Improved electrocardiographic criteria for confirmation of acute myocardial infarction with application in pre-hospital thrombolysis

Prehospital thrombolysis, while associated with improved prognosis in acute myocardial infarction (MI) patients, has limited applicability because of strict electrocardiographic criteria (sum ST elevation at least 1.0 mV). New criteria were developed in a database of computerised ECG's to increase sensitivity of the diagnostic system. Prior to hospital admission standard 12-lead ECG's were recorded with Siemens' SICARD P+ system in 658 consecutive patients with complaints suggestive of infarction (the training set). Mean age of the patients was 64 years, 59% were men. Symptom onset was less than 3 hours in 81%. In 243 patients a final diagnosis of MI was confirmed inhospital. Computerised measurements of the mean beats in every lead were used for logistic regression analysis predicting infarction. The final model included only measures of ST voltage. The findings were confirmed in 414 other patients (the validation set): 128 with proven MI. The combination of sensitivity and specificity was highly dependent on the total sum of ST elevation: 56%95%at 0.6 mV and 35%99%at 1.0 mV. We conclude that (a) sensitivity can hardly be improved without decreasing specificity. However, (b) the number of patients eligible for out of hospital treatment with thrombolysis may safely be increased.