Impact and limitations of 3D computational modelling in transcatheter mitral valve replacement—a two-centre Dutch experience

Abstract

Background

Transcatheter mitral valve replacement (TMVR) has emerged as a minimally invasive alternative to mitral valve surgery for patients at high or prohibitive operative risk. Prospective studies reported favourable outcomes in patients with annulus calcification (valve-in-mitral annulus calcification; ViMAC), failed annuloplasty ring (mitral valve-in-ring; MViR), and bioprosthetic mitral valve dysfunction (mitral valve-in-valve; MViV). Multi-slice computed tomography (MSCT)-derived 3D-modelling and simulations may provide complementary anatomical perspectives for TMVR planning.

Aims

We aimed to illustrate the implementation of MSCT-derived modelling and simulations in the workup of TMVR for ViMAC, MViR, and MViV.

Methods

For this retrospective study, we included all consecutive patients screened for TMVR and compared MSCT data, echocardiographic outcomes and clinical outcomes.

Results

Sixteen out of 41 patients were treated with TMVR (ViMAC n = 9, MViR n = 3, MViV n = 4). Eleven patients were excluded for inappropriate sizing, 4 for anchoring issues and 10 for an unacceptable risk of left ventricular outflow tract obstruction (LVOTO) based on 3D modelling. There were 3 procedure-related deaths and 1 non-procedure-related cardiovascular death during 30 days of follow-up. LVOTO occurred in 3 ViMAC patients and 1 MViR patient, due to deeper valve implantation than planned in 3 patients, and anterior mitral leaflet displacement with recurrent basal septum thickening in 1 patient. TMVR significantly reduced mitral mean gradients as compared with baseline measurements (median mean gradient 9.5 (9.0–11.5) mm Hg before TMVR versus 5.0 (4.5–6.0) mm Hg after TMVR, p = 0.03). There was no residual mitral regurgitation at 30 days.

Conclusion

MSCT-derived 3D modelling and simulation provide valuable anatomical insights for TMVR with transcatheter balloon expandable valves in ViMAC, MViR and MViV. Further planning iterations should target the persistent risk for neo-LVOTO.