New insights into arterial stiffening: does sex matter?

Abstract

This review discusses sexual dimorphism in arterial stiffening, disease pathology interactions, and the influence of sex on mechanisms and pathways. Arterial stiffness predicts cardiovascular mortality independent of blood pressure. Patients with increased arterial stiffness have a 48% higher risk for developing cardiovascular disease. Like other cardiovascular pathologies, arterial stiffness is sexually dimorphic. Young women have lower stiffness than aged-matched men, but this sex difference reverses during normal aging. Estrogen therapy does not attenuate progressive stiffening in postmenopausal women, indicating that currently prescribed drugs do not confer protection. Although remodeling of large arteries is a protective adaptation to higher wall stress, arterial stiffening increases afterload to the left ventricle and transmits higher pulsatile pressure to smaller arteries and target organs. Moreover, an increase in aortic stiffness may precede or exacerbate hypertension, particularly during aging. Additional studies are needed to elucidate the mechanisms by which females are protected from arterial stiffness to provide insight into its mechanisms and, ultimately, therapeutic targets for treating this pathology.

INTRODUCTION

Arterial stiffening is defined as resistance to deformation or a loss of elastic compliance due to changes in the geometry and microstructure of the vascular wall (11, 157). Stiffening alters the mechanical behavior and properties of the vasculature, which can be described as a decrease in distensibility (a change in diameter with pressure) or compliance (the ability to store volume) (155). Furthermore, stiffening encompasses a decrease in elasticity, minimizing the ability to store energy during deformation, which is integral for performing work on the blood during recoil of the artery (60). Compared with smaller muscular arteries, the increased elastin content in large arteries forms a reservoir for storing energy during the cardiac cycle (64). These characteristics of large blood vessels allow them to maintain peripheral perfusion during diastole (Fig. 1). Passive material properties and active cellular processes contribute to the stiffness of the vessel wall (11). Arterial stiffening also increases cardiac afterload and transmits pressure oscillations to end organs, both of which promote cardiovascular morbidity and mortality (113).

Fig. 1.

A: healthy artery. Energy generated by distension of the arterial wall during systole is used during diastole to maintain blood flow and decrease pulse pressure downstream. B: stiffened artery. Poor energy absorption results in higher pulse pressure, reduced flow during diastole, increased cardiac afterload, and end organ damage.

While the elastic properties of arteries have been studied since the late 1800s (175), the first publication containing the term “arterial stiffness” was published 70 yr ago in 1947 (12) and the next did not appear until 1970 (228) (Fig. 2). Interest in this topic steadily rose in the 1990s, and the number of annual publications broke the 100 mark in 1999. A rapid surge in publications started in the 2000s, perhaps in response to the 2003 European guidelines for the management of arterial hypertension, which acknowledged the need for further research on the potential predictive power of large artery compliance (55). Recognition of arterial stiffness as a clinically important variable has lagged in the United States. The National Heart, Lung, and Blood Institute identified this knowledge gap in 2009 and requested applications for grants to study arterial stiffness, resulting in a significant increase in funding on this topic (157). In 2015, the American Heart Association recommended more research on arterial stiffening (208), but still no clinical guidelines exist in the United States (100).

Fig. 2.

Graphical representation of the number of peer-reviewed articles containing the search term “arterial stiffness” alone or with the additional search term “sex” through 2017.

Inclusion of arterial stiffness with standard, well-established risk factors improves the prediction of cardiovascular events. In a 1999, in a study of patients with hypertension, arterial stiffness was found to be an even stronger predictor of cardiovascular mortality than blood pressure (18). The Rotterdam Study found that, in healthy, normotensive patients, arterial stiffness independently predicts stroke and coronary heart disease (135). The Framingham Heart Study supported these results, finding that elevated stiffness correlates with an increased probability for a major cardiovascular event (139). Together, these studies have provided strong clinical evidence that arterial stiffening precedes or accompanies detrimental cardiovascular events, and treatment of this pathology in otherwise healthy patients may provide protection.

Despite the impressive increase in research on “arterial stiffness” in the past decade, addition of the term “sex” to a PubMed search reveals a dire lag in the consideration of sex as an important biological variable in arterial stiffness (Fig. 2). Carotid artery compliance is similar in children but is significantly greater in adult women than men (130). While increased arterial stiffness is lower in young women than men, older women have increased arterial stiffness, particularly with aging (34). In patients with cardiovascular comorbidities, the sexual dimorphism is also reversed, with greater arterial stiffness in women than men. For example, in patients with diastolic heart failure, measures of arterial stiffness are greater in women than men (33). This greater arterial stiffness in women with cardiovascular comorbidities holds true even after adjustment for age, body mass index, mean arterial pressure, and heart rate (177). Given that sex differences play an important role in cardiovascular physiology, inclusion of arterial stiffness as an additional variable during diagnosis may serve as a biomarker for monitoring cardiovascular health.

