Patient-specific computer simulation to predict long-term outcomes after transcatheter aortic valve replacement

Clinical value of CT-derived simulations of transcatheter-aortic-valve-implantation in challenging anatomies the PRECISE-TAVI trial

BACKGROUND

Preprocedural computed tomography planning improves procedural safety and efficacy of transcatheter aortic valve implantation (TAVI). However, contemporary imaging modalities do not account for device-host interactions.

AIMS

This study evaluates the value of preprocedural computer simulation with FEops HEARTguideTM on overall device success in patients with challenging anatomies undergoing TAVI with a contemporary self-expanding supra-annular transcatheter heart valve.

METHODS

This prospective multicenter observational study included patients with a challenging anatomy defined as bicuspid aortic valve, small annulus or severely calcified aortic valve. We compared the heart team's transcatheter heart valve (THV) planning decision based on (1) conventional multislice computed tomography (MSCT) and (2) MSCT imaging with FEops HEARTguideTM simulations. Clinical outcomes and THV performance were followed up to 30 days.

RESULTS

A total of 77 patients were included (median age 79.9 years (IQR 74.2-83.8), 42% male). In 35% of the patients, preprocedural planning changed after FEops HEARTguideTM simulations (change in valve size selection [12%] or target implantation height [23%]). A new permanent pacemaker implantation (PPI) was implanted in 13% and >trace paravalvular leakage (PVL) occurred in 28.5%. The contact pressure index (i.e., simulation output indicating the risk of conduction abnormalities) was significantly higher in patients with a new PPI, compared to those without (16.0% [25th-75th percentile 12.0-21.0] vs. 3.5% [25th-75th percentile 0-11.3], p < 0.01) The predicted PVL was 5.7 mL/s (25th-75th percentile 1.3-11.1) in patients with none-trace PVL, 12.7 (25th-75th percentile 5.5-19.1) in mild PVL and 17.7 (25th-75th percentile 3.6-19.4) in moderate PVL (p = 0.04).

CONCLUSION

FEops HEARTguideTM simulations may provide enhanced insights in the risk for PVL or PPI after TAVI with a self-expanding supra-annular THV in complex anatomies.

Sex-specific aortic valve calcifications in patients undergoing transcatheter aortic valve implantation

AIMS

To study sex-specific differences in the amount and distribution of aortic valve calcification (AVC) and to correlate the AVC load with paravalvular leakage (PVL) post-transcatheter aortic valve intervention (TAVI).

METHODS AND RESULTS

This registry included 1801 patients undergoing TAVI with a Sapien3 or Evolut valve in two tertiary care institutions. Exclusion criteria encompassed prior aortic valve replacement, suboptimal multidetector computed tomography (MDCT) quality, and suboptimal transthoracic echocardiography images. Calcium content and distribution were derived from MDCT. In this study, the median age was 81.7 (25th-75th percentile 77.5-85.3) and 54% male. Men, compared to women, were significantly younger [81.2 (25th-75th percentile 76.5-84.5) vs. 82.4 (78.2-85.9), P ≤ 0.01] and had a larger annulus area [512 mm2 (25th-75th percentile 463-570) vs. 405 mm2 (365-454), P < 0.01] and higher Agatston score [2567 (25th-75th percentile 1657-3913) vs. 1615 (25th-75th percentile 905-2484), P < 0.01]. In total, 1104 patients (61%) had none-trace PVL, 648 (36%) mild PVL, and 49 (3%) moderate PVL post-TAVI. There was no difference in the occurrence of moderate PVL between men and women (3% vs. 3%, P = 0.63). Cut-off values for the Agatston score as predictor for moderate PVL based on the receiver-operating characteristic curve were 4070 (sensitivity 0.73, specificity 0.79) for men and 2341 (sensitivity 0.74, specificity 0.73) for women.

CONCLUSION

AVC is a strong predictor for moderate PVL post-TAVI. Although the AVC load in men is higher compared to women, there is no difference in the incidence of moderate PVL. Sex-specific Agatston score cut-offs to predict moderate PVL were almost double as high in men vs. women.

