Assessment of plantar pressure patterns during early postoperative walking after total hip arthroplasty

Introduction Total Hip Arthroplasty (THA) reduces pain and restores mobility in patients with degeneration hip osteoarthritis. Unilateral load management and gait symmetry are critical after surgery to prevent complications such as dislocation or joint overload [1,2]. Wearable sensors allow continuous, real-time monitoring of gait patterns and load distribution, supporting personalized rehabilitation plans [3,4]. This prospective cohort study evaluates the usage of insole pressure sensors to objectively track postoperative recovery in THA patients by assessing load distribution across the lower limbs, spatiotemporal gait parameters, and changes in loading patterns during the early postoperative phase. Methods The prospective cohort study was conducted on THA patients discharged from Istituto Ortopedico Rizzoli (Bologna, Italy). Assessments occurred after discharge (T0) and 20 days (T1) post-surgery, without intervening rehabilitation advice. At each assessment session, participants performed a 30 second Standing Test, three 10-m walking trials. Subsequently, they reported their levels of pain and fatigue via visual analogue scale (VAS) and Borg scale. Twelve patients (five females) were recruited, with a mean age of 51.3 ± 9.6 years, and mean BMI of 23.8 ± 2.6 kg/m². Kinetic parameters (mean, peak and impulse loads), center of pressure (COP) gait-line parameters and temporal parameters (stance, swing, stride durations) were extracted. Statistical analyses were conducted using a linear mixed-effects model, with experimental time (T0 vs. T1) and z-standardized covariates specified as fixed effects, and subject-specific intercepts and time slopes as random effects. P-values were adjusted via the Benjamini–Hochberg false discovery rate. All analyses were implemented in Python. Results In the operated limb, significant post-intervention improvements were observed in COP gait-line parameters. Gait-line chord increased from 11.3 ± 4.7 cm to 15.5 ± 2.9 cm (p=0.005, d=0.87) and gait-line velocity from 6.4 ± 2.9 cm/s to 15.7 ± 3.8 cm/s (p<0.001, d=3.7). All load and temporal metrics changed significantly (p<0.001): stance time decreased by approximately 50%, from 2119 ± 535 ms to 1040 ± 183 ms (d=–2.4); swing time fell from 632 ± 217 ms to 406 ± 106 ms (d=–1.3); and stride time was reduced from 2749 ± 556 ms to 1446 ± 240 ms (d=–2.7). Concurrently, average and peak loads increased by about 80%, with large effect sizes (d=2.2 and 2.5, respectively). No significant changes were detected in the COP excursion and length, and load impulse (all p>0.05). In the contralateral limb, only COP velocity and chordlength difference increased and both stance time and stride time decreased (all p<0.001). Discussion The three-week follow-up showed significant improvements in operated-limb gait mechanics, notably in foot rollover and load-bearing, with large effect sizes (d>0.8). In particular, enhanced foot rollover and propulsion (gait-line chord and velocity) alongside increases in load-bearing capacity suggest real gains in stability and walking efficiency, which in practice may translate into faster achievement of functional milestones such as independent ambulation and stair negotiation. The small cohort size, short follow-up, and absence of a standardized rehabilitation protocol limit this study.