Association Between Hypnotics and Glycemic Variability Assessed by Continuous Glucose Monitoring Under Real-Life Conditions in Patients With Type 2 Diabetes: A Cross-Sectional Study.

BACKGROUND

Insomnia is common in patients with Type 2 diabetes and can negatively affect glycemic control. However, the effect of hypnotic use on glycemic variability remains unclear. Therefore, we investigated the association between hypnotic use and glycemic variability in patients with Type 2 diabetes.

METHODS

This cross-sectional study enrolled patients with Type 2 diabetes who underwent continuous glucose monitoring (CGM) between June 1, 2017, and February 28, 2022. Patients were classified into six groups based on their insomnia status and hypnotic use: noninsomnia, hypnotic nonusers, benzodiazepine (BZD) users, nonbenzodiazepine (non-BZD) users, orexin receptor antagonist (ORA) users, and melatonin receptor agonist (MRA) users. We used the standard deviation (SD) of glucose, the coefficient of variation (CV) of glucose, and the mean of daily difference (MODD) as indicators of glycemic variability. The independent association between hypnotic use and glycemic variability was assessed using a multiple linear regression model.

RESULTS

A total of 534 patients were included in the analysis (mean age: 67.7 ± 10.1 years old; mean diabetes duration: 14.5 ± 8.4 years). Thirty-seven patients (6.9%) used hypnotics, including BZD (n = 13), non-BZD (n = 10), ORA (n = 11), and MRA (n = 3). The SD was significantly higher in non-BZD users (53.6 mg/dL, 95% confidence interval [CI]: 42.9-64.3) than in the noninsomnia group (40.5 mg/dL, 95% CI: 39.5-41.5). MODD was also significantly higher in non-BZD users (50.1 mg/dL, 95% CI: 38.0-62.1) than in the noninsomnia group (35.6 mg/dL, 95% CI: 34.5-36.7). In contrast, the CV was not significantly different between non-BZD users and the noninsomnia group. When analyzed separately for different times of the day, the nocturnal CV was significantly higher in non-BZD users than in the noninsomnia group.

CONCLUSIONS

The use of non-BZDs was associated with within-day and between-day glycemic variability measured by CGM in patients with Type 2 diabetes.