Diaphragm electromyographic activity as a predictor of weaning failure

Intensive Care Med. 2012 Dec;38(12):2017-25. doi: 10.1007/s00134-012-2700-3. Epub 2012 Sep 26.

Abstract

Purpose: To compare breathing pattern descriptors and diaphragm electromyographic activity (EAdi)-derived indices obtained from a neurally adjusted ventilatory assist catheter during a spontaneous breathing trial (SBT) in patients successfully and unsuccessfully separated from the ventilator and to assess their performance as a potential marker to discriminate these two categories of patients.

Methods: Fifty-seven ready-to-wean patients were included in a prospective observational study. During a 30-min SBT (pressure support 7 cmH(2)O, zero end expiratory pressure), tidal volume (V (T)) and respiratory rate (RR) were obtained from the flow signal at baseline and at 3, 10, 20 and 30 min during the SBT. EAdi-derived indices were simultaneously computed: maximum of the EAdi (EAdi(max)), area under the inspiratory curve of EAdi (EAdi(AUC)), the difference between EAdi(max) and EAdi(min) (∆EAdi), EAdi(max)/V (T), EAdi(AUC)/V (T) and ∆EAdi/V (T). Patients, successfully (success group; n = 35) and unsuccessfully (failure group; n = 22) separated from the ventilator were compared.

Results: At baseline, the breathing pattern was similar in the two groups, whereas EAdi(max) and EAdi(AUC) were significantly lower in the success group (p < 0.05). In the failure group, RR and RR/V (T) increased significantly during the trial, V (T) decreased, whereas EAdi(max) and EAdi(AUC) did not change. At 3 min, the areas under the receiver operating characteristic-curve of RR/V (T) and the EAdi-derived indices to predict weaning outcome were 0.83 for the rapid shallow breathing index (RSBI), 0.84 for EAdi(max)/V (T) , 0.80 for EAdi(AUC)/V (T) (0.80) and 0.82 for ∆EAdi/V (T). The coefficient of variation for V (T) decreased in the failure group while that for EAdi(max) remained unchanged.

Conclusions: EAdi-derived indices provide reliable and early predictors of weaning outcome. However, the performance of these indices is not better than the RR/V (T).

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Diaphragm / physiopathology*
  • Electromyography*
  • Female
  • Humans
  • Interactive Ventilatory Support*
  • Male
  • Middle Aged
  • Prospective Studies
  • ROC Curve
  • Respiratory Function Tests*
  • Respiratory Rate
  • Sensitivity and Specificity
  • Tidal Volume
  • Ventilator Weaning*