Current issues of ACP Journal Club are published in Annals of Internal Medicine


Rapid assay of creatine kinase MB subforms detected acute myocardial infarction

ACP J Club. 1995 Jan-Feb;122:21. doi:10.7326/ACPJC-1995-122-1-021

Source Citation

Puleo PR, Meyer D, Wathen C, et al. Use of a rapid assay of subforms of creatine kinase MB to diagnose or rule out acute myocardial infarction. N Engl J Med. Sep



To determine the accuracy of a rapid assay of subforms of creatine kinase MB (CK-MB) in diagnosing acute myocardial infarction (MI) in the first 6 hours after the onset of chest pain.


Blinded comparison of assay of CK-MB subforms with conventional CK-MB analysis, which was the diagnostic standard.


Emergency department of a hospital in Houston, Texas.


1110 consecutive patients (mean age, 51 y; 51% men) evaluated in the emergency room for chest pain occurring at rest during the previous 24 hours and lasting > 30 minutes. Patients who needed intravenous vasopressor ther-apy or mechanical ventilation, or who had been resuscitated after cardiac arrest, were excluded.

Description of Test and Diagnostic Standard

Blood samples were obtained every 30 to 60 minutes until ≥ 6 hours after the onset of chest pain, and then every 4 hours for up to 48 hours in patients admitted to the hospital. The samples were analyzed for CK-MB subforms and the diagnosis of MI was confirmed by conventional CK-MB analysis. On the basis of the subform analysis, MI was diagnosed when an MB2 level ≥ 1.0 U/L and a ratio of MB2 to MB1 ≥ 1.5 occurred in ≥ 1 sample taken within 6 hours after the onset of symptoms. On the basis of total CK-MB activity, the criteria for MI were a CK-MB level ≥ 14 U/L in ≥ 2 sequential blood samples obtained 6 to 12 hours apart; a single CK-MB level ≥ 14 U/L if this value was 3 times greater than the previous value; and a single CK-MB value ≥ 14 U/L if only 1 sample was analyzed.

Main Outcome Measures

Sensitivity, specificity, and likelihood ratios.

Main Results

The diagnosis of MI was confirmed in 121 patients. The sensitivity of the CK-MB subform assay in detecting MI within 6 hours after symptom onset was 97% (114 of 118 patients; 99% CI, 90% to 99%) compared with a sensitivity of 48% for conventional CK-MB assay. The specificity for subform assay was 94% among patients hospitalized with MI and 96% among those sent home. {Hence, likelihood ratios were 19.0 and 0.04 for positive and negative subform assays, respectively} . (Numbers calculated from data in article.).


A rapid assay of subforms of creatine kinase MB accurately detected myocardial infarction within the first 6 hours after the onset of symptoms.

Sources of funding: American Heart Association Bugher Foundation for Molecular Biology of the Cardiovascular System and National Heart, Lung, and Blood Institute.

For article reprint: Dr. R. Roberts, Baylor College of Medicine, 6535 Fannin, MS F905, Houston, TX 77030. FAX 713-790-4348.


The early determination of whether a patient presenting with chest pain is truly having an MI has 2 important ramifications, only 1 of which is addressed by Puleo and colleagues.

The authors rightly point out that resources are wasted when patients without MIs are placed in units designed to care for patients with MIs. Another important aspect of an early and accurate diagnosis of infarction is to allow for early treatment with thrombolytic drugs. This study clearly shows that CK-MB subforms detect MI within 6 hours after the onset of symptoms. It is clear that patients presenting with "big-bang" infarctions (i.e., significant ST-segment elevation) will need to receive thrombolytic drugs as early as possible and that in this group of patients an elevation of CK-MB subform would be of little value. The management of patients with a good history and equivocal electrocardiographic changes can be more difficult. Thus, although providing valuable information on the sensitivity and specificity of diagnosing MI, the assumption that the test will be clinically useful in all patients with MI may not be true.

It is interesting that the article made no mention of thrombolytic drugs. One wonders how many additional patients with an early increase in subform might have benefited from thrombolytic drugs given on the assumption that the early increase in subform suggested that they were having an MI. Whether electrocardiograms or subforms are better at predicting a benefit from thrombolytic drugs will need to be tested in a larger trial.

Many trials do suggest that patients without obvious MI have a good prognosis. Whether this prognosis can be further improved by this new and potentially more expensive test is unclear. How the test compares with other markers, such as tropanin T, is also unclear, but at least the authors have started the ball rolling.

A. H. Gershlick
University of Leicester Leicester, United Kingdom