
(Figure 4). Only five cases (7.2%) had recurrence, which was early osteoarthritis (26). Therefore, subchondral bone grafting
significantly better than that reported in previous studies (3, 24). was often used as a buffer zone to avoid the harmful effects
High-speed burr can grind the bone ridge in the cavity; hence, it of bone cement (4, 27, 28). Radev et al. (28) performed finite
was convenient to remove the residual tumor tissue around the element analysis and found that as long as there was at least
bone ridge. Electrotome cauterization can sweep the tumor wall 3-mm uniform cancellous bone above the cement, the thermal

FIGURE 4 | Typical imaging manifestations of patients with local recurrence treated with extended curettage. (a) Anteroposterior and lateral radiographs showed that
the proximal and lateral parts of the right tibia had obvious osteolytic foci at the original site. (b) Computed tomography (CT) scan confirmed low-density osteolytic
changes in the original site. (c) Extended curettage, cement filling, and subchondral bone grafting were performed after relapse of GCTB was confirmed by biopsy
(white arrow points to the lesion).

Frontiers in Oncology | www.frontiersin.org 6 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

FIGURE 5 | Typical imaging features of a 38-year-old man with GCT in the distal femur on the right side. Anteroposterior and lateral radiographs (a), CT scan (b), and
MRI (c) of the knee joint showed that the patient had pathological fracture that involved the articular surface, which was defined as Campanacci III GCT. (d)
Anteroposterior radiographs showing that both lower limbs were equal in length and the prosthesis was stable 15 months after segmental excision and artificial
prosthesis reconstruction (white arrows indicate key points).

effect of the cement will not endanger the articular cartilage because it can achieve satisfactory oncological prognosis. So at
and subchondral bone. We used cement filling and subchondral this point, we can reach a consensus that SR can achieve better
bone grafting (5–10 mm) to repair bone defects in the EC group oncological prognosis in the surgical treatment of high-grade
(Figure 3), and the cement was fixed. Only five patients were complex GCTB. But the pros and cons often coexist, although the
found to have secondary early osteoarthritis during follow-up, oncological prognosis of patients undergoing SR was satisfactory,
and surgical treatment was not required. Therefore, we have the functional prognosis of prosthesis replacement can not to
also proven that subchondral bone grafting can avoid the direct ours heart’s content, which was a mechanical reconstruction
damage of cement to cartilage and reduce the incidence of method had some deficiencies.
postoperative complications without affecting the recurrence Prosthetic replacement can make patients recover joint
rate, which was consistent with that reported in previous studies function early without affecting appearance and provide good
(7, 29, 30). joint stability and range of motion, but long-term complications
SR, as an excellent surgical method for oncological prognosis, may engender a heavier burden on patients (5–7, 10). Through
was recommended for GCTB of the proximal fibula (31), distal long-term follow-up study, Bus et al. (36) found that there
radius (32), and part of Campanacci grade III (Figure 5) (5, 29, were high mechanical and structural complications in artificial
30). Medellin et al. (5) reported a lower recurrence rate in patients prostheses of knee joint, and the failure rate of implants
with Campanacci grade III using SR than using EC through a would gradually increase with time. The cumulative rates of
comparative study, and the results were further confirmed by implant failure in 5, 10, and 15 years due to mechanical
Renard et al. (33). Deheshi et al. (34) retrospective analysis of failure were 16.9, 20.7, and 37.9%, respectively. Franklin’s
limb salvage treatment for GCTB in weightbearing long bones (37) long-term study of the effects of cement prosthesis on
revealed that SR were the preferred treatment for patients with periarticular tumors of the distal femur also found a high
severe joint destruction or dislocation, comminuted or intra- risk of revision, reoperation, and infection. These studies
articular fractures. It is also interesting that Balke et al. (35) have confirmed that prosthesis reconstruction may result in
have found that SR is more recommended for recurrent GCTB mechanical and structural complications such as deep infection,

Frontiers in Oncology | www.frontiersin.org 7 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

Our study should be interpreted in light of its limitations.

