Central Sleep Apnoea: Another Important Comorbidity in Heart Failure?

This editorial refers to ‘Advanced heart failure and nocturnal hypoxaemia due to central sleep apnoea are associated with increased serum erythropoietin’ by Calvin et al., on page 354. It is well known that comorbidities play an important role in the pathophysiology, management, and prognosis of chronic heart failure (HF).1,2 In recent years, an increasing amount of data has become available regarding the prominent place of sleep disordered breathing (SDB) in HF.3–6 If an apnoea hypopnoea index (AHI) of >15 is used for the diagnosis of SDB, a consistent prevalence of around 50% is found in various groups of HF patients. Sleep disordered breathing in HF may be classified broadly as either obstructive sleep apnoea (OSA) or central sleep apnoea (CSA). While patients with OSA have a collapse of their upper airway, CSA is more a consequence of HF itself. During periods of shortness of breath, patients with HF tend to hyperventilate and as a result, they will fall below the so-called apnoea threshold and will stop breathing.5 Even in HF patients with a preserved (left ventricular) ejection fraction, it has recently been shown that 30% of patients had a CSA and that its prevalence was related to impaired diastolic function.7 Since a recent analysis demonstrated that the presence of SDB in patients with chronic HF was associated with more than 2-fold increased mortality risk, more information about potential mechanisms in this field is urgently needed.8 Calvin et al.9 present interesting data on CSA in HF patients. Thirty-three patients with a LVEF < 35% and 18 healthy subjects were included in the study. Of the HF patients, 14 had CSA (42%), 4 had OSA and were excluded, while 15 had neither CSA nor OSA and served as a control group. The HF patients with CSA had on average 62% higher erythropoietin (EPO) levels (21.8 mIU/mL) than the healthy controls (13.4 mIU/mL), while levels for HF patients without CSA fell between these two values (16.5 mIU/mL). In addition, the authors found that the duration of nocturnal hypoxia assessed by the percentage of sleep time spent with an arterial oxygen saturation <90% (T90%) was strongly associated with an increased EPO level. Remarkably, there was only a weak and non-significant relationship between the AHI and level of EPO. The authors suggest that the magnitude of hypoxaemia is a more important trigger for increasing EPO than the level of AHI. If this is true, one could question whether the severity of CSA in HF should be based solely on the AHI or that the level of oxygen desaturation should be included as well. Larger studies are needed to investigate this issue further and to see if these results can be replicated. Another important issue in this respect is to understand the significance of these markedly increased EPO levels in HF patients with CSA. It is now well established that EPO concentrations are generally increased in patients with HF, and that these levels carry important independent prognostic value.10 The increase of EPO in HF is probably the result of peripheral under perfusion and ischaemia, particularly in the kidneys, and is considered to be an (endogenous) protective mechanism.11 Significant increases in EPO after myocardial infarction have also been reported.11,12 The highly elevated EPO levels observed in the present study by Calvin et al.9 suggest that the nocturnal hypoxia was so severe, that at the time of measurement (early in the morning, after the test) it was still able to induce an increase in EPO levels. Although most patients will not show this level of oxygen desaturation during the daytime, it would be interesting in further studies to assess EPO levels throughout the daytime as well and to see how this is related to cardiac function. Many therapies have been investigated and are currently available for the treatment of HF patients with CSA. Oxygen,13 acetazolamide,14 continuous positive airway pressure (CPAP),15 bi-level positive airway pressure, and—most recently—adaptive servo ventilation have all shown potentially beneficial effects in these patients.16 Lüthje et al.17 recently demonstrated in this journal that cardiac resynchronization was effective in reducing the number of CSA episodes, which occurred in parallel with an improvement in cardiac function. The largest intervention study so far has been the Canadian CANPAP trial, in which CPAP was compared with placebo in more than 200 HF patients with CSA.15 Although the study did not show an effect on the primary outcome (survival), it had a positive influence on AHI, LVEF, and minimum saturation. In addition, if corrected for an effective reduction in AHI during CPAP (AHI < 5 per h), a survival benefit could be detected.18 In line with OSA patients in whom nocturnal CPAP was effective in decreasing EPO levels,19 we could perform a similar investigation in CSA, and it would be interesting to see whether CPAP (or any other effective treatment) can indeed reduce EPO levels in HF patients with CSA. In such an intervention study, clinical parameters such as cardiac function and patient-reported outcomes such as quality of life should be assessed as well. In summary, the paper by Calvin is interesting since it further emphasizes the importance of sleep disorders, in particular CSA, in patients with HF. It may well be that management of an increasing number of HF patients will be guided increasingly by focusing on their comorbidities,20 and the present data provide new information in this respect. Further studies will be needed to evaluate the impact of nocturnal hypoxic episodes on cardiac and systemic parameters, and in addition larger trials are clearly needed. Conflict of interest: none declared.