Prediction of cardiac arrest in critically ill patients presenting to the emergency department using a machine learning score incorporating heart rate variability compared with the modified early warning score

The emergency department (ED) serves as the first point of hospital contact for many septic patients, where risk-stratification would be invaluable. We devised a combination model incorporating demographic, clinical, and heart rate variability (HRV) parameters, alongside individual variables of the Sequential Organ Failure Assessment (SOFA), Acute Physiology and Chronic Health Evaluation II (APACHE II), and Mortality in Emergency Department Sepsis (MEDS) scores for mortality risk-stratification. ED patients fulfilling systemic inflammatory response syndrome criteria were recruited. National Early Warning Score (NEWS), Modified Early Warning Score (MEWS), quick SOFA (qSOFA), SOFA, APACHE II, and MEDS scores were calculated. For the prediction of 30-day in-hospital mortality, combination model performed with an area under the receiver operating characteristic curve of 0.91 (95% confidence interval (CI): 0.88–0.95), outperforming NEWS (0.70, 95% CI: 0.63–0.77), MEWS (0.61, 95% CI 0.53–0.69), qSOFA (0.70, 95% CI 0.63–0.77), SOFA (0.74, 95% CI: 0.67–0.80), APACHE II (0.76, 95% CI: 0.69–0.82), and MEDS scores (0.86, 95% CI: 0.81–0.90). The combination model had an optimal sensitivity and specificity of 91.4% (95% CI: 81.6–96.5%) and 77.9% (95% CI: 72.6–82.4%), respectively. A combination model incorporating clinical, HRV, and disease severity score variables showed superior predictive ability for the mortality risk-stratification of septic patients presenting at the ED.

Download Full-text

Poor performance of the modified early warning score for predicting mortality in critically ill patients presenting to an emergency department

World Journal of Emergency Medicine ◽

10.5847/wjem.j.issn.1920-8642.2013.04.005 ◽

2013 ◽

Vol 4 (4) ◽

pp. 273 ◽

Cited By ~ 17

Author(s):

Le Onn Ho ◽

Huihua Li ◽

Nur Shahidah ◽

Zhi Xiong Koh ◽

Papia Sultana ◽

...

Keyword(s):

Emergency Department ◽

Critically Ill ◽

Early Warning ◽

Critically Ill Patients ◽

Poor Performance ◽

Early Warning Score

Download Full-text

Attitudes of emergency department physicians and nurses toward implementation of an early warning score to identify critically ill patients: qualitative explanations for failed implementation

CJEM ◽

10.1017/cem.2018.392 ◽

2018 ◽

Vol 21 (2) ◽

pp. 269-273 ◽

Cited By ~ 2

Author(s):

Blair L. Bigham ◽

Teresa Chan ◽

Steven Skitch ◽

Alison Fox-Robichaud

Keyword(s):

Emergency Department ◽

Critically Ill ◽

Early Warning ◽

Critically Ill Patients ◽

Cultural Barriers ◽

Constructivist Grounded Theory ◽

Early Warning Score ◽

Theory Approach ◽

Structured Interviews ◽

Common Time

AbstractBackgroundSepsis, a common, time-sensitive condition, is sometimes not identified at emergency department (ED) triage. The use of early warning scores has been shown to improve sepsis-related screening in other settings.ObjectivesOur objective was to elucidate nurse and physician perceptions with the Hamilton Early Warning Score (HEWS) in combination with the Canadian Triage Acuity Scale.MethodSemi-structured interviews were conducted with nurses, resident physicians and attending physicians to explore perceived feasibility, utility, comfort, barriers, successes, opportunities and accuracy. A constructivist grounded theory approach was used. Transcripts were coded into thematic coding trees.ResultsThe twelve participants did not value the HEWS in the ED because they felt it was not helpful in identifying critically ill patients. We identified five themes; knowledge of sepsis and HEWS, utility of HEWS in emergency triage, utility of HEWS at the bedside, utility in communicating acuity and deterioration, and feasibility and accuracy of data collection. We also found 9 barriers and 7 enablers to the use of early warning score in the ED.ConclusionsIn our emergency departments, we identified potential barriers to implementation of an early warning score. A pre-existing expertise and lexicon related to critically ill patients lessens the perceived utility of an EWS in the ED. Understanding these cultural barriers needs to be addressed through change theory and implementation science.

Download Full-text

Mortality prediction in critically ill patients using machine learning score

IOP Conference Series Materials Science and Engineering ◽

10.1088/1757-899x/788/1/012029 ◽

2020 ◽

Vol 788 ◽

pp. 012029

Author(s):

F Dzaharudin ◽

A M Ralib ◽

U K Jamaludin ◽

M B M Nor ◽

A Tumian ◽

...

Keyword(s):

Machine Learning ◽

Critically Ill ◽

Critically Ill Patients ◽

Mortality Prediction ◽

Learning Score

Download Full-text

News, Heart Rate Variability and Early Warning Score for Oxygen Therapy (EWS.O2): Predictors of Poor Outcomes in ICU Patient Under Spontaneous Ventilation?

