Geometric and Mechanical Changes Along the Length of the Porcine Aorta

Abstract

The aorta, the primary artery responsible for carrying blood away from the heart, exhibits variations in its geometric and mechanical properties along its total length. Although these properties have individually been investigated in various animal and human studies, comprehensive assessments measuring all these parameters simultaneously along the entire aortic length have not been well reported. Consequently, current models predicting the aorta’s in vivo mechanical state are based on separately obtained parameters and thus have inherent limitations. The present study systematically quantified variations in ex vivo diameter and wall thickness, in vivo axial stretch, and ex vivo biaxial mechanical properties along the descending thoracic and abdominal segments of five Yorkshire male porcine aortas. Consistent with earlier findings, diameter decreased from ~13 mm to ~3.6 mm and wall thickness decreased from ~2.05 mm to ~1.02 mm toward distal segments, whereas in vivo axial stretch increased from ~1.15 to ~1.5. Notably, the infrarenal abdominal aorta exhibited significantly higher axial stiffness compared to the mid-descending thoracic region (p < 0.05). Estimated in vivo axial stress was found to increase distally (slope = 0.1, p = 0.12), while estimated in vivo circumferential stress remained relatively constant throughout the porcine aorta’s length (slope = 0.004, p = 0.96). These results suggest that during development, position-specific axial stretches, possibly driven by spinal elongation, induce adaptive remodeling of aortic geometry and material properties to maintain relatively uniform circumferential stresses. The resulting dataset provides a comprehensive, location-specific characterization of aortic geometry and mechanics to support future computational modeling studies.

Introduction

The aorta, the primary artery that carries blood away from the heart, plays a crucial role in maintaining systemic circulation. Along its length, the aorta exhibits marked regional variations in both geometry and mechanical properties (Concannon et al., 2020; Dwivedi et al., 2023; Sokolis, 2023, 2007; Tarraf et al., 2023). These regional differences are particularly important in the context of vascular diseases such as the formation of both aneurysms and atherosclerosis (Chowienczyk, 2015), which often alter the mechanical behaviors of this vessel and can compromise its function.

A well-documented phenomenon in arterial biomechanics is the discrepancy between the aorta’s length measured in vivo and ex vivo, first recognized over a century ago (Fuchs, 1902). Upon excision, arteries undergo significant shortening (contraction), resulting in a shorter ex vivo length compared to their in vivo configuration. The ratio of these lengths in turn, will define the in vivo axial stretch. Early work demonstrated that longitudinal strain substantially affects the biomechanical behavior of the thoracic aorta, influencing both pressure-volume relationships and overall vessel compliance (Bergel, 1961). Furthermore, studies have shown that in vivo axial stretch increases progressively along the aortic length (Guo et al., 2002; Han and Fung, 1995). Similarly, geometrical parameters such as ex vivo wall thickness and diameter decrease distally, trends that have been consistently reported across multiple species (Fung and Liu, 1989; Huang et al., 1996; Kassab, 2006; Moriwaki et al., 2011; Purslow, 1983; Rouleau et al., 2012; Sokolis and Papadodima, 2022), including pigs (Dinardo et al., 2014; Han and Fung, 1995; Ríos-Ruiz et al., 2022; Sokolis, 2007; Zunder, 2021).

Beyond geometric differences, mechanical properties of the aortic wall are also known to show regional variations. Initial studies investigating the mechanical variation primarily employed uniaxial tensile testing and canine models (Cohen et al., 1972; Tanaka and Fung, 1974). More recently, advances in planar biaxial mechanical testing have enabled a more comprehensive characterization of aortic mechanics, particularly of porcine tissue. For example, Zeinali-Davarani et al., 2015 used biaxial testing combined with multiphoton microscopy to link regional mechanical behavior to collagen microstructure along the descending thoracic aorta. They found that ex vivo stiffness increased distally, associated with earlier and more pronounced collagen fiber recruitment. Peña et al., 2015 performed uniaxial and biaxial tests on porcine upper and lower thoracic aorta to study the mechanical behavior of aortic samples. They found that the distal samples were stiffer than the proximal samples ex vivo. In a separate study, Peña et al., 2018 performed biaxial tests comparing the ascending thoracic, descending thoracic, as well as the infrarenal abdominal aorta segments from the same porcine subjects. Using established strain energy functions to quantify behaviors, they reported that the infrarenal aorta was the stiffest and most anisotropic region ex vivo. Their statistical analysis revealed significant differences in circumferential mechanical behaviors between regions, while axial properties appeared relatively position independent.

