Slow-wave sleep and obstructive sleep apnea in patients with type 2 diabetes mellitus

This study aimed to analyze sleep architecture and obstructive sleep apnea (OSA) in patients with type 2 diabetes and clarify the association between sleep characteristics and glycemic control.

All participants underwent metabolism-related laboratory testing and a cross-sectional analysis of nocturnal polysomnography for sleep parameter analysis. Data were analyzed using the chi-squared test, a one-way analysis of variance, and Kruskal-Wallis test to compare the differences among three groups (type 2 diabetes, prediabetes, and control groups). The prevalence of OSA was evaluated using descriptive statistics and comparing the group divided into HbA1c quartiles. Univariate and multivariate linear regression analyses were used to determine factors associated with glycemic control.

Of 75 study participants (age 57.3 ± 4.1 years, 32 men), there were 25 participants each in the type 2 diabetes, prediabetes, and control groups. Participants with type 2 diabetes had significantly decreased slow-wave sleep duration (77.9 ± 30.0 min, p = 0.026) and shortened rapid eye movement sleep latency (median 75 min, p = 0.018) compared with those in the prediabetes and control groups. Forty-five participants (60%) had OSA (apnea-hypopnea index ≥ 5/h), 18 of whom were in the type 2 diabetes group. The prevalence of OSA in this group was 72%. The prevalence of moderate-to-severe OSA was significantly higher in the type 2 diabetes group than in the control group (p = 0.025) and in groups with HbA1c levels of > 6.7% than in groups with HbA1c levels of < 5.3% (p = 0.007). Multiple regression analysis showed that dyslipidemia (β = 0.179, p = 0.000) and slow-wave sleep duration (β =  - 0.113, p = 0.008) were independently associated with the HbA1c level.

Our results suggest that increasing slow-wave sleep is positively associated with glycemic control.