The research, led by Tej Jolly, Arnav Garcha, and Noelia Grande Gutiérrez of Carnegie Mellon University, tackled a stubborn methodological problem. Patient studies have long hinted that the health of the coronary microvasculature, the dense network of arterioles and capillaries downstream of the major arteries, influences both FFR and CFR. But in real patients, microvascular status is tangled up with anatomical variation: every artery has a different lesion length, severity, and branching pattern, making it nearly impossible to isolate what the microcirculation alone is doing. Computational studies have tried to fill the gap, yet most either fix the downstream resistance at a single value or sample only a narrow range of microvascular states, and patient-specific models confound geometry with physiology.

To break that confounding, the team built a multiscale modeling framework that couples three-dimensional computational fluid dynamics of the epicardial coronary arteries with a lumped-parameter network representing the microvascular bed. The three-dimensional component solves the incompressible Navier–Stokes equations over unstructured tetrahedral meshes averaging 3.3 million elements, with boundary-layer refinement near the walls and local refinement around stenoses. Blood was treated as a Newtonian fluid, a reasonable approximation given that simulated shear rates exceeded 100 per second, and flow was assumed laminar, with Reynolds numbers averaging around 200 and peaking at 967 under the most severe stenosis and highest flow. Transient simulations ran for four cardiac cycles at a heart rate of 70 beats per minute, with periodic convergence defined as less than one percent change in outlet pressure and flow between cycles.

The anatomical foundation was equally deliberate. The researchers generated a baseline left coronary artery model using population-mean values for branch diameters, bifurcation angles, and segment lengths drawn from a computed tomography angiography study of 300 adults with zero coronary calcium scores. Onto this fixed skeleton they imposed nine plaque configurations, varying lesion location between the left anterior descending artery and the left circumflex, lesion length between a focal 1.2 centimeters and a diffuse 2.5 centimeters, and stenosis severity across zero, 45, and 60 percent diameter reduction. The 45 percent severity was chosen after exploratory sweeps showed that FFR most often crossed its clinical threshold of 0.8 between 40 and 50 percent narrowing, precisely the borderline zone where diagnostic uncertainty is greatest. Stenoses were modeled as asymmetric constrictions to mimic the eccentric plaque morphology commonly seen in atherosclerosis.

Each of the nine geometries was then run under four hyperemic microvascular resistance states, produced by scaling the microvascular resistor in the lumped network to represent total coronary resistance equal to 24, 43, 62, and 81 percent of its resting value. The baseline of 24 percent reflects the maximum resistance reduction achievable with intravenous adenosine, the hyperemic agent used in clinical FFR measurement, while the elevated states span progressively impaired vasodilation. The resulting hyperemic microvascular resistance values ranged from 1.3 to 8.1 millimeters of mercury per centimeter per second, bracketing the clinical dysfunction threshold of 2.5 and matching pathologically observed ranges. Nine additional resting simulations brought the total to 45, and the simulated cases were verified against a retrospective observational cohort of 299 patients, occupying the same joint distributions of stenotic burden, microvascular resistance, and index values as the clinical data.

The headline finding is strikingly clean: for geometrically identical lesions, raising microvascular resistance increased FFR and decreased CFR. Across all 36 hyperemic simulations, the correlation between microvascular resistance and FFR was strong and positive, while the correlation with CFR was very strong and negative. The mechanism follows directly from the physics. Higher downstream resistance throttles hyperemic flow, and since the pressure drop across a stenosis scales with flow, a throttled flow shrinks the trans-stenotic pressure gradient, making FFR appear more reassuring even as the tissue receives less blood. Meanwhile, the ratio of hyperemic to resting distal velocity collapses toward unity, dragging CFR downward. In short, coronary flow in these models was governed primarily by the microcirculation rather than by the epicardial blockage itself, with flow bands overlapping broadly across lesions of different severities when stratified by stenosis.

