Bioprosthetic aortic valve diameter and thickness are directly related to leaflet fluttering: Results from a combined experimental and computational modeling study.

OBJECTIVE: Bioprosthetic heart valves (BHVs) are commonly used in surgical and percutaneous valve replacement. The durability of percutaneous valve replacement is unknown, but surgical valves have been shown to require reintervention after 10 to 15 years. Further, smaller-diameter surgical BHVs generally experience higher rates of prosthesis-patient mismatch, which leads to higher rates of failure. Bioprosthetic aortic valves can flutter in systole, and fluttering is associated with fatigue and failure in flexible structures. The determinants of flutter in BHVs have not been well characterized, despite their potential to influence durability.
METHODS: We use an experimental pulse duplicator and a computational fluid-structure interaction model of this system to study the role of device geometry on BHV dynamics. The experimental system mimics physiological conditions, and the computational model enables precise control of leaflet biomechanics and flow conditions to isolate the effects of variations in BHV geometry on leaflet dynamics.
RESULTS: Both experimental and computational models demonstrate that smaller-diameter BHVs yield markedly higher leaflet fluttering frequencies across a range of conditions. The computational model also predicts that fluttering frequency is directly related to leaflet thickness. A scaling model is introduced that rationalizes these findings.
CONCLUSIONS: We systematically characterize the influence of BHV diameter and leaflet thickness on fluttering dynamics. Although this study does not determine how flutter influences device durability, increased flutter in smaller-diameter BHVs may explain how prosthesis-patient mismatch could induce BHV leaflet fatigue and failure. Ultimately, understanding the effects of device geometry on leaflet kinematics may lead to more durable valve replacements.

Experimental and computational models quantify influences of device geometry on valve dynamics.

Smaller-diameter bioprosthetic heart valves generate markedly higher fluttering frequencies in both experimental and computational models, independent of operating and flow conditions.

Fluttering can impair the durability of flexible structures. This study demonstrates that bioprosthetic heart valves with smaller diameters and/or thicker leaflets generate higher fluttering frequencies in experimental and computational pulse duplicators. Extensions of this work may ultimately lead to new device design targets or selection guidelines to improve the durability of valve replacement.

See Commentaries on pages 82 and 84.

Bioprosthetic heart valves (BHVs) are widely used for valve replacement because they provide favorable hemodynamics and typically only require patients to receive antiplatelet therapy. The chemically fixated tissues used to construct BHVs can deteriorate over time, and bioprosthetic valves are known to have a durable life span that averages 10 to 15 years. With the recent introduction of transcatheter aortic valve replacement (TAVR), BHV use continues to increase, including in younger and lower surgical risk patients, because of improvements in patient outcomes, progress in valve design, and the growing availability of valve-in-valve TAVR if the first BHV fails. Advancing our understanding of the mechanisms that determine BHV durability remains crucial to improving patient care.

BHV leaflets can flutter,4, 5, 6, 7 and it is well known that fluttering is associated with accelerated fatigue and premature failure in thin flexible structures. The influence of fluttering on durability has been studied in diverse systems,, but the role of fluttering in BHV durability has not been clearly established. Further, prior studies on the influence of valve geometry on leaflet kinematics and on the determinants of BHV flutter are lacking. This study aims to begin to link our more theoretical understanding of the influence of fluttering in the durability of natural and manufactured systems to potential avenues to improve BHV durability by systematically characterizing the roles of device geometry in leaflet kinematics.

Leaflet fluttering is challenging to study in vivo. Consequently, this work uses a well characterized experimental pulse duplicator platform and a computational fluid-structure interaction model of this system to study the effects of valve geometry on leaflet dynamics. Pulse duplicators are widely used to assess prosthetic valve performance. Computational models complement experiments by enabling the assessment of device performance under a broader range of conditions. Further, operating conditions are tightly controlled and trivially reproducible in a computer model, allowing for the elimination of variations both within a given experiment and between different experiments.

