Abstract
The aim of the study was to assess potential associations between obstructive sleep apnea (OSA) and the occurrence of diabetes mellitus and insulin resistance in the elderly. Nondiabetic volunteers (n = 549) with undiagnosed or untreated asymptomatic OSA (66.2+/-1 years at the inclusion) were evaluated as an ancillary study of the PROOF cohort study (n = 1,011). After 7 years follow-up, 494 subjects underwent assessment of fasting insulin and glucose levels. OSA was defined by an apnea-hypopnea index (AHI) of ≥15/h using polygraphy. Diabetes mellitus was defined by a fasting glucose ≥ 1.26 g/L and/or when requiring pharmacological treatment, while insulin resistance corresponded to HOMA-IR ≥ 2. Asymptomatic OSA subjects (men or women) did not display increased risk of incident diabetes (2.8 vs. 3.9%, p = 0.51). However, there was a greater frequency of insulin resistance in subjects with severe OSA (AHI > 30) [OR 2.21; 95% CI (1.22-4.02); p = 0.009]. Furthermore, multiple logistic regression showed that triglycerides levels [OR 1.61; 95% CI (1.10-2.36); p = 0.01] and fasting glycaemia [OR 4.69; 95% CI (1.12-192.78); p = 0.04], but not AHI or oxyhemoglobin desaturation index were independently associated with higher rate of insulin resistance. The deleterious metabolic effect of asymptomatic OSA in the population may be indirectly mediated via perturbations in lipids, and is particularly likely to become manifest in severe apneic subjects with higher glycemic levels.