The Science Journal of the American Association for Respiratory Care

2011 OPEN FORUM Abstracts


Arthur Taft1, Susan Whiddon1, Ramona Herrington2, Manuel Castresana1,3; 1Respiratory Therapy, Georgia Health Sciences University, Augusta, GA; 2Respiratory Therapy, Medical College of Georgia Health Inc., Augusta, GA; 3Anesthesiology and Perioperative Medicine, Georgia Health Sciences University, Augusta, GA

Background: Currently, the best predictors of readiness to wean are the RSBI in conjunction with an SBT. However, some patients pass the weaning evaluation, yet fail removal from mechanical ventilatory support. The IPI is an index of respiratory status derived from capnography and pulse oximetry data (PETCO2, RR, SPO2, PR). This study evaluated the relationship of IPI with results from standard weaning evaluations (RSBI + SBT). Methods: This IRB approved, prospective, observational trial was performed on mechanically ventilated patients. All patients received standard of care, no intervention was performed for the purpose of this study. All subjects had an RSBI evaluation, followed by an SBT, per hospital protocol. IPI was continuously recorded immediately before and during the SBT. Clinicians were blinded to IPI. At the end of each SBT, the staff RT determined the outcome of the SBT (pass or fail). The recorded IPI data was averaged over the first 30 minutes of the SBT, and analyzed to determine the ability of IPI to predict weaning evaluation outcome. Statistical analysis was performed using SAS 9.2, and included two-sample t-tests and ROC analysis. Results: Seventeen subjects (9 males) were studied. Results from 23 SBTs were evaluated, with 13 subjects performing a single SBT and 4 subjects having multiple SBTs. Nineteen SBTs were performed using pressure support and 4 using a T-piece. Of the 23 SBTs, 14 were rated by the staff RT as passing and 9 failing. Of those SBTs classified as failing, mean IPI was lower (7.4 +/- 1.6) than those passing (8.8 +/- 1.1) p < 0.05. ROC analysis revealed a sensitivity of 0.93 and a specificity of 0.56 for a cutpoint of 7.5, with the area under the ROC curve of 0.73. Upon further analysis, the investigators felt that 1 passing SBT could have been rated failing (RT decision was over-ridden by MD) and 2 failing SBTs could have been rated as passing (1 passed SBT but failed ETT leak test, 1 had RR only transiently in the 30's). With these classification changes, IPI was 6.7 +/- 1.3 for failing SBTs and 9.0 +/- 0.7 for passing SBTs, p < 0.01. Sensitivity was 1.0, specificity was 0.75, and the area under the ROC curve was 0.89 for the same cutpoint of 7.5. Conclusions: IPI demonstrates reasonable agreement with clinical evaluation of SBTs by RT staff and may be useful in predicting readiness to discontinue mechanical ventilation. Further study is needed to more clearly define the value of IPI during the weaning process.
Sponsored Research - Educational Grant from Oridion Capnography, Inc