Chronic Knee Dislocation: Reduction, Reconstruction, and Application of a Skeletally Fixed Knee Hinge

Transcription

1 /98/ $02.00/0 THE AMERICAN JOURNAL OF SPORTS MEDICINE, Vol. 26, No American Orthopaedic Society for Sports Medicine Chronic Knee Dislocation: Reduction, Reconstruction, and Application of a Skeletally Fixed Knee Hinge A Report of Two Cases Peter T. Simonian,* MD, Thomas L. Wickiewicz, MD, Robert N. Hotchkiss, MD, and Russell F. Warren, MD From The Sports Medicine Service, The Hospital for Special Surgery, Affiliated with The New York Hospital-Cornell University Medical College, New York, New York A chronic knee dislocation, although extremely rare, is a difficult problem that requires complex management. There are two opposing goals after catastrophic ligament knee injuries: obtaining stability and obtaining range of motion. Achieving both of these goals can be very difficult. 1,2,5,7,10,12 Multiple ligament reconstructions using allograft tissues are currently the recommended method of dealing with acute unstable knee injuries. 1,4,11 Despite modern techniques, recurrent knee laxity or stiffness can be problematic. 1,2,5,7,10,12 These problems are amplified in the case of a chronic fixed dislocation. Recently, there was a report of delayed reduction without reconstruction 24 weeks after a traumatic posterior knee dislocation. 3 The patient was treated with open reduction, maintenance of the reduction with Steinman pins, and placement of the limb in a cylinder cast for 12 weeks. Reconstruction was not attempted because of compromised articular surfaces. The final result yielded a range of motion from 5 to 40. To reduce the chronically dislocated knee, complete releases of scar tissue and capsule are required, placing increased stresses on the reconstruction. In an attempt to decrease stress on the newly reconstructed ligaments and still allow knee motion, a skeletally fixed hinge system can be used. Because of the degree of instability, a limited period of hinge application is thought to be helpful in reestablishing stable motion. For acceptable use, the hinge should provide an anatomic range of motion, offer ease of application, and be tolerated by the patient. We are reporting two cases of chronic, fixed posterior * Address correspondence and reprint requests to Peter T. Simonian, MD, University of Washington Medical Center, Department of Orthopaedic Surgery, 1959 N. E. Pacific Street, Box , Seattle, WA One author has a commercial affiliation with a product named in this study. knee dislocations. Both were treated with extensive release and allograft ligament reconstruction with the addition of a skeletally fixed knee hinge for the first 6 weeks. CASE REPORTS Case 1 A 17-year-old female patient sustained a posterolateral knee dislocation while dismounting a ride at the local fair. The knee was reduced and an arteriogram was reported to be normal. Magnetic resonance imaging revealed disruption of the ACL with an avulsed portion of the tibial plateau in the region of the ACL attachment. The PCL and lateral collateral ligament were also disrupted, and there was a large tibial bone contusion. The knee was reduced at this time. Two weeks after the injury the patient underwent reattachment of the avulsed ACL with suture anchors, repair of the posterior and anterior horns of the medial meniscus, reattachment of the anterior and central portions of the lateral meniscus, and repair of the posterolateral corner including the arcuate and fabellofibular ligaments and the lateral capsule. Her knee was placed in a cast for 5 weeks. At the end of this period, the patient was found to have a posterior knee dislocation and peroneal nerve deficit. Electromyographic testing confirmed these findings. Three months after the injury the patient was referred to our institution with a chronic knee dislocation (Fig. 1) and nerve injury. Examination at this time revealed a fixed posterior dislocation with a 15 arc of motion. In this fixed position, the knee seemed completely unstable to varus stress. The patient also demonstrated a complete common peroneal nerve injury. 591