Key findings of this study are that BHVs with smaller diameter or thicker leaflets show markedly higher fluttering frequencies. A scaling model is proposed that rationalizes both findings. Further, the observed relationship between BHV diameter and fluttering frequency holds under consistent operating conditions (flow rates and pressures) and under consistent flow conditions (characterized by Reynolds number). Although the influence of BHV size on mortality after aortic valve replacement is unknown,, it is known that smaller-diameter BHVs lead to higher rates of prosthesis–patient mismatch (PPM), which, in turn, leads to higher rates of failure.16, 17, 18, 19, 20 However, it is unknown how PPM influences leaflet damage and device failure. Understanding both the determinants and influences of fluttering could ultimately influence patient-specific surgical planning and device selection as well as the design of novel devices, such as polymeric valves.

Methods

Experimental Pulse Duplicator System

Experimental studies used a customized version of the ViVitro Pulse Duplicator System (ViVitro Labs, Inc, Victoria, British Columbia, Canada) (Figure 1, A), which is used and accepted by regulatory agencies, including the US Food and Drug Administration., The customized pulse duplicator includes an electro-optical subsystem to assess projected dynamic valve area (PDVA). We used Labcor DKA valves (Labcor Laboratórios Ltda, Belo Horizonte, Brazil) with diameters 21 mm (DKA015849), 25 mm (DKA015141), and 27 mm (DKA015562), which have bovine pericardial leaflets that are externally wrapped around their frames. Flow and pressure signals are filtered at 100 Hz, and PDVA signals are not filtered. We use 10 consecutive cycles of these experimental signals for each device and report average measurements and cycle-to-cycle variations using confidence intervals. The test fluid was saline, which is accepted under ISO 5840-3 and widely used to assess BHV performance. We perform experiments using a pulse rate of 70 bpm. Additional pressure and flow waveforms were obtained using a glycerin-based blood analog in a commercial ViVitro Pulse Duplicator at 60 bpm.

Figure 1

Experimental and computational pulse duplicators. A, Customized pulse duplicator with electro-optical subsystem for measuring aortic valve projected dynamic valve area. B, Computer model of the aortic valve test section in the pulse duplicator with pericardial bioprosthetic heart valve (BHV) and reduced-order models of the upstream and downstream system components. C, Three-dimensional rendering of the model BHV leaflets. Leaflet kinematics are detailed on the highlighted cross-sections in Figures E5 and E6.

Figure E5

Detailed leaflet kinematics obtained from the computer model with different valve diameters. Time series of leaflet cross sections (see Figure 1, C) for different valve diameters described in Figure 2. Red boxes indicate the times when the peak tip displacement of the leaflet occurs. Note that complex flow patterns result in only quasiperiodic leaflet kinematics. The smaller-diameter valve (21 mm) shows more frequent leaflet bending than the larger-diameter valve (27 mm).

Figure E6

Detailed leaflet kinematics obtained from the computational model with different leaflet thicknesses. Time series of leaflet cross sections (see Figure 1, C) for different valve thicknesses for a fixed valve diameter (25 mm) described in Figure 6. Red boxes indicate the times when the peak tip displacement of the leaflet occurs. The valve with the thickest leaflets (0.6 mm) shows more frequent leaflet bending than the thinnest leaflets (0.2 mm).

Computational Model of BHV Dynamics

Computer simulations used an fluid-structure interaction model of the aortic valve test section of the pulse duplicator (Figure 1, B) described previously and detailed in Appendix 1. (See also Video 1.) We construct a model bovine pericardial BHV with variable diameter and leaflet thickness (Figure 1, C). We use both saline and glycerin in the simulations, with densities ρ = 1.0 and 1.17 g/cm3 and dynamic viscosities μ = 1.0 and 3.6 cP, respectively. Computer simulations use pulse rates consistent with the corresponding experiments.

