The present study shows that in patients with suspected CAD undergoing dobutamine stress echocardiography quantitative tools are not superior to visual wall motion analysis. None of the single parameters investigated was able to identify myocardial ischemia and significant coronary artery disease with a comparable diagnostic accuracy vs. wall motion analysis. The several parameters did not perform differently between themselves. Moreover, these tools are quite time consuming and do not seem to offer a significant diagnostic improvement in the hands of expert echocardiographers.
Comparison with previous studies
There are not many studies comparing multiple quantitative parameters during stress echocardiography. This approach aimed to identify the best quantitative parameter for the diagnosis of myocardial ischemia and significant coronary artery disease. The rationale of applying pulsed wave and color coded Doppler myocardial imaging as well as speckle tracking technique during the same dobutamine stress test is based on the awareness that each of these methods has inherent advantages and limitations. Direct comparison of physiologically different parameters as well as similar markers obtained by different methods should allow the reasonable choice of the most reliable predictor of CAD for practical use. In this, our approach is unique and provides a missing piece of information.
We performed the search of the best predicting parameter of significant stenosis separately for each evaluated myocardial segment, taking into account known base-to-apex and wall-to-wall differencies of myocardial velocity and strain/strain rate
[11, 19–21]. From the list of 98 rest and stress measured and calculated parameters, 40 site-specific markers appeared to be statistically significant predictors of coronary stenosis (AUC >0.5, P≤0.025). Thus, our results confirm the relation of substantial list of quantitative parameters of regional myocardial function to the obstruction of coronary arteries.
In agreement with previously published findings
[6, 8, 11, 22, 23] blunted response of systolic velocity to dobutamine infusion and prolonged time to peak systolic velocity during stress demonstrated by Doppler based methods appeared to be significant predictors of coronary stenosis.
Also, in concordance with previous reports
[7, 24–26] all quantitative techniques employed in the present study provided markers of regional diastolic dysfunction as significant predictors of stenosis. E' wave velocity and E'/A' ratio during stress demonstrated the predictive ability in all coronary territories. Furthermore, we found significant deterioration of local diastolic function in the segments supplied by stenosed RCA even at rest.
In parallel to previous deformation studies
[11–13] longitudinal myocardial deformation was significantly impaired during stress in STENOSED segments. Stress systolic strain, ratio of post-systolic index to systolic and post-systolic strain, absolute and relative changes of post-systolic index appeared to be significant predictors of CAD. Besides, systolic strain and post-systolic index were significantly lower in STENOSED segments already at baseline.
By analogy with the prior clinical studies
[11, 13, 27], longitudinal as well as radial strain rate markers at rest and during stress appeared to be significant predictors of CAD in our data. Stress markers included longitudinal systolic and post-systolic strain rate, absolute changes in longitudinal systolic and post-systolic strain rate. Rest markers included longitudinal post-systolic and radial systolic strain rate in the segments supplied by stenosed RCA and LAD.
However, the predictive ability of discriminated single parameters appeared to be modest: AUC ranged from 0.63 to 0.72. Similar ability to predict significant CAD was reported for strain rate, strain parameters and post-systolic index (AUCs 0.67 – 0.71, 0.64 - 0.66, and 0.60 – 0.63, respectively) in the study of Hanekom et al.
. Remarkably, the majority of informative parameters in the present study were not repetitive from segment to segment, and in the same segment different imaging methods provided various markers. Thereby, analysing the extensive set of quantitative indices we could not distinguish any single robust predictor of coronary stenosis universal for the majority of myocardial segments.
None of the quantitative markers could compare with the visual assessment of DSE in terms of the accuracy of predicting stenosis, which was 86% in the present study (accuracy of single parameters ranged from from 45.8% to 80.0%). In MYDISE investigation similar limited sensitivity (67-69%) and specificity (60-67%) of myocardial systolic velocities before correction by logistic regression models were demonstrated
. In the study of Cain et al.
 the accuracy of myocardial Doppler velocities was lower comparing to wall motion scoring, while strain rate imaging in the report of Voigt et al.
 was found to be comparable with conventional visual assessment. Investigating the accuracy of Doppler-based and two-dimensional strain imaging, Hanekom
 did not find significant differences between quantitative and visual assessment.
Limited value of distinguished indices largely could be attributed to known technical challenges of quantitative imaging: potentially inadequate spatial and temporal resolution, angle-dependency, less accurate tracking of ultrasound speckles at higher heart rates, noise and artefacts
. Mutual interaction of STENOSED and NON-STENOSED segments, possibly, may diminish the differences between local myocardial motion markers of these two groups
. Though strain/strain rate parameters were supposed to overcome the influence of adjacent segments, our data demonstrate interfering greater variability of strain markers.
Current lack of evidence on effective application of quantitative methods in routine practice is reflected in recommendation documents of EAE and ASE
[14, 31]. These new methods are not routinely recommended for detection of myocardial ischemia in stress echocardiography. Recent consensus statement
 also claims that in the majority of areas, including assessment of ischemic myocardium and stress echocardiography, quantitative methodology is not yet ready for routine clinical use.