Aging alters the dampening of pulsatile blood flow in cerebral arteries

Abstract

Excessive pulsatile flow caused by aortic stiffness is thought to be a contributing factor for several cerebrovascular diseases. The main purpose of this study was to describe the dampening of the pulsatile flow from the proximal to the distal cerebral arteries, the effect of aging and sex, and its correlation to aortic stiffness. Forty-five healthy elderly (mean age 71 years) and 49 healthy young (mean age 25 years) were included. Phase-contrast magnetic resonance imaging was used for measuring blood flow pulsatility index and dampening factor (proximal artery pulsatility index/distal artery pulsatility index) in 21 cerebral and extra-cerebral arteries. Aortic stiffness was measured as aortic pulse wave velocity. Cerebral arterial pulsatility index increased due to aging and this was more pronounced in distal segments of cerebral arteries. There was no difference in pulsatility index between women and men. Dampening of pulsatility index was observed in all cerebral arteries in both age groups but was significantly higher in young subjects than in elderly. Pulse wave velocity was not correlated with cerebral arterial pulsatility index. The increased pulsatile flow in elderly together with reduced dampening supports the pulse wave encephalopathy theory, since it implies that a higher pulsatile flow is reaching distal arterial segments in older subjects.

Introduction

Pulsatile stress to highly-perfused organs, such as the brain, has emerged as a possible cause of microvascular damage.^1–4 The pulsatile blood flow, which is created by the heart, is transferred to the aorta and its proximal branches, which are rich in elastin that dampens the pulsations through the Windkessel effect.⁵ With aging, a complex interplay between large and small arteries occurs, and increased large artery stiffness causes an increased pulsatile stress to the brain microvasculature.^6,7 The term “pulse wave encephalopathy” (PWE) is sometimes used for describing how increased pulsatile flow leads to microvascular damage in the brain. Such mechanisms are believed to be involved in the pathophysiology behind leukoariosis,⁸ lacunar infarcts, normal pressure hydrocephalus, mild cognitive impairment, and dementia.^4,9

A common method to investigate aortic stiffness is aortic pulse wave velocity (PWV), which is considered as a predictor of cardiovascular morbidity and mortality.¹⁰ Flow pulsations in cerebral arteries can be investigated using Doppler ultrasonography and phase-contrast magnetic resonance imaging (PCMRI).^11–13 Although Doppler is widely available, it has several drawbacks. The most important is that investigations are limited to the proximal branches of the cerebral arterial tree. Using PCMRI, cerebral blood flow and its pulsations can be investigated without anatomical restrictions in any cerebral artery down to a diameter ∼1.5 mm.¹⁴

Cerebral pulsatile flow can be assessed using Gosling’s pulsatility index (PI).¹⁵ PCMRI makes it possible to investigate proximal as well as distal segments of cerebral and extra-cerebral arteries.¹⁶ This creates a possibility to investigate the dampening capacity of arteries using Gosling’s dampening factor (DF) (i.e., by dividing the PI of a proximal artery by the PI of its distal branch),¹⁵ which can increase our understanding about how the cardiac pulsatile flow is transmitted into cerebral arteries.

The main purpose of this study was to describe the dampening of the pulsatile flow from the proximal to the distal cerebral arteries and its change due to age and sex in 94 healthy subjects. In addition, we investigated the relationship between aortic stiffness and the pulsatile blood flow characteristics of the cerebral and extra-cerebral arteries.

Materials and methods

Subjects

By an ad in the local newspaper, healthy individuals were recruited to this prospective single-center study. The study design has previously been presented in detail.¹⁶ Subjects with cardiac disease, renal disease or vascular disease were not eligible for this study. Patients with a history of hypertension or medication for hypertension were also excluded. One hundred and eleven healthy subjects underwent physical examination (Figure 1). Further exclusion criteria were: mini-mental state estimation (MMSE) score <28 points, ECG-detected ventricular hypertrophy, arrhythmia or previous myocardial infarction. Medications affecting the central nervous system or contraindication for MRI examination (e.g., claustrophobia) were also a reason for exclusion. Figure 1 summarizes the inclusion of the study population and reasons for exclusion. The final study population consisted of 94 subjects that were divided into two groups based on age, 49 subjects in a “healthy young group” (HY, age range 20–30; mean age ± SD, 25 ± 2 years) and 45 subjects in a “healthy elderly group” (HE, age range 64–80; mean age ± SD, 71 ± 4 years).

