Objective. In uncontrolled clinical trials, negative extrathoracic pressure has been shown to be an effective respiratory support. We aimed to assess its role in the context of current neonatal intensive care.

Design. A randomized controlled trial, with sequential analysis of matched pairs of infants. Matching was undertaken by stratified randomization from 15 groups divided according to gestational age, oxygen requirement, and whether patients were intubated at 4 hours of age.

Setting. Two neonatal intensive care units.

Patients. Two hundred forty-four patients (birth weight 1.53 ± 0.69 kg (mean ± SD); gestational age 30.4 ± 3.5 weeks) with respiratory failure.

Interventions. Patients were randomized at 4 hours of age to receive either standard neonatal intensive care, or standard care plus continuous negative extrathoracic pressure (CNEP, -4 to -6 cmH2O) applied within a purpose-designed neonatal incubator.

Outcome Scores. Clinical scores were calculated for each infant at 56 days of age, or death if earlier. Scores included measures for mortality, respiratory outcome, the presence of cerebral ultrasound abnormalities, patent arterial duct, necrotizing enterocolitis, and retinopathy. The treatment given for the higher score for each pair was recorded and the cumulative net number of pairs favoring CNEP plotted in the sequential analysis to provide an ethical early termination strategy. Individual components of the outcome score and other secondary measurements were analyzed on completion of the trial.

Results. The sequential analysis reached a decision boundary after 122 out of a possible maximum of 124 pairs were completed. The overall outcome score showed an overall significant benefit for CNEP. Secondary analysis showed that the use of CNEP was associated with an increase in mortality, cranial ultrasound abnormalities, and pneumothoraces, which were not statistically significant. However, 5% fewer patients were intubated (95% confidence interval [CI], 0-10), and the total duration of oxygen therapy among surviving infants at 56 days was lower (20.5 days, compared with 38.9 in controls; difference 18.4 days, 95% CI 3.8 to 33.0). Among all infants, the mean total duration of oxygen therapy was 18.3 days among CNEP-treated infants compared with 33.6 days among the controls (difference -15.3 days, 95% CI -0.2 to -30.4). This reduction in mean levels is entirely attributable to substantially fewer patients requiring prolonged oxygen therapy, the median duration of treatment being very similar in the two groups. As a result, commensurately fewer surviving infants showed chronic lung disease of prematurity.

Conclusions. The use of continuous negative pressure improves the respiratory outcome for neonates with respiratory failure.

This content is only available via PDF.
You do not currently have access to this content.