The Utility of High Intensity Interval Training to Improve Cognitive Aging in Heart Disease Patients

Adults with cardiovascular disease and heart failure are at higher risk of cognitive decline. Cerebral hypoperfusion appears to be a significant contributor, which can result from vascular dysfunction and impairment of cerebral blood flow regulation. In contrast, higher cardiorespiratory fitness shows protection against brain atrophy, reductions in cerebral blood flow, and cognitive decline. Given that high intensity interval training (HIIT) has been shown to be a potent stimulus for improving cardiorespiratory fitness and peripheral vascular function, its utility for improving cognitive aging is an important area of research. This article will review the physiology related to cerebral blood flow regulation and cognitive decline in adults with cardiovascular disease and heart failure, and how HIIT may provide a more optimal stimulus for improving cognitive aging in this population.


Introduction
Dementia is a debilitating disease that impairs a person's memory and cognitive functions for maintaining independence and normal daily activities. Mild cognitive impairment is a term used to characterize individuals that fall between the cognitive changes of normal aging and dementia but still have the ability to perform usual daily activities and live in an autonomous manner [1]. Globally, it is estimated that around 47 million people are living with dementia-related disease, which is expected to triple by the year 2050 [2]. Vascular dementia and the neurodegenerative condition Alzheimer's disease are the two most common causes of dementia. Although Alzheimer's disease has distinct pathology from vascular dementia (e.g., β-amyloid plaques and neurofibrillary tangles of tau protein), they frequently coexist and can simultaneously contribute to cognitive impairment [3]. Moreover, 30% of Alzheimer's disease cases have been attributed to modifiable vascular risk factors such as hypertension, obesity, diabetes, physical inactivity [4].
Cardiovascular disease (CVD) affects >400 million people worldwide [5] and 120 million people in the United States alone [6]. While the incidence of dementia increases with age, adults with CVD have a higher risk of cognitive impairment [7,8]. Furthermore, a cardiovascular event or the onset of heart failure (HF) accelerates the cognitive decline [9,10]. Reduced blood flow to the brain (i.e., cerebral hypoperfusion) appears to be a significant contributor to cognitive decline [11]. Although reductions in cerebral blood flow (CBF) occur with normal aging [12,13] the age-related decline may be exacerbated by the presence of CVD. Indeed, patients with coronary artery disease have shown reduced CBF and greater rates of brain atrophy compared with age-matched controls [14,15]. Due to medical advancements, patients with CVD are now living for many decades [16]. Therefore, the impact of CVD on the aging brain and cognitive decline is of increasing importance. disrupted regulation of CBF, and impaired protein clearance [40,41]. Adequate CBF is important for maintaining the integrity of the blood brain barrier. Cerebral hypoperfusion, and subsequently reduced oxygen and glucose to the brain, can negatively affect the endothelial and neighboring cells within the neurovascular unit [40,42]. Reduced glucose to the brain can induce excitotoxicity and cause metabolism disruption, mitochondrial dysfunction, activation of proteases/phospholipases, and production of reactive oxidative species, which collectively disrupt blood brain barrier integrity through cell membrane damage and vascular cell death [42]. Breakdown of the blood brain barrier can make neurons susceptible to damage by facilitating the entry of neurotoxic compounds and pathogens that initiate neurodegenerative pathways [43]. Reductions in cardiac function, which decreases cardiac output and systemic perfusion, may also contribute to cerebral hypoperfusion and cognitive impairment [44]. The relationship between cardiac output and middle cerebral artery blood velocity (MCAv) appears to be linear [26,45]. The extent of CBF reduction in patients with HF has been shown to correlate with disease severity [46], and increases in MCAv have been found with improvement in left ventricular ejection fraction following cardiac resynchronization therapy [47]. Although, Hammond et al. [10] found that cognitive decline was similar in HF patients with reduced versus preserved ejection fraction. In HF patients with preserved ejection fraction, the mechanisms for cognitive decline are likely related to obesity, vascular impairment, diastolic dysfunction, chronic neurohormonal activation [48].