Flow Characterization

To normalize flow conditions between devices, we use the peak Reynolds number, which is a ratio of inertial and viscous fluid forces. Here, Qpeak is the peak volumetric flow rate, and D and A are the geometrical diameter and cross-sectional area of the valve. Physiological Reynolds numbers in the aortic root and ascending aorta range from 5000 to 7000, which are in the turbulent flow regime. Notice that maintaining a constant value of Repeak as the device size decreases requires decreasing the flow rate.

Frequency Analysis

Fluttering frequencies are assessed from PDVA and leaflet tip position time series data. We use the MATLAB Signal Processing Toolbox (MathWorks, Inc, Natick, Mass) to determine the power spectral density. Because the highest peak in the power spectral density corresponds to the zero-frequency content, we use the second highest peak to determine the dominant frequency characterizing leaflet fluttering.

Statistical Analysis

We use linear regression to model the relationship between dominant fluttering frequency and valve diameter or leaflet thickness. To quantify goodness of fit, we use the coefficient of determination,in which y are the observed dominant fluttering frequencies for each valve diameter or leaflet thickness, is the average of y, and are the predicted dominant fluttering frequencies from the linear fit for each y.

We also use a scaling model to rationalize relationships between fluttering frequency (f) and valve diameter or leaflet thickness, which both influence orifice area (ie, PDVA) and average leaflet tip displacement (dtip):

A brief derivation of this relation is provided in Appendix 1, G.

Results

We first establish correspondence of the experimental and computational models for the three BHV devices available for experimental analysis. PDVA measurements are available for both experimental and computational platforms, and tip displacement measurements are available in the computational model. The experimental operating conditions are similar for the different devices, and operating conditions for the computational models are consistent with the corresponding experiment in each case. Figure 2 compares the measurements for corresponding experimental and computational models. The simulated pressure and flow rates are in excellent agreement with the experimental data (see Figure E4). The dominant fluttering frequencies from experimental and simulated PDVA signals and simulated tip displacement signals, respectively, are 70.97 ± 2.11 Hz, 59.63 Hz, and 59.26 Hz (21 mm); 32.74 ± 3.14 Hz, 38.62 Hz, and 32.88 Hz (25 mm); and 26.03 ± 1.04 Hz, 21.05 Hz, and 26.32 Hz (27 mm). This demonstrates excellent qualitative agreement and reasonable quantitative agreement in leaflet fluttering frequencies for each device. Further, both experimental and computational results show that much higher fluttering frequencies occur with smaller valve diameters. (See also Video 2.) Experimental and simulated PDVA signals and tip displacement signals show similar fluttering frequency responses, but it is clear that the tip displacement waveforms more directly capture the fluttering dynamics. Consequently, we use tip displacement waveforms for all subsequent spectral analyses. (See also Video 3 and Appendix 1, E.) Figure 3 shows that fluttering frequency is negatively related to BHV diameter, with proportionality coefficients for a linear fit of –5.65 Hz/mm (R = 0.98) and –7.79 Hz/mm (R = 0.96), respectively, for computational and experimental models.

Figure 2

Analysis of experimental and computational leaflet kinematics. Experimental measurements show variations over 10 consecutive cycles, with shaded regions showing where 95% of the data fall. For each available valve diameter, the computer model matches the experimental operating conditions, which are different for each valve. Panels A through C compare simulated results to the experimental data for projected dynamic valve area (PDVA), and insets show the simulated displacement of the leaflet tip from the center of the valve. Panels D through F show frequency analyses. Dominant fluttering frequencies from experimental and simulated PDVA signals and simulated tip displacement signals are, respectively, D, 70.97 ± 2.11 Hz, 59.63 Hz, 59.26 Hz; E, 32.74 ± 3.14 Hz, 38.62 Hz, 32.88 Hz; and F, 26.03 ± 1.04 Hz, 21.05 Hz, 26.32 Hz. Smaller valves clearly show markedly higher fluttering frequencies.

Figure E4

Analysis of experimental and computational pressure and volumetric flow rates. Experimental measurements show variations over 10 consecutive cycles, with shaded regions showing where 95% of the data fall. For each available valve diameter, the computer model matches the experimental operating conditions, which are different for each valve. We compare simulated (A through C) pressure waveforms and (D through F) flow rates to the experimental data. The simulated pressure and flow rates are in excellent agreement with the experimental data.

