Article Text

Nocturnal desaturation in patients with stable heart failure
  1. Section of Cardiology,
  2. Providence Hospital,
  3. 16001 West Nine Mile Road,
  4. Southfield, MI 48075, USA

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Sir,—Staniforth et al correctly pointed out that arterial desaturation is common in patients with congestive heart failure, especially while sleeping.1The desaturation is commonly associated with apnoeic pauses, presumably as a result of depression of the respiratory centres (Cheyne-Stokes episodes). However, the seriousness of this depression requires emphasis.

Nocturnal desaturation in obstructive sleep apnoea has been associated with increased severity of hypertension, congestive heart failure, angina pectoris, atrial fibrillation, complete heart block, and possibly precipitation of myocardial infarction. Cerebral blood flow is reduced with the anoxia of obstructive sleep apnoea.2 Platelet aggregation and activation are increased two- and threefold.3 These changes are normalised with improved oxygenation. Vgontazs and colleagues4 and Entzian and colleagues5 identified severely increased inflammatory cytokines, and tumour necrosis factor α and interleukin 6 with anoxia.

Staniforth et al stated that “only 4% of their subjects were obstructive and no subject fulfilled the diagnostic criteria for obstructive sleep apnoea.” However only 41 (of 104) patients had polysomnography. One wonders about the other patients. Mechanical obstruction should be treated mechanically. The authors emphasise beautifully the serious desaturation that occurs in congestive heart failure, but the need for appropriate treatment gets short shrift.

It appears that all congestive heart failure patients should have nocturnal saturation determined and, if low, properly treated (with supplemental oxygen). Polysomnography is warranted in all who are anoxic with oxygen supplementation. If anoxia is present and oxygen does not correct it, and if an obstructive cause is discovered, continuous positive airway pressure or partial resection of the tongue (where it obstructs the palate) may be helpful.6 The surgery is relatively minor and well tolerated. Weight reduction for the obese is an ideal rarely realised; 20–30% of patients with obstructive sleep apnoea are not obese. Snoring without apnoeic periods may be the only sign of inadequate ventilation.

Most patients with congestive heart failure are hospitalised at some time. As part of their evaluation, documentation of nocturnal oxygen saturation would be worthwhile. If anoxemia is present, the effect of oxygen supplementation could be recorded and failure to oxygenate properly would suggest a mechanical obstruction. Polysomnography could be done to document the need for mechanical relief (with continuous positive airway pressure, weight reduction, or surgery). If oxygen treatment alone relieves the anoxia, then clearly the patient should be given oxygen at home. The effects of anoxemia are so pernicious that we should not be constrained to withhold oxygen if the levels are borderline.

One wonders if the brain damage manifested by the Cheyne-Stokes breathing pattern is not a result of chronic anoxia. I am not sure we know the lowest safe level of cerebral oxygenation. Hopefully, these techniques will improve the very bleak outlook for the congestive heart failure patient.


This letter was shown to the authors, who reply as follows:

The harmful effects of obstructive sleep apnoea are well documented and are summarised in the letter by Ritter. What is poorly recognised, especially in the UK, is the role of central sleep apnoea (Cheyne-Stokes respiration) as an independent marker of poor prognosis in heart failure.1-1 The mechanism behind this is unclear, but our results did not support a causal link between apnoea associated nocturnal desaturation and serious ventricular dysrhythmia. It may be related to activation of the sympathetic nervous system, and we have recently reported the beneficial effects of oxygen treatment on Cheyne-Stokes respiration and sympathetic activation in heart failure.1-2

We accept the criticism that only 41 of our 104 subjects underwent sleep studies. Polysomnography is the gold standard test for diagnosing sleep apnoea, but it is expensive and for that reason we used it only in a representative subgroup of our patients. These studies identified around 3100 apnoeas, 96% of which were central in origin, and no patient met the standard diagnostic criteria for obstructive sleep apnoea. The total number of events we examined was large, and consequently we do not feel that the quality of our data has suffered. Cheynes-Stokes respiration is common in patients with stable controlled heart failure; prevalences of 27−51% are reported in the literature.1-3 1-4 We documented a prevalence of 22%. Our study shows that obstructive sleep apnoea is an unlikely cause of nocturnal desaturation in heart failure patients who do not have a history of pulmonary disease. In this setting, overnight pulse oximetry is an effective screening tool for Cheyne-Stokes respiration. It is important to screen patients with heart failure for Cheyne-Stokes respiration on prognostic grounds. To avoid overdiagnosis, however, the best time to do this may be following discharge when they are clinically stable, rather than during a hospital admission as Ritter suggests. The major advantages of oximetry over polysomnography as a home screening tool are that it is cheap, easily portable, and widely available.

We did not discuss the possible treatment strategies for Cheyne-Stokes respiration in heart failure as the main thrust of our paper was to establish its prevalence, and to demonstrate the usefulness of oximetry as a screening tool. Oxygen, carbon dioxide, continuous positive airway pressure, aminophylline, and sedatives have all been advocated. The advantage of one treatment modality over another remains to be proved, as indeed does the existence of a survival benefit from treating Cheyne-Stokes respiration in heart failure.


  1. 1-1.
  2. 1-2.
  3. 1-3.
  4. 1-4.