Matched increases in cerebral artery shear stress, irrespective of stimulus, induce similar changes in extra-cranial arterial diameter in humans

Abstract

The mechanistic role of arterial shear stress in the regulation of cerebrovascular responses to physiological stimuli (exercise and hypercapnia) is poorly understood. We hypothesised that, if shear stress is a key regulator of arterial dilation, then matched increases in shear, induced by distinct physiological stimuli, would trigger similar dilation of the large extra-cranial arteries. Participants (n = 10) participated in three 30-min experimental interventions, each separated by ≥48 h: (1) mild-hypercapnia (F_ICO₂:∼0.045); (2) submaximal cycling (EX; 60%HRreserve); or (3) resting (time-matched control, CTRL). Blood flow, diameter, and shear rate were assessed (via Duplex ultrasound) in the internal carotid and vertebral arteries (ICA, VA) at baseline, during and following the interventions. Hypercapnia and EX produced similar elevations in blood flow and shear rate through the ICA and VA (p < 0.001), which were both greater than CTRL. Vasodilation of ICA and VA diameter in response to hypercapnia (5.3 ± 0.8 and 4.4 ± 2.0%) and EX (4.7 ± 0.7 and 4.7 ± 2.2%) were similar, and greater than CTRL (p < 0.001). Our findings indicate that matched levels of shear, irrespective of their driving stimulus, induce similar extra-cranial artery dilation. We demonstrate, for the first time in humans, an important mechanistic role for the endothelium in regulating cerebrovascular response to common physiological stimuli in vivo.

Introduction

In conduit arteries such as the coronary,¹ brachial,² radial,³ and femoral,^4,5 it is well established that intra-arterial shear stress regulates endothelium-dependent vasoreactivity. It is also well known that episodic increases in shear can induce anti-atherogenic adaptation in paracrine transduction pathways, including the nitric oxide (NO)-dilator system (reviewed in Green et al.⁶). Shear-mediated stimulation of the endothelium is a key mechanism responsible for vasodilation through these conduit arteries in humans.⁶

Exercise increases blood flow to the brain in humans.⁷ Traditionally, the explanation for this has been attributed to a combination of physiological stimuli, including changes in arterial blood gases, blood pressure, and metabolism.^8,9 Another possible mechanism, however, involves changes in arterial shear stress and consequent endothelium-mediated dilation.¹⁰ Recently, we demonstrated vasodilation of extra-cranial feed arteries as a result of hypercapnia induced increases in brain blood flow and shear stress in vivo.^11,12 However, the impact of shear stress on cerebrovascular responses during exercise has not previously been investigated.

In the current study, we propose that shear stress is involved in regulating cerebrovascular function during exercise in humans. We therefore examined the impact of matched elevations in arterial shear stress, induced by distinct and common physiological mechanisms: exercise or hypercapnia. We examined the hypothesis that similar patterns of dilation of the large extra-cranial arteries would be observed irrespective of the different stimuli induced by exercise or hypercapnia.

Methods

Participants

Ten healthy male participants (age, 25 ± 6 years; weight, 74.14 ± 7.03 kg; height, 2.11 ± 0.21 m) were recruited for the study. All participants were non-smokers, with a body mass index <30, free of cardiovascular, respiratory, cerebrovascular, musculoskeletal and/or metabolic diseases. The study was approved by the University of Western Australia’s Human Research Ethics Committee and was in ethical accordance with the principles established in the declaration of Helsinki. The participants were informed of all experimental procedures and associated risk. Participants provided written informed consent before the commencement of the study.

Experimental design

After inclusion, participants were tested at the Cardiovascular Research Laboratory on three occasions separated by ≥48 h at the same time of day. Each visit consisted of one of three randomly assigned interventional protocols (Control [CTRL], Hypercapnia [CO₂], or Exercise [EX]). Subjects arrived after fasting for a minimum of 6 h, with 24-h abstinence from alcohol, caffeine and vigorous exercise. Upon arrival, participants were instrumented and underwent a 10-min rest period in a semi-recumbent position. Baseline (BL) recordings of the primary outcome measures were then collected during a further 5-min period of quiet rest. Following baseline measures, subjects participated in one of three interventions: (1) 30 min of hypercapnia (see below); (2) 30 min of exercise at ∼60% heart rate (HR) reserve; or (3) 30 min of rest (time-control day). Cerebrovascular (transcranial Doppler and duplex vascular ultrasound) and cardiorespiratory assessments were collected throughout each condition.

