Short sleep ups diabetic retinopathy risk

Short sleep duration appears to be linked to moderate diabetic retinopathy (DR), according to a new Singapore study. Parameters related to obstructive sleep apnoea, as well as high insomnia risk, are also additionally associated with vision-threatening (VT) DR and diabetic macular oedema (DME).

“In our study, we showed that short sleep duration was associated with higher odds of moderate DR. We also found that moderate OSA and [apnoea-hypopnoea index (AHI)] were associated with higher odds of moderate DR and VTDR,” said researchers.

The present cross-sectional analysis included 92 diabetics (mean age, 57.6±8.3 years; 32.6 percent female). Majority of the participants (68 percent; n=63) had less than moderate DR, 32 percent (n=29) had moderate DR, and 18.5 percent (n=17) had VTDR. The mean sleep time was 332.3±68.5 minutes. [Diabetes Res Clin Pract 2019;doi:10.1016/j.diabres.2019.107967]

Multiple logistic regression analysis revealed that moderate DR was significantly correlated with a higher AHI (odds ratio [OR], 1.04, 95 percent confidence interval [CI], 1.00–1.07) and shorter sleep durations (OR, 3.22, 95 percent CI, 1.18–8.79). Adjustments for high AHI and short sleep did not weaken these relationships.

Moderate OSA was also significantly associated with moderate DR (OR, 2.73, 95 percent CI, 1.02–7.31), though this effect was attenuated after adjusting for age and sex (p=0.085).

On the other hand, moderate OSA (OR, 4.73, 95 percent CI, 1.46–14.31) and higher AHI (OR, 1.06, 95 percent CI, 1.02–1.10) were both significantly correlated with VTDR, as was a lower minimum oxygen saturation (OR, 0.89, 0.83–0.96). Further adjustment of the models for body mass index, diabetes type and scores on the Epworth Sleepiness Scale had no significant effect on the primary findings.

Time spent under 90 percent, 80 percent, 70 percent and 60 percent oxygen saturation likewise had no effect on moderate or VTDR.

Moreover, patients who had a high risk of insomnia were at greater odds of DME (OR, 4.01, 95 percent CI, 1.09–14.73). Any, moderate and VTDR had no such relationships with insomnia severity or Epworth Sleepiness Scale scores.

“We were able to objectively confirm the association of short sleep duration with increased prevalence of moderate DR,” the researchers said.

In the present study, patients with type 1 or 2 diabetes were enrolled from the Singapore National Eye Centre; those with cognitive or hearing impairments were ineligible. DR was assessed via retinal images while an in-home, overnight polysomnography monitor was used to evaluate sleep.

Short sleep duration may contribute to the development of DR through worsening glycaemic control, compounding endothelial dysfunction and promoting inflammation, all of which have been implicated in DR, the researchers explained. [Sleep 2018;doi:10.1093/sleep/zsy172; Sleep 2017;doi:10.1093/sleep/zsx156; Atherosclerosis 2017;265:41-46]

The sequence of events may go the other way, too, they added. “Alternatively, DR might cause disturbances in the circadian rhythm resulting in shorter sleep durations. The circadian rhythm in humans is in part regulated by a small population of melanopsin-expressing retinal ganglion cells that have been shown to be lost in more severe DR.” [Invest Ophthalmol Vis Sci 2017;58:2187-2192]

While the mechanisms of effect have yet to be completely elucidated, the present findings contribute to the better understanding of the relationship between sleep and DR.