The Western Washington Intracoronary Streptokinase Trial

We evaluated the relationship between baseline factors defined at 4.6 ± 2.1 hr after onset of acute myocardial infarction and 1 year survival in 245 patients entered in the Western Washington Intracoronary Streptokinase Trial. Univariate statistics identified a significant relationship between 10 of these factors and survival. Multivariate analysis identified three factors as being most closely related to survival: (1) left ventricular ejection fraction (LVEF) (p < .0001), (2) treatment with streptokinase (p = .03), and (3) location of infarction (p = .04). Mathematic models based on this analysis and applied to our patients identified highand low-risk subgroups for 1 year mortality. Patients receiving standard, not interventional, therapy with anterior infarction and an LVEF of 50% or less and those with inferior infarction and an LVEF of 39% or less comprised the high-risk group. For patients receiving standard therapy, 1 year mortality was 41% in the high-risk group and 4% in the lowrisk group. The models illustrated the magnitude of benefit of streptokinase treatment and achievement of complete reperfusion for those at low and high risk. We conclude that LVEF determined in the first hours of acute myocardial infarction is the most important of all baseline factors for prediction of 1 year survival. Mathematic models based on left ventricular function measured as ejection fraction are useful for risk stratification in this setting. Circulation 74, No. 4, 703-711, 1986. THE RISK for recurrent cardiac events in the year after an acute myocardial infarction has been the subject of many investigations.-''0 These studies have demonstrated that certain patient characteristics permit the definition of highand low-risk subgroups for recurrent cardiac events. Typically, risk assessment was performed at the time of hospital discharge,3-10 although several groups have applied the concept of risk assessment during the first 24 hr of hospitalization.' 2 Interventional therapies in the first hours of acute From the Department of Medicine, Division of Cardiology, and the Department of Biostatistics, University of Washington School of Medicine, and the Seattle Veterans Administration Medical Center, Seattle. Dr. Stadius was supported in part by a Research Fellowship from the Washington Affiliate of the American Heart Association and is a recipient of the Clinician-Scientist Award of the American Heart Association, Dallas. Address for correspondence: J. Ward Kennedy, M.D., Division of Cardiology (RG-22), University Hospital, 1959 NE Pacific St., Seattle, WA 98195. Received April 3, 1986; revision accepted July 17, 1986. *Current address: Cardiology Division, Stanford University Medical Center, Stanford, CA 94305. Vol. 74, No. 4, October 1986 myocardial infarction have been applied in hopes of decreasing infarct size and improving subsequent survival. 11-16 Because the myocardial infarction process is progressive over time, 171 1 patients must be selected for interventional therapy at the time of hospitalization. The concept of risk stratification may also be applicable in this setting. If highand low-risk subgroups for subsequent mortality can be defined at the time of hospitalization, the relative risk for subsequent mortality in a given patient can be balanced against the risks and potential benefits of interventional therapy. In this report, we evaluate the relationship between 1 year survival outcome and patient historical, clinical, and angiographic characteristics identified a mean of 4.6 ± 2. 1 hr after the onset of acute myocardial infarction in 245 patients. From this analysis, models defining probability of 1 year survival are developed. Application of these models in our population demonstrates their usefulness as prognostic devices for risk stratification in the first hours of acute myocardial infarction.