Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest

Abstract Background Optimal mean arterial pressure (MAP) range after cardiac arrest remains uncertain. This study aimed to investigate the association between MAP and neurological outcomes during the early post-resuscitation period, with the goal of identifying optimal MAP range associated with favo...

Full description

Saved in:
Bibliographic Details
Main Authors: Sijin Lee, Kwang-Sig Lee, Kap Su Han, Juhyun Song, Sung Woo Lee, Su Jin Kim
Format: Article
Language:English
Published: BMC 2025-07-01
Series:Journal of Intensive Care
Subjects:
Online Access:https://doi.org/10.1186/s40560-025-00814-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849388288690880512
author Sijin Lee
Kwang-Sig Lee
Kap Su Han
Juhyun Song
Sung Woo Lee
Su Jin Kim
author_facet Sijin Lee
Kwang-Sig Lee
Kap Su Han
Juhyun Song
Sung Woo Lee
Su Jin Kim
author_sort Sijin Lee
collection DOAJ
description Abstract Background Optimal mean arterial pressure (MAP) range after cardiac arrest remains uncertain. This study aimed to investigate the association between MAP and neurological outcomes during the early post-resuscitation period, with the goal of identifying optimal MAP range associated with favorable outcomes. Methods This retrospective observational study included 291 post-cardiac arrest patients treated at a tertiary care center. Five machine learning models to predict favorable neurological outcomes using hourly MAP measurements during the first 24 h after return of spontaneous circulation (ROSC) were compared and Random Forest model was selected due to its superior performance. Variable importance and Shapley Additive exPlanations (SHAP) were used to investigate the association between MAP and favorable neurological outcomes. SHAP dependence plots were used to identify optimal MAP ranges associated with favorable outcomes. In addition, individual-level predictions were interpreted using local interpretable model-agnostic explanations (LIME) and SHAP force plots. Results Machine learning analysis showed that MAP were associated with favorable neurological outcomes, with higher variable importance during the first 6 h after ROSC. SHAP analysis revealed an inverted U-shaped relationship between MAP and favorable neurological outcomes, with an optimal threshold of 79.56 mmHg (IQR: 73.70–82.54). This threshold remained consistent across both early (1–6 h: 79.26 mmHg) and later (7–24 h: 80.09 mmHg) hours. Individual-level explanations using SHAP and LIME highlighted that maintaining higher MAP during the early post-resuscitation period contributed positively to outcome predictions. Conclusions Machine learning analysis identified MAP as a major predictor of favorable neurological outcomes, with higher variable importance during the first 6 h after ROSC. MAP showed an inverted U-shaped relationship with favorable neurological outcomes, with an optimal threshold of approximately 80 mmHg.
format Article
id doaj-art-784e1a09616b4cf2acbd8d04a6ac6478
institution Kabale University
issn 2052-0492
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series Journal of Intensive Care
spelling doaj-art-784e1a09616b4cf2acbd8d04a6ac64782025-08-20T03:42:20ZengBMCJournal of Intensive Care2052-04922025-07-0113111110.1186/s40560-025-00814-xOptimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrestSijin Lee0Kwang-Sig Lee1Kap Su Han2Juhyun Song3Sung Woo Lee4Su Jin Kim5Department of Emergency Medicine, Korea University College of Medicine & Anam HospitalAI Center, Korea University College of MedicineDepartment of Emergency Medicine, Korea University College of Medicine & Anam HospitalDepartment of Emergency Medicine, Korea University College of Medicine & Anam HospitalDepartment of Emergency Medicine, Korea University College of Medicine & Anam HospitalDepartment of Emergency Medicine, Korea University College of Medicine & Anam HospitalAbstract Background Optimal mean arterial pressure (MAP) range after cardiac arrest remains uncertain. This study aimed to investigate the association between MAP and neurological outcomes during the early post-resuscitation period, with the goal of identifying optimal MAP range associated with favorable outcomes. Methods This retrospective observational study included 291 post-cardiac arrest patients treated at a tertiary care center. Five machine learning models to predict favorable neurological outcomes using hourly MAP measurements during the first 24 h after return of spontaneous circulation (ROSC) were compared and Random Forest model was selected due to its superior performance. Variable importance and Shapley Additive exPlanations (SHAP) were used to investigate the association between MAP and favorable neurological outcomes. SHAP dependence plots were used to identify optimal MAP ranges associated with favorable outcomes. In addition, individual-level predictions were interpreted using local interpretable model-agnostic explanations (LIME) and SHAP force plots. Results Machine learning analysis showed that MAP were associated with favorable neurological outcomes, with higher variable importance during the first 6 h after ROSC. SHAP analysis revealed an inverted U-shaped relationship between MAP and favorable neurological outcomes, with an optimal threshold of 79.56 mmHg (IQR: 73.70–82.54). This threshold remained consistent across both early (1–6 h: 79.26 mmHg) and later (7–24 h: 80.09 mmHg) hours. Individual-level explanations using SHAP and LIME highlighted that maintaining higher MAP during the early post-resuscitation period contributed positively to outcome predictions. Conclusions Machine learning analysis identified MAP as a major predictor of favorable neurological outcomes, with higher variable importance during the first 6 h after ROSC. MAP showed an inverted U-shaped relationship with favorable neurological outcomes, with an optimal threshold of approximately 80 mmHg.https://doi.org/10.1186/s40560-025-00814-xCardiac arrestPost-cardiac arrest careMean arterial pressureNeurological outcomesExplainable machine learning
spellingShingle Sijin Lee
Kwang-Sig Lee
Kap Su Han
Juhyun Song
Sung Woo Lee
Su Jin Kim
Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
Journal of Intensive Care
Cardiac arrest
Post-cardiac arrest care
Mean arterial pressure
Neurological outcomes
Explainable machine learning
title Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
title_full Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
title_fullStr Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
title_full_unstemmed Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
title_short Optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
title_sort optimal mean arterial pressure for favorable neurological outcomes in patients after cardiac arrest
topic Cardiac arrest
Post-cardiac arrest care
Mean arterial pressure
Neurological outcomes
Explainable machine learning
url https://doi.org/10.1186/s40560-025-00814-x
work_keys_str_mv AT sijinlee optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest
AT kwangsiglee optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest
AT kapsuhan optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest
AT juhyunsong optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest
AT sungwoolee optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest
AT sujinkim optimalmeanarterialpressureforfavorableneurologicaloutcomesinpatientsaftercardiacarrest