Abstract
BACKGROUND: This study evaluated the ability of diaphragmatic excursion (DE), measured 2 h after extubation, to predict the need for resumption of ventilatory support within 48 h in surgical critically ill patients. METHODS: This prospective observational study included adult surgical critically ill patients intubated for > 24 h and extubated after a successful spontaneous breathing trial. Sonographic measurement of the DE was performed 2 h after extubation. Patients were followed up for 48 h after extubation and were divided into reintubation group and successful weaning group. The primary outcome was DE's ability to predict the need for resumption of ventilatory support using the area under receiver characteristic curve (AUC) analysis. RESULTS: Data from 70 patients were analyzed and 25/70 (36%) patients needed reintubation. DE was lower in the reintubation group than the successful weaning group. The AUC (95% confidence interval) for the ability of DE to predict the need for resumption of ventilatory support was 0.98(0.92-1.00) and 0.97(0.89-1.00) for the right and left side, respectively. At cutoff values of 20.8 and 19.8 mm, the right and left DE had positive predictive values of 92% and 88% and negative predictive values of 96% and 93%, respectively. CONCLUSION: Among surgical critically ill patients undergoing weaning from invasive mechanical ventilation, DE obtained 2h after extubation is an accurate predictor for the need for resumption of ventilatory support. Diaphragmatic excursion < 20-21 mm could predict the need for resumption of ventilatory support with a positive predictive value of 88-92% and negative predictive value of 93-96%.