Factors Affecting the Outcomes of Double-Bundle Medial Patellofemoral Ligament Reconstruction for Recurrent Patellar Dislocations Evaluated by Multivariate Analysis

Background: Many factors are involved in causing patellar instability or recurrent patellar dislocations. However, factors affecting the outcomes of anatomic medial patellofemoral ligament (MPFL) reconstruction have not been fully documented. Purpose: To evaluate the rate of recurrent patellar instability after isolated MPFL reconstruction and to elucidate factors affecting the outcomes of isolated MPFL reconstruction using multivariable statistics. Study Design: Case-control study; Level of evidence, 3. Methods: A total of 42 patients (44 knees) who underwent isolated, anatomic MPFL reconstruction using a semitendinosus autograft without any additional patellar stabilization procedures for treating recurrent patellar dislocations were included in this study. All patients were followed for more than 2 years; the mean follow-up period was 3.2 years (range, 2-9 years). Postoperative patellofemoral instability was diagnosed when the patient complained of recurrent patellar subluxations or dislocations or had a positive apprehension sign during follow-up. Preoperative radiographic findings were obtained using plain radiography and computed tomography. Femoral tunnel positions were assessed on postoperative radiographs. The strength of the relationship between postoperative patellofemoral instability and potential risk factors such as preoperative age, sex, body mass index, patellar type, sulcus angle, congruence angle, lateral tilt angle, patellar height, tibial tubercle-trochlear groove (TT-TG) distance, degree of trochlear dysplasia, and femoral bone tunnel position were evaluated by univariate and multivariate logistic regression analyses. Results: At follow-up, 2 knees had experienced a redislocation (4.5%). A positive apprehension sign was still evident in 8 knees (18.2%). Three factors, including the sulcus angle, the congruence angle, and trochlear dysplasia, were extracted by single linear regression analysis. Univariate logistic regression analysis showed that the sulcus angle (odds ratio [OR], 1.11; 95% CI, 1.01-1.22; P = .04) and trochlear dysplasia (OR, 3.04; 95% CI, 1.39-6.63; P = .01) were associated with postoperative patellofemoral instability. Trochlear dysplasia was independently associated with postoperative patellofemoral instability by multivariable logistic regression analysis (P < .05). An increased TT-TG distance exerted a significant effect on the outcomes of MPFL reconstruction, particularly in patients with type D trochlea. Conclusion: Severe trochlear dysplasia is the most important predictor of residual patellofemoral instability after isolated MPFL reconstruction. In addition, an increased TT-TG distance affected the outcomes in patients with type D trochlea. Additional patellar stabilization procedures should be considered for patients with severe trochlear dysplasia and an increased TT-TG distance.