Experimental procedures

Cerebrovascular assessments

Non-invasive insonation via 2 MHz transcranial Doppler ultrasound (TCD; Spencer Technologies, Seattle, WA, USA) was used to assess blood velocity in the middle (MCAv) and posterior (PCAv) cerebral arteries. The cerebral arteries were identified and optimised according to their signal depth, waveform and velocities, according to previously published guidelines.¹³ The MCAv and PCAv were continuously recorded at baseline, throughout the 30 min of each intervention, and during post-intervention assessments.

Blood velocity and diameter of the internal carotid artery (ICA) and vertebral (VA) arteries (VA) were measured using a 10–15 MHz multi-frequency linear array vascular ultrasound (Terason T3200, Teratech, Burlington, MA, USA).¹⁴ Whilst ICA recordings were captured at 10, 20 and 30 min of the intervention, the VA recording was captured at 15 and 25 min of the intervention. Ultrasound assessments were performed by one scanner and required approximately 5 min to achieve a single scan. Thus, a standardised timing of ICA and VA collection was applied to ensure similar time-points were collected throughout each experimental condition. Baseline and post-intervention recordings (5–10 min post) were collected for both the ICA and VA. All of the ICA and VA recordings were screen captured and stored as video files for offline analysis.¹⁵

Cardiorespiratory measurements

All cardiorespiratory variables were sampled continuously throughout the protocol at 1000 Hz via an analogue-to-digital converter (Powerlab, 16/30; ADInstruments, Colorado Springs, CO, USA). Partial pressure of oxygen (P_ETO₂) and carbon dioxide (P_ETCO₂) were assessed using a gas analyser (ADInstruments, Colorado Springs, CO, USA), heart rate (HR) was measured by a 3-lead electrocardiogram (ECG; ADI bioamp ML132), and beat-to-beat blood pressure by finger photoplethysmography (Finometer PRO, Finapres Medical Systems, Amsterdam, the Netherlands).

Experimental interventions

Hypercapnia (CO₂) condition

Hypercapnia was used as a stimulus to elevate cerebral blood flow (CBF) and shear stress for 30 min via breathing an air mixture (4.5% CO₂, 21% O₂, and balanced N₂) through a spirometer connected to a Douglas bag. The participants were seated throughout in a semi-recumbent position. Simultaneous assessment of intracranial velocity (measured by transcranial Doppler of the left MCA and right PCA) and beat-by-beat extracranial blood flow (measurement by Duplex ultrasonography of the ICA and VA) was carried out along with beat-to-beat arterial pressure (Finometer Pro, Amsterdam, Netherlands). Continuous monitoring of MCAv was used as an index of real-time change in cerebrovascular shear stress. If necessary, the concentration of inspired CO₂ was altered (range 3–6%) to achieve the desired increase in arterial shear stress (∼30% above baseline). The decision to use a 30% increase in shear stress was chosen based on pilot data indicating this to be the most reliable increase achievable in both CO₂ and EX protocols. This allowed CO₂ and EX shear to be matched irrespective of study order.

Exercise condition

Participants cycled at submaximal intensity [∼60% heart rate (HR) reserve = 100–120 bpm] for 30 min to maintain a steady increase in CBF measures throughout the exercise condition. As with the hypercapnia intervention, continuous intracranial blood velocity and extra-cranial blood flow were measured at the same time points during the intervention. The MCAv during the 30-min intervention was closely monitored to ensure that an elevated CBF and shear stress was achieved (∼30% above baseline, to match the CO₂ condition).

Time-control condition

Participants sat in a semi-recumbent position for 30 min. Continuous intracranial blood velocity and extra-cranial blood flow were measured at the same time points as the interventions above.

Statistics

Statistical and graphing analysis was performed using GraphPad PRISM 6.01 software (GraphPad Software, LaJolla, CA, USA). All parameters were compared within-subjects using two-way ANOVA repeated-measures. Bonferroni-correction for multiple comparisons was used for all post-hoc analysis. Statistical significance was assumed at p < 0.05. All data are reported as mean ± SD unless otherwise specified.

Results

All cardiorespiratory and cerebrovascular values, for the anterior (ICA, MCA) and posterior (VA, PCA) cerebral arteries, are provided in Tables 1 and 2.

Table 1.

Anterior cerebral blood flow and cardiorespiratory measurements during control (CTRL), carbon dioxide (CO₂), and exercise (EX) interventions (n = 10).

