The consequences of reviewing serial changes in the Minnesota code for diagnosis of acute myocardial infarction

We investigated the effect of reviewing serial electrocardiograms (ECGs) for significance of changes in the Minnesota code, and the influence of this effect on the MONICA diagnosis of myocardial infarction (MI). We used the data from 1340 consecutive admissions to the coronary care units of our MONICA centre and we developed ECG classifications and MONICA diagnoses that were based once on the unreviewed Minnesota code and once on the reviewed code. A comparison of the two ECG classifications showed that codes for evolution of Q-waves and injury currents were much more likely to change as a result of a review (46.8%) than codes for evolution of repolarization changes (12.8%). The review of serial Minnesota codes caused a change of the MONICA diagnosis primarily in the category definite MI (10%). In a blinded clinical assessment, the cases that changed diagnosis were judged to be different from those that remained constant. It is concluded that the use of ECG classification based on unreviewed Minnesota code changes introduces heterogeneity. The significance of this effect depends on the use of the results.