STUDY OBJECTIVE: To analyze the effect of creatine kinase isoenzyme (CK-MB) results on decision making in the evaluation of emergency department patients with chest pain. DESIGN: Prospective, controlled observational study of clinical decision making. SETTING: EDs of two teaching hospitals, a Veterans Affairs medical center, and a medical school university hospital. TYPE OF PARTICIPANTS: Patients more than 30 years old complaining of chest discomfort warranting an ECG. Excluded were hemodynamically unstable patients and patients with ECG evidence of an acute myocardial infarction (AMI). INTERVENTIONS: After the initial assessment including ECG but not CK-MB data, physicians answered questions regarding estimated probability of AMI and disposition plans. CK-MB levels were drawn every hour for as long as three hours (from one to four CK-MBs) with results readily available. Physicians could admit their patients into the hospital at any time. At disposition after reviewing a second ECG and all available CK-MB data, the physicians answered the same questions and rank ordered the contribution to disposition of the following six factors: initial and serial clinical evaluations, initial and serial ECGs, and initial and serial CK-MB enzymes, respectively. The absolute log likelihood ratio test measured the contribution of CK-MB to decision certainty. MEASUREMENTS AND MAIN RESULTS: Three hundred seventy-six patients were studied: 29 (7.7%) with AMI and 347 without AMI (nonAMI). Physicians indicated that CK-MB levels were useful for more than one third of study patients. When considered useful, CK-MB results strengthened the impression of AMI in AMI patients and decreased the impression of AMI for nonAMI patients; CK-MB also correlated with the perceived need for cardiac care unit admission in AMI patients and with a reduced need to admit nonAMI patients. The use of CK-MB results did not significantly increase ED release rates. CONCLUSION: The rapid availability of serial CK-MB results appears to affect decision making in one third of ED patients with chest pain and nondiagnostic ECGs. CK-MB levels appear to complement clinical evaluation of the ED chest pain patient in a manner analogous to the ECG.
STUDY OBJECTIVE: To analyze the effect of creatine kinase isoenzyme (CK-MB) results on decision making in the evaluation of emergency department patients with chest pain. DESIGN: Prospective, controlled observational study of clinical decision making. SETTING: EDs of two teaching hospitals, a Veterans Affairs medical center, and a medical school university hospital. TYPE OF PARTICIPANTS: Patients more than 30 years old complaining of chest discomfort warranting an ECG. Excluded were hemodynamically unstable patients and patients with ECG evidence of an acute myocardial infarction (AMI). INTERVENTIONS: After the initial assessment including ECG but not CK-MB data, physicians answered questions regarding estimated probability of AMI and disposition plans. CK-MB levels were drawn every hour for as long as three hours (from one to four CK-MBs) with results readily available. Physicians could admit their patients into the hospital at any time. At disposition after reviewing a second ECG and all available CK-MB data, the physicians answered the same questions and rank ordered the contribution to disposition of the following six factors: initial and serial clinical evaluations, initial and serial ECGs, and initial and serial CK-MB enzymes, respectively. The absolute log likelihood ratio test measured the contribution of CK-MB to decision certainty. MEASUREMENTS AND MAIN RESULTS: Three hundred seventy-six patients were studied: 29 (7.7%) with AMI and 347 without AMI (nonAMI). Physicians indicated that CK-MB levels were useful for more than one third of study patients. When considered useful, CK-MB results strengthened the impression of AMI in AMI patients and decreased the impression of AMI for nonAMI patients; CK-MB also correlated with the perceived need for cardiac care unit admission in AMI patients and with a reduced need to admit nonAMI patients. The use of CK-MB results did not significantly increase ED release rates. CONCLUSION: The rapid availability of serial CK-MB results appears to affect decision making in one third of ED patients with chest pain and nondiagnostic ECGs. CK-MB levels appear to complement clinical evaluation of the ED chest painpatient in a manner analogous to the ECG.