Patients with COPD may develop sustained hypercapnia, often defined as an awake arterial PCO2 of >45 mm Hg. Other synonymous terms include alveolar hypoventilation or chronic hypercapnic respiratory failure, noting that the specific terminology used may reflect local practice, an assessment of patient severity, or specific insurance requirements. Regardless, available data suggest that hypercapnic COPD patients are at high risk for adverse health outcomes (Yang H, et al. BMJ Open. 2015;5[12]:e008909). Moreover, there appears to have been a growing interest in this population driven by a focus on reducing COPD hospitalizations, increasing recognition of sleep disordered breathing, and progress in potential therapeutic strategies.
There are a number of factors that might drive COPD patients to develop hypercapnia. Lower airway obstruction, expiratory flow limitation and air trapping cause mechanical load on breathing, as well as a trade-off between time spent in inspiration vs prolonged expiration. The function of the diaphragm is impacted by hyperinflation leading to mal-positioning, as well as possibly by local and/or systemic myopathy. The net result is often decreased overall minute ventilation. In terms of gas exchange, increased dead space and ventilation-perfusion mismatching leads to reduced efficiency of ventilation towards CO2 removal. Breathing changes during sleep play an important role, as evidenced by worsened hypercapnia during sleep that can drive chronic CO2 retention (O’Donoghue FJ, et al. Eur Respir J. 2003;21[6]:977). The pathogenesis includes reduced central respiratory drive, increased upper airway resistance and/or obstructive hypopneas and apneas, and respiratory muscle atonia, particularly during REM sleep. The extent to which each of these factors contributes to hypercapnia varies across individual patients, in accordance with the known substantial heterogeneity of COPD. Regardless of underlying traits, patients with COPD who develop hypercapnia have sufficiently severe perturbations to disrupt the normally tight control over CO2 homeostasis.
Nocturnal home noninvasive ventilation (NIV) has been examined as a potential therapeutic strategy for patients with hypercapnic COPD. While older studies have not shown consistent benefits, more recent evidence suggests that NIV can reduce hospitalizations, improve quality of life, and potentially reduce mortality among those with hypercapnic COPD. Accordingly, the American Thoracic Society recently released a clinical practice guideline regarding the use of NIV in patients with chronic stable hypercapnic COPD (Macrea M, et al. Am J Respir Crit Care Med. 2020;202[4]:e74-e87). Recommendations from the guideline included:
1) The use of nocturnal NIV for patients with chronic stable hypercapnic COPD
2) Screening for OSA before initiation of long-term NIV
3) Not using in-hospital initiation of long-term NIV after an episode of acute or chronic hypercapnic respiratory failure, favoring instead reassessment for NIV at 2–4 weeks after resolution
4) Not using an in-laboratory overnight PSG to initially titrate NIV
5) Targeting normalization of PaCO2.
Although it now seems clear that efforts should be made to use NIV in COPD to decrease chronic hypercapnia, there are a number of important questions that remain, particularly surrounding the topic of concurrent OSA, titration, and devices:
• What is the appropriate approach towards patients with suspected concurrent OSA? Most studies of NIV have excluded patients with OSA, or otherwise at higher risk of OSA. Nonetheless, such patients may be common, both based on continued high prevalence of obesity, as well as the potential role that upper airway obstructive events may play towards elevations in CO2 (Resta O., et al. Sleep Breath. 2002;6[1]:11-8). COPD epidemiological studies indicate obesity as a risk factor for several poor outcomes, including severe COPD exacerbation (Lambert AA, et al. Chest. 2017;151[1]:68-77), while studies of COPD and OSA suggest that the presence of hypercapnia defines a high-risk group Jaoude P., Lung. 2014;192:215). Recognizing the potential importance of OSA in this group, ATS guidelines recommend that a general questionnaire-based screening be performed. If screening is positive, the implication would be to perform diagnostic polysomnography to confirm the diagnosis of OSA. However, this may be a challenge for chronically ill patients, and likely would result in delays in NIV initiation. Of note, emerging evidence suggests that home sleep apnea testing (HSAT) might have reasonable accuracy in this group, which may facilitate formal diagnosis. Other concerns in this area include the lack of questionnaire validation in COPD patients.