Similar to many orthopedic oncology studies, our study was
retrospective, the number of patients was limited, and the
follow-up time of patients was inadequate for assessment of
long-term complications. Additionally, the number of patients
in the two groups varies greatly, so there are some biases
in the statistical results, which may weaken the real validity
of the results. Overall, this study comprehensively analyzed
the efficacy of the two methods in the treatment of GCTB
around the knee joint, confirmed the excellent oncological
prognosis of EC and SR, and compared the functional
prognosis of the two methods, which could provide correct
guidance for the surgical treatment of GCTB around the
knee joint.
In conclusion, EC and SR for GCTB around the knee joint
can achieve satisfactory oncological prognosis, but we should
individually select the most suitable surgical method according to
Campanacci grade, age, and long-term complications of patients
FIGURE 6 | The cumulative recurrence-free survival in the Kaplan–Meier curve
was based on local recurrence and 93 cases according to the type of surgery.
and take into account the functional prognosis to ensure excellent
oncological prognosis.

DATA AVAILABILITY STATEMENT

aseptic loosening, fracture of prosthesis stalk, which seriously
affect the life of prosthesis and the prognosis of patients’ function. The datasets generated for this study are available on request to
Considering the age of onset, we can predict that many patients the corresponding author.
will receive revision surgery due to various complications in the
long term, which will increase the financial burden of patients
and sacrifice more joint function. ETHICS STATEMENT
Through this study, we found that the treatment of
The studies involving human participants were reviewed and
GCTB around the knee joint seems to be appropriately
approved by the Research Ethics Committee of Xiangya Hospital.
conservative, giving more patients a joint salvage opportunity.
The patients/participants provided their written informed
Considering that GCTB mostly occurred in individuals aged
consent to participate in this study.
20–40 years, long-term function cannot be guaranteed by SR
and prosthesis replacement. For patients with Campanacci
grade III, we also recommended EC (unless a large mass AUTHOR CONTRIBUTIONS
of peripheral soft tissue was involved or a pathological
fracture involves the articular surface). Even if the patient HH: conceptualization and design of the study, performed the
unfortunately relapses, we can use SR to make up for it. surgical procedures, collected and analyzed the data, prepared
Both SR and EC can effectively reduce the recurrence rate the manuscript, and approved final version of the manuscript.
(Figure 6), but we should balance the recurrence rate and HZ: analyzed the actigraphy data and approved final version
postoperative complications comprehensively when selecting of the manuscript. WL: performed the surgical procedures,
surgical methods for individual patients to maximize the critical revision the manuscript, and approved final version of
functional prognosis of patients on the premise of guaranteeing the manuscript. YL: performed the surgical procedures, screened
excellent oncology prognosis. and included eligible patients, and approved final version of the
We have also acquired some unique insights into the manuscript. CZ: performed the surgical procedures, analyzed
treatment of GCTB through this study: ➀ Oncological the data, and approved final version of the manuscript. QL:
and functional prognosis should be regarded both as conceptualization and design of the study, data collection,
equally important in the treatment of GCTB around statistical analysis, manuscript drafting and revision, and
the knee joint. ➁ Complete removal of the lesion was approved final version of the manuscript.
the fundamental guarantee for oncology prognosis, and
subchondral bone grafting was a good choice to avoid FUNDING
secondary early osteoarthritis. ➂ SR was recommended
for patients with pathological fracture involving the This work was supported by the Provincial Science Foundation
articular surface and lesion that extensively invades the of Hunan (No. 2017JJ3499), the Clinical and Rehabilitation
surrounding tissue. ➃ EC was still preferred for recurrence Research Foundation of Xiangya Hospital and Weiming
as long as the articular surface and peripheral tissue are of Peking University (xywm2015II07), the Fundamental
not involved. Research Funds for the Central Universities (2019zzts019),

Frontiers in Oncology | www.frontiersin.org 8 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

the Key Scientific Research and Innovation Projects of Hunan ACKNOWLEDGMENTS

Postgraduates (CX20190074). Funding was used only for
scientific research in the data collection and analysis. None We would like to give our sincere thanks to Dr. Zhan Liao, Dr.
of the authors received payments or services, either directly Jun wan, Dr. Feng Long, and Dr. Jian Tian for assistance in
or indirectly, that could be perceived to influence or have the data collection and Miss. Qi Qiu (the wife of QL) for her selfless
potential to influence what is written in this work. support during the study.