MEASUREMENTS OF ARTERIAL STIFFNESS

Pulse Pressure

Blood pressure is the conventional method for assessing cardiovascular risk because of its accessibility and high predictive value (100). While systolic pressure increases continuously with aging, diastolic pressure increases until midlife before declining (69). The divergence of systolic and diastolic pressures causes a plateau in mean arterial pressure but an amplification of pulse pressure. Moreover, the increase in pulse pressure is deemed a product of aging-induced arterial stiffening, since it occurs in normotensive and hypertensive patients. Hence, pulse pressure is a better predictor of cardiovascular risk in older patients (75) than in midlife adults (46). Moreover, tracking the longitudinal change in pulse pressure provides an even stronger indication of cardiovascular disease (166). Despite the potential for this biomarker, some meta-analyses failed to find added benefit for including pulse pressure; hence, it has been excluded from current clinical guidelines (100, 117). Pulse pressure should be interpreted with caution, because increased aortic stiffness propagates transmission of energy to distal vessels and does not necessarily increase pulse pressure (82, 191).

Pulse wave velocity (PWV), a measure of the distance traveled by the pulse wave over time, is the gold standard for assessing arterial stiffness (18). The two arterial sites most commonly used are carotid and femoral, although slight variations do exist. Clinical devices used to measure PWV include Complior, SphygmoCor, PulsePen, ultrasound, and magnetic resonance imaging (22, 113). Drawbacks of carotid-femoral PWV include averaging over heterogeneous vascular structures and accurately measuring the distance traveled (212). To circumvent these issues, mathematical models have been developed to estimate PWV using only one arterial site (218) or local PWV of a single arterial segment (162). European guidelines suggest that patients with PWV > 10 m/s should be considered at higher risk for cardiovascular events (212). Average adult values are 7.4 m/s for women and 8.2 m/s for men, and this sex difference is maintained even after correction for age and blood pressure (170). Reference values start at 6.1 (4.6–7.5) m/s for young, healthy individuals but increase with aging and hypertension.

The augmentation index (AIx) is a measurement of arterial stiffness derived from the pressure waveform using applanation tonometry by applying slight pressure on the radial artery. Values are calculated as the difference between the height of the reflected wave and the systolic peak divided by the pulse pressure (113). Radial artery applanation tonometry avoids amplification of peripheral pressure and enables analysis of the central waveform (149). As arteries stiffen, the reflected wave travels faster, resulting in an increased height and, hence, a higher AIx (154). However, clinical studies have shown that decreasing arterial stiffness does not always reduce AIx (116, 140), which may be related to the influence of heart rate on this measurement (222, 223). AIx may have better predictive value when pressure and flow measurements are obtained to separate forward and backward waves (219).

The β-stiffness index [β = ln(SBP/DBP) × D/∆D, where SBP is systolic blood pressure, DBP is diastolic blood pressure, and D is artery diameter (155)] and the cardioankle vascular index (CAVI) are closely related and were developed to be independent of blood pressure. Normal values obtained from the carotid artery are ~10.3 in normotensive men and women (118), but values can be much higher, depending on the artery that is measured (84). Unlike PWV, the β-stiffness index takes into account the blood pressure and is a local measurement that can be used to assess stiffening in both large and small vessels. Because PWV is influenced by blood pressure and the β-stiffness index measures local vascular properties, CAVI overcomes the individual limitations of each by combining the two approaches (186): CAVI involves recording the distance from the brachial artery to the ankle as well as the time difference between the closing of the aortic valve and the arrival of the arterial pressure wave. Not surprisingly, CAVI highly correlates with the β-stiffness index, increases with aging, and is lower in women than men before 70 yr of age (147, 205). Reference values in healthy patients range from 6.6 in young women to 9.4 in older men (147). While this technique is promising, it has been subjected to limited clinical testing and, therefore, has yet to be proven as a universally applicable predictor of cardiovascular outcomes (178).

PATHOLOGIES ASSOCIATED WITH ARTERIAL STIFFENING

Aging

Although aging is not a disease, it correlates more highly with arterial stiffness than any other variable (53). The Baltimore Longitudinal Study of Aging found that PWV increases with each decade of life starting at 40 yr of age (5). Vascular aging occurs across species and in otherwise healthy individuals, but cardiovascular diseases such as hypertension and metabolic syndrome exacerbate these changes (146). Interventional studies have shown that lifestyle modifications, such as caloric restriction, weight loss, and reduced Na⁺ intake, can partially attenuate the impact of aging on arterial stiffening (153). This benefit is found in individuals with cardiovascular risk factors as well as in healthy subjects.

