Author + information
- Frank A. Flachskampf, MD, PhD∗ ()
- ↵∗Institute for Medical Sciences, Uppsala University, Akademiska Sjukhuset, Uppsala, Sweden 75185
I read with interest and benefit the paper by Smith et al. (1) detailing 3-dimensional (3D) echocardiographic strain evaluation of the right ventricle for functional and prognostic assessment in patients with pulmonary hypertension. The authors report that right ventricular 3D endocardial area strain (AS) was the best predictor by receiver-operating characteristic curve analysis of a right ventricular ejection fraction <40% compared with 3D-derived longitudinal strain (LS), circumferential strain (CS), or radial strain; in addition, in a stepwise logistic regression model, it remained the only significant independent prognostic factor of mortality. In the legend for the Central Illustration, endocardial AS was described as the sum of LS and CS. These statements deserve some comment.
First, no head-to-head comparison was performed with 2-dimensional (2D)-based speckle tracking strain measurements of LS and CS. 2D echocardiographic data for speckle tracking have higher temporal and spatial resolution than 3D acquisition; for example, Smith et al. (1) used 3D acquisition rates of 14 to 15 frames/s, compared with the >40 frames/s rates typically requested for 2D speckle tracking.
Second, analytically, AS can be calculated from CS and LS (2). In a simplified analysis (leaving out the effect of somewhat different measuring sites for the different strains), an endocardial rectangular area with sides L and C and area L0 · C0 at end-diastole would deform to an area of (L0 + L0 · LS) · (C0 + C0 · CS) at end-systole. AS then becomes AS = ([L0 + L0 · LS] · [C0 + C0 · CS] − L0 · C0)/L0 · C0, which simplifies to AS = CS + LS + CS · LS.
Because CS and LS are typically negative percents, AS becomes smaller in absolute terms than the sum of CS and LS, which is actually the case in the values reported in the paper. Irrespective of this detail, it is difficult to accept that the composite parameter AS would have fundamentally superior physiological or prognostic value than the component strains that uniquely determine it and can be measured with higher resolution by 2D methods.
- American College of Cardiology Foundation