Abstract
Purpose :
In previous smaller studies, associations have been demonstrated between diabetic retinopathy (DR) and obstructive sleep apnea (OSA), but this has not been tested in national cohorts, and it is not known, if the level of DR acts as an independent predictive marker for OSA. This study aimed to assess the association between DR and OSA, and to evaluate if DR may serve as an independent risk factor of incident OSA.
Methods :
As part of the Ocular And Systemic complications In diabetic retinopathy Study (OASIS), we performed a cross-sectional and 5-year longitudinal registry-based cohort study. Data of exposure and outcome as well as systemic morbidity and use of medications were identified in national registers including the Danish Registry of Diabetic Retinopathy (DiaBase), the Danish National Patient Register, the Danish National Prescription Registry, and the Danish Civil Registration System. The index date was defined as the date of the first DR screening registered in DiaBase.
For cases, we included 153,238 patients with type 2 diabetes, who had attended diabetic eye screening and were registered in DiaBase. Each of these were matched by five control persons without diabetes of the same age and gender (n=746,148).
Present and level-specific DR were used as exposures, and main outcomes were crude, age- and sex-adjusted, and multivariable adjusted odds ratio (OR) for prevalent OSA as well as hazard ratio (HR) for 5-year incident OSA.
Results :
OSA was present in 5.8% and 2.0% of cases and controls. As compared to persons without diabetes, patients with type 2 diabetes were independently more likely to have prevalent OSA (OR 2.01, 95% CI 1.95-2.08) and to develop OSA within five years (HR 1.55, 95% CI 1.46-1.64). In comparison with cases without DR at the index date, those with DR were less likely to have prevalent (OR 0.57, 95% CI 0.52-0.62) and to develop incident OSA (HR 0.86, 95% CI 0.74-0.99).
Conclusions :
In a registry-based national cohort study of 153,238 patients with type 2 diabetes and 746,148 non-diabetes controls, cases had a higher risk of OSA, but the risk of prevalent and 5-year incident OSA were 43% and 14% lower in those with DR, which identifies DR as a marker of reduced risk of OSA.
This is a 2021 ARVO Annual Meeting abstract.