Although several groups have independently assessed in vivo axial stretch, geometry, or biaxial mechanical behavior, few have simultaneously evaluated all these parameters along the full length of the aorta (Guo et al., 2002). A recent review by Badal and Barocas, 2025 has further organized and contextualized the above-mentioned findings, highlighting the need for integrated studies that measure key mechanical and geometrical properties in the same anatomical specimen. Thus, the objective of the present study was to measure ex vivo no load diameter, ex vivo wall thickness, in vivo axial stretch, and biaxial mechanical properties along the porcine aorta in the same animal. We hypothesized that regional differences in geometry and material properties reflect developmental adaptations to positional variations in in vivo axial stretch, potentially driven by factors such as spine growth. Furthermore, we propose that these regional differences contribute to the maintenance of relatively uniform in vivo circumferential stresses along the length of the aorta from these healthy preclinical models.

Methods

Five porcine aortas were harvested from healthy Yorkshire male animals euthanized as part of unrelated studies. The animals ranged in age from 6–8 months and weighed between 70–90 Kg. Because some procedures involved recovery and reanimation of the heart and ascending thoracic aorta, only the descending thoracic and abdominal regions were available for analysis (Fig. 1a–b). Prior to excision, the usable aortic specimens were divided into 9–10 segments along their lengths and marked using pins. The in situ lengths of each segment were recorded. The time interval between death and mechanical testing was kept below 2 hours to minimize degradation and to preserve tissue viability. During this period, the specimens were maintained in a saline bath at room temperature to prevent dehydration and to maintain physiological conditions.

Figure 1.

(a) In situ view of the descending thoracic and abdominal regions of the porcine aorta prior to excision. (b) Excised aorta cleaned and prepared for segmentation and subsequent measurements. (c) Segmented aorta, illustrating clearly marked segments used to assess regional variations along the vessel length. (d) Cross-sectional view of an aortic segment demonstrating the measurements of initial luminal diameter (D₀) and wall thickness (t₀). (e) Square sample from an aortic segment mounted on a biaxial testing apparatus to measure mechanical properties under controlled loading conditions.

After excision and removal of the surrounding tissue, vessel geometry was assessed. Digital calipers were used to measure the ex vivo wall thickness and vessel diameter for each segment. Given the elliptical nature of the aortic cross-section, both major and minor diameters were measured. Each obtained segment length was remeasured ex vivo, and the ratio of the in situ to ex vivo lengths provided the in vivo axial stretch (λ_z).

Subsequently, square samples were excised from the anterior region of each segment and stored in 1X PBS solution at room temperature prior to biaxial mechanical testing. All surrounding tissue was removed prior to testing. Each sample was mounted onto a displacement-controlled biaxial testing apparatus (TA Instruments, Eden Prairie, MN) using fishhooks. A small preload (~0.025 N) was applied to remove slack and define the reference configuration. Displacement was independently controlled along four arms to generate equibiaxial loading, with both axes displaced at a constant rate of 0.2 mm/s. Specimens were stretched to a maximum stretch of approximately 1.55. Forces were recorded using calibrated load cells aligned along two orthogonal directions. Four preconditioning cycles were performed to the maximum value of 1.55 stretch. Testing was conducted in air, while samples were maintained in a saline bath at room temperature between excision and testing to preserve hydration. Forces were recorded using 200N load cells with a resolution of 0.01 N, and actuator displacements were prescribed and measured with a resolution of 0.001 mm. Deformation was quantified based on actuator displacement, and stretch was calculated as the ratio of the current specimen length (initial length plus displacement) to the initial length. The first Piola–Kirchhoff (PK1) stress was computed as the ratio of the measured force to the initial (undeformed) cross-sectional area. All tests were performed under passive conditions without pharmacological stimulation.

