Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032

DOI 10.1007/s00167-009-0721-4

KNEE

Surgical outcomes after traumatic open knee dislocation

Joseph J. King III Æ Douglas L. Cerynik Æ
James A. Blair Æ Susan P. Harding Æ
James A. Tom

Received: 14 July 2008 / Accepted: 7 January 2009 / Published online: 10 February 2009
Ó Springer-Verlag 2009

Abstract The purpose of this study is to describe the While infection rates are high, aggressive, individualized
types of injuries and surgical treatments associated with treatment can lead to satisfactory outcome although full
open knee dislocations and to present the functional out- return to activity is difficult to achieve using current
comes of these patients. Between 2001 and 2005, the treatment methods.
medical records of patients that sustained traumatic open
knee dislocations at our Level 1 Trauma Center were ret- Keywords Open knee dislocation
Functional outcome

rospectively reviewed. Initial surgical intervention was Ligamentous reconstruction
Surgical intervention

performed in all patients including placement of spanning Associated injuries
external fixator, repair of vascular injuries if necessary, and
irrigation and debridement of the open wounds. Ligamen-
tous reconstruction was delayed until after limb salvage. Introduction
The Short Form-12 was the primary outcome measure.
Seven patients (five male, two female) had a mean age of Open knee dislocations represent a reported 5–17% of all
31.9 years (range 22–44) at the time of injury (five right, knee dislocations [1, 6, 7, 11, 14, 16]. The majority of cases
two left). Motorcycle accident was the most common cause of open knee dislocations have been reported with closed
(57%). Follow-up was a mean 27.6 months. The PCL was knee dislocations and the types of associated injuries,
damaged in all patients. Three patients underwent angi- treatments, and expected outcomes are difficult to predict
ography for absent/diminished pulses on initial exam with due to a lack of large reports in the recent literature. While
two requiring operative intervention. Three patients had many case series have presented patients with closed knee
associated common peroneal nerve injury (one iatrogenic). dislocations focusing on their operative treatment and
Ten (10.7) operative procedures were performed per functional outcomes, few studies have described the
patient (range 5–18) with an average of 6.6 debridements methods of treatment and outcomes of open knee disloca-
(range 2–11). Infection rate was 43% with one patient tions. Additionally, the complexity of these dislocations are
undergoing amputation for infection. Good to excellent often seen in the setting of high-energy, polytrauma
results were found in 33% of patients. Most patients (86%) patients further complicating treatment options [13, 15].
report some residual symptomatic or functional deficit. Due The purpose of this study is to describe the types of
to the injury complexity in open traumatic knee disloca- injuries and surgical treatments associated with open knee
tions, the surgical treatment is extensive and challenging. dislocations and to present the outcomes of these patients.

J. J. King III
D. L. Cerynik
J. A. Blair
Materials and methods

S. P. Harding
J. A. Tom (&)
Department of Orthopaedic Surgery, Drexel University College
Between 2001 and 2005, the medical records of seven
of Medicine/Hahnemann University Hospital, 245 N. 15th Street,
MS 420, Philadelphia, PA 19102, USA patients that sustained traumatic open knee dislocations
e-mail: [email protected] who were treated at our institution were retrospectively

123
1028 Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032

reviewed. Knee dislocation was defined as a dislocated (Fig. 1a, b). Subsequent irrigation and debridements were
tibio-femoral joint seen clinically or radiographically by an performed as indicated. Delayed wound closures using
orthopaedist or in the setting of gross knee instability with secondary healing, grafts, or flaps were addressed using
the appropriate history. Demographic data, type and extent standard techniques (Fig. 1c).
of injuries, radiographic imaging, and types of surgical External fixators were removed once the patients were
procedures performed were reviewed and recorded. Con- able to undergo further orthopaedic procedures and their
comitant injuries to the ipsilateral leg were identified and other associated injuries including wound issues were
recorded. The number and types of operative procedures to stabilized (range 9–59 days). After open reduction and
the affected leg and any related complications were noted. internal fixation or intramedullary nailing of ipsilateral
Follow-up data was at the last office visit or by telephone long bone fractures was performed, the ligamentous inju-
interview if the patient was unable to come to the office. ries were addressed. Examination under anesthesia of
Five patients were male and two female. The mean age the knee was then performed at the time of removal of the
at injury was 31.9 years (range 22–44 years). Five knees external fixators. At this point it was determined if the
were right and two knees were left. No patients sustained patient could undergo delayed versus early ligamentous
bilateral knee dislocations. The mechanisms of injury repair or reconstruction depending on the status of the
involved motorcycle accident in four patients, motor wounds and other long bone injuries. Surgical ligamentous
vehicle accident in two patients, and pedestrian struck by procedures included arthroscopic anterior cruciate ligament
vehicle in one patient. All patients were treated at a Level I (ACL) reconstruction, arthroscopic posterior cruciate
Trauma Center using standard trauma protocol. ligament reconstruction (PCL), open medial collateral
All patients whose knees did not reduce prior to pre- ligament (MCL) repair or reconstruction, and open
sentation were reduced in the emergency department or the posterolateral corner (PLC) repair or reconstruction. Post-
operating room within 6 h of initial injury. Three patients’ operatively, a hinged knee brace was used for at least
knees were reduced prior to presentation and four were 4 weeks with physical therapy focused on early passive
reduced at presentation. Immediate arteriography was range of motion.
performed if any question of vascular compromise was The Short Form-12 (SF-12) was used in order to
found on clinical examination. Initial surgical intervention establish a common outcome assessment for patients with
included placement of spanning external fixators on all diverse concomitant injuries. This validated survey pro-
patients, repair of vascular injuries (when indicated) and vided outcome measures with low respondent burden
irrigation and debridement (I and D) of the open wounds which may be compared with population based normative

