Subjective sleepiness in heart failure patients with sleep-related breathing disorder.

BACKGROUND: Previous studies show that sleep-related breathing disorder (SRBD) is common in patients with heart failure (HF) and is associated with increased mortality. This study aimed to determine whether there was significant difference of subjective daytime sleepiness between HF patients with and without SRBD.

METHODS: We enrolled, prospectively, 195 consecutive HF patients with left ventricular ejection fractions (LVEF) < or = 45% and all subjects underwent polysomnography to measure the sleep structure between 2005 and 2008. Patients were then assigned to those with SRBD including obstructive and central sleep apnea (apnea-hypopnea index (AHI) > or = 5/hour of sleep) and those without SRBD (AHI < 5/hour) according to the sleep study. The subjective sleepiness was assessed with Epworth sleepiness scale (ESS).

RESULTS: Among 195 HF patients, the prevalence of obstructive sleep apnea (OSA) was 53% and of central sleep apnea (CSA) was 27%. There was no significant difference of ESS scores between patients without SRBD (NSA) and with SRBD (NSA vs OSA: 6.7 +/- 0.6 vs 7.6 +/- 0.4, P = 0.105 and NSA vs CSA: 6.7 +/- 0.6 vs 7.4 +/- 0.5, P = 0.235, respectively), indicating that SRBD patients had no more subjective daytime sleepiness. Compared with NSA, patients with SRBD had increased arousal index (ArI) (NSA vs OSA: 14.1 +/- 1.4 vs 26.3 +/- 1.5, P < 0.001 and NSA vs CSA: 14.1 +/- 1.4 vs 31.3 +/- 3.5, P < 0.001, respectively), more awake number after sleep onset (NSA vs OSA: 19.2 +/- 1.5 vs 26.2 +/- 1.4, P = 0.01 and NSA vs CSA: 19.2 +/- 1.5 vs 36.9 +/- 4.4, P < 0.001, respectively), and reduced proportion of slow-wave sleep (SWS) (NSA vs OSA: 13.8 +/- 1.7 vs 9.3 +/- 0.7, P = 0.024 and NSA vs CSA: 13.8 +/- 1.7 vs 8.9 +/- 0.9, P = 0.024, respectively).

CONCLUSIONS: OSA and CSA remain common in patients with HF on optimal contemporary therapy. Patients with both HF and SRBD have no significant subjective daytime sleepiness compared with patients without SRBD, despite of significantly increased awake number, arousal and decreased proportion of deep sleep stages. It is not a credible way and means to exclude SRBD in patients with HF according to the absence of subjective daytime sleepiness.