Many constitutive models have been proposed to reproduce the mechanical behavior of soft tissues (Alastrué et al., 2010; Gasser et al., 2005; Hill et al., 2012; Holzapfel et al., 2000). Previous work (Peña et al., 2018) found that the Holzapfel-Gasser-Ogden (HGO) (Holzapfel et al., 2000) model was not able to capture the passive biaxial properties of each specific porcine aorta over a wide range of biaxial deformations. Thus, the Gasser-Ogden-Holzapfel (GOH) (Gasser et al., 2005) model, which is a generalized structure extension of HGO was fitted to the biaxial data for each sample. The strain energy function (ψ_r) for the GOH model is

$${\psi }_r=\frac{C_0}{2}(I_1-3)+\left(\frac{k_1}{2k_2}\right)\sum _{\alpha =1}^2(\text{exp}(k_2{〈E_{\alpha }〉}^2)-1)$$ (1)

$$E_{\alpha }=\kappa (I_1-3)+(1-3\kappa )(I_{4\alpha }-1)$$ (2)

where, I₁ = trC, and I_4α = a_αr•C•a_αr and C is the right Cauchy-Green tensor. C₀ is the ground matrix shear modulus; k₁ is the fiber modulus and k₂ is the exponential coefficient. The formulation assumes two fiber families along the vectors a_αr (α = 1,2) oriented at angled ± γ in the local e₁—e₂ plane, each with a dispersion κ, where $0\le \kappa \le \frac{1}{3}$. When κ = 0, the model is same as the HGO model, showing all fibers are in the direction given by a_αr, while when $\kappa =\frac{1}{3}$, the stress response is isotropic and similar to that used in (Demiray, 1972). The GOH model parameters were fit in MATLAB using the fmincon optimization routine by minimizing the summed squared error between model-predicted and experimentally measured biaxial stress responses over the tested stretch range.

Regional stiffness was characterized by circumferential and axial tangent moduli, obtained as the slope of the GOH-predicted Cauchy stress–stretch curve at prescribed equibiaxial stretches of 1.05, 1.15, and 1.25. To estimate the in vivo stress state for each segment, we employed the thin-walled cylinder approximation using the geometric data, the in vivo axial stretch, and the fitted GOH model parameters. This approximation was used to obtain consistent estimates of the operating wall stress for regional comparison along the vessel length, rather than to resolve transmural stress distributions. The thin-walled approximation is given by:

$${\sigma }_{\theta \theta }=\frac{R\text{P}}{t}=\frac{{\lambda }_{\theta }{R}_{0}P}{{\lambda }_r{t}_0}=\frac{{\lambda }_z{\lambda }_{\theta }^2{R}_{0}P}{t_0}$$ (3)

where, t and R are in vivo thickness and radius, t₀ and R₀ ex vivo (unloaded) values, P is the mean arterial pressure (105mmHg (Reed et al., 2018)), 𝜎_𝜃𝜃 is in vivo circumferential Cauchy stress, and 𝜆_𝜃, 𝜆_𝑧, 𝜆_𝑟 are the circumferential, axial and radial stretches. With λ_z measured experimentally, and incompressibility imposing ${\lambda }_r=\frac{1}{{\lambda }_{\theta }{\lambda }_z}$, the circumferential stress σ_θθ(λ_θ,λ_z, λ_r) can be calculated for any value of λ_θ using the GOH model (Parikh et al., 2024) until the pressure equation is satisfied. After solving for λ_θ from the pressure relation, the corresponding circumferential and axial Cauchy stresses were obtained from the fitted GOH model at that operating point. In particular, the in vivo axial stress was obtained from the axial component of the GOH-predicted Cauchy stress tensor:

(4)

where

$$\frac{\partial E_1}{\partial C_{22}}=\frac{\partial E_2}{\partial C_{22}}=\kappa +(1-3\kappa ){\text{sin}}^2$$ (5)

Statistical analyses were conducted using MATLAB (MathWorks). Linear regression analyses were performed to evaluate relationships between fractional longitudinal position (FLP), which served as the independent variable, and several dependent variables including internal diameter, wall thickness, in vivo axial stretch, and estimated in vivo stresses. Regression parameters were estimated using MATLAB’s built-in functions, with model fits assessed by calculating coefficients of determination and associated p-values. Potential outliers were identified using Cook’s distance (Cook, 1977), with data points exceeding a threshold of four divided by the number of observations considered outliers and excluded from subsequent regression analyses. To compare regional mechanical properties among anatomical groups, linear mixed-effects models were used with anatomical group treated as a fixed effect and pig treated as a random effect, thereby accounting for repeated segment-level measurements obtained from the same aorta. Normality was assessed on the residuals of the mixed-effects models using Lilliefors and Jarque-Bera tests. When the omnibus mixed-effects model indicated a significant group effect (p < 0.05), pairwise group comparisons were performed using contrasts on the fitted model coefficients.

