Objective: To test whether an intensive dietary weight loss program (IN) yields non-inferior symptom improvement vs. knee arthroplasty (KA) in people with knee osteoarthritis eligible for knee arthroplasty.
Design: A non-inferiority, randomised, comparative effectiveness trial involving participants with obesity (body mass index of ≥30 kg/m2) and knee osteoarthritis eligible for knee arthroplasty. After randomisation a 12 week wait list was installed, after which IN participants were offered a 12-week low-calorie meal replacement program with weekly group sessions focusing on behaviour and education supervised by a dietitian and the KA participants were offered surgical insertion of a knee arthroplasty. We aimed to recruit 200 participants. The primary outcome was knee pain and function assessed by the Oxford Knee Score (0-48 score; higher better) 6 months after treatment start, with a prespecified non-inferiority margin of 4 points. Secondary outcomes included change in body weight.
Results: Recruitment target was not reached. 92 participants were randomly assigned to IN (n=45) or KA (n=47). 6 months after treatment start the Oxford Knee Score was 28.2 with IN and 36.5 with KA, with a between-group difference of -8.2 (95%CI, -11.5 to -5.0) favouring KA; the 95% CI did not include the noninferiority margin of 4 points. The mean change in body weight was -11.0 kg (-10.4%) in the IN group and -1.3 kg (-1.2%) in the KA group. The adverse event incidence was similar in the two groups.
Conclusion: Among people with obesity and severe knee osteoarthritis eligible for knee arthroplasty a 12-week supervised intensive weight loss program was not non-inferior to knee arthroplasty, shown as clinically inferior improvements in patient reported knee pain and function after 6 months.
Trial registration: ClinicalTrials.gov number, NCT05172843.
Keywords: Surgery; Weight loss; arthropalsty; clinical trial; comparative effectiveness; diet.
Copyright © 2026. Published by Elsevier Ltd.