The recently published integrative algorithms for echocardiographic grading of native aortic regurgitation (AR) and mitral regurgitation (MR) by the American Society of Echocardiography are consensus based and have not been evaluated. Thus, the aims of the present study were to investigate the feasibility of individual parameters and to evaluate the ability of the algorithms to discriminate severe from moderate regurgitation.Methods:
This prospective study comprised 93 patients with chronic AR (n = 45) and MR (n = 48). All patients underwent echocardiography and cardiovascular magnetic resonance within 4 hours. The algorithms were evaluated using two different definitions for severe regurgitation: (1) a cardiovascular magnetic resonance standard indicating future need for valve surgery and (2) a clinical standard using patients who underwent valve surgery with proven postoperative left ventricular reverse remodeling and improved functional class (AR/MR, n = 26/26).Results:
The feasibility of the criteria in the first step of the algorithm was higher (AR/MR, 95%/91%) compared with the second step using quantitative Doppler parameters (74%/57%). For the AR algorithm, sensitivity was 95% and specificity 44%, whereas for the MR algorithm, sensitivity was 73% and specificity 92%. Among patients with benefit of surgery, the algorithms correctly identified 77%, misclassified 8%, and were inconclusive in 15% of the patients with AR; the corresponding figures were 73%, 15%, and 12% in the patients with MR.Conclusions:
Using cardiovascular magnetic resonance as reference, the recommended algorithms for grading of regurgitation have the ability to rule out severe AR and rule in severe MR. The quantitative Doppler methods are hampered by feasibility issues, and our findings suggest that the decision regarding surgical intervention in symptomatic patients with discordant or inconclusive echocardiographic grading should be based on a consolidated assessment of clinical and multimodality findings.