Diagnosis and treatment of hyperextension bicondylar tibial plateau fractures

From April 2016 to February 2018, 12 cases of hyperextension bicondylar tibial plateau, fractures were treated in our Department of Orthopedics. The study was carried on with the approval of institution’s ethical review board. One patient suffered from severe articular cartilage destruction and was not included in our study as it was treated by total knee arthroplasty. There were six cases of fresh fractures, four cases of old fractures, and one case that was revised by osteotomy and internal fixation because of pain and instability of the knee. There were eight men and three women, with an age range of 19–61 years (average age, 42.4 years). The duration between injury and operation was 27.5 days (5–90 days). The duration between injury and operation of the six fresh fractures was 7.2 days (5–13 days). According to the Schatzker classification, all cases were Schatzker VI type fractures.

All patients underwent X-ray and three -dimensional computed tomography examination of the knee joint before operation, to measure the displacement of the tibial plateau joint surface, the pTSA, and the medial tibial plateau angle (mTPA). Four patients received magnetic resonance imaging (MRI) examination, while all patients received deep intravenous doppler ultrasound examination. Patients with suspected popliteal artery injury underwent computed tomography angiography (CTA) examination of the lower extremities. All patients with fresh fractures received staged treatment. After admission, patients underwent bone traction or trans-articular fixation. Three patients with popliteal artery injury underwent saphenous vein grafting to repair the popliteal artery. All patients underwent open reduction and internal fixation if the soft tissue condition of the affected limb was suitable for surgery. All patients were operated by the same senior surgeons.

After nerve block anesthesia or genanral anesthesia was completed, each patient was placed in the supine position, with a balloon tourniquet on the proximal thigh and a round pillow under the popliteal fossa, which facilitated the opening of the fracture space in front of the tibial plateau fracture. Prophylactic antibiotics were given intravenously at 30 min before the incision. All fractures were fixed with anatomical locking compression plates (Depuy Synthes, USA).

The surgical approach was selected according to the degree of displacement of the posterior tibial cortex; fractures with no or minimal displacement of the posterior tibial cortex (usually ≤2 mm) were operated using the modified anterior midline incision, while fractures with a large displacement (usually > 2 mm) were operated by anterolateral and posteromedial double incisions. Only one case of hyperextension valgus bicondylar tibial plateau fracture with ipsilateral femoral shaft fracture was treated by a parapatellar anterolateral incision.

A longitudinal incision was performed parallel to the anterior midline of the knee, starting from the upper edge of the patellar, passing the lateral edge of the tibial tuberosity and approximately 1 cm lateral to the anterior tibial spine, and then extending approximately 7 cm distally. The full-thickness flap was protected, directly exposed to the deep fascia, stripped to both sides, and then exposed to the anterior edge of the fibular head laterally and the anterior edge of the superficial medial collateral ligament (MCL) medially. The flap was carefully retracted, and the lateral patellar retinaculum was incised, with a tibial tuberosity osteotomy performed occasionally to increase the anterior exposure, to obtain a 50–70-mm-long, 20-mm-wide proximal-end, 15-mm-wide distal end, and 10-mm-thick bone fragment in relation to its insertion in the tibial tubercle. The bone flap was then retracted proximally. Next, the anterior tibial muscle fascia was cut longitudinally, and the tibial is anterior muscle retracted laterally. A medial subperiosteal dissection was then used to reveal the medial fracture line. If the articular surface of the tibial plateau had a gap or step-off, the medial and lateral menisci were elevated to inspect the underlying joint surface. For restoration of sagittal plane alignment, disimpaction and elevation of the anterior metaphysis was performed by pulling the proximal metaphysis of the tibia forward and pulling the distal end of the tibia distally with the knee in the flexion position. The periosteal elevator was often used to reduce the collapsed fragments.

After reduction was accomplished, the Kirschner wire was used for temporary fixation. Allograft bone was used for most cases. The C-arm was adopted to confirm the reduction, and a T-shaped locking plate was placed in front of the MCL medially, and an L-shaped locking plate was placed on the lateral side, to fix the fracture. Once again, the C-arm was used to confirm that the articular surface was smooth, the pTSA was recovered, and the position of internal fixation was satisfactory. The wound was flushed with saline, the drainage tube was indwelled, and the deep fascia was sutured intermittently. The parapatellar retinaculum and the anterior tibialis fascia were repaired, and the superficial fascia and skin was intermittently sutured in favor of drainage.

Double incision surgery was performed mainly according to Firoozabadi’s method. After reduction was confirmed by X-ray, a T-shaped locking plate was then placed on the medial side of the tibial platform, and an L-shaped locking plate was placed on the lateral side.

Postoperatively, patients were instructed to begin range of motion movement of the knee on the first postoperative day. All patients were kept non-weight bearing for three months to avoid early loss of reduction. Toe-touch weight bearing was initiated at 12 weeks. Full weight bearing was not permitted until radiographic signs of primary bone healing, usually present at 12 weeks postoperatively.

All patients were followed for at least three months. The average follow-up was 11.5 months (3–24 months). Standard AP and lateral radiographs were taken to evaluate the mTPA, pTSA, and articular step-off. At the last follow-up, objective knee evaluation and patient-reported functional outcomes were assessed using the Rasmussen knee evaluation system. Complications, including infection, wound irritation, and neurovascular injury, were also noted.

