Background Body mass index (BMI) is an important risk factor for osteoarthritis (OA), however the underlying causal pathways are unclear. Earlier studies showed an association of diabetes mellitus and OA, suggesting involvement of the glucose metabolism. Furthermore in OA the process of non-enzymatic glycation has been described, ultimately leading to advanced glycation end products (AGEs), which have been linked to cartilage damage.
Objectives We investigated whether the association of BMI and OA in the hands, being non-weight bearing joints, can be explained by glucose or insulin concentrations. Furthermore, we investigated whether HbA1c, being an early stage glycation product, was associated with hand OA.
Methods Data from participants of the Netherlands Epidemiology of Obesity (NEO) study, a population-based cohort of men and women aged 45-65 years with an oversampling of individuals with a BMI ≥ 27 kg/m2, were used. Individuals using glucose lowering medication were excluded. BMI was assessed by measured weight in kg and length in cm. Fat mass was assessed in kg using bioelectrical impedance analysis. Glucose (mmol/L), HbA1c (%) and insulin (mU/L) were measured after an overnight fast and the revised HOMA-IR (measure of insulin resistance) was calculated.
Hand OA was defined using the criteria of the American College of Rheumatology; pain was measured using a standardized questionnaire and physical examination of the hands was performed by trained research nurses.
Odds ratios (OR) with 95% confidence intervals (CI) were calculated to associate BMI, glucose, HbA1c, insulin and HOMA-IR with hand OA using logistic regression analyses per standard deviation, stratified for sex and adjusted for age.
Results After exclusion of participants using glucose lowering medication (n=304), data from 4980 participants were analyzed, including 834 individuals with a BMI < 27 kg/m2. Mean age was 56 ± 6 years and 54% was female. Mean BMI of the total study population was 30.4 ± 4.7 kg/m2, mean glucose was 5.62 ± 0.80 mmol/L, mean HbA1c 5.40 ± 0.39 %, median (IQR) insulin was 6.17 (2.47-10.70) mU/L and median (IQR) HOMA-IR 0.83 (0.33-1.43). Hand OA was present in 7% of men and 19% of women.
BMI was associated with hand OA in both men, OR 1.19 (95% CI 1.03-1.38), and women, OR 1.11 (95% CI 1.01-1.23). Glucose was associated with hand OA in men only; OR 1.22 (95% CI 1.08-1.37). Insulin and HOMA-IR were also positively associated with hand OA in men. Next, we assessed the association of HbA1c with hand OA to investigate whether the process of glycation could be involved. HbA1c showed an OR of 1.17 (95% CI 1.04-1.31) for hand OA in men. Remarkably, HbA1c was also associated with hand OA in women; OR 1.12 (95% CI 1.02-1.23). The associations of glucose, HbA1c and HOMA-IR with hand OA remained significant after adjustment for fat mass.
Conclusions BMI was associated with clinical hand OA in both men and women. In men, glucose, insulin and HOMA-IR were associated with hand OA. HbA1c was associated with hand OA in both sexes, independent of fat mass, suggesting involvement of non-enzymatic glycation in the development of hand OA. Strict glucose control might therefore have a beneficial effect on the incidence of hand OA.
Disclosure of Interest None Declared