Risk factors for extubation failure in infants with severe acute bronchiolitis.

OBJECTIVE

To evaluate demographic characteristics, mechanical-ventilation parameters, blood gas values, and ventilatory indexes as predictors of extubation failure in infants with severe acute bronchiolitis.

METHODS

We conducted a prospective observational study from March 2004 to September 2005 with consecutive infants (ages 1-12 months) with severe acute bronchiolitis and considered ready to be extubated. We calculated mean airway pressure and oxygenation index. Before extubation we measured respiratory rate, tidal volume, rapid shallow breathing index, maximal inspiratory pressure, and load/force balance. Arterial blood gases were measured 1 hour before extubation. Extubation was classified as a failure if the infant needed re-intubation within 48 hours.

RESULTS

Extubation failure occurred in 6 (15%) of the 40 extubated infants. The respective median (and interquartile range) age, weight, and days of mechanical ventilation for the extubation-failure and extubation-success groups were: age 5 (3-8) months versus 4 (4-6) months (P = .87), weight 4 (3-5) kg versus 6 (5-7) kg (P < .001), and mechanical ventilation days 8 (6-23) d versus 6 (5-12) d (P = .52). There were no significant differences in arterial blood gas values or mechanical-ventilation parameters between the extubation-success and extubation-failure groups. There were statistically significant differences between the extubation-failure and extubation-success groups for 2 risk factors, weight <or= 4 kg and tidal volume <or= 4 mL/kg, when those risk factors had a large area under the curve of the receiver operating characteristic. Variables that had a large area under the curve were minute volume <or= 0.8 mL/kg/min and maximal inspiratory pressure <or= 50 cm H(2)O. Variables that had a small area under the curve were load/force balance >or= 5 and rapid shallow breathing index >or= 6.7.

CONCLUSIONS

In infants with severe acute bronchiolitis the extubation process is complex because of the combined features of this disease. Pediatric studies have not definitely determined predictive factors, weaning protocols, or ventilatory predictive indexes of extubation failure risk in infants with severe acute bronchiolitis. Lower minute volume and lower maximal inspiratory pressure had large areas under the curve of the receiver operating characteristic for extubation-failure risk in infants with severe acute bronchiolitis.