Obstructive sleep apnoea and incident diabetes mellitus in a Chinese population

Abstract

Introduction: Although clinical and epidemiological studies have reported an association between obstructive sleep apnoea (OSA) and diabetes mellitus (DM), the results remain inconsistent. We hence propose to evaluate the role of OSA on the incidence of DM in a Chinese cohort. Methods: The eligible cohort was drawn from those who received sleep studies between 1 January 2007 and 31 December 2010 in the Department of Medicine, Queen Mary Hospital. Only those without DM at the time of polysomnogram were included. Follow-up data were drawn from the Hospital Authority Clinical Management System up to the day of data entry in this study. Apnoea-hypopnoea index (AHI) from automated scoring of polysomnogram was used in the Cox regression model, controlling for age, gender, body mass index (BMI), daytime sleepiness, and comorbidities. Results: Overall, 101 (14.5%) of 698 subjects experienced DM over a median follow-up of 82 months, giving an incident rate of 2.2 per 100 person-years. In fully adjusted models, AHI was significantly and independently associated with incident DM (hazard ratio=1.006; 95% confidence interval, 1.001-1.011), as were age and BMI. This analysis has not yet taken into account the influence of OSA treatment and manually scored sleep data. Conclusion: AHI, as a measure of OSA severity, predicted incident diabetes, independent of obesity.