Employment of artificial intelligence for an unbiased evaluation regarding the recovery of right ventricular function after mitral valve transcatheter edge-to-edge repair

Vera Fortmeier, Amelie Hesse, Teresa Trenkwalder, Márton Tokodi, Attila Kovács, Elena Rippen, Jule Tervooren, Michelle Fett, Gerhard Harmsen, Shinsuke Yuasa, Moritz Kühlein, Héctor Alfonso Alvarez Covarrubias, Moritz von Scheidt, Ferdinand Roski, Muhammed Gerçek, Tibor Schuster, N. Patrick Mayr, Erion Xhepa, Karl Ludwig Laugwitz, Michael JonerVolker Rudolph, Mark Lachmann

Research output: Contribution to journalArticlepeer-review

Abstract

Aims: Long-standing severe mitral regurgitation (MR) leads to left atrial (LA) enlargement, elevated pulmonary artery pressures, and ultimately right heart failure. While mitral valve transcatheter edge-to-edge repair (M-TEER) alleviates left-sided volume overload, its impact on right ventricular (RV) recovery is unclear. This study aims to use both conventional echocardiography and artificial intelligence to assess the recovery of RV function in patients undergoing M-TEER for severe MR. Methods and results: The change in RV function from baseline to 3-month follow-up was analysed in a dual-centre registry of patients undergoing M-TEER for severe MR. RV function was conventionally assessed by measuring the tricuspid annular plane systolic excursion (TAPSE). Additionally, RV function was evaluated using a deep learning model that predicts RV ejection fraction (RVEF) based on two-dimensional apical four-chamber view echocardiographic videos. Among the 851 patients who underwent M-TEER, the 1-year survival rate was 86.8%. M-TEER resulted in a significant reduction in both LA volume and estimated systolic pulmonary artery pressure (sPAP) levels (median LA volume: from 123 ml [interquartile range, IQR 92–169 ml] to 104 ml [IQR 78–142 ml], p < 0.001; median sPAP: from 46 mmHg [IQR 35–58 mmHg] to 41 mmHg [IQR 32–54 mmHg], p = 0.036). In contrast, TAPSE remained unchanged (median: from 17 mm [IQR 14–21 mm] to 18 mm [IQR 15–21 mm], p = 0.603). The deep learning model confirmed this finding, showing no significant change in predicted RVEF after M-TEER (median: from 43.1% [IQR 39.1–47.4%] to 43.2% [IQR 39.2–47.2%], p = 0.475). Conclusions: While M-TEER improves left-sided haemodynamics, it does not lead to significant RV function recovery, as confirmed by both conventional echocardiography and artificial intelligence. This finding underscores the importance of treating patients before irreversible right heart damage occurs.

Original languageEnglish
JournalEuropean Journal of Heart Failure
DOIs
StateAccepted/In press - 2025

Keywords

  • Deep learning
  • Echocardiography
  • Mitral regurgitation
  • Right ventricular dysfunction
  • Transcatheter edge-to-edge repair

Fingerprint

Dive into the research topics of 'Employment of artificial intelligence for an unbiased evaluation regarding the recovery of right ventricular function after mitral valve transcatheter edge-to-edge repair'. Together they form a unique fingerprint.

Cite this