REPRODUCIBILITY AND AUTOMATIC-MEASUREMENT OF QT DISPERSION

This study investigated interobserver (two observers) and intrasubject (two measurements) reproducibility of QT dispersion from abnormal ele ctrocardiograms in patients with previous myocardial infarction, and c ompared a user-interactive with an automatic measurement system. Stand ard 12-lead electrocardiograms, recorded at 25 mm.s(-1) randomly chose n from 70 patients following myocardial infarction. These were scanned into a personal computer, and specially designed software skeletonize d and joined each image. The images were then available for user-inter active (mouse and computer screen), or automatic measurements using a specially designed algorithm. For all methods reproducibility of the R R interval was excellent (mean absolute errors 3-4 ms, relative errors 0 . 3-0 . 5%). Reproducibility of the mean QT interval was good; intr asubject error was 6 ms (relative error 1 . 4%), interobserver error w as 7 ms (1 . 8%), acid observers' vs automatic measurement errors were 10 and 11 ms (25, 2 . 8%). However QTc dispersion measurements had la rge errors for all methods; intrasubject error was 12 ms (17 . 3%), in terobserver error was 15 ms (22 . 1%), and observers' vs automatic mea surement were errors 30 and 28 ms (35 . 4, 31 . 9%). QT dispersion mea surements rely on the most difficult to measure QT intervals, resultin g in a problem of reproducibility. Any automatic system must not only recognize common T wave morphologies, but also these more difficult T waves, if it is to be useful for measuring QT dispersion. The poor rep roducibility of QT dispersion limits its role as a useful clinical too l, particularly as a predictor of events.