Acute cardiac care & resuscitationResearch article
A Deep Learning Model to Guide Personalized Mechanical Circulatory Support Use in Cardiogenic Shock Patients Undergoing Percutaneous Coronary Intervention (PCI)
- develop and validate a deep learning (DL) model
- multiple hospitals
Authors' conclusion
Among CS patients undergoing PCI, we developed and validated a DL model to predict outcomes and facilitate a patient-centric approach. If it can be externally validated and implemented in clinical practice, it has the potential to distinguish the risk-benefit balance of MCS devices, reduce adverse outcomes, improve survival, and ultimately advance CS care.
Study context
- Population
- Cardiogenic shock (CS) patients undergoing PCI treated with intra-aortic balloon pump (IABP) or microaxial flow-pump (mAFP), 2004–2019CS patients treated with IABP or mAFP
- Participants
- 1,408 patients
- Comparison
- OPtiMCS DL model predicting 30-day mortality, bleeding, acute kidney injury (AKI), 1-year mortality, and 1-year stroke; simulated alternate outcomes by switching between IABP and mAFP
Main finding
As reported in the abstract
Outcome rates: 30-day mortality, bleeding, AKI, 1-year mortality, 1-year stroke
- 30-day mortality
- 31%
- bleeding
- 43%
- AKI
- 35%
- 1-year mortality
- 35%
- 1-year stroke
- 5%
Further findings
OPtiMCS discrimination (AUC)
- AUC, AKI
- 83%
- AUC, mortality
- 98%
- Key predictive features identified (eg, cardiac arrest predicted mortality)
Simulated benefit of IABP switched to mAFP
- Predicted reduction in mortality
- 1.8%
- Predicted reduction in AKI
- 5.2%
- model simulation (eg)
Original publication
A Deep Learning Model to Guide Personalized Mechanical Circulatory Support Use in Cardiogenic Shock Patients Undergoing PCI. JACC Adv. doi:10.1016/j.jacadv.2025.102379
Summarised from the published abstract; every number is checked against it