The Medial Patellofemoral Ligament (MPFL) is the main restraining force against lateral patellar displacement. It is often disrupted following patellar subluxation or dislocation. MPFL reconstruction is frequently performed when conservative management fails and the patient experiences recurrent patellar dislocation. Various MPFL reconstruction procedures have been described in the literature and reported outcomes are encouraging. This study analyses the radiographic outcomes following MPFL reconstruction. From January 2006 to January 2011, 76 consecutive patients (80 knees) with patellar recurrent dislocation underwent medial patellofemoral ligament reconstruction in three large teaching hospitals. Mean follow-up was 31.8 months (range, 13–72). Semitendinosus and gracilis autografts were used for the reconstruction and all procedures were carried out by the senior authors (WL, BR, CW, MB). Plain radiographs (Anteroposterior (AP), Lateral and Skyline) performed preoperatively and postoperatively were used to compare the sulcus angle, congruence angle, lateral patellofemoral angle, trochlear dysplasia (Dejour classification), trochlear boss height and patellar height (Caton-Deschamps ratio). Plain radiographs (Lateral) performed postoperatively were used to evaluate the femoral tunnel placement used for MPFL reconstruction. The sulcus angle improved from 143.2° (122.9–157.7) to 139.3° (115.7–154.6) and the congruence angle improved from 26.7° (−17.5–82.6) to 0.26° (−35.3–7.8). The lateral patellofemoral angle was 7.99° (3.2–19.2) preoperatively and 9.02° (3.2–18.2) postoperatively. The Caton-Deschamps ratio was 1.2 (1.0–1.5) preoperatively and 1.0 (0.8–1.1) postoperatively. Using the Dejour classification of trochlear dysplasia, all preoperative radiographs were considered to be grade C or D and all postoperative radiographs were considered to be grade A or B. Trochlear boss height was 5.9mm (1.8–11.6) preoperatively and 4.7mm (1.6–6.9) postoperatively. 59% of the femoral tunnels were considered to be in a good position on postoperative radiographs. This study displayed a significant improvement in postoperative radiographic parameters, demonstrating the importance of anatomic restoration when performing MPFL reconstruction

Introduction. The medial patellofemoral ligament (MPFL) is the main stabilizer of the patella and therefore mostly reconstructed in the surgical correction of patellofemoral dislocation. Various biomechanical and clinical studies have been conducted on MPFL reconstruction, while the patellofemoral contact pressure (PFCP) which is indicated as one of the predictors of retropatellar osteoarthritis was neglected. Therefore, the aim of this study was to investigate how different MPFL reconstruction approaches affect PFCP. Material & Methods. After radiographic examination and preparation six human cadaveric knee joints (52.1 ± 8.4yrs) were placed in a 6-DOF knee simulator. Three flexion-extension cycles (0–90°) were applied, while the extensor muscles (175N) and an axial joint load (200N) were simulated. PFCP was measured in knee flexion of 0°, 30° and 90° using a calibrated pressure measurement system (K-Scan, Tekscan Inc., USA). The following MPFL conditions were examined: native (P. nat. ), anatomical reconstruction (P. a. ), proximal and distal patellar single-bundle reconstruction (P. p. , P. d. ), proximal and ventral femoral reconstruction (F. p. , F. v. ). The cohesive gracillis graft of each knee was used for MPFL reconstruction. Further, the effect of three different graft pre-tensioning levels (2N, 10N, 20N) on the PFCP were compared. Nonparametric statistical analysis was performed using SPSS (IBM Inc., USA). Results. In 0° knee flexion median PFCP of the native state (P. nat. =0.46MPa) was significantly higher (p=0.04) compared to the ventral femoral fixation state (F. v. =0.24MPa). No significant differences were observed in 30° knee flexion. In 90° knee flexion PCFP of both femoral reconstructions (F. p. =1.26MPa, F. v. =1.12MPa) were significantly higher (p<0.04) compared to the native state (P. nat. =0.43MPa). Graft pre-tensioning had no significant impact (p>0.27) on the PFCP in 0°, 30° and 90° knee flexion for all pre-tensioning levels. Discussion. We investigated the PFCP of different MPFL reconstructions and compared them during continuous joint motion from 0° to 90° knee flexion. While a non-anatomical graft fixation on the femoral side leads to an excessive increase of PFCP (293%), a non-anatomical positioning on the patellar side only showed minor impact on the PFCP. An anatomical MPFL reconstruction showed comparable PFCP to the native joint. In contrast to the literature, we did not find a significant influence of graft pre-tensioning from 2N up to 20N on the PFCP. With respect to all study findings we would recommend to use the anatomical footprints for MPFL reconstruction and a moderate graft pre-tensioning of 2N