By Michael H. Crawford, MD, Editor

SYNOPSIS: A comprehensive cardiorespiratory study of stable systolic heart failure patients showed Cheyne-Stokes breathing in the awake, upright position is related to hypercapnia and is independently associated with a higher risk of cardiac death.

SOURCE: Giannoni A, Gentile F, Sciarrone P, et al. Upright Cheyne-Stokes respiration in patients with heart failure. J Am Coll Cardiol 2020;75:2934-2946.

Supine or sleeping Cheyne-Stokes respiration (CSR) in heart failure subjects is known to be associated with increased severity of disease and a poor prognosis. Less is known about daytime, upright CSR.

Giannoni et al enrolled stable systolic heart failure patients on optimal guideline-based therapy between 2007 and 2018. Those with other comorbidities or on drugs that influence respiratory control were excluded. Each patient underwent two days of intensive baseline testing for cardiopulmonary and neurohormonal characterization. Apnea was defined as cessation of airflow lasting ≥ 10 seconds. Hypopnea was defined as a reduction in airflow ≥ 50% lasting ≥ 10 seconds in association with reduced oxygen saturation (≥ 4% during a 24-hour ambulatory polygraphic cardiorespiratory recording). The authors used the rebreathing technique to assess chemoreflex sensitivity to hypoxia and hypercapnia. Also, researchers performed short-term respiratory response to head up tilt testing. Patients were classified as normal breathing, supine CSR, or upright CSR. Among the 574 patients enrolled, 297 breathed normally, 195 were classified as supine CSR, and 82 were classified as supine and upright CSR.

Compared to those who breathed normally, patients with upright CSR recorded significantly lower ventricular ejection fraction, larger left atria, higher systolic pulmonary artery pressures, higher norepinephrine levels, and higher NT-pro BNP levels. They also recorded lower peak VO2 levels, more oscillatory exercise breathing, a higher ratio of ventilation to CO2 output, more atrial fibrillation, and more moderate to severe mitral regurgitation. In addition, they scored higher on the apnea-hypopnea index and central apnea index vs. normal breathing patients and those classified as supine CSR only. Using a multivariate logistic analysis, the only independent predictor of upright CSR was hypercapnic ventilatory response (hazard ratio [HR], 4.01; 95% confidence interval [CI], 1.54-10.46; P = 0.004). After an eight-year follow-up, upright CSR was independently predictive of a higher risk of cardiac death (HR, 2.39; 95% CI, 1.08-5.29; P = 0.032), as was moderate to severe mitral regurgitation (HR, 3.82). The authors concluded upright CSR is related to an increased sensitivity to hypercapnia, is associated with more severe disease, and is associated with a greater risk of cardiac death.


CSR has long been recognized as a useful bedside observation to confirm the severity of heart failure. The authors of older studies have noted an inverse association between the total cycle length, not just the apnea time, and cardiac output. In more recent studies, investigators have noted CSR is frequent in heart failure patients during polysomnographic studies, categorizing it as a rhythmic subtype of central sleep apnea. It occurs in about half of stable heart failure patients and is associated with a higher risk of ventricular tachycardia and death.

Giannoni et al explored the observation that CSR can occur while awake and even during exercise. They used a simple polygraph to detect CSR in upright patients. They started with 15 minutes of rest. If CSR was observed, they moved on to 15 minutes of head-up tilt. This maneuver uncovered upright CSR in 14% of their patients. These patients featured a worse adrenergic and hemodynamic profile, with the highest levels of norepinephrine, NT-pro BNP, atrial fibrillation, moderate to severe mitral valve regurgitation, and eight-year cardiac mortality. CSR was considered sleep related because the cortical influences on respiration would diminish during sleep, allowing the chemical control of breathing to dominate. Also, fluid redistribution to the lungs while sleeping could exacerbate the problem. Obstructive sleep apnea almost always is sleep related, but CSR also can occur during awake daylight time. The mechanism of upright CSR is not completely clear. In the Giannoni et al study, the only predictor of upright CSR was chemosensitivity to increased CO2, but there may be other contributors. Other studies have shown a relationship between left atrial pressure (LAP) and supine CSR. Perhaps upright positioning reduces LAP and prevents upright CSR in many stable heart failure patients. Upright CSR patients in the Giannoni et al study had more severe mitral regurgitation. Perhaps this maintained high atrial pressures upright. The major limitation of this study is that CSR was assessed only once at intake. There could have been changes in CSR status over the eight-year follow-up. What are the clinical implications of this study? Upright CSR could be detected easily in an office practice and would be of prognostic value. Should the detection of CSR drive more aggressive therapy? Studies of correcting CSR associated with sleep apnea in heart failure patients was not effective at improving outcomes. At this point, there is no specific therapy occasioned by the detection of upright CSR.