The Association Between Sleep Duration, Insulin Sensitivity, and β-Cell Function: The EGIR-RISC Study.

Abstract

In the past decade, over 3 dozen studies reported a relationship between self-reported short sleep and disturbed glucose metabolism. A study with insulin sensitivity assessed according to the gold standard hyperinsulinemic-euglycemic clamp is, however, still missing. To evaluate the cross-sectional association of sleep duration with insulin sensitivity and β-cell function in the European group for the study of insulin resistance (EGIR-RISC) study cohort. We used data from the baseline measurements of the European, multicentre EGIR-RISC study that included 1319 clinically healthy participants. Sleep and physical activity were measured using a single-axis accelerometer. Insulin sensitivity and β-cell function were estimated by hyperinsulinemic-euglycemic clamp and from the oral glucose insulin sensitivity index model, using an oral glucose tolerance test. Associations of sleep duration with insulin sensitivity and β-cell function were analyzed by multiple linear regression, stratified by sex. In our current analysis, we included 788 participants (57% women, age 44 ± 8 y), who had an average sleep duration of 7.3 ± 1.5 hours. In men, we observed an inverted U-shaped association between sleep duration categorized per hour and M/I (in μmol/min per kgFFM/nM per hour) (β-estimate [95% confidence intervals] 41 [2, 80]; P = .04 and β(2)-estimate -3 [-6, -0.2], P = .04) as well as a trend for the oral glucose insulin sensitivity index (in mL/min per kgFFM) (β-estimate [95% confidence intervals] 0.8 [-0.4, 2]; P = .17). In women, we observed a U-shaped association between sleep duration and β-cell function (in pmol/min per m(2)/mM per hour) (β-estimate -45 [-86, -3]; P = .04 and β(2)-estimate 3 [0.2, 6]; P = .04). Sleep duration is associated with insulin sensitivity and β-cell function in a sex-specific manner in clinically healthy people.