184 C OPYRIGHT Ó 2014
BY
T HE J OURNAL
OF
B ONE
AND J OINT
S URGERY, I NCORPORATED
Hinged External Fixation in the Treatment of Knee Dislocations A Prospective Randomized Study James P. Stannard, MD, Clayton W. Nuelle, MD, Gerald McGwin, PhD, and David A. Volgas, MD Investigation performed at the Department of Orthopaedic Surgery, University of Missouri, Columbia, Missouri, and the University of Alabama at Birmingham, Birmingham, Alabama
Background: Our hypothesis was that patients treated with hinged external fixators as an adjunct to multiple-ligament reconstruction would have fewer reconstruction failures than patients treated without external fixation. Methods: In this prospective randomized study, patients with a knee dislocation either underwent ligament reconstruction with placement of an external hinged knee brace following surgery (Group A) or underwent ligament reconstruction with placement of a hinged external fixator (Compass Knee Hinge) for six weeks instead of the brace (Group B). The patients were followed clinically and were evaluated with physical examination, Lysholm and International Knee Documentation Committee knee scores, visual analog scale pain scores, and status regarding return to work and activities. Results: One hundred patients with 103 knee dislocations were enrolled. Seventy-seven patients with seventy-nine dislocations (thirty-two in Group A and forty-seven in Group B), with a minimum follow-up interval of twelve months, were available for evaluation. The mean duration of follow-up was thirty-nine months (range, twelve to eighty-six months). Nine patients (29%) in Group A had failed reconstructions compared with seven (15%) in Group B (p = 0.15). Group-A patients had twenty-two (21%) of 105 reconstructed individual ligaments fail compared with eleven (7%) of 157 reconstructed ligaments in Group B. The difference in ligament failure was significant (p < 0.001; power > 0.8), with more favorable results for the patients managed with the external fixation. Conclusions: Hinged external fixation as a supplement to reconstruction following knee dislocation was associated with fewer failed ligament reconstructions compared with external bracing. Patients presenting with highly unstable knee dislocations should be considered for hinged external fixation to supplement initial reconstructive procedures. Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.
Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.
D
islocation of the knee typically follows a high-energy blunt trauma rather than resulting from an athletic competition injury. Knee dislocations are defined as multiple-ligament injuries, which typically involve both cruciate ligaments. In most cases, the knee has spontaneously re-
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
J Bone Joint Surg Am. 2014;96:184-91
d
http://dx.doi.org/10.2106/JBJS.L.01603
duced and is not dislocated at the time of presentation to the emergency department1-5. Authors of early reports debated the benefits of repair compared with reconstruction of ligaments and acute or delayed operative management and subsequently focused primarily on A commentary by Brett D. Owens, MD, is linked to the online version of this article at jbjs.org.
185 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
TABLE I Characteristics of Patients with Complete Follow-up Total (N = 77) Sex (no. of patients) Male Female
Group A (Control)* (N = 31)
59 18
25 6
Group B (Study)* (N = 46)
34 12
Mean age (yr)
35.0
34.1
35.6
Dislocations Left Right
79 42 37
32 14 18
47 28 19
*Group A had forty-five patients with forty-seven dislocations, and Group B had fifty-five patients with fifty-six dislocations. Follow-up data were available for thirty-one patients (32 dislocations) in Group A and forty-six patients (forty-seven dislocations) in Group B.
achieving ligamentous stability6-8. The rigid stability, however, consequently resulted in sacrificing motion by using casts, braces, or transarticular Steinmann pins during the postoperative period. While this approach frequently resulted in stability, it was also frequently associated with disabling loss of motion and pain as a result of arthrofibrosis. Several authors have noted unfavorable results in patients with multiple injuries who have limited early rehabilitation capability9-11. Most authors believe that a stiff, painful knee leads to more long-term morbidity and dysfunction than an unstable knee8,10,12-16. As a result, many authors now strongly advocate early operative treatment and aggressive motion and rehabilitation despite the risk of failure of ligament repairs or reconstructions. Clinical outcomes following knee dislocations are frequently poor, with arthrofibrosis, stiffness, pain, and recurrent instability being the most frequent complications6,8-10,12,13. Surgeons must choose between two postoperative approaches. The first is early aggressive motion with the potential decrease in arthrofibrosis and improved articular cartilage nutrition but an increased risk of ligament instability. The second option is to restrict knee motion and allow early healing of the reconstructions with a decreased risk of ligament instability, despite an increased risk of arthrofibrosis. The Compass Universal Hinge external fixator (Smith & Nephew, Memphis, Tennessee) was originally developed for use at the elbow to allow early motion without placing articular and/ or ligamentous repairs at increased stress and has been modified to create the Compass Knee Hinge. Two initial case reports documented the use of the Compass Knee Hinge with knee
dislocations17,18. There has been ongoing discussion regarding its use following ligamentous reconstruction after knee dislocations19-21. We hypothesized that knee dislocations supplemented with the Compass Knee Hinge compared with a control group would have equivalent final knee range of motion with fewer failures for ligament reconstructions. Materials and Methods
O
ne hundred patients with 103 knee dislocations were enrolled from August 2000 to January 2007. A power analysis was performed prior to data collection using the Stata program sampsi calculation (StataCorp, College Station, Texas). We assumed a failure rate in the control group of 24% and a failure rate in the Compass Knee Hinge group of 8% on the basis of prior retrospective data. When significance was set at 0.05 and power was set at 0.8, ninety-four subjects were required in each group. Knee dislocations commonly vary between two and four ligament groups (anterior cruciate ligament [ACL], posterior cruciate ligament [PCL], posterolateral corner, or posteromedial corner) torn per patient, with a patient having an average of three torn ligaments. Fifty patients in each study arm would then yield approximately 150 ligament groups per study arm, allowing for an appropriately powered study, even if 20% to 30% of patients did not achieve two years of follow-up. The study was designed as a prospective randomized evaluation of patients with knee dislocations, and it was approved and monitored by the institutional review board at our institution. The trial was registered with ClinicalTrials.gov (NCT00582517).
Study Design Patients were randomized into one of two groups using a computer-generated block randomization scheme. The randomization was based on the order of patient presentation, so each patient was individually randomized to a treatment group, regardless of overall injury severity score or the involvement of multiple
TABLE II Knee Dislocation Grades at Presentation* Grade
Total No. of Knees
Group A (no. of knees)
Group B (no. of knees)
I
2
0
2
II
1
1
0
III
38
13
25
IV
38
18
20
30
*The knees were classified according to the system described by Schenck .
186 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
d
d
IN THE
T R E AT M E N T
OF
d
TABLE III Ligament Reconstruction Failures Total No. of Knees
ACL (no. of ligaments)
PCL (no. of ligaments)
Posterolateral Corner (no. of ligaments)
Posteromedial Corner (no. of ligaments)
Group A
9
9
4
6
3
Group B
7
4
0
5
2
injuries of the extremity. Group-A patients had a staged reconstruction of the acute knee dislocation with placement of a hinged knee brace during the postoperative period. Our preferred protocol involved reconstructing the PCL, the posterolateral corner, and the posteromedial corner three or four weeks following the dislocation. The ACL was reconstructed six weeks following the initial surgery. Group-B patients were treated with the same protocol following the knee dislocation, with the exception that they had a Compass Knee Hinge placed on the injured leg at the end of the initial surgical procedure. The hinge was retained for approximately six weeks and then was removed at the time of ACL reconstruction. All patients underwent surgical treatment and follow-up evaluation by the senior surgeon (J.P.S.). The CONSORT (Consolidated Standards of Reporting Trials) statement has been developed to improve the reporting of randomized controlled 22 trials . We obtained data mentioned in the statement and the flow diagram (Fig. 1).
