Evaluation of a paired creatine kinase test for the diagnosis of acute myocardial infarction in patients with a non-diagnostic electrocardiogram. Issue 3 (May 1997)

• Record Type: Journal Article
• Authors: Hingorani, A D; O'Hanlon, S; Halloran, S P; Wright, J P; Foley, T H
• Is Part Of: Journal of accident & emergency medicine. Volume 14:Issue 3(1997)
• Page Start: 134
• Page End: 138
• Publication Date: 1997-05
• Subjects: Emergency medicine -- Periodicals; 616.025
• DOI: 10.1136/emj.14.3.134
• Languages: English
• ISSNs: 2633-6138

Abstract

Abstract : OBJECTIVE: The rate of rise of total plasma creatine kinase (CK) activity in the first 12 hours from presentation can be used to diagnose acute myocardial infarction. The aim of this study was to evaluate the performance of an abbreviated form of this test in the diagnosis of acute myocardial infarction in patients in whom the initial electrocardiogram was inconclusive. METHODS: Using a protocol that requires only two CK measurements (separated by four hours) to estimate the rate of rise, the performance of the test was investigated using data accrued from 345 consecutive admissions with suspected acute myocardial infarction. RESULTS: A CK increment (delta CK) of > 20% in the first four hours from presentation had a diagnostic sensitivity of 84.4% (95% confidence interval 75.5 to 93.3), specificity of 85.8% (80.1 to 91.5), positive predictive accuracy of 73.0% (62.9 to 83.1), and negative predictive accuracy of 92.4% (87.9 to 96.9). Using more stringent diagnostic criteria (delta CK > 20% and 4 h CK value > 160 U/litre) resulted in an increase in specificity and positive predictive accuracy to 96.5% and 91.1% respectively, and a small reduction in sensitivity and negative predictive accuracy to 79.7% and 91.3%, respectively, 94% of all infarcts were correctly identified using the ECG as the initial investigation and paired CK measurement as an additional test when this was inconclusive. In the 44 patients who received thrombolysis on the basis of an early biochemical diagnosis of acute myocardial infarction, the median time delay (75th centile) to thrombolysis was 10.75 (SD 15.0) hours. CONCLUSIONS: When the presenting ECG is non-diagnostic, sequential sampling of cardiac enzymes is a feasible alternative in the early diagnosis of patients with suspected myocardial infarction, even in the emergency setting. Further studies are required to define the optimal biochemical assay and timed sampling protocol.