Modelling blood flow in patients with heart valve disease using deep learning: A computationally efficient method to expand diagnostic capabilities in clinical routine

Introduction

The computational modelling of blood flow is known to provide vital hemodynamic parameters for diagnosis and treatment-support for patients with valvular heart disease. However, most diagnosis/treatment-support solutions based on flow modelling proposed utilize time- and resource-intensive computational fluid dynamics (CFD) and are therefore difficult to implement into clinical practice. In contrast, deep learning (DL) algorithms provide results quickly with little need for computational power. Thus, modelling blood flow with DL instead of CFD may substantially enhances the usability of flow modelling-based diagnosis/treatment support in clinical routine. In this study, we propose a DL-based approach to compute pressure and wall-shear-stress (WSS) in the aorta and aortic valve of patients with aortic stenosis (AS).

Methods

A total of 103 individual surface models of the aorta and aortic valve were constructed from computed tomography data of AS patients. Based on these surface models, a total of 267 patient-specific, steady-state CFD simulations of aortic flow under various flow rates were performed. Using this simulation data, an artificial neural network (ANN) was trained to compute spatially resolved pressure and WSS using a centerline-based representation. An unseen test subset of 23 cases was used to compare both methods.

Results

ANN and CFD-based computations agreed well with a median relative difference between both methods of 6.0% for pressure and 4.9% for wall-shear-stress. Demonstrating the ability of DL to compute clinically relevant hemodynamic parameters for AS patients, this work presents a possible solution to facilitate the introduction of modelling-based treatment support into clinical practice.

Assessment of valve implantation in the descending aorta as an alternative for aortic regurgitation patients not treatable with conventional procedures

BACKGROUND

Aortic Regurgitation (AR) produces the entrance of an abnormal amount of blood in the left ventricle. This disease is responsible for high morbidity and mortality worldwide and may be caused by an aortic valve dysfunction. Surgical and transcatheter aortic valve replacement (TAVR) are the current options for treating AR. They have replaced older procedures such as Hufnagel's one. However, some physicians have reconsidered this procedure as a less aggressive alternative for patients not eligible for surgical or TAVR. Although Hufnagel suggested a 75% regurgitation reduction when a valve is placed in the descending aorta, a quantification of this value has not been reported.

METHODS

In this paper, CFD/FSI numerical simulation is conducted on an idealized geometry. We quantify the effect of placing a bileaflet mechanical heart valve in the descending aorta on a moderate-severe AR case. A three-element Windkessel model is employed to prescribe pressure outlet boundary conditions. We calculate the resulting flow rates and pressures at the aorta and first-generation vessels. Moreover, we evaluate several indices to assess the improvement due to the valve introduction.

RESULTS AND CONCLUSIONS

Regurgitation fraction (RF) is reduced from 37.5% (without valve) to 18.0% (with valve) in a single cardiac cycle. This reduction clearly shows the remarkable efficacy of the rescued technique. It will further ameliorate the left ventricle function in the long-term. Moreover, the calculations show that the implantation in that location introduces fewer incompatibilities' risks than a conventional one. The proposed methodology can be extended to any particular conditions (pressure waveforms/geometry) and is designed to assess usual clinical parameters employed by physicians.

An artificial intelligence-based platform for automatically estimating time-averaged wall shear stress in the ascending aorta

Aims

Aortopathies are a series of disorders requiring multiple indicators to assess risk. Time-averaged wall shear stress (TAWSS) is currently considered as the primary indicator of aortopathies progression, which can only be calculated by Computational Fluid Dynamics (CFD). However, CFD's complexity and high computational cost, greatly limit its application. The study aimed to construct a deep learning platform which could accurately estimate TAWSS in ascending aorta.

Methods and results

A total of 154 patients who had thoracic computed tomography angiography were included and randomly divided into two parts: training set (90%, n = 139) and testing set (10%, n = 15). TAWSS were calculated via CFD. The artificial intelligence (AI)-based model was trained and assessed using the dice coefficient (DC), normalized mean absolute error (NMAE), and root mean square error (RMSE). Our AI platform brought into correspondence with the manual segmentation (DC = 0.86) and the CFD findings (NMAE, 7.8773% ± 4.7144%; RMSE, 0.0098 ± 0.0097), while saving 12000-fold computational cost.

Conclusion

The high-efficiency and robust AI platform can automatically estimate value and distribution of TAWSS in ascending aorta, which may be suitable for clinical applications and provide potential ideas for CFD-based problem solving.