Surgical Technique Our preferred protocol involved a staged ligament reconstruction, beginning with the PCL, the posterolateral corner, and the posteromedial corner. The ACL is reconstructed six weeks following the initial surgery. Our preferred surgical techniques for each ligament included a tibial inlay technique with two femoral tunnels using Achilles tendon allograft for the PCL; a two-tailed technique for the posterolateral corner anatomically reconstructing the popliteus, fibular collateral ligament, and popliteofibular ligament using tibialis anterior allograft; a medial cruciate ligament (MCL) and posterior oblique ligament reconstruction using anterior or posterior tibialis allograft for the posteromedial corner; and finally an ACL reconstruction using either a bone-patellar tendonbone or hamstring autograft or allograft. The Compass Knee Hinge is an external fixator with a hinge that is placed at the center of rotation of the joint. It consists of two carbon-fiber 5/8 rings bolted to multihole Rancho Cubes that connect to 5 or 6-mm external fixator
Fig. 1
CONSORT diagram shows the flow of patients in the study.
187 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
TABLE IV Activity Level and Work Status Before and After Injury
Preinjury activity* (no. [%] of patients) Sedentary Hobbies Recreational athlete Competitive athlete Postinjury activity at 24 mo† (no. [%] of patients) Level less than preinjury status Previous level of activity Work status‡ No duty Light duty Full duty
Total
Group A (Control)
Group B (Study)
14 (21.2) 22 (33.3) 26 (39.4) 4 (6.1)
5 (19.2) 11 (42.3) 10 (38.5) 0
9 (22.5) 11 (27.5) 16 (40) 4 (10.0)
15 (46.9) 17 (53.1)
5 (27.8) 13 (72.2)
10 (71.4) 4 (28.6)
6 (13.6) 4 (9.1) 34 (77.3)
2 (9.1) 2 (9.1) 18 (81.8)
4 (18.2) 2 (9.1) 16 (72.7)
*The groups include twenty-six patients in Group A and forty patients in Group B for whom data were available. †The groups include eighteen patients in Group A and fourteen in Group B who had complete data available at twenty-four months. ‡The groups include twenty-two patients in Group A and twenty-two patients in Group B who had data available at twenty-four months.
Schanz pins. The rings are connected by two calibrated hinges that are placed on the medial and lateral sides of the joint. The placement of the entire apparatus is based on a centering wire that is temporarily placed at the isometric point on the femur. The hinge is centered by holes that accommodate this centering wire 23 (Fig. 2, Video 1 [online]) . The desired isometric point can be determined with fluoroscopy. With use of the fluoroscope, a perfect lateral view is obtained. The isometric point is located where a line drawn along the anterior aspect of the posterior femoral 24 cortex intersects with the Blumensaat line . This is a crucial step in the procedure, as precise placement of the centering wire is critical to the application and subsequent function of the apparatus. Malalignment of the wire and subse-
quently of the entire apparatus can occur. It is essential to obtain excellent fluoroscopy and to be precise with placement of the wire at the isometric point. After the centering pin is placed, the hinge is completed by placing external fixator pins into the femur and tibia. Generally, two 6-mm pins are placed into the femur. One is placed medially and inferiorly off the ring with a one to three-hole Rancho Cube, and one is placed laterally and inferiorly off of the ring with a two-hole Rancho Cube. Three 5-mm pins are normally placed into the tibia through Rancho Cubes placed inferiorly from the inferior ring. Precise and accurate placement of the hinge is critical to its function to limit knee motion. The Compass Knee Hinge limits motion in the coronal and axial planes, subsequently imparting some limitation to the screw home mechanism in normal knees: axial external rotation
Fig. 2
Intraoperative photograph demonstrating placement of the Compass Knee Hinge with a centering wire placed directly through the isometric point on the femur.
188 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
Fig. 3
Intraoperative photograph demonstrating tibial pin location and final application with the Compass Knee Hinge in place. of the femur on the tibia during knee flexion and internal rotation of the femur on 25 the tibia during knee extension with a static medial femoral condyle . This decreases the overall stress placed on the reconstructed ligaments during healing. The hinge allows motion in the sagittal plane, subsequently allowing physiologic rolling motion of the tibiofemoral joint in flexion and extension. This allows for aggressive rehabilitation without sacrificing stability. A critical issue in placing the Compass Knee Hinge is the timing of the steps in the application on a patient with a multiligamentous knee injury. If repair or reconstruction of either the posteromedial or posterolateral corner is part of the planned surgical procedure, the centering wire should be placed prior to the repair or reconstruction of the injured corner. This is important because the wire is placed through the area of the corner repair or reconstruction, often requiring a number of passes to get the position exactly right. The potential to damage a repair of the posterolateral corner is unacceptably high if the centering pin is placed following repair. The solution is to place the centering wire first, mount the hinge, and place the two femoral external fixation pins through Rancho Cubes. The hinge is then detached from the pins, leaving the pins and Rancho Cubes in place in the femur only. The remainder of the hinge is placed on the back table until the completion of the case. The centering wire is removed, and the remainder of the repair and/or reconstruction of the knee is accomplished. Following closure of all wounds, the hinge is remounted on the two femoral pins through the same holes, ensuring an isometric placement on the knee. The tibial pins are then placed to complete application of the hinge (Fig. 3). The tibial pins should not be placed until the end of the case because they are in the way during the ligamentous reconstruction and/or repairs. Our postoperative protocol for both groups involved a progressive attempt to regain knee motion following the initial procedure. Continuous passive motion machines were used in all patients beginning on postoperative day 1. Patient-controlled analgesia pumps or epidurals were used for early postoperative pain management. Weight-bearing was progressed as tolerated with the knee locked in extension. Rehabilitation included progressive increases in knee motion with quadriceps and hamstring strengthening.
Data obtained included a physical examination for knee motion, stability, and visual analog scale pain score. Stability was categorized as 0 for a knee with no ligamentous laxity, as 11 for 1 to 5 mm of laxity, as 21 for 6 to 10 mm, and as 31 for >10 mm26. Clinical success was defined as a stable knee with either 0 or 11 laxity on examination. Failure was defined as either 21 or 31 laxity on clinical examination. We also noted the status with regard to return to employment and recreational activity. Given the high percentages of poor outcomes following 6,8-10,12,13 treatment of knee dislocations reported in the literature , we classified patients with regard to postinjury return to activities into one of two simplified categories: a full return to their previous level or a return to any level of activity that they deemed to be less than a full return. Return to work was based on three possible categories: no duty, a return to light duty, or a return to the previous level of full duty. Additionally, we evaluated the study subjects with outcome instruments that related to either the knee or the overall health status. Knee-specific instruments 27 included the Lysholm score and the International Knee Documentation Com28 mittee (IKDC) objective and subjective scores . Comprehensive health status was 29 documented using the Short Form-36 (SF-36) health inventory . Data were obtained regarding patient demographics, mechanism of injury, vascular and neurologic injury, associated injuries, associated medical problems, and the need for admission to the intensive care unit following injury. The knee dislo30 cation was classified using the anatomic classification scheme designed by Schenck . Finally, the condition of the soft tissue around the knee was documented at the time 31 32 of injury using the scoring systems of Tscherne and Gustilo et al. .
Statistics Clinical and outcome characteristics were compared between groups using the chi-square test and analysis of variance for categorical and continuous measures, respectively. For non-normally distributed continuous variables, nonparametric statistics were employed. The Fisher exact test was used for analyses of categorical variables when expected cell frequencies were less than five. P values of £0.05 (two-sided) were considered significant.
Source of Funding Postoperative Follow-up Evaluation Patients were scheduled for postoperative follow-up examinations at two, six, twelve, twenty-six, and fifty-two weeks and for annual evaluations following the first year.
Smith & Nephew provided some institutional funding to cover research nurse support for this study. No investigators received salary, consultation payments, or benefits for the completion of this work.
