Automatic Tube Compensation: A Better Weaning Test?

Abstract & Commentary

Synopsis: In this small study, differences in the respiratory rate to tidal volume ratio (RTVR) after 1 hour of spontaneous breathing with ATC were a good predictor of whether patients would remain extubated or require reintubation. But was it really better than other tests?

Source: Cohen JD, et al. Automatic tube compensation-assisted respiratory rate to tidal volume ratio improves the prediction of weaning outcome. Chest. 2002;122:980-984.

The use of automatic tube compensation (ATC) is becoming more common during mechanical ventilation to reduce the work of breathing during spontaneous ventilation. This feature is available on the Evita ventilator manufactured by Dräger (Lubeck, Germany). ATC consists of added pressure support at a level automatically calculated to balance the predicted additional work imposed by the endotracheal tube. The theory is that ATC (at 100%) creates a breathing work condition identical to that present following extubation. The amount of support can be set as a percentage of this optimal value (eg, 50%).

Cohen and associates determined classic weaning parameters with and without ATC (100%) in 35 patients who successfully completed a 1-hour spontaneous breathing trial (SBT) followed by extubation. Twenty-five of the patients remained extubated without signs of respiratory failure, and 10 required reintubation within 48 hours. Demographic information, severity of illness at presentation, cause of respiratory failure, peak airway pressure (Paw), auto-PEEP level, respiratory rate, tidal volume, minute ventilation, occlusion pressure (P.01), heart rate, systolic pressure, and RTVR with and without ATC were determined before and after the SBT (when appropriate).

The only differences between the 2 groups with respect to the assessments made prior to the SBT were in the RTVR with and without ATC. Following the 1-hour SBT, these remained significantly different, as did the peak airway pressure, respiratory rate, and tidal volume. When a multivariate analysis was used on all the data, only the RTVR with and without ATC, both before and after the SBT, and the Paw were significant predictors of success or failure. In each case, the receiver operating characteristic (ROC) value (indicating the degree to which the assessment was different from random in predicting extubation success) was 0.70 or more. The best fit of the ROC curves was obtained when the RTVR on ATC was divided by the Paw following the SBT, which had a ROC value of 0.84, a good balance between sensitivity and specificity. Cohen et al suggest using this index in predicting weaning success.

Comment by Charles G. Durbin, Jr., MD

Adding ATC is suggested as a way to reduce the work of breathing for the patient that is imposed by the endotracheal tube. While some patients seem to be more comfortable on this mode, the actual value of ATC in mechanical ventilation and specifically in weaning has not been determined. Cohen et al suggest a novel way of using this technique to improve prediction of weaning outcome. They had to work hard to do this. While the advanced statistical analysis applied to this data suggests the superiority of a complex series of evaluations, more traditional indices also performed reasonably well. The differences in respiratory rate (22.6 in patients successfully extubated and 28.2 in those requiring reintubation) was also highly predictive of success or failure. In fact, Cohen et al did not test it, but the change in respiratory rate during the SBT appears highly predictive. Likewise, the change in peak airway pressure and the change in tidal volume also appear to be highly predictive. I think it is important to carefully review some of the data in this report before accepting the conclusions suggested.

The RTVT by itself was highly predictive. Adding ATC to this measurement contributed little improvement in prediction (the area under the ROC). It was only when the RTVT on ATC was divided by the Paw following the SBT that a significant improvement in ROC appeared. The same improvement would have occurred using the RTVT without ATC in this ratio, although they did not perform (or at least report) this analysis. A more honest title for this paper would have been: "Dividing the respiratory rate to tidal volume ratio by the peak airway pressure predicts extubation success." This paper’s title appears to be a way of advertising the unique ventilator mode of a particular ventilator and really has nothing to do with the major findings of the study. By applying a series of complicated statistics to derived parameters, Cohen et al were able to manipulate their data to support their bias.

It is important to understand the patient population studied. Although none of the patients would have failed traditional weaning criteria, almost 30% of these patients required reintubation. This is an incredibly high percentage. When this group is evaluated, many of them (40%) had nonrespiratory reasons for reintubation, including new sepsis and poor secretion clearance. The comparison groups thus are not reflective of the usual ICU extubation issues, and the results of the analyses are suspect.

The take-home message from my analysis of this paper is that testing patients for weaning success while on ATC is not necessary. The usual parameters (ie, respiratory rate and RTVT) following a SBT should still be considered the tarnished "gold standards." Energy devoted to improvement in these predictors is probably wasted as new approaches to support of patients with respiratory failure, namely, noninvasive ventilation, are likely to replace many of the current reintubations. 

Dr. Durbin is Professor of Anesthesiology Medical Director of Respiratory Care University of Virginia.