Medial patellofemoral ligament reconstruction combined with biplanar supracondylar femoral derotation osteotomy in recurrent patellar dislocation with increased femoral internal torsion and genu valgum: a retrospective study

The purpose of this study was to evaluate the clinical and radiographic outcomes after medial patellofemoral ligament (MPFL) reconstruction combined with supracondylar biplanar femoral derotation osteotomy (FDO) in recurrent patellar dislocation (RPD) with increased femoral anteversion angle (FAA) and genu valgum. parameters and alignment correction, c) to assess complications associated with bone healing, soft tissue irritation, wound infection, and recurrence of dislocation. It was our hypothesis that patients with RPD associated with increased FAA and genu valgum treated by FDO can achieve satisfactory clinical and radiographic results in the short-term follow-up period.


Page 4/17
Background Recurrent patellar dislocation (RPD) is a complex condition and often associated with multiple contributing factors. Several of these risk factors, including patella alta, genu valgum, disrupted or weaken medial soft tissue, trochlear dysplasia, increased tibial tuberosity and the trochlear groove (TT-TG) distance, and torsional alignment of the femur or tibia [1][2][3][4][5] . Specially, osseous deformities in the coronal and axial plane, such as genu valgum and torsional deformities of the lower extremity, are now considered to be associated with adverse effects on patellofemoral instability. Herein, the increased femoral anteversion angle (FAA) and genu valgum are thought to create a sustaining lateralizing force vector applied on the patella, which might increase excessive loading forces on the reconstructed graft and even lead to patellar redislocation.
At present, there is still much controversy in the surgical techniques of RPD combined with knee valgus deformity. In recent years, various surgical techniques for addressing RPD with genu valgum have been described. Nha et al. [6] demonstrated the satisfactory improvement of knee function of 14 patients (23 knees) who underwent closing-wedge distal femoral osteotomy without MPFL reconstruction. Similarly, Swarup et al. [7] demonstrated that lateral opening wedge distal femoral osteotomy combined with lateral retinacular release yield satisfactory clinical results in this population.
In practice, genu valgum combined with excessive femoral internal torsion remains as the primary risk factors for RPD and is rarely corrected by the surgery simultaneously. Despite the association between mechanical malalignment of multi-plane, soft tissue dynamics and patellar instability, until now, there is a paucity of studies regarding clinical and radiographic results after MPFL reconstruction combined with supracondylar biplanar FDO procedure in this population.
The purpose of this study was to a) analyze the clinical and radiographic results of the MPFL reconstruction combined with biplanar supracondylar FDO procedure, b) to evaluate the differences between pre-and post-operative knee function and radiographic results including effects on patellar parameters and alignment correction, c) to assess complications associated with bone healing, soft tissue irritation, wound infection, and recurrence of dislocation. It was our hypothesis that patients with RPD associated with increased FAA and genu valgum treated by FDO can achieve satisfactory clinical and radiographic results in the short-term follow-up period.

Patients
This study was approved by the Ethics Committee of the A liated Hospital of Southwest University. Between January 2017 and December 2020, a retrospective single-center cohort study was conducted to evaluate the clinical and radiographic outcomes of all included patients who experienced recurrent patellar instability due to genu valgum in the coronal plane and increased FAA in the axial plane. All included patients were treated by MPFL reconstruction combined with supracondylar biplanar FDO procedure and all surgeries were performed by a senior orthopaedic surgeon (Z.L).
Exclusion criteria were as follows: a) tibial rotational deformity; b) patients with open epiphysis; c) patients with lateral compartment knee osteoarthritis; d) patients with previous knee surgery; e) posttraumatic deformities; f) patients who lost to follow up; g) incomplete clinical data.