189 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
Results ne hundred patients with 103 knee dislocations gave informed consent for study enrollment. Forty-five patients with forty-seven dislocations were randomized to the control group (Group A), and fifty-five patients with fifty-six dislocations were randomized to the study group (Group B). A minimum twelve-month duration of follow-up was obtained for thirtyone patients (thirty-two knee dislocations; 68.9%) in Group A and forty-six patients (forty-seven knee dislocations; 83.6%) in Group B. A minimum twenty-four-month follow-up period was obtained for twenty-two patients (twenty-three knee dislocations; 48.9%) in Group A and twenty-two patients (twentythree knee dislocations; 40%) in Group B. Overall, the mean duration of follow-up was thirty-nine months (range, twelve to eighty-six months). The patients in Group A had been followed for a mean of forty-three months (range, fifteen to seventy-five months), while those in Group B had been followed for a mean of thirty-five months (range, twelve to eighty-six months). The seventy-seven patients (77%) who were available for follow-up comprise the cohort for this study. Demographic data, including patient age, sex, and race, were similar between groups (Table I). Knee dislocation grades and mechanisms of injury were also similar between groups (Table II; see Appendix). The Injury Severity Score (ISS)33 ranged from 9 to 41 (mean,16.6). Group A had a mean ISS of 16.2, while Group B had a mean ISS of 16.6. More than 90% of the patients had a normal result on the vascular examination on admission and were followed using a selective arteriography protocol. Seven patients had abnormal results on the vascular examination, with an equal distribution between the groups. Nine patients had open knee dislocations, four in Group A and five in Group B. The majority of patients with closed dislocations (55%) had a Tscherne Grade-2 soft-tissue injury, with an equal distribution between the groups. Only two patients had a Tscherne Grade-3 injury, and both of them were in Group B. Failure of reconstruction was defined as either 21 or 31 laxity of a reconstructed ligament on clinical examination by the primary surgeon during routine follow-up or, in the case of ACL and PCL reconstructions, a side-to-side difference of >3 mm on examination of the ligaments with use of a KT-2000 arthrometer. The reconstructions failed in nine (28%) of the thirty-two knees in Group A and in seven (15%) of forty-seven knees in Group B; the difference was insignificant (p = 0.15). Twenty-two (21%) of 105 individual ligaments reconstructed in Group A failed, whereas eleven (7%) of 157 ligaments reconstructed in Group B failed (Table III). The difference in individual ligament failure was significant (p < 0.001), with more favorable results for the patients who had the external fixation. Group A also had more individual ligament failures per patient, with a mean of 2.44 ligament failures, compared with Group B, which had a mean of 1.57 ligament failures per patient. Final Lysholm knee scores were similar between the two groups. For patients with a minimum follow-up interval of twelve months, the mean score was 87.7 (range, 61 to 100) for Group A compared with 87.2 (range, 69 to 100) for Group B. For those
O
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
with a minimum follow-up of twenty-four months, the mean score was 90.1 for Group A compared with 89.9 for Group B. The IKDC subjective scores were similar between the groups for patients with a minimum follow-up period of twentyfour months. Group-A patients had a mean score of 56.8, while Group-B patients had a mean score of 49.5. The objective scores were also similar, with a normal knee in 50% of Group A and in 45% of Group B. A nearly normal knee was achieved in 25% of Group A and 36% of Group B. When these results are combined, 75% of the patients in Group A and 81% of the patients in Group B had acceptable results. Failing IKDC results (Class C [abnormal] or Class D [severely abnormal]) were present in 25% of Group-A patients and 18% of Group-B patients. Pain was a common finding, as evidenced by the individual visual analog scale scores. Patients in Group A had a mean score of 2.5 (range, 0 to 7) and those in Group B had a mean score of 2.8 (range, 0 to 8). At the time of the final follow-up, 86% of the patients had returned to work. In Group A, 82% of the patients had resumed full duty, 9% had returned to light duty, and 9% did not return to work. In Group B, 73% of the patients had returned to full duty, 9% had resumed light duty, and 18% did not return to work (Table IV). Knee range of motion was good at the final evaluation. The mean arc of motion was 1° to 128°. Group-A patients had a mean range of motion from 1° to 132°, while Group-B patients had a mean range from 2° to 124°. Discussion nee dislocations are severe injuries that are being recognized with increasing frequency as the definition is clarified and awareness of the frequency of spontaneous reduction is heightened5,11,13. Recent studies have noted that knee dislocations can occur without rupture of either the ACL1 or the PCL3,4, although bicruciate injuries are by far more common. Expanded definitions of knee dislocation include gross or radiographically proven dislocation, injury of multiple knee ligaments with multidirectional instability following highenergy trauma, or rupture of both the ACL and PCL when no gross dislocation could be identified13. Associated injuries frequently include acetabular or tibial plateau fractures and popliteal artery and peroneal nerve injuries7-13,16. Open injuries are infrequent but are associated with extensive ligamentous damage and a high rate of vascular and neurologic involvement when they occur26. Nonoperative treatment of multiple ligament injuries has resulted in poor results in up to 77% of patients, and recent literature has shown improved benefit with surgical treatment16,19,34,35. Most authors currently advocate aggressive surgical treatment of knee dislocations6-8,12,13,15,19,24,36. Early surgical intervention is advocated, with both single surgery and staged ligament reconstruction as options16,24,37,38. Arthrofibrosis and the associated loss of knee motion with pain is a major complication associated with multiligamentous knee injuries. Noyes and Barber-Westin reported motion problems in 71% of their patients with acute injuries requiring manipulation under anesthesia6. Shapiro and
K
190 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
Freedman reported arthrofibrosis in 60% of their patients9. Sisto and Warren noted that nine of nineteen patients required manipulation under anesthesia8. Aggressive early knee motion and rehabilitation may help to reduce these problems but may also place repaired and/or reconstructed ligaments under stresses that may be associated with a higher failure rate and recurrent instability. The Compass Knee Hinge provides rigid stability in all planes of motion except the sagittal plane, allowing the patient to obtain flexion and extension. There is minimal rotational stress placed on reconstructed ligaments when the hinge is in place. The hinge also helps to prevent posterior sag during the acute healing phase. Both studies by Richter and Lobenhoffer17 and Simonian et al.18 described patients with chronic knee dislocations in whom a Compass Knee Hinge was successfully used to treat chronic posterior subluxation or dislocation. Our results demonstrate that progressive rehabilitation can be performed with a Compass Knee Hinge without sacrificing stability. Group-A patients had a failure rate of 21% (twenty-two of 105 reconstructed ligaments), whereas Group-B patients had a significantly lower failure rate of 7% (eleven of 157 reconstructed ligaments) (p < 0.05). When patients did experience a ligament failure, they were less likely to have multiple ligaments fail if they were treated with a Compass Knee Hinge than if they were treated with a hinged knee brace. There are several disadvantages to using the Compass Knee Hinge, which must be balanced against the improved stability. The device is expensive, takes approximately thirty minutes to apply, and frequently causes pain at the site of the femoral pins with flexion beyond approximately 60°. Outcome scores were relatively good in both groups after reconstruction of the knees that had recurrent instability. In our study, 86% of patients returned to work, with ranges from 73% to 82% in each treatment arm returning to full duty. This is similar to reported data demonstrating that up to 80% of patients returned to preinjury work status and had overall improved outcome scores after reconstruction24,39. Limitations There are limitations to our study. First, a number of patients were lost to long-term follow-up, thus limiting our overall
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
sample size. The mean duration of follow-up for each treatment group was forty-three months for Group A (control) and thirty-five months for Group B, and our final sample size was seventy-seven patients, which we believe are strong values for long-term follow-up in a trauma population. In addition, the percentage of patients lost to follow-up in Group A (31%) was higher than that of Group B (16.4%). This difference could change some of the data in terms of the overall number of reconstruction failures. Also, despite losing some patients to follow-up, the study was still adequately powered when based on individual ligament reconstructions (power of >0.8). Finally, there is also the potential limitation of a lack of interobserver reliability in grading of ligament laxity, but the primary surgeon was the sole examiner to grade laxity and subsequent ligament failures during follow-up. In conclusion, patients treated with a Compass Knee Hinge had a significantly lower failure rate of repaired and reconstructed ligaments following surgical procedures compared with the control group. We concluded that the Compass Knee Hinge functioned well as an adjunct to aggressive reconstruction following knee dislocation and may be considered as part of the treatment of highly unstable knee dislocations. Appendix A table showing data on the mechanisms of injury is available with the online version of this article as a data supplement at jbjs.org. n
James P. Stannard, MD Clayton W. Nuelle, MD David A. Volgas, MD Department of Orthopaedic Surgery, University of Missouri, 1100 Virginia Avenue, DC953.00, Columbia, MO 65212. E-mail address for J.P. Stannard: [email protected] Gerald McGwin, PhD University of Alabama at Birmingham, 510 South 20th Street, FOT 960, Birmingham, AL 35294-3409
References 1. Bellabarba C, Bush-Joseph CA, Bach BR Jr. Knee dislocation without anterior cruciate ligament disruption. A report of three cases. Am J Knee Surg. 1996 Fall;9(4):167-70. 2. Bratt HD, Newman AP. Complete dislocation of the knee without disruption of both cruciate ligaments. J Trauma. 1993 Mar;34(3):383-9. 3. Cooper DE, Speer KP, Wickiewicz TL, Warren RF. Complete knee dislocation without posterior cruciate ligament disruption. A report of four cases and review of the literature. Clin Orthop Relat Res. 1992 Nov;(284):228-33. 4. Shelbourne KD, Pritchard J, Rettig AC, McCarroll JR, Vanmeter CD. Knee dislocations with intact PCL. Orthop Rev. 1992 May;21(5):607-8: 610-1. 5. Wascher DC, Dvirnak PC, DeCoster TA. Knee dislocation: initial assessment and implications for treatment. J Orthop Trauma. 1997 Oct;11(7):525-9. 6. Noyes FR, Barber-Westin SD. Reconstruction of the anterior and posterior cruciate ligaments after knee dislocation. Use of early protected postoperative motion to decrease arthrofibrosis. Am J Sports Med. 1997 Nov-Dec;25(6):769-78.
