Can Lung Ultrasound Predict Mechanical Ventilation Weaning Outcomes in Critically Ill Adults?

Abstract

This study aimed to assess the ability of lung ultrasound (LUS) to predict mechanical ventilation (MV) weaning outcomes in critically ill adults. Methods: A prospective observational study that analyzed data collected from 50 adult patients, mechanically ventilated for more than 48 hours and eligible for their first spontaneous breathing trial. A LUS was performed prior to and at the end of a 1-hour SBT. To quantify lung aeration, a LUS score was calculated. Patients were divided into two groups according to their response to weaning trials with group A showing successful weaning while group B showing failed weaning. All included patients were followed up to Intensive Care Unit discharge. Results: Weaning failure was observed in 36% of patients. LUS score showed a significant difference between both groups (P < 0.001). Pre-and post-spontaneous breathing trial (post-SBT), LUS scores were significantly higher in the failed weaning group (14.44 ±2.52 and 18.83 ±3.18 respectively) than in the successful weaning group (11.25 ±3.05 and 12.53 ±3.41 respectively). A ROC analysis for the ability of post-SBT LUS score to predict weaning outcomes revealed a significant AUC of 0.911 with a 95% confidence interval ranging from 0.830-0.992 (P < 0.0001). The best cut-off was 14.5, at which sensitivity and specificity were 88.8% and 68.7%, respectively. Conclusions: Our data suggest that LUS can be used as a predictor of MV weaning outcomes in critically ill adults. A post-SBT LUS score cut-off value of 14.5 has a sensitivity and a specificity of 88.8% and 68.7%, respectively.