Predictive models of in-hospital deterioration of Society of Cardiovascular Angiography and Intervention shock stage in patients with acute myocardial infarction initially presenting with stable hemodynamic condition

Background: The Society of Cardiovascular Angiography and Intervention (SCAI) has defined 5 stages of cardiogenic shock (CS). In patients with acute myocardial infarction (AMI) who initially present in stable hemodynamic condition (SCAI CS stage: A or B), CS stages could deteriorate despite therapeutic management. However, deterioration of SCAI CS stages after AMI remains to be fully characterized. Therefore, the current study sought to investigate the frequency and clinical characteristics about deterioration of SCAI CS stages after AMI.

Methods: We retrospectively analyzed 347 patients in a derivation cohort and 163 patients in a validation cohort who had AMI (SCAI shock stage upon arrival: A/B) and underwent percutaneous coronary intervention (PCI) at National Cerebral and Cardiovascular Center, Suita, Japan (enrolment period of study subjects: 2019.07.01-2022.09.30). Deterioration of CS (D-CS) was defined as SCAI shock stage C-E after PCI. Clinical characteristics and in-hospital mortality were compared according to D-CS status. Adjusted hazard ratios (HRs) for in-hospital mortality were calculated with multivariate Cox proportional hazards models that included variables with P<0.10 in univariate models. Uni- and multivariate logistic regression analyses were used to identify predictors of D-CS.

Results: D-CS occurred in 17.3% (60/347) of the derivation cohort. Patients with D-CS had lower systolic blood pressure (BP) (P<0.001) and left ventricular ejection fraction (LVEF) (P<0.001) upon arrival with a higher proportion of initial Thrombolysis in Myocardial Infarction (TIMI) grade flow 0 or 1 (P=0.002). During hospitalization (13.9±9.4 days), D-CS was associated with higher in-hospital mortality [adjusted HR, 12.95; 95% confidence interval (CI): 1.46-114.97; P=0.02]. Initial systolic BP, LVEF, and TIMI grade flow 0 or 1 independently predicted D-CS. The D-CS risk score including these variables satisfactorily predicted D-CS [area under the curve (AUC), 0.749; 95% CI: 0.651-0.848] and in-hospital mortality (AUC, 0.961; 95% CI: 0.914-1.000) in the validation cohort.

Conclusions: D-CS occurred in 17.3% of patients with AMI initially presenting in stable condition and increased the risk of in-hospital mortality. Our D-CS risk score (initial systolic BP, LVEF, and TIMI grade flow) could be helpful to predict D-CS.