A mathematical model for the quantification of mitral regurgitation. Experimental validation in the canine model using contrast echocardiography.

Abstract

BACKGROUND Because the clearance of contrast from the left atrium (LA) relative to the left ventricle (LV) depends on the degree of mitral regurgitation (MR), we hypothesized that a mathematical model can be developed that would provide a quantitative estimation of MR from the washout of contrast from these chambers. METHODS AND RESULTS After mathematically developing the model, we performed experiments in two groups of dogs with the use of contrast echocardiography. Group 1 consisted of nine dogs in which different degrees of MR were produced by creating ischemic LV dysfunction. Contrast was injected into the LV, and MR was graded visually on a scale of from 0 to 4+. Videointensity plots generated from the LA and LV were provided to the model. There was excellent correlation between visual assessment of MR and model-derived regurgitant fraction in the 33 stages: y = 0.16x + 0.002 (r = 0.97, p less than 0.001, SEE = 0.06). To obtain a more quantitative validation, we placed electromagnetic flow probes on the aorta and just cephalad to the mitral annulus in six dogs (group 2) during cardiopulmonary bypass. Different degrees of MR were produced by chordal traction and/or myocardial ischemia. Regurgitant fraction was calculated at each stage from the flow probe and videointensity data. There was excellent correlation between flow probe and model-derived regurgitant fraction (y = 0.90x + 0.03; r = 0.96, p less than 0.001, SEE = 0.06), and close interobserver and intraobserver correlations were noted using flow probe and contrast echocardiographic data. CONCLUSIONS A mathematical model that uses the clearance of contrast from the LA relative to the LV can be used to accurately measure the severity of MR. These findings may have important practical implications for the quantification of MR.