Fig. 1 a Pre-operative image

demonstrating an open knee
dislocation and an open fracture
of the ipsilateral mid-shaft tibia.
b Operating room image after
initial debridement, placement
of spanning external fixation,
and delayed closure of the open
wounds. c Post-multiple
irrigation and debridements,
removal of the spanning
external fixator, and primary
closure of wound with the
addition of skin graft

123
Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032 1029

data. The instruments were administered at follow-up by seven patients (71%) sustained a meniscal tear of at least
asking the patient to report on their current situation as well one horn that was diagnosed intra-operatively.
as to recall their pre-injury status [8]. Changes in mean The most common bony injury observed with the open
scores were evaluated for significance using the two-tailed knee dislocations was a fractured tibia which was present
paired t test. in 5 (71%) patients. In addition, ipsilateral femur fractures,
or floating knees, were also present in two patients who
sustained tibia fractures. Three patients with tibia fractures
Results also sustained fractures to the fibula. Associated intra-
articular fractures were present in two of the five patients
Using the KD Classification, three knees had anterior dis- with fractures.
locations, three had posterior dislocations, and one had a Extensive ipsilateral degloving of the extremity occur-
lateral dislocation (Table 1). Of the seven patients, two red in three patients and included an avulsion of the
patients sustained Type IIIA open dislocations, three patellar tendon. One patient also presented with an exten-
patients had Type IIIB dislocations, and two had Type IIIC sive degloving injury extending from the buttocks to
dislocations (Table 1) if these dislocations are extrapolated ischium requiring a delayed gastrocnemius flap and full
according to the Gustilo classification for open fractures thickness skin graft.
[3]. One patient suffered severe head, chest, and abdominal Three patients were found to have absent or diminished
injuries. dorsalis pedis with or without posterior tibialis pulses on
The PCL was the most commonly injured ligament with initial exam. Angiography demonstrated a popliteal artery
seven (100%) patients sustaining ruptures. The ACL was thrombosis in one patient who subsequently developed
the next most commonly injured ligament (86%). Two compartment syndrome requiring fasciotomies and reverse
patients sustained ACL, PCL, and PLC ruptures and two saphenous vein grafting. One patient demonstrated popli-
had rupture of all the involved ligaments (Table 1). Five of teal artery laceration on arteriogram requiring vascular

Table 1 Patient characteristics, injuries, treatments performed, and outcomes

Pt. # Age, ACL PCL MCL PLC Associated Fasciotomies Injury CPN Popliteal Significant
sex tear Tear tear tear fracturea doneb typec injury artery injury degloving

1 36, M Y Y N Y Y (IA) N IIIB Y (I) N N

2 28, M Y Y Y N Y Y IIIC Y Y Y
3 24, M Y Y – – Y (IA) N IIIB N N Y
4 44, F Y Y N Y N N IIIA N N N
5 32, M Y Y Y Y Y N IIIB Y Y (spasm) Y
6 37, M – Y N N N N IIIA N N N
7 22, F Y Y Y Y Y Y IIIC N Y N
Total 71% 29% 43% 43% 43%
Pt. # KD Total Surgeries Total I and Infection Flap Amputation Follow-up SF-12 physical SF-12 mental
classification performed D’s performed required time (months) score score

