Timing of Acute Kidney Injury in Children on Central vs Peripheral VA ECMO for Septic Shock

Tuesday, March 24, 2026

8:45 AM - 9:45 AM Central Time

Location: Connections Central - RST 15

Salvador Maffei, MD (he/him/his)

Baylor College of Medicine Texas Children's Hospital, Texas

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

Sameer Thadani, MD

Baylor College of Medicine Texas Children's Hospital, Texas

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

Christin Silos, MS

Senior Biostatistics and Data Science Associate
Baylor College of Medicine, Texas

Disclosure information not submitted.
NC

Nancy Chung, MD

Assistant Professor
UT Southwestern Medical Center

Disclosure information not submitted.
SN

Sarah Nelin, MD

Assistant Professor
University of Mississippi Medical Center

Disclosure information not submitted.
KB

Katie Brandewie, MD

Pediatric Critical Care Fellow
Baylor College of Medicine/ Texas Children's Hospital

Disclosure information not submitted.
AA

Ayse Akcan Arikan, MD

Baylor College of Medicine at Texas Childrens Hospital, Texas

Disclosure information not submitted.
AO

Andrea Ontaneda, MD

Assistant Professor
Baylor College of Medicine Texas Children's Hospital, Texas

Disclosure information not submitted.

Introduction: Central veno-arterial extracorporeal membrane oxygenation (VA-ECMO) for septic shock is suggested to improve survival in children compared to peripheral cannulation. Ability to increase flows without additive hemolytic burden has been postulated to reverse organ injury more quickly. Acute kidney injury (AKI) is very common in VA-ECMO, and the comparative frequency and severity of AKI have not been studied between cannulation strategies.

Methods: This is a retrospective cohort study of children supported with VA-ECMO for septic shock admitted to Texas Children’s Hospital between 2014 and 2025. Our primary outcome was the rate of AKI in the first 14 days following ECMO cannulation, comparing peripheral cannulation to central. Secondary outcomes included rates of severe AKI, use of renal replacement therapy (RRT), AKI duration, and hospital mortality. We defined AKI by the KDIGO serum creatinine criteria. Severe AKI was stage 2 or 3 AKI and/or use of RRT.

Results: We included 68 children, median age 58 months (IQR 12, 142), weight 17 kg (11, 40), pSOFA 10 (8,13) at PICU admission and 12 (10,14) on cannulation. 41 (60%) patients were peripherally cannulated. AKI was present prior to ECMO in 54% of peripheral vs 59% central (p=0.65). Day 1 AKI was 73% in peripheral vs 70% in central (p=0.80). Severe AKI was 63% in peripheral vs 56% in central (p=0.52). Median AKI duration was 4 days (IQR 2, 8) in peripheral vs 8 days (1, 11) in central (p=0.35). CRRT use was 59% in peripheral vs 41% in central (p=0.15), with median duration 19 days (3, 29) peripheral and 19 days (9, 30) central (p=0.46). Hospital mortality was 51% in peripheral vs 11% in central (p < 0.001).

Conclusions: Prevalent and incident, AKI was common in our cohort of pediatric patients supported with VA-ECMO for septic shock. AKI, including severe AKI, occurred within 24 hours for both groups. There was no significant difference in the frequency, severity, duration of AKI or the use or duration of RRT between groups, but our power was limited. The reduced hospital mortality in central cannulation is consistent with previously published studies and does not seem to be driven by decreased AKI. Pattern of other organ failures on central vs peripheral VA-ECMO needs further exploration.