Background/Purpose: Adults with osteoarthritis (OA) have greater glucose concentrations and inflammation. It remains unclear if inflammation and glucose homeostasis are related to specific features of early stage OA. Bone marrow lesions (BMLs) and effusion are early features of OA that can change size in < 12 weeks and are associated with knee pain and OA progression. We explored if serum concentrations of impaired glucose homeostasis (glucose, glycated serum protein [GSP]) or inflammation (C-reactive protein; CRP) were associated with the prevalence of knee BMLs or effusion.

Methods: We conducted a cross-sectional study with baseline data from the Osteoarthritis Initiative. We selected participants who had no radiographic knee OA, were at high risk for knee OA, and had sagittal fat-suppressed knee magnetic resonance images and fasting serum. Blinded staff conducted assays for CRP, GSP, and glucose. One reader segmented BML volume using a semi-automated program (intra-tester ICC > 0.86). Two readers used a semi-automated program to segment effusion volume (intra-tester ICC > 0.84). We defined BML prevalence as a knee with a BML volume ≥ 1 cc and effusion prevalence as a knee with an effusion volume ≥ 7.5 cc. We performed separate logistic regression models for these 2 outcomes with each biomarker as a predictor, adjusting for age, sex, body mass index (BMI), and Physical Activity Scale for the Elderly (PASE) scores. We explored interactions between the biomarkers and BMI. When a nonlinear relationship was present we performed piece-wise logistic regressions.

Results: We included 343 participants with a mean age of 59 ± 9 years, BMI of 27.9 ± 4.5 kg/m², PASE score of 171 ± 82, and 64% female. CRP was associated with BML prevalence (odds ratio [OR] = 1.43, 95% confidence interval [CI] = 1.09 to 1.87; Table 1). For effusion, we found an interaction between BMI and CRP: among adults with a BMI < 25 kg/m² there was a trend towards a positive association between CRP and effusion (OR = 1.40, 95% CI = 1.00 to 1.97), while no significant association was observed among people with a BMI ≥ 25 kg/m². We detected a nonlinear relationship between GSP and effusion prevalence: for GSP levels ≥ 5.5cc people with greater concentrations were more likely to have effusion, while for GSP levels < 5.5cc people with lower concentrations were more likely to have effusion (Table 2).

Conclusion: Among individuals without knee OA, CRP is related to the presence of BMLs and effusion among normal weight individuals. Abnormal GSP is associated with effusion. It will be valuable to explore if inflammation and glucose homeostasis are predictive of symptomatic knee OA.