In accordance with the change in mTPA, we divided hyperextension bicondylar tibial plateau fractures into hyperextension varus type (mTPA < 85°), pure hyperextension type (85° ≤ mTPA ≤ 90°) and hyperextension valgus type (mTPA > 90°). It would have been preferable to classify these types by comparing the mTPA of the injured side with that of the contralateral side. However, X-ray of the normal knee was not performed in some patients. Thus, we compared the mTPA of the injured side with normal population data (85–90°) [5]. The hyperextension varus type is highly combined with popliteal artery injury, common peroneal nerve injury, and lateral collateral ligament injury. In the present study, of the six cases of hyperextension varus-type fractures of the tibial plateau, three were complicated with popliteal artery embolism, which is higher than the reported incidence of 12% and 22.2% [1, 3]. There were only two cases with lateral collateral ligament injury in this group, although the actual injury rate may be higher as we did not routinely perform MRI on all patients. The reported incidence of common peroneal nerve injury is as high as 16% [3], while we only observed one case with such injury out of our 11 patients. Meniscus injury occurred in four of 11 cases, three of which were diagnosed during open surgery and were repaired at that time.

These classifications can also be used to instruct the main plating. In principle, the main plate of the hyperextension varus type of tibial plateau fracture should be placed on the anterior medial side, while for the hyperextension valgus type the main plate should be placed on the anterolateral side, and for the pure hyperextension type, both sides should be fixed equally.

Surgical indications for bicondylar tibial plateau fractures vary in the literature [1, 2, 6]. Most authors believe that surgery should be performed if the intra-articular step-off or gap exceeds 2 mm [2, 6]. It was also suggested that a pTSA change > 5°from a normal reference value of 5° should be considered “clinically significant” [7]. By contrast, an angular deformity of > 10° in the sagittal plane was reported as a surgical indication [6]. Further, an angular deformity of > 10° in the coronal plane, a metaphyseal-diaphyseal translation of > 1 cm, open fracture, associated compartment syndrome, associated ligament injury requiring repair, and an associated fracture of the ipsilateral tibia or fibular are also surgical indications of bicondylar tibial plateau fractures [6]. Based on our present findings, the displacement can even fail to meet these criteria. Nevertheless, if the knee is certified to be unstable by the lateral stress test, the procedure should be performed. Our series was mainly based on Hall’s indication criteria [6].

In previous reports, anterolateral and posteromedial double incisions are typically used for hyperextension bicondylar tibial plateau fractures [1, 3]. For fractures with a large displacement of the posterior tibial cortex that require reduction under direct visualization, the posteromedial incision provides excellent exposure of the posteromedial fragments, and facilitates placing of the posterior supporting plate. However, a disadvantage of this approach is that it requires massive stripping of soft tissue to expose the anterior compression fracture, and if the anterior buttressing plate is placed, it is inconvenient to insert screws. Thus, for hyperextension type tibial plateau fractures with no or little posterior displacement, we developed a modified anterior midline incision. The traditional anterior midline incision provides excellent exposure of the whole anterior part of the tibial plateau, although it is associated with more sensory disturbances of the anterior aspect of the knee than that for anterolateral incisions [8]. The traditional anterior midline incision, which usually passes through the medial edge of the tibial tuberosity, is not in line with the anterolateral incision of the knee. However, the single anterolateral approach is difficult to manage for medial plateau fractures. Thus, we combined the advantages of both incisions in our modified anterior midline incision approach.

As most of the hyperextension-type bicondylar tibial plateau fractures are characterized by central depression of the anterior part of the medial plateau, lateral plateau, and the median eminence as a whole, while the cortices of the plateau may lie in situ (similar to the geographic fracture “Tiankeng”), we termed this central depression as “Tiankeng-like collapse.” It is inconvenient to observe the Tiankeng-like collapse from the side, but easy to expose the whole Tiankeng-like collapse from the window behind the patellar ligament. Thus, the patellar ligament can be retracted medially or laterally to reveal the Tiankeng-like collapse through the modified anterior midline incision. For severely comminuted cases or old fractures, tibial tuberosity osteotomy can be performed to increase the anterior exposure of the modified anterior midline incision.

Although there is evidence of a high incidence of poor wound healing for treatment of bicondylar tibial plateau fractures with an anterior median incision, this may be associated with the unhealthy soft tissue caused by the high energy injury and massive stripping of soft tissue required to expose the posterior parts of both the medial and lateral tibial plateau [9]. None of our five cases had wound infection or skin necrosis. This low incidence of infection may be associated with our strict control of indications, gentle operation to protect the skin flaps, and limited peeling range. We also performed full drainage and moderate pressure dressing, which may reduce the occurrence of subcutaneous hematoma in front of the knee joint.

Because of anterior depression of both the medial and lateral plateau, we typically use double plates to fix hyperextension bicondylar tibial plateau fractures. However, most of these fractures are characterized by a Tiankeng-like collapse, as the medial and lateral platforms and the median eminence are collapsed as a whole. Thus, if the depression fragment is not comminuted, it may be fixed by a single plate, although this approach should be certificated by mechanical testing and clinical follow-up. In our series, one case of hyperextension valgus-type fracture was treated with a single lateral locking plate, with no loss of reduction observed at recent follow-up.

Because of the low incidence of hyperextension bicondylar tibial plateau fractures, the sample size of our study was small. Further, during the case analysis, the revision cases, old fracture cases, and fresh fracture cases were analyzed together, which may have affected our findings. Finally, the follow-up time was short, and further long-term assessment of our cases is required.