Acute cardiac care & resuscitationCohort study
Prediction of survival after pediatric cardiac arrest using heart rate variability and machine learning
- retrospective study
- tertiary hospital
- random forest model
- Shapley values
Authors' conclusion
Analysis of HRV at 24 h after ROSC may serve as a strong predictor of 12-month survival after pediatric CA.
Study context
- Population
- Children with cardiac arrest (CA) who achieved return of spontaneous circulation (ROSC) and were admitted to the pediatric intensive care unit (PICU) of a tertiary hospital between 2012 and 2021male: 63.2%, median age: 2.5 [IQR: 0.4–8.0] years · survival 34 · death 42
- Participants
- 76 children
- Exposure
- Random forest model trained on heart rate variability (HRV) parameters from a 5-min electrocardiogram (ECG) segment at 24 h after CA, evaluated against 12-month survival or death
- survivaln = 34
- deathn = 42
- Outcome
- 12-month survival or death (mortality prediction)
- Follow-up
- 12 months
Main finding
As reported in the abstract
accuracy
77.6%
positive predictive value for mortality prediction
0.879
Observational data: associations, not effects.
Further findings
Feature importance (Shapley values)
- Most influential features: frequency-domain parameters total power and very-low frequency (VLF) power, with lower values associated with an increased probability of death
Original publication
Prediction of survival after pediatric cardiac arrest using heart rate variability and machine learning. Resuscitation. Published 21 January 2026. doi:10.1016/j.resuscitation.2026.110983
Summarised from the published abstract; every number is checked against it