Announcement
Thank you for attending THSNA 2026. The virtual meeting is now closed.
Thank you for attending THSNA 2026. The virtual meeting is now closed.
Presentation Details
| Managing Neonatal Central Venous Catheter-Related Thrombosis: A Role for Rivaroxaban Molly C Easterlin1, Mohit Mehra1, Xinwen Li1, Wendy Leung1, Ashwini Lakshmanan2, Shahab Noori1, Philippe Friedlich1, Joseph Miller1, Julie Jaffray3. 1University of Southern California/Children's Hospital Los Angeles, Los Angeles, CA, USA.2Kaiser Permanente Bernard J.Tyson School of Medicine, Pasadena, CA, USA.3University of California, San Diego/Rady Children’s Hospital-San Diego, San Diego, CA, USA |
Abstract
Background: In 2021 the Food and Drug Administration approved the use of rivaroxaban, a direct factor Xa inhibitor, for treatment and secondary prevention of venous thromboembolism (VTE) in patients from birth (at least 37 weeks’ gestation) to 18 years who have received at least five days of parenteral therapy. For many patients, rivaroxaban is the preferred anticoagulant due to oral formulation, lack of laboratory monitoring, and a more predictable dose response compared to more traditional anticoagulants. However, data are sparse on the use of rivaroxaban in neonates. Objectives: To describe the use of rivaroxaban to treat central venous catheter-related thromboses (CVC-RT) and adverse effects among neonatal intensive care unit (NICU) patients in a single center tertiary care center. Methods: Using natural language processing (NLP) of nursing documentation and clinical notes we created a database of every CVC placed in NICU patients between 1/1/2017-9/10/2024. We used a large language model, Qwen 2.5, on internal servers with a custom prompt to identify thrombosis from ultrasound reports. Treatment with anticoagulant medication (heparin, enoxaparin, rivaroxaban) was identified by NLP of medication administration records. Statistical analyses consisted of descriptive statistics. Results: 15 NICU patients (53.5% female; median gestational age at birth 37 weeks, range 23-40 weeks; median post-menstrual age at rivaroxaban initiation 46 weeks, range 39-61 weeks) received rivaroxaban for 18 CVC-RTs (3 patients had 2 CVC-RTs) (Table 1). In total, 45 CVCs were placed during the study period, with an average of 3 per patient. Heparin was initially started in 3 (6.7%) patients prior to enoxaparin and 12 (80.0%) initially started on enoxaparin prior to transitioning to rivaroxaban (Table 2). The median duration of enoxaparin prior to rivaroxaban was 23.3 days. All patients were started on rivaroxaban while inpatient and 6 were continued on it upon discharge. The median total duration of rivaroxaban was 42.0 days (IQR 22.8-92.5). The median total length of any form of anticoagulation was 102 days (IQR 43-184). On initial ultrasound 45% of thromboses were occlusive, 5% near occlusive, 10% partially occlusive, and 40% non-occlusive. By 6 weeks, 8/18 (44%) thromboses resolved, 9/18 (50%) were unchanged, and 1/18 (6%) had decreased in size but persisted. Of the 5 patients whose VTE had not resolved and US was obtained at 3 months, 1 (20%) resolved and 4 (80%) were still present. At 6 months, 3 (60%) had resolved and 2 (40%) showed chronic changes. Three patients developed a second thrombosis associated with a subsequent PICC after having completed treatment for the primary thrombosis, occurring 4-12 weeks after the end of anticoagulation; only 1 was on rivaroxaban prophylaxis at the time. There were no serious bleeding events and 3 minor bleeding events while patients were on rivaroxaban (2 patients with bloody stool, 1 bleeding with vaccines). Conclusions: In this single-center cohort, rivaroxaban was successfully used for treatment of CVC-RT in NICU patients, with no serious bleeding events and acceptable thrombus outcomes. These findings suggest that rivaroxaban may be a feasible and well-tolerated alternative to traditional parenteral anticoagulation in neonates.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.