Functional diagnosis of coronary stenosis using tissue tracking provides best sensitivity and specificity for left circumflex disease: experience from the MYDISE (myocardial Doppler in stress echocardiography) study.

AIMS: To evaluate the diagnostic capacity of quantitative analysis of segmental longitudinal myocardial displacement images (tissue tracking, TT) during dobutamine stress echocardiography for the detection of patients with coronary artery disease (CAD). METHODS AND RESULTS: TT-generated colour-coded maps of systolic segmental longitudinal displacement were obtained by post-processing of echocardiographic data from 105 patients with CAD and 90 low risk individuals selected from MYDISE database. Quantitative analysis of the distribution pattern of segmental displacement during dobutamine stress was most successful when a ratio of basal (high amplitude) to apical (low amplitude) colour-coded displacement bands (B/A ratio) was employed. Applied in four different left ventricular sectors, the B/A ratio provided a significant discrimination of patients with CAD (p<0.05 in the anterior and p<0.001 in the inferior wall) as assessed by receiver operating characteristic analysis. The procedure was most sensitive when applied in inferior wall for the detection of left circumflex coronary artery disease, the B/A ratio of 0.8 giving the best combination of sensitivity (77+/-8%) and specificity (77+/-5%) values. CONCLUSION: Quantification of dobutamine stress echocardiography using TT is an efficient diagnostic approach and a valuable additional modality in functional cardiac imaging for the initial identification of patients suspected for CAD.

Analysis of the effect of temporal filtering in myocardial tissue velocity imaging.

Signal filtering to reduce random noise may compromise the reliability of tissue velocity measurements. This study evaluates the influence of temporal filters on time and velocity variables derived from myocardial tissue velocity images acquired in 15 healthy individuals at a high frame rate (142-184 Hz). Different time and velocity variables from the basal septum were analyzed offline before and after temporal filtering from 20 to 60 milliseconds in 10-millisecond steps using software enabling retrieval of myocardial Doppler velocity and 2-dimensional information from different cardiac locations during the same cardiac cycle. Filtering affected the results by increasing variability of time and by underestimation of velocity variables, the rapid isovolumic events being particularly filter sensitive. In addition, at a certain range of sampling rate, ambiguity of filtering effect was observed. This ought to be considered if an optimal, high-fidelity tissue Doppler velocity signal is to be obtained.

Distribution of left ventricular longitudinal peak systolic strain and impact of low frame rate.

The myocardium has complex 3-D motion that is frequently described using ultrasound (US) Doppler techniques that are limited to recording velocities in one dimension only. Studies using 3-D tagged magnetic resonance show that the myocardium has strain components with varying angles throughout the myocardium. Despite this, there seems to be a belief that the left ventricular longitudinal strain distribution should be homogeneous. When measuring myocardial strain, there are several parameters for the clinician to decide on, one of them being recording frame rate. The current study aims to further investigate the alleged homogeneity of the longitudinal myocardial strain distribution and to discover the impact that the frame rate has on these measurements. Myocardial strain was measured in 43 healthy individuals at different frame rates. Analysis of variance results clearly demonstrate that the strain is not uniformly distributed over the wall; there seems to be an increasing strain from apex toward the base. However, subjects exist with different distributions; thus, it is not possible to conclude that certain strain patterns are normal. Reduced frame rate had a highly significant impact on the measured strain results and it is seen that, at low frame rates, the strain values were reduced.

Tissue motion imaging of the left ventricle--quantification of myocardial strain, velocity, acceleration and displacement in a single image.

AIMS: Several methods of parametric imaging of left ventricular function including tissue velocity imaging (TVI) and strain rate imaging (SRI) have previously been presented, however, they have the limitation that they can, respectively, portray only one physiological myocardial parameter. The aims of this pilot study were to implement and validate tissue motion imaging (TMI) for the first time, a visualization technique which permits acceleration, velocity, displacement and strain to be interpreted quantitatively or semi-quantitatively in a single image. METHODS AND RESULTS: TMI is achieved by the color coding of temporal tissue velocity integrals. The principles behind this technique are validated, and case examples demonstrating its use in the clinical setting are provided. Limitations of the method as well as future applications and improvements are discussed. CONCLUSION: As this method allows representation of a multitude of variables and is visually attractive, it may facilitate more widespread use of myocardial quantitation in everyday practice.

Effect of angular error on tissue Doppler velocities and strain.

One of the major criticisms of ultrasound Doppler is its angle dependency, that is its ability to measure velocity components directly to or from the transducer only. The present article aims to investigate the impact of this angular error in a clinical setting. Apical two- and four-chamber views were recorded in 43 individuals, and the myocardium was marked by hand in each image. We assume that the main direction of the myocardial velocities is longitudinal and correct for the angular error by backprojecting measured velocities onto the longitudinal direction drawn. Strain was calculated from both corrected and uncorrected velocities in 12 segments for each individual. The results indicate that the difference between strain values calculated from corrected and uncorrected velocities is insignificant in 5 segments and within a decimal range in 11 segments. The biggest difference between measured and corrected strain values was found in the apical segments. Strain is also found to be more robust against angular error than velocities because the difference between corrected and uncorrected values is smaller for strain. Considering that there are multiple sources of noise in ultrasound Doppler measurements, the authors conclude that the angular error has so little impact on longitudinal strain that correction for this error can safely be omitted. (ECHOCARDIOGRAPHY, Volume 20, October 2003)