Figure 3

Comparison of linear regressions of fluttering frequency versus valve diameter between simulations and experiments. Blue circles represent dominant fluttering frequency data from simulations that match the different experimental operating conditions of each device. Red triangles represent dominant fluttering frequency data with respect to valve diameters obtained from experimental projected dynamic valve area measurements. Linear regressions demonstrate that both simulation and experiment show negative relations between frequency response and valve diameter, with proportionality coefficients –5.65 Hz/mm and –7.79 Hz/mm, respectively, for simulation (blue solid) and experiment (red dashed).

Figure E7 compares PDVA and tip displacements obtained using the computational model for different valve diameters, as in Figure 2, but now using flow rates and driving and loading pressures that are consistent with the experimental conditions used with the 21 mm valve to eliminate variations in operating conditions. Fluttering frequencies determined from tip displacement waveforms are 59.26 Hz (21 mm), 32.88 Hz (25 mm), and 26.32 Hz (27 mm), respectively, which are identical to the results obtained in Figure 2. (See also Video 4.) The proportionality coefficients for a linear fit is again –5.65 Hz/mm (R = 0.98); see Figure 4, A. Because the frequencies are the same, this clearly demonstrates that variations in leaflet flutter are maintained if operating conditions are normalized across valve sizes.

Figure E7

Analysis of simulated leaflet kinematics for valves with different diameters under consistent operating conditions. A through C, projected dynamic valve area (PDVA) and tip displacements are obtained from the computational models for each valve diameter using volumetric flow rates and pressure loads corresponding to the 21 mm valve in Figure 2, A. D through F, Frequency analyses quantify dominant fluttering frequencies: D, 59.26 Hz; E, 32.88 Hz; and F, 26.32 Hz. These frequencies are identical to those reported in Figure 2.

Figure 4

Linear regressions of fluttering frequency versus valve diameter and leaflet thickness under consistent operating conditions. Blue circles represent dominant fluttering frequency data from simulations with respect to (A) valve diameter and (B) leaflet thickness under consistent operating conditions. Linear regressions demonstrate that simulations show (A) negative relations between frequency response and valve diameter, with proportionality coefficient –5.65 Hz/mm, and (B) positive relations between frequency response and leaflet thickness, with proportionality coefficient 41.1 Hz/mm.

We next use the computational model to consider the effect of leaflet thickness on device kinematics at a fixed device diameter of 25 mm. Figure E8 shows that valves with thicker leaflets open less and flutter at higher frequencies. (See also Video 5.) The dominant fluttering frequencies are 27.40 Hz (0.2 mm), 32.88 Hz (0.4 mm), and 43.84 Hz (0.6 mm). Fluttering frequency is positively related to BHV leaflet thickness, with proportionality coefficients for a linear fit of 41.1 Hz/mm (R = 0.96); see Figure 4, B. The dynamics of the thicker leaflets are consistent with those of a valve of normal thickness and smaller diameter, whereas the thinner leaflets yield kinematics like a valve of normal thickness and larger diameter. (See also Video 6.) Changes in leaflet thicknesses can influence leaflet stresses, and it is well established that larger diastolic stresses, especially near the commissures, are associated with fatigue., Computational stress analyses detailed in Appendix 1, H, recapitulate prior findings, that thinner leaflets experience larger commissural stresses in diastole. Our model is also in agreement with prior results showing that diastolic leaflet stresses decrease with increasing BHV diameter (Figure E10).

Figure E8

Analysis of simulated leaflet kinematics for valves with different leaflet thicknesses under consistent operating condition. A though C, Projected dynamic valve area (PDVA) and tip displacements are obtained from the computational models using the operating condition for the 25 mm valve but with varying leaflet thicknesses. D though F, Frequency analyses quantify the dominant fluttering frequencies: D, 27.40 Hz (0.2 mm); E, 32.88 Hz (0.4 mm); and F, 43.84 Hz (0.6 mm). These results suggest that at a fixed diameter, valves with thinner leaflets flutter at lower frequencies.

