An experimental model of chronic severe mitral regurgitation

A canine model of chronic severe mitral regurgitation showed marked cardiac remodeling.

Figure 1. Representative fluoroscopic image of endovascular avulsion of mitral valve chordae tendinae using endomyocardial biopsy forceps. A, Biopsy forceps in the left ventricular outflow tract; B, transesophageal echocardiography probe in the esophagus.

Figure 2. Representative echocardiographic images of severe mitral regurgitation (MR). A, Transthoracic echocardiographic image (apical 4-chamber view) of the left atrium (LA) with color Doppler showing eccentric MR jet with Coanda effect (wall-hugging, wrap-around jet). B, Transesophageal echocardiographic 3-dimensional image of the mitral valve en face, ie, surgeon's view, showing coaptation defect associated with flail leaflet, imaged during terminal surgery. LV, Left ventricle.

Figure 3. Assessment of atrial arrhythmogenic substrate via transvenous pacing of the right atrial appendage (RAA). Shown are representative electrophysiologic recordings made during a burst pacing attempt that successfully induced an atrial tachyarrhythmia, shown on the electrocardiogram (ECG) and the RAA-sensing electrode.

Figure 4. Echocardiographic measurement of mitral regurgitation (MR): MR remained severe over time. Regurgitant jet area-to-left atrial area (RJA/LAA) ratio (measured via transthoracic echocardiography [TTE]) plotted over the entire time course of the experimental protocol. Mean RJA/LAA values are shown with standard deviation bars. Blue shading indicates TTEs performed during the interval monitoring period, which was of a variable length for each dog. Severe MR was induced and maintained over an average period of 9 months.

Figure 5. Echocardiographic and magnetic resonance imaging (MRI) measurements of volumetric changes to the left atrium (LA) and left ventricle (LV) that occurred over a mean duration of 9 months with severe mitral regurgitation (MR). All values shown were indexed to body surface area as measured at the time of exam, and are shown with standard deviation bars. A, LA volume as measured via transthoracic echocardiography (TTE). Blue shading indicates TTEs performed during the interval monitoring period, which was of a variable length for each dog. B, LA volume as measured via cardiac MRI. C, LV internal diameter as measured via TTE. Blue shading indicates TTEs performed during the interval monitoring period. D, LV end-diastolic volume, as measured via MRI.

Figure 6. Graphical abstract: an experimental model of chronic severe mitral regurgitation. RJA, Regurgitant jet area; LAA, left atrial area; MRI, magnetic resonance imaging; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography; MR, mitral regurgitation; LAV, left atrial volume; LVID, left ventricular end-diastolic internal diameter ; ATA, atrial tachyarrhythmia; LV, left ventricle; LA, left atrium; AF, atrial fibrillation.

Figure E1. Models of left atrial and left ventricular remodeling over time. A and B model percentage change in indexed left atrial (LA) end-systolic volume as measured via transthoracic echocardiography (TTE) versus time since creation of mitral regurgitation (MR); C and D model percentage change in left ventricular end-diastolic internal diameter (LVID) as measured via TTE versus time since MR creation. A and C show simple linear regression models, with the following parameters reported: coefficient of determination (R2), P value as compared with no correlation between percentage change and time since mitral regurgitation (MR) creation, and Akaike information criterion corrected for small sample size (AICC). B and D show locally estimated scatterplot smoothing (LOESS) models, with the following parameters reported: AICC and the smoothing parameter used (smooth). It is possible to perform a simple linear regression to compare time since MR-creation with either left atrial volume (LAV) or LVID. However, in this sample, LAV and LVID are not normally distributed (and therefore should not be modeled using a parametric test). We can use a nonparametric method such as LOESS, which does not make…