		Time (min)
Measure	Condition	BL		10		20		30		Post
		Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	INT
MAP (mmHg)	CTRL	88	11	90	8	90	8	93	8	91	9	**
	CO2	87	7.6	100*ϕ	10	104*ϕ	14	102*ϕ	14	96	11
	EX	82ϕ	11	102*ϕ	10	102*ϕ	9.5	100*ϕ	7.7	86*ѱ	15.8
HR (bpm)	CTRL	64	7	64	9	67	7	67	7	63	7	**
	CO2	66	8	70	9	74	9	70	7	63	7
	EX ϕѱ	70	8	115ϕѱ	11	115ϕѱ	7	118ϕѱ	7	80ϕѱ	11
PetO2 (mmHg)	CTRL	95.2	6.8	97.7	3.2	98.5	5	100.4	6.9	96	6.2	**
	CO2 ϕ	97.4	5	128*ϕ	2.9	130*ϕ	1.9	130 *ϕ	2.3	96	7.8
	EX ѱ	95.2	6.5	96.6ѱ	7.8	95.8ѱ	7.6	96.0ѱ	7.8	96.8	4.1
PetCO2 (mmHg)	CTRL	42.7	4	42.4	2.7	41.9	3.6	41.3	3.6	41.6	3.4	**
	CO2 ϕ	42.3	3	46.9*ϕ	2.4	46.3*ϕ	2.3	46.2*ϕ	2.2	41.3	2.5
	EX	41.6	3.1	44.7*	4.9	44.7*	4.5	44.4ϕ	5.2	40.8	2.4
QICA (ml.min−1)	CTRL	297	24.8	289	20.8	278	21.4	278	24.4	264	24.1	**
	CO2	262	18.8	368*ϕ	23.8	380*ϕ	30.5	354*ϕ	25.1	256	16.1
	EX	274	16.9	361*ϕ	19.7	368*ϕ	29.6	394*ϕ	26	282	24.1
ICA Diam (mm)	CTL	5.2	0.2	5.2	0.2	5.2	0.2	5.2	0.2	5.2	0.2	**
	CO2	5.0 ϕ	0.1	5.3*	0.1	5.3*	0.1	5.4* ϕ	0.1	5.2	0.1
	EX	5.1	0.2	5.2	0.2	5.4*	0.2	5.4* ϕ	0.2	5.3	0.2
ICA Vel (cm.s−1)	CTRL	41.1	1.5	40.8	1.2	39.7	1.7	39.6	1.6	38.3	2.2	**
	CO2	39	2.1	49.5*ϕ	2.8	49.7*ϕ	2.7	49.3*ϕ	2.5	36.9	1.4
	EX	38	1.6	47.5*ϕ	1.5	47.4*ϕ	2.3	49.8*ϕ	2.5	39.7	1.9
ICA shear	CTRL	316	11.3	314	12.1	305	14.2	304	12.9	294	16.6	**
	CO2	309	15.5	372*	18.0	369*ϕ	17.4	362*ϕ	14.2	286	12.2
	EX	297	12.4	366*	14.6	354*ϕ	14.1	371*ϕ	18.0	298	17.0
MCAv (cm.s−1)	CTRL	66	3	66	3	65	3	63	3	63	3	**
	CO2	64	6	83*ϕ	7	83*ϕ	7	82*ϕ	7	68	5
	EX	58	3	75*ϕ	4	75*ϕ	5	74*ϕ	5	60	3
CVC (ml.min−1/mmHg)	CTRL	3.4	0.3	3.2	0.3	3.1	0.3	3.0	0.3	2.9	0.3	NS
	CO2	3.0	0.2	3.7	0.2	3.7	0.2	3.5	0.3	2.7	0.1
	EX	3.5	0.3	3.5	0.2	3.6	0.3	3.9	0.2	3.5	0.4

Time: * = different from BL; p < 0.05, COND: ϕ = different from CTRL; ѱ = different from CO_2; p < 0.01, INT: ** = significant COND × TIME interaction; ϕ = different from CTRL; ѱ = different from CO₂; p < 0.001.

Table 2.

Posterior cerebral blood flow and cardiorespiratory measurements during control (CTRL), carbon dioxide (CO₂), and exercise (EX) (n = 8).