Women have lower arterial stiffness than age-matched men from puberty (3) to menopause (220). After the surge in sex hormones during puberty, stiffness increases in men, while arterial elasticity is maintained or improved in women. Adults of both sexes experience an increase in arterial stiffening with aging, but the increase is steeper in men than women (5). The SardiNIA Study showed that the rate of stiffening parallels increases in blood pressure in women but becomes dissociated in men over time (182). Since changes in stiffness accompany adolescence, it is not surprising that the sexual dimorphism in arterial stiffening disappears after menopause (7). The largest change in female vascular compliance occurs around the menopausal transition (150). In healthy premenopausal women, arterial stiffening measurements, such as pulse pressure and AIx, remain constant over the menstrual cycle (160). However, postmenopausal women have stiffer arteries than aged-matched men, even though mean arterial pressure is similar (32, 151, 215).

Hypertension

Increased systolic pressure is associated with arterial stiffening due to vascular wall degeneration, progressive remodeling, and loss of stored energy (198). However, the temporal relationship between these two pathologies is still under debate. Longitudinal analysis of data from the Genetic Epidemiology Network of Arteriopathy Study shows that higher PWV and AIx are associated with greater increases in blood pressure over time (33). Another longitudinal study of Framingham offspring found that arterial stiffening measured by PWV predicts future hypertensive status (102). In contrast, evidence from young patients in the Bogalusa Heart Study indicates that hypertension precedes changes in arterial stiffening (27). These disparate findings may indicate the importance of age, with hypertension preceding stiffening in the young population but after stiffening in aging patients.

Hypertension is a confounding factor that clouds most clinical studies of the impact of sex on arterial stiffening. However, a few studies have provided evidence that sex impacts the relationship between blood pressure and arterial stiffening. For example, the Australian National Blood Pressure Study found that, despite similar blood pressures, the β-stiffness index and AIx are greater in aging women with hypertension than age-matched men (17). In a longitudinal multiethnic cohort, arterial stiffening is associated with aging and blood pressure in both sexes (200). Race, diabetes, and cholesterol are associated with stiffness in men, but only educational level is a significant predictor in women. Women born to parents with hypertension are at higher risk for increased arterial stiffness than men, suggesting sex differences in genetic factors (21). These data indicate that pathological or environmental factors may be important, depending on sex.

Metabolic Syndrome

After adjustment for age, PWV is higher in patients with metabolic syndrome (143). The elevation is most striking in older patients with multiple risk factors (53). Obesity, hyperglycemia, and hypertension independently contribute to this increase (179). Insulin resistance also plays a part in the development of metabolic syndrome and is associated with lower distensibility in young obese patients (211). Trajectories show that greater body fat and blood pressure in adolescence predict stiffer arteries in adulthood, indicating a cumulative effect of multiple risk factors on arterial elasticity (58).

Type 2 diabetes is associated with an increase in PWV in women but not in men (37), and the arterial stiffness that accompanies metabolic syndrome is more pronounced in women than men (220). Obesity is a more important risk factor for women, since body mass index correlates positively with arterial stiffening in women, but, surprisingly, a negative relationship is found in men (152). The mechanisms that underlie these sex differences in arterial stiffness are complex but may provide insights into potential pathways for promoting healthy vascular aging.

Immune Disorders

Increased PWV and AIx are observed in patients with autoimmune disorders, including Hashimoto’s thyroiditis (197), inflammatory bowel disease, rheumatoid arthritis, and systemic lupus erythematosus (50, 185). In patients with rheumatoid arthritis without clinical signs of cardiovascular disease, 8 wk of treatment with an antibody against TNF-α significantly attenuated vascular inflammatory markers as well as PWV (125). Inflammation also provides a link between the cardiovascular conditions mentioned above and their impact on stiffening. In stroke patients with metabolic syndrome, higher levels of inflammatory markers, such as C-reactive protein, are associated with stiffer arteries (210). Hypertension is also linked to cytokine production and immune cell infiltration (137) and involves accumulation of inflammatory cells in the kidneys and vasculature that induce oxidative stress and end organ damage.

The Gutenberg Health Study found that arterial stiffness correlates with inflammatory markers, including C-reactive protein and interleukins, in both men and women (8). Female subjects of multiple species have more robust immune responses and quicker pathogen clearance than male subjects, but this heightened response may lead to a higher incidence of autoimmune disorders (105). This female phenotype includes greater antibody response, higher B cell populations, and more circulating immunoglobulins, regardless of age (207). Circulating inflammatory markers in female subjects are increased even at baseline, and cell populations differ by sex, with natural killer CD8⁺ cells predominating in male subjects and CD4⁺ cells predominating in female subjects (1). While research on the interactions between inflammation and arterial stiffness is only at the beginning stages, since sex influences the role of immune cells in the development of hypertension (232), it most likely plays a role in arterial stiffness as well.

