Can Baseline MRI Findings Identify Who Responds Better to Early Surgery Versus Exercise and Education in Young Patients With Meniscal Tears? A Subgroup Analysis From the DREAM Trial

OBJECTIVE: To investigate whether MRI findings modified the outcomes of early surgery compared to exercise and education in young patients with meniscal tears. DESIGN: A secondary effect modifier analysis of a randomized controlled trial. METHODS: The primary outcome was change from baseline to 12 months in the mean score of four Knee Injury and Osteoarthritis Outcome Score subscales (KOOS4). Three potential MRI-defined effect modifiers were pre-defined: 1) type of meniscal tear dichotomized to bucket-handle/complex versus simple, 2) meniscus affected (medial or lateral), and 3) presence of knee effusion/synovitis. We used a linear mixed model to investigate the difference in mean change between the treatment groups, stratified by each of the three potential effect modifiers, and estimated the interactions. An adjusted effect difference ≥ 10.0 points (0?100 scale) was considered clinically relevant. RESULTS: Data from all participants (60 in the surgical group and 61 in the exercise group) were analysed. The mean (SD) age was 29.7 (6.6) years, and 28% were female. A potential effect modification was observed for knee effusion/synovitis, with its presence implying an increase of the effect of early surgery by 11 points on KOOS4 (p=0.07). CONCLUSION: Knee effusion/synovitis on MRI potentially modified the treatment effect with a clinically relevant difference in change of KOOS4 in patients with effusion/synovitis, favoring early surgery. We found no indication that patients with bucket-handle/complex versus simple or medial versus lateral tears benefitted more from early surgery.