Figure E10

Stress analyses for 25 mm valves with different leaflet thicknesses. A, Comparison of von Mises stress between valves with a fixed thickness (0.4 mm) and different diameters (21 mm, 25 mm, 27 mm). B, Comparison of von Mises stress between valves with a fixed diameter (25 mm) and different thicknesses (0.2 mm, 0.4 mm, 0.6 mm). The results in panel A indicate that for a fixed thickness, the larger valve experiences smaller stress on the leaflets during diastole. This suggests that larger diameter valves may have an advantage in durability both during systole and diastole. The results in panel B indicate that for a fixed diameter, the thinner valve leaflets experience higher stress loads.

Figure 5 compares predictions of our computer simulations to the simple scaling relation detailed in Appendix 1, G. Both models yield consistent predictions in the relationships between fluttering frequency and valve diameter (Figure 5, A) and leaflet thickness (Figure 5, B).

Figure 5

Comparison of leaflet fluttering frequency versus scaling model for different bioprosthetic heart valve (BHV) diameters and leaflet thicknesses. Panel A compares fluttering frequencies obtained for different BHV diameters to frequencies determined from the scaling law. Panel B compares fluttering frequencies obtained for different BHV leaflet thicknesses to frequencies determined from the scaling law. The results are consistent in both cases.

Experimental data characterizing the role of device geometry on leaflet kinematics are only available for cases that use saline as the test fluid. Consequently, we use our computer model to study leaflet fluttering using parameters consistent with a glycerin-based blood analog, which provides a more physiological Reynolds number than saline. We perform simulations with consistent operating conditions (flow rates and pressure differences) and consistent peak Reynolds numbers for a broad range of device sizes (19, 21, 23, 25, and 27 mm). In these computer experiments, the volumetric flow rate specified at the pump is reduced by the ratio between the control (27 mm) to the valve diameter of interest. This reduces the peak flow rate by the same ratio, yielding the same values of Repeak for each device size. Figure E9 shows that after matching the flow conditions for the 19-, 21-, 23-, and 25-mm cases to the 27-mm case, the fluttering frequencies are 47.86 Hz (19 mm), 35.82 Hz (21 mm), 31.75 Hz (23 mm), 25.24 Hz (25 mm), and 12.12 Hz (27 mm). If we instead match operating conditions, but not flow conditions, the fluttering frequencies are 59.70 Hz (19 mm), 46.75 Hz (21 mm), 32.00 Hz (23 mm), 25.32 Hz (25 mm), and 12.12 Hz (27 mm). (See also Video 7.) Figure 6 compares linear regressions of fluttering frequency with respect to valve diameter for consistent operating and flow conditions. Frequency response is negatively related to valve diameter in both cases, with coefficients –5.83 Hz/mm (R = 0.99) and –4.10 Hz/mm (R = 0.97) for consistent operating and flow conditions, respectively. These results indicate that fluttering frequencies differ markedly with valve diameter, even under identical flow conditions. The proportionality coefficients are essentially the same for glycerin (–5.83 and –4.10 Hz/mm) and for saline (–5.65 Hz/mm), which suggests that relative differences in fluttering frequencies are largely independent of flow conditions and are, instead, determined primarily by device geometry.

Figure E9

Analysis of simulated leaflet kinematics for valves with different diameters at physiological Reynolds numbers with consistent operating and flow conditions. A though E, frequency analyses of leaflet fluttering obtained using a glycerin-based blood analog. Blue solid lines represent the simulation results obtained for a fixed operating condition but varying flow conditions (quantified by Reynolds number [Repeak]). Red dashed lines represent results in which the driving condition is modified to match Repeak. The dominant fluttering frequencies for different Reynolds number cases are: A, 59.70 and 47.86 Hz; B, 46.75 and 35.82 Hz; C, 32.00 and 31.75 Hz; D, 25.32 and 25.24 Hz; and E, 12.12 Hz. Smaller-diameter valves show higher frequency leaflet fluttering.

