SUMMARYBackground: The electrocardiographic (ECG) diagnosis of acute myocardial infarction (MI) should be improved. This might be done either by regarding all 24 aspects (both positive and negative leads), or a subset hereof (for example 19-lead ECG), of the conventional 12-lead ECG or by using additional electrodes. The purpose of the study was to investigate the accuracy of the different ECG methods in diagnosing acute ST-elevation MI.Methods: The study population consisted of 479 patients admitted to Lund University Hospital with acute chest pain. One conventional ECG plus leads V4R, V5R, V8, and V9 were recorded for each patient within 24 hours of admittance. Biochemical markers were used as the "gold standard" for diagnosis of MI. We measured STsegment elevations in the 12-lead, 16-lead, and 24-lead postadmission ECGs as well as in the 12-lead, 19-lead and 24-lead admission ECGs.Results: The sensitivity for detecting acute MI was 28% for the postadmission 12-lead ECG, 33% for the 16-lead ECG, and 37% for the 24-lead ECG. The specificities were 97%, 93%, and 95%, respectively. For admission ECGs, the sensitivity was 33% for the 12-lead ECG, 45% for the 19-lead ECG and 49% for the 24-lead ECG, with specificities of 97%, 96%, and 94%, respectively.Conclusions: The sensitivity for detecting acute MI was higher for the 16-, 19-and 24-lead ECGs than for the conventional 12-lead ECGs. Their specificity, however, was slightly lower. If increased sensitivity for detecting MI is desired, the 24-lead or 19-lead should be used since no additional electrodes are required.