Reply to Tiruvoipati et al.: VENT-AVOID Trial: Avoiding Acute Hypercapnic Respiratory Failure!

From the Authors:

We appreciate the comments by Tiruvoipati and colleagues on the recently published VENT-AVOID (Extracorporeal CO2 Removal with the Hemolung RAS for Mechanical Ventilation Avoidance during Acute Exacerbation of COPD) trial investigating extracorporeal carbon dioxide removal (ECCO₂R) to avoid invasive mechanical ventilation (IMV) in patients with exacerbation of chronic obstructive pulmonary disease (ECOPD) (1). This trial was designed as a pragmatic trial to identify and enroll patients with acute hypercapnic respiratory failure in two different strata: 1) patients initiated on noninvasive ventilation (NIV) at risk of failure and 2) patients already on IMV unable to wean. Tiruvoipati and colleagues failed to recognize that the combined results include patients from both strata. It is important to note that the study is composed mostly of the IMV stratum patients, and these patients did not have evidence of major hypercapnic acidosis because they had been stabilized with IMV on a ventilator, with corresponding changes in their pH and Pa_CO2 (1).

On the basis of the Global Initiative for Chronic Obstructive Lung Disease 2023 report (2), ECOPD requiring hospitalization includes a wide spectrum of severity, which includes both non–life-threatening and life-threatening respiratory failure (2, 3). In our NIV stratum, most of the patients did not have life-threatening acute respiratory failure, but they did have acute hypercapnic respiratory failure with severe CO₂ retention, and, on the basis of existing literature, a physiologic response to ECCO₂R is expected in these patients, and thus they were included in our study (4, 5). The cutoff values suggested by Tiruvoipati and colleagues only include life-threatening situations that would necessitate immediate endotracheal intubation in most clinical situations, and, as a result, such patients would be inappropriate for inclusion in a randomized controlled trial (RCT) aiming to study the avoidance of IMV (2, 3). We also disagree with the assertion made by Tiruvoipati and colleagues that the patients did not have worsening respiratory acidosis at the time of randomization. The median time to randomization for the NIV stratum was 22 hours, whereas in the patients included in the IMV stratum, NIV had already failed. Moreover, it was challenging to implement our protocol in patients with COPD in the NIV stratum because of the dynamic nature of the disease, with many candidates improving rapidly while we were obtaining consent and implementing the protocol. A recent consensus document suggested no decrease in Pa_CO2 or respiratory rate while on NIV as a key initiation criterion for ECCO₂R therapy for patients with ECOPD (6). Most participants also agreed, however, that a pH range from <7.25 to 7.30 was appropriate for ECCO₂R initiation. Notably, Tiruvoipati and colleagues suggested that baseline Pa_CO2 and respiratory rate should be used as key decision factors, but these were favored as triggers by less than 50% of the participants in this consensus document (6).

We agree with Tiruvoipati and colleagues that evidence gathered from observational studies helps us identify appropriate populations to study. But the inherent biases of such studies render them only capable of generating hypotheses. RCTs such as the VENT-AVOID trial are needed to truly test the effectiveness and safety of interventions such as ECCO₂R. The variability in initiation criteria for ECCO₂R used in many of the observational studies probably explains their disparate results and justifies the performance of RCTs with standardized inclusion criteria and outcomes. The VENT-AVOID trial highlights that few patients with ECOPD on NIV are likely to benefit from ECCO₂R therapy, although results could be different in a real-life setting where delays related to study protocol could be avoided. Moving forward, we favor focusing studies on the IMV stratum where there was a trend toward more ventilator-free days during the first 5 days.