Background and objective Serum leptin measures are associated with radiographic knee osteoarthritis, but no studies have examined leptin levels with respect to different measures of knee joint damage from MRI.

Methods Participants in the Michigan Study of Women's Health Across the Nation underwent bilateral knee MRIs at follow-up visit 11 for assessment of cartilage defects, bone marrow lesions, osteophytes, meniscal tears, synovitis and joint effusion. Serum leptin measures were available from baseline, follow-up visits 1 and 3–7.

Results Baseline serum leptin levels were associated with greater odds of having more severe knee joint damage at follow-up visit 11 after adjustment for age, smoking status, menopause status and body mass index residuals. The greatest effect was observed for osteophytes; a 5 ng/ml increase in baseline leptin was associated with 24% higher odds of having larger osteophytes (95% CI 1.17 to 1.32). Correlations with baseline serum leptin were greatest for MRI-assessed osteophytes (r=0.41), followed by effusion (r=0.32), synovitis (r=0.30), cartilage defects (r=0.28), bone marrow lesions (r=0.24) and meniscal abnormalities (r=0.21).

Conclusions Leptin levels 10 years prior to MRI assessment were associated with the presence of cartilage defects, bone marrow lesions, osteophytes, meniscal tears, synovitis and effusion among a population of middle-aged women. Understanding the role that leptin plays in the joint degradation process is critical for development of more targeted interventions for osteoarthritis.