Aims: Cardiac CT is increasingly applied for planning and follow-up of transcatheter aortic valve implantation (TAVI). However, there are no data available on reverse remodelling after TAVI assessed by CT. Therefore, we aimed to evaluate the predictors and the prognostic value of left ventricular (LV) reverse remodelling following TAVI using CT angiography.
Methods and results: We investigated 117 patients with severe, symptomatic aortic stenosis (AS) who underwent CT scanning before and after TAVI procedure with a mean follow-up time of 2.6 years after TAVI. We found a significant reduction in LV mass (LVM) and LVM indexed to body surface area comparing pre- vs. post-TAVI images: 180.5 ± 53.0 vs. 137.1 ± 44.8 g and 99.7 ± 25.4 vs. 75.4 ± 19.9 g/m2, respectively, both P < 0.001. Subclinical leaflet thrombosis (SLT) was detected in 25.6% (30/117) patients. More than 20% reduction in LVM was defined as reverse remodelling and was detected in 62.4% (73/117) of the patients. SLT, change in mean pressure gradient on echocardiography and prior myocardial infarction was independently associated with LV reverse remodelling after adjusting for age, gender, and traditional risk factors (hypertension, body mass index, diabetes mellitus, and hyperlipidaemia): OR = 0.27, P = 0.022 for SLT and OR = 0.22, P = 0.006 for prior myocardial infarction, OR = 1.51, P = 0.004 for 10 mmHg change in mean pressure gradient. Reverse remodelling was independently associated with favourable outcomes (HR = 0.23; P = 0.019).
Conclusion: TAVI resulted in a significant LVM regression on CT. The presence of SLT showed an inverse association with LV reverse remodelling and thus it may hinder the beneficial LV structural changes. Reverse remodelling was associated with improved long-term prognosis.
Keywords: CT angiography; left ventricular reverse remodelling; subclinical leaflet thrombosis; transcatheter aortic valve implantation.
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: [email protected].