Heart Rate Response in OSA: A Clue to Reveal Cardiovascular Benefit from CPAP?

To the Editor:

Obstructive sleep apnea (OSA) is associated with a cluster of serious adverse outcomes, including cardiovascular (CV) morbidity and mortality. However, the protective effect of the first-line therapy of OSA—continuous positive airway pressure (CPAP) against adverse CV outcomes remains controversial (1–3). It is increasingly recognized that the prognostic significance of CPAP on CV events may not be uniform across the entire spectrum of OSA populations. Understanding which phenotypes may experience benefit from CPAP will be essential to provide constructive guidance in clinical practice.

In a recent issue of the Journal, Azarbarzin and colleagues (pp. 766–773) conducted a post hoc analysis of the Randomized Intervention with Continuous Positive Airway Pressure in Coronary Artery Disease and Obstructive Sleep Apnea (RICCADSA) trial (4). They first revealed an optimistic CPAP effect on adverse CV outcomes in selected nonsleepy OSA patients with coronary artery disease (CAD) who exhibited exaggerated pulse rate response to respiratory events (ΔHR) (4). Nevertheless, critical questions remained to be addressed.

Emerging evidence has recognized a low prognostic value of the traditional measure of OSA–apnea–hypopnea index for CV events (3). Instead, a novel metric, “hypoxic burden” proposed by Azarbazin and colleagues was more consistently correlated with CV disease–related mortality in the general population (5). The rationale for the current study (4) was also based on the authors’ prior work in which a subgroup of patients with OSA presenting a higher ΔHR was at an increased risk of CV events, particularly in those with substantial hypoxic burden (6). However, in patients with CAD, the latest study did not engage the impact of hypoxic burden per se and its combination with ΔHR on the long-term CV risk, and did not elucidate whether the CPAP effect would be moderated by hypoxic burden (4).

Moreover, the authors used the pulse rate derived from a pulse oximetry sensor to estimate the heart rate (HR). Pulse oximetry may be chosen over electrocardiogram due to its convenient accessibility and widespread application. However, a simple measure of pulse rate or HR may not fully depict the complex process of autonomous regulatory mechanisms. A high ΔHR may represent a pronounced vagally induced bradycardia during an event and sympathetic response to hypoxemia, and/or a combination of both. Notwithstanding, the actual contribution of sympathetic or parasympathetic activity to ΔHR was not systemically examined. Heart rate variability (HRV) derived from electrocardiogram is generally considered as a reliable and noninvasive measure of autonomic modulation response and adaptation to multiple stimuli in healthy or pathogenic conditions. With more analytical approaches and techniques developing, HRV indices specific to respiratory events could provide additional information on impaired CV alteration related to subclinical CV outcomes.

Another critical issue that requires further clarification lies in the analysis of ΔHR when an arrhythmic heartbeat occurs within the same timescale of respiratory signals. The occurrence rate of concomitant arrhythmias may be high in patients with acute myocardial infarction (accounting for 49.6% of the study population) (4). Although patients with chronic atrial fibrillation were excluded in the sensitivity analysis, the measurement of ΔHR might remain unreliable when other overt arrhythmias coincide with apneas or hypopneas. Furthermore, the baseline levels of diurnal and nocturnal HR were not delineated in the study, so one may suspect that ΔHR appears to merely reflect higher night-to-day variability of HR in general rather than HR responses to obstructive events.

We sincerely recognize that the work of Azarbarzin and colleagues is a valuable contribution to demonstrating the CV benefit of CPAP in the nonsleepy CAD patients exhibiting a higher ΔHR in OSA. However, further studies should provide greater insight into the HRV metrics in response to respiratory events to allow better CV risk stratification, and to clarify whether CPAP therapy would benefit selected patients with both greater ΔHR and hypoxic burden.