Effects of aging and endurance exercise training on cardiorespiratory fitness and cardiac structure and function in healthy midlife and older women

Abstract

Cardiovascular disease (CVD) remains the leading cause of morbidity and mortality in women in developed societies. Unfavorable structural and functional adaptations within the heart and central blood vessels with sedentary aging in women can act as the substrate for the development of debilitating CVD conditions such as heart failure with preserved ejection fraction (HFpEF). The large decline in cardiorespiratory fitness, as indicated by maximal or peak oxygen uptake (V̇o_2max and V̇o_2peak, respectively), that occurs in women as they age significantly affects their health and chronic disease status, as well as the risk of cardiovascular and all-cause mortality. Midlife and older women who have performed structured endurance exercise training for several years or decades of their adult lives exhibit a V̇o_2max and cardiac and vascular structure and function that are on par or even superior to much younger sedentary women. Therefore, regular endurance exercise training appears to be an effective preventative strategy for mitigating the adverse physiological cardiovascular adaptations associated with sedentary aging in women. Herein, we narratively describe the aging and short- and long-term endurance exercise training adaptations in V̇o_2max, cardiac structure, and left ventricular systolic and diastolic function at rest and exercise in midlife and older women. The role of circulating estrogens on cardiac structure and function is described for consideration in the timing of exercise interventions to maximize beneficial adaptations. Current research gaps and potential areas for future investigation to advance our understanding in this critical knowledge area are highlighted.

INTRODUCTION

Cardiovascular disease (CVD) is a leading cause of morbidity and mortality in women in developed nations (1). Ventricular hypertrophy, cardiac and central arterial stiffening, impairments in ventricular relaxation and diastolic suction, and vascular endothelial dysfunction are reported with sedentary but otherwise healthy aging (i.e., the absence of chronic disease) in women (2–8). These age-associated structural and functional adaptations likely set the stage for the development of CVD conditions such as arterial hypertension, coronary heart disease, stroke, and heart failure with preserved ejection fraction (HFpEF) (5, 9–11).

Cardiorespiratory fitness as indicated by maximal or peak oxygen uptake (V̇o_2max and V̇o_2peak, respectively) declines with healthy aging in women (12–15). Numerous population-based epidemiological studies provide compelling evidence of an inverse relationship between cardiorespiratory fitness and cardiovascular-related and all-cause disability and death in healthy women and those with suspected or known CVD, traditional risk factors, or multiple comorbidities (16–23).

Midlife (45–59 yr) and older (≥60 yr) women who have engaged in habitual endurance exercise exhibit a substantially higher V̇o_2max and enhanced cardiovascular function during submaximal and maximal exercise than age-similar, sedentary peers, or even much younger women (4, 24–31). Furthermore, midlife and older endurance-trained women demonstrate lower cardiac and central arterial stiffness, enhanced vasodilatory function and vascular-arterial coupling, and less traditional CVD risk factors, coronary artery plaque prevalence, and calcified plaque volume than their physically inactive but otherwise healthy counterparts (4, 32–36). Together, these observations suggest that endurance-trained midlife and older women provide a unique model of cardiovascular aging (37).

There is evidence of a sex-specific dichotomy in the cardiovascular adaptations to healthy aging and regular endurance exercise training, particularly when training is initiated at an older age (38–50). Thus, a detailed characterization of healthy aging and endurance exercise training-related responses may provide valuable knowledge that could aid in designing evidence-based exercise strategies to enhance health and clinical outcomes in aging women. Moreover, establishing the physiological processes that underpin the key domains of maximal oxygen uptake, cardiac structure, and function at rest and during exercise in endurance-trained and sedentary midlife and older women may provide valuable insight into CVDs in which these factors can be profoundly abnormal, as in the case of HFpEF (51–54).

This narrative review will present previous and current research on V̇o_2max or V̇o_2peak, left ventricular (LV) structure and function at rest and during exercise in midlife and older women who have engaged in regular vigorous endurance exercise training for several years to decades. This review will not focus extensively on the structural and functional vascular adaptations associated with healthy aging and regular endurance exercise training because these topics have been thoroughly reviewed elsewhere (40, 55, 56). To characterize the effects of regular endurance exercise training on V̇o_2max or V̇o_2peak and cardiovascular structure and function in women comprehensively, we will also provide an overview of the influence of healthy aging and relatively short- and long-term (2–12 mo) exercise training on these physiological parameters.

MAXIMAL EXERCISE CAPACITY WITH AGING AND EXERCISE TRAINING

Overview of Maximal and Peak Oxygen Uptake (V̇o_2max or V̇o_2peak)

V̇o_2max reflects the integrative capacity of numerous organ systems (i.e., cardiovascular, respiratory, nervous, and musculoskeletal) to deliver and utilize oxygen during maximal or fatiguing physical exertion (57, 58). In healthy and clinical populations, V̇o_2max is the gold standard index of cardiorespiratory fitness and is commonly assessed during large muscle mass exercises such as walking, running, and cycling. V̇o_2max can be measured directly by assessing respiratory gas volumes and concentrations or estimated using validated equations based on work rate, heart rate, or total exercise time (59).

Absolute V̇o_2max or V̇o_2peak (L·min⁻¹ or mL·min⁻¹), with the latter being the highest epoch of V̇o₂ achieved at peak exercise, is scaled relative to total body mass (mL·kg⁻¹·min⁻¹) to compare populations of different body sizes. This approach is particularly relevant when individuals carry their body mass during physical exertion (i.e., walking or running uphill) (60). Researchers have also scaled V̇o_2max relative to total body or appendicular fat-free mass (mL·kgFFM⁻¹·min⁻¹) to minimize population differences in body composition. This latter approach allows for a more detailed examination of the effects of obesity, aging, chronic disease, sex, and exercise training on V̇o_2max or V̇o_2peak (27, 33, 61–64).

Irrespective of training status, V̇o_2max is 10–20% lower at a similar chronological age in women compared with men (13, 27, 65). Scaling V̇o_2max relative to fat-free mass greatly reduces but does not eliminate sex-related differences (13, 27). Smaller cardiac dimensions and a lower hematocrit and hemoglobin level result in a reduced systemic O₂ transport capacity in women, which largely explains these differences in V̇o_2max between the sexes (27, 58, 66, 67).

There is robust evidence from high-quality, large-scale studies supporting cardiorespiratory fitness as a biomarker of cardiovascular and systemic health status in adults of all ages (59, 68). Epidemiological and clinical studies have shown an inverse, independent, and graded association between V̇o_2max and the risk of cardiovascular and all-cause mortality in women across ages and clinical status (16, 18, 21, 69). There is a twofold increase in mortality for women in the lowest 20% for treadmill exercise capacity or those who fail to reach 85% of the age-predicted estimates for V̇o_2max (21, 69). In a series of studies as part of the Cooper Center Longitudinal Study (CCLS), an estimated maximal exercise capacity expressed in metabolic equivalents (METs) obtained during midlife predicted lifetime CVD mortality and overall health care costs later in life (>65 yr), reinforcing the clinical importance of preserving exercise capacity throughout adulthood (70–72). Based on its significant prognostic and diagnostic capabilities, recent scientific statements have emphasized cardiorespiratory fitness or other measures of exercise capacity as a vital health sign and their inclusion within a clinical assessment for general health and physical function (59, 68, 73).