Results

To assess regional difference of the aortic properties, we used fractional longitudinal position (FLP = x/L), where x is the distance along the aorta from the left subclavian artery and L is the length between the left subclavian artery and the iliac bifurcation. Thus, FLP ranged from 0 at the left subclavian artery to 1 at the bifurcation. Because it was not possible to test every sample at identical FLP values across all animals, segments were pooled into four anatomical groups: FLP 0.20–0.40 (mid-descending thoracic aorta), 0.40–0.60 (lower-descending thoracic aorta), 0.60–0.85 (suprarenal abdominal aorta), and 0.85–1.00 (infrarenal abdominal aorta) (Fig. 1).

Both averaged diameter and wall thickness decreased along the length of the aorta in all five animals studied (Fig. 2a–b). Internal diameter declined from ~ 13 mm in the mid-descending region to ~ 3.6 mm near the iliac bifurcation, and wall thickness fell from ~ 2.05 mm to ~ 1.02 mm (linear regression: p < 0.001 for each animal). In contrast, the in vivo axial stretch ratio increased distally from ~1.15 to ~1.5, consistent with trends reported in other mammalian models (Han and Fung, 1995; Kassab, 2006). Re-plotting the geometric data pairwise (Fig. 2d–f) revealed that larger diameters correlated with thicker walls (linear regression: p < 0.001 for each animal) and that both diameter and thickness were inversely associated with λ_z (diameter vs. λ_z, linear regression: p<0.02 each animal and thickness vs. λ_z, linear regression: p < 0.01 each animal). The ratio of the major diameter to minor diameter showed a minor increase along with the FLP (linear regression: slope = 0.17, p=0.04) (Fig. 2g).

Figure 2.

Variations in geometric parameters and axial stretch along the FLP. (a) Diameter and (b) wall thickness progressively decrease along the aorta with increasing FLP. (c) Conversely, axial stretch increases along the vessel length. (d) Wall thickness positively correlates with luminal diameter. Both (e) diameter and (f) wall thickness inversely correlate with axial stretch. (g) The ratio of major to minor diameters increases with FLP, indicating greater ellipticity in distal segments (slope = 0.17, p = 0.04). Note: Two data points identified as outliers by Cook’s method (Cook, 1977) were excluded from analysis in panel (g).

Fig. 3a–d shows representative PK1 biaxial stress–stretch curves for two different animal in the axial and circumferential directions. As expected, the curves were highly nonlinear, but no clear positional trend was observed in the raw data. Although the GOH model parameters were fit independently for each sample, no consistent relationship between the individual fitted parameters and fractional longitudinal position was observed across animals. Because these parameters are coupled and primarily serve to reproduce the overall constitutive response, regional comparisons were based instead on stress–stretch behavior, tangent moduli, and estimated in vivo stresses, which provide a more direct functional assessment of mechanical variation along the aorta. For readers interested in the GOH fit values, all fitting results are provided in the Supplementary Material.

Figure 3.

(a-b) Axial PK1 stress and (c-d) circumferential PK1 stress plotted against corresponding axial and circumferential stretches for two different Pigs, respectively. Curves illustrate the nonlinear mechanical response of the aortic tissue under equibiaxial loading conditions.

