Abstract
The majority of patients with heart failure have sleep-disordered breathing (SDB)-with central (rather than obstructive) sleep apnoea becoming the predominant form in those with more severe disease. Cyclical apnoeas and hypopnoeas are associated with sleep disturbance, hypoxaemia, haemodynamic changes, and sympathetic activation. Such patients have a worse prognosis than those without SDB. Mask-based therapies of positive airway pressure targeted at SDB can improve measures of sleep quality and partially normalise the sleep and respiratory physiology, but recent randomised trials of cardiovascular outcomes in central sleep apnoea have been neutral or suggested the possibility of harm, likely from increased sudden death. Further randomised outcome studies (with cardiovascular mortality and hospitalisation endpoints) are required to determine whether mask-based treatment for SDB is appropriate for patients with chronic systolic heart failure and obstructive sleep apnoea, for those with heart failure with preserved ejection fraction, and for those with decompensated heart failure. New therapies for sleep apnoea-such as implantable phrenic nerve stimulators-also require robust assessment. No longer can the surrogate endpoints of improvement in respiratory and sleep metrics be taken as adequate therapeutic outcome measures in patients with heart failure and sleep apnoea.