Large artery stiffening is another link between CVD and the brain. Arterial stiffness and wave reflection (the reflection of the pressure pulse wave back to the heart), increases pulse pressure and pulsatile blood flow to the smaller vessels of the arterial tree (notably the brain) [49]. This increased pulse pressure contributes to microvascular pulsatility and hemodynamic stress within the perivascular spaces of the brain, that results in microstructural damage and impairments in β-amyloid clearance [50]. Arterial stiffness is a common cause of cerebral small vessel disease, which affect the arteries, arterioles, capillaries, and veins within the brain [51]. Two common types of cerebral small vessel disease include (1) arteriosclerosis (referred to as arterial stiffening or hypertensive small vessel disease), characterized by the loss of smooth muscle cells, narrowing of the lumen, and thickening of the vessel walls; and (2) cerebral amyloid angiopathy, characterized by the progressive accumulation of β-amyloid protein [51]. This damage to the cerebral vessels can cause ischemia, inflammation, vessel rupture, and disruption of the brain-blood barrier and neural connectivity pathways, that results in pathological brain changes such as white matter hyperintensities, lacunar infarcts, microbleeds, and macroscopic hemorrhage [51,52].
Coronary artery bypass graft surgery (CABG) has been proposed to increase risk of cognitive decline due to intraoperative hypotension, hypoxia, microembolism, and/or inflammatory processes [53]. However, a study by Sweet et al. [54] compared patients following CABG or percutaneous coronary intervention (PCI) with healthy controls, and found a similar degree of decline on neurocognitive tests for both CABG and PCI over 3-weeks, 4-months, and 12-months. Furthermore, Selnes and colleagues conducted an expert review of the topic and concluded 'the extent of pre-existing cerebrovascular and systemic vascular disease' have a greater effect than procedural variables on neurocognitive function in the short-and long-term [55].

Mechanisms for Improving Brain Health with Exercise
There is substantial evidence that exercise training is an effective way to increase cardiorespiratory fitness [56], improve vascular structure and function [38,57], and reduce traditional CVD risk factors and inflammation [58,59]. Moreover, there is accumulating evidence that moderate-vigorous exercise training improves cognitive function, particularly in areas of attention-psychomotor function, executive function, and memory [60,61]. Higher cardiorespiratory fitness, as peak oxygen uptake (peak VO 2 ), has been associated with higher cognitive function [62], attenuation of gray matter volume atrophy [31,63], greater white matter integrity [64], and higher MCAv and cerebrovascular function [62,65,66]. Furthermore, maintaining or increasing cardiorespiratory fitness over time has been shown to reduce the risk of dementia incidence and mortality [67].
While normal aging is associated with a decline in resting levels of CBF, habitual exercisers with higher cardiorespiratory fitness have been shown to have higher resting MCAv levels than age-matched sedentary counterparts [66]. Proposed mechanisms for this involve exercise-induced increases in the recruitment and/or new growth of capillaries (i.e., angiogenesis), as well as beneficial vascular remodelling from repetitive hemodynamic forces on the artery walls (i.e., shear-stress) that occurs with exercise [38,68]. A recent meta-analysis by Smith et al. [18], found no change in resting MCAv with exercise training interventions (2-7 months duration) compared with control. Although, seven of the eight included studies did not measure changes in cerebrovascular conductance index (i.e., MCAv/MAP), which would account for potential changes in arterial blood pressure with exercise training. In contrast, results from The Brain in Motion Study (n = 206 healthy middle-aged adults) [69] showed 6-months of moderate intensity aerobic training significantly improved resting MCAv and reduced cerebrovascular resistance in conjunction with improvements in cognitive functions. Furthermore, Tomoto et al. [70] showed 12-months of moderate-vigorous aerobic training increased total CBF (normalized for total brain mass) and reduced cerebral pulsatility in patients with mild cognitive impairment, which were associated with changes in peak VO 2 .
