Patients

Using a standardized protocol, a retrospective review of our patient databases was undertaken. This yielded 25 consecutive patients with knee dislocation who underwent surgical treatment by a single senior surgeon between January 2002 and October 2010. This study was approved by the ethics committee of the first affiliated hospital of Anhui Medical University, and the informed consents were obtained from all the participants. The diagnosis of knee dislocation was made on the basis of clinical signs and symptoms and magnetic resonance imaging. Exclusion criteria were as follows: open trauma, severe cranial or cerebral injury, vascular injury requiring emergency vascular surgery, associated fractures requiring external fixation, or initial treatment performed at another institution. Ultimately, 19 patients with 20 MLIKs were included in the present study, and all patients were treated surgically with single-stage in situ suture repair.

Surgical technique

All operations were performed 5–10 days after injury. The patient was placed under general anesthesia and positioned supine on the operating table. The uninjured leg was extended, and the hip and knee on the injured side were flexed to 90° and the lateral thigh supported by a solid baffle. General anesthesia with controlled hypotension was used, and no tourniquet was placed. Physical examination and routine arthroscopy were first performed to identify the ligaments injured (Fig. 1). The anterior cruciate ligament (ACL), posterior cruciate ligament (PCL), posterolateral corner (PLC), and medial collateral ligament (MCL) were repaired using the following open surgical technique.

After sterile preparation and draping, an anterior incision was created slightly off the midline to allow for ACL and PCL tunnel placement. The articular space was opened to remove blood clots, and the stumps of ACL and PCL were exposed. The anteromedial and posterolateral bundles of the ACL and PCL were carefully identified. They were then repaired using a running baseball cross stitch (Fig. 2) with 2-0 nonabsorbable COBRAID™ suture (Smith & Nephew, Inc., Andover, MA, USA) for repair. With the knee tightly flexed, an Arthrex guide (Arthrex, Inc., Naples, FL, USA) was used to create the four femoral lead tunnels of the ACL at the center of the two bundles for reattachment of the ACL to the medial wall of the femoral condyle, drilling from anteromedial to posterolateral, and ensuring that the bone bridge between the two tunnels was about 1 cm. The femoral leads tunnels for the PCL, tibial head tunnels for the ACL, and tibial leads tunnels for the PCL were created using the same technique. During preparation of the tibial leads tunnels for the PCL, when the Kirschner wire crossed the posterior tibial cortex, extreme care was taken to avoid injuries to nerves and blood vessels. Next, a suture passer was used to guide the suture lines through each tunnel, simultaneously tightening the ends of the suture lines of the avulsed bundle stump, and then knotting each suture line outside the bone tunnel on the corresponding bone bridge. This completed reattachment of the avulsion stump in situ. Meniscal damage was repaired using sutures or by trimming, depending on the site of the injury.

Next, The PLC was approached via a posterolateral incision, being careful to maintain a 6- to 8-cm skin bridge between the two incisions. The popliteus, popliteofibular ligament, capsule, lateral collateral ligament, iliotibial band, and the biceps femoris were repaired to the femoral epicondyle, fibular head, or the lateral tibia, depending on the site of detachment. After decorticating the bone at the site of insertion, two to five nonabsorbable suture anchors were applied using cross-stitch technique. Complete avulsion of the MCL was repaired through the midline incision using a similar method, and partial injuries were not treated surgically. Before the wound was closed, the knee was brought through a 90-degree range of motion. Anterior and posterior drawer tests and the Lachman test were gently performed to verify stability.

Rehabilitation protocol

There are few reports on postoperative rehabilitation after open repair of MLIK. A hinged knee brace locked in 30° of flexion was used to protect the stability of the injured knee, and a standard rehabilitation protocol was subsequently performed. Patients were allowed to perform quadriceps isometric exercise and straight-leg raise on postoperative day 1. Care was taken to avoid varus and valgus stress in patients who had undergone PLC and/or MCL repairs. Patients began physical therapy 1 week later at our institution on an outpatient basis. The brace could be unlocked and the knee was brought through a full range of motion as tolerated. After 4 weeks, nonweightbearing activities and passive knee-flexion and -extension exercises were begun, gradually increasing the range of flexion from 0 to 120°. The third month postoperatively, closed-chain exercise and hamstring co-contractions were initiated. At postoperative months 4–5, patients began open-chain exercises and walking, partial weightbearing with crutches, while gradually increasing the range of the motion. At 6 months, patients could partially resume daily activities and begin progressive resistive exercise. At the goal of 7 months postoperatively, patients could walk, bearing full weight without crutches.

