In a systematic review of data on drug use and adverse clinical events in infants with birth weights under 2000 g, we observed an association between germinal matrix-intraventricular hemorrhage and the use of heparin to maintain the patency of vascular catheters. Sixty-six infants with germinal-matrix (periventricular) or intraventricular hemorrhage or both (cases) were matched with 254 infants with other conditions (controls), and analysis, taking the matching factors into account, yielded an odds ratio of 14.0 (95 percent confidence interval, 5.4 to 34). When potential confounding factors were taken into account, the odds ratio was 3.9 (1.4 to 11). The association did not appear to vary according to the severity of hemorrhage or to the method of administration or dose of heparin. The data suggest that the routine use of heparin in neonatal intensive care units is associated with a four-fold increase in the risk of germinal matrix-intraventricular hemorrhage. Because of the possibility that confounding may have been incompletely controlled for, the true risk cannot be determined from these data, and a controlled clinical trial of heparin use in low-birth-weight infants is recommended.