MECHANISMS AND PATHWAYS IN ARTERIAL STIFFENING

Arterial stiffening is a multifaceted disorder that involves the interaction of several molecular mechanisms, including sex hormones, extracellular matrix (ECM) alterations, endothelial and smooth muscle cell pathophysiology, oxidative stress, and inflammation. Each of these factors will be discussed in regard to sex differences.

Male Sex Hormones

Testosterone belongs to a family of androgens, including androstenedione and dihydrotestosterone, which bind to the androgen receptor to induce physiological actions. Multiple studies have indicated that low serum testosterone is associated with increasing mortality in men (81). Furthermore, population-based studies show that low total serum testosterone impairs endothelial function in men (54), while administration of testosterone increases endothelium-derived relaxation factor and improves flow-mediated vasodilation in hypogonadal men (16, 68, 187). Androgens also play a role in vascular remodeling, since decreased circulating testosterone levels correlate with increased arterial stiffness (45, 93, 111). Testosterone deficiency also increases intima-media thickness in the common carotid artery in men with metabolic disorders (39). In aging men, hypogonadism is associated with increased arterial stiffening in central and peripheral vessels, while testosterone replacement reduces PWV (226). Conversely, blockade of the production of testosterone as a treatment for prostate cancer increases arterial stiffening (190). These results indicate a protective role for testosterone in men.

Female Sex Hormones

Estrogen binds to estrogen receptor (ER)α and ERβ to initiate genomic responses, while G protein-coupled ER (GPER) induces rapid signaling (171). Which receptor is responsible for the positive vascular effects of estrogen has not been clearly delineated. Genetic deletion of endothelial ERα reduces high-fat diet (HFD)-induced arterial stiffness in both male (126) and female (128) subjects, suggesting that this genomic receptor has adverse effects on vascular remodeling. Conversely, the mesenteric arteries of female mice lacking ERβ are significantly less distensible than those of wild-type control mice, indicating a protective role (49). The ratio of ERs may also be important, since increasing GPER expression or decreasing ERα attenuates the vascular remodeling induced by carotid artery injury (78). Previous results from our laboratory have shown that, during salt-sensitive hypertension, membrane-initiated signaling via GPER has no impact on collagen or elastin but decreases oxidative stress and glycosaminoglycan (GAG) deposition in the medial layer of the aorta (123). Therefore, there is still much to understand about ER signaling and how these multiple receptor pathways collaborate to impact arterial stiffening.

Collagen, elastin, and glycoproteins make up the ECM and provide physical scaffolding for blood vessels as well as a platform for de novo biochemical reactions (180). Collagen contributes to the strength and stability of the arterial wall. Fibrillar collagen types I and III are major components of the intima, media, and adventitia, whereas collagen types I, III, IV, and V are present in the basement membrane. Collagen fibers are continuously deposited and removed from the arterial wall throughout life (112, 199). While elastic fibers form during development and are stretched during somatic growth, collagen fibers develop an undulated pattern, which contributes to their mechanical function. During significant levels of strain, such as an acute increase in blood pressure, collagen fibers straighten to limit overdistension and rupture of vessels. Arterial stiffening is frequently associated with alterations in collagen turnover and may include increased collagen production (189), altered ratio of collagen fiber types (52), and enhanced collagen cross-linking (13).

Elastic fibers provide compliance and resilience to enable storage of energy in the arterial wall during systole to maintain flow during diastole. Elastic fibers are susceptible to proteolysis (92) and mechanical fatigue (119) because they do not regenerate during adulthood (36). Acquired or congenital disruption of elastic fiber integrity induces mechanical consequences, such as loss of resilience, decrease in elastic recoil, and diminished distensibility (61). Moreover, elastic fiber disruption promotes changes in smooth muscle cell phenotype (103), inflammatory status (2), and the cell cycle (138). Increased structural stiffness is frequently reported after loss of elastic fiber integrity (56, 131, 216), but recent biaxial assessments have found that intrinsic material stiffness is maintained while elastic energy storage is decreased (59, 60). Therefore, despite the ability to maintain mechanical homeostasis, physiological vascular function is still compromised with loss of elastic fibers.

GAGs are unbranched polysaccharides that covalently bind to proteins to form proteoglycans. These highly charged hydrophilic macromolecules are secreted into the extracellular space to form a porous hydrated gel. GAGs reinforce collagen and elastic fibers and play an important role in arterial stiffening and mechanohomeostasis (196). The luminal glycocalyx is the first line of defense for the vascular wall, and changes in these proteoglycans impact the mechanics of the underlying endothelium (57). A high-salt diet causes damage to the glycocalyx, allowing Na⁺ to permeate the vascular wall (156). As a result, nitric oxide (NO) production is suppressed, leading to endothelial dysfunction and vascular stiffening (110). In the early stages of vascular injury before changes in collagen are detected, GAGs are upregulated and promote alterations in smooth muscle cell phenotype (221). Elevated GAGs are detected in aortas of spontaneously hypertensive rats before the onset of hypertension and compromise distensibility and mechanosensing in the aorta (172, 173).

