Sleep Abnormalities Among Community Participants At Risk For Heart Failure

Yogesh N.V. Reddy,Susan DeFranco,John Park,Amir Lerman,Virend Somers,Barry Borlaug

DOI: https://doi.org/10.1016/j.cardfail.2023.10.044

IF: 6.592

2024-01-01

Journal of Cardiac Failure

Abstract:Introduction Sleep abnormalities are associated with metabolic health and underlying heart failure (HF). HF with preserved Ejection Fraction (HFpEF) is the most common cause of HF but is severely underdiagnosed given the limitations in current diagnostic criteria. Therefore, we sought to determine self-reported and directly recorded sleep abnormalities using custom fitted beds capable of physiological monitoring (SleepNumber Co) among community-dwelling participants who were at risk of HFpEF. Methods We first developed a regression equation to estimate the probability of HFpEF using age and body mass index in a derivation cohort where the diagnosis of HFpEF was definitively confirmed or refuted conclusively using gold standard diagnosis with exercise right heart catheterization demonstrating elevated exercise pulmonary capillary wedge pressure. This was validated in an independent cohort with similar gold standard diagnosis. We then estimated the probability of HFpEF in community participants at risk for HFpEF and compared directly recorded sleep data averaged over a month of observation in 91954 Sleep Number bed users, as well as self reported Sleep survey data in a subset (n=5861) of these participants. Results The HFpEF prediction algorithm based on age and BMI had good discrimination in the derivation (AUC 0.806 [95% CI 0.761-0.850], p<0.0001) and validation cohort (AUC 0.770 [95% CI 0.730-0.810], p<0.0001). Comparing individuals with a high probability (>75%) of HFpEF (HFpEFhighprob) with those with a low probability (<25%) of HFpEF (HFpEFlowprob), those with HFpEFhighprob were more likely to report taking regular naps during the day (16.4% vs 11.2%, p<0.0001) and never exercise (24.4 vs 17.7%, p<0.0001). The HFpEFhighprob group were also more likely to have self-reported sleep disturbances 7 days a week (36% vs 19%, p<0.0001), nocturia requiring awakening at night (69% vs 45%, p<0.0001), and sleep problems present for more than a year (76% vs 64%, p<0.0001). These sleep disturbances in HFpEFhighprob occurred despite a lower proportion reporting difficulty falling asleep (31% vs 40%, p<0.0001), similar objectively measured average hours slept across groups (median 8.4 [IQR 7.0-10.3] vs 8.6 hours [7.2-10.3], p=0.11) and less awakening episodes per night (median 3 [IQR 3-5] vs 4 [3-5], p=0.0003) suggesting impaired quality of sleep rather than duration. The HFpEFhighprob also demonstrated greater fluctuations in peak and trough sleep motion, respiration and HR indices, and greater HR variability throughout sleeping sessions (p<0.0001 for all) compared to HFpEFlowprob. Conclusion Self reported and objective sleep abnormalities in physiological indices and quality of sleep are commonly observed in community participants at high risk for HFpEF. Given the impaired quality of life in HFpEF, further study of the true prevalence and impact of sleep disorders in HFpEF are urgently needed.

cardiac & cardiovascular systems