		Time (min)
Measure	Condition	BL		15		25		Post
		Mean	SD	Mean	SD	Mean	SD	Mean	SD	Int
MAP (mmHg)	CTRL	92	3	93	3	90	4	93	4	**
	CO2	94	3	102	5	106*ϕ	6	98	5
	EX	87	4	102*ϕ	3	101*ϕ	3	88	4
HR (BPM)	CTRL	63	2	65	3	65	3	64	3	**
	CO2	64	2	69	3	72*ϕ	3	65	3
	EX ѱ	68	2	114*ϕѱ	2	118*ϕѱ	2	80*ϕѱ	3
PetO2 (mmHg)	CTRL	97.5	2.4	104.7	5.3	100	1.4	99.3	1.3	**
	CO2 ϕ	97.5	2.4	129.9*ϕ	0.8	130.3*ϕ	1.3	100.1	0.9
	EX ϕ ѱ	95.9	1.7	94.1 ϕѱ	3.0	96.0ѱ	2.2	97.9	1.6
PetCO2 (mmHg)	CTRL	40.2	2.3	40.4	4.5	40.2	2.2	41	0.9	**
	CO2 ϕ	41.3	1.3	46.4*ϕ	0.8	46.1*ϕ	1	41	0.9
	EXϕ	41.4	1.1	45.7ϕ	1.9	44.5	1.5	40.1	1.1
QVA (ml.min−1)	CTRL	113	20.6	111	21.6	103	19.7	103	17.9	**
	CO2	100	13.1	141*	16.1	142*	20	95	12.9
	EX	90	16.1	140*	26.2	143*	26.1	87	12.3
VA Diam (cm)	CTL	3.9	0.1	3.9	0.2	3.9	0.2	3.9	0.1
	CO2	3.9ϕ	0.2	4.2ϕ	0.2	4.2*ϕ	0.2	4.0ϕ	0.1	**
	EX	3.7ϕ ѱ	0.2	3.9*ѱ	0.2	4.1*ѱ	0.2	3.7ϕѱ	0.1
VA Vel (cm.s−1)	CTRL	27.7	3.3	27.3	3.3	25.5	2.9	24.4*	3	**
	CO2	24.7 ϕ	1.9	31.6* ϕ	2.2	31.2*	2.2	23.4	2
	EX	24.7 ϕ	2.0	31.1* ϕ	2.6	31.5*	2.5	24.1	1.7
VA shear	CTRL	282	28	279	27	261	25	249*	26	**
	CO2	251 ϕ	14	305*ϕ	19	297*ϕ	16	237	17
	EX	268	16	322*ϕ	20	316*ϕ	20	261	16
PCAv (cm.s−1)	CTRL	49	3	47	3	47	3	46	3	**
	CO2 ϕ	51	4	63*ϕ	5	63*ϕ	4	52	4
	EX	48	4	58*ϕѱ	5	56*ϕѱ	4	45*ѱ	3
CVC (ml.min−1/mmHg)	CTRL	1.2	0.2	1.2	0.3	1.1	0.2	0.9	0.1	NS
	CO2	1.1	0.1	1.4	0.2	1.4	0.2	1.0	0.1
	EX	1.1	0.2	1.4	0.3	1.5	0.3	1.0	0.2

Time: * = different from BL; p < 0.05, COND: ϕ = different from CTRL; ѱ = different from CO_2; p < 0.01, INT: ** = significant COND × TIME interaction; ϕ = different from CTRL; ѱ = different from CO₂; p < 0.001.

Cardiorespiratory measures

No changes were observed for any of the cardiorespiratory measures assessed during the 30-min CTRL condition (Tables 1 and 2).

During the CO₂ and EX, MAP increased (p < 0.01), but returned toward baseline values post-intervention (Tables 1 and 2). The magnitude of increase in MAP was similar between the CO₂ and EX conditions, except for the post-intervention time point (EX was lower; p < 0.001). Both CO₂ and EX MAP data were significantly higher throughout the intervention period compared to the CTRL trial (Figure 1; p < 0.001).

Figure 1.

Arterial blood pressure [MAP]) and end-tidal carbon dioxide [P_ETCO₂] at baseline (BL), during (10, 15, 20, 25, 30 min) as well as post exercise during the control (CTRL), carbon dioxide (CO₂) and exercise (EX) conditions. ϕ indicates differences between CO₂ and CTRL (p < 0.001); τ indicates differences between EX and CTRL (p < 0.001).