To highlight regional differences, stresses at λ = 1.05, 1.15, and 1.25 were averaged within each FLP group. Both circumferential and axial mean stresses rose from proximal to distal segments (except the suprarenal), with the infrarenal region showing the highest values (Fig. 4a–b). Both circumferential and axial tangent moduli at λ = 1.05, 1.15, and 1.25 also generally increased from proximal to distal segments, with the infrarenal abdominal aorta exhibiting the highest values and the suprarenal abdominal aorta showing a lower stiffness relative to the infrarenal region (Fig. 4c–d). Linear mixed-effects analysis, with anatomical group treated as a fixed effect and pig as a random effect, showed significant regional effects for axial stress at all three stretches (λ = 1.05: p = 0.0449, λ = 1.15: p = 0.0099, λ = 1.25: p = 0.0081), whereas circumferential stress did not show significant regional effects at the omnibus model level (λ = 1.05: p = 0.31, λ = 1.15: p = 0.137, λ = 1.25: p = 0.112). Similarly, axial tangent modulus showed significant regional effects at all three stretches (λ = 1.05: p = 0.0102, λ = 1.15: p = 0.0073, λ = 1.25: p = 0.0093), whereas circumferential tangent modulus did not reach significance at the omnibus model level (λ = 1.05: p = 0.0710, λ = 1.15: p = 0.0877, λ = 1.25: p = 0.1077).

Figure 4.

(a) Axial and (b) circumferential first PK1 stress measured at specified equibiaxial stretch values across different aortic locations. Corresponding values of the (c) axial and (d) circumferential tangent modulus (TM) are shown, highlighting differences in local stiffness along the aortic length. The error bars represent standard deviation.

The estimated in vivo axial stress is shown in Fig. 5a, and circumferential stress is shown in Fig. 5b. The estimated axial stress increases with FLP, however the increase is statistically insignificant (linear regression: slope = 0.1, p=0.12). Higher-order polynomial regressions were also examined for the relationship between FLP and estimated in vivo axial stress. However, neither quadratic nor cubic terms significantly improved the fit relative to the linear model (nested-model F-tests: p = 0.476 and p = 0.212, respectively), and the linear model was therefore retained. On the other hand, estimations of the in vivo circumferential Cauchy stress do not exhibit consistent positional variability across the pigs. A linear fit across all data shows no evidence of significant association (linear regression: slope = 0.004, p=0.96), suggesting that, on average, circumferential stress remains relatively constant along the length of the studied aortic samples.

Figure 5.

Estimated (a) axial and (b) circumferential stresses plotted against FLP regional trends. Outliers identified using Cook’s method (Cook, 1977) were excluded from the analyses to ensure robustness of the presented trends.

Discussion

The most significant outcome of this preclinical research study was the generation of a unique dataset of geometries, mechanical behaviors, and in vivo axial stretches at different locations along the length of the aorta for swine. This dataset as a whole provided in the Supplemental Data. Estimating accurate in vivo wall stresses along the length of the descending aorta necessitates matched geometric and mechanical data. (Han and Fung, 1995) previously calculated axial stresses using their geometric parameters for canine aorta, but they used uniaxial data from a different study (Tanaka and Fung, 1974) potentially introducing significant inaccuracies. Additionally, their estimations ignored vessel pressurizations and circumferential loading. Yet, the Han and Fung study reported that the estimated axial stresses increased along with the length of aorta, and the axial tensions remained nearly constant. Dai et al., 2021 applied thin-walled cylindrical assumptions to porcine descending aortas at multiple pressures, but overlooked the actual axial stretches by assuming axial stress equaled half of the circumferential stress and treated radius–thickness ratios as constant across pressures. Such assumptions are likely inappropriate for compliant soft tissues. Similarly, Badal and Barocas, 2025 combined data from multiple studies to estimate in vivo stresses concluding that net wall stress increases distally. In contrast with those studies, our preclinical approach here combined geometrical and mechanical data from the same specimens and explicitly incorporated in vivo axial stretch, ensuring more accurate stress estimations.

Various aspects of the present work can be compared to previous reports. Multiple studies have consistently demonstrated a progressive decrease in luminal diameter and wall thickness from the thoracic to the abdominal aorta (Fung and Liu, 1989; Huang et al., 1996; Kassab, 2006; Moriwaki et al., 2011; Ríos-Ruiz et al., 2022; Sokolis, 2007; Sokolis and Papadodima, 2022). Our findings confirm this established trend, showing a monotonic reduction in both diameter and thickness as FLP increases (Fig. 2a–b). This tapering aligns with the uniform wall shear stress hypothesis, under which endothelial-mediated growth and remodeling occurs in response to reduced distal blood flow and associated wall shear stress.