7. Roman PD, Hopson CN, Zenni EJ Jr. Traumatic dislocation of the knee: a report of 30 cases and literature review. Orthop Rev. 1987 Dec;16(12):917-24. 8. Sisto DJ, Warren RF. Complete knee dislocation. A follow-up study of operative treatment. Clin Orthop Relat Res. 1985 Sep;(198):94-101. 9. Shapiro MS, Freedman EL. Allograft reconstruction of the anterior and posterior cruciate ligaments after traumatic knee dislocation. Am J Sports Med. 1995 SepOct;23(5):580-7. 10. Shelbourne KD, Porter DA, Clingman JA, McCarroll JR, Rettig AC. Low-velocity knee dislocation. Orthop Rev. 1991 Nov;20(11):995-1004. 11. Wascher DC, Becker JR, Dexter JG, Blevins FT. Reconstruction of the anterior and posterior cruciate ligaments after knee dislocation. Results using fresh-frozen nonirradiated allografts. Am J Sports Med. 1999 MarApr;27(2):189-96. 12. Cole BJ, Harner CD. The multiple ligament injured knee. Clin Sports Med. 1999 Jan;18(1):241-62.
191 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
13. Yeh WL, Tu YK, Su JY, Hsu RW. Knee dislocation: treatment of high-velocity knee dislocation. J Trauma. 1999 Apr;46(4):693-701. 14. Levy BA, Dajani KA, Whelan DB, Stannard JP, Fanelli GC, Stuart MJ, Boyd JL, MacDonald PA, Marx RG. Decision making in the multiligament-injured knee: an evidence-based systematic review. Arthroscopy. 2009 Apr;25(4):430-8. 15. Chhabra A, Cha PS, Rihn JA, Cole B, Bennett CH, Waltrip RL, Harner CD. Surgical management of knee dislocations. Surgical technique. J Bone Joint Surg Am. 2005 Mar;87(Pt 1)(Suppl 1):1-21. 16. Howells NR, Brunton LR, Robinson J, Porteus AJ, Eldridge JD, Murray JR. Acute knee dislocation: an evidence based approach to the management of the multiligament injured knee. Injury. 2011 Nov;42(11):1198-204. Epub 2010 Dec 14. 17. Richter M, Lobenhoffer P. Chronic posterior knee dislocation: treatment with arthrolysis, posterior cruciate ligament reconstruction and hinged external fixation device. Injury. 1998 Sep;29(7):546-9. 18. Simonian PT, Wickiewicz TL, Hotchkiss RN, Warren RF. Chronic knee dislocation: reduction, reconstruction, and application of a skeletally fixed knee hinge. A report of two cases. Am J Sports Med. 1998 Jul-Aug;26(4):591-6. 19. Meyers MH, Moore TM, Harvey JP Jr. Traumatic dislocation of the knee joint. J Bone Joint Surg Am. 1975 Apr;57(3):430-3. 20. Stannard JP, Sheils TM, McGwin G, Volgas DA, Alonso JE. Use of a hinged external knee fixator after surgery for knee dislocation. Arthroscopy. 2003 Jul-Aug;19(6):626-31. 21. Fitzpatrick DC, Sommers MB, Kam BC, Marsh JL, Bottlang M. Knee stability after articulated external fixation. Am J Sports Med. 2005 Nov;33(11):1735-41. Epub 2005 Aug 10. 22. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA.1996;276:637-9. 23. Stannard JP, Schmidt AH, Kregor PJ. Surgical Treatment of Orthopaedic Trauma. New York: Thieme Medical Publishers; 2007. 24. Stannard JP, Hammond A, Tunmire D, Clayton M, Johnson C, Moura C. Determining the isometric point of the knee. J Knee Surg. 2012 Mar;25(1):71-4. 25. Dennis DA, Komistek RD, Mahfouz MR, Walker SA, Tucker A. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin Orthop Relat Res. 2004 Nov;(428):180-9. 26. Wright DG, Covey DC, Born CT, Sadasivan KK. Open dislocation of the knee. J Orthop Trauma. 1995 Apr;9(2):135-40.
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
27. Lysholm J, Gillquist J. Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. Am J Sports Med. 1982 MayJun;10(3):150-4. 28. Irrgang JJ, Anderson AF, Boland AL, Harner CD, Kurosaka M, Neyret P, Richmond JC, Shelborne KD. Development and validation of the International Knee Documentation Committee subjective knee form. Am J Sports Med. 2001 SepOct;29(5):600-13. 29. Patel AA, Donegan D, Albert T. The 36-item short form. J Am Acad Orthop Surg. 2007 Feb;15(2):126-34. 30. Schenck RC Jr. The dislocated knee. Instr Course Lect. 1994;43:127-36. 31. Tull F, Borrelli J Jr. Soft-tissue injury associated with closed fractures: evaluation and management. J Am Acad Orthop Surg. 2003 Nov-Dec;11(6):431-8. 32. Gustilo RB, Merkow RL, Templeman D. The management of open fractures. J Bone Joint Surg Am. 1990 Feb;72(2):299-304. 33. Baker SP, O’Neill B, Haddon W Jr, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974 Mar;14(3):187-96. 34. Engebretsen L, Risberg MA, Robertson B, Ludvigsen TC, Johansen S. Outcome after knee dislocations: a 2-9 years follow-up of 85 consecutive patients. Knee Surg Sports Traumatol Arthrosc. 2009 Sep;17(9):1013-26. Epub 2009 Jul 16. 35. Strobel MJ, Schulz MS, Petersen WJ, Eichhorn HJ. Combined anterior cruciate ligament, posterior cruciate ligament, and posterolateral corner reconstruction with autogenous hamstring grafts in chronic instabilities. Arthroscopy. 2006 Feb;22(2):182-92. 36. Levy BA, Fanelli GC, Whelan DB, Stannard JP, MacDonald PA, Boyd JL, Marx RG, Stuart MJ; Knee Dislocation Study Group. Controversies in the treatment of knee dislocations and multiligament reconstruction. J Am Acad Orthop Surg. 2009 Apr;17(4):197-206. 37. Bin SI, Nam TS. Surgical outcome of 2-stage management of multiple knee ligament injuries after knee dislocation. Arthroscopy. 2007 Oct;23(10):1066-72. 38. Miller MD. Re: primary repair of knee dislocations: results in 25 patients (28 knees) at a mean follow-up of four years. J Orthop Trauma. 2007 Feb;21(2):97; discussion 97-8. 39. Jenkins PJ, Clifton R, Gillespie GN, Will EM, Keating JF. Strength and function recovery after multiple-ligament reconstruction of the knee. Injury. 2011 Dec;42(12):1426-9. Epub 2011 Apr 09.
