In the first place, we wish to express our appreciation for the comments and interest shown by Gil et al.1 towards our work.2 Also, we wish to thank the editor for giving us the opportunity to answer back. Although it is true that the model of ventilator was never included in our study, we do know that they were all specific ventilators for non-invasive ventilation. Also, although the programming of the different parameters involved varies from one model to the next, like they say, the actual ventilation support comes from the support pressure that results from programming. Regarding the EPAP (expiratory positive airway pressure) and CPAP (continuous positive airway pressure) values, their mechanical meaning is different, and it is true that this variable could have been shown separately on the table, which is why we wish to share the results now. CPAP values (cmsH2O): mean (SD) was 7.36 (2.20) in the NIV-HEU group (non-invasive ventilation at the hospital emergency unit)<12h and 6.71 (1.98) in the NIV-HEU group 12≥h; P value=.526; EPAP values (cmsH2O): mean (SD) was 6.67 (1.45) in the NIV-HEU group<12h and 6.36 (1.12) in the NIV-HEU group 12≥h; P value=.223.

The variables associated with the patients’ clinical situation and arterial blood gas test were collected. They were grouped by type of acute respiratory failure, specific arterial blood gas test data, arterial oxygen tension, carbon dioxide, and pH. No significant differences were seen between the 2 groups under comparison. Poor secretion control or interphase rejection were not specifically included.

One of the main aspects of the VNICat registry (Non-invasive ventilation in Catalonia)3 was that only 17% of the cases recruited (N=27) were referred to hospitalization units with non-invasive ventilation still as part of the ventilation support therapy. This circumstance seems routine at the emergency services like other studies conducted in our setting reveal.4 Although this situation can be taken as a failure, it is not a technical failure, which is what we assessed in our study, but as an organization failure of the flows of participant centers. Therefore, what we say in our conclusion is of paramount importance, that the flows of these patients need to become standardized based on the resources available to make sure that a proper transition takes place. Actually, let’s go one step further. This should not be based on the resources available at all, but, as the go-to health professionals, we need to demand the existence and availability of these resources.5