Pulmonary hemorrhage risk in infants with a clinically diagnosed patent ductus arteriosus: a retrospective cohort study.

OBJECTIVE: To determine if an early, clinically detectable patent ductus arteriosus (PDA) was associated with pulmonary hemorrhage (PH) in infants who received rescue artificial surfactant therapy.
METHODS: This retrospective cohort study of 233 low birth weight infants (< or = 1700 g) who received artificial surfactant therapy for respiratory distress syndrome compared antenatal and postnatal characteristics of infants with PH and without PH. Pulmonary hemorrhage was defined by an onset of bright red blood from the endotracheal tube in quantities that resulted in increased ventilatory support and a new infiltrate on a chest radiograph.
RESULTS: Pulmonary hemorrhage occurred in 6% (15/233) of the infants. Thirty-three percent (5/15) of the infants with PH died within 14 days of the hemorrhage. Of the 15 PH, 73% occurred within 48 hours of the first surfactant dose. Pulmonary hemorrhage was more common in male infants and infants of mothers who received antibiotic therapy during labor (P < or = .04). Infants with PH received surfactant earlier than those without PH (P = .04). Nursery events or therapies occurring following surfactant therapy that were associated with PH included: little improvement in ventilatory efficiency index (P = .01), dopamine infusion (P = .04), and the presence of a clinically detectable PDA before, or at the time of, the PH [60% (9/15) vs 33% (71/217), P = .03]. After adjusting for severity of illness before surfactant therapy, risk of PH remained greater in infants who developed symptoms of a PDA. Dopamine support appeared to modify the association between PDA and PH.
CONCLUSIONS: In this retrospective cohort study, pulmonary hemorrhage, was associated with the presence of a clinically detectable patent ductus arteriosus before, or at the time of, pulmonary hemorrhage.