Physiological predictors of respiratory and cough assistance needs after extubation

Nicolas Terzi, Frédéric Lofaso, Romain Masson, Pascal Beuret, Hervé Normand, Edith Dumanowski, Line Falaize, Bertrand Sauneuf, Cédric Daubin, Jennifer Brunet, Djillali Annane, Jean-Jacques Parienti, David Orlikowski

Identifying patients at high risk of post-extubation acute respiratory failure requiring respiratory or mechanical cough assistance remains challenging. Here, our primary aim was to evaluate the accuracy of easily collected parameters obtained before or just after extubation in predicting the risk of post-extubation acute respiratory failure requiring, at best, noninvasive mechanical ventilation (NIV) and/or mechanical cough assistance and, at worst, reintubation after extubation. We conducted a multicenter prospective, open-label, observational study from April 2012 through April 2015. Patients who passed a weaning test after at least 72 h of endotracheal mechanical ventilation (MV) were included. Just before extubation, spirometry and maximal pressures were measured by a technician. The results were not disclosed to the bedside physicians. Patients were followed until discharge or death. Among 3458 patients admitted to the ICU, 730 received endotracheal MV for longer than 72 h and were then extubated; among these, 130 were included. At inclusion, the 130 patients had mean ICU stay and endotracheal MV durations both equal to 11 ± 4.2 days. After extubation, 36 patients required curative NIV, 7 both curative NIV and mechanical cough assistance, and 8 only mechanical cough assistance; 6 patients, all of whom first received NIV, required reintubation within 48 h. The group that required NIV after extubation had a significantly higher proportion of patients with chronic respiratory disease (P = 0.015), longer endotracheal MV duration at inclusion, and lower Medical Research Council (MRC) score (P = 0.02, P = 0.01, and P = 0.004, respectively). By multivariate analysis, forced vital capacity (FVC) and peak cough expiratory flow (PCEF) were independently associated with (NIV) and/or mechanical cough assistance and/or reintubation after extubation. Areas under the ROC curves for pre-extubation PCEF and FVC were 0.71 and 0.76, respectively. In conclusion, FVC measured before extubation correlates closely with FVC after extubation and may serve as an objective predictor of post-extubation respiratory failure requiring NIV and/or mechanical cough assistance and/or reintubation in heterogeneous populations of medical ICU patients. ClinicalTrials.gov as #NCT01564745.