Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

Overview

Journal Res Sq

Date 2024 Jan 23

PMID 38260374

Authors

Andrew Kalra

Preetham Bachina

Benjamin L Shou

Jaeho Hwang

Meylakh Barshay

Shreyas Kulkarni

Isaac Sears

Carsten Eickhoff

Christian A Bermudez

Daniel Brodie

Corey E Ventetuolo

Bo Soo Kim

Glenn J R Whitman

Adeel Abbasi

Sung-Min Cho

Affiliations

Soon will be listed here.

Abstract

Objective: To determine if machine learning (ML) can predict acute brain injury (ABI) and identify modifiable risk factors for ABI in venoarterial extracorporeal membrane oxygenation (VA-ECMO) patients.

Design: Retrospective cohort study of the Extracorporeal Life Support Organization (ELSO) Registry (2009-2021).

Setting: International, multicenter registry study of 676 ECMO centers.

Patients: Adults (≥18 years) supported with VA-ECMO or extracorporeal cardiopulmonary resuscitation (ECPR).

Interventions: None.

Measurements And Main Results: Our primary outcome was ABI: central nervous system (CNS) ischemia, intracranial hemorrhage (ICH), brain death, and seizures. We utilized Random Forest, CatBoost, LightGBM and XGBoost ML algorithms (10-fold leave-one-out cross-validation) to predict and identify features most important for ABI. We extracted 65 total features: demographics, pre-ECMO/on-ECMO laboratory values, and pre-ECMO/on-ECMO settings.Of 35,855 VA-ECMO (non-ECPR) patients (median age=57.8 years, 66% male), 7.7% (n=2,769) experienced ABI. In VA-ECMO (non-ECPR), the area under the receiver-operator characteristics curves (AUC-ROC) to predict ABI, CNS ischemia, and ICH was 0.67, 0.67, and 0.62, respectively. The true positive, true negative, false positive, false negative, positive, and negative predictive values were 33%, 88%, 12%, 67%, 18%, and 94%, respectively for ABI. Longer ECMO duration, higher 24h ECMO pump flow, and higher on-ECMO PaO were associated with ABI.Of 10,775 ECPR patients (median age=57.1 years, 68% male), 16.5% (n=1,787) experienced ABI. The AUC-ROC for ABI, CNS ischemia, and ICH was 0.72, 0.73, and 0.69, respectively. The true positive, true negative, false positive, false negative, positive, and negative predictive values were 61%, 70%, 30%, 39%, 29% and 90%, respectively, for ABI. Longer ECMO duration, younger age, and higher 24h ECMO pump flow were associated with ABI.

Conclusions: This is the largest study predicting neurological complications on sufficiently powered international ECMO cohorts. Longer ECMO duration and higher 24h pump flow were associated with ABI in both non-ECPR and ECPR VA-ECMO.

References

Schmidt M, Zogheib E, Roze H, Repesse X, Lebreton G, Luyt C . The PRESERVE mortality risk score and analysis of long-term outcomes after extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. Intensive Care Med. 2013; 39(10):1704-13. PMC: 7094902. DOI: 10.1007/s00134-013-3037-2. View

Shah N, Farhat A, Tweed J, Wang Z, Lee J, McBeth R . Neural Networks to Predict Radiographic Brain Injury in Pediatric Patients Treated with Extracorporeal Membrane Oxygenation. J Clin Med. 2020; 9(9). PMC: 7565544. DOI: 10.3390/jcm9092718. View

Pavlushkov E, Berman M, Valchanov K . Cannulation techniques for extracorporeal life support. Ann Transl Med. 2017; 5(4):70. PMC: 5337209. DOI: 10.21037/atm.2016.11.47. View

Kalra A, Shou B, Zhao D, Wilcox C, Keller S, Whitman G . Racial and ethnical discrepancy in hypoxemia detection in patients on extracorporeal membrane oxygenation. JTCVS Open. 2023; 14:145-170. PMC: 10328809. DOI: 10.1016/j.xjon.2023.02.011. View

Booke H, Zacharowski K, Adam E, Raimann F, Bauer F, Flinspach A . Cardiopulmonary resuscitation in veno-venous-ECMO patients-A retrospective study on incidence, causes and outcome. PLoS One. 2023; 18(8):e0290083. PMC: 10420365. DOI: 10.1371/journal.pone.0290083. View

Cho S, Canner J, Chiarini G, Calligy K, Caturegli G, Rycus P . Modifiable Risk Factors and Mortality From Ischemic and Hemorrhagic Strokes in Patients Receiving Venoarterial Extracorporeal Membrane Oxygenation: Results From the Extracorporeal Life Support Organization Registry. Crit Care Med. 2020; 48(10):e897-e905. PMC: 8170265. DOI: 10.1097/CCM.0000000000004498. View