BY
T HE J OURNAL
OF
B ONE
AND J OINT
S URGERY, I NCORPORATED
Hinged External Fixation in the Treatment of Knee Dislocations A Prospective Randomized Study James P. Stannard, MD, Clayton W. Nuelle, MD, Gerald McGwin, PhD, and David A. Volgas, MD Investigation performed at the Department of Orthopaedic Surgery, University of Missouri, Columbia, Missouri, and the University of Alabama at Birmingham, Birmingham, Alabama
Background: Our hypothesis was that patients treated with hinged external fixators as an adjunct to multiple-ligament reconstruction would have fewer reconstruction failures than patients treated without external fixation. Methods: In this prospective randomized study, patients with a knee dislocation either underwent ligament reconstruction with placement of an external hinged knee brace following surgery (Group A) or underwent ligament reconstruction with placement of a hinged external fixator (Compass Knee Hinge) for six weeks instead of the brace (Group B). The patients were followed clinically and were evaluated with physical examination, Lysholm and International Knee Documentation Committee knee scores, visual analog scale pain scores, and status regarding return to work and activities. Results: One hundred patients with 103 knee dislocations were enrolled. Seventy-seven patients with seventy-nine dislocations (thirty-two in Group A and forty-seven in Group B), with a minimum follow-up interval of twelve months, were available for evaluation. The mean duration of follow-up was thirty-nine months (range, twelve to eighty-six months). Nine patients (29%) in Group A had failed reconstructions compared with seven (15%) in Group B (p = 0.15). Group-A patients had twenty-two (21%) of 105 reconstructed individual ligaments fail compared with eleven (7%) of 157 reconstructed ligaments in Group B. The difference in ligament failure was significant (p < 0.001; power > 0.8), with more favorable results for the patients managed with the external fixation. Conclusions: Hinged external fixation as a supplement to reconstruction following knee dislocation was associated with fewer failed ligament reconstructions compared with external bracing. Patients presenting with highly unstable knee dislocations should be considered for hinged external fixation to supplement initial reconstructive procedures. Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.
Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.
D
islocation of the knee typically follows a high-energy blunt trauma rather than resulting from an athletic competition injury. Knee dislocations are defined as multiple-ligament injuries, which typically involve both cruciate ligaments. In most cases, the knee has spontaneously re-
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
J Bone Joint Surg Am. 2014;96:184-91
d
http://dx.doi.org/10.2106/JBJS.L.01603
duced and is not dislocated at the time of presentation to the emergency department1-5. Authors of early reports debated the benefits of repair compared with reconstruction of ligaments and acute or delayed operative management and subsequently focused primarily on A commentary by Brett D. Owens, MD, is linked to the online version of this article at jbjs.org.
185 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
TABLE I Characteristics of Patients with Complete Follow-up Total (N = 77) Sex (no. of patients) Male Female
Group A (Control)* (N = 31)
59 18
25 6
Group B (Study)* (N = 46)
34 12
Mean age (yr)
35.0
34.1
35.6
Dislocations Left Right
79 42 37
32 14 18
47 28 19
*Group A had forty-five patients with forty-seven dislocations, and Group B had fifty-five patients with fifty-six dislocations. Follow-up data were available for thirty-one patients (32 dislocations) in Group A and forty-six patients (forty-seven dislocations) in Group B.
achieving ligamentous stability6-8. The rigid stability, however, consequently resulted in sacrificing motion by using casts, braces, or transarticular Steinmann pins during the postoperative period. While this approach frequently resulted in stability, it was also frequently associated with disabling loss of motion and pain as a result of arthrofibrosis. Several authors have noted unfavorable results in patients with multiple injuries who have limited early rehabilitation capability9-11. Most authors believe that a stiff, painful knee leads to more long-term morbidity and dysfunction than an unstable knee8,10,12-16. As a result, many authors now strongly advocate early operative treatment and aggressive motion and rehabilitation despite the risk of failure of ligament repairs or reconstructions. Clinical outcomes following knee dislocations are frequently poor, with arthrofibrosis, stiffness, pain, and recurrent instability being the most frequent complications6,8-10,12,13. Surgeons must choose between two postoperative approaches. The first is early aggressive motion with the potential decrease in arthrofibrosis and improved articular cartilage nutrition but an increased risk of ligament instability. The second option is to restrict knee motion and allow early healing of the reconstructions with a decreased risk of ligament instability, despite an increased risk of arthrofibrosis. The Compass Universal Hinge external fixator (Smith & Nephew, Memphis, Tennessee) was originally developed for use at the elbow to allow early motion without placing articular and/ or ligamentous repairs at increased stress and has been modified to create the Compass Knee Hinge. Two initial case reports documented the use of the Compass Knee Hinge with knee
dislocations17,18. There has been ongoing discussion regarding its use following ligamentous reconstruction after knee dislocations19-21. We hypothesized that knee dislocations supplemented with the Compass Knee Hinge compared with a control group would have equivalent final knee range of motion with fewer failures for ligament reconstructions. Materials and Methods
O
ne hundred patients with 103 knee dislocations were enrolled from August 2000 to January 2007. A power analysis was performed prior to data collection using the Stata program sampsi calculation (StataCorp, College Station, Texas). We assumed a failure rate in the control group of 24% and a failure rate in the Compass Knee Hinge group of 8% on the basis of prior retrospective data. When significance was set at 0.05 and power was set at 0.8, ninety-four subjects were required in each group. Knee dislocations commonly vary between two and four ligament groups (anterior cruciate ligament [ACL], posterior cruciate ligament [PCL], posterolateral corner, or posteromedial corner) torn per patient, with a patient having an average of three torn ligaments. Fifty patients in each study arm would then yield approximately 150 ligament groups per study arm, allowing for an appropriately powered study, even if 20% to 30% of patients did not achieve two years of follow-up. The study was designed as a prospective randomized evaluation of patients with knee dislocations, and it was approved and monitored by the institutional review board at our institution. The trial was registered with ClinicalTrials.gov (NCT00582517).
Study Design Patients were randomized into one of two groups using a computer-generated block randomization scheme. The randomization was based on the order of patient presentation, so each patient was individually randomized to a treatment group, regardless of overall injury severity score or the involvement of multiple
TABLE II Knee Dislocation Grades at Presentation* Grade
Total No. of Knees
Group A (no. of knees)
Group B (no. of knees)
I
2
0
2
II
1
1
0
III
38
13
25
IV
38
18
20
30
*The knees were classified according to the system described by Schenck .
186 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
d
d
IN THE
T R E AT M E N T
OF
d
TABLE III Ligament Reconstruction Failures Total No. of Knees
ACL (no. of ligaments)
PCL (no. of ligaments)
Posterolateral Corner (no. of ligaments)
Posteromedial Corner (no. of ligaments)
Group A
9
9
4
6
3
Group B
7
4
0
5
2
injuries of the extremity. Group-A patients had a staged reconstruction of the acute knee dislocation with placement of a hinged knee brace during the postoperative period. Our preferred protocol involved reconstructing the PCL, the posterolateral corner, and the posteromedial corner three or four weeks following the dislocation. The ACL was reconstructed six weeks following the initial surgery. Group-B patients were treated with the same protocol following the knee dislocation, with the exception that they had a Compass Knee Hinge placed on the injured leg at the end of the initial surgical procedure. The hinge was retained for approximately six weeks and then was removed at the time of ACL reconstruction. All patients underwent surgical treatment and follow-up evaluation by the senior surgeon (J.P.S.). The CONSORT (Consolidated Standards of Reporting Trials) statement has been developed to improve the reporting of randomized controlled 22 trials . We obtained data mentioned in the statement and the flow diagram (Fig. 1).
Surgical Technique Our preferred protocol involved a staged ligament reconstruction, beginning with the PCL, the posterolateral corner, and the posteromedial corner. The ACL is reconstructed six weeks following the initial surgery. Our preferred surgical techniques for each ligament included a tibial inlay technique with two femoral tunnels using Achilles tendon allograft for the PCL; a two-tailed technique for the posterolateral corner anatomically reconstructing the popliteus, fibular collateral ligament, and popliteofibular ligament using tibialis anterior allograft; a medial cruciate ligament (MCL) and posterior oblique ligament reconstruction using anterior or posterior tibialis allograft for the posteromedial corner; and finally an ACL reconstruction using either a bone-patellar tendonbone or hamstring autograft or allograft. The Compass Knee Hinge is an external fixator with a hinge that is placed at the center of rotation of the joint. It consists of two carbon-fiber 5/8 rings bolted to multihole Rancho Cubes that connect to 5 or 6-mm external fixator
Fig. 1
CONSORT diagram shows the flow of patients in the study.