Figure 1.

Study population and reason for exclusion. One hundred eleven subjects were included for physical examination, where 10 were excluded. After MRI examination three subjects were excluded. The final number of study population consisted of 49 healthy young and 45 healthy elderly subjects.

Oral and written information about the study was given to the subjects and a written consent was obtained from all subjects. The ethical review board of Umeå University (IRB) approved the study and was performed in accordance with the guidelines of the declaration of Helsinki.

Magnetic resonance imaging

The MRI examination was performed using a 3-Tesla scanner (GE Discovery MR 750, Waukesha, WI) with a 32-channel head coil. The duration of the examination was approximately 1 h. Sagittal T1 sequence and a three-dimensional time of flight sequence (3DTOF) were performed for identifying cerebral and vascular pathology, describing the morphology of the cerebral arteries and for planning PCMRI measurements. Fazekas scoring was performed on axial T2 sequence in HE by an experienced neuroradiologist (RB) that was blinded for all clinical and PCMRI data. The subjects were scored as: none (0), mild (1), moderate (2), or severe (≥3) leukoariosis.¹⁷ Fazekas scoring was not performed in HY subjects, under the assumption that leukoariosis was absent.

A 2D fast phase-contrast sequence with retrospective cardiac gating was performed to assess the blood flow velocities of the cerebral arteries. The following parameters were used: TR/TE = 9/5 ms, slice thickness 3–5 mm, 15° flip angle, 180 mm × 180 mm field-of-view, 0.35 mm × 0.35 mm in-plane resolution, six views-per-segment, and two averages. The PCMRI data were reconstructed to provide 32 phases throughout one entire cardiac cycle as phase (velocity-mapped) and magnitude (anatomic) images. The velocity encoding ranged from 20 cm/s in the middle meningeal artery to 120 cm/s in the middle cerebral artery.

Blood flow rate (BFR) was measured in the following arteries: 1. Anterior cerebral circulation: Internal carotid artery (ICA), middle cerebral artery (MCA), distal MCA (MCA_dist), anterior cerebral artery (ACA), and the distal ACA (ACA_dist). 2. Posterior cerebral circulation: vertebral artery (VA), basilar artery (BA), and posterior cerebral artery (PCA). 3. Extra-cerebral arteries were defined as the external carotid artery (ECA), middle meningeal artery (MMA), and the ophthalmic artery (OA).

Measured parameters

MRI flow measurements

PCMRI data were analyzed using Segment v1.8 software (http://medviso.com) by one investigator (L.Z.). Using the magnitude image, the maximum cross-section area during the cardiac cycle was chosen, where a region of interest (ROI) was manually drawn around the boundary of the artery of interest. The cross-sectional area of the ROI was kept constant during the cardiac cycle. The BFR (mL/min) was automatically computed as the product between mean velocity and the cross-sectional area. The mean, maximum (systolic) and minimum (diastolic) BFRs during the cardiac cycle were assessed (Figure 2). The total cerebral blood flow was calculated by summing BFR in the ICAs and VAs. For investigating post-processing test-retest variability for blood flow measurements in 10 randomly subjects (5 HY), two independent observers (K.A. and L.Z.) measured BFR in ICA, MCA, ACA, and PCA on left side. The interclass correlation coefficient was used¹⁸ and an inter-rater variability of >0.9 was achieved.¹⁶ A corresponding analysis for PI on the same data gave an interclass correlation coefficient of 0.96.

Figure 2.

Flow rate of internal carotid artery. Blood flow rate of left internal carotid artery in an elderly subject during two repeated cardiac cycles, demonstrating how pulsatility index was achieved. The figure shows the maximum, minimum, and means blood flow rates.