Sex Differences in the ECM

Few studies have directly compared the arterial ECM composition of men with that of women. However, theoretical investigation indicates that aging has a greater effect on ECM stiffness in men (10). Estrogen deficiency promotes ECM remodeling that is associated with arterial stiffening, including an increase in collagen synthesis and a decrease in degradation (65, 193, 230), while estrogen administration to ovariectomized rats decreases the total amount of both collagen and elastin (63). Femoral arteries from female mice lacking endothelial ERα show no alterations in collagen or the internal elastic lamina (128). Male mesenteric arteries with endothelial-specific deletion of ERα have more elastin and collagen as well as decreased activity of matrix metalloproteinases, which break down the ECM (126). On the other hand, mesenteric arteries from female mice lacking ERβ have more elastin than wild-type control mice (49). Estrogen decreases GAGs in the aorta of ovariectomized rabbits and attenuates carotid injury by reducing proteoglycan deposition (4, 161). The mechanism for these estrogenic effects may be transcriptional, as estrogen reduces gene expression for proteoglycans as well as the enzyme involved in their synthesis (70, 136). Alternatively, since proteoglycan synthesis is inhibited by antioxidants in vascular smooth muscle, estrogen may alter GAGs indirectly by reducing oxidative stress (26). Along these lines, we found that in vivo GPER activation attenuates salt-induced increases in aortic GAGs in hypertensive mRen2 female rats and highly correlates with tissue oxidant levels (123). The male sex hormone testosterone inhibits collagen synthesis in cardiac fibroblasts (225) but increases GAGs in vascular smooth muscle cells (83). In cultured aortic smooth muscle cells, estrogen, progesterone, and testosterone decrease collagen, while only the female sex hormones increase elastin (148). The significance of this latter finding is unclear, since functional elastic fibers are not synthesized in adulthood (36). These data indicate mixed results with regard to the impact of sex hormones on ECM factors and are most likely complicated by the presence of estrogens and androgens in both male and female subjects as well as the presence of multiple receptor signaling pathways.

Vascular Smooth Muscle Cells

Intrinsic changes in the structural properties and phenotype of vascular smooth muscle cells contribute to arterial stiffening, yet there has been limited work assessing this cell type. Vascular smooth muscle cells play a critical role in vascular function by regulating tone, pressure, and flow. Moreover, aging, oxidative stress, and hypertension directly impact vascular smooth muscle cells by increasing apoptosis (231) and inducing cytoskeletal stiffening (169, 184). Mechanical homeostasis between smooth muscle and the ECM impacts cell phenotype, as well as the structure and content of the ECM, during remodeling (94). Additional investigation is needed to fully understand this relationship and the molecular mechanisms that contribute to arterial stiffening.

Transcriptional profiling of aortas from young and old male and female monkeys reveals higher expression of genes regulating the ECM as well as genes that play a role in switching smooth muscle cells from the contractile to the synthetic phenotype in aging male monkeys (167). In addition, the observation that aging-induced alterations in stiffness were absent after actin depolymerization (169) emphasized the importance of smooth muscle cell phenotype. Sex differences also impact the proliferation of vascular smooth muscle cells. While both male and female rat aortic smooth muscle cells express ERs, the impact of genetic deletion of these receptors has, in some cases, opposite effects on cell proliferation (91). Sexual dimorphism in this cell type is also found in response to induced apoptosis: more resistant cells and increased organization of actin fibers in female subjects (201). Therefore, in smooth muscle cells, genomic as well as hormonal programming most likely contributes to sex differences in function.

Endothelial Dysfunction and NO

Arterial stiffening and endothelial dysfunction are frequently observed simultaneously (188) and are key components of vascular aging (183). Endothelial dysfunction is characterized by impaired NO release and may be due to decreased availability of substrate and/or cofactors necessary for enzymatic action by NO synthases (23). This “uncoupling” event also results in oxidative stress (44), both of which promote vascular remodeling. Systemic inhibition of NO synthase increases the carotid β-stiffness index in men and postmenopausal women within 15 min (203). Similarly, inhibition of NO synthase for 3 wk increases PWV in spontaneously hypertensive rats (98). In the reverse direction, structural changes in the vascular wall increase endothelial permeability, allowing infiltration of inflammatory molecules and exacerbating stiffness (96). Arterial stiffening is attenuated when either the NO substrate l-arginine or the NO precursor nitrite is increased (188). The endothelial dysfunction observed in autoimmune disorders most likely contributes to the increased prevalence of stiffness in this patient population (204).

