P1721Determination of a predictive score of cardiogenic shock in acute coronary syndrome

Abstract Introduction Cardiogenic shock is one of the leading causes of death in patients with acute coronary syndrome (ACS), reaching in-hospital mortality rates of 50%. Purpose To identify a predictive score of cardiogenic shock in patients with ACS. Methods We performed a retrospective, descriptive and correlational study encompassing patients admitted with ACS in a Cardiology service from 1st October 2010 to 1st October 2018. Demographic factors, risk factors, antecedents and clinical characteristics were analyzed. The correlation between the categorical variables was performed by the Chi-square test, while the T-Student test was applied to the continuous variables, with a significance level of 95%. Independent predictors of cardiogenic shock were identified through a binary logistic regression analysis, considering p=0,05. Then, a discriminatory function was applied using the Wilks lambda test to determine the discriminant score of the analized groups. Statistical analysis was conducted with SPSS 24.0. Results During this period, 4458 patients were admitted with ACS and 74 (1,7%) developed cardiogenic shock. In this subgroup, 59,5% were over 65 years of age, 63,5% were male and 93,2% presented with acute myocardial infarction with ST segment elevation. Also, 83,8% were in sinus rhythm at admission, 22,7% had creatinine>1,5 mg/dL and 17,9% had left ventricular ejection fraction (LVEF) <30%. The in-hospital mortality rate was 51,4%. LVEF <30% (p=0,018), creatinine>1,5 mg/dL (p=0,044) and absence of sinus rhythm at admission (p=0,041) were independent predictors of cardiogenic shock. A predictive score of this complication in patients with ACS was determined using the formula: 1,723 + 1,505 x (creatinine>1,5) + 4,483 x (LVEF <30%) – 2,094 x (sinus rhythm at admission). A cutoff of 0,58 was obtained with 44,4% sensitivity, 85,2% specificity and 85% discriminative power. Conclusion Cardiogenic shock occurred in 1,7% of patients admitted with ACS and was associated with a high mortality rate. We determined a predictive score of this complication with a good discriminative power, which included LVEF <30%, creatinine >1,5 mg/dL and the rhythm on admission's electrocardiogram. By taking into account clinical variables, this score can be used at a very early stage of admission, allowing risk stratification of developing cardiogenic shock in each patient. However, it needs validation to be applied in clinical practice.