Associations among preoperative MRI features and functional status following arthroscopic partial meniscectomy.

BACKGROUND: Arthroscopic partial meniscectomy (APM) is the most frequently performed orthopedic procedure. Functional outcomes of APM are variable, particularly among patients with underlying knee osteoarthritis. While most patients undergoing APM have knee magnetic resonance imaging (MRI) performed preoperatively, the prognostic value of knee MRI in predicting the functional outcomes of APM has not been evaluated. METHODS: We studied patients who had APM performed by one of five participating surgeons at one institution in 2002. The preoperative MRI scans of these patients were assessed using a standardized rating system by an independent observer who was not involved in the care of the patients and who was blinded to patient outcomes. Patients completed a questionnaire in the summer of 2003, 6-18 months postoperatively. The questionnaire included the Knee Injury and Osteoarthritis Outcome Score (KOOS) and items on satisfaction with the results of surgery. We used bivariate and multivariate techniques to evaluate the associations between MRI findings, other preoperative findings, and the functional status and pain scales of the KOOS. RESULTS: Eighty-three patients were included in the analyses. The outcome of surgery was variable with average KOOS functional score of 77 and range of 15-100. One-quarter of patients were somewhat or very dissatisfied with the results of surgery and 17% were using a cane at the time of follow-up. In bivariate analyses, preoperative predictors of KOOS function score at follow-up included preoperative functional status and several MRI findings including the extent of cartilage damage, bone marrow edema in the medial compartment, and length of the tear. Multivariate analyses showed that after adjusting for preoperative functional status, the extent of cartilage signal abnormality in the medial compartment on MRI remained an independent predictor of functional status, 6-18 months following surgery. Specifically, preoperative functional status explained 21% of the variability in follow-up KOOS functional status score and the extent of medial tibial cartilage damage on MRI explained an additional 16%. Analyses of knee pain 1 year following APM yielded similar findings, with preoperative functional status accounting for 17% of the variability in pain scores and medial tibial cartilage damage accounting for an additional 13%. CONCLUSIONS: Preoperative MRI findings of cartilage damage have independent prognostic value in predicting the functional outcome of APM. This study was limited by a cross-sectional design with retrospective recall of preoperative functional status. Thus, the findings need to be confirmed in prospective investigations.