Abstract
BACKGROUND: Despite the advent of innovative knee prosthesis design, a consistent first-option knee implant design in total knee arthroplasty (TKA) remained unsettled. This study aimed to compare the clinical effects among posterior-stabilized (PS), cruciate-retaining (CR), bi-cruciate substituting (BCS), and bi-cruciate retaining designs for primary TKA. METHODS: Electronic databases were systematically searched to identify eligible randomized controlled trials (RCTs) and cohort studies from inception up to July 30, 2021. The primary outcomes were the range of knee motion (ROM), and the secondary outcomes were the patient-reported outcome measures (PROMs) and complication and revision rates. Confidence in evidence was assessed using Confidence in Network Meta-Analysis. The Bayesian network meta-analysis was performed for synthesis. RESULTS: A total of 15 RCTs and 18 cohort studies involving 3520 knees were included. The heterogeneity and inconsistency were acceptable. There was a significant difference in ROM at the early follow-up when PS was compared with CR (mean difference [MD] = 3.17, 95% confidence interval [CI] 0.07, 7.18) and BCS was compared with CR (MD = 9.69, 95% CI 2.18, 17.51). But at the long-term follow-up, there was no significant difference in ROM in any one knee implant compared with the others. No significant increase was found in the PROMs and complication and revision rates at the final follow-up time. CONCLUSIONS: At early follow-up after TKA, PS and BCS knee implants significantly outperform the CR knee implant in ROM. But in the long run, the available evidence suggests different knee prostheses could make no difference in clinical outcomes after TKA with extended follow-up.