Direct and indirect quantification of mitral regurgitation with cardiovascular magnetic resonance, and the effect of heart rate variability

Quantifying mitral regurgitation with cardiovascular magnetic resonance (CMR) involves indirect calculation, which increases the potential for error. We examined a direct quantification method using velocity mapping across the mitral valve, which may be less susceptible to error, and also examined the effect of heart rate variability on both techniques. Fifty-five patients underwent mitral regurgitation quantification with CMR by the direct method and two indirect methods-the standard method subtracting aortic flow (assessed by velocity mapping) from left ventricular stroke volume (assessed by cine imaging) and the 'volumetric' method using the difference between left and right ventricular stroke volumes. The methods were compared using Bland-Altman analyses. Patients with low heart rate variability (beat-to-beat variability < 30 bpm; n = 44) showed good agreement between direct and indirect methods (95% confidence limits for the difference between measurements +/- 16.7ml/11.8% regurgitant fraction for the standard method; +/- 21.7ml/15.4% for the volumetric method), with no significant offset (mean difference +2.8 ml/+1.9% for standard and +3.1ml/+2.3% for volumetric methods). Patients with high heart rate variability (> 30 bpm; n = 11) showed poor agreement between techniques (95% limits +/- 80.3ml/56.0%) and significant offset (mean difference +31.7ml/+19.5%). Direct quantification of mitral regurgitation with CMR compares well with indirect methods for patients with low heart rate variability, involves fewer calculations and is quick. All CMR measurements that use velocity mapping may be inaccurate, however, in patients with highly irregular rhythms and should be avoided in these patients.