Aging Effects

Cross-sectional reports indicate that V̇o_2max relative to total body mass declines 4–10% per decade in healthy women of increasing age (12, 14, 15, 30). However, a couple of longitudinal examinations have shown that V̇o_2peak does not decline linearly but accelerates with each successive decade of adulthood. For example, data from the Baltimore Longitudinal Study of Aging (BLSA) in a cohort of 375 highly screened women aged in their 20s to 80s showed that the mean per-decade decline in V̇o_2peak scaled to total body mass expressed as relative percentage change increases >3 fold, going from a −6.4% change in young adulthood (<30 yr) to −21.1% in older age (≥70 yr) (13). Supporting these findings, both the Aerobics Center Longitudinal Study (ACLS) of >3,400 women studied over 30 yr and the HUNT study of >750 women examined over 10 yr both reported an accelerated per-decade reduction in V̇o_2peak between 20 and 80 yr, although the effect in those >70 yr differed between the two studies, plateauing in one and accelerating in the other (65, 74).

In addition to nonmodifiable physiological adaptations (i.e., a reduction in maximal heart rate), the development of common cardiovascular and noncardiovascular chronic medical conditions and modifiable lifestyle factors such as physical inactivity that commonly occur with aging can accelerate the reduction in V̇o_2max or other indices of cardiorespiratory fitness (75). In terms of physical activity levels, self-reported and directly assessed physical activity levels in women decline over adulthood, resulting in many midlife and older women failing to reach the minimum recommendations for weekly physical activity (76–79). Moreover, the total amount of moderate and vigorous-intensity physical activity is greatly diminished in older versus younger adults, with older women preferring low-intensity physical activity (80, 81). Although low-intensity exercise can infer cardiovascular health benefits and a reduction in the risk of CVD morbidity and mortality, this physiological stimulus may not be sufficient to preserve V̇o_2max with aging.

Exercise Training Effects

In midlife and older endurance-trained women, V̇o_2max relative to total body mass is substantially higher (32–83%) compared with their physically inactive peers and is more comparable with healthy untrained young adults (24, 26–28, 31, 33, 82). A substantial difference (22–52%) remains between endurance-trained and untrained midlife and older women when V̇o_2max is scaled relative to fat-free mass (26, 27, 31, 33, 82). These latter findings stress two important concepts concerning oxygen delivery and utilization with healthy aging and endurance exercise training in women. First, they indicate the remarkable capacity for oxygen delivery and utilization during vigorous exercise in endurance-trained midlife and older women. Second, they highlight the importance of regularly engaging in endurance exercise training with aging to preserve and enhance skeletal muscle mass and capillary and mitochondrial density to maximize oxygen extraction and utilization in active peripheral tissue during exercise (31, 83, 84).

Although there is an inevitable loss of V̇o_2max over a lifetime despite maintaining a large exercise volume, multiple meta-analyses have shown that V̇o_2max is significantly higher in habitually endurance-trained women across all ages (30, 85–88) (Fig. 1A). Thus, regular engagement in endurance exercise training should be considered a cornerstone of any strategy to mitigate the unfavorable age-associated adaptations of O₂ delivery and extraction pathways in women.

Figure 1.

A: even in the absence of chronic disease, V̇o_2max progressively declines with aging (blue-shaded area). Habitual endurance exercise that is consistent with or above current physical activity recommendations is associated with a greater V̇o_2max across the lifespan (red-shaded area). Endurance exercise training started late in life results in an increase in V̇o_2max despite only limited left ventricular (LV) structural adaptations (hatched area). B: the stroke volume and stroke index responses for submaximal and maximal exercise are shifted upward in lifelong (>20 yr) endurance-trained women compared with untrained age-similar controls. The age-related effect on exercise stroke volume has yet to be clearly delineated. Data extracted from Carrick-Ranson et al. (33). C: noninvasive indices of LV relaxation and diastolic suction are not significantly enhanced in lifelong endurance-trained vs. untrained age-similar controls across multiple LV loading conditions (Vp, propagation velocity displayed, unpublished data). D: LV mass index is significantly larger in those who have performed lifelong endurance exercise compared with untrained age-similar controls. Data extracted from Carrick-Ranson et al. (33). E: lifelong endurance-trained women demonstrate a rightward shift and a flatter diastolic p-v curve, indicating increased LV chamber compliance and distensibility compared with untrained age-similar controls. Data extracted from Carrick-Ranson et al. (33). Gray symbols and lines represent young untrained women, blue symbols and lines represent midlife and older untrained women, redline symbols and lines represent lifelong endurance-trained women.

V̇o_2max represents one of the most responsive physiological parameters to physical training in adults of all ages (89, 90). Although there is large interindividual variability in the response of V̇o_2max with regular training, programs lasting several months increase V̇o_2max in healthy midlife and older women to a similar relative level to that reported in younger women (Table 1) (41, 46, 89, 91, 102–108). Multiple months of regular endurance exercise training in healthy and frail women in their eighth and ninth decades suggest that aging and frailty do not prevent the ability to confer measurable improvements in V̇o_2max with the application of a sufficient training stimulus (97, 109).

Table 1.

Physiological adaptations to exercise training in women based on the period that endurance exercise training is initiated

Parameter	Initiated at Midlife	Initiated at Older Age
Cardiorespiratory fitness
Exercise capacity (V̇o2max; V̇o2peak; peak METs, total exercise time)	↑ (91–96)	↑ (41, 46, 97–100)
LV function during exercise
Maximal exercise cardiac output	↑ (92, 93)	↔ (41, 46, 97, 99)
Maximal exercise stroke volume	↑ (92, 93)	↔ (41, 46, 97, 99)
Maximal exercise v-a coupling	ND	ND
Submaximal exercise stroke volume	↑ (absolute workload) (91, 92) ↑ (low intensity), ↔ (moderate intensity) (relative workload) (101)	↔ (absolute workload) (46) ↑ (relative workload) (41)
Submaximal exercise heart rate	↓ (absolute workload) (91–93) ↔ (relative workload) (93, 101)	↓ (absolute workload) (46) ↓ (relative workload) (41)
Submaximal exercise v-a coupling	ND	ND
Cardiac structure and function
LV mass	↑ (94)	↔ (98)
LV cavity dimensions	↑ (94)	↔ (98, 100)
LV chamber stiffness	↓ (95)	ND
Resting LV relaxation	↔ (94)	ND

↑, increased; ↓, decreased; ↔, no change; LV, left ventricular; ND, no data in women only cohorts; V̇o_2max, maximal oxygen uptake; V̇o_2peak, peak oxygen uptake; v-a coupling, ventricular-arterial coupling.

CARDIAC STRUCTURE AND FUNCTION AT REST AND EXERCISE WITH AGING AND EXERCISE TRAINING

To preface the discussion on the effects of aging and endurance exercise training on exercising cardiac function in women, the responses to acute exercise in healthy individuals are presented herein.

LV FUNCTION AND HEART RATE RESPONSE DURING ACUTE INCREMENTAL EXERCISE

Based on the Fick equation, V̇o₂ during exercise is determined by O₂ delivery (cardiac output, Q̇) and utilization by working tissue [systemic oxygen content difference, (a-v) DO₂]. Increases in heart rate and stroke volume are precisely regulated to ensure that Q̇ is tightly coupled to the metabolic demand of working skeletal muscle, which receives a sizable proportion (∼70–80%) of Q̇ during maximal exercise (110–113). Irrespective of age, sex, or physical training status, the slope of the relationship between systemic Q̇ and V̇o₂ is consistent above resting V̇o₂ levels (ΔQ̇/ΔV̇o₂: 5–6.0 L·L⁻¹) (33, 61, 63, 114). The Q̇ achieved at maximal exercise effort represents the primary determinant of an individual’s exercise capacity (57, 58, 113, 115).

Heart rate increases directly with exercise intensity up to a maximal rate, with aging, genetics, and the presence of CVD influencing the maximal value achieved (112, 113, 115–117). Experimental pharmacological blockade studies have shown that the initial increase in heart rate during low-intensity exercise is predominately due to vagal withdrawal. Above ∼100 beats/min, increases in heart rate are more influenced by increasing sympathetic activity (118, 119). After the cessation of exercise, the initial rapid reduction in heart rate is due to the restoration of vagal tone, whereas longer recovery durations reflect the balance between vagal tone restoration and sympathetic tone withdrawal. Because of its clear association with vagal activity, heart rate recovery after exercise may be used as a measure of cardiovascular regulation and health (120).

