S 182 Osteoarthritis and Cartilage Vol . 16 Supplement 4 418 THE CLINICAL ASSOCIATIONS OF KNEE CARTILAGE DEFECTS IN AN OBESE POPULATION
F. Eckstein,C. Hutchinson,L. Sharma,M. Marshall,R. Maciewicz,C. Taylor,J. Waterton,S. Cahue,A. Holmes,T. Williams,M. Hudelmaier,G. Vincent
Abstract:Purpose: To generate hypothesis regarding the associations of knee cartilage defects in an obese population with and without OA. Methods: Obese subjects (BMI> 30) were recruited from laparoscopic adjustable gastric banding or exercise and diet weight loss programs. ACR clinical criteria for OA knee were determined. MRI assessment was conducted on a 3T machine (Magnetom Trio; Siemens, Erlangen, Germany) at baseline. Cartilage defects were graded according to an established protocol at medial tibial, medial femoral, lateral tibial, lateral femoral and patellar sites (grade 0−4). Composite scores for medial tibiofemoral, lateral tibiofemoral and whole compartments were obtained. All subjects completed a general health and musculoskeletal (MSK) questionnaire, WOMAC, SF-36, AQoL, Moorehead-Ardelt Quality of Life Questionnaire II (developed in an obese population), and Multidimensional Assessment of Fatigue. Physical assessment was conducted by a single assessor. The MSK questionnaire assessed joint pain; activity level at work, home and leisure; medication use; previous injury and previous surgery. Physical assessment parameters included: range of knee and hip motion; crepitus; alignment; quadriceps strength; waist, hip and knee circumference. Correlations between cartilage defect scores and MSK outcomes were obtained (Stata software). Regression analysis corrected for age, gender, BMI and presence of knee OA. Results: There were 111 subjects (78 women and 33 men) with mean BMI 39.9±5.8 and mean age 51±12 years. Fifty-three (48%) subjects met ACR criteria for knee OA. Knee cartilage defects were significantly associated with BMI (p< 0.03), presence of knee OA (p< 0.0001), knee pain (p< 0.05), and duration of knee pain (p< 0.01) in the medial tibiofemoral and whole compartments. There was statistically significant association with BMI and the lateral tibiofemoral compartment (p< 0.01). Significant associations were also seen with physical function, role physical, bodily pain and general health on SF-36 (p< 0.05); pain, stiffness and function on WOMAC (p< 0.0001); and medication use and independence subscale (p< 0.02) on AQoL for both the medial tibiofemoral and whole compartments. The association between knee cartilage defects, and BMI (p< 0.04) and knee OA (p< 0.0001) remained significant in the medial tibiofemoral and whole compartments on regression analysis. ICCs for intra-observer reproducibility were >0.98 for all compartments. Conclusions: Knee cartilage defects scores are significantly associated with BMI, knee OA, the presence and duration of knee pain. Cartilage defects are also significantly associated with a number of quality of life measures indicating clinical significance. They provide a time-efficient means of assessing articular cartilage which is reproducible. With the rising prevalence of obesity, it is important to assess the natural history of cartilage defects in the obese population, a group at risk for both incident and prevalent OA.
Medicine