Indicators of fatigue and of prolonged weaning from mechanical ventilation in surgical patients.

Abstract

Indicators of weaning success have been tested primarily in patients who have been ventilated for short periods of time, and they may not be as accurate in cases where support has been required for longer than a few days. In patients requiring longer periods of support it is difficult to estimate the likelihood of successful liberation. Therefore we evaluated established weaning indices for their accuracy in surgical patients who required > or = 72 hours of mechanical ventilation. Surgical patients who required mechanical ventilation for > or = 72 hours were prospectively followed (over 6 months). We obtained standard indices of ventilatory function daily once patients were ready to wean. These indices included the respiratory rate/tidal volume ratio (RSBI), the maximal inspiratory pressure, and the minute ventilation. The duration of weaning and explicitly defined episodes of fatigue were the outcomes of interest. Statistical analyses evaluated the multiple factors that might influence the duration of weaning. Ninety-five patients (66% trauma; 34% surgery) survived to begin weaning, and 93% were liberated. The median duration of mechanical ventilation prior to weaning was 4 days (range 3-16 days), and the median duration of weaning was 3 days (range 0-56 days). Fatigue occurred in 36 patients and was not reliably predicted by any of the weaning measurements. However, a RSBI of > 105 on the first day of weaning was associated with prolonged weaning. By multivariate analysis, an RSBI of > 105 on the first day of weaning predicted prolonged weaning (hazard ratio 1.9; p = 0.03). After 72 hours of mechanical ventilation, clinical fatigue and successful liberation are not reliably predicted by standard indices of respiratory muscle strength and reserve. However, an RSBI of >105 observed once the patient is ready to wean is associated with prolonged weaning.