Previous studies that had employed uniaxial (Krüger et al., 2016; Sokolis, 2007; Tanaka and Fung, 1974) and biaxial (Rouleau et al., 2012; Shahbad et al., 2025) testing have generally reported increased aortic stiffness with distal distance from the heart, except for Guo and Kassab, 2003, who identified the diaphragm-adjacent region of the descending thoracic aorta as the stiffest in a mouse model using uniaxial testing. Zeinali-Davarani et al., 2015 and Peña et al., 2018 performed biaxial tests on porcine aortas, though at relatively broad intervals (proximal, mid-proximal, mid-distal, and distal thoracic regions in Zeinali-Davarani et al., 2015, and ascending thoracic, descending thoracic, and infrarenal abdominal regions in Peña et al., 2018), both groups concluding similarly that the distal regions were stiffer. Our raw stress–stretch data did not elicit an explicit monotonic relationship with FLP (Fig. 3); however, regional grouping revealed that the infrarenal abdominal aorta was significantly stiffer axially compared to the mid-descending thoracic aorta (p < 0.05), although circumferential stiffness differences were not significant. The relatively lower stiffness observed in the suprarenal abdominal aorta may result from altered loading conditions at the renal bifurcation, where the branching renal arteries modify both axial and hemodynamic stresses. Given the absence of specific suprarenal samples in earlier studies and their use of broader regional pooling, their conclusions about distal stiffening effectively align with our findings.

The mechanical properties of arterial tissue are strongly influenced by extracellular matrix composition, with increased collagen content generally associated with higher stiffness, as collagen is the stiffest structural constituent of the arterial wall (Xu and Shi, 2014). Sokolis, 2007 quantified extracellular matrix composition along the length of the porcine aorta and reported a progressive increase in collagen content with distance from the heart, a trend that is broadly consistent with the elevated axial stiffness observed in distal segments in the present study. However, collagen content alone does not fully explain the distinct mechanical behavior of the suprarenal abdominal aorta. Although collagen abundance may be elevated in this region, local differences in collagen organization, fiber recruitment, dispersion, or interactions with elastin and smooth muscle may play a dominant role in governing its mechanical response. Because the present study did not include direct compositional or microstructural measurements, these interpretations remain speculative, and future studies combining biaxial testing with histological or microstructural analyses will be required to elucidate the mechanisms underlying the suprarenal region’s distinct behavior.

Our calculated in vivo stresses show a steady increase in axial stress with FLP, whereas circumferential stress remains constant (Fig. 4). From the perspective of tensional homeostasis (Flynn et al., 2013; Stamenović and Smith, 2020), elevated axial stress in the IAA necessitates additional collagen fibers to maintain target fiber stress. Histological studies confirm that collagen content increases in distal aortic segments (Liyanage et al., 2022; Sokolis, 2007), and given collagen’s nonlinear stiffness (Weisbecker et al., 2012), its higher level under higher axial stress provides an explanation for the observed increase in axial tangent modulus toward the iliac bifurcation.

The uniform circumferential stress hypothesis (Kassab, 2006) suggests that arteries remodel to maintain consistent circumferential wall stress, thus achieving tensional homeostasis (Flynn et al., 2013; Stamenović and Smith, 2020). Our data support this hypothesis, showing nearly constant circumferential stress along the length of the swine descending aortic. Additionally, the observed distal thinning can be partially attributed to the maintenance of uniform circumferential stress. According to Laplace’s law, as the diameter decreases in response to regulated flow and endothelial signaling, wall thickness must also diminish proportionally to sustain stable circumferential stress, maintaining approximately 230 kPa throughout the aorta.

Our findings suggest a possible mechanism: during early development, rapid growth—particularly elongation of the spine (Chatelain et al., 2024)—imposes region-specific in vivo axial stretch on the aorta. In response, the aorta grows and remodels accordingly, adjusting its geometric and mechanical parameters to maintain near-uniform circumferential stress levels. This adaptive development enables the aorta to accommodate changes in the mechanical environment throughout growth. Recent research has further indicated that in vivo axial stretch decreases with age in human abdominal aortas (Horny et al., 2014, 2011). Thus, it can be speculated that while axial length is externally imposed, in vivo axial stretch itself is modulated by tissue remodeling due to aging. The greater than 1 in vivo axial stretch suggests that the remodeling does not “keep up” with the imposed stretch, or that coupling between the directions and the pressure-imposed non-zero circumferential stress and the axial stress requires a greater than unity axial stretch or both.

