Identification of novel vascular risk factors is necessary to understand vascular contributions to the progression of cognitive decline and dementia (Gorelick et al., 2011). Specifically, investigating mechanisms underlying the reduction in cerebral blood flow and changes in blood pressure (BP) during the presymptomatic phase of dementia could help to elucidate the link between vascular risk and dementia (Snyder et al., 2015). Impaired peripheral and cerebral vascular regulation as a consequence of chronic hypertension could contribute to the reduction in cerebral blood flow and progression of cognitive decline (Farkas and Luiten, 2001). Indeed, vascular dysregulation and impaired hemodynamic responses to stress may occur early in the progression of cognitive decline, prior to clinically-relevant declines in cognitive function (Iadecola, 2004; Iturria-Medina et al., 2016). In fact, vascular risk scores have been used to predict future risk of dementia 20 years later in middle-aged adults (Kivipelto et al., 2006). Further, in individuals with Alzheimer’s disease (AD), vascular factors including systolic hypertension and angina at the time of AD diagnosis are associated with a more rapid decline in cognitive function (Mielke et al., 2007).

White matter hyperintensities (WMH) are areas of increased brightness and signal density observed on MRI scans indicative of small vessel cerebrovascular disease which are associated with cognitive decline and increased severity of dementia in individuals with AD (Heyman et al., 1998). Longitudinal studies suggest an association between WMH burden and cardiovascular risk factors (Breteler et al., 1994; Jeerakathil et al., 2004), lower total cerebral perfusion (Vernooij et al., 2008), and lower cerebral blood flow (Promjunyakul et al., 2018). Indeed, cerebral blood flow is lower in areas with WMH compared to areas without WMH (Brickman et al., 2009; Stewart et al., 2021). Cerebral blood flow is also lower in adults with WMH compared to those without (Fazekas et al., 1988; O’Sullivan et al., 2002). In addition, lower middle cerebral artery blood velocity (MCAv) and higher cerebral pulsatility index (PI) has been reported in adults with greater severity of WMH burden (Puglisi et al., 2018). Importantly, during the presymptomatic phase of dementia, WMH volumes are associated with increased risk of progression to MCI (Soldan et al., 2020).

Longitudinal studies have demonstrated that elevated BP and systolic blood pressure (SBP) variability in midlife increases the risk of developing WMH (Dufouil et al., 2001; Havlik et al., 2002; Wartolowska and Webb, 2021). In addition to vascular risk factors, the physiological response to repeated acute or chronic physiological and psychological stressors likely contribute to future risk of disease (McEwen and Stellar, 1993). The response to acute physiological stressors may therefore reveal dysfunction in the systemic and cerebral circulation that could affect white matter health. For example, a greater increase in BP in response to mental stress is associated with WMH burden (Waldstein et al., 2004) and impaired hypercapnic cerebrovascular reactivity is associated with white matter damage (Sam et al., 2016). Isometric handgrip (IHG) exercise represents an acute sympathoexcitatory stressor indicative of activities of daily living as heart rate (HR), BP, and sympathetic nerve activity increase in response to sustained isometric contractions (Lind and McNicol, 1967; Mark et al., 1985). Post-exercise ischemia occludes blood flow to the previously contracted muscle leading to sustained elevations in BP and sympathetic activity, in the absence of muscle contraction (Rowell and O’Leary, 1990). Together, IHG exercise and post-exercise ischemia provide insight into the cardiovascular responses to muscle mechanoreflex and metaboreflex activation. Isometric and rhythmic handgrip exercise responses have been used to evaluate cardiovascular risk in a variety of populations (Ide et al., 1998; Giller et al., 2000; Ranadive et al., 2017; Miller et al., 2019; Tarumi et al., 2021). We have previously reported that women with a history of preeclampsia (who have an elevated risk of cerebrovascular disease and cognitive decline) demonstrated a greater cerebral blood flow velocity response to IHG exercise (Miller et al., 2019), despite similar increases in BP (Ranadive et al., 2017), compared with women without a history of preeclampsia. However, the cardiovascular and cerebrovascular response to IHG exercise and post-exercise ischemia have not been linked to white matter health. The purpose of this study was to evaluate associations between cardiovascular and cerebrovascular responses to sympathoexcitatory stimuli and WMH burden in a cohort of cognitively unimpaired middle-aged and older adults. We hypothesized that both a greater cardiovascular and cerebrovascular response to IHG exercise and post-exercise ischemia would be associated with greater WMH burden.