187 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
TABLE IV Activity Level and Work Status Before and After Injury
Preinjury activity* (no. [%] of patients) Sedentary Hobbies Recreational athlete Competitive athlete Postinjury activity at 24 mo† (no. [%] of patients) Level less than preinjury status Previous level of activity Work status‡ No duty Light duty Full duty
Total
Group A (Control)
Group B (Study)
14 (21.2) 22 (33.3) 26 (39.4) 4 (6.1)
5 (19.2) 11 (42.3) 10 (38.5) 0
9 (22.5) 11 (27.5) 16 (40) 4 (10.0)
15 (46.9) 17 (53.1)
5 (27.8) 13 (72.2)
10 (71.4) 4 (28.6)
6 (13.6) 4 (9.1) 34 (77.3)
2 (9.1) 2 (9.1) 18 (81.8)
4 (18.2) 2 (9.1) 16 (72.7)
*The groups include twenty-six patients in Group A and forty patients in Group B for whom data were available. †The groups include eighteen patients in Group A and fourteen in Group B who had complete data available at twenty-four months. ‡The groups include twenty-two patients in Group A and twenty-two patients in Group B who had data available at twenty-four months.
Schanz pins. The rings are connected by two calibrated hinges that are placed on the medial and lateral sides of the joint. The placement of the entire apparatus is based on a centering wire that is temporarily placed at the isometric point on the femur. The hinge is centered by holes that accommodate this centering wire 23 (Fig. 2, Video 1 [online]) . The desired isometric point can be determined with fluoroscopy. With use of the fluoroscope, a perfect lateral view is obtained. The isometric point is located where a line drawn along the anterior aspect of the posterior femoral 24 cortex intersects with the Blumensaat line . This is a crucial step in the procedure, as precise placement of the centering wire is critical to the application and subsequent function of the apparatus. Malalignment of the wire and subse-
quently of the entire apparatus can occur. It is essential to obtain excellent fluoroscopy and to be precise with placement of the wire at the isometric point. After the centering pin is placed, the hinge is completed by placing external fixator pins into the femur and tibia. Generally, two 6-mm pins are placed into the femur. One is placed medially and inferiorly off the ring with a one to three-hole Rancho Cube, and one is placed laterally and inferiorly off of the ring with a two-hole Rancho Cube. Three 5-mm pins are normally placed into the tibia through Rancho Cubes placed inferiorly from the inferior ring. Precise and accurate placement of the hinge is critical to its function to limit knee motion. The Compass Knee Hinge limits motion in the coronal and axial planes, subsequently imparting some limitation to the screw home mechanism in normal knees: axial external rotation
Fig. 2
Intraoperative photograph demonstrating placement of the Compass Knee Hinge with a centering wire placed directly through the isometric point on the femur.
188 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
Fig. 3
Intraoperative photograph demonstrating tibial pin location and final application with the Compass Knee Hinge in place. of the femur on the tibia during knee flexion and internal rotation of the femur on 25 the tibia during knee extension with a static medial femoral condyle . This decreases the overall stress placed on the reconstructed ligaments during healing. The hinge allows motion in the sagittal plane, subsequently allowing physiologic rolling motion of the tibiofemoral joint in flexion and extension. This allows for aggressive rehabilitation without sacrificing stability. A critical issue in placing the Compass Knee Hinge is the timing of the steps in the application on a patient with a multiligamentous knee injury. If repair or reconstruction of either the posteromedial or posterolateral corner is part of the planned surgical procedure, the centering wire should be placed prior to the repair or reconstruction of the injured corner. This is important because the wire is placed through the area of the corner repair or reconstruction, often requiring a number of passes to get the position exactly right. The potential to damage a repair of the posterolateral corner is unacceptably high if the centering pin is placed following repair. The solution is to place the centering wire first, mount the hinge, and place the two femoral external fixation pins through Rancho Cubes. The hinge is then detached from the pins, leaving the pins and Rancho Cubes in place in the femur only. The remainder of the hinge is placed on the back table until the completion of the case. The centering wire is removed, and the remainder of the repair and/or reconstruction of the knee is accomplished. Following closure of all wounds, the hinge is remounted on the two femoral pins through the same holes, ensuring an isometric placement on the knee. The tibial pins are then placed to complete application of the hinge (Fig. 3). The tibial pins should not be placed until the end of the case because they are in the way during the ligamentous reconstruction and/or repairs. Our postoperative protocol for both groups involved a progressive attempt to regain knee motion following the initial procedure. Continuous passive motion machines were used in all patients beginning on postoperative day 1. Patient-controlled analgesia pumps or epidurals were used for early postoperative pain management. Weight-bearing was progressed as tolerated with the knee locked in extension. Rehabilitation included progressive increases in knee motion with quadriceps and hamstring strengthening.
Data obtained included a physical examination for knee motion, stability, and visual analog scale pain score. Stability was categorized as 0 for a knee with no ligamentous laxity, as 11 for 1 to 5 mm of laxity, as 21 for 6 to 10 mm, and as 31 for >10 mm26. Clinical success was defined as a stable knee with either 0 or 11 laxity on examination. Failure was defined as either 21 or 31 laxity on clinical examination. We also noted the status with regard to return to employment and recreational activity. Given the high percentages of poor outcomes following 6,8-10,12,13 treatment of knee dislocations reported in the literature , we classified patients with regard to postinjury return to activities into one of two simplified categories: a full return to their previous level or a return to any level of activity that they deemed to be less than a full return. Return to work was based on three possible categories: no duty, a return to light duty, or a return to the previous level of full duty. Additionally, we evaluated the study subjects with outcome instruments that related to either the knee or the overall health status. Knee-specific instruments 27 included the Lysholm score and the International Knee Documentation Com28 mittee (IKDC) objective and subjective scores . Comprehensive health status was 29 documented using the Short Form-36 (SF-36) health inventory . Data were obtained regarding patient demographics, mechanism of injury, vascular and neurologic injury, associated injuries, associated medical problems, and the need for admission to the intensive care unit following injury. The knee dislo30 cation was classified using the anatomic classification scheme designed by Schenck . Finally, the condition of the soft tissue around the knee was documented at the time 31 32 of injury using the scoring systems of Tscherne and Gustilo et al. .
Statistics Clinical and outcome characteristics were compared between groups using the chi-square test and analysis of variance for categorical and continuous measures, respectively. For non-normally distributed continuous variables, nonparametric statistics were employed. The Fisher exact test was used for analyses of categorical variables when expected cell frequencies were less than five. P values of £0.05 (two-sided) were considered significant.
Source of Funding Postoperative Follow-up Evaluation Patients were scheduled for postoperative follow-up examinations at two, six, twelve, twenty-six, and fifty-two weeks and for annual evaluations following the first year.
Smith & Nephew provided some institutional funding to cover research nurse support for this study. No investigators received salary, consultation payments, or benefits for the completion of this work.
