Sleep quality measures like sleep efficiency (OR 0.69; 95% CI 0.68-0.70) showed stronger associations with atrial fibrillation than OSA severity measures like AHI4% (OR 1.20; 95% CI 1.19-1.22).
Observational (n=526,523)
Are sleep quality measures more strongly associated with atrial fibrillation risk than obstructive sleep apnea severity or hypoxemia in patients undergoing home sleep apnea testing?
Sleep quality and fragmentation measures derived from home sleep apnea tests are more strongly associated with atrial fibrillation risk than traditional OSA severity or hypoxemia metrics.
Estimación del efecto: OR 0.69 (95% CI 0.68-0.70)
Abstract Introduction Obstructive sleep apnea (OSA) and atrial fibrillation (AFib) have a well established relationship, sharing many common risk factors such as sex, age, obesity, diabetes. The two conditions are highly co-prevalent: 10.5% of patients with OSA have AFib and 45.9% of patients with AFib have OSA. This work examines the relationship between markers of OSA severity and sleep quality with AFib burden in a large, representative clinical sleep cohort. Methods N=526,523 Type IV home sleep apnea tests (HSAT) with 4-12 hours of photoplethysmography (PPG) recording were analyzed. A deep learning model was applied to classify AFib epoch-by-epoch then compute a subject-level AFib Burden. Respiratory instability measures (apnea-hypopnea index (AHI), oxygen-desaturation index (ODI), hypoxic-burden index (HBI)), hypoxemia measures (SpO2 time-under 95% (T95), 90% (T90), 85% (T85)), sleep quality and disturbance measures (sleep efficiency (SE), total sleep time (TST), awakenings after sleep onset (AASO), sleep fragmentation index (SFI; stage transitions count), heart rate variability (HRV) and demographics were extracted. Univariate logistic regression quantified association between each predictor and AFib; odds ratios (ORs) and 95% confidence intervals are reported, with ORs for continuous predictors representing the change per standard deviation. Results OSA severity indices were significantly associated with AFib. The strongest associations (ORs) for respiratory instability were AHI4% (1.20, 1.19–1.22), AHI3% (1.19, 1.17–1.21), ODI4% (1.16, 1.15–1.18), HBI4% (1.14, 1.13–1.16), HBI3% (1.14, 1.13–1.15), and ODI3% (1.14, 1.12–1.15). Hypoxemia measures showed weaker associations: T85% (1.07, 1.05–1.08) and T80% (1.03, 1.01–1.04). Sleep quality measures were more strongly associated with AFib than OSA severity. Sleep efficiency (0.69, 0.68–0.70) and total sleep time (0.77, 0.75–0.78) were protective, while awakenings after sleep onset (1.26, 1.25–1.27) and sleep fragmentation (1.17, 1.15–1.18) indices were positively associated. HRV during sleep (3.52, 3.47–3.56) and during wake (3.57, 3.52–3.62) was strongly associated. Female sex was protective (0.59, 0.57–0.61) whereas age increased risk (2.44, 2.37–2.52). Conclusion OSA severity and sleep fragmentation measures showed significant and moderately strong associations with AFib, highlighting OSA’s role as an important risk factor. Unexpectedly, sleep quality measures demonstrated stronger associations than measures of respiratory instability and hypoxemia. These findings have direct clinical relevance and implications when discussing AFib risk with patients undergoing sleep studies. Support (if any)
Wodnicki et al. (Fri,) conducted a observational in Obstructive sleep apnea and atrial fibrillation (n=526,523). Home sleep apnea test factors (sleep quality and OSA severity) was evaluated on Atrial fibrillation burden (OR 0.69, 95% CI 0.68-0.70). Sleep quality measures like sleep efficiency (OR 0.69; 95% CI 0.68-0.70) showed stronger associations with atrial fibrillation than OSA severity measures like AHI4% (OR 1.20; 95% CI 1.19-1.22).