17 September 2007
Objectives: Conventionally, the detection of prior myocardial infarction (MI) is based on QRS abnormalities, which may ignore non-Q-wave MI (NQMI). We aimed at finding automatically applicable quantitative ECG variables for diagnosing prior MI. Methods: Body surface potential mapping (BSPM) was registered and automatically analyzed in 144 patients with prior MI and in 75 healthy controls. The MI was defined according to its age as recent or old, and Q-wave status as Q-wave MI (QMI) or NQMI. Results: The QRSSTT integral, the STT integral and the T-wave apex amplitude applied in single, selected leads were found to be the optimal parameters in the detection of prior MI. The areas under the receiver-operating characteristic curves (AUC) were 89% for each, and detection was equal in old and recent MI (AUCs from 87 to 90%), and in QMI and NQMI (AUCs from 88 to 90%). Conclusions: The quantitative, automatically applicable single-lead variables comprising ventricular repolarization was effective in detecting prior MI, irrespective of the time elapsed from MI or the Q-wave status. These variables could be suitable for population studies and health screening purposes and are applicable to automatic ECG diagnostics of prior MI.