Association of reduced REM sleep with mortality in adults with coronary artery disease and obstructive sleep apnea in the RICCADSA cohort

TL;DR

Reduced REM sleep is linked to higher mortality in heart patients with sleep apnea, suggesting it could help identify high-risk individuals.

Contribution

This study shows that reduced REM sleep independently predicts mortality in CAD patients with OSA, beyond traditional risk factors.

Findings

Participants with reduced REM sleep had a 12.8% mortality rate versus 4.4% in others.

Reduced REM sleep predicted mortality (hazard ratio 2.39) after adjusting for age, sex, BMI, and CPAP use.

Adjusting for additional factors like TST and AHI did not change the association.

Abstract

Reduced rapid-eye movement (REM) sleep has been linked to increased mortality in the general population. We investigated whether diminished REM sleep is associated with higher mortality in adults with coronary artery disease (CAD) and obstructive sleep apnea (OSA).

This secondary analysis of the RICCADSA trial included 356 revascularized CAD patients with OSA (apnea–hypopnea index [AHI] ≥ 15 events/h) and total sleep time (TST) ≥ 240 min on baseline polysomnography. Reduced REM sleep was defined as the lowest quartile of REM percentage. Cox proportional hazards models assessed the association between reduced REM sleep and mortality over a median 4.7-year follow-up.

The lowest REM quartile corresponded to 8.7% of TST. Participants with reduced REM sleep (n = 86) were older (66.0 ± 8.1 vs. 63.0 ± 8.0 years; p = 0.035), had higher BMI (29.8 ± 4.6 vs. 28.7 ± 3.8 kg/m²; p = 0.010), shorter TST (369 ± 77 vs. 497 ± 69 min; p < 0.001), less slow-wave sleep (5.2 ± 7.0% vs. 8.1 ± 10.0%; p = 0.007), and higher AHI (54.4 ± 26.3 vs. 35.6 ± 20.1 events/h; p < 0.001) than those with REM ≥ 8.7% (n = 270). Mortality was 12.8% in the reduced REM group versus 4.4% in the higher REM group (p = 0.006). Reduced REM sleep independently predicted mortality (hazard ratio 2.39; 95% CI 1.03–5.56; p = 0.043) after adjustment for age, sex, BMI, and CPAP allocation. Further adjustment for TST, slow-wave sleep, baseline AHI, coronary bypass surgery, atrial fibrillation, and REM–AHI interaction did not alter the association.

Reduced REM sleep independently predicted higher all-cause mortality in revascularized CAD patients with OSA. Identifying diminished REM sleep may help identify particularly vulnerable patients.

The online version contains supplementary material available at 10.1007/s11325-026-03614-1.

Figures