The threshold analysis carries the most clinical weight. Because treatment decisions hinge on binary cutoffs—FFR of 0.8 and CFR of 2—the team fitted regression models within narrow bands around those values. Near the FFR threshold, the standardized coefficients for microvascular resistance and hyperemic stenosis resistance were comparable in magnitude and both highly significant, with models explaining over 90 percent of the variance. This means that near-threshold FFR readings reflect both the epicardial lesion and the microvascular state, so a shift across the cutoff cannot be uniquely attributed to plaque severity. Near the CFR threshold, by contrast, microvascular resistance dominated, consistent with CFR’s role as the gold standard for microvascular function. The response was also nonlinear: the transition from healthy to the first impaired microvascular state produced far larger changes in flow and FFR than transitions deeper into dysfunction, with sensitivity to microvascular resistance dropping by roughly 77 percent for FFR and 71 percent for CFR as both indices approached a plateau once hyperemic flow was already severely limited.

Perhaps most provocatively, the simulations reproduced both modes of FFR–CFR discordance in identical anatomies purely by varying microvascular resistance. The same plaque could land in the preserved-FFR, reduced-CFR quadrant or the reduced-FFR, preserved-CFR quadrant depending on the microvascular state alone, and preserved FFR with reduced CFR was consistently associated with higher microvascular resistance. Notably, lesion length showed no association with the direction of discordance in a Fisher’s exact test, suggesting that the focal-versus-diffuse distinction, often invoked to explain index disagreements, may be an inadequate surrogate for physiological state. The study also found that elevated microvascular resistance expanded the arterial wall area exposed to low wall shear stress at the LAD–LCx bifurcation, an atherogenic condition, and shrank areas of high shear stress linked to cap thinning and plaque rupture, with the bifurcation effects persisting even after controlling for bulk flow rate.

The implications reach from the catheterization lab to the computational pipeline. Clinically, lesions with FFR at or below 0.50 remain significant regardless of microvascular status, but when microvascular resistance is elevated, an FFR above 0.8 can coexist with a pathologically low CFR, and severe microvascular disease can push CFR below threshold even with no stenosis at all and an FFR near 1. In such cases, flow-oriented evidence from perfusion imaging or CFR itself should carry more weight. Conversely, when microvascular resistance is low and flow runs high, pressure-based indices can be penalized, and flow-normalized metrics such as hyperemic stenosis resistance become especially informative. For the growing field of CT-derived FFR, the message is that distal boundary conditions are not mere numerical settings—they encode microvascular health, and uncertainty in them propagates directly into predicted indices. The authors argue that hyperemic microvascular resistance should be reported routinely alongside FFR and CFR, both in the clinic and in computational studies, so that geometric effects are not mistaken for physiological ones and so that cross-study comparisons become reproducible. Future work will extend the framework to patient-specific coronary trees, compliant walls, branch-specific and time-varying microvascular parameterizations, and prospective validation against simultaneous invasive pressure-flow measurements and perfusion imaging.

Subject of Research: The effect of hyperemic microvascular resistance on coronary hemodynamic diagnostic indices such as FFR and CFR, investigated through multiscale computational fluid dynamics simulation.

Article Title: Impact of Microvascular Resistance on Coronary Hemodynamic Diagnostic Indices: A Multiscale Simulation Study

Article References: Impact of Microvascular Resistance on Coronary Hemodynamic Diagnostic Indices: A Multiscale Simulation Study. (n.d.). https://doi.org/10.1007/s10439-026-04378-1

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04378-1

Keywords: coronary artery disease, fractional flow reserve, coronary flow reserve, microvascular resistance, computational fluid dynamics, coronary hemodynamics, lumped-parameter network, stenosis, wall shear stress, revascularization, CT-FFR, cardiovascular modeling

The research tackled a long-standing puzzle in vascular biology: why women fare disproportionately badly as they age. Although both sexes experience arterial stiffening and neurological decline with age, women show a higher incidence of cardiovascular disease and many neurological conditions after menopause compared with age-matched men. The link between arterial stiffness and mortality is nearly twice as strong in women as in men, and cognitive decline progresses more steeply in elderly females than in elderly males. Because aging-related vascular stiffening can alter the blood–brain barrier and brain function, the investigators reasoned that the earliest changes in the vessels of the neck—the carotid arteries and jugular veins—might be tightly coupled to biochemical remodeling inside the brain.