The CO₂ and CTRL conditions did not elicit any changes in HR. In contrast, the EX condition was associated with an elevated HR compared to baseline values and by comparison to the other conditions during the intervention period. Both P_ETO₂ and P_ETCO₂ were elevated throughout the CO₂ intervention (p < 0.001), compared to baseline values and also when compared to the CTRL condition. EX had a small increase on P_ETCO₂ compared to baseline values, but not to the same extent as the CO₂ condition.

Cerebrovascular measures

No changes in either MCAv or PCAv were observed during CTRL. When all time points were considered, similar mean increases, from BL, were observed in response to CO₂ (28.1 ± 3.4 and 23.3 ± 2.8%) and EX (27.7 ± 3.8 and 19.6 ± .3.0%) in the in MCAv and PCAv, respectively (p < 0.001). All MCAv and PCAv values were greater during the EX and CO₂ conditions relative to the CTRL data (Figure 2; p < 0.001). Both the EX and CO₂ conditions stimulated similar magnitudes of increase in blood flow, blood velocity and shear rate (p < 0.001) in the ICA and VA, relative to baseline value. Furthermore, blood flow, velocity, and shear rate data collected under the CO₂ and EX conditions were elevated compared to CTRL throughout the intervention period (Figure 2; p < 0.001).

Figure 2.

Middle (MCAv) and posterior (PCAv) cerebral blood flow velocity (panels a and c), internal carotid (ICA) and vertebral (VA) artery Shear (panels b and d), pre (BL) during (10, 15, 25, 30 min, respectively) and POST (5 and 10 min, respectively) of the control (CTRL), carbon dioxide (CO₂) and exercise (EX) interventions. ϕ indicates differences between CO₂ and CTRL (p < 0.001); τ indicates differences between EX and CTRL (p < 0.001).

No changes in ICA or VA diameter were observed during CTRL. In contrast, ICA and VA diameters increased from baseline values during both CO₂ and EX (Table 1; p < 0.001). The increases in diameter, in both arteries, were higher than the changes observed in the CTRL condition at all time-points, except for the VA diameters 10 min post intervention (Figure 3). There were no differences in the changes in diameter observed when the CO₂ and EX conditions were compared.

Figure 3.

Changes in diameter in the internal carotid (ICA, panel a), and vertebral (VA, panel b) arteries during (10, 15, 20, 25 and 30 min, respectively) as well as POST (5 and 10 min, respectively) control (CTRL) carbon dioxide (CO₂) and exercise (EX) interventions. Straight line connectors indicate significance between experimental and control conditions (p < 0.001).

Linear regression

Linear regression analysis between the peak changes in diameter and shear stress indicated a significant correlation coefficient during the CO₂ (R²= 0.47; p < 0.01) and EX (R²= 0.34; p < 0.01) conditions (see Figure 4). Multiple regression analysis using changes in blood pressure as a covariate to predict the relationship between shear stress and ICA dilation during the CO₂ condition revealed an enhanced correlation coefficient (R²= 0.61; p < 0.01); however, the addition of CO₂ as a further covariate had no impact. In contrast, multiple regression analysis during the EX condition, using changes in CO₂ as a covariate revealed an enhanced correlation coefficient between the peak shear and dilation (R²= 0.53; p < 0.01). There was no additional impact of change in BP.

Figure 4.

Relationship between changes in arterial shear stress and dilation of the internal carotid artery during the carbon dioxide (CO₂; r = 0.69) and exercise (EX; r = 0.58309) conditions. Control data are included in both the CO₂ and EX figures (p < 0.01). Dotted lines represent confidence intervals.

Discussion

Our findings indicate that matched increases in cerebrovascular shear stress, irrespective of the eliciting stimulus (CO₂ or EX), induce similar increases in extra-cranial artery diameter in humans. This study therefore provides novel evidence that arterial shear stress, alongside changes in blood gases and pressure, is an important stimulus regulating cerebral conduit artery dilation in vivo. Given that repetitive episodic increases in shear stress beneficially impact arterial function and structure,⁶ our novel observation provides a mechanistic basis for linking exercise, physical activity and cerebrovascular health in humans.