The effect of habitual exercise and cardiorespiratory fitness on cerebrovascular function has also been studied, by assessing cerebrovascular reactivity (i.e., changes in CBF or MCAv with changes in arterial CO 2 ). Similar to resting CBF or MCAv, cerebrovascular reactivity decreases with age [13], however higher levels of habitual exercise and cardiorespiratory fitness are associated with greater cerebrovascular reactivity [71,72]. Furthermore, several studies have shown aerobic exercise training (compared with control), can improve cerebrovascular reactivity in middle-aged healthy adults [73], stroke survivors [74], and breast cancer survivors (high intensity group) [75]. Concomitantly, two of these studies demonstrated substantial improvements in peak VO 2 (~19%) [74,75], and Ivey et al. [74] found a positive correlation (r = 0.55) between changes in peak VO 2 and changes in cerebrovascular reactivity. In contrast, The Brain in Motion Study [69] found no improvement in cerebrovascular reactivity with 6-months of moderate intensity aerobic training. This study showed a significant, but small improvement in peak VO 2 (~7%). Therefore, substantial improvements in cardiorespiratory fitness (as peak VO 2 ) may be necessary for eliciting changes in cerebrovascular reactivity. Although, a study by Thomas et al. [76] found cerebrovascular reactivity was reduced in elderly Masters athletes compared with elderly sedentary males. The authors suggested the lower cerebrovascular reactivity may be related to vascular desensitization from chronic exposure to higher CO 2 levels with exercise training [76]. Over a short-term period, we have found no effect of repeated hypercapnia (CO 2 ) exposures on cerebrovascular reactivity in healthy adults [77].
Higher levels of cardiorespiratory fitness have also been positively associated with volumes of gray matter and white matter in older adults, specifically the attenuation of the age-related atrophy rates in the frontal, temporal, and parietal regions [63,78]. Therefore, exercise may contribute to a greater 'reserve' or threshold against the clinical manifestations of cognitive impairment. This concept of 'reserve' emerged due to individual differences in the susceptibility of cognitive impairment as well as discontinuity between cognitive outcomes and neuropathology (e.g., β-amyloid burden & neurofibrillary tangles of Tau) [31,79]. Stern [80] discusses the concept of 'brain reserve', whereby individuals with greater brain volumes may tolerate higher levels of neuropathology without affecting cognitive function; and 'cognitive reserve' whereby cognitively normal adults tolerate a higher level of neuropathology through pre-existing cognitive processes or compensatory approaches [80]. Exercise and stimulating environments may contribute to brain and cognitive reserve through the upregulation of neurotrophic growth factors such as brain-derived neurotrophic factor and insulin-like growth factor, which promote neurogenesis (i.e., the formation and maintenance of neurons) and neuroplasticity (i.e., the improvement of brain structure and function) [81,82]. While life-long exercise and higher levels of cardiorespiratory fitness are associated with less atrophy of gray matter volumes, the evidence for exercise training increasing gray matter volumes is less convincing and may be specific to the left hippocampus region [83].

Effect of Exercise Training on Brain Health in CVD
Previous reviews that have investigated the effect of cardiac rehabilitation on cognitive function [53,84] have found promising but limited evidence for exercise-based cardiac rehabilitation on cognitive function. A literature search of MEDLINE and Scopus was recently conducted by Taylor [85] identifying fourteen exercise training studies in patients with CVD and/or HF assessing brain-related outcomes (i.e., cognitive function, CBF, cerebral artery blood velocity, cerebrovascular function, and brain imaging markers). A summary of the study characteristics and brain-related outcomes for thirteen chronic exercise training studies (≥14 days) are outlined in Table 1. One study by Anazodo et al. published their results as two manuscripts including outcomes of cognitive function and brain structure [15] and CBF [14]. Twelve of thirteen exercise training studies were performed within a cardiac rehabilitation setting. Cardiac rehabilitation is internationally recognized as a class 1A recommendation for patients following a cardiovascular-related event or procedure, which promotes intensive CVD risk factor modification and exercise training [86,87].