To capture that coupling, the team combined three complementary technologies in a single cohort of healthy female C57BL/6 mice. Ultrasound imaging, performed with a high-frequency transducer on a Vevo 3100 system, allowed them to measure vessel diameters and blood velocities in living animals, from which they calculated wall shear stress, circumferential cyclic strain, pulsatility index, and volumetric flow in the carotid artery. Histology then quantified the balance of elastin and collagen—the two structural proteins that determine whether an artery snaps back like a rubber band or behaves like a rigid pipe. Finally, two mass spectrometry approaches, liquid chromatography mass spectrometry and matrix-assisted laser desorption ionization mass spectrometry imaging, charted the lipid landscape of the carotid artery, jugular vein, and brain, mapping exactly where age-sensitive lipids reside in tissue.

The ultrasound results were unambiguous. Middle-aged females, at 52 weeks of age, showed a roughly 23.5 percent drop in average carotid velocity during systole compared with 12-week-old young females, and a 27.8 percent reduction in wall shear stress during systole, a difference that was highly statistically significant. Average volumetric flow, normalized to body weight, plummeted by about 55.2 percent with age. Because the reduction in shear stress was driven primarily by the decline in systolic velocity, the authors attribute it to age-related changes in vascular compliance and cardiac output. Circumferential cyclic strain and pulsatility index trended downward but did not reach statistical significance, while the jugular vein diameter grew about 11.6 percent larger in the older animals—evidence that even veins remodel with age.

Under the microscope, the structural counterpart of those functional losses came into focus. The elastin-to-collagen ratio in the carotid artery fell in middle-aged mice relative to young controls, a shift that reflects the fundamental biomechanics of aging arteries. Elastin fibers, which allow vessels to stretch and recoil, replenish at an extremely low rate throughout life; as they dwindle, stiffer collagen fibers accumulate in relative terms. A lower elastin-to-collagen ratio means diminished arterial compliance, limiting the vessel’s ability to dampen the pulsatile pressure generated by each heartbeat. When large arteries lose that damping capacity, the delicate microvasculature surrounding the brain is exposed to higher pulsatile energies than it can safely withstand, a mechanism that previous work has linked to damage of small vessels and the blood–brain barrier. Notably, the jugular vein showed the opposite trend, with a modest increase in its elastin-to-collagen ratio, suggesting that arteries and veins age along distinct structural trajectories.

The lipidomic data added a molecular dimension to this picture. In the brain, ten lipid headgroups—including hexosylceramides, phosphatidic acid, phosphatidylcholine, phosphatidylethanolamine, phosphatidylglycerol, phosphatidylinositol, phosphatidylserine, and sulfated hexosylceramides—decreased significantly with age, while N-acylethanolamines increased, possibly as a compensatory response to age-related stress or inflammation given their known neuroprotective roles. These reductions matter because the brain is among the most lipid-rich organs in the body. Phospholipids protect cells from damage, sphingolipids stabilize the myelin that insulates nerves, and low levels of these molecules can signal neuronal vulnerability. In the carotid artery, cardiolipins and lysophosphatidylethanolamines rose with age, changes the authors link to mitochondrial dysfunction and membrane remodeling within the vessel wall. A triglyceride species, TG 45:9, accumulated in the aged carotid—consistent with epidemiological evidence that triglycerides are the blood lipid most strongly associated with arterial stiffness—while the signaling lipid DG 55:10 declined, hinting at impaired endothelial responsiveness.

Perhaps the study’s most consequential technical insight came from comparing whole-brain and region-specific analyses. When the researchers ran an unsupervised principal component analysis on mass spectrometry imaging data from the entire brain, young and middle-aged animals overlapped almost completely, with only a faint suggestion of age-related variation. But when the same analysis was restricted to the hippocampus—the memory center known to be exquisitely sensitive to metabolic and mitochondrial stress—the two age groups separated cleanly, with minimal overlap. The dominant contributor to that separation was a cardiolipin-associated ion at mass-to-charge ratio 1489.73, which diminished in the middle-aged hippocampus. Cardiolipins are essential components of mitochondrial membranes, and their loss fits established models of mitochondrial dysfunction in brain aging. The lesson is methodological as much as biological: bulk analysis of the whole brain masks regional lipid remodeling that may drive cognitive decline, and imaging-based approaches can expose changes that homogenized tissue cannot.