REFERENCES treatment of tumors of the musculoskeletal system. Clin Orthop Relat Res.
(1993) 286:241–6. doi: 10.1097/00003086-199301000-00035
1. van der Heijden L, van de Sande MA, Heineken AC, Fiocco M, 17. Flandry F, Hunt JP, Terry GC, Hughston JC. Analysis of subjective knee
Nelissen RG, Dijkstra PD. Mid-term outcome after curettage with complaints using visual analog scales. Am J Sports Med. (1991) 19:112–
polymethylmethacrylate for giant cell tumor around the knee: higher 8. doi: 10.1177/036354659101900204
risk of radiographic osteoarthritis? J Bone Joint Surg Am. (2013) 18. Kellgren J, Lawrence J. Radiological assessment of osteo-arthrosis. Ann Rheum
95:e159. doi: 10.2106/JBJS.M.00066 Dis. (1957) 16:494–502 doi: 10.1136/ard.16.4.494
2. Unni KK, Inwards CY. Dahlin’s Bone Tumors: General Aspects and Data on 19. Prosser GH, Baloch KG, Tillman RM, Carter SR, Grimer RJ. Does
10,165 Cases. Rochester, MN: Lippincott Williams & Wilkins (2010). curettage without adjuvant therapy provide low recurrence rates in
3. Niu X, Zhang Q, Hao L, Ding Y, Li Y, Xu H, et al. Giant cell tumor of the giant-cell tumors of bone? Clin Orthop Relat Res. (2005) 435:211–
extremity: retrospective analysis of 621 Chinese patients from one institution. 8. doi: 10.1097/01.blo.0000160024.06739.ff
J Bone Joint Surg Am. (2012) 94:461–7. doi: 10.2106/JBJS.J.01922 20. Algawahmed H, Turcotte R, Farrokhyar F, Ghert M. High-speed burring with
4. Ayerza MA, Aponte-Tinao LA, Farfalli GL, Restrepo CA, Muscolo DL. Joint and without the use of surgical adjuvants in the intralesional management
preservation after extensive curettage of knee giant cell tumors. Clin Orthop of giant cell tumor of bone: a systematic review and meta-analysis. Sarcoma.
Relat Res. (2009) 467:2845–51. doi: 10.1007/s11999-009-0913-8 (2010) 2010:1–5. doi: 10.1155/2010/586090
5. Medellin MR, Fujiwara T, Tillman RM, Jeys LM, Gregory J, Stevenson JD, 21. Lin WH, Lan TY, Chen CY, Wu K, Yang RS. Similar local control between
et al. Prognostic factors for local recurrence in extremity-located giant cell phenol- and ethanol-treated giant cell tumors of bone. Clin Orthop Relat Res.
tumours of bone with pathological fracture. Bone Joint J. (2018) 100-B:1626– (2011) 469:3200–8. doi: 10.1007/s11999-011-1962-3
32. doi: 10.1302/0301-620X.100B12.BJJ-2018-0189.R2 22. Durr HR, Maier M, Jansson V, Baur A, Refior HJ. Phenol as an adjuvant for
6. Grimer RJ, Aydin BK, Wafa H, Carter SR, Jeys L, Abudu A, et al. Very long- local control in the treatment of giant cell tumour of the bone. Eur J Surg
term outcomes after endoprosthetic replacement for malignant tumours of Oncol. (1999) 25:610–8. doi: 10.1053/ejso.1999.0716
bone. Bone Joint J. (2016) 98-B:857–64. doi: 10.1302/0301-620X.98B6.37417 23. Malawer MM, Bickels J, Meller I, Buch RG, Henshaw RM,
7. Salunke AA, Chen Y, Chen X, Tan JH, Singh G, Tai BC, et al. Does Kollender Y. Cryosurgery in the treatment of giant cell tumor.
pathological fracture affect the rate of local recurrence in patients with a A long-term followup study. Clin Orthop Relat Res. (1999)
giant cell tumour of bone?: a meta-analysis. Bone Joint J. (2015) 97-B:1566– 359:176–88. doi: 10.1097/00003086-199902000-00019
71. doi: 10.1302/0301-620X.97B11.35326 24. Takeuchi A, Suwanpramote P, Yamamoto N, Shirai T, Hayashi K, Kimura
8. Hu P, Zhao L, Zhang H, Yu X, Wang Z, Ye Z, et al. Recurrence rates and H, et al. Mid- to long-term clinical outcome of giant cell tumor of bone
risk factors for primary giant cell tumors around the knee: a multicentre treated with calcium phosphate cement following thorough curettage and
retrospective study in China. Sci Rep. (2016) 6:36332. doi: 10.1038/srep36332 phenolization. J Surg Oncol. (2018) 117:1232–8. doi: 10.1002/jso.24971
