🧵In our paper “Intraventricular Vector Flow Imaging with Blood Speckle Tracking in Adults: Feasibility, Normal Physiology and Mechanisms in Healthy Volunteers” pubmed.ncbi.nlm.nih.gov/34620522/ , findings were not only qualitative as described in previous threads, but also quantitative.
1/ Measures of kinetic energy (KE), vorticity (VO), energy loss (EL) and pressure gradients (PG) can be calculated. The figure from the paper shows the curves from all subjects in the study. What does these measures mean, and are they likely to add useful information?
2/ Starting with kinetic energy, this is the kinetic energy per volume, and can be integrated from the individual velocity vectors. As this was integrated over the 2D area only, the energy is given in J/m. The upper panel shows all subjects, the lower a curve from one subject.
3/ It's obvious that the peaks of KE is related to the in- and outflow phases where velocity peaks (and higher at the LV basis. This means that the pW velocities represent the resultant, and integrated KE represents mostly the same (squared).
4/ So chasing KE as a new measure, just because it's new, without establishing standards the method, and without any sound hypothesis that it gives added information seems unsound.😈 But what about Vorticity, which is a completely new measure?
5/ VO is the rotation of the blood in each point of the image, and is related to the complexity of the blood flow. It is calculated by the mean curl or momentum of the blood velocity field over the LV, and is given in Hz. However, still method dependent, with no gold standard.
6/ Which means
A: It can't be validated against other methods (like MR), B: Nor are values derived from different #echofirst methods comparable. However, qualitative information may be physiologically interesting, especially if they are consistent with other findings:
7/ VO peak seems to be somewhat more dispersed than KE, uncertain whether this difference in variability is methodological or biological. However, VO seems to peak close to, but slightly later than KE.
8/ This is consistent with qualitative findings that shows that the vortex is created by the interaction of inflow with the AV-plane motion subsequent to the inflow itself.
9/ Vorticity decreases in diastasis, but apparently less than KE, again consistent with the presence of a vortex through diastasis, where the septal flow closes the MV, and the lateral part conserves momentum for late filling.
10/ vorticity peaks again after peak KE of late filling, but this time extends into pre ejection, where the septal flow aligns the momentum with the ejection and closes the MV, while the lateral flow slowly fades, along with the whole vortex.
11/ vorticity decreasing during ejection where flow is mainly laminar, except at the very end where a slight apical momentum is imparted by the apical AV-plane motion.
12/ - and finally reaching minimum as the volume shift during iVR imparts a unidirectional flow gradient before early filling.
12/ VO is a new measure, and findings seems consistent with the qualitative evaluation of vortex formation, as well as colour flow and pw Doppler, so the physiology is credible.
13/ Caveats:
A:Actual peak values will differ with method.
B: There is no gold standard, ref. method will also be methodologically different, phantom validation might be possible though.
14/ I don't see the value of research stampeding along trying to run VO through the "sausage factory" of generating normal values or comparing patients with controls just because this is a new (and "sexy"?) method.
15/ so far, VO has confirmed what we see from flow patterns, and qualitative evaluation of the patterns and curve forms / timing, may give new information (like strain and strain rate) about physiology and pathophysiology, I'm specially optimistic about load.
16/ As you see, there are more measures available from the method, I'll probably return with a new thread about pressure gradients and maybe energy loss later.
🧵 on the physiology of regional strain. While the added value of global longitudinal strain is doubtful, compared to MAPSE, strain and strain rate was, and is still a method for visualising inequalities of *Regional* systolic shortening.
1/ B-mode provides all the necessary information where changes are large and obvious, as in this case, but regional strain/strain rate may be of added value in giving some added physiological information
2/ in other cases, where changes are small, strain and strain rate may add diagnostic value.
🧵While it is doubtful that GLS adds anything to MAPSE in global function, strain (and strain rate) are useful to assess differences in regional function, both in CHD and dyssynchrony. Regional myocardial work, however, doesn't seem to add information.
1/ The typical finding in CHD is delayed onset and reduced magnitude of systolic shortening, and post systolic shortening in affected segments as seen here in the apex (white curve - compare normal magenta curve).
2/ In segmental length-pressure loops, using a reconstructed pressure curve, the width of the loop at AVC, corresponds to the strain at AVC. But any regional difference in the loops, must be a function of the difference in strain, as all segments relate to the same pressure curve
🧵What’s layer strain, and what does it mean? With speckle tracking, the ROI can be divided into layers, and strain measured selectively in each layer, both longitudinal strain in apical views, and circumferential strain in short axis views. But what is this actually?
Strains differ between layers, but the difference is NOT due to differences in fibre function in the different layers, it is again a function of geometry.
1/ 1/ starting with circumferential strain, which is conceptually easiest, in the HUNT study, with linear strains, we found outer Sc ca 13%, midwall Sc ca 23% and endocardial Sc 36%, so there is a clear gradient of Sc across the wall. pubmed.ncbi.nlm.nih.gov/31673384/
🧵Strain is defined in three dimensions; longitudinal, circumferential and transmural (radial). Each strain component defines deformation in one dimension. It is, however, absurd to consider the three components independently, or as reflection of shortening of specific fibres.
1/ Any three-dimensional object is defined by a three dimensional coordinate system. The simplest is the cartesian system of xyz. In the LV myocardium, being more of a a hollow ellipsoid, the longitudinal, circumferential and transmural directions are more convenient.
2/ Thus, systolic deformation of a 3D object occurs along the three axes, simultaneously. With some incompressibility (not necessarily total), deformation in one direction must relate to deformation in the two other, expansion in one usually follows shrinking in the two others.
However, in the HUNT study, we found no significant sex differences in MAPSE (although a trend, p=0.1), but in a large study of 1266 subjects, the difference was small < 0.05mm - far below measurement limit). pubmed.ncbi.nlm.nih.gov/29399886/ Why, when both are long axis function?
1/ In our study, we compared GLS derived from segmental values by our software, with MAPSE normalised for the corresponding end diastolic wall length (straight line) and non-normalised MAPSE pubmed.ncbi.nlm.nih.gov/29399886/
🧵What is GLS? 1/ It is evident that it is some measure of the systolic LV longitudinal shortening, normalised for the diastolic LV length, after the basic Lagrangian formula S = (L-L0)/L0
But how do we chose the numerator and denominator? 2/ The simplest measure would be LV systolic shortening / LV end diastolic length. In the HUNT 3 study, strain by this method was -17.1%. pubmed.ncbi.nlm.nih.gov/32978265/
LV shortening can be approximated by MAPSE, so GLS is similar to MAPSE normalised for LV diastolic length. Average MAPSE of sep-lat was similar to average of sep-ant-lat-inf within the measurement accuracy in the HUNT3 study. pubmed.ncbi.nlm.nih.gov/29399886/