LETTER

To the Editor: We read with interest the recent study by Garot et al. (1), which not only shows that systolic left ventricular twist changes after myocardial infarction but also introduces a new methodology, based on Doppler myocardial imaging, to quantify ventricular twist. The methodology is based on the appearance of a "black zone" in the tissue Doppler short-axis image that, according to the authors, represents a myocardial region without circumferential motion and therefore acts as a myocardial tag in the Doppler image. Measurement of the position of such a tag at both end diastole and end systole would indeed allow us to define the twist angle  as defined in this study.

However, we believe that such a myocardial Doppler tag does not exist. By definition, it is impossible to have a region of zero circumferential velocity, i.e., a Doppler tag moving in the circumferential direction during systole. The origin of the black zone is, in our opinion, different from the one set out in the report. Indeed, current Doppler techniques only measure the velocity component along the image line (2). The black zone [or zone of zero (measured) velocity] thus represents a region where motion is purely perpendicular to the image line. Because the in-plane velocity vector is the sum of the radial and the circumferential velocity components, this implies that the black zone represents the region where the sum of these components is perpendicular to the image line (see Fig. 1). The angulation of this black zone with respect to the 90° position, as defined in their study, is thus a measure of the circumferential velocity component. Indeed, it is this circumferential velocity component that determines the amount of angulation (compare Figs. 1 and 2). This implies, however, that the angulation measures the speed of rotation, rather than the rotation itself. The fact that this instantaneous rotational velocity was not equal to zero at end systole, as one might expect, can most likely be attributed to the inaccurate definition of end systole (the time resolution of the data sets was in the order of 30 ms).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   A: the in-plane velocity vector consists of a radial (VR) and circumferential (VC) component. B: because tissue Doppler can currently only assess the component of the in-plane velocity vector along the image line, a typical red-blue pattern appears. The black zone represents the region where this component is zero.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   The same images as in Fig. 1 but without the rotation of the ventricle, i.e., VC = 0.

The fact that the measured angles with ultrasound correlated in five patients to the values found with MRI tagging is, therefore, coincidental because MRI tagging extracts the true ventricular rotation (and not the speed of rotation). Of course, if the instantaneous speed of rotation increases, total rotation is likely (but not necessarily) to increase as well. Therefore, the angulation measurement is likely to correlate with the true ventricular rotation (which was apparent from the good correlation with the MRI tagging data).

Thus, in our opinion, the proposed methodology to assess myocardial twist does not measure true ventricular twist but rather the instantaneous velocity of twist. Of course, notwithstanding the different nature of the physical property measured, the conclusion that this parameter can be useful for the quantification of global left ventricular function holds.