Airway Pressure Release Ventilation in Acute Respiratory Failure Due to COVID-19: When One Door Closes.

To the Editor:

We read with great interest the article by Ibarra-Estrada et al (1) published in a recent issue of Critical Care Medicine on a randomized controlled trial using airway pressure release ventilation in acute respiratory failure due to coronavirus disease 2019 (COVID-19). While the authors present many important results, we would like to emphasize several statistical issues that should be taken under consideration when interpreting the findings of this clinical trial. Too often, clinicians and investigators are quick to dismiss interventions that, when subjected to the rigors of statistical testing, fail to meet the sacred threshold of p value of less than 0.05 without considering the many intricacies of study design that could have contributed to the trial’s demise. One of these important considerations is the inclusion of too few patients, leading to insufficient statistical power (2). The authors appropriately included a power calculation in the main body of the article, as well as the study protocol in the supplementary materials. However, aside from mentioning that the study was underpowered to detect a difference in mortality, the authors do not acknowledge that the recruitment presented was underpowered to detect the hypothesized difference in the primary outcome of ventilator-free days. Another important aspect in calculating statistical power and limiting the probability of an error of omission (false negative) is appropriately estimating the difference in outcome between the treatment and control arms of the trial. The authors assumed a 4-day difference in ventilator-free days as clinically relevant, which translates to a 33% change in outcome. This assumption raises multiple concerns. First, the minimal clinically important difference in ventilator-free days has not been established. Second, the Prospective, Randomized, Multi-Center Trial of 12 mL/kg vs. 6 mL/kg Tidal Volume Positive Pressure Ventilation for Treatment of Acute Lung Injury and Acute Respiratory Distress Syndrome (ARMA) trial, the hallmark trial in mechanical ventilation for acute respiratory failure, which included more patients by an order of magnitude, only showed a 25% change in outcome (3). In this case, it is more appropriate to describe the findings as inconclusive rather than negative. We acknowledge, however, that despite the concerns regarding statistical power, the interpretation that the intervention did not result in benefit may persist. The ECMO to Rescue Lung Injury in Severe ARDS (EOLIA) trial (4), which assessed the efficacy of venovenous extracorporeal membrane oxygenation, suffered a similar fate. However, subsequent post hoc Bayesian analysis (5) demonstrated the posterior probability for benefit that ranged up to 99%. Thus, given accumulating evidence for ventilator strategies that can maintain lung stability, perhaps a more nuanced approach to interpretation of the results of this clinical trial is warranted. Ultimately, future clinical trials may yet establish airway pressure release ventilation as lacking benefit in acute respiratory failure due to COVID-19. But based on the results of the present trial, the door should not be closed.