BACKGROUND: Sleep disordered breathing (SDB) is common in patients with heart failure with reduced ejection fraction (HFrEF). An increased apnea-hypopnea index (AHI) is associated with poor outcomes. We examined whether an analysis of nocturnal desaturations (NDs) can improve the risk stratification.
METHODS: Three-hundred seventy-six consecutive patients with stable chronic HFrEF and LVEF ≤ 45% were prospectively screened using polygraphy. Sleep apnea (SA) was defined as an AHI ≥ 15. The mean age was 59 ± 13 years, the mean LVEF was 30 ± 6%, and the median AHI was 18 [IQR: 9.33). The composite end-point of death, heart transplantation or LV assistance occurred in 98 patients (26%) within 3 years. Minimal oxygen saturation (MOS) during sleep, the number of desaturations <90%/h and the time spent with oxygen saturation <90% were significantly associated with adverse events (adjusted HR 1.25 [1.03-1.52], 1.25 [1.03-1.53], and 1.28 [1.04-1.59]), whereas the AHI was not (1.10 [0.86-1.39]). The best MOS cut-off value for poor outcomes was ≤ 88%. The patients with an MOS ≤ 88% had a significantly higher event rate (31.9%) than those with an MOS >88% (15.6%; p<0.01). The risk assessment using an MOS of ≤ 88% in addition to established prognostic markers yielded a net reclassification index (NRI) of nearly 6% and was particularly useful in the subgroup of patients with events (NRI: 8.4%).
CONCLUSIONS: In HFrEF patients, ND ≤ 88% appears to be predictive of adverse events, independent of the presence of SA. This suggests that the risk assessment in HFrEF should also include ND in top of AHI.