FAILURE OF INTRAVENOUS INFUSIONS IN NEONATES

The median life expectancy (survival) of 286 peripheral intravenous infusion sites in 105 babies in a children's intensive care unit was 36 h. Unadjusted univariate survival analysis revealed that dextrose infusions and the initial infusions received by a baby had longer survival than total parenteral nutrition (TPN) infusions and later infusions respectively. Also infusions with cloxacillin and penicillin survived for longer than average while infusions with phenytoin had reduced survival. Gestational age, weight, infusion site, other drugs, co-infusion of Intralipid with TPN solutions and neutralization of TPN did not influence survival of infusions. Multivariate survival analysis confirmed the findings for TPN and penicillin but not for cloxacillin, phenytoin or later infusions. Multivariate analysis also suggested that survival was improved with ampicillin and aminophylline and worse for leg sites, for older babies and for infusions in which the fluids were given at greater rates. It also indicated that neutralization of TPN improved survival.