2 592 Simonian et al. American Journal of Sports Medicine Four months after the injury the patient was taken to the operating room. Examination under anesthesia revealed a fixed posterolateral knee dislocation with a range of motion of 10 to 40. The patient underwent anteromedial parapatellar arthrotomy with complete lysis of adhesions and scar tissue and subsequent reduction of the knee. Extensive circumferential releases were required to reduce the tibia; the intercondylar and posterolateral tibial regions required the most debridement. The remains of the ACL and PCL were excised. During release of the lateral side, a suture was found and removed from around the peroneal nerve. A careful neurolysis was performed. The remains of the lateral collateral ligament and the popliteal tendon were removed. The medial and lateral menisci were almost completely absent. The damaged peripheral meniscal rim was scarred to the posterior capsule. The meniscofemoral ligament was resected. There was also scarring between the distal anterior femur and the extensor mechanism that was released, allowing knee flexion beyond 90. Open reconstruction of the ACL was performed using a bone-patellar tendon-bone allograft with endoscopic-type passage, and reconstruction of the PCL was done using a bone-patellar tendon-bone allograft with a transtibial and femoral single drill hole technique. This patient also had posterolateral reconstruction of the popliteus tendon and the lateral collateral ligament reconstructed with a split Y-type Achilles tendon allograft. The sequence of graft fixation was the PCL, ACL, and then the posterolateral split graft. The posterolateral graft was secured with the knee in 70 of flexion and neutral rotation. An anterior compartment release was performed. The peroneal nerve was explored and found to be encased in scar tissue with a suture through it. The lateral retinaculum was not closed because the lateral patellar retinaculum would be excessively tight. An external fixator compass hinge was then applied after determining the isometric points on the medial and lateral femoral condyles (see Hinge Application for details). Fluoroscopy was used to determine the adequacy of reduction of the knee throughout a range of motion from 0 to 90. The patient used a continuous passive motion machine and remained nonweightbearing. Almost 6 weeks after the reconstruction, the hinge was removed. Examination under anesthesia revealed a range of motion of 0 to 90 and excellent stability to anteroposterior stress as well as varus-valgus stress. At this point the patient began weightbearing and used a prefabricated functional ACL brace (Legend, DonJoy, Smith & Nephew, Inc., Carlsbad, California). Fifteen months after the reconstruction, the patient had a range of motion from 5 to 105. Stability was symmetric with the opposite side. The peroneal nerve had partially recovered. She now had Figure 1. Lateral (A) and AP (B) radiographs of the patient at the time of referral to our institution demonstrate chronic knee dislocation. These radiographs were taken 3 months after the injury and 10 weeks after the initial surgery. See text for details.

3 Vol. 26, No. 4, 1998 Chronic Knee Dislocation 593 the ability to actively dorsiflex and evert the foot, but the strength was decreased at about 4/5. An AP radiograph revealed symmetric reduction with a reduction in the medial joint space (Fig. 2A). Comparison lateral radiographs revealed nearly symmetric centering of the joint with the tibia slightly anterior relative to the femur at 90 of flexion (Fig. 2, B and C). Case 2 The second patient was also a 17-year-old woman. She sustained a posterior knee dislocation after stepping off a curb and sustaining a hyperextension injury of the knee. She was initially treated with closed reduction and cast immobilization. She was seen by an orthopaedic surgeon from another institution 5 months after the injury and had a fixed posterior knee dislocation. On examination, the physician was unable to reduce the knee. Examination also demonstrated instability of the PCL, ACL, and lateral collateral ligament. Magnetic resonance imaging revealed ACL, PCL, and lateral collateral ligament injuries as well as a displaced lateral meniscus tear. The patient was taken to the operating room by the outside surgeon. Examination under anesthesia revealed a grade IV posterior drawer test result, a positive sag sign, a posterolateral drawer sign, a reverse pivot shift, and increased external rotation. Fluoroscopic examination revealed 50% subluxation of the tibia throughout the range of motion. Arthroscopic examination revealed a disruption of the ACL and PCL and scarring of the displaced portion of the lateral meniscus. There was a large amount of posterior scar formation. The articular surfaces demonstrated no fullthickness lesions. The patient was referred to our institution 1 month later. Examination revealed a hollow in the anterior knee region and a range of motion of 10 to 90. The neurovascular examination was intact. The tibia could not be reduced in any position. There was gross posterior instability and an increased posterolateral spin both at 30 and 90 of flexion. The knee was stable to varus stress. The patella was mildly tender, and she could actively fire her quadriceps and hamstring muscles. Radiographs demonstrated a fixed posterior dislocation in extension and flexion (Fig. 3). Comparison lateral views of the uninvolved knee were also obtained. The MRI scans confirmed injury to the ACL, PCL, and posterolateral knee structures. Eight months after the injury, the patient was taken to the operating room. The examination under anesthesia revealed a range of motion from 10 to 110, with the knee dislocated throughout the range of motion. At surgery, the medial femoral condyle and medial and lateral tibial plateau articular surfaces were normal. There was grade I chondromalacia in the regions of the patella and femoral trochlea and grade II changes at the lateral femoral condyle. The medial meniscus was without injury. The lateral meniscus had become deformed because of the dislocated position of the knee. The patient underwent arthrotomy with extensive lysis of adhesions and scar tissue. Extensive releases were required to reduce the tibia; the intercondylar and posterior tibial regions required the most debridement, including release of the semimembranosus muscle. Open reconstruction of the ACL was done using a bonepatellar tendon-bone allograft with endoscopic-type passage, and reconstruction of the PCL was done using a Figure 2. Fourteen months after reduction and reconstruction with hinge application, the AP radiograph (A) reveals symmetric reduction with a reduction in the medial joint space. Comparison lateral radiographs reveal symmetric centering of the joint with the tibia slightly anterior relative to the femur at 90 of flexion between the injured (B) and noninjured (C) knees.