Taylor [85] reported a major limitation within the available studies is the lack of a control group, and therefore it cannot be determined whether improvements in cognition following a cardiac-related event would occur naturally without exercise training or could be from learning effects related to the cognitive tests. Stanek et al. [90] found improvements in attentionpsychomotor function and verbal memory exceeded those of practice effects but improvements in global and executive function were similar to practice effects. Tanne et al. [88] found significant improvements in measures of psychomotor and executive functions compared with controls that could not complete the exercise training intervention, but no differences for improvements in global function (MMSE) or other neurocognitive tests. Finally, Fujiyoshi and colleagues [95] found patients attending monthly cardiac rehabilitation over 6-months had significantly greater improvements in global cognition and executive function than patients attending less than monthly cardiac rehabilitation as a control group.          From the available studies it is difficult to determine whether intensity of the exercise interventions plays a role in improving cognitive functions. Six of the thirteen studies did not report details of exercise intensity. However, the majority of included studies reported significant improvements in exercise capacity for the exercise groups (Table 1). Furthermore, several studies reported significant correlations between change in exercise capacity and change in cognitive domains, including change in peak metabolic equivalents (METs) and attention-executive function [89], change in peak METs and verbal memory [90], and change in submaximal test METs and working memory [96]. The two studies that found no improvement in cognition, also found no significant improvements in peak VO 2 [15,94].

Brain Structure and Cerebrovascular Function
The effect of exercise training on brain structure was only measured by one study [15]. Using magnetic resonance imaging (MRI), Anazodo et al. [15] found significant bilateral improvements in gray matter volume within the frontal lobe, middle temporal gyrus, supplementary motor area during cardiac rehabilitation, which were areas that showed significant atrophy compared with healthy controls at baseline. One study investigated the effect of exercise training on CBF using MRI arterial spin labelling [14], and three studies measured the effect on cerebral artery blood velocity using transcranial doppler ultrasound (TCD) [88,90,99]. One study also used near-infrared spectroscopy as a marker of cerebral oxygenation [96]. In the same MRI study above, Anazodo et al. [14] found significant improvements in regional gray matter CBF bilaterally within the Anterior Cingulate by 30%, but no significant change in global CBF. Of note, the Anterior Cingulate lies within the medial aspect of the frontal lobe and is known to have an important role in executive functions relating to emotional control, focused problem-solving, error recognition, and adaptive responses to changing conditions [100]. Using TCD, none of the three studies [88,90,99] found a significant change in resting MCAv, and Smith et al. [99] found a decrease in resting posterior cerebral artery velocity (PCAv). However, Stanek et al. [90] found a significant improvement in resting anterior cerebral artery velocity (ACAv). Notably, the ACA branches into the pericallosal artery to supply CBF to the Anterior Cingulate [101], where Anazodo et al. [14] found a significant increase in CBF. Stanek et al. [90] also found that higher ACAv and MCAv at baseline was associated with greater improvements in visuospatial working memory.
In terms of cerebrovascular regulation, Tanne et al. [88] measured MCAv cerebrovascular reactivity with TCD using a breath-hold index and found no significant improvements in either the exercise group or control. It is intriguing that this study found improvements in cardiovascular hemodynamics (cardiac index and systemic vascular resistance) but this did not translate into improvements in MCAv or cerebrovascular reactivity in patients with HF [88]. This study also demonstrated significant improvements in exercise capacity with 6-min walk test and treadmill METs, however changes in cardiorespiratory fitness as peak VO 2 were not directly measured. Smith et al. [99] measured cerebrovascular regulation during submaximal and maximal exercise before and after 12-weeks of exercise-based cardiac rehabilitation in patients with left ventricular assist devices (LVADs). Despite a reduction in resting PCAv following exercise training, patients with LVAD achieved greater increases in PCAv during submaximal and maximal exercise, compared with before training. Similarly, following exercise training patients had higher flow through the internal carotid artery during submaximal exercise, although this did not lead to improved regulation of MCAv. Overall, there is very limited evidence available assessing the effect of exercise training on brain structure and cerebrovascular function in patients with CVD.

