Complement Activation in Pediatric Patients Undergoing Extracorporeal Membrane Oxygenation (ECMO)

Sunday, March 22, 2026

9:45 AM - 10:45 AM Central Time

Location: Connections Central - RST 18

Elizabeth Grogan, MD (she/her/hers)

Ohio

Disclosure(s): No relevant financial relationship(s) to disclose.
JB

Jianli Bi, MD, PhD

Senior Research Scientist, Cardiovascular Research Shared Resource Core Manager
Nationwide Children's Hospital at The Ohio State University

Disclosure information not submitted.
RH

Rejuan Haque, PhD, MS

Biostatistician
Ohio State University

Disclosure information not submitted.
JF

Joshua Frazier, MD

Associate Professor, Division of Pediatric Critical Care, Director of Pediatric ECMO Program
Nationwide Children's Hospital at The Ohio State University

Disclosure information not submitted.
MB

Molly Ball, MD

Associate Professor, Division of Neonatology, Director of Neonatal ECMO
Nationwide Children's Hospital at The Ohio State University

Disclosure information not submitted.
CY

Chack-Yung Yu, D.Phil.

Professor, Center for Microbe and Immunity Research
The Research Institute at Nationwide Children's Hospital at The Ohio State University

Disclosure information not submitted.
AY

Andrew Yates, MD

Professor, Division of Cardiology and Pediatric Critical Care
Nationwide Children's Hospital at The Ohio State University

Disclosure information not submitted.

Introduction: ECMO supports critically ill patients but triggers systemic inflammation causing morbidity and mortality. One driver of inflammation is the complement pathway. The terminal product of the complement pathway is the pro-inflammatory membrane attack complex (C5b-9). We hypothesize that following ECMO cannulation of neonatal and pediatric patients, (1) C5b-9 will rapidly increase and subsequently decrease to pre-cannulation levels within 24h, (2) there will be no difference in levels between age groups, and (3) higher peak levels will be seen in non-survivors.

Methods: C5b-9 levels were analyzed using ELISAs prior to cannulation and at 2h, 4h and 6h, and 1d and 2d post cannulation. A log transformed linear effect model was used to determine estimated mean values. A Mann-Whitney test compared survivors to non-survivors. Data presented as estimated mean (CI) or median (IQR).

Results: Ten neonatal (≤ 28d) and ten pediatric (29d – 18y) patients were included. Neonatal median age was 2 (1, 4)d and pediatric median age was 3 (1, 4)y. 13 patients survived to discharge. Estimated mean C5b-9 plasma levels peaked at 2h post cannulation and returned to pre-cannulation levels by 1d post cannulation; there were no differences between age groups at any time point. Pre-cannulation levels increased at 2h post cannulation in neonatal (646 (313, 1332)ng/mL to 1273 (617, 2626)ng/mL, p=0.01) and pediatric (1070 (519, 2208)ng/mL to 1807 (870, 3751)ng/mL, p=0.13) patients, and subsequently decreased by 1d post cannulation in neonatal (1273 (617, 2626)ng/mL to 601 (289, 1247)ng/mL, p=0.01) and pediatric (1807 (870, 3751)ng/mL to 891 (429, 1849)ng/mL, p=0.02) patients. Median C5b-9 levels at 2h post cannulation trended higher in non-survivors (2979 (1106, 9453)ng/mL v 1093 (776, 1463)ng/mL, p=0.07).

Conclusions: C5b-9 levels peak within 2h of cannulation and then decline to pre-cannulation levels by 24h. In our pilot study, 2h post cannulation levels trended higher in non-survivors. These results suggest that ECMO triggers rapid complement pathway activation and identify complement pathway inhibitors, such as eculizumab, as potential therapeutic targets to improve outcomes.