9. Blackley HR, Wunder JS, Davis AM, White LM, Kandel R, Bell RS. Treatment 25. Bini SA, Gill K, Johnston JO. Giant cell tumor of bone. Curettage
of giant-cell tumors of long bones with curettage and bone-grafting. J and cement reconstruction. Clin Orthop Relat Res. (1995)
Bone Joint Surg Am. (1999) 81:811–20. doi: 10.2106/00004623-199906000- 245–50. doi: 10.1097/00003086-199512000-00037
00008 26. Xu HR, Niu XH, Zhang Q, Hao L, Ding Y, Li Y. Subchondral
10. Myers GJ, Abudu AT, Carter SR, Tillman RM, Grimer RJ. Endoprosthetic bone grafting reduces degenerative change of knee joint in patients
replacement of the distal femur for bone tumours: long-term results. of giant cell tumor of bone. Chin Med J. (2013) 126:3053–6.
J Bone Joint Surg Br. (2007) 89:521–6. doi: 10.1302/0301-620X.89B4. doi: 10.3760/cma.j.issn.0366-6999.20121150
18631 27. Frassica FJ, Sim FH, Pritchard DJ, Chao EY. Subchondral replacement:
11. Becker WT, Dohle J, Bernd L, Braun A, Cserhati M, Enderle A, et al. a comparative analysis of reconstruction with methyl methacrylate or
Local recurrence of giant cell tumor of bone after intralesional treatment autogenous bone graft. Chir Organi Mov. (1990) 75(1 Suppl):189–90.
with and without adjuvant therapy. J Bone Joint Surg Am. (2008) 90:1060– 28. Radev BR, Kase JA, Askew MJ, Weiner SD. Potential for thermal damage
7. doi: 10.2106/JBJS.D.02771 to articular cartilage by PMMA reconstruction of a bone cavity following
12. O’Donnell RJ, Springfield DS, Motwani HK, Ready JE, Gebhardt tumor excision: a finite element study. J Biomech. (2009) 42:1120–
MC, Mankin HJ. Recurrence of giant-cell tumors of the long 6. doi: 10.1016/j.jbiomech.2009.02.005
bones after curettage and packing with cement. J Bone Joint 29. Wang HC, Chien SH, Lin GT. Management of grade III giant cell tumors of
Surg Am. (1994) 76:1827–33. doi: 10.2106/00004623-199412000- bones. J Surg Oncol. (2005) 92:46–51. doi: 10.1002/jso.20338
00009 30. Tsukamoto S, Mavrogenis AF, Tanzi P, Leone G, Righi A, Akahane M, et al.
13. Benevenia J, Rivero SM, Moore J, Ippolito JA, Siegerman DA, Similar local recurrence but better function with curettage versus resection for
Beebe KS, et al. Supplemental bone grafting in giant cell tumor bone giant cell tumor and pathological fracture at presentation. J Surg Oncol.
of the extremity reduces nononcologic complications. Clin (2019) 119:864–72. doi: 10.1002/jso.25391
Orthop Relat Res. (2017) 475:776–83. doi: 10.1007/s11999-016- 31. Faezypour H, Davis AM, Griffin AM, Bell RS. Giant cell tumor of the proximal
4755-x fibula: surgical management. J Surg Oncol. (1996) 61:34–7. doi: 10.1002/
14. Szalay K, Antal I, Kiss J, Szendroi M. Comparison of the degenerative (SICI)1096-9098(199601)61:1<34::AID-JSO8>3.0.CO;2-T
changes in weight-bearing joints following cementing or grafting techniques 32. Liu Q, Luo W, Zhang C, Liao Z, Liu Y, He H. How to optimize the
in giant cell tumour patients: medium-term results. Int Orthop. (2006) 30:505– therapeutic effect of free autogenous fibula graft and wrist arthroplasty
9. doi: 10.1007/s00264-006-0190-z for giant cell tumors of distal radius? Jpn J Clin Oncol. (2019) 49:656–
15. Campanacci M, Baldini N, Boriani S, Sudanese A. Giant-cell tumor of bone. J 63. doi: 10.1093/jjco/hyz045
Bone Joint Surg Am. (1987) 69:106–14. 33. Renard AJ, Veth RP, Pruszczynski M, Wobbes T, Lemmens JA, van Horn JR.
16. Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ. A system Giant cell tumor of bone: oncologic and functional results. J Surg Oncol. (1994)
for the functional evaluation of reconstructive procedures after surgical 57:243–51. doi: 10.1002/jso.2930570408

