Discussion
To our knowledge, this paper is the first comparison of MDCT to TEE for the purpose of TMVr planning. Current generation MDCT scanners, with at least 64-slice technology capable of generating images with sub-millimetric spatial resolution allow for comprehensive evaluation of complex mitral anatomy. Quantitative information including leaflet length and mitral apparatus dimensions can be obtained with MPR, while volume rendering depicts a 3D assessment of the valve. In our study, we found that patients evaluated for TMVR/r with MDCT had similar findings of the mitral valve and valvular apparatus compared to TEE.
Our results found that measurements between MDCT and TEE were not significantly different for MVA, flail width, commissural diameter, AP diameter, posterior leaflet length and leaflet thickness. Previously, Fuechtner et al found that MDCT can diagnose mitral valve prolapse with high accuracy. Our study also found that MDCT can accurately identify the pathological leaflet scallop location, discriminate between flail vs billowing leaflets and characterize leaflet thickening. Shanks et al, also found that 3D TEE has comparable mitral valve geometry to MDCT (including similar posterior leaflet length measurement). Sizing of the mitral annulus is traditionally done with MDCT, though our study found that measurements were similar among imaging modalities. Additionally, annular calcification is best assessed with MDCT, yet MAC severity was also comparable.
We did find a discrepancy of the flail gap measurements. It is unclear why MDCT tended to overestimate the flail gap. Prior studies have felt that the four-chamber view on MDCT resulted in an overestimation of billowing in prolapsed valves [5]. Our CT readers therefore used the two- and three- chamber reformations on MDCT to measure flail gap, however we still found a significant difference. We cannot rule out the possibility that MDCT is more accurate than TEE, and perhaps we did not appreciate the full extent of flail on TEE. Mitral regurgitation may be a dynamic process and the volume load (with IV contrast) for MDCT along with the sedation given during TEE may also lead to real life changes in the mitral pathology measurements. We therefore can’t exclude the possibility that the differences found were accurate. Identification of cleft leaflet is best recognized with 3D TEE. There were two patients in our study with cleft leaflet, and MDCT was only able to detect the cleft in one of them. The slit-like appearance of a cleft leaflet can be difficult to see on 2D TTE or TEE. Cleft leaflets are identified on 3D TEE. With MDCT, using MPR projections is difficult to assess cleft leaflet, but with VR 3D imaging we were able to identify one of the cleft leaflets. However, adjusting the window of the 3D rendered image can make assessment of thin/soft tissue structures of the mitral leaflet a challenge, particularly clefts or pseudoclefts.