Aims Dynamic left ventricular outflow tract obstruction (LVOTO) is a hemodynamically significant complication following transcatheter aortic valve replacement (TAVR) that remains difficult to predict with conventional transthoracic echocardiography (TTE). We examined whether a deep learning (DL) model developed for LVOTO detection in hypertrophic cardiomyopathy (HCM) could predict post-TAVR LVOTO from pre-TAVR TTE in patients with severe aortic stenosis (AS).

Methods and Results In this retrospective study of 302 consecutive patients undergoing TAVR for severe AS, a pre-trained DL model was applied to pre-TAVR TTE to generate a patient-level DL index of LVOTO (DLi-LVOTO; range 0–100). Post-TAVR LVOTO was defined as a peak pressure gradient ≥30 mmHg on follow-up TTE. Logistic regression and receiver operating characteristic analyses assessed the association and discriminative performance of DLi-LVOTO. Pre-TAVR LVOTO was present in 32 patients (10.6%) and post-TAVR LVOTO in 35 (11.6%). Follow-up TTE was performed at a median of 47 days (IQR 37–63) after TAVR, with the majority of TTE (216 of 302, 71.5%) performed within 2 months. DLi-LVOTO was significantly higher in patients with LVOTO at both pre- and post-TAVR TTE (all p<0.001). In multivariable analysis, DLi-LVOTO remained independently associated with post-TAVR LVOTO even after adjusting for conventional TTE parameters and pre-TAVR LVOTO (adjusted OR 1.29, 95% CI 1.06–1.56 per 10-score increase, p=0.011), with an AUROC of 0.78 (95% CI 0.72–0.85). Among patients without pre-TAVR LVOTO, DLi-LVOTO retained independent predictive value (adjusted OR 1.56, 95% CI 1.19–2.06, p=0.001; AUROC 0.84, 95% CI 0.77–0.91).

Conclusion A DL model originally trained in HCM patients independently predicts post-TAVR LVOTO from pre-TAVR TTE, including in patients without pre-existing LVOTO, suggesting it captures hemodynamic features beyond conventional echocardiographic assessment.

Y.E.Y, J.K., J.J., Y.J., and S.A.L. are currently affiliated with Ontact Health Co, Ltd. J.J., J.K., and S.A.L are co-inventors on a patent related to this work filed by Ontact Health Co, Ltd (Method for Providing Information for the Prediction of Left Ventricular Outflow Tract Obstruction in Hypertrophic Cardiomyopathy). Y.E.Y holds equity in Ontact Health Co, Ltd. The other authors have no conflicts of interest to declare.

This work was supported by a grant from the Institute of Information & communications Technology Planning & Evaluation (IITP) funded by the Korea government (Ministry of Science and ICT) (No.2022000972, Development of a Flexible Mobile Healthcare Software Platform Using 5G MEC).

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

The study protocol complied with the Declaration of Helsinki and was approved by the Institutional Review Board of Seoul National University Bundang Hospital (approval number B-2602-1029-102). The requirement for informed consent was waived due to the retrospective design of the study.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

The data underlying this study cannot be made publicly available due to ethical restrictions set by the IRB of the study institution; i.e., public availability would compromise patient confidentiality and participant privacy. Please contact the corresponding author (yeonyeeyoongmail.com) to request the minimal anonymized dataset. Researchers with additional inquiries about the deep learning model developed in this study are also encouraged to reach out to the corresponding author.