Chronic Obstructive Pulmonary Disease and Breathing during Sleep. A Strain in the Neck.

Although sleep is often considered a time of rest and recuperation, in many ways, sleep presents a challenge for patients with chronic lung disease (1). Although data are relatively sparse, disrupted sleep in individuals with chronic obstructive pulmonary disease (COPD) is a both a common complaint and an objective finding (2, 3). In addition, normal physiological changes during sleep such as increased upper airway resistance, decreased lung volumes, ventilation-perfusion mismatching, and decreased respiratory drive may result in substantially worsened breathing during sleep among patients with COPD (4). Importantly, it is increasingly appreciated that these issues with sleep and breathing may have far-reaching effects: poor sleep has been linked to an increased risk for exacerbations (5), and sleep-disordered breathing (both sleep apnea and hypoventilation) has been convincingly shown to lead to rehospitalizations and mortality, although mechanisms remain unclear (6–9).

In this innovative study published in this issue of the Journal, Redolfi and colleagues (pp. 414–422) examined whether respiratory accessory muscle use quantified as neck inspiratory muscle (NIM) activity was present during sleep among patients with a recent severe COPD exacerbation, whether its presence was associated with signs of sleep disruption, and whether it could predict recurrent exacerbation requiring hospitalization (10). They found that many patients had evidence of NIM during sleep, and there were no observed differences in demographics, lung function, or hypoventilation/hypercapnia between those with and without NIM, although awake oxygenation was worse in those with NIM activity. Second, those with NIM had more disrupted sleep, quantified by EEG changes (i.e., persistent high-frequency activity during sleep). Third, those with NIM activity were more likely to be rehospitalized. The effect was most pronounced in those with “permanent” NIM activity, meaning they had NIM all throughout sleep, including during REM sleep. Such unusual physiology has also been noted in those with neuromuscular disease (11), highlighting that in the setting of disease, respiratory muscle activity patterns may differ substantially from normal physiology.

What are the potential explanations for these findings? Perhaps NIM activity is indeed causative and does lead to sleep disruption as hypothesized, which leads to a risk for severe exacerbation via mechanisms yet unknown. However, another possibility is that the finding of poor sleep in those with NIM activity and a higher risk for readmission is correlative. With respect to NIM activity, this activation of accessory muscles presumably serves to compensate for inadequate ventilation relative to respiratory drive, which may reflect diaphragm dysfunction as well as poor respiratory mechanics and high intrinsic drive to breathe. Given similar levels of lung function and gas exchange between groups, another explanation is that those with NIM activity have unrecognized severe respiratory muscle (i.e., diaphragm) dysfunction, and that NIM activity is indeed compensatory, but simply insufficient to stabilize breathing long-term. Finally, NIM activity may be a marker of individuals who are “sicker” in some way that we are not capturing with our usual measures of COPD severity. Indeed, those with NIM had lower PaO during wakefulness.

Importantly, those with NIM activity might represent a group that would benefit from nocturnal noninvasive ventilation (NIV). Conventional criteria for NIV has required the presence of daytime hypercapnia (i.e., development of chronic respiratory failure). Indeed, reduction in PaCO has been linked to usefulness of NIV (6, 7), whereas those without daytime hypercapnia have not had clear benefit (12). However, breathing strategies vary across patients with COPD, classically conceptualized as the blue bloater versus pink puffer. Might patients with NIM activity be those who benefit from nocturnal NIV to offload overtaxed respiratory muscles, potentially reducing their risk for adverse outcomes? Even in those with hypercapnia, optimal targets for NIV titration are not known, and perhaps elimination of NIM activity might represent a rational goal.

What are the barriers to implementing NIM measurement? A noninvasive measure to identify a previously unrecognized high-risk group certainly has appeal. However, obtaining polysomnography after each COPD exacerbation may not be feasible in many centers. Similarly, there are a lack of standards for measurement of NIM in the clinical setting. Technician training would be needed, along with software packages capable of quantifying NIM activity and removing artifacts such as electrocardiogram signals. Limited channel polygraphy recordings incorporating an EMG channel might be sufficient. In addition, Redolfi’s study did find differences in NIM activity during wake, although outcomes were not assessed on the basis of awake activity. Finally, until high-quality trials of NIV (and perhaps other therapeutic strategies) are available, it is not exactly clear whether we can prevent poor outcomes in this nonhypercapnic group. Nonetheless, this study provides additional evidence that sleep and breathing at the same time are difficult in some patients with advanced COPD and should encourage additional investigation toward diagnostic and therapeutic strategies.