While arterial stiffening is associated with lower NO bioavailability, it is well established that estrogen increases NO in the vasculature (25). Estrogen increases NO bioavailability and decreases vascular tone (24) and negates arterial stiffening in postmenopausal women (72, 181) and ovariectomized rats (206). Although estrogen improves endothelial function, cross-sectional studies on early-menopausal women have indicated that it does not decrease carotid intima-media thickness (14). In postmenopausal women, administration of estrogen or the NO synthase cofactor tetrahydrobiopterin improves carotid artery compliance (142). Interestingly, tetrahydrobiopterin provides no additional benefit when administered with estrogen, suggesting that these treatments share a similar mechanism. Estrogen increases endothelial NO through membrane-initiated signaling (85) via ERα (28) and/or GPER (120) as well as transcriptionally by upregulating mRNA for endothelial NO synthase (106). Estrogen regulates inducible NO synthase in a bidirectional manner, with ERα repressing and ERβ increasing its transcription (209). Acute signaling by GPER also increases NO production in brain microvascular endothelial cells (6). Interestingly, the increased NO was associated with increased cell stiffness, as measured by atomic force microscopy. The antiaging effects of resveratrol are also associated with increased NO bioavailability, which may be, at least in part, mediated by ERα (107). These studies have suggested a close relationship between ER signaling and NO bioavailability, which may at times be difficult to separate.

Oxidative Stress

Increases in reactive oxygen species (ROS), including superoxide and hydrogen peroxide, are detrimental to vascular function and remodeling (71). Sources of ROS include NADPH oxidases in endothelial and smooth muscle cells as well as mitochondrial production. In addition to the intimal and medial layers of the vasculature, perivascular adipose tissue is another source of ROS that may become particularly important during aging (67). ROS directly induce vascular damage and decrease NO bioavailability by combining with NO to form peroxynitrite (80). Reductions in superoxide levels (67) and inhibition of mitochondrial ROS (73) attenuate aortic PWV in aging mice.

While ROS contribute to vascular remodeling, estrogen protects against oxidative damage (224). Infusion of the antioxidant ascorbic acid improves carotid artery compliance in postmenopausal women, indicating that menopause-induced oxidative damage promotes arterial stiffening (89). Mechanisms for estrogenic protection from oxidative stress include decreased NADPH oxidase activity, increased superoxide dismutase activity, stimulation of NO production to quench ROS, and neutralization of excess ROS by the phenolic hydroxyl group of estrogen (109). Selective activation of the GPER reduces oxidative stress in endothelial cells (20) and pancreatic β-cells (108), and our work in the salt-sensitive mRen2 female rat provides evidence that GPER reduces ROS and arterial remodeling in vivo (121, 123).

Inflammation

Changes in the inflammatory profile of the vasculature may promote arterial stiffening. Expression of proinflammatory cytokines is increased in isolated coronary arteries from 25-mo-old Fischer-344 rats (35), whereas IL-6, TNF-α, and monocyte chemoattractant protein-1 are increased in vascular endothelial cells from aging humans (47). Androgen receptors are present in the thymus, and testosterone has suppressive effects on the immune system by decreasing production of both B and T cells (158). ERs are expressed in the thymus, peripheral T cells, and bone marrow and normalize immune function by increasing T helper cells and lowering B cell production in ovariectomized animal models (158). Estrogen exhibits protective anti-inflammatory properties, such as its ability to reduce production of proinflammatory cytokines, including IL-1β, IL-6, and TNF-α (164, 227). Production of TNFα and IL-1β increases during the naturally occurring drops in estrogen during the ovarian cycle, and ovariectomy increases production of TNF-α as well as IL-17, a key modulator of IL-1, IL-6, and TNF-α (38). In addition to the impact of estrogen, there is evidence for a suppressive effect of androgens on male macrophage activity that may contribute to sex differences in the inflammatory response (74, 163).

ANIMAL MODELS OF ARTERIAL STIFFNESS

Disease modeling in animals is important for understanding cardiovascular pathophysiology and advancing translational research. However, designing a “pure” model of arterial stiffness is complex, because arterial stiffening develops gradually over time. Instead, arterial stiffening is usually induced by dietary or genetic manipulations, aging, or hypertension.

Few studies have addressed the impact of estrogen on arterial stiffening in female subjects in the absence of other pathologies. In ovariectomized Sprague-Dawley rats, low-dose estrogen increases the distensibility of coronary but not mesenteric arteries (229). Estrogen was also found to attenuate aging-induced increases in mesenteric artery wall thickness and restore vascular matrix metalloproteinase activity (230). While studies in ER global knockout mice have not been performed, genetic deletion of steroid receptor coactivator-1, which associates with and promotes the actions of the nuclear ERs, results in increased blood pressure and aortic stiffness (90). These studies have indicated that estrogen and, in particular, the transcriptional actions of estrogen may significantly protect from arterial stiffening. Many questions remain, however, regarding the receptors and signaling pathways that mediate the actions of this hormone.

