Abstract
AIMS: To determine the most effective resistance training (RT) type and the optimal intensity, duration, and number of repetitions for patients with knee osteoarthritis (KOA). METHODS: PubMed, MEDLINE, Embase, CENTRAL, and Web of Science were searched from inception to April 2025 for randomized controlled trials (RCTs) of exercise for KOA. Bayesian network meta-analyses were conducted. The outcomes of pain, stiffness, and function were expressed as standardized mean differences (SMDs) with 95% credible intervals (CrI). RESULTS: The analysis included 46 RCTs with 3463 participants. High-speed RT was identified as the most effective type, significantly improving pain (SMD: -1.35, 95% CrI: -1.89 to -0.81), stiffness (SMD: -1.26, 95% CrI: -1.81 to -0.70), and function (SMD: 1.70, 95% CrI: 0.96 to 2.45). Dose-response analysis indicated that moderate-intensity RT (43-47% 1RM) for 35-37 weeks with 610-640 weekly repetitions was optimal for reducing pain (SMD: -0.76, 95% CrI: -1.03 to -0.44) and improving function (SMD: 1.29, 95% CrI: 0.92 to 1.68). For stiffness, a higher number of repetitions and shorter duration (12 weeks, 1200 weekly repetitions) were more effective (SMD: -1.11, 95% CrI: -1.58 to -0.64). CONCLUSIONS: This study identifies high-speed RT as the most effective type for managing KOA and provides optimal RT dosages for different symptoms. These findings emphasize the need for tailored RT programs and provide crucial insights for clinical guidelines and individualized patient care. PROSPERO NUMBER: CRD42024535806.