Is There Rationale for Higher Intensity Exercise?
The magnitude of change in cardiorespiratory fitness [89,90,96] and vascular function [102] during cardiac rehabilitation has been shown to significantly correlate with changes in cognitive function. High intensity interval training (HIIT) is a potent stimulus for improving cardiorespiratory fitness (as peak VO 2 ) and vascular function compared with moderate intensity continuous training (MICT) [103][104][105]. Importantly, the superior effects of HIIT on peak VO 2 and vascular function compared with MICT also occurs in patients with CVD [106][107][108][109], and HIIT has been shown to be safe and feasible in patients attending cardiac rehabilitation [110][111][112]. Furthermore, in patients with HF, HIIT has shown greater improvements in cardiac function compared with MICT [106,113]. Therefore, we propose that greater improvements in cardiorespiratory fitness, vascular function, and cardiac perfusion with HIIT, has potential to translate into greater adaptations for cerebrovascular health and cognitive function (see Figure 1). 70% of peak heart rate; fairly light effort) [109,[116][117][118]. High-volume HIIT protocols have demonstrated superior improvements on health outcomes compared with MICT (particularly for peak VO2) while low-volume HIIT protocols typically show comparable improvements with MICT [105,115]. A HIIT protocol can be applied to various forms of exercise based on an individual's preference and capabilities, such as walking hills, jogging, cycling, rowing ergometer, elliptical, swimming, or aerobics/dance [119]. A proposed mechanism for the superior effect of HIIT on peripheral vascular function, when compared with MICT, is that higher intensity exercise provokes greater blood flow and shear stress stimulus that allows for greater vascular adaptation and upregulation of vasodilatory prostaglandins [13,38] and nitric oxide [38,120]. While this evidence has emerged mainly from studies of the forearm artery [104], it is reasonable to infer that greater shear stress and therefore vascular adaptation, would also occur in the cerebral vasculature. Indeed, Ogoh et al. [121] showed significantly higher internal carotid artery velocity (i.e., shear stress) for HIIT compared with MICT during acute exercise and recovery. Moreover, Klein et al. [122] found significantly greater total MCAv (i.e., shear stress) during an acute bout of HIIT compared with MICT for younger and older adults. Historically our understanding of CBF changes with cycle exercise has involved an inverted U-shape pattern, whereby MCAv and internal carotid artery flow increases with exercise intensity to ~60% of maximal VO2 and then decreases toward baseline levels with higher intensities due to hyperventilation-induced hypocapnia (reduced CO2) and cerebral vasoconstriction [123]. Therefore, exercise at higher intensities seemed counterintuitive to increasing CBF and shear stress within the cerebral circulation [124]. However, combining intervals of high intensity exercise with short intervals of active recovery (i.e., HIIT) appears to lead to a greater accumulation of CBF and shear stress than MICT [121,122]. Furthermore, studies examining MCAv during rowing exercise [125] and running [126] have shown that MCAv may not exhibit the same inverted U-shape pattern as cycling exercise. With running, Furlong et al. [126] showed a continual increase in MCAv across the full range of intensities up to 95% VO2max, compared with the usual inverted U-shape pattern for cycling (peaking at 65% VO2max). Faull et al. [125] found similar patterns of MCAv response with increasing intensities of rowing exercise, although this was not compared with cycling exercise. Taken together, these studies suggest acute bouts of higher intensity exercise provide a greater dose of shear stress, that may promote greater cerebrovascular adaptations