Abstract Background Cardiogenic shock (CS) after acute myocardial infarction (AMI) is a life-threatening condition with an high rate of in-hospital mortality. Purpose This study aims to 1) describe predictors of in-hospital mortality; 2) evaluate ten years mortality temporal trend in our Cardiac Intensive Care Unit (CICU); 3) assess the feasibility of CARDSHOCK risk score in our population; 4) elaborate a simpler version of CARDSHOCK risk score. Methods All consecutive patients with CS after AMI admitted at our CICU from March 2012 to July 2021 were included in this single-centre retrospective study. Results We included 167 patients [males 67%; age 71 (61–80) years] with ischemic CS. Patients had severe LV dysfunction in 66%. Baseline serum lactate was 5.2 (3.1–8.8) mmol/L. All patients required inotropes: 71% required dopamine [mean dose 5.6 (2.4–11,3) mcg/kg/min], 65% required noradrenaline [mean dose 0.10 (0.05–0.18) mcg/kg/min], 32% required dobutamine [mean dose 4.5 (2.2–15.9) mcg/kg/min]; 17.4% received levosimendan alone [mean dose 0.1 mcg/kg/min]. Mechanical cardiac support (MCS) was pursued in 91.1% [65% IABP, 23% Impella CP, 4% VA-ECMO]. From March 2012 to July 2021 we observed a significative temporal trend mortality reduction (OR=0.90, 95% CI: 0.84–0.96, p=0.0015), in particular in-hospital mortality has reduced from 57% of first time-quartile to 29% of the fourth quartile (Figure 1). In addition we also noted a significant increase in Impella catheter use (p=0.0005) with a consequent reduction of IABP (p=0.01), a reduction in dopamine administration (p=0.0007) and a greater use of dobutamine and levosimendan (p=0.015 and p=0.0001). In our population of AMI-CS patients CARDSHOCK risk score was a reliable in-hospital mortality predictor tool (OR 1.11; 95% CI, 1.06–1.17; p=0.00011). After the multivariate analysis only ejection fraction (EF) at baseline (OR=0.99, 95% CI: 0.98–0.99, p=0.009), lactate level at presentation (OR=1.03, 95% CI: 1.01–1.06, p=0.015) and presence of three-vessels coronary artery disease (OR=0.73, 95% CI: 0.59–0.90, p=0.0038) resulted to be in-hospital mortality predictors. For this reason, a prediction model composed by those three variables was created which exhibited better predictive performance for in-hospital mortality than Cardshock risk score (AUC of 0.94 vs AUC of 0.72 respectively, p=0.015) (Figure 2). Conclusions In our retrospective single-centre study a significant reduction of mortality through the years is observed, probably due to more extensive use of micro axial pumps and better manipulation of inotropic drug therapies. The use of Cardshock risk score has been proven to be a feasible tool in prediction on in-hospital mortality also in our sample composed only of AMI-CS patients. In addition, a more simplified risk score made up of only three clinical variables demonstrates at least the same predictive performance. Future validation in a larger population could be advisable to validate the simplified score. Funding Acknowledgement Type of funding sources: None.