Beyond confirming established results, our comprehensive dataset provides a crucial baseline for future modeling studies. Over recent decades, various continuum (Rodriguez et al., 1994), discrete (Humphrey and Rajagopal, 2002), combined (Cyron et al., 2016; Gacek et al., 2022), and bio-chemo-mechanical (Flanary and Barocas, 2023; Irons and Humphrey, 2020) models have investigated arterial growth and remodeling. Typically, these models assume a baseline healthy geometry derived from prior experimental or theoretical studies, subsequently applying perturbations to simulate disease progression. Yet, such studies rarely address how the healthy baseline parameters are established during development or account for regional variations along the aortic length. We suggest here that our data bridges this critical gap by providing detailed, experimentally verified benchmarks for establishing accurate healthy-state preclinical models.

Previous modeling studies by Alford et al., 2008 and Alford and Taber, 2008 successfully predicted variations in opening angles and in vivo axial stretch, comparing their computational results to experimental data from Fung and Liu, 1989 in rats and Guo et al., 2002 in C57bl/6 mice. Nevertheless, those studies did not include biaxial mechanical properties, and their reliance on mixed datasets from different studies and species introduces inherent inconsistencies. The comprehensive experimental dataset presented here addresses these limitations, providing robust and consistent benchmarks for validating future computational predictions of aortic mechanical and geometric properties across regions. Opening angle was not measured here, and future studies incorporating residual strain measurements together with biaxial mechanics would further strengthen region-specific constitutive and growth-and-remodeling models.

While the present study was conducted using porcine aortas, and direct quantitative mapping to human aortic properties is not appropriate, several of the regional trends observed here are consistent with prior human studies. In particular, healthy human aortas exhibit a progressive decrease in diameter and wall thickness along their length, along with regional variations in mechanical stiffness and in vivo axial stretch (Chen et al., 2024; Sokolis et al., 2020, 2017). Similarly, the increase in stiffness toward distal segments observed in this study aligns with reported trends in human aortic tissue (Kamenskiy et al., 2014). These qualitative consistencies suggest that, despite interspecies differences, the underlying biomechanical principles governing regional adaptation may be conserved. As such, the dataset presented here provides a valuable and internally consistent benchmark for developing, calibrating, and validating computational models of arterial mechanics. Importantly, such models can incorporate species-specific parameters where needed, while still leveraging the spatial relationships and coupled measurements reported in this study.

The present study employs planar biaxial testing, a widely used and well-validated approach for characterizing the intrinsic material behavior of arterial tissue under controlled multiaxial loading (Peña et al., 2015; Zunder, 2021). As with any ex vivo methodology, planar biaxial testing involves simplifying assumptions. Because the aorta is a cylindrical structure, flattening the tissue alters vessel curvature and releases residual stresses present in the intact wall. In addition, planar testing does not preserve through-thickness stress gradients or axial–circumferential coupling associated with pressurization of intact vessels. Inflation testing of arterial segments provides a complementary, physiologically representative loading configuration by maintaining native geometry and stress states. Importantly, inflation-based measurements reported by Dwivedi et al., 2023 similarly demonstrate an increase in aortic stiffness with distance from the heart, consistent with the regional trends observed here, indicating that the distal stiffening reported in this study is not an artifact of the testing modality.

A further limitation of this study is that only equibiaxial loading was applied. Previous studies have shown that applying multiple loading ratios (e.g., 1:1, 1:2, 2:1) provides a more complete characterization of soft tissue anisotropy and fiber recruitment (Ahuja et al., 2018; Wang et al., 2016). In addition, estimation of in vivo stresses relied on a thin-walled cylinder approximation, which neglects wall heterogeneity and transmural stress variations. These assumptions may affect the absolute stress estimates but are unlikely to alter the relative regional trends reported here.