189 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
Results ne hundred patients with 103 knee dislocations gave informed consent for study enrollment. Forty-five patients with forty-seven dislocations were randomized to the control group (Group A), and fifty-five patients with fifty-six dislocations were randomized to the study group (Group B). A minimum twelve-month duration of follow-up was obtained for thirtyone patients (thirty-two knee dislocations; 68.9%) in Group A and forty-six patients (forty-seven knee dislocations; 83.6%) in Group B. A minimum twenty-four-month follow-up period was obtained for twenty-two patients (twenty-three knee dislocations; 48.9%) in Group A and twenty-two patients (twentythree knee dislocations; 40%) in Group B. Overall, the mean duration of follow-up was thirty-nine months (range, twelve to eighty-six months). The patients in Group A had been followed for a mean of forty-three months (range, fifteen to seventy-five months), while those in Group B had been followed for a mean of thirty-five months (range, twelve to eighty-six months). The seventy-seven patients (77%) who were available for follow-up comprise the cohort for this study. Demographic data, including patient age, sex, and race, were similar between groups (Table I). Knee dislocation grades and mechanisms of injury were also similar between groups (Table II; see Appendix). The Injury Severity Score (ISS)33 ranged from 9 to 41 (mean,16.6). Group A had a mean ISS of 16.2, while Group B had a mean ISS of 16.6. More than 90% of the patients had a normal result on the vascular examination on admission and were followed using a selective arteriography protocol. Seven patients had abnormal results on the vascular examination, with an equal distribution between the groups. Nine patients had open knee dislocations, four in Group A and five in Group B. The majority of patients with closed dislocations (55%) had a Tscherne Grade-2 soft-tissue injury, with an equal distribution between the groups. Only two patients had a Tscherne Grade-3 injury, and both of them were in Group B. Failure of reconstruction was defined as either 21 or 31 laxity of a reconstructed ligament on clinical examination by the primary surgeon during routine follow-up or, in the case of ACL and PCL reconstructions, a side-to-side difference of >3 mm on examination of the ligaments with use of a KT-2000 arthrometer. The reconstructions failed in nine (28%) of the thirty-two knees in Group A and in seven (15%) of forty-seven knees in Group B; the difference was insignificant (p = 0.15). Twenty-two (21%) of 105 individual ligaments reconstructed in Group A failed, whereas eleven (7%) of 157 ligaments reconstructed in Group B failed (Table III). The difference in individual ligament failure was significant (p < 0.001), with more favorable results for the patients who had the external fixation. Group A also had more individual ligament failures per patient, with a mean of 2.44 ligament failures, compared with Group B, which had a mean of 1.57 ligament failures per patient. Final Lysholm knee scores were similar between the two groups. For patients with a minimum follow-up interval of twelve months, the mean score was 87.7 (range, 61 to 100) for Group A compared with 87.2 (range, 69 to 100) for Group B. For those
O
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
with a minimum follow-up of twenty-four months, the mean score was 90.1 for Group A compared with 89.9 for Group B. The IKDC subjective scores were similar between the groups for patients with a minimum follow-up period of twentyfour months. Group-A patients had a mean score of 56.8, while Group-B patients had a mean score of 49.5. The objective scores were also similar, with a normal knee in 50% of Group A and in 45% of Group B. A nearly normal knee was achieved in 25% of Group A and 36% of Group B. When these results are combined, 75% of the patients in Group A and 81% of the patients in Group B had acceptable results. Failing IKDC results (Class C [abnormal] or Class D [severely abnormal]) were present in 25% of Group-A patients and 18% of Group-B patients. Pain was a common finding, as evidenced by the individual visual analog scale scores. Patients in Group A had a mean score of 2.5 (range, 0 to 7) and those in Group B had a mean score of 2.8 (range, 0 to 8). At the time of the final follow-up, 86% of the patients had returned to work. In Group A, 82% of the patients had resumed full duty, 9% had returned to light duty, and 9% did not return to work. In Group B, 73% of the patients had returned to full duty, 9% had resumed light duty, and 18% did not return to work (Table IV). Knee range of motion was good at the final evaluation. The mean arc of motion was 1° to 128°. Group-A patients had a mean range of motion from 1° to 132°, while Group-B patients had a mean range from 2° to 124°. Discussion nee dislocations are severe injuries that are being recognized with increasing frequency as the definition is clarified and awareness of the frequency of spontaneous reduction is heightened5,11,13. Recent studies have noted that knee dislocations can occur without rupture of either the ACL1 or the PCL3,4, although bicruciate injuries are by far more common. Expanded definitions of knee dislocation include gross or radiographically proven dislocation, injury of multiple knee ligaments with multidirectional instability following highenergy trauma, or rupture of both the ACL and PCL when no gross dislocation could be identified13. Associated injuries frequently include acetabular or tibial plateau fractures and popliteal artery and peroneal nerve injuries7-13,16. Open injuries are infrequent but are associated with extensive ligamentous damage and a high rate of vascular and neurologic involvement when they occur26. Nonoperative treatment of multiple ligament injuries has resulted in poor results in up to 77% of patients, and recent literature has shown improved benefit with surgical treatment16,19,34,35. Most authors currently advocate aggressive surgical treatment of knee dislocations6-8,12,13,15,19,24,36. Early surgical intervention is advocated, with both single surgery and staged ligament reconstruction as options16,24,37,38. Arthrofibrosis and the associated loss of knee motion with pain is a major complication associated with multiligamentous knee injuries. Noyes and Barber-Westin reported motion problems in 71% of their patients with acute injuries requiring manipulation under anesthesia6. Shapiro and
K
190 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
Freedman reported arthrofibrosis in 60% of their patients9. Sisto and Warren noted that nine of nineteen patients required manipulation under anesthesia8. Aggressive early knee motion and rehabilitation may help to reduce these problems but may also place repaired and/or reconstructed ligaments under stresses that may be associated with a higher failure rate and recurrent instability. The Compass Knee Hinge provides rigid stability in all planes of motion except the sagittal plane, allowing the patient to obtain flexion and extension. There is minimal rotational stress placed on reconstructed ligaments when the hinge is in place. The hinge also helps to prevent posterior sag during the acute healing phase. Both studies by Richter and Lobenhoffer17 and Simonian et al.18 described patients with chronic knee dislocations in whom a Compass Knee Hinge was successfully used to treat chronic posterior subluxation or dislocation. Our results demonstrate that progressive rehabilitation can be performed with a Compass Knee Hinge without sacrificing stability. Group-A patients had a failure rate of 21% (twenty-two of 105 reconstructed ligaments), whereas Group-B patients had a significantly lower failure rate of 7% (eleven of 157 reconstructed ligaments) (p < 0.05). When patients did experience a ligament failure, they were less likely to have multiple ligaments fail if they were treated with a Compass Knee Hinge than if they were treated with a hinged knee brace. There are several disadvantages to using the Compass Knee Hinge, which must be balanced against the improved stability. The device is expensive, takes approximately thirty minutes to apply, and frequently causes pain at the site of the femoral pins with flexion beyond approximately 60°. Outcome scores were relatively good in both groups after reconstruction of the knees that had recurrent instability. In our study, 86% of patients returned to work, with ranges from 73% to 82% in each treatment arm returning to full duty. This is similar to reported data demonstrating that up to 80% of patients returned to preinjury work status and had overall improved outcome scores after reconstruction24,39. Limitations There are limitations to our study. First, a number of patients were lost to long-term follow-up, thus limiting our overall
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
sample size. The mean duration of follow-up for each treatment group was forty-three months for Group A (control) and thirty-five months for Group B, and our final sample size was seventy-seven patients, which we believe are strong values for long-term follow-up in a trauma population. In addition, the percentage of patients lost to follow-up in Group A (31%) was higher than that of Group B (16.4%). This difference could change some of the data in terms of the overall number of reconstruction failures. Also, despite losing some patients to follow-up, the study was still adequately powered when based on individual ligament reconstructions (power of >0.8). Finally, there is also the potential limitation of a lack of interobserver reliability in grading of ligament laxity, but the primary surgeon was the sole examiner to grade laxity and subsequent ligament failures during follow-up. In conclusion, patients treated with a Compass Knee Hinge had a significantly lower failure rate of repaired and reconstructed ligaments following surgical procedures compared with the control group. We concluded that the Compass Knee Hinge functioned well as an adjunct to aggressive reconstruction following knee dislocation and may be considered as part of the treatment of highly unstable knee dislocations. Appendix A table showing data on the mechanisms of injury is available with the online version of this article as a data supplement at jbjs.org. n
James P. Stannard, MD Clayton W. Nuelle, MD David A. Volgas, MD Department of Orthopaedic Surgery, University of Missouri, 1100 Virginia Avenue, DC953.00, Columbia, MO 65212. E-mail address for J.P. Stannard: [email protected] Gerald McGwin, PhD University of Alabama at Birmingham, 510 South 20th Street, FOT 960, Birmingham, AL 35294-3409
References 1. Bellabarba C, Bush-Joseph CA, Bach BR Jr. Knee dislocation without anterior cruciate ligament disruption. A report of three cases. Am J Knee Surg. 1996 Fall;9(4):167-70. 2. Bratt HD, Newman AP. Complete dislocation of the knee without disruption of both cruciate ligaments. J Trauma. 1993 Mar;34(3):383-9. 3. Cooper DE, Speer KP, Wickiewicz TL, Warren RF. Complete knee dislocation without posterior cruciate ligament disruption. A report of four cases and review of the literature. Clin Orthop Relat Res. 1992 Nov;(284):228-33. 4. Shelbourne KD, Pritchard J, Rettig AC, McCarroll JR, Vanmeter CD. Knee dislocations with intact PCL. Orthop Rev. 1992 May;21(5):607-8: 610-1. 5. Wascher DC, Dvirnak PC, DeCoster TA. Knee dislocation: initial assessment and implications for treatment. J Orthop Trauma. 1997 Oct;11(7):525-9. 6. Noyes FR, Barber-Westin SD. Reconstruction of the anterior and posterior cruciate ligaments after knee dislocation. Use of early protected postoperative motion to decrease arthrofibrosis. Am J Sports Med. 1997 Nov-Dec;25(6):769-78.