The primary aim of this study was to evaluate associations between cardiovascular and cerebrovascular responses to sympathoexcitatory stimuli and WMH burden. The novel finding of this study was that a lower percent change in HR, BP, and CVRi in response to IHG exercise was associated with greater WMH burden in middle-aged and older adults, which was opposite of our original hypothesis. These results persisted when accounting for potentially confounding variables including age at MRI, sex, APOEε4 status, and work performed during the IHG exercise protocol. Our results indicate that individuals with greater WMH burden may have impaired cardiovascular and cerebrovascular responses to sympathoexcitatory stress. Together, our findings suggest that blunted HR, BP and CVRi in response to acute sympathoexcitatory stress are associated with greater WMH burden in cognitively unimpaired middle-aged and older adults.

We have previously reported a 10-14% increase in HR, 7-20% increase in MAP, and 9-13% increase in MCAv in response to IHG exercise at 30% of MVC until fatigue in postmenopausal women with or without a history of hypertensive pregnancy (Ranadive et al., 2017; Miller et al., 2019). Consistent with our previous work and others, we observed similar sustained increases in cardiovascular (Lewis et al., 1983; Sander et al., 2010; Ranadive et al., 2017; Lee et al., 2021) and cerebrovascular variables (Miller et al., 2019) during IHG exercise and post-exercise ischemia. A unique aspect of this study is the inclusion of cerebral hemodynamics (MCAv and CVRi) to assist in interpretation of hemodynamic changes in the cerebral vessels in response to two sympathoexcitatory stressors. In response to IHG exercise at 40% MVC, we observed a 25% increase in HR, a 27% increase in MAP, a 5% increase in MCAv, and a 22% increase in CVRi during the final tertile of IHG exercise. Compared to our previous work, the larger increases in HR, MAP, SBP, and DBP observed in this study are likely due to the inclusion of both men and women as well as a greater relative workload (40 vs. 30% MVC). Our findings build upon our previous work in this area by including a larger sample size of both sexes and evaluating associations between cardiovascular and cerebrovascular responses to acute sympathoexcitatory stimuli and WMH. To our knowledge, this is the first study to examine associations between cardiovascular and cerebrovascular responses to sympathoexcitatory stimuli and WMH burden. All participants in this study were enrolled in Wisconsin ADRC cohorts. Participants enrolled in these cohorts are part of a larger longitudinal study of over 1,200 individuals focused on early detection of AD, identification of both protective and risk factors, and developing strategies to delay onset and progression of AD (Johnson et al., 2017). The participants recruited for this study were relatively young and were healthy, free of cardiovascular disease (other than controlled hypertension n = 13), dementia of any kind, and history of clinically significant stroke, cerebrovascular disease, or other major neurological disorders. Therefore, collecting physiological data in this cohort provides insight beyond aging alone. As a result, the stressors used in this study in cognitively unimpaired middle-aged and older adults have the potential to serve as useful tools for early detection and identification of risk factors associated with cognitive decline and dementia.

