Development and validation of a risk score in acute myocardial infarction-related cardiogenic shock

Aims Mortality in patients with acute myocardial infarction-related cardiogenic shock (AMICS) is high, but a widely accepted tool for individual risk assessment is lacking. A reliable prediction model could assist in clinical decision-making, patient selection for clinical trials, and comparison of AMICS populations. Therefore, the aim of this study was to develop and externally validate a prediction model for 30-day mortality in AMICS patients.Methods and results This retrospective cohort study included patients from 2017 to 2021 (development cohort) and 2010-2017 (validation cohort). Patients with AMICS undergoing percutaneous coronary intervention in The Netherlands were identified using the Netherlands Heart Registration. International validation was performed in the Danish Retroshock registry. The main outcome was 30-day mortality. Among 2261 patients, the median age was 67 years [interquartile range (IQR) 58-75], and 1649 (73%) were male. The mortality rate at 30 days was 39% (n = 886). Significant predictors for mortality were: initial lactate, glucose, renal function, haemoglobin, age, blood pressure, heart rate, intubation prior to PCI, intervention in the left main coronary artery, and successful revascularization. The AUC of the initial model was 0.81 (0.79-0.83). The external validation cohort included 1393 patients with 1050 (75%) male and a median age of 67 years (IQR 59-75). The 30-day mortality rate was 49% (n = 680). The model showed good performance on the external validation with an AUC of 0.73 (0.70-0.76).Conclusion A prediction model was developed and externally validated using data from two large national registries. The model demonstrated good performance and is suitable for clinical decision-making and quality purposes in AMICS.