7. Roman PD, Hopson CN, Zenni EJ Jr. Traumatic dislocation of the knee: a report of 30 cases and literature review. Orthop Rev. 1987 Dec;16(12):917-24. 8. Sisto DJ, Warren RF. Complete knee dislocation. A follow-up study of operative treatment. Clin Orthop Relat Res. 1985 Sep;(198):94-101. 9. Shapiro MS, Freedman EL. Allograft reconstruction of the anterior and posterior cruciate ligaments after traumatic knee dislocation. Am J Sports Med. 1995 SepOct;23(5):580-7. 10. Shelbourne KD, Porter DA, Clingman JA, McCarroll JR, Rettig AC. Low-velocity knee dislocation. Orthop Rev. 1991 Nov;20(11):995-1004. 11. Wascher DC, Becker JR, Dexter JG, Blevins FT. Reconstruction of the anterior and posterior cruciate ligaments after knee dislocation. Results using fresh-frozen nonirradiated allografts. Am J Sports Med. 1999 MarApr;27(2):189-96. 12. Cole BJ, Harner CD. The multiple ligament injured knee. Clin Sports Med. 1999 Jan;18(1):241-62.
191 TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG V O LU M E 96-A N U M B E R 3 F E B R UA RY 5, 2 014 d
d
d
13. Yeh WL, Tu YK, Su JY, Hsu RW. Knee dislocation: treatment of high-velocity knee dislocation. J Trauma. 1999 Apr;46(4):693-701. 14. Levy BA, Dajani KA, Whelan DB, Stannard JP, Fanelli GC, Stuart MJ, Boyd JL, MacDonald PA, Marx RG. Decision making in the multiligament-injured knee: an evidence-based systematic review. Arthroscopy. 2009 Apr;25(4):430-8. 15. Chhabra A, Cha PS, Rihn JA, Cole B, Bennett CH, Waltrip RL, Harner CD. Surgical management of knee dislocations. Surgical technique. J Bone Joint Surg Am. 2005 Mar;87(Pt 1)(Suppl 1):1-21. 16. Howells NR, Brunton LR, Robinson J, Porteus AJ, Eldridge JD, Murray JR. Acute knee dislocation: an evidence based approach to the management of the multiligament injured knee. Injury. 2011 Nov;42(11):1198-204. Epub 2010 Dec 14. 17. Richter M, Lobenhoffer P. Chronic posterior knee dislocation: treatment with arthrolysis, posterior cruciate ligament reconstruction and hinged external fixation device. Injury. 1998 Sep;29(7):546-9. 18. Simonian PT, Wickiewicz TL, Hotchkiss RN, Warren RF. Chronic knee dislocation: reduction, reconstruction, and application of a skeletally fixed knee hinge. A report of two cases. Am J Sports Med. 1998 Jul-Aug;26(4):591-6. 19. Meyers MH, Moore TM, Harvey JP Jr. Traumatic dislocation of the knee joint. J Bone Joint Surg Am. 1975 Apr;57(3):430-3. 20. Stannard JP, Sheils TM, McGwin G, Volgas DA, Alonso JE. Use of a hinged external knee fixator after surgery for knee dislocation. Arthroscopy. 2003 Jul-Aug;19(6):626-31. 21. Fitzpatrick DC, Sommers MB, Kam BC, Marsh JL, Bottlang M. Knee stability after articulated external fixation. Am J Sports Med. 2005 Nov;33(11):1735-41. Epub 2005 Aug 10. 22. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA.1996;276:637-9. 23. Stannard JP, Schmidt AH, Kregor PJ. Surgical Treatment of Orthopaedic Trauma. New York: Thieme Medical Publishers; 2007. 24. Stannard JP, Hammond A, Tunmire D, Clayton M, Johnson C, Moura C. Determining the isometric point of the knee. J Knee Surg. 2012 Mar;25(1):71-4. 25. Dennis DA, Komistek RD, Mahfouz MR, Walker SA, Tucker A. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin Orthop Relat Res. 2004 Nov;(428):180-9. 26. Wright DG, Covey DC, Born CT, Sadasivan KK. Open dislocation of the knee. J Orthop Trauma. 1995 Apr;9(2):135-40.
H I N G E D E X T E R N A L F I X AT I O N K N E E D I S L O C AT I O N S
IN THE
T R E AT M E N T
OF
27. Lysholm J, Gillquist J. Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. Am J Sports Med. 1982 MayJun;10(3):150-4. 28. Irrgang JJ, Anderson AF, Boland AL, Harner CD, Kurosaka M, Neyret P, Richmond JC, Shelborne KD. Development and validation of the International Knee Documentation Committee subjective knee form. Am J Sports Med. 2001 SepOct;29(5):600-13. 29. Patel AA, Donegan D, Albert T. The 36-item short form. J Am Acad Orthop Surg. 2007 Feb;15(2):126-34. 30. Schenck RC Jr. The dislocated knee. Instr Course Lect. 1994;43:127-36. 31. Tull F, Borrelli J Jr. Soft-tissue injury associated with closed fractures: evaluation and management. J Am Acad Orthop Surg. 2003 Nov-Dec;11(6):431-8. 32. Gustilo RB, Merkow RL, Templeman D. The management of open fractures. J Bone Joint Surg Am. 1990 Feb;72(2):299-304. 33. Baker SP, O’Neill B, Haddon W Jr, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974 Mar;14(3):187-96. 34. Engebretsen L, Risberg MA, Robertson B, Ludvigsen TC, Johansen S. Outcome after knee dislocations: a 2-9 years follow-up of 85 consecutive patients. Knee Surg Sports Traumatol Arthrosc. 2009 Sep;17(9):1013-26. Epub 2009 Jul 16. 35. Strobel MJ, Schulz MS, Petersen WJ, Eichhorn HJ. Combined anterior cruciate ligament, posterior cruciate ligament, and posterolateral corner reconstruction with autogenous hamstring grafts in chronic instabilities. Arthroscopy. 2006 Feb;22(2):182-92. 36. Levy BA, Fanelli GC, Whelan DB, Stannard JP, MacDonald PA, Boyd JL, Marx RG, Stuart MJ; Knee Dislocation Study Group. Controversies in the treatment of knee dislocations and multiligament reconstruction. J Am Acad Orthop Surg. 2009 Apr;17(4):197-206. 37. Bin SI, Nam TS. Surgical outcome of 2-stage management of multiple knee ligament injuries after knee dislocation. Arthroscopy. 2007 Oct;23(10):1066-72. 38. Miller MD. Re: primary repair of knee dislocations: results in 25 patients (28 knees) at a mean follow-up of four years. J Orthop Trauma. 2007 Feb;21(2):97; discussion 97-8. 39. Jenkins PJ, Clifton R, Gillespie GN, Will EM, Keating JF. Strength and function recovery after multiple-ligament reconstruction of the knee. Injury. 2011 Dec;42(12):1426-9. Epub 2011 Apr 09.