It is well established that endothelium-dependent mechanisms play a key role in shear stress-mediated vasodilation of coronary and peripheral arteries in humans.^1,3–5,16 These studies have established a role for paracrine hormones such as NO and prostacyclin in vasomotor regulation, for example, in response to exercise. In animals, classic studies have established that in vivo increases in flow and shear stress through cerebral conduit arteries (e.g. basilar artery) trigger endothelial-mediated vasodilation.¹⁷ Recently, in a well-controlled mouse model, Raignault et al.¹⁸ illustrated that the cerebrovascular endothelium optimally couples shear stress to eNOS-mediated dilation under physiological pulse pressures, in contrast to static flow conditions. These findings strongly infer that changes in the diameter of cerebral arteries are responsive to a pulsatile environment and that shear stress sensitivity and consequent production of NO are optimised under in vivo conditions.¹⁸ It is also well established that carotid artery remodelling as a result of chronic changes in flow is dependent upon the presence of a functional endothelial layer,¹⁹ implicating shear- and NO-mediated mechanisms in arterial structural adaptation. Collectively these findings, in animals, support a key role for shear-mediated endothelial-dependent dilation of extra- and intra-cranial cerebral conduit arteries, although it is also true that some studies indicate a role for non-shear mediated NO-dependent (i.e. eNOS) pathways in cerebral microvascular dilation.^20–22 Few studies, however, have addressed the impact of changes in arterial shear on human cerebral vasodilator function in vivo.

Recently we characterised the time-course and response of ICA dilation to changes in shear using duplex ultrasound with high temporal resolution edge-detection and wall-tracking in healthy subjects.¹¹ We concluded that, in response to a sustained hypercapnic stimulus (5 min), dilation of the ICA occurs subsequent to marked increases in intra-arterial blood flow and shear stress. Carter et al.¹⁰ observed a significant relationship (R²= 0.46) between the peak shear response and the magnitude of ICA dilation during hypercapnic stimulus. In a subsequent follow-up experiment,¹¹ a more transient increase in P_ETCO₂ (30 s) triggered similar ICA vasodilation to those observed during a sustained 4-min stimulus. Hoiland et al.¹¹ also observed a significant relationship (R²= 0.44) between shear stress and the magnitude of ICA dilation following the transient hypercapnic bolus. Since the transient changes in P_ETCO₂ elicited vasodilation following the hypercapnic stimulus, the notion of shear-mediated dilation of the ICA was further confirmed. The current study advances these findings by illustrating that similar changes in shear stress during moderate intensity exercise – and hypercapnia – elicit similar increases in extra-cranial artery dilation of both the ICA and VA in humans. This is further corroborated by the observed significant relationship (Figure 4) between the peak changes in shear stress and the maximal dilation of the ICA during exercise (R²= 0.34) and CO₂ (R²= 0.47). Such correlations between ICA diameter and change in BP, or in P_ETCO₂, were not significant.

This study is one of the few to have assessed extra-cranial artery responses to exercise in humans. Whilst some previous studies have assessed blood flow responses in the ICA during exercise,^7,23,24 these have not specifically focused on changes in arterial diameter or shear stress. Furthermore, to the best of our knowledge, no study has reported changes in ICA and VA diameter or shear during and following a commonly undertaken bout of exercise. We observed significant ICA and VA dilation during exercise in all subjects. Our findings therefore raise the possibility that exercise-mediated and shear-driven increases in the bioavailability of endothelium-derived dilators such as NO may counteract other vasoactive pathways, such as the increase in sympathetic drive associated with some forms of exercise. Indeed, functional sympatholysis is a well-established phenomenon in skeletal muscle arterioles during exercise.^25,26 However, the role of sympatholysis in the brain is unclear, and mechanism(s) of vasomotion and adrenoceptor distribution differ from the large to smaller cerebral arteries.²⁷

This is the first study that has assessed VA diameter and shear relationships during EX, CO₂ and CTRL conditions in humans. Our findings indicate that the vertebral arteries exhibit dilator responses which are similar to those observed in the ICA. Only two previous studies, to our knowledge, have recorded VA blood flows during exercise in humans, but these did not address changes observed in either diameter or shear.^7,24 When considered in concert with our contemporaneous MCA and PCA observations, our ICA and VA data therefore indicate that cerebral shear and artery dilation occur in both the anterior and posterior cerebral circulation in response to both EX and CO₂ in humans.

An interesting secondary finding of the present study relates to change in shear and artery diameter following the cessation of exercise. We continued data collection in the post-exercise period, at different time-points in the ICA (5 min post) and VA (10 min post). Dilation remained elevated 5 min post-exercise, whereas it had returned to near resting baseline levels 10-min post-exertion. These findings provide some insight into the time-course of change in extra-cranial artery diameter following a physiological relevant exercise duration and intensity.

