Serial Diaphragm Ultrasonography to Predict Successful Discontinuation of Mechanical Ventilation

Introduction: Diaphragm excursion and contraction velocity measured using ultrasonography have been used to assess diaphragm function. We aimed to evaluate the performance of diaphragm ultrasonography during weaning from mechanical ventilation (MV). Methods: Diaphragm ultrasonography was performed on 73 mechanically ventilated patients who were being considered for extubation on three separate occasions: (1) on assist control mode (A/C) during consistent patient triggered ventilation, (2) following 30 min during a spontaneous breathing trial (SBT), (3) 4-24 h following extubation. Right hemidiaphragm excursion and contraction velocity were measured on A/C, during SBT, and following extubation. These measurements were correlated with the outcome of extubation. Results: Twenty patients failed extubation: 6 of whom required re-intubation and 14 of whom required non-invasive ventilatory support. During SBT, the mean diaphragm excursions were 1.7 +/- 0.82 cm in the group who failed extubation compared to 2.1 +/- 0.9 cm in the group who were successfully extubated (p = 0.06). To predict successful extubation, a decrease in diaphragm excursion of < 16.4% between A/C and SBT had a sensitivity of 84.9% and a specificity of 65%. The area under curve (AUC) for receiver operative characteristics for above cut-off was 0.75. Diaphragm contraction velocity performed poorly in predicting weaning outcome. Conclusion: Diaphragm excursion measured during SBT is an imperfect predictor of the outcome of extubation. Maintenance of diaphragm excursion between A/C and SBT has good performance characteristics by AUC analysis. Diaphragm contraction velocity has poor ability to predict outcome of extubation.