Statistics from Altmetric.com
Sir,—We read with interest O’Sullivanet al’s article “Abnormal ventricular activation and repolarisation during dobutamine stress echocardiography in coronary artery disease”.1 In that study patients with coronary artery disease (CAD) had an increased QTc interval but not QT dispersion during positive dobutamine exercise stress testing. QT interval is a measure of the duration of ventricular repolarisation,2 and is sensitive to myocardial ischaemia. Therefore, the authors suggest that there are repolarisation abnormalities in patients with CAD during positive dobutamine stress testing. The interlead variation of QT interval in the 12 lead ECG referred to as QT dispersion reflects heterogeneity in the duration of myocardial repolarisation.2 O’Sullivanet al showed that QT dispersion was not heart rate dependent and did not change with increasing heart rate in control subjects. They found that QT dispersion decreased, but not significantly, during positive dobutamine stress testing in patients with CAD; they did not discuss QT dispersion as a marker of repolarisation abnormalities.
In contrast to O’Sullivan et al, many investigators have found increased QT dispersion during exercise induced ischaemia in patients with CAD.3-5 In one study, myocardial ischaemia induced by incremental atrial pacing caused an acute increase in QT dispersion.3 Furthermore, increased QT dispersion was found during positive stress testing in patients with CAD.4 5 We also performed a study related to QT dispersion during positive exercise test in patients with CAD. Our study group comprised 58 patients with CAD and 40 healthy volunteers. We found that QT dispersion at rest was profoundly increased in the CAD group compared with controls, and the percentage change in QT dispersion with exercise was significantly higher in the CAD group (table 1).
Our understanding from these studies under ischaemic conditions, is that non-homogeneity of repolarisation, and therefore QT dispersion, would be expected to increase further. O’Sullivanet al found mean QT dispersion values in patients with CAD and healthy subjects of 98 ms and 82 ms, respectively. Data from many studies suggest that QT dispersion at rest varies from 30–60 ms in healthy subjects and 60–80 ms in patients with CAD.2 Therefore, O’Sullivan et al’s findings are in conflict with most other studies and further work is required to investigate the effects of stress test induced ischaemia on QT dispersion.
This letter was shown to the authors, who reply as follows:
We thank Dr Aytemir and colleagues for their comments. In fact, effects on QT dispersion formed only a minor part of our study, and we note that there are differences between our results and those subsequently appearing in the literature. We believe that these differences are methodological. As we stated, it was our aim to determine all ECG intervals objectively using previously developed algorithms on digitised records. In all three studies quoted by Aytemiret al, measurements were made manually from the ECG as printed in analogue form, and leads were arbitrarily excluded from the analysis. As we included data from all 12 leads in all subjects studied, it is not surprising that we obtained larger values for QT dispersion.
While QT dispersion measurements appear to have shown some promise in predicting ventricular tachycardia or mortality after a previous myocardial infarction, this is not a uniform finding.1-1 The use of QT dispersion is based on the idea that spatial differences in depolarisation within the heart translate into differences between ECG leads. Theoretical backing for this idea is lacking, and convincing evidence that apparent QT dispersion is anything other than the instantaneous electrical vector being perpendicular to the lead in question has not been produced. While the possibility of dispersion of depolarisation within the myocardium remains an important one, we concur with recently expressed opinion1-2 that methods in current use based on QT dispersion are not reliable enough to detect it. Nevertheless, if it were possible to demonstrate a correlation between dispersion of mechanical activity, which can readily be detected by simple non-invasive methods, and QT dispersion, this might go some way towards validating the electrocardiographic approach. We were unable to do so.