From the Authors:

We thank Dr. Jha for his reading of our study (1) and for his valuable and insightful comments. We wish to address some of the comments made by Dr. Jha.

First, Dr. Jha points out that our exclusion criteria did not mention risk factors for failure of liberation from mechanical ventilation. However, we would like to stress that patients with overt congestive heart failure at the time of liberation from mechanical ventilation could not be enrolled until clinicians estimated that another reason might explain weaning failure. In our study, we purposely did not select patients on the basis of the presence of diaphragm function, for two reasons: first, it has been established that diaphragm dysfunction is present in a majority of patients at the time of liberation from mechanical ventilation (2), and second, liberation from ventilation depends on the balance of respiratory muscle load and capacity, and any improvement in diaphragm function is likely to facilitate safe extubation despite the lack of severe diaphragm dysfunction. Dr. Jha mentions that “differential distribution of lung collapse, atelectasis, lung fibrosis, diastolic dysfunction and pulmonary hypertension in the control and treatment arm could have affected the outcomes.” In addition, Dr. Jha notes that in our study, there were several risk factors for diastolic dysfunction. We completely agree with this comment, and we believe that the randomization process was the best way to allocate equal proportions of patients with lung and cardiac diseases to the treatment and control groups. We also agree with Dr. Jha that lung ultrasound–based aeration score and echocardiography are interesting tools in this context, as reported in a recent study from our group (3). However, in our multicenter study, it was not deemed feasible to ask investigators to perform echocardiography and lung ultrasound. Regarding the assessment of regional variation in aeration by electrical impedance tomography, only a few centers in the world possess this technology.

Second, Dr. Jha underlines that “patients with overt congestive heart failure were to be excluded, however, the authors reported congestive heart failure in 9% of patients and valvular heart disease in 19% of patients in the treatment arm.” We would like to clarify that only patients with overt congestive heart failure at the time of eligibility screening were not enrolled, but if clinicians could deal with fluid overload, patients were reassessed and eventually included despite the presence of chronic heart disease. We do not see any reason that would have required the exclusion of patients with chronic heart disease from our study. Indeed, we believe that the opposite would have been unethical.

Third, we appreciate the Dr. Jha’s physiological description of maximum inspiratory pressure. We share his interpretation regarding the recruitment of extradiaphragmatic inspiratory muscles in the generation of maximal inspiratory pressure. As reported in several publications from our group (4–6), Dr. Jha is absolutely correct in saying that twitch transdiaphragmatic pressure is the reference method to specifically assess diaphragm function. Nonetheless, Dr. Jha should fairly recognize that measuring diaphragm function according to twitch pressure is simply not possible at the scale of an international multicenter trial. Following Dr. Jha’s reasoning, the fact that there was a significant increase in maximal inspiratory pressure in the treatment group and not in the control group works in favor of the treatment.

Last, Dr. Jha rightly points out the heterogeneity of our population, in particular the fact that half of the patients were tracheostomized. As suggested by Dr. Jha, we provide here a sensitivity analysis pertaining to the tracheostomized patients. Fifty-two patients were tracheotomized at study entry. Among them, weaning was successful in 79.8% in the treatment group and 72.4% in the control group. Forty-six patients had endotracheal tubes. Among them, weaning was successful in 82.1% in the treatment group and 76.0% in the control group. Further studies will be required to confirm these findings.