042 Assessment of valve haemodynamics, reverse ventricular remodelling and myocardial fibrosis following transcatheter aortic valve implantation compared to surgical aortic valve replacement. A cardiovascular magnetic resonance study

Introduction One-year survival post transcatheter aortic valve implantation (TAVI) is non-inferior to surgical aortic valve replacement (SAVR) in patients with severe aortic stenosis (AS) and high operative risk. The effects of TAVI on valve haemodynamics and subsequent ventricular reverse remodelling relative to SAVR are less certain, and the impact of myocardial fibrosis (MF) is unknown. These would be expected to impact on longer-term outcome. Our aim was to use cardiovascular magnetic resonance (CMR) imaging to assess the 6-month post-operative aortic valve haemodynamics, reverse ventricular remodelling, and myocardial fibrosis changes following TAVI compared to SAVR. Secondary aims were to identify predictors of impaired left ventricular reverse remodelling and to establish the importance of pre-operative myocardial fibrosis on clinical outcomes. Methods 77 high-risk AS patients referred for TAVI or SAVR were prospectively recruited. 50 patients (25 TAVI, 25 SAVR) completed baseline and 6-month post-operative 1.5 Tesla CMR scans. Multi-slice, multi-phase cine imaging was performed to cover the entire left ventricle. Phase contrast (velocity encoded) imaging was used to quantify aortic mean gradient and % regurgitation. Late gadolinium enhancement was performed 10 min after the administration of 0.2 mmol/kg of Gadoteric acid (Doteram, Guerbet, SA, Villepinte). TAVI used the third generation CoreValve revalving system. In the surgical patients 96% had bioprosthetic valves. Results Patients were matched for gender, BMI, AS severity and the majority of comorbidities, but not for age (80±6 vs 73±7 years, p=0.001) or EuroSCORE (22±14 vs 7±3, p<0.001). Aortic valve mean pressure gradient decreased to a greater degree (p=0.017) 6 months post-TAVI (50±16 vs 21±8 mm Hg, p<0.001) compared to SAVR (55±20 vs 35±13 mm Hg, p<0.001). AR was reduced by 8% in each group, only reaching statistical significance for TAVI (p=0.003). Post-operative ventricular end-systolic volumes (ESVI) and mass improved in both groups (p<0.05), additionally SAVR reduced end-diastolic volumes (EDVI, p<0.001) and TAVI increased ejection fraction (EF, p=0.01), Abstract 042 table 1. Concentric remodelling (mass/EDV: 0.88±0.2 vs 0.73±0.2, p<0.001) and geometric wall function (thickness and thickening), improved post-TAVI (p<0.001) but not post-SAVR (p>0.05), Abstract 042 figure 1. MF burden and low EF were associated with greater post-operative remodelling in both groups by univariate analysis. EF remained an independent predictor on multivariate analysis (p<0.001). MF showed evidence of regression post-TAVI (p=0.04) but not post-SAVR. Abstract 042 Table 1 TAVI SAVR p Value† Baseline 6 months Baseline 6 months Function LVEDVI, ml/m2 94±18 90±20 92±19 74±12 0.04 LVESVI, ml/m2 46±18 41±17 44±22 32±6 0.19 LVSVI, ml/m2 48±10 50±10 49±8 42±7 0.14 LVEF, % 52±12 56±10 55±11 57±8 0.57 LVMI, g/m2 82±20 65±17 74±11 59±8 0.35 LVM/LVEDV, g/ml 0.88±0.2 0.73±0.2 0.80±0.1 0.81±0.2 <0.001 Aortic valve MPG, mm Hg 50±16 21±8 55±20 35±13 0.017 AR fraction, % 16±11 8±6 18±7 10±11 0.46 Abstract 042 Figure 1 Conclusion In high-risk AS patients, TAVI compared to SAVR produced a greater improvement in the aortic valve pressure gradient, concentric LV reverse remodelling, geometric wall function and MF. EF rather than MF was a more powerful predictor of this process.