There are several limitations of the current experiment. We recruited and studied young male subjects. The impact of cyclical changes in sex hormones on shear-mediated cerebrovascular function and health would be an important follow-up study in women (as would aging and menopause). Although our findings suggest that exercise or hypercapnia elicited shear-mediated endothelial dependent dilation of the ICA and VA, we did not utilise a NO antagonist to address specific endothelial pathways. However, a previous study following a sublingual administration of an NO donor (nitroglycerin spray) observed vasodilation in the common carotid artery²⁸ that was of similar magnitudes to the dilation that we observed in the ICA. Future studies involving more invasive approaches, for example endothelial denudation and/or pharmacological blockade, would advance our understanding of the endothelial dependency of extra-cranial dilation. Similarly, experiments involving deliberate manipulation of the magnitude of shear (e.g. different levels of exercise or P_ETCO₂) and assessment of consequential diameter change in extra-cranial arteries are germane. It is unlikely that our findings are explained solely as a result of the increases in blood pressure, per se. Increases in intra-arterial pressure, and associated transmural pressure, are typically associated with myogenic-mediated vasoconstriction²⁸; however, it is unknown how this directly influences shear-mediated vasodilation in extracranial arteries. Our multiple linear regression analysis suggests that the principal determinant of change in arterial diameter was shear stress, but that the addition of changes in BP during the hypercapnic condition contributed to shear-mediated dilation. Also, P_ETCO₂ was somewhat elevated above baseline in each condition, and linear regression analysis between ICA dilation and shear stress during exercise was enhanced when changes in P_ETCO₂ were included as a covariate. We are therefore unable to exclude the possibility that CO₂ contributes to extra-cranial dilation. However, we have recently provided evidence^10,11 that the influence of CO₂ on extra-cranial vasodilation is due, in part, to shear stress. Indeed, when a bolus of CO₂ is administered and arterial diameter is measured following restoration of P_ETCO₂ and BP to normal values, vasodilation occurs. This vasodilation is approximately 50% in magnitude of the vasodilation observed during a steady state, or persistent CO₂ stimulus. Therefore, while data to support direct vasodilatory mechanisms on dilation in the ICA or VA do not currently exist in humans, a decrease in small intracranial arteriole resistance and consequent increase in extracranial shear stress clearly possesses a significant role in cerebrovascular regulation. Stimulation of the peripheral chemoreceptors is associated with sympathoexcitation, and usually associated with systemic arterial vasoconstriction rather than the dilation we observed. Finally, we acknowledge that the limitations of TCD impair our ability to extrapolate our extracranial diameter responses to changes in shear stress, to intracranial changes that may have simultaneously occurred in the MCA and PCA. However, the intention of our study was to focus on the ICA and VA responses, with intracranial measures collected primarily in order to confirm that matching of shear in each intervention was successful.

Conclusion

We have demonstrated, for the first time in humans, that matched increases in shear stress through the extra-cranial cerebral conduit arteries in response to distinct stimuli induce similar levels of vasodilation. Alongside previous evidence, largely derived from animal and in situ preparations,^29,30 our findings regarding vasodilator changes in both the anterior and posterior extra-cranial conduit arteries suggest an important mechanistic role for the endothelium in regulating cerebrovascular function in humans in response to both 30 min of moderate hypercapnia and exercise. Since previous studies indicate that some clinical populations (stroke, Alzheimer’s, etc.)^31,32 exhibit impaired carotid and MCA responses to physiological challenges, our cerebral flow-mediated dilation approach may prove useful in the characterisation of cerebrovascular health in humans. Interventions that enhance endothelial function may mitigate age-related declines in cerebrovascular vasodilator function in humans.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: KJS was supported by a Natural Sciences and Engineering Council of Canada Post-Doctoral Fellowship. RLH was supported by a UBC School of Medicine Friedman’s Scholarship. PNA is a Canadian Research Chair in Cerebrovascular Physiologist. DJG is a National Health and Medical Research Council Principal Research Fellow (1080914). This work was supported by NHMRC Project grant (no. APP1045204)

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors’ contributions

KJS and RLH were involved in the experimental design, data collection, analysis, and manuscript preparation. HM and RG were involved in the experimental design, data collection analysis, and editing and reviewing of the final manuscript. PNA, LN and DJG were involved in the experimental design, interpretation of data, manuscript preparation and review.