Surgical technique
Supracondylar biplanar femoral derotation osteotomy Supracondylar biplanar FDO was performed according to a method described previously by Hinterwimmer et al. [9] , in which they developed an anterior closed-wedge technique of biplanar supracondylar DFO for patellofemoral malalignment.
Patients received general anesthesia and were placed on a radiolucent operating table in a supine position, and a tourniquet was used to the proximal thigh. First, an arthroscopic examination was performed at the beginning of surgery to evaluate the patellar tracking under the direct visualization.
Second, all cases were performed by the standard medial subvastus approach through a longitudinal skin incision of approximately 10-12 cm in length to access the medial metaphysis of the distal femur. The subcutaneous tissue and fascia were separated and the vastus medialis was stripped with a blunt Hohmann retractor to protect the neurovascular behind the femoral shaft. The axial osteotomy was performed perpendicular to the femoral shaft axis and involved the posterior two-thirds of the femur. The frontal plane osteotomy runs in an oblique direction from the superior margin of the axial osteotomy to the anterior femoral cortex (Fig.2a). After that, the rst cut of the anterior plane osteotomy is performed with an oscillating saw from medial to lateral with the saw blade slightly angled towards the oor. Both the medial and contralateral lateral cortex is completely cut through. For derotation osteotomy, the target correction rotation angle calculated in preoperative planning and was marked using two Kirschner wires intra-operatively under the uoroscopic guidance. The lateral cortex was cut completely and two Steinmann nails were placed proximally and distally to the osteotomy site to allow the distal femur external rotated to the predetermined angle. After that, a medial-based anterior wedge is produced by a second anterior saw cut inferior to the rst wedge (Fig.2b). The wedge is removed (Fig.2c), and the gap created by anterior osteotomy is closed by rotating residual fragments using the two Schanz screws (Fig.2d).
After that, patellar tracking was re-evaluated under arthroscopic visualization after osteosynthesis with Tomo x-locking plate xation (Depuy Synthes, Umkirch, Germany). Of note, it is important to place the Tomo x-locking plate xation in a position where they do not impede the later positioning of the femoral tunnel of MPFL. Typical case was presented in Fig.3.

MPFL reconstruction
MPFL double-bundle anatomical reconstruction using autologous semitendinosus tendon autograft was performed in all patients and our surgical technique was similar to Schoettle et al. [10] . Two bony grooves were drilled in the patellar medial edge, which were placed into the center and the upper inner corner of the patellar medial edge, respectively. The graft xed in patella side by two anchors inserted into the bony grooves. Subsequently, a femoral tunnel was made at the femoral insertion site of the MPFL, and the graft was xed by a bioabsorbable interference screw with the knee in 20° to 30° of exion.

Rehabilitation protocol
All patients started functional exercises within 24 hours post-operation, including active circummovements of ankle and isometric quadriceps muscle training. Patients were instructed to begin partial weight-bearing with crutches for the rst 4 weeks, and strengthening exercise of vastus medialis muscle was encouraged. Full-weight bearing without limitation was allowed at 6-8 weeks postoperatively.

Functional scores evaluation
Functional scores, including visual analogue scale (VAS), Kujala score [11] , International Knee Documentation Committee (IKDC) score [12] , Lysholm score [13] , and Tegner activity score [14] were used to evaluate knee function preoperatively and at the last follow-up after surgery.

Radiographic assessment
Preoperative radiological analysis prior to surgery, including full-leg standing anteroposterior (AP) radiographs of lower extremity, lateral views, and standardized hip-knee-ankle computed tomography (CT) scans were used to evaluate the mechanical lateral distal femoral angle (mLDFA), trochlear dysplasia, TT-TG distance, patellar height, and tibial and femoral torsion deformities in the axial plane.
The mLDFA was measured on the full-leg standing AP radiograph and was de nes as a lateral angle between a line connecting the center of the femoral head to the center of the knee and a line tangent to distal femoral condylar. Trochlear dysplasia and TT-TG distance were evaluated by axial CT images. The TT-TG distance was measured between the two parallel lines that perpendicular to the posterior condylar tangents on superimposed axial slices: a line passed the most cephalad point of the tibial tubercle and a line passed through the deepest point of the trochlear groove [8]. Caton-Deschamps index (CDI) was measured on lateral radiographs to evaluate the patellar alta [15].
Tibial and femoral torsion deformity was measured on reconstructed 3-dimensional axial CT images utilizing the method described by Takagi et al. [16], as shown in Fig.1. Tibial torsion was de ned as the angel between a line connecting the posterior tibial condyles and a line connecting the midpoints of the medial and lateral malleoli. The FAA was de ned as the angle between a line connecting the midpoint of the femoral neck and the center of the femoral head and a line connecting the most posterior points of the medial and lateral femoral condyles.
Statistical analysis SPSS software (version 24.0, IBM Corp., USA) was performed for statistical analysis. Clinical and radiographic results were present with mean and standard deviation (SD). Paired t test was used to compare the differences in the pre-and post-operative clinical and radiographic outcome data. For all tests, p values < 0.05 were considered statistically signi cant.

