Prediction of cardiovascular death by means of clinical and exercise test variables in patients selected for cardiac catheterization.

The objective of this report is the development of a population-specific prediction rule based on clinical and exercise test data that would estimate the risk of cardiovascular death in patients selected for cardiac catheterization. Prospective data and follow-up information were obtained from patients who underwent cardiac catheterization soon after clinical assessment and exercise testing. Males (n = 588) referred for evaluation of coronary heart disease from 1984 to 1990 were selected after exclusion of patients with significant valvular heart disease and patients with prior cardiac surgery. Half had a prior myocardial infarction and half complained of typical angina pectoris. All patients performed a treadmill test and were selected for clinical reasons to undergo coronary angiography within 3 months. Over a mean follow-up period of 2.5 years (+/- 1.4 years), there were 39 cardiovascular deaths and 45 nonfatal myocardial infarctions. The Cox proportional hazards model demonstrated the following characteristics to be statistically significant independent predictors of time until cardiovascular death: history of congestive heart failure (hazards ratio of 4), ST depression on the resting ECG (hazards ratio of 3), and a drop in systolic blood pressure below the resting value during exercise (hazards ratio of 5). Exercise-induced ST depression was not associated with either death or nonfatal myocardial infarction. A simple score based on one item of clinical information (history of congestive heart failure), a resting ECG finding (ST depression), and an exercise test response (exertional hypotension) stratified our patients for 4 years after testing from 75% with a low risk (annual cardiac mortality rate of 1%), 17% with a moderate risk (annual mortality rate of 7%), and 1% with a high risk (annual cardiac mortality rate of 12%, with a hazards ratio of 20 and 95% confidence interval from 6 to 70X). It was concluded that the variables available from the usual noninvasive workup of patients with known or suspected coronary artery disease enable prediction of risk of cardiovascular death. Three quarters of those usually undergoing cardiac catheterization can be identified by simple noninvasive variables as being at such low risk that invasive intervention is unlikely to improve prognosis.