Left ventricular morphology and systolic function in sleep-disordered breathing: the Sleep Heart Health Study.

Abstract

BACKGROUND
Whether sleep-disordered breathing (SDB) is a risk factor for left ventricular (LV) hypertrophy and dysfunction is controversial. We assessed the relation of SDB to LV morphology and systolic function in a community-based sample of middle-aged and older adults.
METHODS AND RESULTS
The present study was a cross-sectional observational study of 2058 Sleep Heart Health Study participants (mean age 65+/-12 years; 58% women; 44% ethnic minorities) who had technically adequate echocardiograms. A polysomnographically derived apnea-hypopnea index (AHI) and hypoxemia index (percent of sleep time with oxyhemoglobin saturation < 90%) were used to quantify SDB severity. LV mass index was significantly associated with both AHI and hypoxemia index after adjustment for age, sex, ethnicity, study site, body mass index, current and prior smoking, alcohol consumption, systolic blood pressure, antihypertensive medication use, diabetes mellitus, and prevalent myocardial infarction. Adjusted LV mass index was 41.3 (SD 9.90) g/m(2.7) in participants with AHI < 5 (n=957) and 44.1 (SD 9.90) g/m(2.7) in participants with AHI > or = 30 (n=84) events per hour. Compared with participants with AHI < 5, those with AHI > or = 30 had an adjusted odds ratio of 1.78 (95% confidence interval 1.14 to 2.79) for LV hypertrophy. A higher AHI and higher hypoxemia index were also associated with larger LV diastolic dimension and lower LV ejection fraction, with a trend toward lower LV fractional shortening. LV wall thickness was significantly associated with the hypoxemia index but not with AHI. Left atrial diameter was not associated with either SDB measure.
CONCLUSIONS
In a community-based cohort, SDB is associated with echocardiographic evidence of increased LV mass and reduced LV systolic function.