1 Lateral 12 11 Y N N 20 21 61
2 Anterior 16 9 Y N Y 4 – –
3 Anterior 9 9 N Y N 22 48 50
4 Posterior 5 2 N N N 32 57 55
5 Posterior 12 7 Y Y N 24 52 53
6 Anterior 5 2 N Y N 68 25 29
7 Posterior 16 6 N N N 23 35 52
Total Avg 10.7 Avg 6.6 43% 43% 14% Avg 27.6 Avg 40 Avg 50

Pt. patient, M male, F female, Y yes, N no, ACL anterior cruciate ligament, PCL posterior cruciate ligament, MCL medial collateral ligament,
PLC postero-lateral corner, CPN common peroneal nerve, IA intra-articular, I iatrogenic, KD knee dislocation, I and D irrigation and
debridement, Avg average
a
Fracture of the ipsilateral mid-shaft to distal femur, proximal to mid-shaft tibia, or fibula
b
Lower leg fasciotomies only
c
Gustilo classification for open fractures applied in this setting

123
1030 Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032

repair using reverse saphenous vein grafting and fascioto- more than one residual symptom. Four patients (57%)
mies for impending compartment syndrome. One patient reported returning to some form of employment or previ-
was found to have vasospasm on angiography and required ous educational pursuits, with one patient even returning to
no vascular intervention. work as a driver.
One patient sustained a complete loss of common
peroneal nerve function secondary to excavation of the
popliteal fossa during injury. One patient experienced a Discussion
persistent common peroneal nerve palsy though no evi-
dence of nerve destruction was present on visualization. Management of open knee dislocations is complex and
An average of 10.7 operative procedures were per- arduous. Each patient must be evaluated individually and
formed per patient (range 5–18). The average number of the treatment based on the type and extent of the patient’s
irrigation and debridements per patient was 6.6 (range injuries. While closed knee dislocations require extensive
2–11). Delayed wound closure for the extremities included treatment, the addition of contamination in open injuries
a gastrocnemius flap (1 patient), soleus flaps (2 patients), makes the treatment course even more difficult. In addition,
split-thickness skin grafts (5 patients, 10 total), and dermal patients with open knee dislocations often have associated
scaffolds (2 patients). traumatic injuries further complicating treatment options
Delayed reconstruction of ruptured ligaments was per- [18]. In our experience, adequate irrigation and debride-
formed in two patients. Primary repair was performed in ments, appropriate bony fixation, followed by ligamentous
one patient. Three patients were not offered extensive early repair can yield satisfactory functional outcomes for
repair or late reconstruction secondary to extensive ipsi- patients with these severe injuries. The individualization
lateral and associated injuries. Two of these patients of treatment of open knee dislocations is paramount to a
underwent surgery to improve range of motion after non- successful result as these injuries can severely compro-
operative treatment of the ligamentous injuries. One patient mise function and are associated with a high rate of
had a unicondylar replacement before a total knee arthro- complications.
plasty was performed 4 and 6 years after the initial injury, Vascular injury to the popliteal artery or vein can be a
respectively. devastating problem in these patients and requires imme-
One patient experienced temporary iatrogenic peroneal diate diagnosis and management [11]. Ultimately 8–22% of
nerve palsy subsequent to open reduction and internal patients with knee dislocations have a vascular injury
fixation of the tibia. One patient continues to have persis- requiring repair [1, 9, 10, 12, 14]. A slightly higher per-
tent foot drop although no injury to the peroneal nerve was centage has been reported with open knee dislocations
identified at the time of surgery. The patient with extensive (26%) [18] which was similar to our study (29%). The
excavation of the popliteal fossa went on to an above knee protocol used at our institution is that any patient with
amputation secondary to vascular insufficiency and deep absent or diminished distal pulses after reduction of the
infection. dislocated side compared to the contralateral side receives
Infection occurred in three of the seven patients (43%) a prompt vascular examination including arteriogram.
(Table 1). Two of the infections resolved with repeat Diminished pedal pulses can be caused by arterial transec-
irrigation and debridements and a 6-week course of tion, vasospasm, thrombosis, compression by extravascular
intravenous antibiotics. One patient with a significant hematoma, mural hematoma, or intimal flaps [5]. Patients
vascular injury ultimately failed limb salvage treatment with normal vascular exams have been shown to have a low
for infection and eventually underwent an above knee probability of having vascular injury requiring surgery [5].
amputation. Abnormal distal pulses were present in three of our seven
Final follow-up was an average 27.6 months (range patients (43%) which is similar to the rates found in the
4–68 months). Six patients were ambulating, though five literature [5]. Two of these patients underwent vascular
patients complained of difficulties, after multiple operative surgery for injuries diagnosed on angiography. In open and
procedures to restore function and increase range of closed knee dislocations, close monitoring for compartment
motion. One patient was lost to follow-up though report- syndrome with serial exams should be done due to the high
edly ambulating with an above knee prosthesis. risk of rising compartment pressures with such an extensive
At final follow-up, SF-12 physical and mental compo- soft tissue injury. Delayed diagnosis and treatment of
nent scores decreased an average 17.2 points (P = 0.04) popliteal artery injury can lead to higher complication rates
and 5.2 points (P = 0.35), respectively. Most patients [2, 5].
(85.7%) reported residual symptoms, including pain, Injury to the common peroneal nerve is another common
ongoing knee instability, loss of range of motion, and finding in knee dislocations affecting from 5 to 20% of all
complications of nerve palsies. No patient complained of patients [4, 7, 11, 12, 14–16] although rates as high as 75%

