Extended-phase anticoagulant treatment of acute venous thromboembolism in children: a cohort study from the EINSTEIN-Jr phase 3 trial

Background Extended-phase anticoagulation of venous thromboembolism in children is not well documented nor systematically reported. Previously, we reported on recurrent venous thromboembolism and bleeding during acute-phase anticoagulation in EINSTEIN-Jr, a randomised controlled study in 500 children with venous thromboembolism comparing rivaroxaban to standard anticoagulants. The aim of the present study was to evaluate the efficacy and safety of extended-phase anticoagulant therapy in children and to characterise factors associated with the decision to extend anticoagulation.Methods Children aged 17 years or younger, who were enrolled in the EINSTEIN-Jr trial (NCT02234843) from 107 paediatric hospitals in 28 countries, and who had previously completed a 3-month acute anticoagulation treatment phase (1-month in children <2 years with catheter-related venous thromboembolism) for acute venous thromboembolism within the trial were included in this cohort study. After completion of the preceding acute anticoagulation treatment phase, children could extend study treatment for up to 9 months (or up to 2 months for children <2 years with catheter-related venous thromboembolism). Study anticoagulants were bodyweight-adjusted rivaroxaban (tablets or suspension) in a 20 mg equivalent dose or standard anticoagulants (heparin or vitamin K antagonist). The main outcomes were suspected recurrent venous thromboembolism (primary efficacy outcome) and clinically relevant bleeding (principal safety outcome), both confirmed or refuted by appropriate objective testing. Cumulative incidences of efficacy and safety outcomes are reported for children who received extended anticoagulation within the framework of the study. We also compared demographic and clinical characteristics of those administered any extended-phase anticoagulation (whether within or outside the framework of the study) with those not administered extended-phase anticoagulation, applying multivariable logistic regression.Findings 248 (51%) children received extended-phase anticoagulation between Nov 14, 2014, and Jan 15, 2019, 214 within the study and 34 outside the framework of the study. During extended-phase anticoagulant treatment, recurrent venous thromboembolism occurred in three (1%) of the 214 children within the study (cumulative incidence 3·0%; 95% CI 0·9-9·8). Clinically relevant non-major bleeding occurred in four (2%) of 214 children (3·3%; 1·2-9·2). Fatal venous thromboembolism or major bleeding did not occur. Outcome rates were similar with rivaroxaban or standard anticoagulants. Symptomatic index venous thromboembolism (odds ratio 1·88; 95% CI 1·14-3·11), unprovoked venous thromboembolism or persistent risk factor (2·16; 1·46-3·19), and residual thrombosis on repeat imaging (3·79; 2·52-5·71) were associated with the decision to extend anticoagulation.Interpretation Incidences of recurrent venous thromboembolism and bleeding during extended-phase anticoagulant treatment were low and similar to those observed during acute-phase treatment and adult studies on extended-phase anticoagulant treatment, providing valuable information for clinical practice on extended anticoagulation in children.Funding Bayer and Janssen Research & Development.