Follow-up evaluations

In all, 17 patients (18 knees) were followed for a mean of 4.8 ± 1.3 years (range, 2.4–7.3 years). Patients were examined by an independent senior orthopedic resident.

To assess clinical outcomes, we used a visual analog scale (VAS) score from 0 (no pain) to ten (worst pain) and a patient satisfaction score of 10 to 0, with a higher score indicating greater satisfaction. Self-administered questionnaires, including the Lysholm score [16], the Tegner score [17], and the 36-item Short-Form Health Survey® (SF-36) total score, were also used to evaluate clinical outcome. Finally, the Meyers functional rating was used to determine postoperative function [18, 19].

Range of motion and knee stability were also evaluated on physical examination. Range of motion was measured using standard goniometry; loss of flexion and extension were calculated by comparing the injured knee with the uninjured knee. The two patients with bilateral knee dislocation were excluded from this analysis of range of motion. To estimate the laxity of the ACL and PCL, a KT1000™ arthrometer (MEDmetric® Corp., San Diego, CA, USA) was used to perform Lachman and posterior drawer tests, respectively. Collateral ligament laxity was tested clinically by applying varus or valgus stress in extension and 30° of flexion. The stability of the PLC was tested using the Cooper asymmetry test (Dial test), which was performed in 30 and 90° of flexion.

Postoperative complications, including deep vein thrombosis (DVT), infection, suture granuloma, re-rupture, fibrosis, common peroneal nerve palsy, and heterotopic bone were recorded.

Statistical analysis

Statistical analysis was performed using SPSS for Windows, Version 13.0 (SPSS Inc., Chicago, IL, USA). All data are presented as mean ± standard deviation. The results were analyzed using a Student’s t-test, and significance was defined as P < 0.05for 95 % confidence.

Patients and epidemiological profiles

All patients had closed injuries that were reduced in the emergency room. The ligamentous injuries were on the left in seven cases, on the right in nine cases, and bilateral in one case. The mechanism of the injury was a motor vehicle accident in 11 patients (12 knees), a direct hit in three patients, fall from a height in two patients, and from football in one patient. Patients were categorized by combinations of ligament injuries using the modified Schenck system, as follows: KD III-M, n = 3; KD III-L, n = 4; KD IV, n = 9; and KD V, n = 2.

Twelve patients (13 knees, 72.2 %) had at least partial injury of the common peroneal nerve. Seven (53.8 %) of these knees had partial sensory loss, five (38.5 %) had partial sensory and motor loss, and one (7.7 %) had complete sensory and motor loss. Associated meniscus tear was also common in our study group, and only two patients had significant fractures with ipsilateral tibial plateau fracture (one Schatzker type I and one Schatzker type III).

Clinical outcomes

At final follow-up, mean VAS pain score was 2.4 ± 0.9 (range, 1.0–4.3); 15 patients (88.2 %) had a VAS pain score ≤3. Final patient satisfaction mean score was 8.0 ± 1.1 (range, 5.1–9.2), and only one patient had a satisfaction score <6. At final follow-up, mean total SF-36 score was 85.5 ± 10.4 (range, 57–100), and only one patient scored <60.

Mean postoperative Lysholm score was 87.5 ± 7.7 (range, 71–95). No patient scored below 60, and only one patient scored below 70. Mean preinjury Tegner activity score injury was 5.6 ± 1.4 (range, 3–9); postoperatively it decreased to 3.4 ± 1.7 (range, 1–6) demonstrating significant improvement (P < 0.01). The Meyers functional rating yielded one excellent, eight good, six fair, and two poor results. Of the 17 patients analyzed, 12 (70.6 %) were able to return to their previous work with little to no activity modification, four (23.5 %) were able to do light duty only, and only one patient (5.9 %), who suffered bilateral knee dislocation, was unable to return to work.

Complications

No DVT, compartment syndrome, iatrogenic neurovascular compromise, or deep infection was identified in our patients. Two patients experienced wound problems, including infection and fat liquefaction, which resolved after treatment with antibiotics and dressing changes. Three of the patients had arthrofibrosis: one received stereoarthrolysis under anesthesia 20 weeks postoperatively, and the other two received arthroscopic lysis of adhesions, one 33 weeks and the other 45 weeks postoperatively. Mild heterotopic bone formation developed in the knee of the patient with a Schatzker type III tibial plateau fracture and in the right knee of the patient with bilateral knee dislocations. Although the patient who suffered tibial plateau fracture had good knee joint activity, she suffered the pain around her knee joint, and the pain associated with climate change. It may be caused by the scar around her knee joint. The other patient had poor result might due to not well rehabilitation. As the patient suffered bilateral knee dislocations, he needed pay more attention to rehabilitation than other patients.