Venous thromboembolism in children with central nervous system tumors: Comparison of an institutional cohort to a national administrative database

BACKGROUND: Central nervous system (CNS) tumors are the second most common malignancy of childhood, and published data on venous thromboembolism (VTE) rate and risk factors for these patients are outdated or incomplete. Here, we determine the cumulative incidence and risk factors for VTE in this population. PROCEDURE: VTE diagnosis and associated clinical risk factors were abstracted and analyzed for two cohorts of children (0-21 years) diagnosed with CNS tumors between January 1, 2010 to September 30, 2018. The first study was a retrospective single institution cohort study. The initial observations were confirmed across multiple pediatric hospitals using the Pediatric Health Information System (PHIS) administrative database. RESULTS: The single-institution cohort included 338 patients aged 3 days to 20.9 years (median age, 8.6 years); VTE developed in eight (2.4%) patients. The PHIS cohort included 17 634 patients aged from 0 to 21.9 years (median: 9.5 years); VTE developed in 354 (2.0%) patients. Univariate analysis for the single-institution cohort identified central venous catheter (CVC) placement as a risk factor for VTE (odds ratio [OR] 8.40, 95% confidence interval [CI] 1.43-49.41, P = .0186). Multivariable analysis of the PHIS dataset identified CVC placement (OR 1.97, 95% CI 1.57-2.46; P < .0001), obesity (OR 2.96, 95% CI 1.21-7.26; P = .0177), and more than one hospital admission (OR 3.54, 95% CI 2.69-4.64; P < .0001) as significant predictors of VTE. VTE diagnosis was not associated with increased mortality in either cohort. CONCLUSIONS: The VTE rate in children with CNS tumors is low (2%). CVC placement was identified as a modifiable risk factor in both cohorts.