with exercise training. HIIT involves alternating bouts of high intensity exercise interspersed with bouts of active or passive recovery. Various protocols of HIIT have been studied in clinical and healthy populations, ranging from short-duration HIIT protocols (work intervals < 1-min), medium-duration HIIT protocols (work intervals 1-3 min), and long-duration HIIT protocols (work intervals ≥ 3 min) [114,115]. Furthermore, HIIT protocols accumulating ≥ 15-min of high intensity exercise in total have been defined as high-volume HIIT [105]. HIIT should not be confused with sprint interval training, which instead involves brief maximal or supramaximal efforts (≥100% peak oxygen uptake or work capacity) typically with shorter work intervals than HIIT and a greater proportion of recovery time [103]. A well-known example of a high-volume HIIT protocol is the 4x4 Norwegian protocol involving four bouts of 4-min high intensity intervals (at~85-95% of peak heart rate; hard to very hard effort) interspersed with 3-min lower intensity recovery intervals (at~60-70% of peak heart rate; fairly light effort) [109,[116][117][118]. High-volume HIIT protocols have demonstrated superior improvements on health outcomes compared with MICT (particularly for peak VO 2 ) while low-volume HIIT protocols typically show comparable improvements with MICT [105,115]. A HIIT protocol can be applied to various forms of exercise based on an individual's preference and capabilities, such as walking hills, jogging, cycling, rowing ergometer, elliptical, swimming, or aerobics/dance [119].
A proposed mechanism for the superior effect of HIIT on peripheral vascular function, when compared with MICT, is that higher intensity exercise provokes greater blood flow and shear stress stimulus that allows for greater vascular adaptation and upregulation of vasodilatory prostaglandins [13,38] and nitric oxide [38,120]. While this evidence has emerged mainly from studies of the forearm artery [104], it is reasonable to infer that greater shear stress and therefore vascular adaptation, would also occur in the cerebral vasculature. Indeed, Ogoh et al. [121] showed significantly higher internal carotid artery velocity (i.e., shear stress) for HIIT compared with MICT during acute exercise and recovery. Moreover, Klein et al. [122] found significantly greater total MCAv (i.e., shear stress) during an acute bout of HIIT compared with MICT for younger and older adults. Historically our understanding of CBF changes with cycle exercise has involved an inverted U-shape pattern, whereby MCAv and internal carotid artery flow increases with exercise intensity tõ 60% of maximal VO 2 and then decreases toward baseline levels with higher intensities due to hyperventilation-induced hypocapnia (reduced CO 2 ) and cerebral vasoconstriction [123]. Therefore, exercise at higher intensities seemed counterintuitive to increasing CBF and shear stress within the cerebral circulation [124]. However, combining intervals of high intensity exercise with short intervals of active recovery (i.e., HIIT) appears to lead to a greater accumulation of CBF and shear stress than MICT [121,122]. Furthermore, studies examining MCAv during rowing exercise [125] and running [126] have shown that MCAv may not exhibit the same inverted U-shape pattern as cycling exercise. With running, Furlong et al. [126] showed a continual increase in MCAv across the full range of intensities up to 95% VO 2 max, compared with the usual inverted U-shape pattern for cycling (peaking at 65% VO 2 max). Faull et al. [125] found similar patterns of MCAv response with increasing intensities of rowing exercise, although this was not compared with cycling exercise. Taken together, these studies suggest acute bouts of higher intensity exercise provide a greater dose of shear stress, that may promote greater cerebrovascular adaptations with exercise training.
