Comparison of ECG variables of dispersion of ventricular repolarization with direct myocardial repolarization measurements in the human heart

Abstract

Introduction: QT dispersion (QTD) from the 12-lead ECG has been widely adopted as a noninvasive index of dispersion of ventricular repolarization (DVR). QTD, however, has never been validated by direct comparison with myocardial DVR in the human heart.

Methods and results: Monophasic action potential (MAP) recordings obtained in an earlier study were retrospectively matched with 12-lead ECGs available from within 24 hours of the invasive procedure. MAPs were available from an average of 8+/-3 left endocardial sites in 4 patients with left ventricular hypertrophy (LVH) and 7 patients with normal ECGs, and 6+/-2 epicardial sites in 3 patients of each group during normal ventricular activation. Local repolarization time (RT) was determined as MAP duration at 90% repolarization plus the local activation time. Dispersion of RT was calculated as the difference between the earliest and latest RT. ECGs were digitized and analyzed with recently described interactive QTD analysis software. In addition to standard QTD (defined as QTmax-QTmin), all currently proposed ECG dispersion variables were compared and correlated with the invasive measurements of DVR. QTD exhibited a reasonable correlation with dispersion of RT (R = 0.67; P < 0.01). Several other variables designed to measure DVR exhibited a similar, but not better, correlation. Among them, the QT peak/QT end ratio in V3 (R = -0.72; P < 0.01) and averaged over all analyzable leads (R = -0.59; P < 0.01) exhibited a good correlation with dispersion of RT, which was further improved when endocardial measurements were considered alone. T area measures did not correlate with dispersion of RT, but discriminated LVH.

Conclusion: DVR can be assessed by means of a 12-lead surface ECG. Several of the variables under study exhibit a similar accuracy in determination of true myocardial dispersion of repolarization. Variables involving the terminal part of repolarization, such as the QT peak/QT ratio, even from a single lead, may add to the determination of DVR from the human heart.