A multi-institutional registry of pediatric hospital-acquired thrombosis cases: The Children's Hospital-Acquired Thrombosis (CHAT) project.

BACKGROUND: Pediatric hospital-acquired venous thromboembolism (HA-VTE) rates have increased dramatically. To achieve generalizable knowledge in the derivation and validation of HA-VTE risk factors and risk prediction models and inform future risk-stratified prevention strategies, multi-institutional studies are needed.

OBJECTIVES: This paper presents an investigator-initiated, multicenter pediatric case-cohort study designed to identify risk factors for HA-VTE to create a HA-VTE risk prediction model.

METHODS: A registry, which houses pertinent variables from HA-VTE subjects and non-HA-VTE controls, was created for the Children's Hospital-Acquired Thrombosis (CHAT) study. Specific variables from the registry associated with HA-VTE risk will be identified using multivariable regression to create a pediatric HA-VTE risk prediction model to be prospectively validated.

RESULTS: Seven large pediatric institutions have entered over 600 HA-VTE subjects aged 0-21years of age into the registry. Subjects showed a male predominance (57%), a median age of three years (IQR 0.3-13) and were most likely admitted to an intensive care unit (57%) at VTE diagnosis. Median time to HA-VTE was 10days after admission. The most prevalent risk factors include central venous catheters (80%), surgery (43%), systemic steroids (31%), congenital heart disease (27%), infection (14%) and cancer (13%).

CONCLUSIONS: CHAT, with its creation of a risk prediction model with prospective validation using the CHAT registry, is a novel study design and will be the first step in identifying safe and effective strategies to decrease HA-VTE in children by helping define the highest risk population for initial, or more aggressive, thromboprophylaxis efforts.