731 - Pre-ECMO Factors Associated with Severe Bleeding on ECMO

Monday, May 1, 2023

9:30 AM – 11:30 AM ET

Poster Number: 731
Publication Number: 731.401

Krystin Wong, University of Hawaii, John A. Burns School of Medicine, Honolulu, HI, United States; Laura A. Borruso, University of Hawaii/Tripler Army Medical Center, Honolulu, HI, United States; Sherri Sommer-Candelario, Kapi'olani Medical Center for Women and Children, Honolulu, HI, United States; Kyle Tadaki, Kapi'olani Medical Center for Women and Children, Honolulu, HI, United States; Stefanie Ching, Kapi'olani Medical Center for Women and Children, Honolulu, HI, United States; Sheree Kuo, University of Hawaii, John A. Burns School of Medicine, Honolulu, HI, United States; Venkataraman Balaraman, University of Hawaii, John A. Burns School of Medicine, Honolulu, HI, United States; Len Y. Tanaka, University of Hawaii, John A. Burns School of Medicine, Honolulu, HI, United States

Presenting Author(s)

Krystin Wong, BA

Medical Student
University of Hawaii, John A. Burns School of Medicine
Honolulu, Hawaii, United States

Background: Bleeding and thrombosis are common complications of ECMO. Heparin, the primary anticoagulant for ECMO, works primarily via a heparin-antithrombin III complex. Maintaining adequate antithrombin III (AT3) levels and careful titration of heparin infusion on ECMO are standard practice to achieve optimal anticoagulation while avoiding excessive bleeding and thrombosis formation. Despite protocols outlining heparin and AT3 management, severe bleeding can occur while on ECMO.

Objective: Our objective was to identify patient factors, coagulation measures and/or inflammatory markers in the first 48 hours of ECMO associated with the development of significant bleeding on ECMO.

Design/Methods: This study was approved by the Institutional Review Board at Hawai‘i Pacific Health. Single center retrospective chart review of ECMO patients aged birth to 18 years between 2014 and 2022 was performed. Patients who were taken off ECMO or died within 24 hours of cannulation were excluded. Patient demographics, initial coagulation measures including antithrombin III (AT3), prothrombin (PT), partial-thromboplastin time (PTT), and platelets, heparin dose and pump flow rate during the first 48 hours of ECMO were collected. Outcomes examined included intracranial hemorrhage (ICH), gastrointestinal (GI) and cannula site bleeding, survival from ECMO and to discharge. Descriptive statistics, linear and multiple regression analyses were performed using R version 4.2.1.

Results:
A total of 23 neonatal and 20 pediatric patients were included in our study. The majority of the patients received veno-arterial ECMO. Heparin doses and ECMO flow rates were similar at 24 and 48 hours. There were twenty bleeding events identified (5 cannula site, 3 GI, and 12 ICH). Most cannula site bleeds occurred in pediatric patients (80%), while the majority of ICH occurred in neonates (75%). Initial PT and PTT were significantly higher in patients with ICH, while no correlation was found with initial AT3 levels. Following multiple regression analysis, initial PTT was found to have a stronger correlation with ICH.

Conclusion(s): Consistent with previous studies, our ECMO patients who developed ICH had higher initial PT and PTT levels. Therefore, alternate heparin dosing at the time of cannulation should be considered in neonatal patients with elevated PT/PTT levels to potentially prevent ICH. Timely correction of low AT3 levels may not affect ICH risk in neonatal ECMO patients. Larger prospective studies are needed to determine optimal anticoagulation management to minimize risk of ICH, especially in neonatal patients.