Obstructive Sleep Apneoa Is Associated With Incident Pre-diabetes In A Population Cohort Of Men

Rationale: There have been concomitant increases in the prevalence of obstructive sleep apnea, diabetes and obesity over the past twodecades. There are plausible pathways, such as chronic intermittent hypoxemia, and sleep fragmentation, through which untreated OSAmay contribute to the development of diabetes. While clinic studies report an association of untreated OSA and diabetes, a recentsystematic review found little evidence of a longitudinal association in population studies. The aim of this study was to examine therelationship between previously undiagnosed OSA and diabetes and pre-diabetes in community-dwelling men aged ≥ 40 years.Methods: The Men Androgen Inflammation Lifestyle Environment and Stress (MAILES) Study is a longitudinal population-based cohortassessing diabetes status over time. At cohort follow-up only (2011-12), n=837 underwent full in-home unattended polysomnography(PSG, Embletta X100). Clinic assessments at both stages included anthropometry, fasting plasma glucose (FPG), glycated haemoglobin(HbA1c) and self-reported chronic conditions (including diabetes) and risk factors. Diabetes was defined as self-reported physiciandiagnosis/ FPG ≥7.0 mmol/L/ HbA1c ≥ 6.5%/ diabetes medication use, and pre-diabetes was FPG 6.1-6.9 or HbA1c 5.7-6.4 (in those withoutdiabetes). Cross sectional analyses examined relationships between PSG indices and prevalent and incident diabetes/pre-diabetes cases.To determine longitudinal relationships in the absence of baseline PSG data, we identified probable longstanding OSA, defined as apneahypopnea index (AHI) ≥10/h at follow-up who also had OSA symptoms at baseline (2006-7) and stable weight over the follow-up period.Results: In cross-sectional analyses, significant age adjusted odds ratios (95% CI) for prevalent diabetes [2.42 (1.41-4.17)] or incidentdiabetes [2.15 (1.04-4.43)] associated with severe OSA (AHI≥30) did not persist after additional adjustment for obesity. Longitudinalanalyses showed that longstanding OSA was not significantly associated with incident diabetes [age-adjusted OR: 1.55 (0.74-3.23)] but wassignificantly associated with incident pre-diabetes [1.83 (1.00-3.35), p=0.049] after adjustment for age, obesity, smoking, physical activity,sleepiness and grip strength. There were no significant relationships of oxygen desaturation index (3%) ≥16 or time oxygensaturation<90% ≥4%, persisting after age and obesity adjustment.Conclusions: Longstanding OSA may be independently associated with the development of pre-diabetes or impaired glycaemic control inmiddle-age and older men