Abstract
Several meta-analyses of surgical versus non-operative treatment of femoroacetabular impingement syndrome (FAIS) have been published, but reliable evidence is still lacking. The aim of this meta-analysis of randomized controlled trials (RCTs) was to assess the outcomes of FAIS patients treated conservatively compared with those treated with hip arthroscopy (HAS). PubMed, CENTRAL of the Cochrane Library, Epistemonikos, and Embase databases were searched up to March 31, 2025. Quality was assessed using the Cochrane Risk of Bias 2 tool, the level of evidence for each outcome parameter was determined using the GRADE system, and publication bias was presented in funnel plots. In a common effect and random effects meta-analysis, mean differences (MDs) between the conservative treatment group and the HAS group were calculated with 95% confidence intervals (CIs) using the Hartung-Knapp-Sidik-Jonkman heterogeneity estimator. A total of 7 RCTs with a total of 489 patients in the conservative treatment group and 484 patients in the HAS group met the inclusion criteria. Of the 7 RCTs included, four were assessed as having a low risk of bias, one as having a moderate risk of bias, and two as having a high risk of bias. The outcomes "post-intervention functional MCID" and "iHOT at ≤ 12 months post-intervention" had a high level of evidence, and the outcome "HOS-ADL at ≤ 8 months post-intervention" had a moderate level of evidence. No significant publication bias was detected for any outcome. The HAS group had a statistically significant 0.85 higher post-intervention functional MCID (common effect model: MD: 0.85 CIs 0.53-1.17; random effects model: MD: 0.85 CIs 0.64-1.06; I (2) = 0%; τ (2) = 0.02; p = 0.96) and a statistically significant 10.74 higher iHOT at ≤ 12 months post-intervention than the conservative treatment group (common effect model: MD: 10.74 CIs 7.06 to 14.42; random effects model: MD: 10.98 CIs 6.62 to 15.34; I (2) = 0%; τ (2) = 7.52; p = 0.62). There was no difference between the HAS group and the conservative treatment group in HOS-ADL at ≤ 8 months post-intervention (common effect model: MD: 5.62 CIs 1.76 to 9.48; random effects model: MD: 4.10 CIs -12.31 to 20.50; I (2) = 69%; τ (2) = 29.88; p = 0.04). This meta-analysis using high-quality statistical methods showed a statistically significant higher post-intervention functional MCID and iHOT at ≤ 12 months post-intervention in favor of the HAS group compared to the conservative treatment group. HOS-ADL at ≤ 8 months post-intervention showed no differences.