4 594 Simonian et al. American Journal of Sports Medicine The patient had an anterior compartment release and the peroneal nerve was explored and found to be intact. An external fixator compass hinge was then applied after determining the isometric points on the medial and lateral femoral condyles (see Hinge Application for details). Fluoroscopy was used to determine the adequacy of reduction of the knee throughout a range of motion. The patient used a continuous passive motion machine and remained nonweightbearing. Almost 6 weeks after the reconstruction, the hinge was removed. Examination under anesthesia revealed a range of motion of 0 to 100 with symmetric stability to anteroposterior and posterolateral stress. At this point, weightbearing was begun and she was given a prefabricated functional ACL brace (Legend). Three months after the reconstruction her range of motion had increased to 0 to 120 and she was able to bear full weight. At 30 of flexion, the patient had 5 mm of total anteroposterior translation with firm end points. At 90 of flexion, the drop back was 1 and the tibia was flush with the femur. She had no patellar entrapment. Six months after reconstruction she began early functional sports activities. Her range of motion was unchanged. The degree of joint translation at 30 and 90 was unchanged. At this point, she was weaned from her brace. Twelve months after the reconstruction, the examination findings were unchanged, the patient had returned to a moderate level of sports activities and had discontinued the use of the brace. Radiographs at this time revealed preservation of the joint space and excellent centering of the knee joint (Fig. 4) as compared with the uninvolved knee. HINGE DESCRIPTION Figure 3. In case 2, 6 months after the injury radiographs demonstrated a fixed posterior dislocation in extension (A) and flexion (B). bone-patellar tendon-bone allograft with a transtibial and femoral single drill hole technique. The patient also had posterolateral reconstruction of the popliteus tendon and the lateral collateral ligament reconstructed with a split Y-type Achilles tendon allograft. The sequence of graft fixation was the PCL, ACL, and then the posterolateral split graft. The posterolateral graft was secured with the knee in 70 of flexion and neutral rotation. The hinge we used was designed for the elbow (Compass Elbow Hinge; Smith & Nephew Orthopaedics Inc., Memphis, Tennessee). A prototype has been designed for universal application to the ankle, elbow, or knee (Compass Universal Hinge) (Fig. 5). The prototype has a polyetherimide body with 7 of fixed valgus angulation. To accommodate the anatomic valgus alignment of the joint, the superior and inferior ends of the hinge can be reversed by simply turning the hinge over for use on the left or right knee. The extension-flexion range when applied to the knee is 0 to 120. This diameter of the stainless steel rings will allow enough clearance for anterior thigh soft tissues in most patients. The hinges are single axis and designed for ease of application through a centering hole in each. HINGE APPLICATION To place the hinge in an optimal position, placement of an axis or centering pin is critical. The placement of the centering pin depends on finding the most isometric point on the medial and lateral femoral condyles. This is done by placing a pin in the middle of both the medial and lateral collateral ligament insertions on the tibia and fibula, respectively, 3 cm distal to the joint line. One end of a suture

5 Vol. 26, No. 4, 1998 Chronic Knee Dislocation 595 Figure 5. Photograph of a prototype hinge that has been designed for universal application to the ankle, elbow, or knee. is placed around each of these pins and the other end is placed proximally on the medial and lateral femoral condyles; the specific point is then identified on both the medial and lateral femoral condyles where the suture does not lengthen or shorten through a range of knee motion. Once these points are identified and marked on the femoral condyles, the centering pin is placed from lateral to medial through both of them. The hinge s centering holes are then placed over the centering pin to assure optimal placement of the hinge. Two 5.0-mm Schantz pins are placed in both the femur and tibia through the semicircular rings of the hinge to secure it to the bones. The semicircular rings allow for multiple choices for Schantz pin placement; attempts are made to avoid pin placement through the quadriceps muscle and its extensor mechanism. The hinge is always placed with the knee in full Figure 4. Twelve months after reduction and reconstruction with hinge application, the AP radiograph (A) demonstrates maintenance of the joint space. Comparison lateral radiographs (B and C) reveal symmetric centering of the joint with tibia slightly anterior relative to the femur in extension (B) and flexion (C).

