Explaining recurrent thrombotic events and predicting bleeding in critically ill neonates diagnosed with umbilical central venous catheter-related portal vein thrombosis

Abstract

Background

The short-term risk of recurrent venous thrombotic events (VTEs) and bleeding in hospitalized neonates after umbilical venous catheter (UVC)-related portal vein thrombosis (PVT) is unknown.

Objectives

The primary objective was to determine the incidence of recurrent VTEs in neonates with UVC-related PVT who required subsequent central vascular access. Secondary objectives included investigating the incidence of bleeding events, variables that explained VTE recurrence, and bleeding predictors.

Methods

This longitudinal retrospective cohort study included neonates diagnosed with UVC-related PVT at the Hospital for Sick Children between 2016 and 2024. An explanatory model for VTE recurrence (using mixed models for repeated observations) and an exploratory predictive model for bleeding events (using logistic regression) were developed.

Results

Three hundred neonates requiring 684 subsequent central venous accesses were included. In total, 78 recurrent events were identified in 62 patients at median age 32 days (Q1-Q3, 14-67 days) for the first recurrence. Incidence proportion of recurrent VTEs was 21%; incidence rate was 0.45 per 100-catheter days. Throughout follow-up, 216 neonates (72%) sustained 328 bleeding events; 50% of patients experienced major bleeding at median age 5 days (Q1-Q3, 3-15 days), including microbleeds and small central nervous system bleeds. Younger age at vascular access placement and larger catheter diameter explained VTE recurrence risk. The exploratory model suggested that gestational age, sex, and underlying conditions predicted bleeding risk.

Conclusion

There was a high frequency of bleeding and thrombosis in this cohort of acutely ill neonates. Strategies to decrease recurrent thrombosis risk are relevant to avoid thrombotic complications during the highest bleeding-risk period.