Reproducibility of dobutamine digital stress echocardiography

Abstract

The aim of this study was to investigate the temporal variability and interobserver agreement of dobutamine digital stress echocardiography. We performed two dobutamine stress echocardiographic studies (dobutamine up to 40 μg/kg/min and atropine up to 1 mg) in 15 patients with previous myocardial infarction at a mean of 19 days apart. Two observers assessed the wall motion using a six-point score in a 16-segment model and calculated the wall motion score index at rest and at peak stress by using a quad screen display. Analysis of the wall motion was performed separately on the day after each dobutamine stress test (analysis A), and all images from the two serial studies in the same patient were simultaneously retrieved and compared side-by-side in the same view (analysis B). The mean values of heart rate and blood pressure were comparable for each stage in the two studies except for the heart rate at rest. Regarding the presence and absence of positive findings of dobutamine stress echocardiography, interobserver agreement was 93% (70% to 99% with 95% confidence limits, kappa value 0.86) in the patients and 93% (70% to 99% with 95% confidence limits, kappa value 0.80) in the three major vascular regions with the use of analysis A. These values did not improve with the use of analysis B. The agreement of the temporal variability was 93% (70% to 99% with 95% confidence limits, kappa value 0.86) in the patients and 84% (71% to 92% with 95% confidence limits, kappa value 0.66) in the vascular regions with the use of analysis A. These values further improved with the use of analysis B. With the comparison of the wall motion score index, interobserver variability showed a correlation coefficient of 0.88 at rest and 0.90 at peak stress with analysis A and 0.78 and 0.82, respectively, with analysis B. Corresponding analysis of temporal variability showed correlation coefficients of 0.99 at rest and 0.99 at peak stress when both analyses were used. Although dobutamine digital stress echocardiography has good reproducibility and negligible interobserver variability, even if the digital quad screen format is used, it requires strict standardization of the reading criteria and the objective measurements of wall motion in the expansion of this test to the evaluation of the changes in left ventricular function during more than two serial studies in the same patient. (J Am Soc Echocardiogr 1997;10:344–51.)