ロボット支援/手動THAの多次元比較：REML-HKおよびマルチモデルを用いた主要転帰の系統的レビューとメタアナリシス

Total hip arthroplasty (THA) is a standard treatment for advanced hip disorders; however, conventional manual THA (M-THA) is potentially limited by suboptimal prosthesis positioning, a prolonged learning curve, and an elevated risk of complications. Robot-assisted THA (R-THA), with its enhanced accuracy in prosthesis alignment and intraoperative feedback capabilities, has been proposed to address these shortcomings. Nevertheless, the current literature on the differences in clinical outcomes between R-THA and M-THA remains inconsistent and fragmented, warranting a systematic review and meta-analysis. This systematic review and meta-analysis aimed to evaluate the comparative effectiveness of R-THA versus M-THA regarding surgical accuracy, perioperative outcomes, and complications.

A comprehensive search was performed by 7 general databases prior to November 2024, and studies were collected if they were involved in the comparison between robot-assisted THA and manual THA in the surgical outcomes, including learning curve, operation time, prosthesis placement accuracy (cup anteversion and inclination), Harris Hip Score (HHS), and complications. The R 4.5.1 software was used to calculate the 95% confidence intervals with standardised mean differences and Cohen's d values for continuous variables, as well as odds ratios for binary variables. The methodological bias and potential heterogeneity sources of included studies were evaluated.

A total of 37 studies were analysed in this study. Compared with M-THA, R-THA showed the better accuracy in prosthesis anteversion (OR 2.67; 95% CI, 1.15-6.19) and inclination angles (OR 2.71; 95% CI, 1.47-4.98), a shorter learning curve inflection point (18.48 vs. 63.53), a lower periprosthetic fracture rate (OR 0.54; 95% CI, 0.38-0.77) and a lower incidence of leg length-discrepancy (LLD) (MD -1.42; 95% CI, -2.19--0.66), but at the cost of a longer operation time (MD 18.16; 95% CI, 8.81-27.51). Moreover, there was no significant difference between the 2 groups in the other complications and HHS (MD -2.04; 95% CI, -6.84-2.76).

R-THA shortens the learning curve, improves implant positioning accuracy, and reduces LLD, although it is associated with longer operative time, while functional outcomes and complication rates remain largely comparable to M-THA.PROSPERO registration number: CRD42025636461.