Aubron C, Cheng A, Pilcher D, Leong T, Magrin G, Cooper D . Factors associated with outcomes of patients on extracorporeal membrane oxygenation support: a 5-year cohort study. Crit Care. 2013; 17(2):R73. PMC: 4056036. DOI: 10.1186/cc12681. View

Akin S, Caliskan K, Soliman O, Muslem R, Guven G, van Thiel R . A novel mortality risk score predicting intensive care mortality in cardiogenic shock patients treated with veno-arterial extracorporeal membrane oxygenation. J Crit Care. 2019; 55:35-41. DOI: 10.1016/j.jcrc.2019.09.017. View

Shou B, Ong C, Premraj L, Brown P, Tonna J, Dalton H . Arterial oxygen and carbon dioxide tension and acute brain injury in extracorporeal cardiopulmonary resuscitation patients: Analysis of the extracorporeal life support organization registry. J Heart Lung Transplant. 2022; 42(4):503-511. PMC: 10050131. DOI: 10.1016/j.healun.2022.10.019. View

10.

Kalra A, Shou B, Zhao D, Wilcox C, Keller S, Kim B . Extracorporeal Membrane Oxygenation Physiological Factors Influence Pulse Oximetry and Arterial Oxygen Saturation Discrepancies. Ann Thorac Surg. 2023; 117(6):1221-1228. PMC: 10959762. DOI: 10.1016/j.athoracsur.2023.09.019. View

11.

Schmidt M, Burrell A, Roberts L, Bailey M, Sheldrake J, Rycus P . Predicting survival after ECMO for refractory cardiogenic shock: the survival after veno-arterial-ECMO (SAVE)-score. Eur Heart J. 2015; 36(33):2246-56. DOI: 10.1093/eurheartj/ehv194. View

12.

Muller G, Flecher E, Lebreton G, Luyt C, Trouillet J, Brechot N . The ENCOURAGE mortality risk score and analysis of long-term outcomes after VA-ECMO for acute myocardial infarction with cardiogenic shock. Intensive Care Med. 2016; 42(3):370-378. DOI: 10.1007/s00134-016-4223-9. View

13.

Abbasi A, Karasu Y, Li C, Sodha N, Eickhoff C, Ventetuolo C . Machine learning to predict hemorrhage and thrombosis during extracorporeal membrane oxygenation. Crit Care. 2020; 24(1):689. PMC: 7727105. DOI: 10.1186/s13054-020-03403-6. View

14.

Carlson J, Etchill E, Whitman G, Kim B, Choi C, Tonna J . Early withdrawal of life sustaining therapy in extracorporeal cardiopulmonary resuscitation (ECPR): Results from the Extracorporeal Life Support Organization registry. Resuscitation. 2022; 179:71-77. PMC: 9530029. DOI: 10.1016/j.resuscitation.2022.07.038. View

15.

Ong C, Etchill E, Dong J, Shou B, Shelley L, Giuliano K . Neuromonitoring detects brain injury in patients receiving extracorporeal membrane oxygenation support. J Thorac Cardiovasc Surg. 2021; 165(6):2104-2110.e1. DOI: 10.1016/j.jtcvs.2021.09.063. View

16.

Mamdani M, Slutsky A . Artificial intelligence in intensive care medicine. Intensive Care Med. 2020; 47(2):147-149. DOI: 10.1007/s00134-020-06203-2. View

17.

Hu L, Li L . Using Tree-Based Machine Learning for Health Studies: Literature Review and Case Series. Int J Environ Res Public Health. 2022; 19(23). PMC: 9736500. DOI: 10.3390/ijerph192316080. View

18.

Carlson J, Etchill E, Enriquez C, Peeler A, Whitman G, Choi C . Population Characteristics and Markers for Withdrawal of Life-Sustaining Therapy in Patients on Extracorporeal Membrane Oxygenation. J Cardiothorac Vasc Anesth. 2021; 36(3):833-839. PMC: 9435591. DOI: 10.1053/j.jvca.2021.04.040. View

19.

Peigh G, Cavarocchi N, Keith S, Hirose H . Simple new risk score model for adult cardiac extracorporeal membrane oxygenation: simple cardiac ECMO score. J Surg Res. 2015; 198(2):273-9. DOI: 10.1016/j.jss.2015.04.044. View

20.

Hafner S, Beloncle F, Koch A, Radermacher P, Asfar P . Hyperoxia in intensive care, emergency, and peri-operative medicine: Dr. Jekyll or Mr. Hyde? A 2015 update. Ann Intensive Care. 2015; 5(1):42. PMC: 4653126. DOI: 10.1186/s13613-015-0084-6. View