P1485Left atrial reservoir function is associated with major adverse cardiac events in patients undergoing transarterial valve implantation for isolated severe aortic stenosis

• Published in: European heart journal 2019-10, Vol.40 (Supplement_1)
• Main Authors: Weber, J, Petillo, F, Pollack, S, Petrossian, G, Robinson, N, Thomas, S, Barasch, E
• Format: Article
• Language: English
• ISSN: 0195-668X; 1522-9645
• DOI: 10.1093/eurheartj/ehz748.0250

Abstract Background Left atrial (LA) reservoir function as measured by LA global longitudinal strain (LAGS) is an independent predictor of left ventricular (LV) performance and has prognostic value. Purpose To evaluate by speckle tracking echocardiography (STE), LAGS and other myocardial deformation indices changes after transarterial valve implantation (TAVI) for severe isolated aortic stenosis (AS) in relation to the outcome measures. Methods Of 995 pts who underwent TAVI at our Institution between 2017–18, 120 (age = 82.8±7.7 years, 74% female, AVAi = 0.37±0.09 cm2/m2, LVEF = 61.6±11.3%, no > than 2+ mitral or aortic regurgitation, all in NSR) underwent 2-D echocardiography and STE, pre (21±34 days) and post (16±27 days) TAVI. LAGS was measured at QRS onset, and LV global longitudinal strain (LVGS) and RV free wall strain were recorded. The velocity index = peak vel LVOT/AV. Phillips IE 33 scanners (frame rates 60–80 Hz) were used and one observer analyzed data on QLAB software. The median follow-up was 208 days (range 20–763). The outcome variable was a composite of death, atrial fibrillation and hospitalization for heart failure (MACE). Univariate and multivariable logistic regression were used to determine independent predictors of LA, LV and RV free wall global strain changes (covariates; age, sex, BSA, LVEF, systolic blood pressure, LA volume index) and, separately, for predictors of MACE (covariates; age, sex, AVA index, LVEF and E/e'). Intra- and interclass correlation coefficients (ICC) were calculated. Results The intra- and inter-observer ICC was 0.70–0.90 and 0.90–0.95, respectively. In the absence of LA volume change, LAGS improved post TAVI in 54% of pts. Overall, mean change was 2.2±11.6% (95% CI; 0.05, 4.3) and it was significantly associated in multivariable analysis with RV free wall strain (OR=2.7, 95% CI; 1.2, 6), velocity index (OR=0.4, 95% CI; 0.2, 1), LVEF (OR= 0.3, 95% CI; 0.2, 0.8) and LVGS (OR=3.8, 95% CI; 1.4, 10), yielding together an AUC of 0.90. LVGS improved in 64% of pts by −2.8±7.5%, (95% CI: −4.2, −1.5) and the velocity index independently predicted the LVGS change (OR = 0.6; 95% CI: 0.4, 0.9). The other deformation indices did not significantly change. At follow-up, there were 6 hospitalizations for heart failure, 5 atrial fibrillation events and 6 deaths. At multivariable logistic regression analysis, post TAVI LAGS was the only variable independently predicting MACE (OR (in units of 1%) = 0.90, 95% CI; 0.82, 0.98), es