Acute cardiac care & resuscitationCohort study
Predicting recurrent cardiac arrest within one year after surviving in-hospital cardiac arrest using a machine learning model
- Extreme gradient boosting (XGBoost) model was trained to predict the outcome
- Swedish registry for cardiopulmonary resuscitation (SRCR)
- 2010 to 2021
Authors' conclusion
In this dataset, an XGBoost model could predict recurrent cardiac arrest or death within one year after IHCA with a modest performance and good accuracy. If these results can be validated, this model could potentially be used clinically to assess the risk of another cardiac arrest in survivors of IHCA.
Study context
- Population
- Patients (>18 year) discharged alive after in-hospital cardiac arrest (IHCA)
- Participants
- 7,302 included patients
Main finding
As reported in the abstract
Recurrent cardiac arrest or death · within one year after IHCA
Patients who had developed the outcome
22%
Observational data: associations, not effects.
Further findings
Model performance (receiver operating characteristics and area under the curve (ROC-AUC))
- Best performing model with 1241 variables, ROC-AUC
- 0.72 (95% CI 0.70-0.75)
Feature importance
- Features of greatest importance were highest serum creatinine measured before IHCA, age and unknown civil status
- Comorbidity and cardiac arrest circumstances were important classes of features
Original publication
Predicting recurrent cardiac arrest within one year after surviving in-hospital cardiac arrest using a machine learning model. Resuscitation. Published 6 February 2026. doi:10.1016/j.resuscitation.2026.111006
Summarised from the published abstract; every number is checked against it