When examining the exercise heart rate response at the same absolute workload (i.e., watts, V̇o₂ relative to total body mass), the absolute or relative change and the value achieved can vary with aging, sex, maximal exercise capacity (i.e., V̇o_2max or V̇o_2peak), and chronic disease status; however, at the same relative workload, namely, the percentage of V̇o_2max (%), the response is essentially superimposable in groups varying in age, sex, and V̇o_2max levels (Fig. 3B in Ref. 61 and Fig. 1 in Ref. 63).

Although the absolute levels and magnitudes of change differ due to variations in venous return, stroke volume increases from rest to maximal exercise in both the supine and upright postures in women (121–126). The larger stroke volume during supine and upright exercise is mediated through an increase in LV end-diastolic volume (LVEDV) and a reduction in LV end-systolic volume (LVESV), which results in an overall increase in LV ejection fraction (LVEF) (121, 122, 124, 126). Increases in LVEDV predominately occur at low-to-moderate intensities of submaximal exercise to facilitate the recruitment of the Frank–Starling mechanism, whereas reductions in LVESV are more pronounced at higher intensities of submaximal and maximal exercise (121).

A recent prospective study in a population of women varying in age, physical activity levels, V̇o_2peak, and clinical disease status showed that women who were in the highest quartile for LVEDV (i.e., had the largest LVEDV) at supine rest exhibited the greatest ability to augment stroke volume from supine rest to maximal exercise (122). Conversely, those in the lowest quartile for resting supine LVEDV demonstrated the smallest augmentation in stroke volume during exercise.

Systolic blood pressure increases proportionally with intensity during incremental upright cycling or treadmill exercise, whereas diastolic blood pressure stays the same or decreases slightly in women (39, 127, 128). Akin to resting values, systolic blood pressure, diastolic blood pressure, and mean arterial pressure at maximal upright cycle or treadmill exercise are increased with age in women (39, 124, 126, 127, 129). In 47 rigorously screened middle-aged women (mean age ∼53 yr), higher systolic blood pressure at maximal exercise was associated with a greater degree of cardiac and central arterial stiffness, which suggests the systolic blood pressure at maximal exercise may serve as a noninvasive indicator of early pathophysiological changes in cardiovascular structure and function in middle-aged women (128).

According to the Frank–Starling mechanism, the pressure that fills the ventricle during diastole influences stroke volume over a wide range of LV filling pressures (130). LV filling pressure during exercise would be influenced by both intracardiac (ventricular geometry, chamber stiffness, amount of relaxation, and diastolic suction) and extracardiac (pericardium and pulmonary constraint, total blood volume, arterial stiffness) factors. The magnitude of the rise in LV filling pressure carries important physiological information about the functioning of the LV during large changes in venous return. Beneficial adaptations in intra- and extracardiac factors allow endurance-trained individuals to maximize the Frank–Starling mechanism to produce a large stroke volume while minimizing the rise in LV diastolic filling pressure during submaximal and maximal exercise (58, 131–136). Conversely, severe abnormalities in intra- and extracardiac factors result in a large increase in LV filling pressures despite a profoundly impaired LVEDV and stroke volume response, as is the case with HFpEF (52, 54, 137–140). In later sections, we will address the important structural and functional cardiac properties that can enhance the Frank–Starling mechanism during exercise in endurance-trained midlife and older women.

CHRONOTROPIC RESPONSES AND LV FUNCTION AT REST AND DURING EXERCISE WITH AGING AND EXERCISE TRAINING

Aging Effects

Although findings may vary among studies due to differences in participant characteristics and measurement techniques, findings from the BLSA study indicate that supine resting Q̇ and stroke volume are not significantly affected by age (126).

Cross-sectional studies have shown that maximal Q̇ declines with healthy aging in women without a significant change in the relationship between Q̇ and V̇o₂ (27, 33, 41, 47, 62–64, 141–145). The BLSA project showed that maximal Q̇ relative to BSA (L·m⁻²·min⁻¹) is ∼15–20% smaller in women aged in the eighth decade of life compared with the third decade of life (126). This finding is similar in magnitude to some studies; however, others have reported much larger changes in women with aging (27, 41, 47, 62, 144). Interestingly, evidence from cross-sectional studies indicates that the absolute and relative reduction in maximal Q̇ and stroke volume with healthy aging is smaller in women than men, potentially suggesting better preservation of LV function during vigorous exercise (47, 146). Additional prospective longitudinal studies are required to confirm this contention.

Cross-sectional and longitudinal examinations and meta-analyses indicate that maximal heart rate declines with aging in men and women (27, 41, 116, 126, 146, 147). In men, the decline in maximal heart rate with healthy aging results largely from a reduction in intrinsic heart rate, although a decrease in β-adrenergic responsiveness does contribute (148, 149). Based on the previous findings of sex differences in cardiac autonomic regulation with aging at rest and exercise, women-focused examinations to elucidate whether there may be potential implications for exercise function and postexercise recovery at different stages of adulthood are needed (150–152).

The age-related changes in the maximal stroke volume in women are not well defined due to the heterogeneous findings among studies (27, 33, 41, 47, 62, 91, 126, 142, 143, 145, 146, 153). The small amount of available data from studies that have investigated the volumetric components of stroke volume has shown that LVEDV is unaltered or larger, whereas LVESV is larger with age at peak supine or upright exercise (124, 126, 153).

In women with small, stiff, and slow-relaxing hearts, which may be common in older women, there may be a reduced ability to increase cardiac filling volume and stroke volume during exercise without provoking significant rises in LV filling pressure (154). Two recent cross-sectional studies have shown that pulmonary capillary wedge pressure (PCWP) is elevated during submaximal and maximal exercise in older versus younger mixed-sex cohorts (155, 156). The characterization of the effects of aging on filling and emptying volumes and ventricular and systemic pressures in healthy women would increase the current understanding of LV function and regulation in aging women during acute exercise (101). From these data, chronic disease-related changes in these parameters could be investigated and clarified.

Exercise Training Effects

Resting heart rate is significantly lower in endurance-trained midlife and older women irrespective of posture (e.g., supine, seated upright, or orthostatic) (27, 33). Resting Q̇ is not changed after endurance exercise training, as resting stroke volume is increased whereas heart rate is reduced (27, 33).

Because oxygen demand for an absolute submaximal exercise workload is not affected by training status, Q̇ is not significantly different between habitually endurance-trained and nonendurance-trained individuals. However, heart rate is substantially lower at any given absolute workload in habitually endurance-trained midlife and older women compared with their untrained counterparts (27, 33). This lower heart rate response is secondary to a larger stroke volume in habitually endurance-trained women but may also involve additional autonomic and nonautonomic factors (157–159).

A large Q̇ during vigorous exercise is a hallmark characteristic of habitually endurance-trained midlife and older women. Ogawa et al. (27) reported a ∼20% larger maximal Q̇ during treadmill exercise in endurance-trained women aged >50 yr than in slightly older untrained controls. Several subsequent cross-sectional investigations using similar measurement techniques have confirmed these findings, reinforcing the superior LV exercise function in endurance-trained women (24–26, 33, 61).

A larger stroke volume is exclusively responsible for the greater Q̇ during maximal exercise in habitually endurance-trained women (24–27, 33, 61) (Fig. 1B). Although no studies have directly examined diastolic and systolic volumes during exercise in healthy midlife and older women, there is supporting evidence from cross-sectional studies for enhanced LV diastolic filling in endurance-trained women (29, 33, 160–162). The intra- and extracardiac factors that would facilitate a larger stroke volume will be discussed in more depth in the following sections.

