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
Critical Care, 06/22/2012
Ong MEH et al. – The authors found Machine Learning (ML) scores to be more accurate than modified early warning score (MEWS) in predicting cardiac arrest within 72hours. There is potential to develop bedside devices for risk stratification based on cardiac arrest prediction.Methods
- The authors conducted a prospective observational study of critically ill patients (Patient Acuity Category Scale 1 and 2) in an emergency department of a tertiary hospital.
- At presentation, HRV parameters generated from a 5–minute electrocardiogram recording are incorporated with age and vital signs to generate the ML score for each patient.
- The patients are then followed up for outcomes of cardiac arrest or death.
- From June 2006 to June 2008, the authors enrolled 925 patients.
- The area under the receiver operating characteristic (AUROC) curve for ML scores in predicting cardiac arrest within 72hours is 0.781 as compared to 0.680 for MEWS (difference in AUROC (95% CI): 0.101 (0.006, 0.197)).
- As for in–hospital death, the area under the curve for ML score is 0.741 as compared to 0.693 for MEWS (difference in AUROC (95% CI): 0.048 (–0.023, 0.119).
- A cutoff ML score of [greater than or equal to] 60 predicted cardiac arrest with a sensitivity of 84.1%, specificity of 72.3% and NPV of 98.8%.
- A cutoff MEWS of [greater than or equal to] 3 predicted cardiac arrest with a sensitivity of 74.4%, specificity of 54.2% and NPV of 97.8%.