Background: Application of the traditional diagnostic criteria validated for exercise testing may be inappropriate when dobutamine infusion is associated with radionuclide ventriculography (RNV). The objective of this study was to establish appropriate diagnostic criteria for evaluation of dobutamine stress testing with RNV for the detection of early postinfarction ischemia.
Methods: RNV was performed at baseline and during dobutamine infusion in 10 control subjects and in 30 patients who were studied within 1 week after uncomplicated myocardial infarction. Several quantitative parameters including left ventricular ejection fraction (EF), regional EF, and absolute change in global and regional EF were tested. In addition, regional wall-motion changes were scored by visual analysis. The limit of normal response for each quantitative parameter was defined as the 95th percentile of the distribution observed in control subjects. On the basis of predischarge clinical evaluation and exercise stress testing, patients were grouped as having evidence of residual ischemia (group 1, 15 patients) or no evidence of ischemia (group 2, 15 patients). Sensitivity, specificity, and accuracy in the detection of postinfarction ischemia were calculated for each parameter, and empiric receiver-operating characteristic curves were generated.
Results: The limits of the normal response to dobutamine infusion in the control subjects were found to be a 7.5% and a 12.5% increase in global EF at low and high dose, respectively. Median baseline EF was significantly lower in patients than in control subjects (Mann-Whitney U test: P < .001). There were no differences in resting EF between group 1 and group 2 patients. However, median high-dose EF and increase in EF were significantly lower in group 1 as compared with group 2 patients (56% vs 72% and 5% vs 17%, respectively; Mann-Whitney U test: P < .0001 for both). A biphasic (up-and-down) response with initial increase at low dose followed by decrease or no change in EF at high dose was observed in 66% of group 1 patients and only in 13% of group 2 subjects. The best criterion for detection of postinfarction ischemia was a change of <12.5% in global EF at high-dose dobutamine infusion, with 80% predictive accuracy. The traditional criterion, a change of <5 % or a decrease in EF, yielded an accuracy of only 63%. The association of a change of <12.5% with a biphasic response increased accuracy to 87%, with a sensitivity of 93% and a specificity of 80%. Quantitative regional wall motion analysis showed greater diagnostic accuracy than visual wall motion scoring (83% vs 70%).
Conclusions: Application of appropriate criteria and stepwise quantitative analysis of RNV during dobutamine infusion may increase accuracy in the detection of early postinfarction ischemia.