Figure 6

Linear regression of fluttering frequency versus valve diameter at physiological Reynolds numbers with consistent operating and flow conditions. Blue circles represent dominant fluttering frequency data with respect to valve diameters obtained for a fixed operating condition but varying flow conditions (quantified by Reynolds number). Red triangles represent dominant fluttering frequency data with respect to valve diameters in which the driving condition is modified to match Reynolds number. Linear regressions show negative relations between frequency response and valve diameter, with proportionality coefficients –5.83 Hz/mm (blue solid) and –4.10 Hz/mm (red dashed) for consistent operating conditions and flow conditions, respectively.

Discussion

Using both experimental and computational models, we consistently find that smaller-diameter pericardial aortic valves show substantially higher leaflet fluttering frequencies. Further, our computer model predicts that at a fixed device diameter, thinner leaflets will yield lower fluttering frequencies than thicker leaflets under consistent volumetric flow rates and pressure differences. Differences in operating or flow conditions can impact fluttering dynamics, but we confirm that under similar operating conditions, fluttering frequency is negatively related to the valve diameter, with proportionality coefficients from a linear regression of –5.65 Hz/mm (R = 0.98) and –5.83 Hz/mm (R = 0.99), respectively, for saline and glycerin. Differences in BHV fluttering have been ascribed to variations in pressures and flow rates in the clinical literature. Our tests using a pulse duplicator clearly indicate that fluttering frequencies differ with valve geometry, even under identical operating or flow conditions, as summarized in Figure 7.

Figure 7

Graphical abstract that summarizes methodology, main results, and clinical implications. We leverage both (A) experimental and (B) computational platforms (C) to analyze the dominant leaflet flutter frequency. Our main results suggest that as valve diameter decreases and leaflet thickness increases the flutter frequency increases independent of flow and operating conditions. Our proposed scaling model rationalizes these results. This study may help us better understand geometrical and mechanical factors to optimize bioprosthetic heart valve (BHV) design.

Taken together with studies on the influence of flutter on the durability of other thin structures,8, 9, 10 our results suggest that the high frequency fluttering in smaller valves may provide a mechanistic explanation for prior clinical observations that aortic valve replacement using small BHVs leads to earlier device failure., Because our computer model suggests that for a fixed device diameter and pressure load, thicker leaflets show higher fluttering frequencies, an intriguing prediction of this study is that an approach to reducing leaflet flutter could be to use thinner biomaterials. Although there are limits in terms of what can be done with fixated tissues, some possibilities include using different fixation pressures to achieve different stiffnesses, or designing different fiber distributions. Additional experimental, computational, and in vivo studies are clearly needed, but taken together with prior results, showing that diastolic leaflet stresses are minimized in larger-diameter devices with thicker leaflets, our results suggest the hypothesis that durability will be maximized by choosing the largest possible BHV diameter along with a leaflet thickness that is optimized for both systolic and diastolic conditions.

To our knowledge, the only prior works to characterize the role of BHV geometry on fluttering dynamics are those of Avelar and colleagues, who used an experimental system to study fluttering in several sizes of bovine and porcine pericardial BHVs, and of Johnson and colleagues, who used a computational model to predict that higher fluttering frequencies occur with thinner leaflets. The study by Avelar and colleagues did not control for variations in device diameter, leaflet thickness or biomechanics, or operating conditions, and it considered only steady flow conditions. In contrast, the experimental platform used in this study provides more physiological pulsatile operating conditions, and the computational platform provides precise control over device properties and operating conditions, enabling more comprehensive assessments of the influences of device geometry on leaflet dynamics. The study by Johnson and colleagues is purely computational and does not include comparisons to either in vitro or in vivo data. That model also describes the leaflet mechanics using shell theory, which may alter model predictions compared with a volumetric leaflet model like that used in this study.

