Sleep biosignature of Type 2 diabetes: a case-control study.

Abstract

To determine whether or not the sleep disturbances associated with Type 2 diabetes affect the structure of sleep. We designed a case-control study in 76 patients with Type 2 diabetes and 76 control subjects without Type 2 diabetes, matched by age, gender, BMI and waist and neck circumferences. A subgroup of 32 patients with Type 2 diabetes was also matched with 64 control subjects without Type 2 diabetes according to apnoea-hypopnoea index score. Examination included an overnight full polysomnography. No differences in the percentage of time spent in either rapid eye movement or non-rapid eye movement sleep were observed between groups; however, patients with Type 2 diabetes had more microarousal events during sleep than control subjects [41.4 (total range 4.0-104.4) vs 20.7 (total range 1.3-94.5) events/h; P < 0.001]. These differences were mainly observed during the non-rapid eye movement sleep [7.4 (total range 0-107.2) vs 0.2 (total range 0-65.2) events/h; P < 0.001]. In addition, sleep variables related to oxygen saturation measures, such as the percentage of time spent with oxygen saturation ≤90%, were significantly greater during the rapid eye movement sleep in patients with Type 2 diabetes [20.3 (total range 0-99.2) vs. 10.5 (total range 0-94.0)%; P = 0.047]. This pattern was maintained in the subgroup of patients matched by apnoea-hypopnaea index. Finally, stepwise regression analyses showed that apnoea-hypopnoea index, the presence of Type 2 diabetes and fasting plasma glucose value were independently associated with the number of microarousals (R2 =0.667). Type 2 diabetes is associated with an altered sleep structure, with different effects according to rapid eye movement (increase in nocturnal hypoxia) or non-rapid eye movement (increase in sleep fragmentation) sleep.