Thrombosis in infants in the neonatal intensive care unit: Analysis of a large national database

Thrombosis in the neonatal population is rare, but increasing. Its incidence and management are not well understood.

To investigate the incidence, associated factors, and management of thrombosis in the neonatal intensive care unit (NICU) population.

We performed a retrospective cohort study of infants admitted to a Pediatrix Medical Group-affiliated NICU from 1997 through 2015. We determined the prevalence of venous and arterial thrombosis, and assessed demographic characteristics and known risk factors. Categorical variables were compared with the Pearson χ² test and continuous variables with Wilcoxon rank-sum tests. Stepwise logistic regression was used to identify associated factors. The primary outcome was incidence of thrombosis. Secondary analyses investigated correlations between clinical and demographic characteristics and thrombosis.

Among 1 158 755 infants, we identified 2367 (0.20%) diagnosed with thrombosis. In a multivariable regression analysis, prematurity, male sex, congenital heart disease, sepsis, ventilator support, vasopressor receipt, central venous catheter, invasive procedures, and receipt of erythropoietin were associated with increased risk of thrombosis, while Black race and Hispanic ethnicity were associated with reduced risk. The majority of infants diagnosed with thrombosis (73%) received no anticoagulation, but anticoagulant use in infants with thrombosis was higher than those without (27% versus 0.2%, P < .001). Thrombosis in infants was associated with higher mortality (11% versus 2%, P < .001) and longer hospital stays (57 days, [interquartile range (IQR) 28--100] versus 10 days, [IQR 6--22], P < .001).

In the largest national study to date, we found that thrombosis in NICU patients is associated with prematurity, low birth weight, sepsis, and invasive procedures.