How intense is effective? Exploring aerobic exercise intensity for knee osteoarthritis through a Bayesian network meta-analysis

Introduction Knee osteoarthritis (KOA) is a degenerative disease of the articular cartilage, characterized by pain, stiffness, and functional limitation. Exercise is one of the main treatment options. Aerobic exercise has shown favorable effects in these patients. Aims Determine which intensity of aerobic exercise (AE) is most effective for these patients. Methods Searches were conducted in 7 databases, complemented by citation and manual searches. Eligible studies were randomized controlled trials with an AE-only group. Outcome measures were knee pain intensity, walking performance, sit-to-stand performance, combined sit-to-stand and walking performance, perceived knee stiffness and disability. Methodological quality and risk of bias were assessed. Two Bayesian random-effects network meta-analysis models were employed for comparing different AE intensities, reporting standardized mean differences (Hedges’ g) with 95% credible intervals. The certainty of evidence was evaluated using the adapted GRADE framework for network meta-analysis exploring risk of bias, indirectness, heterogeneity, publication bias, intransitivity, incoherence and imprecision of the estimate. Results Fifteen studies were finally included. They had an average of 5.93 points in the PEDro scale and a risk of bias including “some concerns” and “high risk”. Two meta-analyses (pain intensity and walking performance) were conducted. Comparisons between AE intensities presented non-significant and imprecise results, classified with a very low certainty of evidence. Conclusions Very low certainty of evidence prevents clear recommendations on the optimal dose of AE. Current findings are limited by high risk of bias, imprecise credible intervales, and reliance on indirect comparisons. Clinicians should strictly individualize intensity based on patient characteristics, preferences, and clinical response, and combine AE with other interventions (physiotherapy, self-management, education, medication control). Medium-to-long term AE programs are advisable given their additional benefits on frequent comorbidities. More high-quality trials directly comparing AE intensities are needed.