123
Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032 1031

have been reported [10]. However, nerve function is The amputation rate after vascular injury has been
regained in a majority of patients with closed dislocations. reported at 55% with an even higher rate in patients not
Harner et al. [4] only had one case of permanent nerve undergoing vascular repair within the first 8 h after injury
palsy in 31 cases of closed knee dislocations (3%). Sixty- [2]. Hollis et al. [5] reported an amputation rate of 8% with
seven percent of the nerve injuries in another study com- traumatic knee dislocations due to irreparable vascular
pletely recovered [10]. Injury to the tibial nerve has been injury. Some other studies reported less problems with
described as well [12]. A higher percentage of neurologic vascular injury and had no amputations [7, 10–12]. In the
injury has been reported in open knee dislocations (37%) largest study of open knee dislocations, the amputation rate
[18]. was 16% with two-thirds of these amputated secondary to
The goal of ligamentous reconstruction is to restore knee subsequent infection despite adequate debridement [18].
stability and regain as much function of the knee as pos- The most common functional complaints after operative
sible. Patients with complete ligamentous reconstructions treatment of knee dislocations are instability, stiffness, a
after closed knee dislocations have reported good stability, limp, locking, and swelling. Manipulation or arthroscopic
but on the other hand can have to stiffness due to soft tissue lysis of adhesions can provide increased range of motion in
scarring [1, 4, 10]. Timing of ligamentous reconstruction in stiff knees [14, 15, 17]. Harner et al. [4] and Chhabra et al.
closed knee dislocations is a controversial topic with some [1] reported that patients treated for ligamentous injury
studies showing improved results with a ligamentous repair within 3 weeks of a closed knee dislocation had higher
within 3–4 weeks [1, 4]. Allograft reconstruction is the subjective functional scores, but similar knee range of
method of choice for early reconstruction of severe injuries motion compared to those treated greater than 3 weeks
and for delayed reconstruction due to the absence of donor after injury. Two studies demonstrated higher subjective
site injury as occurs with autograft reconstruction. Primary functional scores with patients treated with ligamentous
repair of avulsion injuries and ligamentous tears close to reconstruction within 14 days post-injury compared with
bone is advocated in patients undergoing surgery early. In those treated after 14 days, but the difference was not
addition, the contamination associated with open injuries, significant [7, 10] The best function obtained in a patient
as in our patient population, makes delayed ligamentous with an open dislocation in the largest study of open dis-
repair necessary to decrease the risk of infection and aid in locations was a moderate functional disability with
wound closure in order to prevent subsequent failure of the significant clinical signs and moderate symptoms [18]. We
repair unlike in patients with closed knee dislocations. All have had improved functional results compared to this
of our patients had delayed ligamentous reconstruction. study in patients with open knee dislocations.
In addition, the presence and treatment of ipsilateral and With the use of a validated scoring system (SF-12
contralateral fractures are important factors that affect the physical component), good to excellent results were
outcome in patients with knee dislocations. One study achieved in two of six patients (33%). Both patients had
reported that 11 of 28 dislocated knees (39%) had associ- Type IIIB dislocations with subsequent ligamentous repair
ated fractures although the incidence of open injuries was or reconstruction. Despite their injuries, both patients
not reported [10]. Another study reported that 31% of reported being relatively pain-free and had returned to
patients had associated ipsilateral fractures of the tibia or work without restrictions, even though one patient suffered
femur [12]. Associated ipsilateral knee fractures were a complete peroneal nerve palsy. Fair to poor results were
found in 16% of cases closed knee dislocation in a met- obtained in four of six patients. Three of the four patients
analysis [15]. reported that their main limitation was pain while the last
The aim of acute surgical treatment is to preserve the patient reported a loss of range of motion.
limb, repair the most severely damaged structures of the The main limitations of this study are its retrospective
knee, and close the open wound. All patients in our study design. The use of recall to obtain pre-injury functional
underwent multiple irrigation and debridements in order to scores is not ideal for analysis of improvement of function
attempt to prevent infection and most required subsequent due to recall bias, but actual pre-injury function is difficult
surgeries for wound coverage and closure. to determine by other means in the trauma setting. Another
The infection rate in this study was 43% for open limitation of this study is its lack of a standardized
traumatic knee dislocations. One study of open traumatic treatment protocol; however, this is difficult given the
knee dislocations reported an infection rate of 42% despite nature of this type of injury and associated injuries.
the use of immediate intravenous antibiotics and multiple Not only are open traumatic knee dislocations rare, they
debridements [18]. The risk of infection in these open also are associated with heterogeneous injuries where the
injuries is an important determinant of optimization of treatment varies on the presence and extent of damage to
definitive treatment of the bony and ligamentous injuries the vasculature, bones, nerves, tendons, and ligaments.
seen in this patient population. This complexity and rarity makes large studies with

