Predicting In-Hospital Mortality in Patients With Acute MI

Study Questions:

What clinical characteristics predict in-hospital mortality for patients with acute myocardial infarction (AMI)?


The authors developed and validated a clinical risk model of in-hospital mortality after AMI (non–ST-segment elevation [NSTEMI] and ST-segment elevation AMI [STEMI]) using data from the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry–Get With the Guidelines (GWTG) database from January 2012 through December 2013. The final cohort consisted of 243,440 patients and was randomly divided into a derivation cohort (60%; n = 145,952) and a validation cohort (40%; n = 97,288). Mortality was defined as all-cause mortality during hospitalization. Data collected included patient demographics; presenting features, including cardiac arrest; prehospital, in-hospital, and discharge therapies; laboratory tests; procedures; and in-hospital outcomes. Hierarchical logistic regression was used to generate a risk model.


The derivation and validation cohorts did not differ significantly, and each included 39% STEMI. In multivariable analysis, age, presenting heart rate, presenting blood pressure, presentation after cardiac arrest, presentation in cardiogenic shock, presentation in heart failure, STEMI, creatinine clearance, and troponin ratio were independently associated with in-hospital mortality. The final model had high discrimination in both the derivation and validation cohorts, with a C-statistic of 0.88 for both groups, and was used to develop an integer score based on the above variables. Mortality rates in patients with risk scores <30, 30-39, 40-49, 50-59, and >59 were 0.4%, 1.7%, 5.5%, 18.5%, and 49.5%, respectively. The model performed well in STEMI, NSTEMI, no cardiac arrest, and cardiac arrest subgroups (C-statistic 0.895, 0.851, 0.848, and 0.788, respectively).


The authors developed a focused, nine-variable risk model and score for risk stratification of in-hospital mortality for contemporary patients with AMI.


This risk model was derived and validated on a contemporary cohort of patients, the vast majority of whom underwent primary percutaneous coronary intervention for STEMI and received guideline-recommended pharmacological therapies. Overall, this model shows mortality is relatively low, with the exception of patients with very high risk features such as cardiac arrest and cardiogenic shock, and this score may be, in the longer term, more useful for evaluating mortality outcomes rather than aiding in clinical decision making. While at present this new risk model has only been validated using ACTION Registry-GWTG data (a voluntary registry), it is an important addition to other mortality risk models developed for AMI patients.

Clinical Topics: Acute Coronary Syndromes, Arrhythmias and Clinical EP, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Prevention, Implantable Devices, SCD/Ventricular Arrhythmias, Acute Heart Failure, Interventions and ACS

Keywords: ACTION Registry, Acute Coronary Syndrome, Blood Pressure, Creatinine, Heart Arrest, Heart Failure, Hospital Mortality, Myocardial Infarction, Percutaneous Coronary Intervention, Primary Prevention, Shock, Cardiogenic, Troponin

< Back to Listings