Tying everything together, Pearson correlation analysis revealed moderate to strong associations—correlation coefficients exceeding 0.6 in magnitude—between vascular biomechanical measures such as wall shear stress and circumferential cyclic strain and specific lipid classes across tissues, including glycerolipids, phospholipids, and sphingolipids. The authors are careful to stress that these correlations do not establish causality, and because neuronal and glial endpoints were not directly measured, any link between vascular changes and downstream neurological consequences remains speculative. They also acknowledge limitations: the effect sizes were relatively small, the 52-week-old mice represent a middle-aged rather than truly aged cohort, and even careful use of anatomical landmarks could not eliminate all positional variability in imaging and biochemical sampling. More advanced aging or emerging pathology may be required to produce larger, more disruptive shifts in vascular metrics.

Even with those caveats, the study delivers a compelling integrated portrait of neurovascular aging in females. Declining blood flow, falling wall shear stress, stiffening arterial walls, and wholesale lipid remodeling in both vessel and brain appear together in otherwise healthy animals, well before overt disease. That timing matters: if neck vascular dysfunction genuinely precedes brain biochemical alteration, as related work from the same group suggests, the vasculature may offer an early warning system—and a potential intervention target—for the cognitive diseases that disproportionately afflict aging women. By establishing this baseline in disease-free mice, the team has laid groundwork for future mechanistic studies in models of neurodegeneration, and has underscored a point increasingly hard to ignore: brain aging cannot be understood without also understanding the blood vessels that feed it.

Subject of Research: The effects of aging on neck vascular function, vascular and brain lipid composition, and brain biochemistry in female mice.

Article Title: Impact of aging on neck vasculature and brain biochemistry in female mice

Article References: Jones, A. R., Jarrahi, A., Karpowich, K., Moody, E., Renner, L., Brown, L. P., Tressler, C. M., & Crouch, A. C. (2026). Impact of aging on neck vasculature and brain biochemistry in female mice. Physiological Reports, 14(17), Article e71059. https://doi.org/10.14814/phy2.71059

Image Credits: AI Generated

DOI: 10.14814/phy2.71059

Keywords: aging, carotid artery, wall shear stress, arterial stiffness, lipidomics, mass spectrometry imaging, hippocampus, cardiolipin, female mice, neurovascular, sphingolipids, brain biochemistry

The innovation at the heart of the work is a novel geometric model built on a unit-step-function formulation. Unlike classical stenosis models that impose a single cosine, Gaussian, or parabolic profile on a symmetric narrowing, the new framework allows the upper and lower arterial walls to be prescribed independently, capturing true bilateral asymmetry. Surface irregularities, the micro-undulations and protrusions that pathological specimens display, are encoded through independently tunable parameters for each wall. Crucially, phase parameters control the relative axial alignment of the opposing irregularities, permitting systematic exploration of in-phase configurations, where wall peaks align with peaks, and out-of-phase configurations, where a peak on one wall faces a trough on the other. The model also enforces first-order differentiability at the junctions between diseased and healthy lumen, eliminating artificial slope discontinuities that would destabilize analytical and numerical solutions.

Blood itself is treated in an unusually sophisticated way. In vessels 80 to 500 micrometers in diameter, the size of red blood cells, averaging about 7.65 micrometers, becomes comparable to the vessel dimension, and the fluid’s microstructure cannot be ignored. The study therefore models blood as a couple-stress fluid, a formulation first proposed by V.K. Stokes in 1966 that incorporates higher-order stresses associated with particle rotation. This captures the micro-rotation of blood cells and their interaction with surrounding structures, effects that conventional shear-thinning or viscoelastic models miss entirely. The couple-stress parameter, defined as the ratio of vessel radius to an intrinsic microstructural length scale, falls physiologically between roughly 5 and 30 for small arteries, and the analysis employs representative values of 5, 8, and 20.

The flow problem is solved under lubrication and low-Reynolds-number approximations, appropriate because lesion lengths in coronary imaging are several times the local lumen width and because inertial effects are negligible at these scales. The governing equations reduce to a fourth-order differential equation in the transverse direction, and the closed-form solution yields an explicit expression for the pressure gradient as a function of local lumen width and the couple-stress parameter. In the Newtonian limit, as the couple-stress effect vanishes, the solution gracefully recovers the classical plane Poiseuille relation, a rigorous check on the mathematics. The resulting formulas reveal that effective resistance exceeds the classical Newtonian value, particularly in regions where the lumen is most constricted.