BFR in ICA, VA, and ECA were measured at the level of the first and second cervical vertebrae, all based on the same PCMRI data. In MCA, a perpendicular plane was set across the artery at M1 level, that is, the first segment of MCA before branching. ACA was measured at A1 level, that is, proximal to the anterior communicating artery and PCA at P2 segment, that is, distal to the aperture of the posterior communicating artery. Distal cerebral arteries, MCA_dist and ACA_dist, were measured by choosing the largest visualized MCA branch at the level of the Sylvian fissure and the two largest visible pericallosal branches of ACA, respectively. Finally, flow measurements for BA were assessed above the anterior inferior cerebellar arteries and MMA at the level of the foramina spinosa.

Blood pressure and PWV measurements

The subjects were investigated in a quiet room after resting for at least 5 min. Blood pressure was measured in left arm in sitting position (Omron HEM757; Omron Matsusaka Co. Ltd., Mie, Japan). Aortic PWV (m/s) was measured using the Arteriograph oscillometric device (TensioMed Kft., Budapest, Hungary).¹⁹ This is a noninvasive method using an upper arm cuff. The distance from the sternal notch to the upper edge of pubic bone was manually measured as an estimation of the aortic length. Initially the Arteriograph measures the systolic and diastolic blood pressure and then deflates. A new measurement starts within a few seconds and the device again inflates. The diastolic pressure is measured and the device then inflates to approximately 35 mmHg above the actual systolic blood pressure causing occlusion of the brachial artery. The signals at both pressure levels are recorded for approximately eight seconds and transmitted to a computer. The Arteriograph measures the time interval between the peaks of the direct systolic wave and the reflected systolic wave. Two measurements were done and the mean of the two values was registered.

Pulsatility index

PI was calculated based on BFR in (mL/min) according to the equation used by Gosling¹⁵

$$\text{PI}=\frac{{\text{BFR}}_{\text{syst}}{-\text{BFR}}_{\text{diast}}}{{\text{BFR}}_{\text{mean}}};$$

where BFR_syst, BFR_diast, and BFR_mean are the systolic, diastolic, and mean BFR, respectively (Figure 2).

Dampening factor

For investigating how the pulsatile flow was dampened from the proximal to the distal arterial segments, Gosling’s DF was used¹⁵

$$\text{DF}=\frac{{\text{PI}}_{\text{proximal}}}{{\text{PI}}_{\text{distal}}}$$

The results were expressed in absolute values. ICA, VA, and ECA were defined as proximal feeding arteries. DFs for arterial branches of ICA (MCA, ACA, MCA_dist, ACA_dist, and OA) were calculated by dividing PI of ICA with its distal branches. BA and PCA were calculated in the same way using VA as proximal artery. All P2 segments of PCA where a fetal type variation occurred (n = 19) were excluded from the calculation of DF of PCA. This is because PCA pulsatile blood flow arises from ICA and not from the vertebrobasilar arterial system. In subjects with only unilateral fetal type PCA (n = 17) and no other anatomical variation (e.g., hypoplastic ACA or bilateral fetal type PCA) a subanalysis was made to compare DF with the contralateral “normal” side. Finally, MMA was calculated using ECA as proximal artery.

Statistical analysis

PASW Statistics version 22 (IBM, Chicago, USA) was used to analyze the data. PI was calculated as the mean of right and left of the paired artery to limit the number of comparisons. If PI was missing for a paired artery, the PI of the contralateral artery was used. A univariate general linear model was used to analyze the effect of age and sex (used in the same statistical model) on PI and DF.

To investigate how unilateral fetal type PCA affects PI and DF, a paired t-test was used for comparing the ipsilateral side with the contralateral side. The correlations between PI and PWV were analyzed using a partial correlation with age and sex as control variables. Results were expressed in mean ± SD. Significant threshold was set as p < 0.05. To correct for multi-comparisons, the false discovery rate procedure was used.²⁰

Results

The subject characteristics are shown in Table 1. PI could be measured in 1893 arteries. Eighty-one arteries (4.2%) could not be measured due to technical problem or anatomical variations (e.g., hypoplasia or nonvisible artery on the MRI sequences). PWV could not be measured in one subject due to technical problems.