The relationship between elevated BP at rest and WMH burden has been previously reported (van Dijk et al., 2004) and elevated BP and BP variability during the middle-aged years are associated with increased WMH burden (Dufouil et al., 2001; Havlik et al., 2002; Wartolowska and Webb, 2021). Elevated BP at midlife is also associated with increased risk of cognitive decline (Swan et al., 1998; DeCarli et al., 2001), dementia (McGrath et al., 2017; Walker et al., 2019), and AD (Lennon et al., 2019; Ou et al., 2020). However, epidemiological and longitudinal studies only provide insight into associations between BP values at rest and WMH burden or retrospective analysis based on historical BP measurements. Cardiovascular and cerebrovascular responses to acute or chronic physiological stimuli in healthy adults may reveal dysfunction in the systemic and cerebral circulation prior to significant damage to or changes in brain volumes. Indeed, greater BP responses to mental stress are associated with poor performance on cognitive challenges (Waldstein and Katzel, 2005) and WMH burden (Waldstein et al., 2004). Acceleration of WMH burden occurs prior to presentation of MCI (Silbert et al., 2012) and midlife represents a unique period during the presymptomatic phase of dementia in which intervention may be beneficial to delay the onset of cognitive decline. As such, identifying potential mechanisms contributing to the increase in WMH burden during midlife may be important for understanding the progression from normal cognition to presentation of symptoms of cognitive decline.

Isometric handgrip exercise followed by a period of post-exercise ischemia represent acute physiological stimuli that elicit a marked increase in HR, BP, and sympathetic nervous system activity. In agreement with our original hypothesis, the expected response to a sympathoexcitatory stimulus is a substantial increase in HR, BP, and CVRi. Indeed, we observed a 25% increase in HR, 27% increase in MAP, 25% increase in SBP, 26% increase in DBP, and a 22% increase in CVRi during the final tertile of IHG exercise. Moreover, each of these variables remained elevated during post-exercise ischemia. Contrary to our hypothesis, however, a lower percent change in HR, MAP, SBP, DBP, and CVRi in response to IHG exercise was associated with greater WMH burden in middle-aged and older adults. One explanation for these findings is that failure to increase resistance in the cerebral circulation (i.e., CVRi) may lead to propagation of highly pulsatile blood flow into the delicate microcirculation suggesting that individuals with greater WMH burden may have impaired regulation of cerebral blood flow. Along these lines, following an acute hypertensive stimulus (a single bout of resistance training), older adults demonstrated a greater increase in cerebral PI despite no change in mean MCAv compared with young adults (Rosenberg et al., 2020). These results suggest that cerebral blood flow regulation may be impaired following an acute hypertensive stimulus in older adults, which may lead to greater transmission of pulsatile flow into the microcirculation and increased risk of end-organ damage.

Consistent with this idea, carotid artery PP, carotid PI, cerebral PI, and aortic stiffness are associated with increased risk for silent subcortical infarcts (Mitchell et al., 2011). Additionally, aortic stiffness is associated with higher WMH volume and carotid PI is associated with lower gray and white matter volumes (Mitchell et al., 2011). In the cerebral circulation, cerebral PI is positively associated with greater WMH volume in middle-aged and older adults (Tarumi et al., 2014). These findings suggest that aortic stiffness and transmission of pulsatile flow into the microcirculation may lead to quantifiable changes in brain volume and increased WMH burden. A recent study conducted by Tarumi et al. in young adults investigated cardiovascular variables and cerebral blood flow in response to repeated bouts of rhythmic handgrip exercise using phase-contrast MRI (Tarumi et al., 2021). Using a similar relative intensity of handgrip exercise (30-40% MVC), HR, BP, CVRi, cerebral blood flow, and respiratory rate increased during rhythmic handgrip exercise despite no change in vessel cross-sectional area (Tarumi et al., 2021). Thus, higher resistance in the cerebral vessels may attenuate an increase in cerebral blood flow and prevent cerebral hyperperfusion during rhythmic handgrip exercise, which may represent a compensatory myogenic response to a hypertensive stimulus in order to dampen pulsatile flow. Failure to increase resistance in the cerebral vessels (indicated by a smaller increase in CVRi) during an acute sympathoexcitatory stimulus suggests a potential lack of active vasoconstriction occurring in the cerebral circulation that may allow for propagation of pulsatile flow into the microcirculation. Broadly, acute instances of physiological stressors that affect BP and cerebral blood flow, experienced across the lifespan, may affect white matter health. We report that individuals who demonstrated greater increases in HR, BP, and CVRi in response to IHG exercise had lower WMH burden, which was opposite of our hypothesis. It is possible that increases in CVRi during a sympathoexcitatory stress, indicating active vasoconstriction, may protect the cerebral circulation from increases in pulsatile flow. For example, in spontaneously hypertensive rats, smooth muscle hypertrophy in the cerebral arteries occurs, leading to increased cerebrovascular resistance (Tayebati et al., 2012). Changes in structure and function in the cerebral circulation may protect the brain from high perfusion pressures during periods of elevated BP and augmented cerebral blood flow (Faraci and Heistad, 1990; Tayebati et al., 2012). Alternatively, individuals with greater WMH burden or reduced white matter integrity (which was not measured in the present study) may have impaired cardiovascular responses to sympathoexcitatory stress. However, this is unlikely as WMH volumes in the present study are low, likely due to the average age of participants in this study being only ∼63 years.

