B-lines by lung ultrasound as a predictor of re-intubation in mechanically ventilated patients with heart failure.

Abstract

There have been few studies on predictors of weaning failure from MV in patients with heart failure (HF). We sought to investigate the predictive value of B-lines measured by lung ultrasound (LUS) on the risk of weaning failure from mechanical ventilation (MV) and in-hospital outcomes. This was a single-center, prospective observational study that included HF patients who were on invasive MV. LUS was performed immediate before ventilator weaning. A positive LUS exam was defined as the observation of two or more regions that had three or more count of B-lines located bilaterally on the thorax. The primary outcome was early MV weaning failure, defined as re-intubation within 72 h. A total of 146 consecutive patients (mean age 70 years; 65.8% male) were enrolled. The total count of B-lines was a median of 10 and correlated with NT-pro-BNP level (r2 = 0.132, p < 0.001). Early weaning failure was significantly higher in the positive LUS group (9 out of 64, 14.1%) than the negative LUS group (2 out of 82, 2.4%) (p = 0.011). The rate of total re-intubation during the hospital stay (p = 0.004), duration of intensive care unit stay (p = 0.004), and hospital stay (p = 0.010) were greater in the positive LUS group. The negative predictive value (NPV) of positive LUS was 97.6% for the primary outcome. B-lines measured by LUS can predict the risk of weaning failure. Considering the high NPV of positive LUS, it may help guide the decision of weaning in patients on invasive MV due to acute decompensated HF.