Impact of left atrial strain assessed with feature-tracking computed tomography on long-term mortality after transcatheter aortic valve implantation
Abstract Background Aortic stenosis (AS) induces left atrial (LA) remodeling through the increase of left ventricular (LV) filling pressure. Peak left atrial longitudinal strain (PALS) has been proposed as a prognostic marker in patients with AS. Novel feature-tracking (FT) software allows to assess LA strain from multidetector computed tomography (MDCT) dataset. Purpose To investigate the association between PALS using FT MDCT and moratlity in patients who underwent transcatheter aortic valve implantation (TAVI). Methods A total of 369 Patients (mean 80±7 years, 51% male) who underwent preprocedual MDCT before TAVI and had suitable data for measureing PALS using dedicated FT software were included. Patients were classified into 4 groups according to PALS quartiles; PALS more than 19.3% (Q1), 19.3% or less to more than 15.0% (Q2), 15.0% or less to more than 9.1% (Q3), and 9.1% or less (Q4). The primary outcome was all-caurse mortality. Results During median follow-up of 45 [22 - 68] months, 124 patients (34%) were died. On multivariable Cox regression analysis, PALS is an independently associated with all-cause mortality (HR: 0.958 [95% CI: 0.925–0.993], P=0.006). Kaplan-Meier analysis showed the worse outcome of the quatile with more impaired PALS (Logrank P<0.001). Compared to Q1, Q3 and Q4 had higher risk of mortality after TAVI (HR: 2.475 [95% CI: 1.411–4.340] for Q3, HR: 3.253 [95% CI: 1.878–5.633] for Q4). Conclusion In this retrospective study, PALS measured with FT MDCT was strongly associated with all-cause mortality after TAVI. LA functial assessment using MDCT may have a importan role for risk stratification in patients referred to TAVI. FUNDunding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): ESC research grant 2018 K-M curve according to PALS quartiles