10.1164/ajrccm-conference.2020.201.1_meetingabstracts.a1145 ◽

2020 ◽

Author(s):

F. Lellouche ◽

N. Ferrière ◽

L. Bodenes ◽

V. Pateau ◽

J. Doulou ◽

...

Keyword(s):

Heart Rate ◽

Heart Rate Variability ◽

Early Warning ◽

Oxygen Therapy ◽

Spontaneous Ventilation ◽

Early Warning Score ◽

Poor Outcomes

Download Full-text

Evaluation of peripheral perfusion index and heart rate variability as early predictors for intradialytic hypotension in critically ill patients

BMC Anesthesiology ◽

10.1186/s12871-019-0917-1 ◽

2019 ◽

Vol 19 (1) ◽

Author(s):

Hanan Mostafa ◽

Mohamed Shaban ◽

Ahmed Hasanin ◽

Hassan Mohamed ◽

Shymaa Fathy ◽

...

Keyword(s):

Heart Rate ◽

Heart Rate Variability ◽

Pulmonary Oedema ◽

Critically Ill ◽

Critically Ill Patients ◽

Perfusion Index ◽

Intradialytic Hypotension ◽

Peripheral Perfusion ◽

Hemodialysis Session ◽

Prophylactic Measures

Abstract Background Intradialytic hypotension is a serious complication during renal replacement therapy in critically ill patients. Early prediction of intradialytic hypotension could allow adequate prophylactic measures. In this study we evaluated the ability of peripheral perfusion index (PPI) and heart rate variability (HRV) to predict intradialytic hypotension. Methods A prospective observational study included 36 critically ill patients with acute kidney injury during their first session of intermittent hemodialysis. In addition to basic vital signs, PPI was measured using Radical-7 (Masimo) device. Electrical cardiometry (ICON) device was used for measuring cardiac output, systemic vascular resistance, and HRV. All hemodynamic values were recorded at the following time points: 30 min before the hemodialysis session, 15 min before the start of hemodialysis session, every 5 min during the session, and 15 min after the conclusion of the session. The ability of all variables to predict intradialytic hypotension was assessed through area under receiver operating characteristic (AUROC) curve calculation. Results Twenty-three patients (64%) had intradialytic hypotension. Patients with pulmonary oedema showed higher risk for development of intradialytic hypotension {Odds ratio (95% CI): 13.75(1.4–136)}. Each of baseline HRV, and baseline PPI showed good predictive properties for intradialytic hypotension {AUROC (95% CI): 0.761(0.59–0.88)}, and 0.721(0.547–0.857)} respectively. Conclusions Each of low PPI, low HRV, and the presence of pulmonary oedema are good predictors of intradialytic hypotension.

Download Full-text

Heart Rate Variability: Utility in Critically Ill Patients?

10.1164/ajrccm-conference.2020.201.1_meetingabstracts.a1772 ◽

2020 ◽

Author(s):

G.L. Jones ◽

V. Patel ◽

A. Achunair ◽

D.J. Patel ◽

J. Chiong ◽

...

Keyword(s):

Heart Rate ◽

Heart Rate Variability ◽

Critically Ill ◽

Critically Ill Patients

Download Full-text

Using machine learning tools to predict outcomes for emergency department intensive care unit patients

Scientific Reports ◽

10.1038/s41598-020-77548-3 ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Qiangrong Zhai ◽

Zi Lin ◽

Hongxia Ge ◽

Yang Liang ◽

Nan Li ◽

...

Keyword(s):

Machine Learning ◽

Intensive Care Unit ◽

Emergency Department ◽

Intensive Care ◽

Critically Ill ◽

Critically Ill Patients ◽

Predictive Accuracy ◽

Medical Center ◽

Learning Tools ◽

Critical Patients

AbstractThe number of critically ill patients has increased globally along with the rise in emergency visits. Mortality prediction for critical patients is vital for emergency care, which affects the distribution of emergency resources. Traditional scoring systems are designed for all emergency patients using a classic mathematical method, but risk factors in critically ill patients have complex interactions, so traditional scoring cannot as readily apply to them. As an accurate model for predicting the mortality of emergency department critically ill patients is lacking, this study’s objective was to develop a scoring system using machine learning optimized for the unique case of critical patients in emergency departments. We conducted a retrospective cohort study in a tertiary medical center in Beijing, China. Patients over 16 years old were included if they were alive when they entered the emergency department intensive care unit system from February 2015 and December 2015. Mortality up to 7 days after admission into the emergency department was considered as the primary outcome, and 1624 cases were included to derive the models. Prospective factors included previous diseases, physiologic parameters, and laboratory results. Several machine learning tools were built for 7-day mortality using these factors, for which their predictive accuracy (sensitivity and specificity) was evaluated by area under the curve (AUC). The AUCs were 0.794, 0.840, 0.849 and 0.822 respectively, for the SVM, GBDT, XGBoost and logistic regression model. In comparison with the SAPS 3 model (AUC = 0.826), the discriminatory capability of the newer machine learning methods, XGBoost in particular, is demonstrated to be more reliable for predicting outcomes for emergency department intensive care unit patients.

Download Full-text