Thoracic fluid content by electrical cardiometry versus diaphragmatic excursion by ultrasound for the prediction of weaning success in patients with lung congestion.

BACKGROUND: Predicting the weaning outcomes is critical, since premature or delayed extubation is associated with an increased risk of mortality. This study aimed to compare two physiological indices, thoracic fluid content (TFC) and diaphragmatic excursion (DE), for predicting weaning success in mechanically ventilated patients. METHODS: This observational cohort study involved 100 mechanically ventilated patients with congested lungs who were eligible for weaning. Patients' TFC and DE were measured using electrical cardiometry and ultrasonography, respectively, before starting the spontaneous breathing trial. Following extubation, patients were grouped into successful and failed-weaning groups, with failure defined as reintubation or a need for non-invasive ventilation within 48 hours. Respiratory and cardiovascular variables were compared. The receiver operating characteristic (ROC) curve was used to assess the ability of TFC and DE to predict weaning success. RESULTS: Successful weaning occurred in 73 patients (73%) and failed weaning occurred in 27 patients (27%). The two groups' baseline characteristics were comparable; however, TFC and DE were significantly different between the failed- and successful-weaning groups (P<0.001). The area under the ROC curve (AUC) exhibited moderate predictive abilities of both the TFC and DE in predicting weaning success (AUC, 0.805, cutoff <40 kΩ-1 and AUC, 0.774, cutoff >1.45 cm). In the cardiac patient subgroup, TFC exhibited high predictive ability (AUC, 0.861), but DE did not achieve comparable results (AUC, 0.750). CONCLUSIONS: Both TFC and DE are significant predictors for successful weaning from mechanical ventilators. In particular, a TFC of <40 kΩ-1 demonstrated an excellent ability to predict weaning success in patients with low ejection fraction.