Abbreviated anticoagulation regimen with direct oral anticoagulants is safe and efficacious in the management of infection-associated cerebral sinovenous thrombosis in children

Abstract Introduction There has been a dramatic increase in the rate of venous thrombo-embolism (VTE) amongst hospitalized children in the United States. Although recent landmark pediatric trials have allowed evidence-based treatment decisions, real-world data validating the safety and efficacy of these therapeutic approaches, particularly in the management of high-risk VTE such as cerebral sinovenous thrombosis (CSVT), remain limited. The principal objective of the study is to investigate the safety and efficacy of an abbreviated course (<12 weeks) of anticoagulation using direct oral anticoagulants (DOACs) in children with infection-associated CSVT. Methods Permission for this retrospective cohort study was obtained from the Institutional Review Board. Three prospectively maintained institutional databases - neuroradiology, neurosurgery, and anticoagulation were used to identify eligible patients, defined as children (≤ 18 years of age) with infection-associated CSVT who received care at our quaternary care center between 1/1/2018 and 12/31/2024. Clinical characteristics, treatment, and radiological outcomes were abstracted from medical records. Safety (clinically relevant bleeding) and efficacy (progressive/recurrent thrombosis) were defined per criteria established by the International Society on Thrombosis and Haemostasis. Descriptive statistical analyses were conducted to summarize the data. Fisher's Exact Test was used to assess differences in the use of DOACs and abbreviated treatment courses across diagnosis years. Results Imaging data for 2991 patients was screened for inclusion. Two hundred and three patients were diagnosed with CSVT between 1/1/2018 and 12/31/2024, of which 47 (23%) patients had infection-associated CSVT. Median age (IQR) at diagnosis of infection-associated CSVT was 6.7 (3.9 – 9.8) years, and median duration of hospitalization was 14 (10-19) days. Thirty-five patients (75%) were admitted to the ICU, and 9 (19%) required mechanical ventilation. Head and neck region infections included: 28 (38%) otomastoiditis, 15 (22%) sinusitis, 4 (6%) neck infection, and 16 (23%) CNS infection. Five (11%) patients had evidence of venous infarction, and 6 (13%) had evidence of hemorrhage on initial diagnostic imaging. All patients were treated with antibiotics, and 45 (96%) received anticoagulation. Pharmaco-mechanical thrombectomy ± stent placement was performed in 4 (9%) patients, of whom two patients had recurrent thrombosis within 1 week of the procedure. Median time interval between diagnosis and starting anticoagulation was 1 (0-2) day. Forty-five (96%) patients were started on unfractionated heparin (UFH) or low molecular weight heparin, and 15 (32%) were transitioned to rivaroxaban or apixaban. Children diagnosed with infection-associated CSVT between 2022 – 2024 were more likely to receive a DOAC compared to those diagnosed between 2018 – 2021, though this was not statistically significant (41% versus 9%; p=0.07). Median duration of anticoagulation for the entire cohort was 7.8 (4.3 – 13.8) weeks. Thirty-one (74%) patients received <12 weeks of anticoagulation, and 11 (26%) received ≥ 12 weeks. Children diagnosed with infection-associated CSVT between 2022 – 2024 were not more likely to receive an abbreviated course of anticoagulation compared to those diagnosed between 2018 – 2021 (75% versus 70%; p=1). Three (6.4%) patients developed progressive thrombosis within the first week of diagnosis, while on UFH. Only one patient (3%) developed worsening hemorrhage within 24 hours of starting UFH. No patient developed recurrent/ progressive thrombosis or clinically relevant bleeding while receiving DOACs. Twenty-seven (57%) patients underwent thrombophilia testing of whom one (4%) was noted to be heterozygous for the factor V Leiden (G1691A) variant. Forty-six (98%) patients underwent follow-up imaging at our institution. Median time from diagnosis to most recent imaging was 14.1 (6.7 – 25.7) weeks. Forty-four (96%) had complete/partial thrombus resolution, and 2 (4%) had stable thrombus burden. One patient died of an underlying malignancy. Conclusion Our data suggests that an abbreviated course of anticoagulation using DOACs is safe and efficacious in the management of infection-associated CSVT in children. Additionally, thrombophilia work-up is low yield and likely not indicated in this cohort.