There are no published studies examining the effect of HIIT on cerebrovascular outcomes in patients with CVD. In non-cardiac populations, Whittaker et al. [127] conducted a systematic review on the acute and chronic effects of HIIT on cerebrovascular outcomes. Only three studies assessed the chronic effect of HIIT on cerebrovascular outcomes, including cerebral oxygenation using NIRS in healthy older adults [128], dynamic cerebral autoregulation in endurance training males [129], and cerebrovascular reactivity in breast cancer survivors [75]. Overall, none of the studies found a significant interaction effect between time and exercise intensity on any cerebrovascular outcomes, however it should be highlighted that all studies were small and not adequately powered to detect significant differences between groups. The pilot study by Northey et al. [75] in breast cancer survivors showed that 12-weeks of HIIT resulted in moderate-large positive effects on resting MCAv (effect size (ES) = 0.86) and MCAv cerebrovascular reactivity (ES = 0.72) compared with control, and a moderate effect on MCAv cerebrovascular reactivity (ES = 0.54) compared with MICT. Drapeau et al. [129] found a subtle reduction in dynamic cerebral autoregulation after 6-weeks of HIIT in endurance training athletes, although there was no comparison to a moderate intensity group or control group.
To date, only one study by Lee et al. [94] has compared the effect of HIIT and MICT on cognitive function during cardiac rehabilitation and found no significant improvement in either group for any cognitive domains (Table 1). Although, recruitment and retention challenges resulted in a small sample size (n = 7 per group) which was underpowered to detect differences between groups. This study also found no significant group differences for peak VO 2 improvement between HIIT and MICT despite using a highvolume 4 × 4-min HIIT protocol (i.e., greater dose than MICT); and the improvement in peak VO 2 with HIIT (1.3 mL/kg/min) was 3-fold lower in comparison to the peak VO 2 improvements with HIIT in other studies involving patients with coronary artery disease (mean = 4.6 mL/kg/min) [106]. A reason for this may be that most of the exercise training was performed in a home-based setting with one supervised exercise session per week. Several small studies have compared HIIT and MICT on cognitive function in non-cardiac populations. A study in older adults [130] found improvement in working memory with 12-weeks of HIIT but not MICT. In younger adults [131], 6-weeks of HIIT but not MICT showed improvement in executive function response time. The pilot study by Northey et al. [75] comparing HIIT and MICT with no-exercise control in breast cancer survivors, found moderate-large positive effects for HIIT in comparison to MICT for executive function (ES = 0.55) and working memory (ES = 1.41) and in comparison to no-exercise control for episodic working (ES = 0.76), executive function (ES = 0.75), and working memory (ES = 0.81). Of note, the improvement in peak VO 2 significantly increased with HIIT (19%) but not MICT (6%) or control (3%), which supports the notion that the stimulus of the intervention on cardiorespiratory fitness is likely important for changes in cognitive functions and cerebrovascular outcomes. More studies are needed comparing exercise of different intensities on measures of cognitive function, cerebrovascular outcomes, and brain structure.

Conclusions
Patients with CVD and HF are at increased risk of cognitive decline. Exercise training, predominantly during cardiac rehabilitation, consistently shows improvements in global and attention-psychomotor cognitive functions, but the effect on executive function and memory is less clear. Although, an important caveat is the lack of control groups and therefore cannot be definitive that improvements did not occur from practice effects or naturally with time following a cardiac event. However, the degree of cardiorespiratory fitness improvement with exercise training appears to be an important mediator (or marker of adequate training stimulus) for changes in cognitive function and cerebrovascular outcomes in healthy and clinical populations. Since HIIT is a potent stimulus for improving cardiorespiratory fitness, this supports the hypothesis that HIIT may also promote superior improvements in cognitive function and cerebrovascular outcomes. Furthermore, the greater accumulation of CBF and shear stress during HIIT compared to MICT, may allow for greater vascular adaptations and improvements in cerebrovascular outcomes and cognitive function. Currently, there is a very limited amount of research investigating the influence of exercise training intensity on cognitive function and cerebrovascular outcomes, and there are no studies in CVD or HF populations. Well-designed exercise training studies are warranted to investigate the effect of HIIT on cognitive function and cerebrovascular outcomes in patients with CVD and HF; and determine the optimal exercise prescription for improving brain health in this high-risk population.