New approaches to electrocardiographic evaluation of acute myocardial ischemia

Large interindividual variability in the electrocardiogram (ECG) interferes with the detection of changes caused by acute myocardial ischemia. Serial analysis, comparing the acute ECG with a reference ECG (a previously made non-acute ECG in the same patient), is a guideline recommended strategy for this problem. However, operational procedures for this comparison are lacking. This thesis focuses on the interpretation of serial differences in specific ECG variables (which we termed subtraction electrocardiography), for the detection of myocardial ischemia.
In the STAFF III database, in which acute ischemia was induced by elective coronary angioplasty, both ΔST (difference vector between the acute and reference ST vectors at the J point) and ΔVG (difference vector between the acute and reference ventricular gradient vectors) behaved similarly during myocardial ischemia and had increased sensitivity for its detection compared to STEMI criteria. Additionally, the onset QRS could act as an objective synchronization point for the comparison of two ECGs.
We found that the validity of reference ECGs decreases with time. Additionally, in apparently healthy individuals in the multi-ethnic HELIUS population, striking differences were found in the prevalence of routine ECGs that falsely qualify as evidence of ischemia across ethnicities, sexes and age groups.
ΔST and ΔVG performed well in detecting myocardial ischemia in a case-control patient mix composed from STAFF III patients and non-ischemic outpatient-clinic patients. In patients prehospitally presenting with acute complaints suggestive of myocardial ischemia, subtraction electrocardiography was tested using logistic regression of 28 QRST-difference descriptors (including ΔST and ΔVG), yielding an impressive diagnostic performance.