Patient demographics
A total of 13 patients (13 knees) who treated by double-bundle anatomical MPFL reconstruction combined with biplanar supracondylar FDO were include in the study. Of the 13 patients, 4 cases were males and 9 cases were females. The average age of the included patients at the time of surgery was 18.7 years (range, 15-29 years). The average body mass index (BMI) was 25.7±3.6 (21.8-31.2). The average follow-up period was 14.7 months (range, 12-26 months).
Pre-operative standing full-leg AP and lateral radiographs and CT scans for hip-knee-ankle were available for all cases, and patients demographics were presented in Table 1.

Complications
No cases of patients developed wound infection, soft tissue irritation, and recurrence of subluxation or dislocation during the follow-up period after surgery. Bone healing at the osteotomy site was achieved in all cases, and all patients regained full extension and exion and no limited range of motion was observed.

Discussion
Patella instability associated with genu valgum treated by supracondylar distal femoral osteotomy have been reported in several studies [17][18][19]. However, the clinical signi cance and potential advantages of MPFL reconstruction combined with biplanar supracondylar FDO procedure in RPD with increased FAA and genu valgum have not yet been identi ed. In this retrospective study, MPFL reconstruction combined with biplanar supracondylar FDO procedure achieved signi cant functional improvement after surgery in knee function scores (VAS, Kujala score, IKDC score, Lysholm score, and Tegner scores) and satisfactory radiographic outcomes (FAA, TT-TG, and mLDFA) in patellar instability with increased FAA and genu valgum, and no recurrence of dislocation cases had been found within the follow-up period.
Several orthopaedic surgeons have emphasized the role that osseous deformities of the axial and coronal plane acts as a signi cantly higher risk factor for patellofemoral maltracking. Dejour et al. [8] identi ed that patients with patellofemoral instability had a higher value of FAA than healthy controls (15.6 vs. 10.8) following CT evaluation. Similarly, Zhang et al. [20] have reported that the adverse effects of increased femoral internal torsion on reconstructed MPFL, especially in patients when the FAA greater than 30°, which could be partially explained by the fact that the excessive lateralizing force vector acting on the patella due to the increased Q angle [21]. Recently, biomechanical studies further demonstrated that the adverse effect of isolated MPFL reconstruction for patellar instability associated with increased FAA. Kaiser et al. [22] revealed that isolated MPFL reconstruction for patella instability is insu cient for higher degrees of FAA, which indicated that increased FAA may result in a persistent lateral force vector on the patella.
Due to these abnormal biomechanics of osseous deformity, it is vital to identify these underlying predisposing risk factors and to early make intervention for RPD. Despite isolated anatomical MPFL reconstruction is considered to be a standard treatment for patellofemoral instability with satisfactory results [23,24], subsequent studies have demonstrated that a high rate of subjective dissatisfaction in patients with increased femoral internal torsion [25]. Supracondylar femoral derotation osteotomy as an isolated procedure has been shown good clinical outcomes for RPD with increased FAA [26,27]. However, isolated FDO procedure at the distal femur may increase the risk of graft failure when ignoring the correction of the knee valgus deformity, because the laterally-oriented vector forces applied to patellofemoral joint which can result in excessive tension into the reconstructed MPFL graft [5,28].
Recently, there has been a great focus of investigating the effect of derotation femoral osteotomy on the changes of coronal alignment. Nelitz et al. [29] reported that FDO procedure tend to result in an increased valgus angulation in the frontal plane due to a decreased mLDFA. Similarly, Konrads et al. [30] also identi ed that supracondylar femoral external osteotomy would lead to valgus deformity of the coronal limb alignment, which may be attributed to the reorientation of the femoral antecurvature and the femoral neck. Thus, isolated FDO procedure may not address the problem involving increased FAA combined with genu valgum in the presence of RPD patients.
Despite a biplanar supracondylar DFO has been performed in the correction of valgus deformity and excessive femoral internal torsion simultaneously, none of this group of patients showed signs for delayed union or non-union of the osteotomy, which was comparable to Imhoff et al. [26] on a combined varus and external rotation producing distal femoral osteotomy.
There are several limitations in this study. First, the number of patients was relatively small and the follow-up period was short. However, RPD patients both with increased FAA and genu valgum are rare. Second, a missing comparative group of patients who treated by other surgical techniques. However, considering that the signi cant functional improvement and absence of redislocation, FDO combined with MPFL reconstruction may be a treatment option for RPD with increased FAA and genu valgum. Third, second-look arthroscopic evaluation was not performed to evaluate the changes of trochlear and retropatellar cartilage. All the data and material involving this article will be available upon request by send an e-mail to the rst author.