Frontiers in Oncology | www.frontiersin.org 9 September 2019 | Volume 9 | Article 946

He et al. GCTB Around the Knee Joint

34. Deheshi BM, Jaffer SN, Griffin AM, Ferguson PC, Bell RS, Wunder JS. Joint for periarticular tumors of the distal femur. Knee. (2016) 23:167–
salvage for pathologic fracture of giant cell tumor of the lower extremity. Clin 72. doi: 10.1016/j.knee.2015.08.010
Orthop Relat Res. (2007) 459:96–104. doi: 10.1097/BLO.0b013e31805d85e4
35. Balke M, Ahrens H, Streitbuerger A, Koehler G, Winkelmann W, Gosheger Conflict of Interest: The authors declare that the research was conducted in the
G, et al. Treatment options for recurrent giant cell tumors of bone. absence of any commercial or financial relationships that could be construed as a
J Cancer Res Clin Oncol. (2009) 135:149–58. doi: 10.1007/s00432-008- potential conflict of interest.
0427-x
36. Bus MP, van de Sande MA, Fiocco M, Schaap GR, Bramer JA, Dijkstra Copyright © 2019 He, Zeng, Luo, Liu, Zhang and Liu. This is an open-access article
PD. What are the long-term results of MUTARS R
modular endoprostheses distributed under the terms of the Creative Commons Attribution License (CC BY).
for reconstruction of tumor resection of the distal femur and proximal The use, distribution or reproduction in other forums is permitted, provided the
tibia? Clin Orthop Relat Res. (2017) 475:708–18. doi: 10.1007/s11999-015- original author(s) and the copyright owner(s) are credited and that the original
4644-8 publication in this journal is cited, in accordance with accepted academic practice.
37. Houdek MT, Wagner ER, Wilke BK, Wyles CC, Taunton MJ, Sim No use, distribution or reproduction is permitted which does not comply with these
FH. Long term outcomes of cemented endoprosthetic reconstruction terms.

Frontiers in Oncology | www.frontiersin.org 10 September 2019 | Volume 9 | Article 946