6 596 Simonian et al. American Journal of Sports Medicine extension. This is done because the hinge is most reproducibly placed in vivo with the knee in full extension. DISCUSSION The two cases reported here are unique because they were chronic, fixed posterior knee dislocations that underwent reconstruction and skeletal hinge application. Henshaw et al. 3 reported a case of delayed reduction of 24 weeks of a traumatic knee dislocation without reconstruction. The patient was treated with open reduction, maintenance of the reduction with Steinman pins, and placement in a cylinder cast for 12 weeks. Reconstruction was not attempted because of compromised articular surfaces. The final result yielded a range of motion from 5 to 40. Each of the cases was approached in a systematic fashion. The first step was to reduce the knee joint. In both cases this required extensive releases about the knee, primarily in the posterior, lateral, and intercondylar regions. The next step was to obtain stability through a balanced reconstruction. This was achieved by first reestablishing the central hinge through reconstruction of the ACL and PCL maintaining the knee in the reduced neutral position with radiographic confirmation. Next the posterolateral instability was addressed. Because of the complete releases required, the final step was to provide a form of protection that still allowed a functional range of motion. This was provided by the skeletal hinge. The idea of protecting a knee and still allowing controlled motion with multiple disrupted ligaments or after ligament reconstruction is not new. 8,9 Hinges have been incorporated into casts. 8,9 Many hinged braces are also available. With modern reconstruction techniques after acutely reduced knee dislocations, these braces are often used successfully. The case of a chronic knee dislocation represents a more complicated situation. The need for application of the skeletal hinge was secondary to the extensive release of the scar tissue and capsule needed for reduction of the knee. By using skeletal fixation to fix the hinge, we postulated that mismatch between the knee and the single-axis hinge motion could be minimized, protecting the reconstruction. However, the problems of infection and patient tolerance with an external fixation system cannot be minimized. In the two cases presented here neither of these problems occurred. We used a hinge initially designed for the elbow (Compass Elbow Hinge). Application of any hinge system requires placement of the hinge at the most isometric point of the joint. Using customized knee braces, Regalbuto et al. 6 demonstrated that disruptions of anatomic motion were much larger when the brace hinges were offset 12 mm from the ideal placement. Posterior placement resulted in the lowest force and anterior placement the highest. They concluded that mismatch of knee motion to brace motion would probably lead to abnormal ligament lengths and tensions and other internal mechanical changes. Similar findings have been reported by others. 13 In the present study, accurate placement of the hinge was done with a centering pin. The technique we used for finding these isometric points, observing the changes in length of a suture as the knee is placed through a range of motion, has been used successfully in our clinical practice when reconstructing the medial and lateral ligaments. Also we chose to place the hinge with the knee in full extension; this was done because this is the technically most reproducible and easiest position in which to place the hinge on a patient. Placement of an external hinge must be accurate to allow the most anatomic knee motion. This is likely not possible with a simple hinge through the full range of motion. However, there may be a limited range of anatomic motion that a simple hinge will allow before knee and hinge mismatch occurs. The results of the two cases support this. REFERENCES 1. Almekinders LC, Logan TC: Results following treatment of traumatic dislocations of the knee joint. Clin Orthop 284: , Frassica FJ, Sim FH, Staeheli JW, et al: Dislocation of the knee. Clin Orthop 263: , Henshaw RM, Shapiro MS, Oppenheim WL: Delayed reduction of traumatic knee dislocation. A case report and literature review. Clin Orthop 330: , Meyers MH, Moore TM, Harvey JP Jr: Traumatic dislocation of the knee joint [Follow-up notes on article previously published in the Journal]. J Bone Joint Surg 57A: , Montgomery JB: Dislocation of the knee. Orthop Clin North Am 18: , Regalbuto MA, Rovick JS, Walker PS: The forces in a knee brace as a function of hinge design and placement. Am J Sports Med 17: , Roman PM, Hopson CN, Zenni EJ Jr: Traumatic dislocation of the knee: A report of 30 cases and literature review. Orthop Rev 16: , Salacz T, Nemes G, Wagner J: Postoperative treatment of knee ligament injuries using a hinged plaster cast [in Hungarian]. Magy Traumatol Orthop Helyreallito Seb 27: , Sandberg S, Nilsson B, Westlin N: Hinged cast after knee ligament surgery. Am J Sports Med 15: , Shields L, Mital M, Cave EF: Complete dislocation of the knee: Experience at the Massachusetts General Hospital. J Trauma 9: , Sisto DJ, Warren RF: Complete knee dislocation. A follow-up study of operative treatment. Clin Orthop 198: , Thomsen PB, Rud B, Jensen UH: Stability and motion after traumatic dislocation of the knee. Acta Orthop Scand 55: , Walker PS, Rovick JS, Robertson DD: The effects of knee brace hinge design and placement on joint mechanics. J Biomech 21: , 1988