Debate has existed as to the relative contributions of surfactant therapy or patent ductus arteriosus (PDA) to the cause of pulmonary hemorrhage (PH). To assess the effect of early closure of PDA on both PH and bronchopulmonary dysplasia (BPD), we conducted a prospective, double-blind, randomized, placebo-controlled trial of early indomethacin (INDO) in 276 infants of BW 500-1249g who received synthetic surfactant during 1992-95. Infants were randomized on day 1 to receive either placebo or INDO (0.2, 0.1, 0.1 mg/kg by 48 hrs). PH included any episode of non-traumatic blood-tinged pulmonary secretions in the 1st 28d; severity of PH was assessed by # and character of episodes & radiographic/clinical worsening. Among all infants (both groups), 68 cases of PH were detected and were associated with increased risk of death and IVH; 85% of PH began in 1st 4d. 59 cases had >1 episode and 21 had frank blood. Treatment groups were balanced for BW (879±193g), gender, race, RDS, surfactant doses & timing, & time of 1st treatment dose (mean 9 hrs). Findings: Table Both groups did not differ in NEC, perforation, IVH, renal function or overall survival, although survival was improved by INDO in the sub-group <750 gm (65% vs 43%, P=0.06). Fatal PH occurred in 4 placebo and 0 INDO infants. Logistic regression indicated that predictors of developing PH were symptomatic PDA (O.R. 5.5, 95% C.I. 2.7-11.0), BW (1.3, 1.04-1.5 per - 100g), and RDS (7.4, 0.9-59.4) (controlled for # of surfactant doses & timing), INDO group infants with PH more commonly had symptomatic PDA (71% vs 21%, P=0.0001) and PDA relapse. We conclude that in surfactant-treated tiny infants: (1) PH is associated with adverse outcomes (2) PDA and BW are strong predictors of PH, & (3) early indomethacin reduces PH but not BPD.Supported by AHA-MC.
