Abstract
This meta-analysis compared the gait patterns of unicompartmental knee arthroplasty (UKA) patients and total knee arthroplasty (TKA) patients during level walking by evaluating the kinetics, kinematics, and spatiotemporal parameters. Studies were included in the meta-analysis if they assessed the vertical ground reaction force (GRF), joint moment at stance, flexion at initial contact, flexion at swing, overall range of motion (ROM), coronal knee angle at stance, walking speed, cadence, and stride length in UKA patients or TKA patients. Seven non-randomized studies met the criteria for inclusion in this meta-analysis. UKA patients and TKA patients were similar in terms of vertical GRF (95% CI: -0.36 to 0.20; P = 0.60), joint moment (95% CI: -0.55 to 0.63; P = 0.90), kinematic outcomes (95% CI: -0.72 to 1.02; P = 0.74), walking speed (95% CI: -0.27 to 0.81; P = 0.32), and cadence (95% CI: -0.14 to 0.68; P = 0.20). In contrast, the stride length (95% CI: 0.01 to 0.80; P = 0.04) differed significantly between groups. Subgroup analyses revealed that the pooled data were similar between the groups: 1st maximum (heel strike), -0.18 BW (P = 0.53); 1st minimum (mid-stance), -0.43 BW (P = 0.08); and 2nd maximum (toe off), -0.03 BW (P = 0.87). On gait analysis, there were no significant differences in vertical GRF, joint moment at stance, overall kinematics, walking speed, or cadence between UKA patients and TKA patients during level walking. However, the TKA group had significantly shorter stride length than UKA patients. Although the comparison was inconclusive in determining which types of knee arthroplasty offered the closest approximation to normal gait, we consider it important to provide better rehabilitation programs to reduce the abnormal stride length in TKA patients compared to UKA patients.