Abstract
BACKGROUND: Osteoarthritis(OA) is a major source of pain, disability, and socioeconomic cost in worldwide. However, there is no effective means for the early diagnosis of OA, nor can it accurately predict the progress of OA. To develop and validate a novel nomogram to predict the radiographic progression of mild to moderate OA based on three-dimensional(3D)-MRI bone shape and bone shape change during 24 months. METHOD: Analysis of publicly available data from the Foundation for the National Institutes of Health (FNIH) OA Biomarkers Consortium. Radiographic progression was defined as minimum radiographic narrowing of the medial tibiofemoral joint space of ≥ 0.7 mm from baseline at 24, 36, or 48 months. There were 297 knees with radiographic progression and 303 without. The bone shapes of the tibia, femur, and patella were evaluated by 3D-MRI at the baseline and at 24 months. Two nomograms were separately established by multivariate logistic regression analysis using clinical risk factors, bone shape at baseline (nomogram 0), or bone shape change at 24 months (nomogram Δ24). The discrimination, calibration, and usefulness were selected to evaluate the nomograms. RESULTS: There were significant differences between groups in baseline Kellgren-Lawrence (KL) grade, gender, age, and tibia, femur, and patella shape. The areas under the curve (AUC) of nomogram 0 and nomogram Δ24 were 0.66 and 0.75 (p < 0.05), with accuracy of 0.62 and 0.69, respectively. Both nomograms had good calibration. The decision curve analysis ( DCA) showed that nomogram Δ24 had greater clinical usefulness than nomogram 0 when the risk threshold ranged from 0.04 to 0.86. CONCLUSIONS: Nomograms based on 3D-MRI bone shape change were useful for predicting the radiographic progression of mild to moderate OA.