Comparison of weight loss interventions in overweight and obese adults with knee osteoarthritis: A systematic review and network meta-analysis of randomized trials
Arashi Shahid, Aricia Jieqi Thirumaran, Robin Christensen, V. Venkatesha, Marius Henriksen, Jocelyn L. Bowden, David J. Hunter
Abstract
OBJECTIVE: To ascertain the comparative effectiveness of weight-loss strategies for osteoarthritis (OA) to develop rational treatment algorithms aimed at improving OA-related symptoms in overweight/obese individuals. DESIGN: Medline, Embase, CINAHL, Scopus, and Web of Science were searched from inception to June 2023 for observational studies and randomized trials. Network meta-analyses were performed using a frequentist approach. Effect sizes for pain and function were computed as standardized mean differences, while change in body weight was computed as mean differences. RESULTS: 13 RCTs on knee OA (KOA) (2800 participants) with 7 interventions: diet (D); exercise (E); diet and exercise (DE); pharmacological (L); psychological (P); psychological, diet, and exercise (PDE); and Mediterranean diets (M) were networked. For weight change (kg), all interventions significantly outperformed control comparators, with effect sizes ranging from -11.2 (95% CI, -16.0, -6.5 kg) for the most effective approach (PDE) to -4.7 (95% CI, -6.7, -2.7 kg) for the least effective approach (DE). In terms of pain (0-20 scale), only DE outperformed control comparators (-2.2, 95% CI: -4.1, -0.21), whereas PDE was not superior to control comparators (-3.9, 95% CI: -8.4, 0.5) in improving the pain. Regardless of the chosen intervention, prediction intervals from meta-regression analysis indicate that significant pain relief may be anticipated when patients achieve at least a weight reduction of 7%. CONCLUSIONS: PDE and DE interventions may offer the most effective approach for weight loss, potentially leading to improvements in pain and physical function among overweight/obese individuals with KOA if they achieve more than 7% weight loss.