Risk Determination in Pediatric Patients for Venous Thromboembolism

Objective: Cases of hospital-acquired acute deep venous thrombosis (HA-DVT) and pulmonary embolism (HA-PE) within the pediatric population are rare. Risk assessment models (RAMs) are used to identify a patient’s likelihood of developing venous thromboembolism (VTE). When VTE risk exceeds bleeding risk of low dose anticoagulants, patients are prescribed thromboprophylaxis. Within this study, we aim to highlight previously identified risk factors for pediatric HA-VTE by comparing three different RAMs: Caprini, Children's Hospital-Acquired Thrombosis (CHAT) Consortium Admission, and the Peds-Clot clinical Decision Rule (PCDR).

Methods: This retrospective case-control study focused on pediatric patients between the ages of 12 and 22 admitted to a single institution from 2017 to 2022. Cases of acute VTE (defined as HA-DVT and/or PE) were matched to randomly selected controls 2:1 matched on length of stay. Total scores for Caprini, CHAT, and PCDR RAMs were calculated using clinical data extracted from chart review. Cases and controls were compared using chi² and Mann-Whitney U test for categorical and continuous outcomes, respectively.

Results: The total incidence of acute VTE amongst adolescent and young adult inpatients was 0.2% (33/16,186). Cases were found to have a significantly higher BMI compared to controls (38.8±7.9 vs 22.4±6.2, p<0.001). No difference in age was found for cases compared to controls (mean age 17.2±2.2 vs 16.3±2.4, p=0.64). The PCDR scores were slightly different between cases and controls (2.32 vs 1.24, p=0.001) as were the modified Caprini scores (6.5 vs 2, p<0.001) whereas the CHAT predicted probability demonstrated a near ten-fold enrichment in risk amongst cases compared to controls (9.3% v. 0.8%, p=0.02). The risk factor determined to be significant across all RAM was the placement of a central venous catheter.

Conclusions: Cases for our study had an extremely low incidence rate, underscoring the importance of a robust risk assessment model capable of defining a high-risk population. Even amongst the highest risk population, (those who developed acute HA-VTE), the average calculated risk in the best-performing RAM was only found to be 9.3%. Currently available pediatric VTE RAMs may lead to overuse of thromboprophylaxis when broadly applied.