The diagnosis of acute myocardial infarction in the emergency department; Part 2.

At present, routine use of cardiac enzymes in the emergency department (ED) cannot be justified, except possibly as a final screen prior to discharge. Computer-derived predictive instruments do not surpass the physician's diagnostic sensitivity for acute myocardial infarction (AMI), but do demonstrate significantly higher specificity. Limited data exist on the utility of echocardiography and thallium scanning in the ED. Methods of triaging patients on the basis of prognosis are well supported in the literature. The physician's high diagnostic sensitivity is maintained at the cost of significant numbers of admissions who subsequently rule out for AMI. No single clinical variable or combination of clinical variables can reliably confirm or exclude AMI in the ED. Ultimately, the physician's clinical assessment must remain the final determinant of the necessity for admission. However, judicious use of prediction rules and prognostic indicators should improve resource utilization.