CPAP vs Intubation and Mechanical Ventilation for Preterm Infants
Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet J, Carlin JB, for the COIN Trial Investigators
N Engl J Med. 2008;358:700-708

The authors note that despite advances in neonatal care for premature infants, the incidence of bronchopulmonary dysplasia (BPD) has not improved. Previous evidence had suggested that some infants who required ventilatory support might do well with continuous positive airway pressure (CPAP) instead of intubation, but a large randomized trial was needed.

Morley and colleagues conducted a multicenter trial conducted in several countries. Eligible subjects were born between 25 and 28 weeks' gestation, were born at a trial hospital, and were breathing spontaneously at 5 minutes after birth but required respiratory support.

Infants were randomized equally to the 2 arms -- nasal CPAP or intubation and mechanical ventilation. Infants initially receiving CPAP could be intubated later if they experienced clinical deterioration to prespecified points. Infants in the CPAP arm did not receive surfactant unless they were later intubated.

The outcome of interest was a combined one: rate of either death or BPD. The authors also evaluated multiple secondary outcomes, including need for supplemental oxygen at 28 days of life and surfactant use.

The study enrolled 610 infants from 1999 to 2006. The groups were similar with only a slight predominance of males in the intubation group (49% males in CPAP group vs 56% males in intubation group).

Of all subjects, approximately 36% met the primary outcome. There was no statistical difference between rates of primary outcome between the 2 groups -- 33.9% of CPAP infants vs 38.9% of intubated infants, odds ratio 0.80 (95% confidence interval: 0.58-1.12). Rates of death were similar between the 2 groups at 6.5% for CPAP and 5.9% for intubated infants. Stratification by gestational age did not alter these general results.

Forty-six percent of CPAP infants later required intubation, but surfactant use was overall much less in the CPAP group. Pneumothorax occurred in 9% of CPAP infants but only in 3% of intubated infants.

The authors conclude that there was no difference in the rate of primary outcome between premature infants treated with CPAP vs intubation and mechanical ventilation.

This trial was powered to detect a one-third reduction in primary outcome, so the number of enrolled infants is not an issue for the lack of difference between the 2 groups. At this point, these data would suggest that CPAP is an option for treating premature infants who require respiratory support, but one can't say conclusively that it seems to offer clinical (outcome) advantages. Longer-term follow-up will also be of potential interest in the CPAP group because these infants were less likely to still require oxygen at 28 days of life. The long-term nature of lung problems resulting from BPD, and the less traumatic approach of CPAP make looking at longer-term pulmonary outcomes of interest in this group of patients as well.

Abstract