A limitation of this study is that our experimental studies analyze the performance of only Labcor bovine pericardial BHVs, which are not approved by the Food and Drug Administration. However, our computational model uses a generic pericardial BHV leaflet biomechanics model that was not tuned in any way to match the properties of the Labcor BHVs. In addition, experimental data in Appendix 1, I, demonstrates that 25 mm Labcor and Edwards Perimount valves generate similar fluttering frequencies (32.74 ± 3.14 Hz and 29.62 ± 4.7 Hz, respectively). Future studies are needed to evaluate valves from different manufacturers that use alternative construction techniques. This study also does not systematically examine the role of heart rate or rhythm. Another limitation is that we use a rigid aortic root model, which could influence fluttering. In future work, we plan to study leaflet kinematics in cryopreserved aortic root grafts. We also note that prior studies describe fluttering in native aortic valves,, and Chin and colleagues suggest that aortic valve systolic flutter can be used as a screening test for severe aortic stenosis. This study is limited to BHVs and does not address native valves, which possess different material properties than pericardial BHVs. Further, native valve leaflets are living tissue that may not be impacted by fluttering in the same way as chemically fixated tissues.

Conclusions

Ultimately, our goal is to optimize BHV design by understanding the geometrical and mechanical factors that govern BHV leaflet fluttering and its influence on leaflet durability. Although the present study does not reveal the underlying mechanisms that determine leaflet durability, we do demonstrate that a simple scaling model can rationalize our findings that relate fluttering frequency to BHV diameter and leaflet thickness. Understanding the effects of device geometry on leaflet kinematics, and ultimately its effect on leaflet durability, may help improve guidelines for BHV selection. This is potentially highly relevant for both surgical and transcatheter valve replacement. In surgical valve replacement, for instance, our results suggest that there may be a role for aortic root enlargement in improving BHV durability. Several studies have reported on the effects of aortic root enlargement in improving hemodynamics and alleviating PPM by using larger valves., Similarly, this study has potential implications in identifying factors that influence the durability of TAVR. In TAVR valves, it is desirable to use thinner biomaterials to improve device deliverability. Further, it is known that TAVR prostheses can produce improved forward flow hemodynamics with larger effective orifice areas compared with surgical valves. However, they also experience fatigue and failure, especially at smaller diameters., Further work is needed to determine whether the current findings apply to TAVR devices. It also is possible to extend these platforms to study other unique aspects of TAVR devices, including the effect of incompletely expanded TAVR valves that can demonstrate pin-wheeling of the leaflets. This platform can also be used to study the leaflet kinematics of the BHVs in the mitral and tricuspid position. Understanding the precise role of fluttering on device fatigue and failure requires further study, but device designs may ultimately aim to balance BHV diameter and leaflet thickness to optimize device durability.

Conflict of Interest Statement

Dr Vavalle serves as a consultant for Edwards Lifesciences. All other authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.

Table E1

Calibrated parameters for the reduced-order models

	Saline
	Rc (mm Hg/mL/s)	Rp (mm Hg/mL/s)	C (mL/mm Hg)	R1 (mm Hg/mL/s)	R2 (mm Hg/mL/s)	Cvia (mL/mm Hg)
21 mm (DKA015849)	0.0184	1.2382	1.0360	0.3	0.15	0.1
25 mm (DKA015141)	0.0246	1.1983	1.1241	0.3	0.15	0.1
27 mm (DKA015562)	0.0255	1.978	1.1243	0.3	0.15	0.1

Reduced-order model parameters that characterize the pulse-duplicator system components both upstream and downstream of the aortic test section, including the resistance and compliance of the pump, the viscoelastic impedance adapter subsystem, and the left ventricular chamber for both cases, as well as mitral valve and left atrial chamber for blood analog case. The parameters are calibrated using experimental pressure and flow data obtained from the pulse duplicator. These calibrations, which are done for each experimental condition, are performed independently from the three-dimensional fluid-structure interaction model of the aortic valve test section.