Predictive factors for the outcome of noninvasive ventilation in pediatric acute respiratory failure.

OBJECTIVES

To identify success and failure prognostic signs of noninvasive ventilation in pediatric acute respiratory failure. Noninvasive ventilation constitutes an alternative treatment for pediatric acute respiratory failure. However, tracheal intubation should not be delayed when considered necessary.

DESIGN

Prospective, noncontrolled, clinical study.

SETTING

Pediatric intensive care unit in a university hospital.

PATIENTS

Children (age range, 1 month-16 yrs) with moderate-to-severe acute respiratory failure who received noninvasive ventilation during a 4-year period. Failure was defined as the need for tracheal intubation.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Nine (19.1%) of 47 patients needed tracheal intubation between the third and 87th hour after the start of treatment (33.6 ± 29.6 hrs). Failure was associated with the younger age group (4 ± 3.3 yrs vs. 7.7 ± 5 yrs, p < .04), acute respiratory distress syndrome (failure/acute respiratory distress syndrome: 5 of 10 vs. failure/non acute respiratory distress syndrome: 4 of 37, p = .013), and worsening radiographic images taken at 24 hrs and/or 48-72 hrs (p = .001 and p < .001, respectively). A significant reduction in heart rate was observed between the second and fourth hour after starting noninvasive ventilation (130 ± 25.8 bpm vs. 116 ± 27.7 bpm, p < .001) and Pco2 (54.1 ± 19.5 torr vs. 48.6 ± 14.3 torr; 7.21 ± 2.6 vs. 6.48 ± 1.91 kPa, p < .007) in the success group. The failure group had a higher rate of breathing assistance, both initial and maximal. In the multivariant analysis, only maximum mean airway pressure and Fio2 formed part of the success/failure discriminant function with a cutoff point of 11.5 and 0.57, respectively.

CONCLUSIONS

Modifications in a patient's respiratory assistance were made depending on the clinical, blood gas, and radiologic evolution of the patient. Mean airway pressure and Fio2 values of >11.5 and 0.6, respectively, predict failure and possibly set the limit above the patient's risk of delayed intubation increases.