This retrospective matched cohort study analyzed data from the Osteoarthritis Initiative, comparing 430 individuals who achieved clinically significant weight loss (>5% of body weight) with 860 weight-stable controls over 24 months. The population included individuals with (ROA) and without (non-ROA) radiographic osteoarthritis, with 164 cases/328 controls in the non-ROA group and 266 cases/532 controls in the ROA group. The primary outcome was the 2-year change in the novel Cartilage Thickness Score (CTh-Score).
In individuals without radiographic OA, the median delta CTh-Score was 1.58 (IQR 0.61, 6.53) in the weight loss group versus 3.14 (IQR 0.44, 7.12) in controls (p=0.005), representing approximately 50% lower progression. Similarly, in those with radiographic OA, the median delta was 1.69 (IQR 0.97, 6.71) versus 2.90 (IQR 0.19, 7.38) (p=0.004), also showing about 50% slower progression. No between-group differences were detected for changes in joint space width or regional mean cartilage thickness.
For patient-reported outcomes, a trend toward greater pain improvement was observed in the ROA weight loss group (delta KOOS Pain: 2.75 vs 0.00, p=0.06). Safety and tolerability data were not reported. Key limitations include the observational design, which precludes definitive causal inference, and the use of a novel composite score (CTh-Score) whose clinical meaning requires further validation. The lack of effect on conventional structural metrics (JSW, regional thickness) warrants caution.
In practice, these findings support the role of weight management as a potential structural disease-modifying strategy in knee OA. However, clinicians should interpret the association conservatively, recognizing that the evidence is observational and the primary outcome measure is investigational. Weight loss recommendations should remain grounded in established benefits for symptoms and function.
View Original Abstract ↓
Purpose To determine whether clinically significant weight loss (>5% of body weight) is associated with slower 2-year knee cartilage degeneration in individuals with and without radiographic osteoarthritis. This study used a cartilage structural assessment score derived from the spatial distribution of cartilage thickness, referred to as the Cartilage Thickness Score (CTh-Score). It is based on cartilage thickness patterns and scores the cartilage between 0 and 100, with higher scores indicating greater severity. Methods We conducted a retrospective matched cohort study within the Osteoarthritis Initiative. High-resolution cartilage thickness maps (CTh-Maps), along with their corresponding CTh-Score, were extracted from a public repository. Participants with complete radiographic and MRI data at baseline and 24 months were stratified by baseline Kellgren-Lawrence (KL) grade into non-radiographic OA (non-ROA; KL<2) and radiographic OA (ROA; KL>=2). Within strata, cases (>5% 2-year weight loss) were propensity score-matched 1:2 to weight-stable controls on age, sex, height, weight, KL grade, joint space width (JSW), KOOS Pain, baseline CTh-Score, and mean cartilage thickness in the medial and lateral femoral and tibial compartments. The primary outcome was 2-year change (delta) in CTh-Score, where higher values indicate worsening. Secondary outcomes were delta JSW, delta regional mean cartilage thickness, and delta KOOS Pain. Non-parametric tests were used. Results We included 164 cases and 328 controls in non-ROA, and 266 cases and 532 controls in ROA. Median (interquartile range) weight loss was -6.10 kg (-8.90, -4.70) versus +0.30 kg (-1.30, 2.20) in non-ROA and -6.80 kg (-9.10, -5.02) versus +0.40 kg (-1.40, 2.82) in ROA (both p<0.001). Weight loss was associated with significantly smaller 2-year increases in CTh-Score: in non-ROA, median 1.58 (0.61, 6.53) vs 3.14 (0.44, 7.12) (p=0.005); in ROA, median 1.69 (0.97, 6.71) vs 2.90 (0.19, 7.38) (p=0.004). No between-group differences were detected for delta JSW or delta regional mean cartilage thickness in any of the 4 ROIs. A trend toward greater KOOS Pain improvement with weight loss was observed in ROA: 2.75 (-3.35, 13.40) vs 0.00 (-5.60, 8.40) (p=0.06). Conclusions Achieving >5% weight loss over 2 years is associated with approximately 50% lower progression in median cartilage degeneration, as assessed by CTh-Score, in both non-ROA and ROA. No change was observed with conventional structural metrics. These findings support weight management as a structural disease-modifying strategy and highlight CTh-Score as a sensitive endpoint.