Many population-based studies have shown that a lower heart rate at rest is strongly and independently associated with a significantly lower risk for CVD-related and all-cause mortality (163). One study showed that exercise training for 9–12 mo results in a lowering of the resting heart rate by an average of 9 beats/min in previously sedentary midlife and older women (46). However, others have reported a more modest change (0–3 beats/min) with a shorter duration of exercise training (8–12 wk) (91, 164, 165).

In contrast to lifelong or habitual endurance exercise findings, a higher V̇o_2max is not consistently associated with a larger maximal or peak exercise stroke volume, or Q̇ in previously untrained midlife and older women (Table 1). Maximal stroke volume was unchanged after 9–12 mo of training (pre: 70 ± 2, post: 70 ± 2 mL), whereas only a very modest 3 mL increase was reported (pre: 106.2 ± 14.6, post: 109.5 ± 16.6 mL) after 12 wk of training (41, 46). However, in a younger cohort of postmenopausal women (mean age 55 yr), a ∼9 mL increase in maximal stroke volume was reported after a 20-wk training program (92).

As described for habitually endurance-trained women, a reduction in heart rate and an increase in stroke volume during steady-state submaximal exercise are recognized as hallmark adaptations to endurance exercise training (166). In women aged >50 yr, findings supporting a larger submaximal exercise stroke volume postintervention are equivocal. Murias et al. (41) reported an 8 mL larger stroke volume (pre: 98.4 ± 17.4, post: 106.1 ± 22.7 mL) at a similar absolute workload with 12 wk of endurance exercise in older women (69 ± 7 yr). Kilbom and Astrand (91), Wilmore and coworkers (93), and Green et al. (92) reported a 5–9 mL increase in stroke volume during submaximal exercise workloads after 7–20 wk of exercise training in women aged in their 50s and early 60s. However, Spina et al. (46) reported only a trivial increase in stroke volume at ∼50% of pre-and-post-V̇o_2max (pre: 83 ± 4, post: 84 ± 3 mL) after 9–12 mo of endurance exercise in women aged over 60 yr (Table 1). Based on these findings, it could be concluded that more sizable effects on exercise stroke volume are more likely to occur when exercise is initiated early in midlife compared with at an older age.

Based on several observations of modest changes in LV dimensions, diastolic filling rates, and β-adrenergic responsiveness with exercise training, some research groups have asserted that cardiac adaptability to endurance exercise training is diminished with advanced aging in previously untrained women (45, 98, 102). Although the mechanistic pathways are poorly understood, changes in endogenous sex hormones with menopause may impair cellular and hormonal processes involved in the exercise training-related adaptations of cardiac structure (34, 167, 168). Although hormone-replacement studies in postmenopausal women have generally produced negative results, this area still needs further research in humans as the timing and route of administration for hormone replacement may have important effects on cardiac structure and function outcomes (167, 168).

VENTRICULAR-ARTERIAL COUPLING AT REST AND DURING EXERCISE WITH AGING AND EXERCISE TRAINING

Overview of Ventricular-Arterial Coupling

The dynamic interaction of the left ventricle and the arterial system [ventricular-arterial coupling (v-a coupling)] represents a key determinant of ventricular performance and energetics at rest and exercise (169, 170). Analysis of v-a coupling allows for the description of ventricular properties and the modulation of the elastance and resistance of the arterial system at rest and with exercise.

The concept of v-a coupling is defined by the ratio of arterial elastance (Ea) to ventricular systolic elastance (Ees, sometimes denoted Elv in the literature). Ea represents an integrative index of the total arterial load imposed on the LV from the entire arterial tree. Ea incorporates the properties of systemic arterial elastance, peripheral vascular resistance, characteristic impedance, and systolic and diastolic time intervals (171–174). Ees is an index of LV systolic performance derived from inotropic efficiency, as indicated by the slope of the end-systolic volume/pressure relationship, that also incorporates functional, structural, and geometric characteristics of the LV (175, 176). Optimal v-a coupling occurs when there is an appropriate matching of Ea and Ees, so dynamic adjustments in stroke volume can occur in response to potentially large changes in venous return without marked fluctuations in arterial pressure (177, 178).

Ea and Ees are modified from rest to exercise conditions to ensure sufficient flow and pressure to support oxygen delivery to metabolically active tissue. However, even with “normal” LV function, there appears to be a degree of uncoupling during exercise. Although Ees is consistently shown to increase during exercise, the Ea response has been less consistent across studies (33, 61, 169, 171, 177, 179–185). The inconsistent results reported for Ea may be partially explained by methodological variations among previous studies, with differences in subject age, sex, health and disease status, exercise capacity, as well as exercise intensity, all having the potential to significantly influence central and peripheral hemodynamics and heart rate during exercise.

Aging Effects

A cross-sectional examination of >2,000 people reported that noninvasive measures of resting Ea, Ees, and ventricular diastolic (Ed) elastance are higher (i.e., increased stiffness or reduced compliance) in healthy women aged >45 yr compared with those <45 (50). The tandem elevation of Ea and Ees results in the preservation of the resting Ea/Ees ratio with aging (50, 186). More recently, data from 129 (43 women, 86 men) people aged 21–96 studied longitudinally over an average period of 12 yr (range 2–29 yr) as part of the BLSA cohort showed that both Ea and Ees increased with aging. However, Ea increased to a greater degree than Ees, with a notable increase in Ea after the age of 50 yr (187). Although both sexes experience a similar degree of increase in Ea, women experience a smaller change in Ees compared with men (187). This latter finding suggests that the age-associated changes in ventricular and arterial elastance may differ between the sexes, leading to differences in ventricular performance with advanced age.

Possible explanations for the increase in arterial elastance (i.e., increased stiffness or reduced compliance) with aging include structural and functional adaptations of the arterial tree, such as changes in vascular diameter, wall thickness and stiffness, and endothelial dysfunction (64, 188, 189). A reduction in myocyte number with concomitant hypertrophy of the residual myocytes and accumulation of extracellular matrix collagen are putative mechanisms for the increase in ventricular elastance with aging (5, 190–192). Animal studies have shown that estrogen can influence the signaling and molecular pathways involved in these physiological processes (193).

Arterial and ventricular elastance are increased during submaximal and maximal exercise in older versus younger normotensive women. Najjar et al. showed that the Ea/Ees ratio was higher and changed less from rest to submaximal and maximal exercise in healthy women >60 yr compared with those <40 yr. This increased Ea/Ees ratio with age resulted from a much greater increase in Ea than Ees. This v-a coupling response in older women was associated with a smaller stroke volume, higher systolic blood pressure, and a blunted and overall lower LVEF during exercise (169). Others have also reported an age-related increase in Ea at low, moderate, and maximal-intensity exercise in women aged >60 yr, compared with young and middle-aged women (33). This response in Ea in older women was associated with a smaller exercise reserve for stroke volume, systemic arterial elastance, and systemic vascular resistance. Collectively, these findings highlight that age-associated modifications in Ea and Ees alter v-a coupling during exercise, ultimately resulting in a less efficient stroke volume response during exercise.

Acute administration of verapamil positively influenced Ea, systemic vascular resistance, and stroke volume during exercise in elderly individuals and extended submaximal but not maximal exercise tolerance (194). Likewise, sodium nitroprusside lowered stroke work during exercise but did not improve maximal exercise capacity in healthy older adults (195). Thus, these findings provide preliminary evidence that v-a coupling may influence stroke volume regulation during exercise and potentially submaximal exercise tolerance in healthy aging adults.

Exercise Training Effects

The aging-related increase in resting Ea is absent in men and women who have performed lifelong exercise training (2, 62, 172, 196). In addition to a lower resting heart rate, parameters indicative of the functional and structural properties of the large elastic arterial vessels are better in lifelong exercisers. For example, Shibata and Levine (197) showed that central pulse wave velocity, carotid β-stiffness index, and biological aortic age were consistently lower in a small group (6 women, 5 men; mean age ∼71 yr) of high-performing lifelong exercisers compared with age-similar sedentary but overall healthy counterparts (4 women, 6 men; mean age ∼68 yr).

Enhanced ventricular and central arterial compliance and lower systemic resistance would improve v-a coupling during submaximal and maximal exercise in endurance-trained adults (8 women, 17 men) aged >60 versus untrained controls (10 women, 15 men) (61). Supporting this contention, the change in Ea is smaller, and the overall value is lower during moderate and vigorous-intensity exercise in women who have performed lifelong endurance exercise compared with untrained age-similar and younger women (33).

In nine (3 women, 6 men) previously sedentary older men and women, 1 yr of progressive and intensive endurance exercise training improved Ea at rest and during maximal exercise (182). A follow-up study in the same cohort of seniors showed that supine rest heart rate, Ea, systemic arterial compliance, and vascular resistance were improved post-training, whereas biological aortic age, which conceptually reflects the structural changes of the aortic vessel wall, was not significantly changed (172). When stroke volume was restored to pre-exercise training levels using lower body negative pressure, training-related improvements in Ea, systemic arterial compliance, and vascular resistance were eliminated. These findings suggest that improvements in Ea post-training in previously sedentary older adults are due to changes in the functional modulation of LV afterload secondary to an increase in stroke volume rather than structural changes within the walls of the central blood vessels (172). Stroke volume expansion can decrease sympathetic activity, which lowers peripheral vascular resistance and increases systemic arterial compliance by reducing smooth muscle tone (198). Shear stress from a high blood flow rate increases endothelial nitric oxide production and lowers smooth muscle tone (199). Together, these findings suggest that endurance exercise training-related adaptations in stroke volume and the central and peripheral arterial vasculature enhance v-a coupling during exercise in previously untrained older adults.

LV SYSTOLIC PERFORMANCE WITH AGING AND EXERCISE TRAINING

Aging Effects

Despite its limitations, a large body of evidence supports the clinical utility of LVEF as a parameter of LV systolic function (200). In a large population-based study, resting LVEF was higher in women than men across adulthood (201). This higher resting LVEF in women occurs independently of changes in LVEDV and stroke volume (202). In >1,100 individuals aged between 67 and 89 including those with and without chronic diseases (e.g., diabetes mellitus, CVD, hyperlipidemia, and obesity), women demonstrated greater global longitudinal and circumferential strain and torsion at resting conditions than men (203). These findings highlight that sex-related differences in resting LV systolic function that occur with healthy aging must be carefully considered when interpreting any training-related adaptations.

At maximal exercise, older adults’ LVEF is lower than that of young adults (124, 126, 169, 195, 204–206). Several studies investigating sex and age-related differences have highlighted that older women exhibit a smaller LVEF reserve (Δ from rest to maximal effort) during supine and upright exercise than age-similar men due to a smaller reduction in ESV (123, 124, 126, 169). A smaller Ea/Ees ratio reserve (Δ seated upright rest to upright maximal effort) in older women suggests suboptimal v-a coupling in older women (169). Factors that may contribute to the smaller change in LVEF in women during exercise may include a reduced intrinsic cardiac contractility, a blunted arterial vasodilator capacity leading to increased cardiac afterload and reduced β-adrenergic responsiveness (98, 169, 184).

Exercise Training Effects

Pre-recruitable stroke work across a wide range of LV filling pressures is not significantly different in endurance-trained and sedentary older adults, suggesting an equivalent contractile function (2, 33, 207). Moreover, LVEF at maximal exercise is unaltered in midlife and older men who have performed high levels of habitual vigorous endurance exercise training for 4–42 yr (161, 208). However, others have shown that resting global longitudinal systolic strain is preserved in a mixed-sex cohort of lifelong endurance exercisers compared with young sedentary controls (209). This enhanced LV strain response was related to a larger LVEDV, highlighting the preload dependence of measures of LV systolic function (209). However, it is unclear whether sex modulated the effects of lifelong exercise on longitudinal systolic strain measurements, as the sample size was too small for such an analysis.

At maximal supine exercise, LVEF increased on average by 3–8% with 6–8 mo of endurance exercise training in older men (161, 204). Another study reported that indices of LV systolic and contractile function during maximal exercise improved in previously sedentary older men after 12 mo of endurance exercise training (210). Moreover, LV contractile function is higher with incremental doses of β-adrenergic stimulation (isoproterenol) in older men after performing endurance exercise training for 9 mo (211). However, neither LVEF nor contractile function, as indicated by the LVESV-systolic blood pressure relationship, was enhanced with short-term (3 mo) or longer-term exercise training (9–12 mo) in older women, which suggests a lack of β-adrenergic-mediated change in LV systolic and contractile function in women (99, 168). Although the mechanisms that underpin this lack of adaptation remain unresolved, cardiac tissue sensitivity to catecholaminergic stimuli with both aging and exercise training may differ between men and women (45, 46, 98, 99, 168, 211).

FACTORS THAT INFLUENCE LV DIASTOLIC FILLING DURING EXERCISE IN ENDURANCE-TRAINED MIDLIFE AND OLDER WOMEN

Although not directly tested to date, the increased stroke volume during exercise in endurance-trained midlife and older women likely results from enhanced diastolic filling (i.e., a larger LVEDV). This section will focus on several cardiac properties that may enhance diastolic filling during exercise in endurance-trained women. Where possible, evidence from studies directly investigating endurance-trained midlife and older women will be presented.

LV MORPHOLOGY AND MASS WITH AGING AND EXERCISE TRAINING

Aging Effects

Autopsy data suggest that cardiac mass is preserved with aging in women in their second to ninth decade of adulthood (48). Likewise, cross-sectional imaging studies in individuals without overt heart disease have reported that LV mass in absolute levels or relative to BSA remains relatively stable or increases slightly with advancing age in women (47, 49, 50, 201, 212–214). However, with aging, LV wall thickness increases whereas chamber size (or volume) decreases, resulting in a concentric remodeling pattern (50, 201, 213–216).

Exercise Training Effects

Endurance exercise delivers an acute volume load to the heart, which, when performed repeatedly, results in the enlargement of LV mass and ventricular dimensions that markedly improve an individual’s ability to perform physical work (217–219). Possibly due to the difficulties of recruiting a suitable cohort of highly trained participants, investigations of cardiac morphology in midlife and older women are rare compared with the copious amount of data obtained from examining older men and younger men and women (217–219). One cross-sectional examination of postmenopausal former competitive Swedish elite endurance athletes and healthy controls showed that LV cavity dimensions and volume relative to BSA were ∼12% and ∼17% larger in endurance-trained women (220). Nearly all the ex-athletes in this research were still performing regular recreational training between 2 and 10 h weekly, which, if intensive enough, may have been a sufficient hemodynamic stimulus to preserve the cardiac size and volumes in these women.

Two recent investigations have reported a large LV mass relative to BSA (g·m⁻²) in postmenopausal amateur marathon runners and within a cohort of competitive women athletes competing in endurance, sprint, and strength events (221, 222). However, the lack of a healthy untrained control group in either study limits the interpretation and translation of these findings.

It was recently reported that women aged ≥60 yr who had regularly performed at least 4 weekly exercise sessions of at least 30 min per session for >25 yr had a larger LV mass and LVEDV relative to BSA than age-similar sedentary women (33) (Fig. 1D). Moreover, the ratio of LV mass to LVEDV was not different between the two groups, suggesting a balanced increase in cardiac mass and volume with endurance exercise training. These findings are consistent with eccentric remodeling (physiological hypertrophy), the hallmark pattern of the “athlete’s heart” in high-performing endurance-trained women (219). Although our results originate from a cross-sectional examination, which does not permit a causal relationship between training history and physiological adaptation to be established, a robust hemodynamic stimulus from multiple years and decades of regular vigorous exercise likely stimulated physiological hypertrophy signaling pathways in endurance-trained midlife and older women.

In previously sedentary women aged in their 60s and 70s, exercise training for 3–11 mo failed to increase LV mass or chamber size (Table 1) (98, 100, 102). It is unclear why regular and vigorous exercise performed over multiple months did not increase these parameters in older women. Since the hormones associated with menopause can influence LV remodeling (discussed in effects of estrogen on lv morphology and mass), it is plausible that sex hormone levels were insufficient when regular exercise was initiated (38, 223). Although there was a variable response, LV chamber size was significantly increased after 8 mo of endurance exercise training involving three sessions weekly in postmenopausal women (>5 yr since menopause) (224). Likewise, a more recent study showed that after 12 wk of high-intensity exercise training, LV mass and chamber diameter were significantly increased in both premenopausal and recently postmenopausal women (>12 mo since last menstrual period) (94). Thus, it is plausible that the estrogen signaling pathways for physiological LV hypertrophy may still be functional shortly after early menopause and gradually downregulate with time.

An exercise timing hypothesis has recently been put forth in support of this claim, based on a recent clinical study that showed hormone replacement therapy had more favorable effects on the progression of atherosclerotic plaques in early postmenopause (6 yr since the last menstrual period) compared with late postmenopause (10 yr since last menstrual period) (225, 226). Divergent findings on central and peripheral vascular adaptations to exercise training in midlife and older women were used to provide the rationale for the exercise time hypothesis. This hypothesis suggests that the beneficial arterial adaptations with exercise training would be more pronounced if initiated before menopause or around the menopausal transition due to the beneficial effects of estrogen signaling pathways on cardiovascular structure and function and the potential of preventing or reversing the early progression of pathophysiological adaptations before they are established. Although the development of this hypothesis was focused on the ability to improve the efficacy of exercise training in vascular adaptations, similar principles may apply to cardiac adaptations, particularly if the deleterious adaptation is underpinned by metabolic dysfunction or inflammatory processes associated with aging (227).

Another consideration could be that the training period or hemodynamic stimulus may have been insufficient to elicit significant cardiac adaptations in older women. Although studies have indicated that exercise training for 3–12 mo is sufficient to elicit LV adaptations in young men and women and older men, older women may require more intense exercise training over a longer period to stimulate physiological hypertrophy pathways (38, 228–230).

LV CHAMBER STIFFNESS AND DISTENSIBILITY WITH AGING AND EXERCISE TRAINING

Overview of Cardiac Chamber Stiffness (Compliance) and Distensibility

The LV diastolic stiffness, or its reciprocal compliance, is a principal determinant of the diastolic pressure-volume (p-v) relationship (231, 232). The LV chamber stiffness is largely determined by the relationship between the relatively compliant cardiac muscle and the comparatively less compliant connective tissue and extracellular matrix.

The LV chamber diastolic stiffness is determined by measuring LVEDV and LV filling pressure, typically PCWP, which is an accurate indicator of LV end-diastolic pressure (LVEDP) in the absence of severe valvular disease, during several experimental conditions in which LV preload is increased and/or decreased (133, 232, 233). The diastolic p-v relationship is curvilinear, with the slope increasing as the ventricle fills (231, 232). With very large diastolic filling volumes, as seen in elite endurance athletes, the diastolic p-v relationship is likely influenced by pericardial and pulmonary restraint (58, 132, 231). The LV chamber distensibility, another parameter that reflects the mechanical properties of the heart during diastole, is indicated by the LVEDV at a given filling pressure. A rightward shift of LVEDV for any given filling pressure would indicate increased LV distensibility.

Over the last three decades, a Dallas-based group has produced a series of reports on rigorously screened healthy men and women that have provided critical insight into the effects of healthy aging and relatively short-term or lifelong endurance exercise on cardiovascular structure and function (2, 3, 33, 182, 228, 234).

Aging Effects

Using invasive determination of LV filling pressures and concomitant measurement of LV volumes by echocardiography, Arbab-Zadeh et al. (2) showed a leftward shift and a steeper slope of the LV diastolic p-v curve suggestive of reduced distensibility and increased LV chamber stiffness (reduced compliance) in 12 sedentary seniors (6 women, 6 men; mean age ∼70 yr) compared with a similar number of sedentary young adults (7 women, 7 men; mean age ∼29 yr). Consistent with these outcomes, a follow-up retrospective analysis found a smaller, stiffer LV chamber in 35 older women (mean age ∼66 yr) compared with 22 young and middle-aged women (mean age ∼48 yr) (33) (Fig. 1E). Fujimoto et al. (3) showed that LV chamber stiffening manifests during late middle age (50–64 yr) in sedentary but otherwise healthy groups of men and women. Collectively, these findings suggest that LV chamber stiffness increases with aging and that the transition to late middle age appears to be a period of adulthood in which significant LV stiffening occurs.

The cellular and molecular processes that underpin the age-related increase in LV chamber stiffness may involve multiple signaling pathways. Based on rodent and human models, extracellular matrix remodeling and the accumulation of intramyocardial triglycerides and collagen cross-linked proteins are potentially involved (235–240).

Exercise Training Effects

In addition to demonstrating notable aging effects, Arbab-Zadeh et al. (2) showed the age-related increase in LV chamber stiffness was eliminated in Master endurance athletes (6 women, 6 men; average age 67 yr) who had performed near-daily exercise training plus sanctioned competitive events for virtually their whole adult lives. When transmural pressure was examined to assess the contribution of extraventricular forces (i.e., pericardial constraint), the marked differences in LV chamber stiffness between the athletes and sedentary adults remained. Thus, habitual, vigorous endurance exercise training appears to be an effective countermeasure to age-related changes in LV chamber stiffening.

To more accurately quantify the exercise dose that may protect against the age-associated increase in LV chamber stiffness, a follow-up investigation examined over 100 adults (32 women, 70 men) aged ≥60 yr who had a consistent history of performing different weekly frequencies of exercise training (at least 30 min of exercise per session) for >25 yr on LV chamber stiffness. While near-daily exercise (6–7 exercise sessions/wk) plus competition provided the greatest benefits, 4–5 exercise sessions/wk (committed exercisers) effectively protected participants from the detrimental effects of aging on LV stiffening. An exercise frequency of 4–5 sessions/wk is more practical to perform and represents a more acceptable population health message for physical activity recommendations (207). Additional analyses showed that this dose of exercise was sufficient to enhance V̇o_2max and other aspects of cardiac and central arterial function during alterations in venous return (61, 196, 241). Interestingly, this research also showed that 2–3 exercise sessions per week did not benefit LV chamber stiffness more than individuals who had been sedentary (0–1 exercise session/wk) during an equivalent period.

A recent retrospective analysis of endurance-trained and untrained women aged 20–75 showed that the LV diastolic p-v curve was shifted rightward and was less steep in endurance-trained older women compared with sedentary young and age-similar controls, indicating enhanced LV chamber compliance (33) (Fig. 1E). The relationship between transmural pressures and LVEDV was investigated in these women (2, 3, 207, 242), and the disparity in LV chamber compliance between trained and untrained women remained, verifying that factors other than extraventricular forces contributed to the enhanced cardiac compliance in these women (33). These findings confirmed that lifelong exercise protects LV chamber compliance in women similarly to men.

The mechanistic underpinnings of the exercise-mediated protection of LV chamber compliance in aging humans are not well understood; however, preventing or minimizing the age-related pathophysiological cellular and molecular processes previously detailed could explain the preserved LV chamber compliance with habitual vigorous lifelong endurance exercise training in humans.

Although tissue biopsies would provide the most convincing mechanistic evidence, this approach is not practical for healthy human volunteers. Instead, noninvasive imaging and biomarker measurements have produced equivocal findings on the relative importance of these mechanisms with regular and vigorous lifelong exercise (237, 243). As a result, the mechanisms that mediate the favorable modulatory effects of habitual endurance exercise on age-related changes in LV chamber stiffness remain relatively unknown in humans.

Whether regular endurance exercise training can reverse age-related LV chamber stiffening in previously sedentary older adults has been investigated. In nine adults (3 women, 6 men) aged ≥65 yr, 12 mo of progressive vigorous exercise, encompassing over 200 min weekly of moderate and high-intensity endurance exercise training, imparted favorable effects on V̇o_2max, maximal Q̇ and Ea, and LV mass; however, LV chamber stiffness was not changed postintervention (182). A follow-up randomized controlled trial in 57 (37 women, 20 men) aged >60 yr showed that a year of moderate-intensity endurance exercise training combined with an advanced glycation end product crosslink breaker (Alagebrium) reversed 10–15 yr of the aging process (234). However, supporting the earlier work of Fujimoto et al. (182), exercise training alone as a countermeasure strategy was unsuccessful in reducing LV chamber stiffness. Together, these findings appear to indicate that adaptations in LV chamber stiffness in previously sedentary older adults are very modest.

However, more promising findings have been reported for enhancing LV chamber compliance when exercise training is initiated earlier in life. Howden et al. (38) reported a rightward shift and flattening of the LV pressure-volume curve, indicating a reduction of LV chamber stiffness (improved LV chamber compliance) and distensibility after 1 yr of strenuous endurance exercise training in a small number of young women (n = 5, mean age ∼31 yr). Based on previous findings indicating that LV chamber stiffening manifests during midlife, a 2-yr exercise-based study aimed to initiate exercise training during this period. After the 2-yr intervention period, there was a significant reduction in LV chamber stiffness and increased distensibility (95). The authors reported that the female participants, who made up 56% of the exercise training group, did not respond differently to the magnitude of change in LV chamber stiffness with training from men. This finding suggests that midlife may be a potential “sweet spot” for women to restore to prevent age-related changes in LV chamber compliance.

LV RELAXATION AND DIASTOLIC SUCTION WITH AGING AND EXERCISE TRAINING

Overview of Ventricular Relaxation and Diastolic Suction

The rate and extent of myocardial relaxation influence the diastolic p-v relationship differently (231). The rate of relaxation will influence the diastolic filling rate during the rapid early filling phase due to its effect on the transmitral pressure gradient. The extent of the relaxation influences the entire diastolic p-v curve relationship due to its effect on equilibrium length and volume. An impaired extent of relaxation would result in a smaller equilibrium volume, leading to a higher diastolic pressure at a given LVEDV. Consequently, higher pressure at the same volume would result in an upward shift of the entire diastolic p-v curve relationship (231).

Diastolic suction results from muscle fiber compression to lengths shorter than their equilibrium length during systole (231). This end-systolic state stores elastic energy in the myocardium, which is subsequently released during relaxation, resulting in the elastic recoil of the ventricle. These cardiac mechanics create transmitral and intraventricular pressure gradients that actively suck blood into the ventricle from the atrium (244, 245). Rapid relaxation and vigorous diastolic suction aid early ventricular filling in the face of dramatically reduced diastolic filling times with exercise tachycardia (231, 246, 247). Abnormalities in these ventricular properties, as seen in heart failure, lead to a smaller stroke volume response and an upward shift of the entire diastolic p-v curve relationship during exercise (248).

Aging Effects

Due to the dynamic processes that influence diastolic filling, assessing relaxation and diastolic suction properties can be extremely difficult (231). Multiple imaging studies in men and women report that indices of LV relaxation and diastolic suction (peak early mitral annular velocity [E′], isovolumetric relaxation time [IVRT], time constant of isovolumic pressure decay [τ], diastolic untwisting rate, propagation velocity [Vp], intraventricular pressure gradients [IVPG)]) are slowed at rest with healthy aging (6, 7, 135, 249–253).

In healthy older adults, IVRT, E′, IVPG, and Vp were not restored to levels of young adults in experiments in which LV preload was altered using lower body negative pressure and saline infusion (7, 135). Similar findings were found using the same experimental approach in 70 healthy men and women aged 21–77 yr (6). Together, these findings suggest that these changes in dynamic diastolic relaxation and suction parameters are likely due to changes in intrinsic ventricular cellular and molecular processes with myocyte calcium handling, rather than potential age-related differences in LV filling conditions (6, 233).

Age-related changes in relaxation and diastolic suction indices are not attenuated during submaximal exercise when examined in exclusively male cohorts (253–255). Thus, it is unclear whether these parameters are altered during submaximal exercise with aging in women.

Exercise Training Effects

A larger exercise LVEDV in older athletes has raised interest in investigating whether age-related changes in ventricular relaxation and diastolic suction properties are prevented or attenuated with habitual vigorous endurance exercise. Although some studies have shown modest improvements in dynamic LV diastolic filling, much of the evidence in predominately midlife and older men indicates no substantial differences for a range of ventricular relaxation and diastolic suction indices at rest between endurance-trained and untrained adults (7, 61, 135, 254–258). Results from studies that collected indices of LV diastolic relaxation and suction across a wide range of LV filling pressures showed that the LV relaxation and diastolic suction properties were not different between endurance-trained and untrained men and women aged >60 yr (7, 135). Likewise, indices of LV relaxation and diastolic suction assessed during submaximal exercise are generally not superior in endurance-trained individuals (135, 253–255). There are no studies that have directly examined any exercise training effects on these indices in trained and untrained older women, and therefore it is unclear whether these diastolic ventricular properties are essential to improved exercise LV function in exercise-trained women.

MENOPAUSE TRANSITION PHASE—AN ACCELERATED AGING PHENOTYPE?

The menopause transition period is widely recognized as a critical period in which there is an accelerated increase in a woman’s CVD risk profile. However, delineating the effects of menopause is challenging, as the reductions in circulating estrogen associated with menopause temporally overlap and interact with age-related structural and functional adaptations of the cardiovascular system. In addition, understanding the effects of menopause on adaptations in cardiovascular structure and function is further challenged by methodological approaches such as reporting the cohort’s chronological age and not the time since menopause.

The powerful and diverse effects of estrogen on the cardiovascular system are achieved via binding to several receptors, including the nuclear estrogen receptors (ERs), ERα, and ERβ, and a new class of membrane G protein-coupled receptor, GPR30, referred to as GPER (259, 260).

EFFECTS OF ESTROGEN ON LV MORPHOLOGY AND MASS

Estrogen has been shown to modulate multiple genomic and nongenomic signaling molecules and pathways involved in cardiomyocyte hypertrophy and apoptosis (261). Estrogen enhances the degradation of calcineurin (a central molecule for pathological hypertrophy), hypertrophic transcription factor (nuclear factors of activated T cells; NF-AT), and mitogen-activated protein kinases (MAPK) signaling, leading to an attenuation of afterload-related ventricular hypertrophy (261). Estrogen also blocks other molecular mechanisms, including angiotensin II or endothelin-1 pathways that regulate cardiomyocyte hypertrophy and remodeling genes (262).

Cross-sectional studies have examined premenopausal and early postmenopausal women (amenorrhoeic ≥ 9 mo) similar in age (±5 yr) to minimize the confounding effects of aging on LV structure and reported a greater degree of concentric LV remodeling and impaired dynamic diastolic function in the postmenopausal women (263, 264). A longitudinal examination that collected measurements of cardiac structure over 30 yr, and analyzed pre- and postmenopausal trends, noted more rapid increases in LV mass and concentric LV remodeling after menopause, suggesting that changes in sex hormones with menopause influence cardiac structural changes. A follow-up longitudinal study of the same population reported that the age of natural menopause was inversely associated with a greater change in LV structural parameters (LV mass, LV mass/volume ratio, and relative wall thickness). However, further analyses showed that these significant effects were basically eliminated when accounting for traditional CVD risk factors (e.g., smoking, obesity, systolic blood pressure, physical activity, total cholesterol, and HDL cholesterol) established before menopause. Thus, these findings suggest that premenopausal CVD risk factors predispose women to a greater degree of LV structural adaptation during the menopause transition (265).

EFFECTS OF ESTROGEN ON LV CHAMBER STIFFNESS AND DISTENSIBILITY

Changes within the myocardium and/or LV hypertrophy are potential causes of the age-associated stiffening of the heart (3). In addition to modulating multiple molecular pathways involved in cardiomyocyte hypertrophy, estrogen also influences changes in myocardial tissue composition, including the deposition of fibrotic proteins, through signaling pathways involving transforming growth factor-β (TGFβ) and angiotensin II. The deposition of collagen in interstitial and perivascular spaces ultimately increases passive cardiac stiffness (261, 266–269). It is also thought that estrogen may influence the regulation of titin, the giant filamentous protein within the sarcomere that exerts a prominent influence on the passive and active properties of cardiac muscle. Diabetes-related rodent models have shown that estrogen can influence the ratio of titin’s two isoforms [N2B (stiff type) and N2BA (compliant type)], which leads to changes in passive LV diastolic stiffness (269, 270). Therefore, the menopause transition phase may be an important target for therapeutic strategies that reduce the adverse structural adaptations of the heart with aging.

EFFECTS OF ESTROGEN ON LV RELAXATION AND DIASTOLIC SUCTION

Estrogen can modulate cardiomyocyte relaxation via several proteins involved in the regulation of calcium (Ca²⁺) homeostasis (261). Rodent models have shown that estrogen influences the expression and activity of cardiac Ca²⁺ and ion membrane and intracellular channels, including L-type channels, sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase 2a (SERCA2a) ryanodine receptors, and protein kinase A (PKA) (271–273). Loss of ovarian estradiol is associated with decreased expression and activity of SERCA2a and ATPase resulting in decreased Ca²⁺ reuptake by the sarcoplasmic reticulum in cardiomyocytes and impaired myocyte relaxation in diastole (274).

Collectively, these findings highlight the potentially significant effect of estrogen on diastolic filling properties in women and thus reaffirm the previous point that the menopause transition period may represent a key phase of adulthood in which therapeutic strategies, including exercise training, may be particularly beneficial to establishing or preserving the favorable effects of estrogen on the heart.

FUTURE DIRECTIONS

Given the sparse data and the potential for sizable benefits in health promotion, the role of regular physical activity and exercise-based strategies in preserving or improving physical function, physiological function, and chronic disease risk in midlife and older women is a ripe area for future research. Several research areas would provide important scientific information that could translate into professional and clinical practice (Table 2). First, investigations should be directed at providing a better understanding of the fundamental O₂ delivery and tissue utilization mechanisms of exercise intolerance in midlife and older women. Second, establish the mechanisms of cardiovascular, skeletal muscle, and autonomic nervous system adaptations to short- and long-term endurance exercise training. Third, the development of novel, multimodal exercise training programs that aim to elicit adaptations in multiple organ systems in previously sedentary midlife and older women. The prescribed exercise dose should be suitable to achieve the desired metabolic, cardiovascular, skeletal muscle, or neural adaptations. As further research elucidates the dose-response relationship for biological and physiological adaptations with exercise-based interventions, this information will inform the design of prescriptions for enhancing important health and physiological outcomes in women’s populations. Fourth, while not discussed in detail in this review, detailed physiological studies examining the effects of endurance training on factors that influence cardiometabolic health in midlife and older women are required given that arterial hypertension, obesity, and diabetes are common comorbidities associated with HFpEF in women (9). Fifth, safety and adherence aspects should be examined, with barriers established for those who have been consistent exercisers to continue and those who are looking to initiate exercise programs later in life (275). Finally, devise strategies to improve the translation of findings garnered from aging and exercise research to clinical populations to promote health, well-being, functionality, and quality of life in women with CVD and other chronic diseases.

Table 2.

Research gaps and future directions

	Research Gap	Future Direction
Exercise intolerance	Physiological mechanisms that contribute to exercise intolerance	Define cardiovascular, skeletal muscle, and autonomic function during acute submaximal and peak/maximal exercise in midlife and older women
Physiological adaptions to training	Physiological adaptations to relatively short-term (1–6 mo), long-term (6 mo–1 yr) habitual (>5 yr), and lifelong (>20 yr) endurance training	Define cardiovascular, skeletal muscle, and autonomic nervous system adaptations to short-term, long-term, habitual, and lifelong exercise training in midlife and older women
Cardiac adaptations	Cardiac adaptations to endurance training in different stages of adulthood	Identify the cardiac adaptations to relatively short- and long-term habitual endurance training in young, midlife, and older women.
Ventricular-arterial coupling (v-a coupling)	v-a coupling adaptations with aging, menopausal status, and exercise training	Identify the v-a coupling changes with aging, menopausal status, and exercise training.
Exercise training prescription	Optimal prescription of physical activity and exercise training programs to elicit targeted physiological and health benefits	Identify the training modalities that optimally elicit targeted adaptations in the cardiovascular, skeletal muscle and autonomic nervous systems in midlife and older women
Dose-response effect	The exercise dose-response effect for targeted physiological adaptations for different modalities of physical training	Determine the dose-response effects of different exercise modalities on adaptations in V̇o2max, cardiovascular, skeletal muscle, and autonomic nervous systems in midlife and older women
Menopausal effects	The effect of menopausal status on cardiac adaptations to exercise training including testing the exercise timing hypothesis.	Determine whether menopausal status influences targeted training adaptations
Perceived barriers	The perceived barriers to initiating or continuation of physical activity or exercise training programs.	Define the perceived barriers to both initiating or continuing physical activity or exercise training programs in midlife and older women
Adherence, tolerance, and safety of physical activity and exercise programs	Adherence, tolerance, and safety of endurance-focused and multi-modality exercise training programs	Examinations of adherence, safety, and tolerance to physical activity and exercise training programs in midlife and older women
Exercise programs for special populations	Physiological adaptations to short and longer-term multi-modality exercise training programs in the oldest-old, frail, or those with chronic diseases	Determine the cardiovascular, skeletal muscle, and autonomic nervous system adaptations to different multimodal exercise training programs in special populations of midlife and older women

SUMMARY

In conclusion, adverse modifications of cardiovascular structure and function and a substantially reduced V̇o_2max and cardiac reserve during exercise are common observations in sedentary older women. Conversely, V̇o_2max, cardiac structure, and function at rest and exercise is markedly enhanced in women who have performed regular endurance exercise for several years or decades. Thus, habitual endurance exercise is an effective strategy for improving V̇o_2max and the structure and function of the cardiovascular system in aging women. For untrained women, endurance exercise should be initiated before or shortly after menopause to maximize the beneficial gains in V̇o_2max and structural and functional cardiovascular adaptations.

GRANTS

G.C.-R. holds a Hugo Charitable Trust Fellowship. E.J.H. is a holder of a National Heart Foundation Future Leader Fellowship under Grant No. 102536.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

G.C.-R. prepared figures; G.C.-R., E.J.H., T.L.B., B.D.L., and S.A.R. drafted manuscript; G.C.-R., E.J.H., T.L.B., B.D.L., and S.A.R. edited and revised manuscript; G.C.-R. approved final version of manuscript.