Gait recovery after total hip arthroplasty: insights from principal component analysis of pre- and post-surgical walking performance

Abstract

Total hip arthroplasty (THA) alleviates pain and restores function in hip osteoarthritis, yet gait abnormalities may persist. We tested two hypotheses: (H1) THA patients would improve from preoperative (M0) to six months postoperative (M6) in principal component analysis (PCA)–derived gait outcomes; (H2) despite improvement, patients at M6 would still differ from healthy adults. Full-body overground kinematics from 50 THA patients (64.9 ± 9.9 years) and 50 healthy adults (64.5 ± 8.9 years) were decomposed using PCA into coordinated movement components (“principal movements,” PMs). Each PM was quantified by the relative explained variance (rVAR; structural contribution), the number of zero-crossings of PM accelerations (N; control reversals/smoothness), and the root-mean-square of PM accelerations (RMS; acceleration magnitude). The first five PMs (PM_1–5) mapped to recognizable phases: PM₁—forward progression/propulsion; PM₂—swing-phase movement; PM₃—single-leg support with mediolateral stabilization; PM₄—single-leg support with trunk rotation; PM₅—double-support transition. Consistent with H1, from M0 to M6 patients showed increased PM₂ rVAR (p = 0.005), reduced N across all PMs (all p ≤ 0.003), and reduced RMS in PM_3–5 (p ≤ 0.001). Consistent with H2, at M6 patients still exhibited a trend in higher rVAR (unadjusted p = 0.022; not significant after multiplicity control) and significantly higher RMS (p = 0.002) in PM₃ versus healthy adults. These findings indicate that THA yields substantial gains over six months—particularly in swing-phase structure and overall smoothness—yet phase-specific deviations persist during single-leg support (mediolateral stabilization), highlighting targets for extended rehabilitation to improve walking performance.