Early postoperative gait changes after total hip arthroplasty in patients with hip osteoarthritis and osteonecrosis of the femoral head: an exploratory comparative cohort study

Total hip arthroplasty (THA) is a standard treatment for end-stage hip osteoarthritis (OA) and osteonecrosis of the femoral head (ONFH). Although both conditions can cause gait impairment, direct evidence comparing early postoperative gait changes between these etiologically distinct groups remains limited. This exploratory study compared pre-to-postoperative changes in gait quality, spatiotemporal parameters, and multiplanar joint kinematics between OA and ONFH patients after THA. Forty-eight patients undergoing unilateral robotic-assisted THA, including 22 with OA and 26 with ONFH, and 20 healthy controls underwent three-dimensional gait analysis. Patients were assessed preoperatively and at a mean follow-up of 4.19 ± 0.75 months. Gait profile score (GPS), gait variable scores (GVSs), spatiotemporal parameters, and sagittal, coronal, and transverse range of motion (ROM) were analyzed. Linear mixed-effects models adjusted for age, sex, and follow-up duration were used to evaluate Group × Time interactions, with false discovery rate correction applied separately within each outcome domain. No significant FDR-adjusted Group × Time interaction was detected for GPS or any GVS component. For GPS, the between-group difference in pre-to-postoperative change was 0.66 (95% CI − 1.37 to 2.69; P interact = 0.517, P adj = 0.939). No ROM parameter showed a significant FDR-adjusted Group × Time interaction across sagittal, coronal, or transverse planes. Among the spatiotemporal outcomes, contralateral stance proportion showed a between-group difference in change of 3.64 percentage points (95% CI 1.11 to 6.17 percentage points; P adj = 0.046). This interaction was attenuated and no longer remained significant after FDR correction in the walking-speed-adjusted sensitivity analysis (difference in change, 2.69 percentage points; P adj = 0.145). Both patient groups showed residual postoperative deviations from healthy-control reference values in several gait and kinematic parameters. In this exploratory single-center cohort, no clear diagnosis-related differential change was detected for GPS/GVS or multiplanar ROM after THA. Overall, no clear evidence of diagnosis-related differences in early gait recovery was detected, while the isolated contralateral stance finding warrants further confirmation.