Five-Year Structural Changes in the Knee Among Patients With Meniscal Tear and Osteoarthritis: Data From a Randomized Controlled Trial of Arthroscopic Partial Meniscectomy Versus Physical Therapy.

Abstract

To estimate the risk of magnetic resonance imaging (MRI)-based structural changes in knee osteoarthritis (OA) among individuals with meniscal tear and knee OA, using MRIs obtained at baseline and 18 and 60 months after randomization in a randomized controlled trial of arthroscopic partial meniscectomy (APM) versus physical therapy (PT). We used data from the Meniscal Tear in Osteoarthritis Research (METEOR) trial. MRIs were read using the MRI OA Knee Score (MOAKS). We used linear mixed-effects models to examine the association between treatment group and continuous MOAKS summary scores, and Poisson regression to assess categorical changes in knee joint structure. Analyses assessed changes in OA between baseline and month 18 and between months 18 and 60. We performed both intention-to-treat and as-treated analyses. The analytic sample included 302 participants. For both treatment groups, more OA changes were seen during the early interval than during the later interval. ITT analysis revealed that, between baseline and month 18, APM was significantly associated with an increased risk of having a worsening cartilage surface area score, involving both any worsening across all knee joint subregions (risk ratio [RR] 1.35 [95% confidence interval (95% CI) 1.14, 1.61]) and the number of subregions damaged (RR 1.44 [95% CI 1.13, 1.85]) having a worsening effusion-synovitis score (RR 2.62 [95% CI 1.32, 5.21]), and having ≥1 additional subregion with osteophytes (RR 1.24 [95% CI 1.02, 1.50]). Significant associations were detected between months 18 and 60 only for having any subregion with a worsening osteophyte score (RR 1.28 [95% CI 1.04, 1.58]). These findings suggest that the association between APM and MRI-based structural changes in knee OA is most apparent during the initial 18 months after surgery. The reason for attenuation of this association over longer follow-up merits further investigation.