• Should patients with OSA be managed differently than those without OSA? A diagnosis of OSA might impact several subsequent management decisions related to appropriate NIV therapy and titration. Patients with OSA have increased upper airway collapsibility, which might necessitate higher EPAP support than the minimal EPAP used in NIV trials with non-OSA patients (often fixed at 4 cm water). Potential strategies for optimizing EPAP include use of an NIV device with auto-titrating EPAP, titration in the sleep laboratory (discussed below), or outpatient titration based on clinical parameters and subsequent device download follow-up. On the other hand, one might consider all patients to be at risk for upper airway obstruction and need for additional EPAP titration, which would obviate the need for OSA diagnostic testing.
• What is the role of the sleep laboratory towards successful titration? The inpatient hospital setting has been the traditional site to initiate home NIV in some institutions but is highly resource intensive and increasingly impractical in many health systems. On the other hand, advances in home remote device monitoring now provide the clinician with the ability to examine daily usage, estimated leak, tidal volumes, respiratory rate, and other parameters – often reported as recently as the prior night. In addition, setting changes can be made via these remote monitoring tools (for nonventilator devices), allowing titration to be performed over time on outpatients. Several studies support the effectiveness of this approach over hospital titration in neuromuscular disease and now in COPD (Duiverman ML, et al. Thorax. 2020;75[3]:244-52). Similarly, data suggest that titration under polysomnographic guidance might not be necessary (Patout M, Arbane G, Cuvelier A, Muir JF, Hart N, Murphy PB. Polysomnography versus limited respiratory monitoring and nurse-led titration to optimize non-invasive ventilation set-up: a pilot randomised clinical trial. Thorax. 2019;74:83-86).
Limitations towards the sleep lab as the site of initial titration include waiting time, cost and insurance coverage, and the need to accommodate issues such as impaired mobility or reliance on a caretaker. In addition, titration goals must be clearly outlined in protocols and via staff training specific to NIV. The sleep laboratory may be most appropriately utilized in the minority of patients in whom outpatient titration is unsuccessful. Relatively common issues that might be best addressed in the lab setting include excessive mask leaks, residual apneas and hypopneas, failure to control CO2, or other sleep complaints. In general, studies should probably be focused primarily on titrating EPAP to alleviate upper airway obstructive events. The goals in terms of IPAP titration (or ventilation titration, in the case of “VAPS” modes) are less clear, and overly aggressive increases may complicate the picture with excessive leaks or airway obstruction due to glottic closure. Attempting to accomplish “too much” often leads to a study with limited utility. In contrast, simply performing the study in the patient’s home settings can provide useful diagnostic information regarding the problem one is trying to solve.
• When and where should one initiate NIV following a severe COPD exacerbation? In contrast to the ATS guidelines, the European Respiratory Society guidelines suggest that patients recovering from severe COPD exacerbations be initiated on NIV during that hospitalization, noting that this is a group at high risk for early rehospitalization and mortality (Ergan B, et al. Eur Respir J. 2019;54[3]:1901003). ATS guidelines had the concern of unnecessary start of NIV in those who might normalize their CO2 after recovery, and the possibility of prolonging hospitalizations for titration. For the clinician, the decision will probably be individualized based on risk and available resources. For patients with frequent ICU admissions and/or difficulty with close outpatient follow-up, earlier NIV initiation is certainly a reasonable approach, but adherence and effectiveness remains a concern and, thus, more data are needed.
• Which patients should receive a bedside respiratory assist device (RAD, i.e., BIPAP machine) vs. a noninvasive ventilator? Two classes of devices can be used for home NIV. While both can provide similar positive pressure ventilation, ventilators are designed as life support with alarms and batteries, and may have modes not otherwise available (e.g., auto-titrating EPAP). On the other hand, RAD devices are more convenient for patients and less expensive, but difficult qualification requirements (particularly for devices capable of Bilevel ST or VAPS) have likely resulted in their underutilization. CHEST is spearheading an effort to reconsider Medicare coverage determinations (current rules are from 1998), which will hopefully better align device qualification requirements with emerging evidence regarding patient needs and preferences.
Home non-invasive ventilation can improve outcomes in these high-risk patients with hypercapnic COPD, and the new clinical practice guidelines are an important step in outlining appropriate management. Further progress is needed to delineate an individualized approach based on underlying patient pathophysiology, COPD manifestations/phenotypes, and systems-based practice considerations.
Dr. Orr is Assistant Professor, Division of Pulmonary, Critical Care, and Sleep Medicine, UC San Diego.