Airway occlusion pressure and breathing pattern as predictors of weaning outcome.

Airway occlusion pressure (P0.1) and the ratio of breathing frequency (f) to tidal volume (VT) (f/Vt) are good predictors of weaning outcome. However, the specificity of f/VT in predicting weaning success is relatively low. We postulated that the product of P0.1 and f/VT (P0.1*f/VT) would better predict weaning outcome than either variable alone. In 45 male patients, we prospectively evaluated P0.1*f/VT, P0.1, and f/VT in predicting weaning outcome. The threshold values of each variable were determined from published data. The sensitivity, specificity, and positive and negative predictive values in detecting weaning success, and the area under the receiver operating characteristic (ROC) curves were calculated. Ten (22%) of the 45 patients failed weaning. P0.1*f/VT yielded the highest specificity and positive and negative predictive values. P0.1*f/VT, P0.1, and f/VT were all highly sensitive (0.97); but they were less specific, 0.60 for P0.1*f/Vt and 0.40 for P0.1 and f/VT. The areas under the ROC curves for P0.1*f/VT, P0.1, and f/VT were not significantly different. We conclude that P0.1*f/VT has equivalent sensitivity as P0.1 and f/VT. P0.1 slightly improves the specificity of f/VT in predicting weaning success.