Thrombosis risk factor assessment and implications for prevention in critically ill children

To describe nursing compliance with a computer-based pediatric thrombosis risk assessment tool; to generate an estimate of risk factors present in our population; and to explore relationships between risk factors and confirmed thrombotic events.

Institutional review board-approved prospective, observational cohort study.

Pediatric intensive care unit within a tertiary care children's hospital.

All infants and children admitted to the pediatric intensive care unit during a 6-month study period (January 1, 2010-June 30, 2010).

Eight hundred admissions were enrolled, representing 742 patients. Thrombosis risk assessment scores were recorded for 707 admissions (88% of total). Mean age = 6.95 ± 6 yrs, mean weight = 28 ± 23 kg, 45% female. A total of 32 thrombi (14 prehospital and 18 in-hospital) were present in the study group. This translated to an overall occurrence rate of 4.3% (1.9% for prehospital and 2.4% for in-hospital). Logistic regression identified that for every 1-point increase in total thrombosis score, the risk of developing a symptomatic thrombus increased by 1.57-fold (95% confidence interval 0.192-5.5) to 2.12-fold (95% confidence interval 0.175-18.34), for prehospital and in-hospital thrombi, respectively (p < .05). The most important risk factors identified for development of any thrombus were thrombophilia (acquired or inherited) (p < .001), presence of a central catheter (p = .01), and age <1 or >14 yrs (p = .052).

Incorporation of a scoring system into the bedside nursing assessment flow sheet was successful and identified children at risk for in-hospital thrombosis. The overall score appears to be most indicative of thrombus risk. These data may serve as a platform for future development of routine screening and possible interventional trials in critically ill children.