Background

  • Non-invasive positive pressure ventilation (NIPPV) is commonly used to assist with respiration in exacerbations of chronic obstructive pulmonary disease (COPD) by offering support in both oxygenation and ventilation. This treatment modality can cause significant discomfort for some patients, often resulting in treatment failure. 
  • High-velocity nasal insufflation (HVNI) is an alternative, more comfortable mode of respiratory support that is non-inferior to NIPPV with respect to the need for intubation in undifferentiated respiratory failure.1 Its use has not been studied in COPD exacerbations specifically, which often present with significant hypercapnia and/or acidemia and thus may benefit more from the additional ventilation support that NIPPV offers.

Clinical Question

  • In patients with moderate COPD exacerbations, is HVNI non-inferior to NIPPV with respect to patients’ dyspnea severity after four hours of treatment? 

Population

  • 68 patients across 7 different emergency departments
  • Known or strongly suspected history of COPD 
  • Blood gas analysis with partial pressure of carbon dioxide (pCO2) of 60mmHg or higher, and serum pH between 7.00 and 7.35

Intervention

  • Randomization to 4 hours of HVNI (initial flow rate of 30L/min, FiO2 of 0.50) or NIPPV (initial settings of 10-12 cmH2O end-inspiratory pressure, 5-6 cmH2O end-expiratory pressure, with goal tidal volumes of 6-8 mL/kg of ideal body weight)
  • Both interventions were titrated according to standard practice to optimize their effects.

Outcomes

  • The primary outcome of interest was the degree of patient-perceived dyspnea at 4 hours, measured by the modified Borg scale, a validated measure scaled from 0 (no dyspnea) to 10 (unbearable dyspnea). 
  • Secondary outcomes of interest were vital signs, oxygen saturation, pCO2, pH, patient discomfort level, and need for endotracheal intubation, all measured at 4 hours.

Design

  • Prospective, randomized noninferiority trial

Excluded

  • Patients with altered mental status, immediate need for endotracheal intubation, cardiac or respiratory arrest, or determination by the treating clinician that the patient was too clinically unstable to participate.
  • Patients with a known intolerance to either NIPPV or HVNI. 
  • Patients with an alternate cause of dyspnea (pneumonia, congestive heart failure, neurologic dysfunction, or toxic exposure). 
  • Patients under age 18 or patients who were pregnant.

Primary Results

  • With respect to the primary outcome, HVNI was noninferior to NIPPV.
  • There was no difference between the two groups with respect to changes in venous pCO2 and pH – a decrease in pCO2 and an increase in pH was observed in both groups. 
  • Patients who received HVNI reported greater levels of comfort than patients who received NIPPV (Mean comfort score difference of 23.12, 95% CI 7.41-38.82. Scores ranged from 0-100).
  • No patients were intubated, but patients who received HVNI were more likely to be admitted to the ICU (38% of HVNI group compared to 27% of the NIPPV group). This difference was statistically significant.

Strengths

  • Overall, this was a well-conducted study. It had well-defined, objective inclusion and exclusion criteria. They also met their prespecified enrollment goal. 
  • Adherence to treatment within the two groups was also very high – greater than 90% of the patients in each group received the allocated intervention, and reasons for non-adherence were reasonable and well-documented.
  • Patient-centered outcomes. 

Limitations

  • The primary outcome of this study was a subjective measure. 
  • The study did not follow patients beyond the 4-hour mark. 
  • The patients included in this study were not as sick as the inclusion criteria suggest. For example, while the study’s inclusion criteria allowed for a pH of 7.00 to 7.35, only 10% of the patients enrolled had a pH of less than 7.20. 
    • This significantly limits the generalizability of these findings to patients with more significant acidemia, who are also the patients who (at least presumably) would benefit from the additional ventilatory support NIPPV offers. 
  • The study was not powered to detect inferiority in terms of treatment failure (i.e. respiratory arrest/intubation). However, this event would be rare in this patient population of less sick patients and would have required a much larger sample.

Author's Conclusions

  • HVNI was non-inferior to NIPPV in patients with moderate COPD exacerbations for dyspnea at 4 hours.

Our Conclusions

  • This study adds to the growing evidence2,3 that HVNI offers more than simply oxygenation, as equivalent reductions in acidemia and hypercapnia were observed in both groups. Animal model studies have hypothesized that the high-velocity insufflation washes excess carbon dioxide out of the anatomic dead space, thus helping to decrease hypercapnia.4 
  • Another study last summer looked at a similar population of patients, except in the ICU setting instead of the ED setting, and found significantly higher intubation rates in patients randomized to HVNI.5 This study found no such differences, but it was not designed or powered to detect them.  
  • While it wasn’t a primary outcome, the 10% absolute increase in ICU admissions in the HVNI group is hard to overlook. It makes it difficult to argue that these two treatments are equivalent for this population.

Potential Impact To Current Practice

  • This study suggests that HVNI offers more ventilation support than we may have thought initially, and it is a reasonable alternative for patients struggling to tolerate the discomfort of NIPPV. However, the increased rates of ICU admission in the HVNI group and the absence of adequate power to detect differences in intubation rates make it difficult for this study to unseat NIPPV as the preferred means of respiratory support for patients with acute COPD exacerbations.

Read More

References:

  1. Doshi PB, Whittle JS, Bublewicz M, et al. High velocity nasal insufflation in the treatment of acute respiratory failure: a randomized clinical trial. Annals Emerg Med. 2018;72:73-83. 
  2. Doshi PB, Whittle JS, Dungan G, et al. The ventilatory effect of high velocity nasal insufflation compared to non-­ invasive positive-pressure ventilation in the treatment of hypercapneic respiratory failure: a subgroup analysis. Heart Lung. 2020;49:610-615.
  3. Lee MK, Choi J, Park B, et al. High flow nasal cannulae oxygen therapy in acute-­ moderate hypercapnic respiratory failure. Clin Respir J. 2018;12:2046-2056.
  4. Moller W, Celik G, Feng S, et al. Nasal high flow clears anatomical dead space in upper airway models. J Appl Physiol. 2015;118:1525-1532.
  5. Tan D, Wang B, Cao P, et al. High flow nasal cannula oxygen therapy versus non-­ invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease with acute-­ moderate hypercapnic respiratory failure: a randomized controlled non-­ inferiority trial. Crit Care. 2024;28:250-259.