The model’s most striking results concern the phase relationship between opposing lesions. When irregular lesions overlap in-phase, with a 47.37 percent axial overlap and mild roughness, the pressure gradient plunges to roughly negative 12 at the throat for the strongest couple-stress case, producing a single broad trough and a predictable bell-shaped rise in wall shear stress. But simply shifting the upper wall by half a unit, inverting the central region to out-of-phase alignment and increasing overlap to 55.55 percent, nearly doubles the peak adverse pressure gradient to negative 24.38, a 99 percent increase relative to the comparable in-phase case. The tortuous channel forces blood through successive accelerations and decelerations, generating sharp, deep spikes in the pressure gradient that dominate the flow field.

Quantitative tables reinforce the message that lesion extent alone is a poor predictor of hemodynamic burden. In one comparison, an out-of-phase configuration with 80 percent overlap and 63.5 percent throat occlusion produced a maximum adverse pressure gradient approximately 231 percent higher and a total pressure drop about 51.8 percent higher than an in-phase baseline, despite having a shorter overlap interval than a rival in-phase case with 89.47 percent overlap. The authors conclude that out-of-phase alignment combined with greater throat occlusion amplifies resistance more strongly than overlap extent alone, helping explain why anatomically similar plaques can impose markedly different functional consequences in patients.

Wall shear stress tells an equally consequential story. In an in-phase configuration with 68.42 percent overlap, shear stress forms a broad bell-shaped peak, reaching a maximum of 9.37 at the throat. In the out-of-phase counterpart with 60 percent overlap, peak shear stress jumps to 12.61, about 34.7 percent higher, while the spatially averaged shear stress remains nearly identical. This divergence means out-of-phase geometries concentrate extreme mechanical loads at discrete micro-constrictions rather than distributing them evenly, imposing localized endothelial injury disproportionate to what aggregate flow indices would suggest. Chronic exposure to such abnormal shear promotes a pro-inflammatory endothelial phenotype, stimulating production of enzymes that degrade elastin and collagen, weakening the arterial wall, and potentially fostering microaneurysm formation with attendant rupture risk.

The downstream clinical cascade modeled by the study is sobering. The marked pressure drop across an irregular stenosis represents irreversible energy loss through viscous dissipation, leaving downstream blood at a permanent pressure deficit even after the gradient normalizes. This weakened driving pressure threatens perfusion of metabolically active organs such as the brain, kidneys, and myocardium, potentially triggering compensatory flow redistribution or vascular steal phenomena. To compensate, the heart must generate higher pressures, a chronic strain linked to cardiac hypertrophy and heightened susceptibility to heart failure. Meanwhile, low and oscillatory shear in the pre- and post-stenotic zones creates athero-prone environments that promote inflammatory cell adhesion and LDL cholesterol infiltration, seeding new lesions or enlarging existing ones in a self-reinforcing cycle of vascular decline.

Context matters here: cardiovascular diseases account for roughly 31 percent of global deaths according to the World Health Organization, and while large-artery atherosclerosis is well studied, small penetrating artery disease remains comparatively neglected. Recent reviews have highlighted persistent challenges in defining, classifying, and treating cerebral small vessel disease, and neuropathological assessment of brain arteriolosclerosis still relies on semi-quantitative scales with only moderate reliability. By providing a mechanistic analytical tool that reflects realistic lesion morphology, the new model could complement imaging-based assessment and contribute to improved risk stratification in atherosclerotic disease, particularly if geometric parameters can one day be extracted from high-resolution clinical imaging.

The study’s broader warning is that smooth-wall simulations may systematically underestimate hemodynamic variations linked to endothelial dysfunction and atherogenesis. Surface irregularities, though modest in amplitude, add remarkable fluctuations to pressure and shear patterns and amplify the damage potential of disturbed flow. Conversely, the couple-stress effect proves nearly negligible in wide, laminar regions but acts as a powerful amplifier wherever geometry concentrates velocity gradients, underscoring the intimate coupling between blood rheology and lesion topology. As a first-order mechanistic foundation for functional lesion assessment, this work marks a significant step toward understanding why the smallest vessels in the body, when diseased, can exert some of the largest influences on human health.

