Koo, Patrick

Background: The relationship between obstructive sleep apnea (OSA) and heart failure (HF) has been under-researched especially in postmenopausal women. We therefore evaluated relationship between OSA risk factors and HFpEF and HFrEF in post-menopausal women. Methods: We performed a prospective analysis of a subset of participants who had adjudicated heart failure outcomes (n=42,362) in the Women Health Initiative Observational, Clinical Trial, and Extension Studies (1998-Present). The cohort was followed over an average of 13.4 years. Inverse probability weighting was employed to account for potential selection bias. Cox proportional hazards regression was used to examine the association between OSA risk factors and time to first hospitalized HF. Type of heart failure was determined using the ejection fraction (EF) obtained from 2D echocardiography. EF of ≥45% was categorized as HFpEF, and EF of < 45% was categorized as HFrEF. Models were adjusted for age, race/ethnicity, education, income, marital status, systolic blood pressure, waist-to-hip ratio, diabetes, coronary heart disease, atrial fibrillation, use of hormone replacement therapy, use of sleep medications, modified Charlson comorbidity index, smoking, alcohol consumption, physical activity, and hysterectomy. We also created an OSA summary score (obesity, snoring, poor sleep quality, sleep fragmentation, daytime sleepiness, and hypertension) based on the Berlin questionnaire, which reliably predicts OSA, to examine its relationship with HF. Results: Of the 42,362 women, 1,054 (2.49%) had preserved EF, and 631 (1.49%) had reduced EF. Four of the 6 risk factors (obesity (HR=1.51, 95% CI 1.29-1.76), snoring (HR=1.23, 95% CI 1.04-1.45), sleep fragmentation (HR=1.15, 95% CI 1.01-1.31), and hypertension (HR=1.46, 95% CI 1.31-1.62)) were associated HFpEF after adjusting for confounders. Each additional OSA risk factor in an OSA summary score compared to no risk factors significantly increased the risk of HFpEF in a dose-response fashion (HR=1.36, 1.61, 2.01, 1.97, 2.02, and 2.74 for scores of 1-6, respectively; P trend <0.001) and not HFrEF (P trend =0.26). Only hypertension was associated with HFrEF (HR=1.39, 95% CI 1.22-1.60). Conclusion: Having more OSA risk factors increases the risk of HFpEF but not HFrEF in postmenopausal women. Early recognition and management of OSA risk factors may play an important role in reducing risk of HFpEF in this population.