Risk-prediction tool for identifying hospitalized children with a predisposition for development of venous thromboembolism: Peds-Clot clinical Decision Rule

Abstract

Background: The prevalence of VTE is increasing in tertiary pediatric hospitals. Identification of high-risk populations using uniform criteria is required to develop evidence-based VTE prevention guidelines.

Objective: To develop a VTE risk prediction rule, the Peds-Clot clinical Decision Rule (PCDR), to identify high-risk children who were at increased risk of developing VTE.

Methods: This retrospective case-control study developed the PCDR using a derivation cohort (173 cases, 346 controls) and validated it on a separate validation cohort (100 cases, 100 controls). A uniform data collection strategy was applied to derive both the samples. Conditional logistic regression analyses were used to develop a risk-prediction model. Each significant predictor was assigned a score based on its beta coefficient and the PCDR was developed. ROC curves were derived to test the performance of the PCDR.

Results: Characteristics of derivation and validation cohorts were comparable. Six risk factors (positive blood stream infection, central venous catheter, direct admission to ICU/NICU, hospitalization for ≥ 7 days, immobilization for > 72 h, and use of birth control pills) formed the final risk prediction model (risk score range, 0.5-9.5). A risk score of 3 or more identified high-risk children at a sensitivity of 70% and specificity of 80% and AUC of 0.852 (95% confidence interval, 0.814-0.890). The application of a risk score to the validation sample showed sensitivity 57% and specificity 88% and an AUC of 0.875 (95% confidence interval, 0.82-0.924).

Conclusion: Incorporation of the PCDR in routine clinical care can be an attractive strategy to identify high-risk hospitalized children with a predisposition for VTE. The clinical utility of the PCDR needs validation in prospective studies.