Secondary Logo

Institutional members access full text with Ovid®

Share this article on:

An Analysis of Risk Factors for Hemolysis in Children on Extracorporeal Membrane Oxygenation*

Okochi, Shunpei, MD1; Cheung, Eva W., MD2; Barton, Sunjay, BS3; Zenilman, Ariela, MD1; Shakoor, Aqsa, MD1; Street, Cherease, BS1; Streltsova, Svetlana, RN, MSN, CCRN4; Chan, Christine, CCP5; Brewer, Michael P., CCP5; Middlesworth, William, MD1

doi: 10.1097/PCC.0000000000001699
Extracorporeal Support

Objectives: Hemolysis is a known complication of pediatric extracorporeal membrane oxygenation associated with renal failure and mortality. We sought to identify predictors of hemolysis in pediatric extracorporeal membrane oxygenation patients and determine its influence on outcomes.

Design: Retrospective, single-center study.

Setting: Urban, quaternary care center pediatric and neonatal ICU.

Patients: Ninety-six patients requiring extracorporeal membrane oxygenation.

Interventions: None.

Measurements and Main Results: Daily measurements of plasma-free hemoglobin were obtained while patients were on extracorporeal membrane oxygenation. Patients with a prior extracorporeal membrane oxygenation run, on extracorporeal membrane oxygenation for less than 24 hours, or without complete medical records were excluded from the study. Ninety-six patients met inclusion criteria, of which, 25 patients (26%) had plasma-free hemoglobin greater than 30 mg/dL. Of those patients, 15 of 25(60%) had plasma-free hemoglobin greater than 50 mg/dL, and 21 of 25(84%) occurred during the first 7 days on extracorporeal membrane oxygenation. Compared with patients without hemolysis, those with hemolysis were younger (0.2 mo [0.06–3.2 mo] vs 8.2 mo [0.6–86 mo]; p < 0.001), had a higher pericannulation international normalized ratio (3.9 [3.5–5.5] vs 2.6 [1.8–3.7]; p = 0.003), lower pericannulation platelet count (33 × 103/μL [22–42 × 103/μL] vs 61 × 103/μL [38–86 × 103/μL]; p < 0.001), and had a less negative inlet pressure (–3.5 mm Hg [–14 to 11.5 mm Hg] vs –19 mm Hg [–47 to 0 mm Hg]; p = 0.01). A greater proportion of patients with hemolysis had a heparin assay less than 0.2 mg/dL (50% vs 17%; p = 0.001) and had fluid removal via slow continuous ultrafiltration (32% vs 6%; p < 0.001). Patients with hemolysis had increased risk of in-hospital mortality (odds ratio 10.0; 95% CI 3.4–32; p < 0.001). On multivariable analysis, continuous ultrafiltration (odds ratio, 8.0; 95% CI, 1.9–42; p = 0.007) and pericannulation international normalized ratio greater than 3.5 (odds ratio, 7.2; 95% CI, 2.3–26; p = 0.001) were significantly associated with hemolysis.

Conclusions: Hemolysis is a common complication of pediatric extracorporeal membrane oxygenation. We found that patients with hemolysis (plasma-free hemoglobin > 30 mg/dL) had a 10-fold increase in in-hospital mortality. In our study cohort, hemolysis was associated with continuous ultrafiltration use, but not continuous renal replacement therapy. Additionally, our results suggest that the degree of coagulopathy (international normalized ratio > 3.5) at the time of cannulation influences hemolysis. Additional prospective studies are necessary to define further strategies to prevent hemolysis and improve outcomes in pediatric extracorporeal membrane oxygenation patients.

1Department of Surgery, New York - Presbyterian, Columbia University Herbert and Florence Irving Medical Center, New York, NY.

2Department of Pediatrics, New York - Presbyterian, Columbia University Herbert and Florence Irving Medical Center, New York, NY.

3College of Physicians and Surgeons, New York - Presbyterian, Columbia University Herbert and Florence Irving Medical Center, New York, NY.

4Department of Nursing, New York - Presbyterian, Columbia University Herbert and Florence Irving Medical Center, New York, NY.

5Cardiovascular Perfusion, New York - Presbyterian, Columbia University Herbert and Florence Irving Medical Center, New York, NY.

*See also p. 1089.

This work was performed at Columbia University Herbert and Florence Irving Medical Center, Morgan-Stanley Children’s Hospital of New York-Presbyterian.

The authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article, E-mail:so2462@cumc.columbia.edu

©2018The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies