Secondary Logo

Journal Logo


The Role of Exercise-Induced Cardiovascular Adaptation in Brain Health

Tarumi, Takashi1,2; Zhang, Rong1,2,3

Author Information
Exercise and Sport Sciences Reviews: October 2015 - Volume 43 - Issue 4 - p 181-189
doi: 10.1249/JES.0000000000000063
  • Free


We are facing an unprecedented aging of population in the history of humanity. Life expectancy has nearly doubled during the past two centuries, whereas birth rate has steadily declined in developed countries. Advanced age is a major risk factor for chronic noncommunicable diseases including Alzheimer disease (AD), the most common type of late-life dementia. In 2010, AD affected about 33.9 million people worldwide, and its prevalence is expected to triple by 2050 if no effective prevention or treatments are developed. Currently, there is no cure for AD; however, population-based studies revealed that approximately 30% of AD cases may be related to modifiable risk factors such as physical inactivity and cardiovascular risk factors (23).

Compelling evidence suggests that habitual aerobic exercise attenuates age-related cognitive decline that is linked to the preservation of brain structure (8,12,38). The mechanisms underlying exercise-related improvements in brain structure and function are not well understood and likely to be multifactorial. One possible mechanism is that cardiovascular adaptations to endurance exercise ameliorate brain health through attenuation of age-related arterial stiffness and/or endothelial dysfunction (30,31). In addition, exercise-related reductions in other cardiovascular risk factors may contribute to the reduced risk of cognitive impairment and dementia.

Another important question is whether there is a dose-response relationship between the intensity of exercise and brain health. We currently do not know whether more exercise would naturally lead to a better brain health or such benefits would plateau or even have deleterious effects beyond a certain “dose” of exercise. The existing evidence suggests the presence of a hormetic relation between the intensity of exercise and brain structure and function, such that strenuous exercise performed without an adequate recovery may be deleterious to brain health (2,13). To gain insights into these critical questions, we recently investigated master athletes (MA), a unique group of middle-aged and older adults who have participated in long-term or lifelong endurance exercise training.


Mounting evidence suggests that regular aerobic exercise attenuates age-related cognitive decline (8). Cognitive performance peaks during early adulthood and gradually decreases after late 20s to 30s (24). Specifically, the tasks requiring working memory and attention-executive function such as reasoning, perceptual speed, and spatial visualization experience an earlier loss, whereas performance on tasks involving the crystallized intelligence including vocabulary and general world knowledge is relatively spared until much later (24).

Investigating two independent samples of MA, we found higher cognitive performance in memory and executive function compared with their sedentary peers (31,36). Figure 1 shows the result of neurocognitive assessments in middle-aged MA and sedentary adults. In this study (31), besides age, sex, and education level, lifestyle factors such as daily nutritional intake and sleep quality were similar between the groups. MA have participated in vigorous endurance exercise training for more than 1 h d-1. According to an age- and sex-adjusted regression analysis established by the American College of Sports Medicine, cardiorespiratory fitness of these MA, as assessed by maximal oxygen uptake (V˙O2max), was more than 90 percentile whereas that of sedentary participants was approximately 15 percentile of the general population. In this study, the effect size of MA, as estimated by the standardized difference between the two groups (Cohen d), was 0.75, 0.61, and 0.84 in episodic memory, attention-executive function, and total cognitive composite scores, respectively. This suggests a medium to large effect size of vigorous aerobic exercise training on improving cognitive function, consistent with a previous meta-analysis and systematic review of exercise training and cognition (8). Another study of older MA with similar levels of cardiorespiratory fitness also exhibited better cognitive performance in global intelligence and executive function when compared with age, sex, and education level-matched sedentary older adults (36).

Figure 1
Figure 1:
Endurance-trained master athletes demonstrated a higher cognitive performance in episodic memory, attention-executive function, and total composite scores than age-matched sedentary adults. The sample-specific Z score was calculated for each of the cognitive domains. The numbers inside bar graphs are mean ± standard error. (Reprinted from (31). Copyright © 2013 Wolters Kluwer Health. Used with permission.)

In addition, a systematic review of prospective studies demonstrated that efficacy of exercise training on cognitive function increased when performed longer than 6 months, with each exercise session lasting 30 to 45 min (8). The favorable effect of exercise training was greater when performed earlier than later in life and when combined with strength exercise compared with aerobic exercise alone (8).


The human brain contains approximately 100 billion neurons and consists of the gray (GM) and white (WM) matter. The GM is composed mainly of neuronal cell bodies with branched projections (dendrites). The dendrites receive electrochemical stimuli from the neighboring and remote neurons via synapse and propagate them to the cell body. Subsequently, the neuronal axons in WM transmit electrical impulse to the surrounding neurons and function as a relay between the adjacent and remote brain areas as well as between the brain and peripheral organs. The axons are covered by myelin, which increases the conduction velocity of electrical impulse and plays a critical role for efficient neuronal communications. With advancing age, brain atrophy, beginning as early as one’s 20s to 30s of age, combined with axonal demyelination that commensurately slows propagation of electrical impulses, results in a loss of cognitive efficiency. In contrast, habitual aerobic exercise may prevent or at least in part attenuate these age-related deteriorations of GM and WM physiological and cellular functions, therefore improving cognitive performance.

Brain Volume

Regular aerobic exercise may increase or preserve regional brain volume at areas affected by aging and associated with cognitive impairment. Tseng et al. (36) acquired high-resolution T1 magnetic resonance (MR) images to examine brain volume in MA. Using a voxel-based morphometric approach, MA exhibited greater volume at several cortical areas associated with motor control and visuospatial function than sedentary older adults (Fig. 2). Specifically, these regions were the Brodmann areas 7 (parietal lobe) and 19 (visual cortex) and the culmen (anterior portion of the cerebellum). They also found a greater concentration of WM at the parietal, occipital, and temporal lobes. Finally, V˙O2max was correlated positively with the precuneus GM volume as well as the subgyral frontal and occipital WM volume, when analyzed in all study participants (36).

Figure 2
Figure 2:
Master athletes exhibited higher gray matter volume in the (A) SII/SPL, (B) V2/IOG, and (C) higher white matter volume in the O3-WM than age-matched sedentary adults. The color bar illustrates t statistic scores. The neurological coordinates are based on the Montreal Neurological Institute (MNI) space. SII, secondary sensorimotor cortex; SPL, superior parietal lobule; V2, secondary visual cortex; IOG, inferior occipital gyrus; O3-WM, inferior occipital gyrus white matter; A, anterior; and P, posterior. (Reprinted from (36). Copyright © 2013 Wiley Online Library. Used with permission.)

Regional brain volume may increase after a relatively short period of moderate-intensity aerobic exercise. Sedentary older adults who participated in 6 to 12 months of a walking program demonstrated greater volume of the hippocampus as well as several regions of the frontal lobe, including anterior cingulate cortex, supplementary motor area, and middle frontal gyrus, when compared with the control group that performed stretching (9,12). In addition, volume of anterior WM tracts, particularly the genu of the corpus callosum, increased in the walking but not the stretching group (9).

Brain WM Neuronal Fiber Integrity

The structural integrity of WM neuronal fibers can be assessed by diffusion tensor imaging (DTI). Using the motion of water molecules in the brain tissues as an endogenous probe, DTI provides diffusion metrics that reflect an overall integrity as well as biological characteristic of WM fiber tracts. Specifically, fractional anisotropy (FA) and mean diffusivity (MD) provide a measure of water diffusive directionality and magnitude respectively, in that a higher FA and/or a lower MD generally reflect a better integrity of WM fiber tracts. These DTI metrics have been used to detect early abnormalities which may precede WM lesions (22).

MA demonstrated higher FA and lower MD in the WM tracts associated with motor control and coordination when compared with their sedentary peers (Fig. 3) (35). Specifically, these tracts included superior corona radiata, superior longitudinal fasciculus, superior longitudinal fasciculus, inferior fronto-occipital fasciculus, and posterior thalamic radiation. In addition, FA measured from the left superior and inferior longitudinal fasciculi was correlated positively with V˙O2max in all subjects. Furthermore, these MA exhibited better WM tract integrity in the cingulum, which carries and integrates memory information from the hippocampus. Damage to this tract is associated with mild cognitive impairment and AD. Finally, subcortical WM lesion volume was smaller in MA than in the control group.

Figure 3
Figure 3:
Master athletes exhibited higher fractional anisotropy (A–D) and lower mean diffusivity (E–F) of brain white matter neuronal fiber tracts. These tracts are (A) right superior corona radiata, (B) right superior longitudinal fasciculus, (C) left superior longitudinal fasciculus, (D) right inferior fronto-occipital fasciculus, (E) left posterior thalamic radiation, and (F) left cingulum hippocampus. The white matter tracts are shown by green pixels, whereas the red pixels show disruptions of white matter fiber tracks in sedentary elderly compared with master athletes. A scatter plot (G) shows the correlation of maximal oxygen uptake (V˙O2max) with fractional anisotropy (FA) measured from the left superior longitudinal fasciculus (SLF) and left inferior longitudinal fasciculus (ILF). A, anterior; P, posterior; L, left; R, right. [Adapted from (35). Copyright © 2013 Elsevier. Used with permission.]

To date, there are few studies that investigated the effect of aerobic exercise training on WM structural integrity using DTI. Voss et al. (38) conducted a randomized controlled trial of 1-yr walking program in sedentary community-dwelling older adults. In this study, the exercise group did not show group-level improvement in WM integrity or cognitive function compared with the control subjects who performed stretching. However, within the walking group, there was a positive correlation between the individual improvements in V˙O2max and FA measured from the frontal and temporal lobe WM. More recently, Svatkova et al. (27) conducted a 6-month interventional trial of moderate-intensity cycling in the group of healthy participants as well as patients with schizophrenia. After the program, both groups exhibited elevations in FA at the areas associated with motor function. Collectively, these findings suggest that brain WM integrity is related to cardiorespiratory fitness and those WM areas responsible for motor function likely are sensitive to exercise training.


The voluntary contractions of skeletal muscle during dynamic exercise dramatically increase the metabolic demand and challenge whole-body homeostasis. In response, the body must coordinate its compensatory mechanisms that deliver adequate oxygen and nutrients to the working muscles. Such compensatory responses take place in multiple organs, particularly the cardiovascular system, and minimize the disruptions of homeostasis. During exercise, heart rate and cardiac contractility increase to augment cardiac output, which is the systemic source of blood flow distributed to all organs. For efficient substrate delivery, blood flow is redistributed to the working muscles during dynamic exercise via an integrative mechanism of nervous, hormonal, and humoral systems. Consequently, dynamic exercise performed at moderate to high intensity can increase cardiac output and skeletal muscle blood flow by eightfold and 100-fold, respectively, whereas brain blood flow may increase by approximately 10% to 20% above the resting values during moderate-intensity exercise (16,25).

Repeated bouts of aerobic exercise during a prolonged period improve the cardiovascular control of systemic perfusion and counteract the degenerative processes of aging and/or disease (16). Because the brain lacks intracellular energy storage and heavily relies on the vascular supply of oxygen and nutrients, exercise-related improvements in cardiovascular control of cerebral blood flow (CBF) may play a crucial role in maintaining normal brain function and structure. Although CBF is coupled closely with the metabolic demand of neurons, hypoperfusion and/or impaired regulation of CBF attributed to vascular disease or dysfunction can cause neuronal dysfunction (17). Below, we discuss the key components of cardiovascular system that may benefit from aerobic exercise training and confer the favorable effects on brain structure and function (Fig. 4).

Figure 4
Figure 4:
A hypothetical diagram illustrating the effect of aerobic exercise training on cerebral blood flow (CBF) regulations. “Normal” CBF is maintained by the hierarchical layers of hemodynamic support extending from the systemic circulation to the local capillary levels. Although not shown in this diagram, there are overlaps and interactions between each of the layers. For example, cerebral autoregulation and vasomotor reactivity may also take place at the level of large conduit arteries (e.g., internal carotid and vertebral arteries). Arterial baroreflex may interact with cerebral autoregulation. Please see the text for detailed explanations. Arrows next to vascular phenotype represent the change that may occur after aerobic exercise training. Currently, potential effects of aerobic exercise training on the cerebral vasculature, such as cerebral autoregulation, vasomotor reactivity, and microcirculatory function, remain largely unknown.

The Ventricular-Central Arterial Coupling

Regular aerobic exercise alters chronotropic and inotropic control of cardiac output. Specifically, endurance training decreases heart rate at rest by shifting autonomic balance of sympathetic and parasympathetic innervations to the heart. This is accompanied by an eccentric remodeling of cardiac chambers, which increases left ventricular end-diastolic filling and stroke volume. Because of a close coupling with metabolism, cardiac output at rest may not change in healthy individuals after exercise training.

In contrast, patients with heart failure may benefit from aerobic exercise training by improving cardiac function at rest. According to the Framingham Heart Study, systemic hypoperfusion resulting from depressed cardiac index (i.e., cardiac output divided by body surface area) is associated with an elevated risk of incident dementia and AD (19). Conversely, an 18-wk treadmill program improved cognitive performance in attention and psychomotor speed, accompanied by increased cardiac index at rest, in patients with severe congestive heart failure (28).

Habitual aerobic exercise attenuates age-related stiffening of central elastic arteries (e.g., aorta and carotid arteries) (26), which may improve the efficiency of left ventricular ejection of stroke volume. The left ventricle generates systolic blood pressure (SBP) by ejecting a stroke volume against a hydraulic load (i.e., aortic impedance). Aortic impedance is determined by its morphological characteristics such as diameter and wall elasticity. With age and/or presence of cardiovascular risk factors, the aorta stiffens; arterial wave reflections returning from the peripheral vascular beds arrive prematurely to the heart; and consequently aortic impedance rises. These changes necessitate the left ventricle to increase contractility and maintain stroke volume and cardiac output. With endurance training, reductions in central arterial stiffness and cardiac afterload decrease the contractile energy required to generate the equivalent stroke volume, accompanied by a lower SBP. Importantly, SBP is an established risk factor for stroke and cognitive impairment.

Exercise-related reductions in central arterial stiffness may attenuate the transmission of excessive blood pressure pulsatility into the brain (32). Cerebral circulation has low vascular resistance/impedance and, thus, may be vulnerable to excessive hemodynamic pulsatility transmitted into the microcirculations. In young healthy adults, SBP generated by the left ventricle is dampened effectively by the Windkessel function of central arteries that expand and recoil with cardiac pulsations. With arterial stiffening and chronic exposure to high blood pressure pulsatility, cerebral resistance vessels may adapt by increasing vascular resistance. This maladaptation has been shown in animal studies where higher pulse pressure is associated with hypertrophic remodeling of cerebral arterioles (4). Elevated cerebrovascular resistance may not only increase the risk of ischemia (7,33) but also impair a clearance of neuronal waste products (e.g., amyloid-β) (18). Consistent with these notions, MA demonstrated greater distensibility of the carotid artery that was correlated positively with a higher cerebral perfusion in the occipitoparietal area (31).

Regular aerobic exercise may improve blood pressure control via enhanced arterial baroreflex function (1). Arterial baroreceptors, a type of mechanoreceptor located in the aortic arch and carotid sinus, monitor changes in blood pressure via mechanical distortion of the vessel walls. The baroreceptor afferents are relayed to the brainstem, which in turn sends out autonomic efferent to the heart and blood vessels to control blood pressure. The brain is particularly sensitive to changes in blood pressure because of its low resistant vascular bed. Therefore, baroreflex-mediated control of blood pressure is likely important for maintaining stable CBF, possibly working together with cerebral autoregulation (discussed below) (37). Regular aerobic exercise decreases stiffness of the barosensory arteries and thereby restores baroreflex sensitivity (1,26,31). Our recent study exhibited that baroreflex sensitivity is associated with the structural integrity of WM neuronal fibers, as assessed by DTI, and cognitive performance in executive function (i.e., Trail Making Test B minus A) in older adults (Fig. 5) (29).

Figure 5
Figure 5:
Associations among cardiovagal baroreflex sensitivity (BRS), fractional anisotropy measured from brain white matter neuronal fiber tracts, and Trail Making Test B−A performance in older adults. The BRS was assessed during acute changes in arterial blood pressure using the established pharmacological method (modified Oxford technique). The brain images on the top (A) exhibit the area of neuronal fiber tracts that were significantly associated with BRS based on voxelwise statistics. The color bar illustrates P value of the association. At the bottom (B), scatter plots show the correlations of BRS (left) and Trail Making Test B−A performance (right) with individual FA values that were extracted from the significant white matter tracts highlighted by voxelwise analysis (A). [Adapted from (29). Copyright © 2015 Elsevier. Used with permission.]

Collectively, exercise-related increase in central arterial elasticity may have favorable effects on CBF regulations, including attenuation in the transmissions of excessive blood pressure pulsatility, reduction in cerebrovascular resistance, and stable blood pressure control and CBF homeostasis. These data further suggest that central elastic arteries function as a key vascular component that bridges the heart to the brain.

Large Conduit Artery: Endothelial Function and Atherosclerosis

The main function of conduit arteries is the delivery and regulation of blood flow to the peripheral organs. In the brain, a large portion of vascular resistance is located outside of the parenchyma including the pial arterioles as well as the large extracranial (i.e., internal carotid and vertebral arteries) and intracranial arteries, whereas cerebral penetrating arterioles and capillaries account for the remaining (21,39). This necessitates a coordinated regulation of cerebrovascular resistance both inside and outside of the parenchyma to maintain adequate blood supply to the active neurons (i.e., neurovascular coupling). In this regard, vascular endothelial cells are situated ideally at the intima of the arterial wall, release vasoactive substances in response to blood shear stimuli as well as neuronal and blood-borne chemicals, and protect the vessel walls from atherogenesis.

Regular aerobic exercise improves vascular endothelial function via an upregulation of nitric oxide bioavailability (26). The nitric oxide regulates vasomotor tone during rest and functional activations and also inhibits atherogenesis by reducing oxidative modification of low-density lipoprotein cholesterol and preventing the proliferation of vascular smooth muscle cells. Atherosclerosis, a pathological condition characterized by the buildup of cholesterol and inflammatory substances in the arterial walls, occludes blood flow and is associated with AD pathology (6). Thus, exercise-related ameliorations in endothelial function as well as a reduced risk of atherosclerosis may facilitate cerebral perfusion.

Cerebral Resistance Vessel: Cerebral Autoregulation and Vasomotor Reactivity

Cerebral autoregulation (CA) and vasomotor reactivity (CVMR) provide a local control of CBF. CA is a protective function of cerebral resistance vessels that keep CBF relatively constant in the face of changes in cerebral perfusion pressure. Although uncertain, age and endurance training seem to have fewer effects on the CA compared with the peripheral vascular beds (5). For example, MA demonstrated the presence of similar dynamic CA compared with sedentary adults, as assessed by transfer function analysis of arterial pressure and CBF velocity measured in the middle cerebral artery during a repeated sit-stand maneuver (1). Currently, we do not know whether steady-state CA also is less influenced by age or exercise training.

There are inconsistent findings on the effects of age and regular aerobic exercise on CVMR. In our study of MA (40), CVMR was assessed by transcranial Doppler using a modified rebreathing technique that induces incremental elevations in end-tidal CO2, a proxy of arterial pCO2. We found a higher CVMR during hypercapnia in sedentary and endurance-trained older adults than in younger subjects, whereas no difference was observed between the older groups. Testing a similar sample using functional MR imaging, blood oxygen level-dependent (BOLD) responses to steady-state hypercapnia were attenuated in MA compared with that observed in sedentary older adults (34) (Fig. 6). In contrast, other studies reported higher CVMR in exercise-trained adults than in sedentary subjects, as assessed by transcranial Doppler during steady-state hypercapnia (3). Therefore, these inconclusive findings necessitate the standardization of techniques to assess CVMR. Of note, neither transcranial Doppler nor BOLD measures CBF per se, and rebreathing and steady-state hypercapnia may have different effects on cerebrovascular beds and brain tissues. A recent study using multimodal methods to quantify CBF (i.e., arterial spin labeling and BOLD) revealed a significant contribution of basal cerebrovascular tension to CVMR (15). As such, elevated cerebrovascular tension under resting conditions may increase hypercapnic vasodilatory reserve while decreasing hypocapnic vasoconstrictor capacity, consistent with our previous observations (40).

Figure 6
Figure 6:
Panel A shows cerebrovascular reactivity (CVR) to steady-state hypercapnia at regions of interest (ROI), as assessed by blood oxygen level dependence (BOLD) using functional magnetic resonance imaging. Master athletes (MA) demonstrated lower levels of CVR in all of the ROI than the sedentary elderly (SE) and young control (YC) groups. The CVR was calculated as a percentage change in BOLD divided by an absolute change in end-tidal CO2 (in millimeters mercury) relative to the baseline level. Panel B shows the result of voxelwise comparison of CVR between the MA and SE groups. Consistent with the ROI analysis, MA showed a lower CVR throughout the cortex (red), and there was no cluster in which CVR was greater in MA than in SE subjects. Fro, frontal lobe; Tem, temporal lobe; Par, parietal lobe; Occ, occipital lobe; Cer, cerebellum; Ins, insula; Sub, subcortical gray matter. Error bars represent standard error. (Reprinted from (34). Copyright © 2013 Wiley Online Library. Used with permission.)

Cerebral Microcirculation: Substrate Exchange and Waste Clearance

CBF ultimately perfuses neuronal tissues at the level of the capillary where substrate exchange and waste product clearance occur across the blood-brain barrier. Regional cerebral perfusion is coupled tightly with metabolic demand of neuronal tissues (neurovascular coupling), which presents a high level of temporal and spatial heterogeneity. Among the important areas of brain that are highly active during rest and also involved in AD pathology are the prefrontal lobe, medial temporal lobe including the hippocampus, posterior cingulate cortex (PCC), and precuneus. These brain areas function as a neural network hub in mediating and processing information flow and also serve as the main components of the Default-Mode-Network that contributes to learning and memory consolidations (14).

MA demonstrated an enhancement of CBF at the PCC and precuneus compared with sedentary older adults (34) (Fig. 7). Using the arterial spin labeling technique, whole-brain and regional CBF were measured. Our initial analysis of whole-brain CBF showed similar levels between the sedentary and endurance-trained older subjects, whereas younger subjects demonstrated higher levels of CBF than the older groups. Next, we calculated relative CBF, which represents a ratio of regional CBF to the whole-brain value and normalizes the effect of interindividual differences in global CBF. This analysis revealed that MA have significant elevations in relative CBF at the PCC and precuneus compared with young and older sedentary subjects. Therefore, these findings suggested that aerobic exercise training selectively preserve blood supply in the PCC and precuneus by attenuating the age effect.

Figure 7
Figure 7:
In panel A, voxelwise analysis revealed a higher level of relative cerebral blood flow (rCBF) in master athletes (MA) than in sedentary elderly (SE) subjects (cluster size, 250). The cluster was located in the posterior cingulate cortex (PCC) and precuneus of the Default-Mode-Network. No clusters showed MA < SE. Results are shown in glass brain view. Panel B shows group-averaged rCBF in the area highlighted by panel A. The rCBF values in MA were higher than those in the SE and young control (YC) groups. In panel C, absolute CBF in the same area showed a significant reduction in SE compared with YC; however, MA and YC showed similar levels. These data suggest that aerobic exercise training selectively preserves CBF in the PCC/precuneus region by attenuating the age effect. The rCBF was calculated as a ratio of regional CBF to the whole-brain value. CBF was measured by arterial spin labeling technique.*P < 0.05, **P < 0.005. Error bars represent standard error. (Reprinted from (34). Copyright © 2013 Wiley Online Library. Used with permission.)


MA as well as sedentary adults who have undergone aerobic exercise training even for a short period of 6 months to a year demonstrated a similar magnitude of improvements in cognitive performance when compared with their control groups (8,31,36). These improvements appeared mainly in the memory and executive function. On the other hand, brain structural adaptations to exercise training seemed somewhat different between the cross-sectional and longitudinal studies. For example, those sedentary adults who completed a short-term exercise training program demonstrated increases in hippocampal and prefrontal volume as well as frontal and temporal WM integrity (9,12,38). In contrast, MA exhibited preservations of the parietal lobe volume and WM integrity, which is not only limited to the frontal or temporal areas (35,36). Although it is important to acknowledge the limitations of different study designs and MA may differ from the general population in terms of genetic and/or lifestyle factors, these findings collectively suggest that aerobic exercise training for both a short and a longer duration has favorable effects on brain structure and function. Prospective trials that investigate the lifelong effects of exercise training are difficult to conduct; however, future studies combining cross-sectional and interventional study design may elucidate further the relation between the duration of exercise training and brain structure and function.

How is exercise intensity linked with brain health in general and cerebrovascular health in particular? A meta-analysis of cohort and case-control studies reported a dose-dependent inverse linear relation between physical activity intensity and stroke incidence or mortality (11). Specifically, moderately active individuals had a 20% lower risk and highly active individuals had a 27% lower risk of stroke incidence or mortality than the low-active individuals (11). However, this meta-analysis was limited by different definitions of physical activity intensity used in individual studies, and a few prospective studies reported a curvilinear relation between exercise intensity and stroke risk. For example, Harvard University alumni study showed a U-shape relation between relative risks of stroke and the estimated weekly energy expenditure accounted for by physical activity (20). Particularly, the effect of physical activity on reducing stroke risk steadily increased up to the energy expenditure of 3000 kcal wk-1 but attenuated thereafter. In addition, physical activity performed with moderate intensity such as walking 20 km wk-1 or more or stair climbing (≥4.5 metabolic equivalents) was associated with the nadir of the U-shape relation.

Emerging evidence suggests that strenuous endurance exercise leads to brain injury when performed without adequate recovery (2,13). Freund et al. (13) measured global GM volume of ultramarathon runners before, twice during, and 8 months after the race using MR imaging. The race took over 4487 km (2788 miles) in 64 days without a rest day. During the race, those runners experienced an average of approximately 6% reduction in global GM volume accompanied by a significant weight loss. The authors estimated that this magnitude of the volume reductions equates to approximately 30 yr of age-related atrophy, which normally progresses at an annual rate of approximately 0.2%. After 8 months of the race, the reduction in GM volume recovered back to the baseline level. Other studies also reported the link between prolonged endurance exercise and an elevated risk of cerebral lesions or edema (2).

The mechanism underlying the adverse effects of strenuous exercise on the brain is not clear. Strenuous exercise substantially increases systemic catabolic burden, inflammatory responses, and risks of cardiovascular injury, which may affect brain structure and function. Under normal physiological conditions, the brain secures its energy intake to support the high metabolic demand of neuronal tissues via neurovascular coupling and a relatively constant blood supply. Conversely, a prolonged high-intensity exercise can deplete intracellular energy storage of active skeletal muscles (e.g., phosphocreatine and glycogen). Thereafter, those working muscles need to rely on oxidative phosphorylation of adenosine triphosphate, which may not be able to sustain a high exercise intensity (16). These elevations in metabolic demand can lead to systemic catabolic conditions, which may compromise brain structure and function via elevated stress hormone levels (e.g., cortisol), changes in electrolyte balance (e.g., hyponatremia), inflammation and edema, hypoxia, and oxidative stress (2). Nevertheless, this apparent hormetic relation between the intensity of exercise and brain health needs to be investigated further.

Currently, there is no clear evidence of a dose-response relation or any consensus on the optimal dose of exercise training that may prevent or slow the age-related functional and structural deteriorations of the brain. It is likely that such an “optimal” dose is individually based and determined by the environmental and genetic factors, including age, sex, cardiovascular disease risk, and presence or absence of brain-derived neurotrophic factor Val66Met or apolipoprotein E polymorphisms (10). Furthermore, most studies to date focused on older subjects and there are not enough data to support whether exercise training in middle age, for example, would modify the trajectory of cognitive aging, although midlife vascular risk factors (e.g., hypertension, obesity) elevate an incidence of late-life dementia (23). For future work, much research is needed to identify and design a personalized “optimal program” of exercise training for heterogeneous populations, including the individuals who have comorbidities or chronic diseases.


Current evidence suggests that regular aerobic exercise performed at moderate to high intensity attenuates an age-related reduction in regional brain volume, deterioration of WM integrity, and cognitive decline. Cardiovascular adaptations to aerobic exercise training represented by reductions in arterial stiffness and improvements in endothelial function make a favorable systemic and cerebral hemodynamic environment (milieu) where the brain may benefit from the improvements in arterial pressure regulation, blood flow homeostasis, and metabolic waste clearance. However, strenuous exercise performed without adequate recovery may have deleterious effects as manifested by GM atrophy and WM lesions, which suggests the presence of a hormetic relation between the intensity of exercise and brain health. Currently, there are no effective prevention and treatment modalities for dementia; however, improvement of cardiovascular health, particularly via regular aerobic exercise, may prevent or slow age-related cognitive decline and delay the onset of dementia, thus improving quality of life and extending a healthy life span.

This work was supported by the National Institutes of Health (R01AG033106, R01HL102457, and P30AG012300) and the American Heart Association (14POST20140013). The authors thank Jonathan Riley and Erin Howden for editing and revising the manuscript.


1. Aengevaeren VL, Claassen JA, Levine BD, Zhang R. Cardiac baroreflex function and dynamic cerebral autoregulation in elderly masters athletes. J. Appl. Physiol. (1985). 2013; 114(2): 195–202.
2. Ayus JC, Varon J, Arieff AI. Hyponatremia, cerebral edema, and noncardiogenic pulmonary edema in marathon runners. Ann. Intern. Med. 2000; 132(9): 711–4.
3. Bailey DM, Marley CJ, Brugniaux JV, et al. Elevated aerobic fitness sustained throughout the adult lifespan is associated with improved cerebral hemodynamics. Stroke. 2013; 44(11): 3235–8.
4. Baumbach GL. Effects of increased pulse pressure on cerebral arterioles. Hypertension. 1996; 27(2): 159–67.
5. Carey BJ, Eames PJ, Blake MJ, Panerai RB, Potter JF. Dynamic cerebral autoregulation is unaffected by aging. Stroke. 2000; 31(12): 2895–900.
6. Casserly I, Topol E. Convergence of atherosclerosis and Alzheimer’s disease: inflammation, cholesterol, and misfolded proteins. Lancet. 2004; 363(9415): 1139–46.
7. Clark LR, Nation DA, Wierenga CE, et al. Elevated cerebrovascular resistance index is associated with cognitive dysfunction in the very-old. Alzheimer Res. Ther. 2015; 7(1): 3.
8. Colcombe S, Kramer AF. Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychol. Sci. 2003; 14(2): 125–30.
9. Colcombe SJ, Erickson KI, Scalf PE, et al. Aerobic exercise training increases brain volume in aging humans. J. Gerontol. A Biol. Sci. Med. Sci. 2006; 61(11): 1166–70.
10. Cotman CW, Berchtold NC, Christie LA. Exercise builds brain health: key roles of growth factor cascades and inflammation. Trends Neurosci. 2007; 30(9): 464–72.
11. Lee CD, Folsom AR, Blair SN. Physical activity and stroke risk: a meta-analysis. Stroke. 2003; 34(10): 2475–81.
12. Erickson KI, Voss MW, Prakash RS, et al. Exercise training increases size of hippocampus and improves memory. Proc. Natl. Acad. Sci. U. S. A. 2011; 108(7): 3017–22.
13. Freund W, Faust S, Birklein F, et al. Substantial and reversible brain gray matter reduction but no acute brain lesions in ultramarathon runners: experience from the TransEurope-FootRace Project. BMC Med. 2012; 10: 170.
14. Greicius MD, Krasnow B, Reiss AL, Menon V. Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. Proc. Natl. Acad. Sci. U. S. A. 2003; 100(1): 253–8.
15. Halani S, Kwinta JB, Golestani AM, Khatamian YB, Chen JJ. Comparing cerebrovascular reactivity measured using BOLD and cerebral blood flow MRI: the effect of basal vascular tension on vasodilatory and vasoconstrictive reactivity. Neuroimage. 2015; 110: 110–23.
16. Hawley JA, Hargreaves M, Joyner MJ, Zierath JR. Integrative biology of exercise. Cell. 2014; 159(4): 738–49.
17. Hossmann KA. Viability thresholds and the penumbra of focal ischemia. Ann. Neurol. 1994; 36(4): 557–65.
18. Iliff JJ, Wang M, Zeppenfeld DM, et al. Cerebral arterial pulsation drives paravascular CSF–interstitial fluid exchange in the murine brain. J. Neurosci. 2013; 33(46): 18190–9.
19. Jefferson AL, Beiser AS, Himali JJ, et al. Low cardiac index is associated with incident dementia and Alzheimer disease: the Framingham Heart Study. Circulation. 2015; 131(15): 1333–9.
20. Lee IM, Paffenbarger RS Jr. Physical activity and stroke incidence: the Harvard Alumni Health Study. Stroke. 1998; 29(10): 2049–54.
21. Lewis NC, Smith KJ, Bain AR, Wildfong KW, Numan T, Ainslie PN. Impact of transient hypotension on regional cerebral blood flow in humans. Clin. Sci. (Lond.). 2015; 129(2): 169–78.
22. Mori S, Zhang J. Principles of diffusion tensor imaging and its applications to basic neuroscience research. Neuron. 2006; 51(5): 527–39.
23. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. Lancet Neurol. 2014; 13(8): 788–94.
24. Salthouse TA. When does age-related cognitive decline begin? Neurobiol. Aging. 2009; 30(4): 507–14.
25. Sato K, Ogoh S, Hirasawa A, Oue A, Sadamoto T. The distribution of blood flow in the carotid and vertebral arteries during dynamic exercise in humans. J. Physiol. 2011; 589(Pt 11): 2847–56.
26. Seals DR, DeSouza CA, Donato AJ, Tanaka H. Habitual exercise and arterial aging. J. Appl. Physiol. (1985). 2008; 105(4): 1323–32.
27. Svatkova A, Mandl RC, Scheewe TW, Cahn W, Kahn RS, Pol HEH. Physical exercise keeps the brain connected: biking increases white matter integrity in patients with schizophrenia and healthy controls. Schizophr. Bull. 2015; 41(4): 869–78.
28. Tanne D, Freimark D, Poreh A, et al. Cognitive functions in severe congestive heart failure before and after an exercise training program. Int. J. Cardiol. 2005; 103(2): 145–9.
29. Tarumi T, de Jong DL, Zhu DC, et al. Central artery stiffness, baroreflex sensitivity, and brain white matter neuronal fiber integrity in older adults. Neuroimage. 2015; 110: 162–70.
30. Tarumi T, Gonzales MM, Fallow B, et al. Cerebral/peripheral vascular reactivity and neurocognition in middle-age athletes. Med. Sci. Sports Exerc. 2015 [Epub ahead of print].
31. Tarumi T, Gonzales MM, Fallow B, et al. Central artery stiffness, neuropsychological function, and cerebral perfusion in sedentary and endurance-trained middle-aged adults. J. Hypertens. 2013; 31(12): 2400–9.
32. Tarumi T, Ayaz Khan M, Liu J, et al. Cerebral hemodynamics in normal aging: central artery stiffness, wave reflection, and pressure pulsatility. J. Cereb. Blood Flow Metab. 2014; 34(6): 971–8.
33. Tarumi T, Shah F, Tanaka H, Haley AP. Association between central elastic artery stiffness and cerebral perfusion in deep subcortical gray and white matter. Am. J. Hypertens. 2011; 24(10): 1108–13.
34. Thomas BP, Yezhuvath US, Tseng BY, et al. Life-long aerobic exercise preserved baseline cerebral blood flow but reduced vascular reactivity to CO2. J. Magn. Reson. Imaging. 2013; 38(5): 1177–83.
35. Tseng B, Gundapuneedi T, Khan MA, et al. White matter integrity in physically fit older adults. Neuroimage. 2013; 82: 510–6.
36. Tseng BY, Uh J, Rossetti HC, et al. Masters athletes exhibit larger regional brain volume and better cognitive performance than sedentary older adults. J. Magn. Reson. Imaging. 2013; 38(5): 1169–76.
37. Tzeng YC, Lucas SJ, Atkinson G, Willie CK, Ainslie PN. Fundamental relationships between arterial baroreflex sensitivity and dynamic cerebral autoregulation in humans. J. Appl. Physiol. (1985). 2010; 108(5): 1162–8.
38. Voss MW, Heo S, Prakash RS, et al. The influence of aerobic fitness on cerebral white matter integrity and cognitive function in older adults: results of a one-year exercise intervention. Hum. Brain Mapp. 2013; 34(11): 2972–85.
39. Willie CK, Macleod DB, Shaw AD, et al. Regional brain blood flow in man during acute changes in arterial blood gases. J. Physiol. 2012; 590(Pt 14): 3261–75.
40. Zhu YS, Tarumi T, Tseng BY, Palmer DM, Levine BD, Zhang R. Cerebral vasomotor reactivity during hypo- and hypercapnia in sedentary elderly and masters athletes. J. Cereb. Blood Flow Metab. 2013; 33(8): 1190–6.

age; master athletes; aerobic exercise training; cerebral blood flow; cardiovascular function; cognitive function

© 2015 American College of Sports Medicine