Subject of Research: Mathematical modeling of blood flow in stenosed small arteries with irregular luminal surfaces

Article Title: A novel geometric model of blood flow in stenosed small arteries with luminal surface irregularities

Article References: Elogail, M. (2026). A novel geometric model of blood flow in stenosed small arteries with luminal surface irregularities. Results in Physics, Article 108751. https://doi.org/10.1016/j.rinp.2026.108751

Image Credits: AI Generated

DOI: 10.1016/j.rinp.2026.108751

Keywords: small arteries, stenosis, couple-stress fluid, hemodynamics, wall shear stress, pressure gradient, arteriolosclerosis, cerebral small vessel disease, lubrication approximation, non-Newtonian blood flow, atherosclerosis, mathematical modeling

The biological stage for this idea is set by decades of research into wall shear stress, the tangential frictional force that flowing blood exerts on the endothelial cells lining the vessel lumen. In straight arteries, flow is predominantly laminar and shear stress remains unidirectional and pulsatile within a physiological range of roughly 10 to 20 dynes per square centimeter. Under these benign conditions, endothelial cells stay quiescent and secrete protective molecules, including nitric oxide and prostaglandins, while suppressing vasoconstrictors such as endothelin. They also ramp up expression of the transcription factor KLF2, which coordinates many of these protective gene programs. Trouble begins where vessel geometry changes—at branches, stenoses, and aneurysms—where secondary flows generate complex, multidirectional, and oscillatory shear patterns. Since the 1980s, researchers have recognized that regions exposed to low, oscillating shear stress correlate with the focal development of atherosclerosis and intimal hyperplasia, the pathological thickening of the vessel’s inner layer.

To study how these mechanical signals translate into disease, the authors turn to an unusual but clinically vital model: the arteriovenous fistula, or AVF. An AVF is a surgically created connection between an artery and a vein in the arm, and it serves as the preferred vascular access for hemodialysis patients, whose blood must be routed through an extracorporeal circuit. The surgery abruptly transforms the hemodynamic environment of both vessels. The vein, previously accustomed to low, gentle flow, is suddenly subjected to arterial pressures and flow rates many times higher than normal. In many patients the vessels adapt gracefully, enlarging and thickening in a process called outward remodeling. But in a large fraction of patients, the disturbed flow instead triggers intimal hyperplasia—smooth muscle cell proliferation and migration that narrows the vein over months to a few years, reduces blood flow, and ultimately causes the fistula to fail. Because the onset of these changes is precisely known, AVFs offer researchers a rare window into the earliest stages of flow-induced vascular disease.

The group’s computational work began with medical image-based computational fluid dynamics simulations of patient-specific AVFs. One of their earliest and most striking observations was that the massive increase in blood flow generates turbulent-like conditions at the anastomosis and in the juxta-anastomotic vein. The venous wall, previously exposed to steady unidirectional shear, suddenly experiences rapidly fluctuating forces varying in both magnitude and direction. The team showed that the venous segments most affected by intimal hyperplasia were precisely those bathed in this unsteady, transitional flow, suggesting that the chaotic mechanical environment pushes endothelial cells into a signaling mode that converts smooth muscle cells from a quiescent state into a proliferative, migratory phenotype.

But shear stress told only part of the story. Extending their analysis beyond the fluid domain, the researchers employed fluid-structure interaction simulations, which couple blood flow to the mechanical behavior of the vessel wall, using the open-source turtleFSI solver. These preliminary models revealed something unexpected: the unsteady flow near the anastomosis induces rapid motion of the vessel wall itself, resembling mechanical vibrations with frequencies ranging from about 50–70 Hz to more than 200 Hz. This finding reframed the problem. Disturbed flow was not only altering the frictional environment of endothelial cells; it was physically exciting the entire wall, meaning every cell within it—endothelial cells, smooth muscle cells, and fibroblasts—was being subjected to a high-frequency mechanical stimulus. The team hypothesized that these vibrations might directly influence vascular cell biology, potentially opening entirely new therapeutic avenues if the responsible pathways could be identified.

