Risk Factors for Neonatal Thrombosis in the Neonatal Intensive Care Unit -a Case Control Study

Background

Neonates admitted to the neonatal intensive care unit (NICU) are an age group most susceptible to thrombosis in pediatrics. Besides central access devices (CAD), maternal and neonatal factors are reported to be associated with thrombosis. This has not been well investigated because of the relatively rare incidence of thrombosis and the inherent heterogeneity of thrombotic events. An assessment of the impact of individual risk factors is an essential step, as appropriate risk stratification is fundamental to evidence based thromboprophylaxis policy.

Objective

To identify risk factors associated with thrombosis in sick neonates admitted to the NICU.

Methods

A case-control study was conducted using the Children's Hospital Neonatal Database (CHND) dataset with neonates admitted to the NICU at Ann and Robert H. Lurie Children's Hospital and Nationwide Children's hospital between Jan 2010 and June 2013. Cases were neonates diagnosed with either arterial or venous thrombosis during their NICU stay and controls were matched to the cases in the same patient pool in a 1:4 ratio on the basis of gestational age and presence or absence of CAD. Neonates with a less than 72-hour stays in the NICU or complex congenital heart defects needing surgical repair as well as re-admission data were excluded. Chi-square tests were performed to compare characteristics as well as potential risk factors between cases and controls. A conditional multivariate logistic regression analysis included potential risk factors with p-value<0.1 in chi-square tests and with clinical relevance. Local IRB approval was obtained at both sites.

Results

A total of 47 cases were identified in 4,122 NICU patients (11.4 per 1,000 patients). There were 32 (68%) males and 27 (57.5%) preterm neonates with thrombosis. On univariate analysis blood stream infections (BSI) and prolonged mechanical ventilation were significantly more common in cases than controls Table 1. A conditional multivariate analysis showed that prolonged mechanical ventilation was independently associated with higher risk of thrombosis (OR 3.03 [95% CI: 1.29, 7.09], p value 0.01 Table 2).

Conclusions

The incidence of thrombosis appears to be 5 fold higher than that previously reported in a Canadian registry. After matching for CAD and GA, prolonged mechanical ventilation represents an independent risk factor of thrombosis in neonates. This is the largest study of systematic assessment of risk factors in neonates with mechanical ventilation being reported as a risk factor independent of CAD. Larger multi-centered data should confirm the study results for developing evidence-based risk stratification protocols and thrombosis prevention strategies.