T wave normalization during dobutamine stress testing in patients with non-Q wave myocardial infarction. A marker of myocardial ischaemia?

Abstract

Conflicting results in a heterogenous patient population have been described on the functional significance of stress-induced T wave normalization in the EGG. The aim of this study was to evaluate the relationship between T wave normalization during dobutamine stress testing and stress-induced ischaemia evaluated by echocardiography and myocardial perfusion scintigraphy in patients with previous non-Q wave myocardial infarction. Among 520 patients who underwent dobutamine stress testing in conjunction with simultaneous echocardiography and 201 thallium or sestamibi SPECT for evaluation of suspected myocardial ischaemia, 36 were selected according to the following criteria: previous non-Q wave myocardial infarction, normal QRS, negative T waves in two or more ECG leads and no significant ST segment depression or elevation at rest or during stress. Diagnosis of ischaemia relied upon the occurrence of reversible perfusion defects by scintigraphy and stress-induced wall motion abnormalities by echocardiography. During the test, T wave normalization (defined as a resting negative T wave becoming upright in one or more ECG leads during stress) occurred in 20 patients (group 1), while in 16 patients the T waves remained negative (group 2). The prevalence of ischaemia was higher in group 1 than in group 2 both by scintigraphy (85% vs 38%, P = 0.004) and by echocardiography (70% vs 32%, P = 0.02). The sensitivity, specificity and accuracy of T wave normalization in the detection of ischaemia were 74%, 77% and 75% by SPECT and 74%, 65% and 69% by echocardiography respectively. Conclusion In patients with non-Q wave myocardial infarction and suspected myocardial ischaemia, T wave normalization without concomitant ECG changes during dobutamine stress testing is associated with a higher prevalence of ischaemia compared to patients with persistent T wave inversion. This ECG finding should not be disregarded as a marker of ischaemia in that particular patient population.