In agreement with the findings from Tarumi et al., we report significant increases in HR, BP, MCAv, and CVRi during IHG exercise in middle-aged and older adults. Yet, the cardiovascular and cerebrovascular responses were highly variable. While IHG exercise was performed at the same relative intensity for all participants, we observed a wide range of responses in HR (3-63% increase), MAP (8-57% increase), MCAv (−20 to + 23% change), and CVRi (−4 to + 80% change) during the final tertile of IHG exercise. In agreement with our findings, the percent change in MAP and MCAv in response to rhythmic handgrip exercise appears to be heterogeneous (Giller et al., 2000) whereby MCAv may decrease in some individuals. Our findings indicate that middle-aged and older adults demonstrate varying responses to an acute sympathoexcitatory stimulus performed at the same relative intensity. IHG exercise represents an acute sympathoexcitatory stimulus that may be comparable to carrying an object for an extended period of time. As these types of activities are performed frequently throughout activities of daily living, evaluating the cardiovascular and cerebrovascular response to repeated bouts of acute sympathoexcitatory stress may be important for understanding changes in the brain prior to presentation of symptoms of cognitive decline. Altered expected responses to acute physiological stressors such as IHG exercise, which induce brief, large increases in BP, may be associated with measurable changes in brain volume and WMH.

Hypertension is associated with increased risk of developing WMH (Dufouil et al., 2001). Although we may be underpowered to adequately control for controlled hypertension (n = 13, 19% of participants) in our population of middle-aged and older adults, associations with WMH fraction were evaluated with controlled hypertension included in the linear model (in addition to age at MRI, sex, APOE ε4 status, and work performed during the IHG exercise protocol). After adjusting for controlled hypertension, negative associations between the percent change in HR, MAP, SBP, and DBP during IHG exercise and WMH fraction remained significant. Yet, there was no longer an association between the percent change in CVRi during IHG exercise and WMH fraction. Although the individuals with controlled hypertension were evenly distributed throughout the entire group for the variables of interest, previous work has suggested that older hypertensive adults demonstrate greater increases in raw MAP values and sympathetic nerve activity in response to IHG exercise performed at 30 and 40% MVC compared with normotensive older adults (Delaney et al., 2010). Additionally, older hypertensive adults exhibit a more rapid increase in MAP and sympathetic nerve activity (within the first 10s of muscular contraction) compared with normotensive older adults (Greaney et al., 2015). In the aforementioned studies, older adults in the hypertensive group currently taking antihypertensive medication (80% of participants) were instructed to refrain from taking medication for two days prior to the experimental study (Delaney et al., 2010; Greaney et al., 2015). In the present study, there were no differences in the change in raw MAP and CVRi values between adults with controlled hypertension and those who were normotensive (P > 0.05 for both), although the aims of the present study did not include evaluating the effects of controlled hypertension. One possible explanation for the discrepancy between previous studies (Delaney et al., 2010; Greaney et al., 2015) and our findings is that individuals with uncontrolled hypertension were excluded from our study. Additionally, participants in the present study were instructed to take prescription medication on the day of the study, which could have influenced the response to IHG and post-exercise ischemia. Nevertheless, when controlled hypertension was added to the linear model, the inverse association between the percent change in HR, MAP, SBP, and DBP during IHG exercise and WMH fraction remained significant (P < 0.05 for all; Supplementary Table 1). These results suggest that regardless of controlled hypertension status, an attenuated increase in HR and BP in response to IHG exercise is associated with greater WMH burden in middle-aged and older adults.