Consumption of a HFD promotes arterial stiffening in male Wistar rats (165), young and old male B6D2F1 mice (87), and male C57BL/6J mice (15). Although a side-by-side comparison of both sexes was not performed, female C57BL/6J mice are susceptible to HFD-induced arterial stiffness, and either mineralocorticoid receptor blockade (41) or genetic deletion of the endothelial mineralocorticoid receptor (101) ameliorates the damage. Additional studies using this same model have shown benefits for exercise (159), inhibition of dipeptidyl peptidase-4 (127), and amiloride treatment (132). Contrary to the well-known benefits of vascular ER signaling, deletion of ERα from endothelial cells attenuates vascular stiffening in female (128) and male (126) mice. In contrast to studies in healthy animals, these results indicate that genomic estrogen signaling in endothelial cells may promote, rather than attenuate, arterial stiffening. Since the protective effect of female sex in rats with metabolic syndrome is absent during feeding of a HFD (95), these dietary models may not be optimal for studying sex differences in arterial stiffening.

Multiple genetic mouse models show the significant impact of ECM factors on arterial stiffening. Few studies have examined the impact of sex, and many have found a lack of sex differences. Large arteries from mice heterozygous for the elastin gene (Eln^+/−) have increased blood pressure and more smooth muscle cell layers with smaller luminal diameters, which normalize wall stress in the face of decreased elastin (56). Analysis of sex differences in this mouse model found no significant differences in blood pressure, pulse pressure, length of the ascending aorta, and axial stretch (114). The impact of genetic deletion of other proteins that are critical for the formation of mature elastic fibers, such as fibulin-4 (115) and fibulin-5 (217), has also been characterized; fibulin-5 shows a phenotype of structural stiffness and decreased energy storage that is independent of sex (60). The aortas of male and female fibrillin-1 hypomorphic mice are similarly stiffer, with smaller internal diameters and disorganized elastic fibers (131). In addition to a lack of sex differences at baseline, exercise training of fibrillin-1 hypomorphic mice induces no sex differences in vascular structure (133). Although genetic deletion of these ECM components affects arterial stiffening to the same degree in male and female animals, the relatively few studies indicate the need for additional work, including studies of other ECM components, addressing the potential impact on sexual dimorphism in arterial stiffening.

Aging is a nonmodifiable factor that contributes to arterial stiffening. Remarkably, the longest-lived rodent, the naked mole rat, has no age-related cardiovascular diseases and no detectable increase in PWV (77). In contrast, PWV is increased in <26-mo-old male C57BL/6 mice and associated with alterations in NO (188). Arterial stiffening also manifests in macaque monkeys during aging, along with alterations in collagen and elastin (167). This nonhuman primate study provides strong evidence of sex differences, with female monkeys having less stiffness and maintaining elastin density, despite equal pulse pressures. Analysis of gene expression from these aortas found that while >600 genes were impacted by aging, <5% of these altered genes were the same in male and female monkeys (168). Another nonrodent animal model that displays sex differences in PWV is the domestic fowl (176). Interestingly, while young female chicks have less stiffness than age-matched male chicks, this sex difference is absent in the prepubertal age group but returns in sexually mature chickens. Therefore, aging-induced mechanisms, as well as sex differences in the clinical measurements, may be necessary to diagnose vascular changes.

Multiple models of hypertension show increases in arterial stiffness, including animals infused with angiotensin II (51) or N-nitro-l-arginine methyl ester (66) and genetic models such as the spontaneously hypertensive rat (31). Unfortunately, the majority of these studies have not included female animals, and even fewer studies have directly compared the two sexes. In the Dahl salt-sensitive rat model, male and female rats display a similar increase in PWV from 3 to 6 mo of age (88), but the quantitative trait loci associated with arterial stiffening are sex-specific (40). In studies focused on arterial stiffening in female subjects, estrogen was found to be protective. For example, when SU-5416 and chronic hypoxia were used to induce pulmonary hypertension, compliance decreased in ovariectomized female mice but not in mice supplemented with estrogen (122). While animal models provide data on key mechanisms involved in arterial stiffening, major gaps remain, because relatively few models have been used to study sex differences.

