Insomnia Increases Insulin Resistance and Insulin Secretion in Elderly People

To the Editor: Elderly people with insomnia showed high insulin resistance and high insulin secretion after an oral glucose tolerance test (OGTT). Decreased insulin secretion is not a cause of diabets but rather a consequence of insulin resistance in elderly people with insomnia. Aging is common factor known to contribute to the development of diabetes mellitus1 and insomnia.2 Difficulties initiating and maintaining sleep are associated with diabetes mellitus,3 and diabetes mellitus itself can impair the quantity and quality of sleep conversely.4 Although diabetes mellitus and insomnia are closely correlated with one another, it is controversial whether low insulin secretion or high insulin resistance more strongly influences the beginning of diabetes mellitus in elderly people with insomnia. The OGTT is the most sensitive measure of glycemic status, but no data exist to evaluate glycemic status using OGTT in elderly people with and without insomnia. To clarify the relationship between insomnia and glucose tolerance, the 75-g OGTT was used to evaluate elderly people with and without insomnia. In August 2006 and 2007 at the healthcare center of Tosa Town, 402 subjects aged 75 and older were recruited. Written informed consent for participation in the study was obtained from participants. Participants were divided into three groups according to the Japanese Society of Sleep Research definition and frequency of hypnotic drug use. Insomnia was defined as a complaint of insomnia with duration of at least 3 months, with hypnotic drugs used almost every day. Poor sleep was defined as a moderate to severe complaint of difficulty falling asleep, difficulty staying asleep, early final awaking, or unrefreshing sleep and occasional use of hypnotic drugs. Normal sleep was defined as the absence of any symptom and non-use of hypnotic drugs. Glycemic status was evaluated using the 75-g OGTT, and the functions of insulin resistance and insulin secretion were calculated using the following equations: homeostatic model assessment (HOMA)-R=fasting plasma glucose (mg/dL) × fasting plasma insulin (μU/mL)/405; HOMA-β=fasting plasma insulin (μU/mL) × 360/(fasting plasma glucose (mg/dL)−63). The insulin increase occurring after OGTT was calculated as a marker of insulin secretion. Paired t-tests and analyses of variance were used for analyses (JMP, SAS Institute, Inc., Cary, NC). Variables are presented as means with 95% confidence intervals (CIs). P<.05 was considered statistically significant. Subject characteristics are presented in Table 1. The mean age was 73.8. Female participants constituted 63.0% of the sample and had a higher prevalence of poor sleep. Body weight, body mass index, waist measurements, and fasting plasma glucose were not different between the three groups, although mean glucose concentration after OGTT (insomnia: 153.4 mg/dL, 95% CI=134.6–172.1; poor sleep: 145.4 mg/dL, 95% CI=134.0–156.9; normal sleep: 141.5 mg/dL, 95% CI=135.0–147.9, P=.009) was significantly higher in the insomnia group than in the other groups. Insulin before OGTT (insomnia: 8.2 μU/mL, 95% CI=6.9–9.4; poor sleep: 5.4 μU/mL, 95% CI=4.7–6.2; normal sleep: 5.1 μU/mL, 95% CI=4.7–5.6), insulin after OGTT (insomnia: 67.9 μU/mL, 95% CI=55.2–80.5; poor sleep: 43.7 μU/mL, 95% CI=36.0–51.4; normal sleep: 42.6 μU/mL, 95% CI=38.2–46.9), and insulin secretion according to OGTT (insomnia: 59.7 μU/mL, 95% CI=47.6–71.8; poor sleep: 38.3 μU/mL, 95% CI=30.9–45.7; normal sleep: 37.4 μU/mL, 95% CI=33.3–41.6, P<.001) were all significantly higher in the insomnia group. Insulin secretion after OGTT was significantly higher in the insomnia group. HOMA-R (insomnia: 2.11, 95% CI=1.76–2.46; poor sleep: 1.34, 95% CI=1.13–1.55; normal sleep: 1.26, 95% CI=1.14–1.38) and HOMA-β (insomnia: 83.4, 95% CI=69.9–96.1; poor sleep: 60.2, 95% CI=52.1–68.0; normal sleep: 57.6, 95% CI=53.2–62.0, P<.001) were significantly higher in the insomnia group. Data for glycosylated hemoglobin were not different between the three groups. The main finding of this study was that insomnia was associated with high insulin resistance and high insulin secretion. Sleep disturbances induced insulin resistance through several potential mechanisms, including sympathetic overactivity5 and altered secretion of counterregulatory hormones during sleep.6 Sympathetic overactivity and high counterregulatory hormones during sleep induced insulin resistance and thereby caused hyperinsulinemia in the morning.7 Therefore, the insulin increase after glucose loading shown in these results probably occurred to compensate for low insulin action caused by insulin resistance. A previous study reported that restricted sleeping induced high fasting plasma glucose and low insulin secretion.8, 9 These results appear to conflict with those reported herein, although insulin secretion was estimated according to intravenous glucose infusion in the earlier study. The current study investigated insulin secretion using OGTT, involving incretin effects, which engenders enhanced insulin secretion after oral glucose loading.10 Differences between these data and the data from the previous study might be attributable to incretin effects. The data from the current study indicate that elderly people with insomnia might have greater insulin resistance but that insulin secretion is prevented through incretin effects. In conclusion, insulin resistance is more influential than low insulin secretion on the glycemic status of elderly people with insomnia.