Cargando…

Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study

INTRODUCTION: At hospital arrival, early prognostication for children after out-of-hospital cardiac arrest (OHCA) might help clinicians formulate strategies, particularly in the emergency department. In this study, we aimed to develop a simple and generally applicable bedside tool for predicting out...

Descripción completa

Detalles Bibliográficos
Autores principales: Goto, Yoshikazu, Maeda, Tetsuo, Nakatsu-Goto, Yumiko
Formato: Online Artículo Texto
Lenguaje:English
Publicado: BioMed Central 2014
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4227055/
https://www.ncbi.nlm.nih.gov/pubmed/24972847
http://dx.doi.org/10.1186/cc13951
_version_ 1782343725361397760
author Goto, Yoshikazu
Maeda, Tetsuo
Nakatsu-Goto, Yumiko
author_facet Goto, Yoshikazu
Maeda, Tetsuo
Nakatsu-Goto, Yumiko
author_sort Goto, Yoshikazu
collection PubMed
description INTRODUCTION: At hospital arrival, early prognostication for children after out-of-hospital cardiac arrest (OHCA) might help clinicians formulate strategies, particularly in the emergency department. In this study, we aimed to develop a simple and generally applicable bedside tool for predicting outcomes in children after cardiac arrest. METHODS: We analyzed data of 5,379 children who had undergone OHCA. The data were extracted from a prospectively recorded, nationwide, Utstein-style Japanese database. The primary endpoint was survival with favorable neurological outcome (Cerebral Performance Category (CPC) scale categories 1 and 2) at 1 month after OHCA. We developed a decision tree prediction model by using data from a 2-year period (2008 to 2009, n = 3,693), and the data were validated using external data from 2010 (n = 1,686). RESULTS: Recursive partitioning analysis for 11 predictors in the development cohort indicated that the best single predictor for CPC 1 and 2 at 1 month was the prehospital return of spontaneous circulation (ROSC). The next predictor for children with prehospital ROSC was an initial shockable rhythm. For children without prehospital ROSC, the next best predictor was a witnessed arrest. Use of a simple decision tree prediction model permitted stratification into four outcome prediction groups: good (prehospital ROSC and initial shockable rhythm), moderately good (prehospital ROSC and initial nonshockable rhythm), poor (prehospital non-ROSC and witnessed arrest) and very poor (prehospital non-ROSC and unwitnessed arrest). By using this model, we identified patient groups ranging from 0.2% to 66.2% for 1-month CPC 1 and 2 probabilities. The validated decision tree prediction model demonstrated a sensitivity of 69.7% (95% confidence interval (CI) = 58.7% to 78.9%), a specificity of 95.2% (95% CI = 94.1% to 96.2%) and an area under the receiver operating characteristic curve of 0.88 (95% CI = 0.87 to 0.90) for predicting 1-month CPC 1 and 2. CONCLUSIONS: With our decision tree prediction model using three prehospital variables (prehospital ROSC, initial shockable rhythm and witnessed arrest), children can be readily stratified into four groups after OHCA. This simple prediction model for evaluating children after OHCA may provide clinicians with a practical bedside tool for counseling families and making management decisions soon after patient arrival at the hospital.
format Online
Article
Text
id pubmed-4227055
institution National Center for Biotechnology Information
language English
publishDate 2014
publisher BioMed Central
record_format MEDLINE/PubMed
spelling pubmed-42270552014-11-12 Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study Goto, Yoshikazu Maeda, Tetsuo Nakatsu-Goto, Yumiko Crit Care Research INTRODUCTION: At hospital arrival, early prognostication for children after out-of-hospital cardiac arrest (OHCA) might help clinicians formulate strategies, particularly in the emergency department. In this study, we aimed to develop a simple and generally applicable bedside tool for predicting outcomes in children after cardiac arrest. METHODS: We analyzed data of 5,379 children who had undergone OHCA. The data were extracted from a prospectively recorded, nationwide, Utstein-style Japanese database. The primary endpoint was survival with favorable neurological outcome (Cerebral Performance Category (CPC) scale categories 1 and 2) at 1 month after OHCA. We developed a decision tree prediction model by using data from a 2-year period (2008 to 2009, n = 3,693), and the data were validated using external data from 2010 (n = 1,686). RESULTS: Recursive partitioning analysis for 11 predictors in the development cohort indicated that the best single predictor for CPC 1 and 2 at 1 month was the prehospital return of spontaneous circulation (ROSC). The next predictor for children with prehospital ROSC was an initial shockable rhythm. For children without prehospital ROSC, the next best predictor was a witnessed arrest. Use of a simple decision tree prediction model permitted stratification into four outcome prediction groups: good (prehospital ROSC and initial shockable rhythm), moderately good (prehospital ROSC and initial nonshockable rhythm), poor (prehospital non-ROSC and witnessed arrest) and very poor (prehospital non-ROSC and unwitnessed arrest). By using this model, we identified patient groups ranging from 0.2% to 66.2% for 1-month CPC 1 and 2 probabilities. The validated decision tree prediction model demonstrated a sensitivity of 69.7% (95% confidence interval (CI) = 58.7% to 78.9%), a specificity of 95.2% (95% CI = 94.1% to 96.2%) and an area under the receiver operating characteristic curve of 0.88 (95% CI = 0.87 to 0.90) for predicting 1-month CPC 1 and 2. CONCLUSIONS: With our decision tree prediction model using three prehospital variables (prehospital ROSC, initial shockable rhythm and witnessed arrest), children can be readily stratified into four groups after OHCA. This simple prediction model for evaluating children after OHCA may provide clinicians with a practical bedside tool for counseling families and making management decisions soon after patient arrival at the hospital. BioMed Central 2014 2014-06-27 /pmc/articles/PMC4227055/ /pubmed/24972847 http://dx.doi.org/10.1186/cc13951 Text en Copyright © 2014 Goto et al.; licensee BioMed Central Ltd. http://creativecommons.org/licenses/by/4.0 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
spellingShingle Research
Goto, Yoshikazu
Maeda, Tetsuo
Nakatsu-Goto, Yumiko
Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title_full Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title_fullStr Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title_full_unstemmed Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title_short Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
title_sort decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study
topic Research
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4227055/
https://www.ncbi.nlm.nih.gov/pubmed/24972847
http://dx.doi.org/10.1186/cc13951
work_keys_str_mv AT gotoyoshikazu decisiontreemodelforpredictinglongtermoutcomesinchildrenwithoutofhospitalcardiacarrestanationwidepopulationbasedobservationalstudy
AT maedatetsuo decisiontreemodelforpredictinglongtermoutcomesinchildrenwithoutofhospitalcardiacarrestanationwidepopulationbasedobservationalstudy
AT nakatsugotoyumiko decisiontreemodelforpredictinglongtermoutcomesinchildrenwithoutofhospitalcardiacarrestanationwidepopulationbasedobservationalstudy