123
1032 Knee Surg Sports Traumatol Arthrosc (2009) 17:1027–1032

standardized treatments difficult. In this study, patients 5. Hollis J, Daley B (2005) 10-Year review of knee dislocations: is
were treated based on the type of injuries and in a manner arteriography always necessary? J Trauma 59:672–676
6. Ibrahim S, Ahmad F, Salah M et al (2008) Surgical management
that the orthopaedic surgeons thought would be the best of traumatic knee dislocation. Arthroscopy 24:178–187
treatment individualized for each patient. 7. Liow R, McNicholas M, Keating J et al (2003) Ligament repair
and reconstruction in traumatic dislocation of the knee. J Bone
Joint Surg Br 85:845–851
8. Mancuso C, Charlson M (1995) Does recollection error threaten
Conclusion the validity of cross-sectional studies of effectiveness? Med Care
33:AS77–AS88
Open knee dislocations are often the result of high-energy 9. Meyers M, Harvey J Jr (1971) Traumatic dislocation of the knee
trauma. They can be associated with other limb-threatening joint. A study of eighteen cases. J Bone Joint Surg Am 53:16–29
10. Owens B, Neault M, Benson E et al (2007) Primary repair of knee
injuries which include multiple ligamentous disruption, dislocations: results in 25 patients (28 knees) at a mean follow-up
neurovascular injury, and open fractures. Due to the com- of four years. J Orthop Trauma 21:92–96
plexity of the injuries in open traumatic knee dislocations, 11. Reckling F, Peltier L (2004) Acute knee dislocations and their
the surgical treatment is extensive and challenging. Careful complications. Clin Orthop Relat Res 422:135–141
12. Richter M, Bosch U, Wippermann B et al (2002) Comparison of
evaluation and management requires an integrated multi- surgical repair or reconstruction of the cruciate ligaments versus
disciplinary team approach. Treatment should be individ- nonsurgical treatment in patients with traumatic knee disloca-
ualized for each patient to optimize patient satisfaction and tions. Am J Sports Med 30:718–727
functional outcome. 13. Rihn J, Groff Y, Harner C et al (2004) The acutely dislocated
knee: evaluation and management. J Am Acad Orthop Surg
12:334–346
14. Rios A, Villa A, Fahandezh H et al (2003) Results after treatment
of traumatic knee dislocations: a report of 26 cases. J Trauma
References 55:489–494
15. Robertson A, Nutton R, Keating J (2006) Dislocation of the knee.
1. Chhabra A, Cha P, Rihn J et al (2005) Surgical management of J Bone Joint Surg Br 88:706–711
knee dislocations: surgical technique. J Bone Joint Surg Am 16. Taylor A, Arden G, Rainey H (1972) Traumatic dislocation of the
87(Suppl 1):1–21 knee. A report of forty-three cases with special reference to
2. Green N, Allen B (1977) Vascular injuries associated with dis- conservative treatment. J Bone Joint Surg Br 54:96–102
location of the knee. J Bone Joint Surg Am 59:236–239 17. Tzurbakis M, Diamantopoulos A, Xenakis T et al (2006) Surgical
3. Gustilo R, Williams D (1984) Problems in the management of treatment of multiple knee ligament injuries in 44 patients:
type III (severe) open fractures: a new classification of type III 2–8 years follow-up results. Knee Surg Sports Traumatol Arth-
open fractures. J Trauma Injury Infect Crit Care 24:5 rosc 14:739–749
4. Harner C, Waltrip R, Bennett C et al (2004) Surgical manage- 18. Wright D, Covey D, Born C et al (1995) Open dislocation of the
ment of knee dislocations. J Bone Joint Surg Am 86-A:262–273 knee. J Orthop Trauma 9:135–140

123