Indomethacin tocolysis and risk of necrotizing enterocolitis

To investigate the possible association of indomethacin tocolysis with neonatal necrotizing enterocolitis.

A case-control study was performed for the period November 1, 1997, through May 1, 1999. All cases of proven necrotizing enterocolitis were ascertained, and four controls for each case were randomly identified from all Special Care Nursery admissions before 37 weeks' gestation without necrotizing enterocolitis during that same period.

During the 18-month period there were 24 cases of necrotizing enterocolitis out of 10,200 deliveries. Infants with necrotizing enterocolitis were more preterm (29.7 +/- 3.9 compared with 32.7 +/- 6.0 weeks; P =.03) and had lower birth weights (1453 +/- 777 compared with 1820 +/- 678 g; P =.02) compared with controls (n = 96). Respiratory distress syndrome (RDS) and sepsis were both significantly associated with an increased risk of necrotizing enterocolitis: 16 of 24 cases compared with 40 of 96 controls had RDS (odds ratio [OR] 2.8, 95% confidence interval [CI] 1.0, 8.3) and 14 of 24 cases compared with 11 of 96 controls were septic (OR 10.8, 3.4, 95% CI 34.2). Indomethacin as a single agent was not associated with necrotizing enterocolitis (OR 1.0, 95% CI 0.2, 4.8). Using a logistic regression model, necrotizing enterocolitis was strongly associated with sepsis (adjusted OR 8.5, 95% CI 2.2, 32.5). When sepsis was removed from the model, double-agent tocolytic therapy was significantly associated with necrotizing enterocolitis (adjusted OR 6.9, 95% CI 1.1, 43.6).

Tocolysis with indomethacin as a single agent was not associated with necrotizing enterocolitis in this case-control study. Combination tocolytic therapy may be a marker for subclinical infection and not causally related to necrotizing enterocolitis.