Abstract
PURPOSE: Non-surgical treatment of primary patellar dislocation has a high risk of recurrent dislocation; thus, we tried to identify injuries in which sites of the medial patellofemoral ligament (MPFL) were most associated with recurrent dislocation by analyzing relevant original literature in order to provide improved suggestions on early surgical treatment. METHODS: According to the preset retrieval strategy, the original studies were retrieved until January 2020 using MEDLINE, Embase and Cochrane Library. Review Manager 5.3 software was used to summarize and compare the differences of recurrent dislocation of MPFL injuries at different attachments. RESULTS: Although the incidence of recurrent patellar dislocation at the femoral attachment of MPFL was higher overall (femoral only vs. patellar only vs. combined: 37.6% vs. 32.3% vs. 35.8%), no statistical difference was found among the three groups (femoral only vs. patellar only, RR = 1.32 [95% CI 0.89-1.95]; P = 0.17) (femoral only vs. combined, RR = 1.15 [95% CI 0.59-2.22]; P = 0.68) (patellar only vs. combined, RR = 0.94 [95% CI 0.69-1.29]; P = 0.72). In addition, the sulcus angle of recurrent dislocation group is significantly greater than that in the non-recurrent dislocation group (MD = 3.06 [95% CI 0.42-5.70]; P = 0.02). CONCLUSIONS: Based on the pooled data collected from the original studies available, the risk of recurrent patellar dislocation due to damage to the MPFL at different sites did not differ. Additionally, the sulcus angle in the group with recurrent dislocation was considerably higher when comparing with the group without recurrent dislocation, that is, the shallower and flatter of the trochlear groove, the higher the risk of recurrent patellar dislocation.