UC Irvine

Background: Pediatric hospital-acquired venous thromboembolism (HA-VTE) has increased over the past ten years, with central venous catheters (CVC) being the strongest risk factor. Current tools are not sufficient to predict VTE risk at this time. The utility of biomarkers in predicting CVC-related VTE has been minimally explored.

Aims: Determine the utility of microparticles (MPs), factor VIII activity (FVIII) and thrombin generation (TG) in prospectively predicting VTE occurrence in hospitalized children with CVCs.

Methods: In this cohort pilot study, 42 hospitalized acutely ill children needing CVC placement (1 month to 21 years) and 42 age-matched healthy controls were enrolled. Venous samples were collected prior to or within 24 hours of CVC placement and processed using a strict protocol to minimize pre-analytical variables. MPs were measured using factor Xa initiated clot-based assay. FVIII was measured using a one-stage clot-based assay. TG was measured using the calibrated automated thrombogram.

Results: There were three CVC-related VTE events (7%) in our cohort. Xa clotting time (XaCT) ratio was significantly lower, while FVIII, peak thrombin (peak), estimated thrombin potential (ETP) and velocity index (VI) were significantly higher in patients with CVC-related VTE, as compared to healthy controls and patients without CVC-related VTE. Sensitivity/specificity analysis revealed optimal cutoff values for XaCT ratio (0.75), FVIII (370), ETP (1680), Peak (315) and VI (130), with AUC values of all biomarker ROC curves >0.9.

Conclusion: MPs, FVIII and TG can potentially predict pediatric CVC-related VTE in a prospective fashion. Further studies are needed to explore if stratification according to VTE risk will guide preventative efforts in this patient population.