The present meta-analysis updated the previous literature. However, the meta-analysis also has limitations. Firstly, all the included studies showed some weaknesses, mainly related to low-quality RCT design regarding factors such as age, gender, surgical, and rehabilitative interventions. Three studies [
22,
26,
28] reported solely on patients younger than 16 years, one study [
27] contained no information on patient age, and the remaining six studies included patients of varying ages, including children, adolescents, and adults; this may have affected the outcomes, as age is reportedly correlated with the incidence of primary and recurrent patellar dislocation [
36]. Secondly, the risk of bias scores reflects that some studies are at high risk, although all data were extracted from RCTs. We attempted to resolve the problem by performing a subgroup analysis by type of surgery. However, it is difficult to compare studies using a single surgical technique with studies using multiple surgical methods to treat APPD. In the study by Nikku et al. [
20], 63 patients had the medial retinaculum in a variety of ways: suture, duplication, or additional MPFL augmentation. Among them, 54 underwent a LRR. Seven patients underwent only a LRR. In the study by Palmu et al. [
22], MPFL repair was performed in 29 knees, of which 25 had a LRR. Seven patients underwent LRR alone. Petri et al. [
6] and Regalado et al. [
26] realigned the extensor mechanism with the Roux-Goldthwait procedure, which is used to manage recurrent patellar dislocation [
37]. It is unclear why APPD was often treated with more than one type of surgical technique several years ago. One possible reason for this is that this type of injury is relatively uncommon, and no surgical method has been definitively proven to be effective. Thus, various procedures were performed simultaneously in order to surgically rectify an MPFL rupture and subsequent patellar instability. Another possible reason is that patients with APPD are often linked with anatomic abnormality, and those congenital malformations may need to be corrected during ligamentous repair in order to improve outcomes. In a word, surgical confounders discussed above should also be considered in future clinical trials. Previous meta-analyses have also discussed other limitations, including the small number of trials, diverse demographics, and lack of allocation concealment methods [
15‐
19].