In order to determine CVRi, we measured the cerebral hemodynamic response to IHG exercise in the MCA using TCD. A major assumption of TCD is that the MCA diameter remains constant at rest and during a stimulus. Previous work evaluating MCAv and MCA diameter in response to rhythmic handgrip exercise has demonstrated that MCA diameter may decrease by ∼ 2% in adults aged 20-59 years old suggesting active vasoconstriction in the cerebral circulation during rhythmic handgrip exercise (Verbree et al., 2017). However, the cross sectional area of the internal carotid and vertebral arteries do not change in response to rhythmic handgrip exercise performed at 30-40% MVC in young adults (Tarumi et al., 2021). It is possible that IHG exercise causes vasoconstriction of the MCA which may cause an increase in MCAv despite little or no change in cerebral blood flow (Giller et al., 2000), though the effect of isometric versus rhythmic handgrip exercise on MCA diameter is currently unknown. While the present study evaluated MCAv in response to IHG exercise in a robust sample size of middle-aged and older adults, the cerebrovascular response to IHG exercise and potential associations with WMH burden should be interpreted with caution. Additionally, all participants performed the IHG exercise protocol using their left hand, regardless of handedness. Due to the experimental setup, we evaluated MCAv in the left (ipsilateral) MCA. When MCAv was evaluated in both the right and left MCA during rhythmic handgrip exercise in adults aged 21-43 years old, previous studies have shown that MCAv increased in the contralateral side only (Jørgensen et al., 1993; Linkis et al., 1995). In the present study, we report a significant, albeit variable, increase in MCAv in the ipsilateral MCA during IHG exercise in adults aged 55-69 years old, which we interpret as a global response. It is possible that we could have observed a greater increase in MCAv in the contralateral (right) MCA during IHG exercise, and follow-up studies should perform bilateral MCA assessments. Lastly, arterial CO₂ is a powerful regulator of cerebral vascular tone (Brian, 1998) and we utilized ETCO₂ as a surrogate measure of arterial CO₂. A reduction in ETCO₂ may cause cerebral vasoconstriction and potentially a decrease in cerebral blood flow. We observed a small (1-2 mmHg), but significant, reduction in ETCO₂ during the final tertile of IHG exercise which persisted during post-exercise ischemia. Similarly, reductions in ETCO₂ have also been observed during isometric (Muza et al., 1983; Fontana et al., 1993), and rhythmic handgrip exercise (Jørgensen et al., 1993; Tarumi et al., 2021). It is possible that a 1-2 mmHg decrease in ETCO₂ could have resulted in a decrease in MCAv in this study.

In summary, IHG exercise performed at 40% MVC until fatigue elicited significant increases in HR, BP, MCAv, and CVRi in cognitively unimpaired middle-aged and older adults. Individuals with greater WMH burden demonstrated a lower percent change in HR, BP, and CVRi in response to IHG exercise. Together, these results indicate an inverse association between WMH burden and cardiovascular and cerebrovascular responses to a sympathoexcitatory stimulus.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by University of Wisconsin-Madison Institutional Review Board. The patients/participants provided their written informed consent to participate in this study.

JB conceived and designed the research. AP, KM, AC, NE, AH, and JB performed the experiments. AP analyzed the data. AP, KC, and JB interpreted the results of experiments. AP prepared the figures. AP and JB drafted the manuscript. All authors edited and revised the manuscript, and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.