Relief of dyspnoea after TAVI is associated with improvement in left ventricular systolic function
C S Venema, K Panaou, K H Van Bergeijk, J A Krikken, H W Van der Werf, T L Braber, A F M Van den Heuvel, Y L Douglas, E Lipsic, A A Voors, J J WykrzykowskaAbstract
Background
Relief of symptoms is a key goal of transcatheter aortic valve implantation (TAVI) for severe, symptomatic aortic stenosis. Patients with heart failure with preserved ejection fraction (HFpEF) are at higher risk of residual symptoms after TAVI, but whether the myocardial response after TAVI is associated with symptomatic relief remains unclear. We hypothesized that symptomatic relief relates to improvement in cardiac structure and function.
Purpose
To determine whether symptomatic response measured by New York Heart Association (NYHA) class is associated with longitudinal changes in echocardiographic measures. In patients with preserved left ventricular ejection fraction (LVEF ≥ 50%), we evaluated whether HFpEF is associated with an unfavourable echocardiographic response post-TAVI.
Methods
TAVI patients underwent NYHA assessment at baseline and at 30 days. Symptomatic relief was defined as NYHA class I at 30 days. Echocardiograms at baseline, before discharge and beyond 3 months post-TAVI were collected and analysed using an automated, validated, deep-learning workflow (Us2.ai, Singapore). HFpEF was defined as an H2FPEF score ≥ 6. Trajectories of echocardiographic parameters over time were modelled using linear mixed-effects models using a spline term for time to allow for non-linear trajectories. Differences were compared at 1 month before TAVI, and 6 and 12 months after TAVI using contrast analysis.
Results
Among 1,334 patients undergoing TAVI (median age 81 years [IQR: 76 – 84 years], 51% women), 933 (70%) achieved symptomatic relief at 30 days. Patients with persistent symptoms had lower peak aortic jet velocity before TAVI (3.86 vs. 4.08 m/s, p < 0.001). Symptomatic relief was significantly associated with improvement in left ventricular (LV) systolic function, measured by LVEF and global longitudinal strain (-16.0% in NYHA I vs. -14.9% in NYHA II-IV at 12 months, p = 0.017; Figure). In addition, improvement in RV systolic function (TAPSE) was associated with symptomatic relief. In contrast, change in diastolic function was not associated with symptomatic response. Also, no differences were observed in the hemodynamic effect of TAVI on the aortic valve parameters. Patients with persistent symptoms had higher NT-proBNP levels at 6 and 12 months post-TAVI. In patients with preserved LVEF (n = 707), 18% (n = 125) had HFpEF. Despite similar LVEF at baseline, patients with HFpEF had significantly worse LV systolic function post-TAVI.
Conclusion
In patients undergoing TAVI, symptomatic relief is associated with improvement in LV and RV systolic function. Persistent dyspnoea may be related to underlying HFpEF, which was also associated with worse LV systolic function post-TAVI. Trials targeting HFpEF therapy post-TAVI are warranted.Trajectories by symptomatic responseFor image description, please refer to the figure legend and surrounding text.Trajectories by HFpEF vs. no HFpEFFor image description, please refer to the figure legend and surrounding text.