In reply:
We appreciated the comments from Mathews et al regarding our article in Respiratory Care.1 We would like to add some comments and clarify some issues about their opinions.
The study was randomized (as described in the methods section), and because of its characteristics, it could not be blinded. Therefore, the small differences between the groups were a result of randomization. However, the number of subjects included was enough to answer the main question and achieve our conclusions.
In the cited article, table 11 shows that the distribution of the subjects was broad, including neurological (trauma or non-trauma), medical, and surgical patients. The proposed classification by Mathews et al for “neurologic and non-neurologic” diseases sounds artificial, since it does not consider important clinical situations (such as COPD) and the fact that the neurological population can include young patients with traumatic brain injury and elderly patients with stroke.
According to the study protocol (see methods section), all subjects with COPD were placed on noninvasive ventilation immediately after extubation. This approach is reasonably well described in the literature.2–4 The 15% extubation rate is in line with the international literature.5 We should not compare oranges with apples: In Esteban et al6 the failure rate was 25% of the total number of subjects, which is comparable with that found in our study.
The statistical analysis shows that all methods had comparable abilities to predict extubation success or failure, with values comparable with those in the literature.7,8 The fact that the incidence of tracheostomy was larger in a group does not mean greater efficiency in predicting extubation failure or success: The decision to perform tracheotomy includes several issues, like consciousness level, underlying medical conditions, and etiology of respiratory failure.9
Therefore, Mathews et al share our conclusions, that proportional assist ventilation plus is a safe method and is efficient to perform a spontaneous breathing trial, comparable with other existing methods (T-tube and pressure support ventilation), and a clinical option for clinicians and respiratory practitioners in the ICU.
Footnotes
The authors have disclosed no conflicts of interest.
- Copyright © 2016 by Daedalus Enterprises