Right Ventricular (RV) dysfunction is the primary cause of acute RV failure in pulmonary arterial hypertension (PAH). RV failure in turn is the cause of at least 70% of deaths in PAH patients and is almost universally correlated to poor prognosis. RV function is routinely assessed with echocardiography, which allows for dynamic visualization of RV motion as well as assessment of major cardiac structures. As part of a routine clinical exam, echo-derived global longitudinal strain (GLS) is measured from a two-dimensional echo (2DE) image; strain is a mechanical measure of cardiac deformation during RV contraction. GLS has been incorporated into recent echo guidelines, is emerging as a useful measure of RV systolic function and has prognostic significance. GLS tends to be an earlier measure of ventricular dysfunction, with reduced values showing up before global function (as measured by ejection fraction, EF) is significantly reduced. However, the 2DE modality by which GLS is obtained is by definition confined to a 2D plane, and thus cannot fully visualize contractions over the asymmetrically shaped RV. We believe that analysis of three-dimensional echo (3DE) images can overcome this limitations and provide 3D surface (3DS) strain upon the entire RV surface. Aside from providing longitudinal & circumferential strains in this way, 3DS strain analysis additionally yields shear strain as well as the two normal principal component (PC) strains. PC strains provide not only magnitude but also direction, and may offer additional insight into RV remodeling and potentially even fiber direction, which in turn may enable remodeling-specific therapeutic strategies. Together, the components of 3DS strain add substantial additional and novel information regarding how the RV deforms in three dimensions and should allow for earlier detection of ventricular function and in turn, earlier management changes for PAH patients.
Kendall Hunter is a recovering computational mechanist who now, as a bioengineer, enjoys computing and interpreting vector (blood velocity, vorticity) and tensor (tissue strain) fields from clinical images. In his free time he learns new machine learning algorithms to better detect patterns in clinical data and assembles LEGO Technic models.