Comparing Highs and Flows in Patients With COPD With Chronic Hypercapnic Respiratory Failure

Out-patient or home management of patients with severe COPD following a hospitalization or severe exacerbation remains challenging. These patients are at risk for respiratory failure or are in a state of incipient respiratory failure that can be difficult to manage. Aside from supplemental oxygen, additional adjuncts to therapy have been the subject of decades of investigation^1-3 with nocturnal ventilatory support and variable success. Marked improvement in outcomes were realized with a shift in treatment strategies to noninvasive ventilation (NIV) providing more intensive support with a reduction in P_aCO2.^4-6 This experience has formed the basis of mostly conditional recommendations on NIV for chronic hypercapnic patients with COPD.⁷

A stronger endorsement was tempered by questions about concomitant sleep disorders, mask issues, training, adherence to therapy, and the process of initiation of support with some uncertainty about the P_aCO2 targets for reduction, whether reduction be normalization or just below a threshold such as 48 mm Hg. More recent experience suggests home initiation is possible, but small numbers of actual participants remain a concern.^8,9 The difficulties of application of NIV cannot be overstated as benefit is only achieved with consistent use. Successful NIV trials required close monitoring, adjustment, and adherence to achieve P_aCO2 reduction targets and benefit.

Heated, humidified high-flow nasal cannula (HFNC) oxygen does not provide ventilation but has other benefits that may provide an effective option for home management. Some of the mechanisms of action seen with NIV include reduction in respiratory muscle load, diaphragmatic activity, breathing pattern, and correction of hypercapnia. Similar effects have been seen with HFNC, with reduction in work of breathing, improved ventilatory pattern, low level PEEP providing a balance to intrinsic PEEP, and reduction in deadspace and decrease in P_aCO2.^10,11 In groups of patients with advanced lung disease including COPD, clinical benefits have been seen with reduction in exacerbations, or improved functional capacity, and quality-of-life measures. There may be better adherence to HFNC since it is easier to initiate and use, and is probably more comfortable than NIV, increasing the likelihood of continued use and benefit.

Whereas both modalities have been available for several years, there is a paucity of direct comparison between the two, and any clinical trial would require years for completion. Clinical management questions may be addressed with a network meta-analysis, which extends evaluation beyond the traditional head-to-head comparisons. A network meta-analysis permits comparison between > 2 interventions and between interventions not directly evaluated in a clinical trial. Pitre and colleagues¹² conducted and report on their network meta-analysis of NIV, HFNC, and standard care that included long-term oxygen therapy. This involved any type of NIV and HFNC, with different devices, modes of positive pressure, levels of support, and mask interfaces in comparison to HFNC at different flows. They evaluated a total of 24 randomized control trials (1,824 subjects), but only one directly compared NIV to HFNC (102 subjects).¹³

Further insights can be gleaned from the summary of their reviewed trials. These were male-dominated trials of older participants (> 65 y), with severe COPD, baseline FEV₁ generally < 30%, and baseline hypercapnia (P_aCO2 ≥ 50 mm Hg), representing subjects likely to experience higher numbers of exacerbations and hospitalizations. Whereas there is some uncertainty about their effect on mortality (NIV relative risk 0.82 [95% CI 0.66–1.00] and HFNC relative risk 1.20 [95% CI 0.63–2.28]), both modalities reduced exacerbations compared to standard care (NIV relative risk 0.71 [95% CI 0.58–0.87] and HFNC relative risk 0.77 [95% CI 0.68–0.88]). Neither definitively reduced the risk of hospitalization, but quality-of-life scores were better for HFNC compared to standard care (HFNC mean difference –7.01 [95% CI −12.27 to −1.77] and NIV mean difference −3.81 [95% CI −10.00 to 3.23]). In the only head-to-head comparison of both, neither demonstrated benefit over the other.

These are different treatment strategies producing comparable results. NIV provides ventilation which in turn may produce larger reductions in P_aCO2 and have a greater impact on mortality. Both provide enough benefit to reduce exacerbations but not hospitalizations. HFNC is easier to use and likely more comfortable than NIV, which may explain the better quality-of-life measures.^14-16 A question may arise about blinding of treatment and the potential for bias. Both are mechanical interventions that are not easily subject to any sort of blinding given differences in device and interfaces. There are also unanswered questions with underlying sleep disorders and whether benefit may be related to simultaneous treatment of sleep-disordered breathing.

However, both modalities provide advantages over standard care, which includes long-term oxygen therapy. This would suggest that either would be an acceptable adjunct for the very severe hypercapnic patients with COPD. There are undoubtedly patient subgroups who may fare better with one modality over the other. One might envision that the more severely hypercapnic patient may be better served with NIV where higher inspiratory pressures might be needed, whereas HFNC and its humidification may have more of a role in those with airway clearance issues. Then again, one size will never fit all, and adjustments may be needed on a case-by-case basis.

However, the differences between modalities seem modest, and a clinical trial to determine the most optimal therapy may require hundreds of participants and years to complete. In addition, it may not be necessary to establish superiority of one over the other as both are viable options of support that can be used in end-stage COPD.¹⁷ Whereas a large-scale clinical trial comparing NIV and HFNC would be ideal, current comparisons have focused on acute hypercapnic respiratory failure, with no comparison trials in chronic hypercapnic respiratory failure on the horizon. Given this reality, the results from Pitre and colleagues may be the best evidence going forward for some time.¹² Any support is better than none. If the goal is reducing mortality, the question remains unanswered. Both NIV and HFNC effectively reduce exacerbations, but to improve quality of life, it may be best to go with the flow.