INTERVENTIONS FOR ARTERIAL STIFFENING

While there are no drugs indicated for counteracting arterial stiffening, many antihypertensive drugs simultaneously improve elasticity. While many studies have included women, few were powered to assess whether sex is a factor. In a cohort containing men and women, the angiotensin receptor blocker eprosartan had greater benefits in terms of arterial stiffening than the β-blocker atenolol, despite similar blood pressures (42). Similarly, in men and women with resistant hypertension, addition of the angiotensin receptor blocker valsartan to a regimen containing an angiotensin-converting enzyme inhibitor enhanced the reduction in both blood pressure and arterial stiffening (124). In contrast to these studies showing the importance of renin-angiotensin system blockade, a study of only postmenopausal women found that valsartan and the Ca²⁺ channel blocker amlodipine equally reduce arterial stiffening (86). Therefore, additional studies are needed to determine whether certain antihypertensive medications are more effective in attenuating arterial stiffening in women than men.

The impact of menopausal hormone therapy on large artery remodeling is varied. Arterial stiffness is decreased with estrogen treatment in combination with the synthetic progestins drospirenone (214) and dydrogesterone (134). The combination of estrogen and progestin has a positive impact on the carotid intima-to-media ratio in healthy women (144) and women with mild hypercholesterolemia (76), but a reduction in PWV does not accompany the improvements in carotid measurements. Similarly, treatment with estrogen alone does not improve PWV after 12 wk (9) or 4 mo (43). In contrast, administration of the phytoestrogen resveratrol for 12 wk significantly reduces CAVI in patients with type 2 diabetes (97) but not in healthy individuals (192). These mixed results indicate the need for knowledge of the estrogenic signaling pathways that attenuate arterial stiffening, so that more selective approaches can be tested.

In contrast to the controversy surrounding estrogen therapy, clear evidence indicates that testosterone improves arterial elasticity and does not increase overall cardiovascular risk (30). However, based on a relatively small number of studies that show increased cardiovascular events in patients taking this therapy, the United States Food and Drug Administration requires a warning for increased cardiovascular risk on testosterone prescriptions (213). While most studies that have assessed arterial stiffening in male and female subjects still fail to measure and report hormonal status, one recent analysis from the Multi-Ethnic Study of Atherosclerosis found that higher levels of testosterone in women and, conversely, estradiol in men are associated with greater aortic stiffness (202). More adequate assessment of the impact of aging-induced decreases in sex hormones on arterial stiffness is needed in future studies.

Exercise improves multiple measures of arterial health in humans (99) and animal models (79). Sex differences emerge when the type and intensity of exercise and its impact on arterial stiffening are considered. While cardiorespiratory fitness and physical activity are associated with better arterial elasticity in both sexes, leisure activities have adverse effects on stiffening only in men, an effect most likely resulting from the predominance of sedentary leisure in men compared with active leisure in women (19). PWV after acute exercise is higher in men than women (29, 48), but chronic aerobic exercise in men reduces arterial stiffening to a greater extent than resistance training (141, 145). In fact, resistance training acutely increases PWV to a similar extent in young men and women (104). High-intensity resistance training for 8 wk, on the other hand, improves endothelial function but does not impact PWV in young and older women (174). A combination of resistance and aerobic exercise significantly decreased arterial stiffness in postmenopausal women (62, 195), and this regimen also reduced stiffness, along with blood pressure and insulin resistance, in adolescent girls (194). Together, these studies have emphasized that even the type of exercise may need to be tailored according to sex to obtain the optimal reduction in arterial stiffness.

Perspectives

While reducing blood pressure improves health outcomes, inclusion of other vascular measurements may provide more information about the underlying vascular pathology. Additional research is needed to understand the cellular mechanisms that precede and follow changes in arterial stiffening. Integration of measurements of arterial stiffening into clinical practice will provide more opportunities to improve methodology and a platform for additional clinical studies. Moreover, the impact of sex and sex hormones on arterial stiffening should be acknowledged, with separate acceptable ranges defined for each stage of life. Since arterial stiffening arises from several etiologies (Fig. 3), assessment of familial genetic inheritance and environmental exposure should provide further insights into potential targets for interventions aimed at reversing arterial stiffening. Unresolved issues, such as the significance of assessing local PWV on specific vessels (e.g., carotid artery vs. carotid-femoral PWV), can improve risk stratification on cardiovascular disease.

Fig. 3.

Graphical abstract denoting interactions of sex, disease, and mechanisms in arterial stiffness. ER, estrogen receptor; GPER, G protein-coupled ER; AR, androgen receptor.

GRANTS

This work was supported by National Institutes of Health Grant 5-R01-HL-133619 (to S. H. Lindsey).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

B.O.O., M.A.Z., and S.H.L. interpreted results of experiments; B.O.O. and S.H.L. prepared figures; B.O.O., M.A.Z., G.L.C., C.M.A., K.M.G., K.S.M., and S.H.L. drafted manuscript; B.O.O., M.A.Z., G.L.C., C.M.A., K.M.G., K.S.M., and S.H.L. edited and revised manuscript; B.O.O., M.A.Z., K.S.M., and S.H.L. approved final version of manuscript.