Although research on high-frequency mechanical stimulation of vascular cells remains limited, the available evidence lends support to the hypothesis. In vivo and in vitro studies consistently show that vibrations modulate vascular cell phenotype and gene expression, inducing cytoskeletal reorganization, altered proliferative activity, and changed secretion of signaling molecules. These responses are associated with vascular dysfunction, disruption of the internal elastic lamina, and activation of inflammatory pathways. Several mechanotransduction pathways have been implicated, including ERK1/2 signaling in smooth muscle cells, which promotes intimal thickening and recruits inflammatory cells to sites of injury, along with vibration-responsive regulators such as Syn4, VEGF, KLF2, ICAM-1, and NFATc3. Together, these findings provide a plausible mechanistic basis for vibration-driven stenosis.

To test the idea more rigorously, the team built a substantially more realistic FSI model. They replaced assumed steady-state inlet flows with patient-specific, visit-specific pulsatile boundary conditions, modeled wall mechanics with a three-term Mooney–Rivlin constitutive law calibrated on experimental literature data, assigned distinct material properties to artery and vein, and accounted for the mechanical influence of surrounding perivascular tissue through Robin boundary conditions. Applying this refined model to a patient whose radio-cephalic AVF failed one and a half years after surgery, monitored at five time points with magnetic resonance imaging and ultrasound, they found that flow instabilities produced elevated vibration amplitudes and high-frequency strain in the initial segment of the cephalic vein—exactly the region where stenosis later developed. Vibration intensity peaked at the time point just before stenosis formation, hinting at a causal role in triggering the pathological remodeling.

The most compelling evidence came from a one-year longitudinal study of six patients with native distal radio-cephalic AVFs, tracked with MRI and Doppler ultrasound and simulated at multiple time points. Two patients maintained adequate patency, while four developed complications—two stenoses and two excessive dilatations. Fistulas that remained patent showed negligible high-frequency components in wall displacement spectrograms throughout the year. Every fistula that later developed complications, by contrast, exhibited high-frequency wall vibrations, and these signatures were prominent before the onset of adverse remodeling, whether it took the form of narrowing or excessive enlargement. Remarkably, distinct spectral patterns distinguished the two outcomes: patients destined for intimal hyperplasia displayed two narrow, well-defined frequency bands that shifted higher and weakened after stenosis onset, while those developing excessive dilatation showed a single prominent band at a lower frequency. The vibrations appeared to encode, in advance, the type of remodeling the vessel would undergo.

Finally, the team brought the computational frequencies into the laboratory. Using a purpose-designed vibrational loading device, they exposed monolayers of endothelial cells and smooth muscle cells to vertical vibrations at 75 and 150 Hz, frequencies matching bands observed in the FSI simulations. Cells vibrated at 150 Hz migrated significantly farther and faster over six hours than controls. Smooth muscle cells vibrated for 24 hours showed a marked increase in Ki67-positive nuclei, a marker of proliferation. Most intriguingly, smooth muscle cells cultured in conditioned medium from endothelial cells previously exposed to 75 or 150 Hz vibrations also proliferated more vigorously, indicating that vibration changes endothelial secretion of soluble factors that then drive smooth muscle cell growth. Taken together, the simulations, longitudinal imaging, and in vitro experiments converge on a provocative conclusion: high-frequency wall vibrations are not a byproduct of disturbed flow but a candidate mechanobiological stimulus in their own right, one that acts across the entire vessel wall and on all vascular cell types. The authors call for a paradigm shift in vascular research—looking beyond endothelial shear stress toward vibration-related forces—which could reshape how clinicians predict AVF failure and, perhaps, how future drugs are designed to protect failing blood vessels.

Subject of Research: Flow-induced high-frequency vibrations of the vascular wall as a mechanobiological stimulus driving vascular remodeling and intimal hyperplasia in arteriovenous fistulas.

Article Title: Flow-Induced High-frequency Vascular Wall Vibrations: A New Mechanobiological Stimulus for Vascular Remodeling?

Article References: Flow-Induced High-frequency Vascular Wall Vibrations: A New Mechanobiological Stimulus for Vascular Remodeling?. (n.d.). https://doi.org/10.1007/s10439-026-04360-x

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04360-x

Keywords: arteriovenous fistula, hemodialysis, wall shear stress, mechanobiology, fluid-structure interaction, intimal hyperplasia, vascular remodeling, endothelial cells, smooth muscle cells, high-frequency vibrations, computational fluid dynamics, vascular access failure