New aspects on the assessment of left ventricular dyssynchrony by tissue Doppler echocardiography: comparison of myocardial velocity vs. displacement curves

• Published in: The International Journal of Cardiovascular Imaging Vol. 25; no. 7; pp. 699 - 704
• Main Authors: Bogunovic, N., Hering, D., van Buuren, F., Welge, D., Lamp, B., Horstkotte, D., Faber, L.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.10.2009; Springer Nature B.V
• Subjects: Cardiac Imaging; Cardiac Pacing, Artificial; Cardiology; Clinical Competence; Echocardiography, Doppler, Color; Heart Failure, Systolic - diagnostic imaging; Heart Failure, Systolic - physiopathology; Heart Failure, Systolic - therapy; Heart Rate; Humans; Imaging; Medicine; Medicine & Public Health; Myocardial Contraction; Observer Variation; Original Paper; Patient Selection; Predictive Value of Tests; Radiology; Reproducibility of Results; Time Factors; Ventricular Dysfunction, Left - diagnostic imaging; Ventricular Dysfunction, Left - physiopathology; Ventricular Dysfunction, Left - therapy; Left ventricular dyssynchrony; Tissue Doppler
• ISSN: 1569-5794; 1573-0743; 1875-8312
• DOI: 10.1007/s10554-009-9484-8

The purpose of this study is to assess the incremental value of tissue Doppler (TDI) derived displacement curves (TDint) compared to TDI velocity curves (TDvel) for the evaluation of left ventricular (LV) dyssynchrony (LVD). About 25 patients (pts.) with systolic heart failure were studied by TDvel and TDint. Four TDI sample volumes were placed at the basal and four at the mid ventricular myocardium, utilising two imaging planes. LV dyssynchrony (LVD) was defined as an interregional delay of >40 ms corrected for heart rate. 10 pts. had synchronous contraction, 15 pts. LVD as defined by two experts (EC). To determine diagnostic accuracy and intra-observer variability two identical sets of 100 documents (25 pts. × two imaging planes × two modalities) were produced and presented in random order to one trained (TR) and two untrained (UR) readers. The TR more frequently classified documents as unreadable (7.5 vs. 3.5%, P  < 0.05) but more often as correct, i.e., consistent with EC (72.0 vs. 57.8%, P  < 0.001). 8.7% of the documents were classified as unreadable using TDvel, 1.0% when applying TDint ( P  < 0.001). The mean value of correct classification of all 3 readers was 54.3% (TDvel only), 70.7% (TDint only), and 77.7% (combining both modalities), ( P  < 0.001). The kappa value for TR and TDint was 0.68, for TDvel 0.29. For UR, kappa did not differ (TDint: 0.58; TDvel 0.51). TDint is superior to TDvel in accuracy, reproducibility, and applicability for skilled and unskilled investigators when evaluating LVD by TDI. The combined application of TDint and TDvel is optimal.