Table 1.

Clinical features of the included subjects.

	HE (n = 45)	HY (n = 49)	p-Value
Clinical features
Age (y)	71 ± 4	25 ± 2	< 0.001
Sex (F/M)	23/22	27/22	0.26
Height (cm)	172 ± 10	174 ± 10	0.79
Weight (kg)	72 ± 14	72 ± 14	0.88
Waist (cm)	86 ± 11	77 ± 10	< 0.001
Hip (cm)	99 ± 7	100 ± 7	0.64
BMI (kg/m2)	24 ± 3	24 ± 3	0.25
MMSE (points)	29.7 ± 0.5	29.3 ± 0.7	0.002
Cerebral and vascular features
Systolic BP (mmHg)	140 ± 14	124 ± 10	< 0.001
Diastolic BP (mmHg)	84 ± 7	73 ± 6	< 0.001
HR (bpm)	69 ± 8	65 ± 12	0.53
MAP (mmHg)	103 ± 9	91 ± 7	< 0.001
Brachial PP (mmHg)	57 ± 13	51 ± 9	0.01
PWV (m/s)	9.3 ± 1.6	6.2 ± 1.2	< 0.001
tCBF (mL/min)	657 ± 94	771 ± 112	< 0.001
Fazekas score ≥ 2	7 (16%)	NA	NA

Note: Healthy subjects were included into two age groups, (HY, age range 20–30; mean age ± SD, 25 ± 2 years and HE, age range 64–80; mean age ± SD, 71 ± 4 years). Values are means ± standard deviation or number of subjects. HY: healthy young; HE: healthy elderly; F: female; M: male; BMI: body mass index; MMSE: Mini-Mental State Examination; BP: blood pressure; MAP: mean arterial pressure; HR: heart rate, measured during MRI examination; PWV: pulse wave velocity; PP: pulse pressure; tCBF: total cerebral blood flow; NA: not applicable.

PI in the individual arteries

PIs in HE and HY for cerebral and extra-cerebral arteries are shown in Table 2. A significant increase in PI was found for all cerebral arteries except for VA, when comparing HE and HY. For extra-cerebral arteries, PI in OA was significantly increased in HE, PI in ECA was higher in HY and PI in MMA did not differ significantly between the two age groups (Table 2).

Table 2.

Pulsatility index for cerebral and extra-cerebral arteries in healthy elderly and healthy young.

Arteries	HE (n = 49) PI mean ± SD	HY (n = 45) PI mean ± SD	p-Value
ICA	0.96 ± 0.15	0.84 ± 0.13	<0.001
MCA	0.89 ± 0.12	0.72 ± 0.11	<0.001
MCAdist	0.84 ± 0.13	0.65 ± 0.10	<0.001
ACA	0.88 ± 0.14	0.67 ± 0.12	<0.001
ACAdist	0.77 ± 0.12	0.58 ± 0.11	<0.001
VA	1.11 ± 0.18	1.07 ± 0.22	0.31
BA	0.88 ± 0.12	0.71 ± 0.12	<0.001
PCA	0.85 ± 0.13	0.64 ± 0.10	<0.001
ECA	1.57 ± 0.24	1.92 ± 0.46	<0.001
MMA	1.48 ± 0.32	1.35 ± 0.30	0.06
OA	1.20 ± 0.22	1.10 ± 0.26	0.04

Note: PI: pulsatility index; ICA: internal carotid artery; VA: vertebral artery; MCA: middle cerebral artery; MCA_dist: distal branch of middle cerebral artery; ACA: anterior cerebral artery; ACA_dist: distal branch of anterior cerebral artery; BA: basilar artery; PCA: posterior cerebral artery; ECA: external carotid artery; MMA: middle meningeal artery; OA: ophthalmic artery.

There was no difference in PI for any investigated artery when comparing women and men (supplementary table).

In subjects with fetal type PCA, there was no significant difference in PI for ACA, MCA, and PCA between the ipsilateral and the contralateral side (ACA: p = 0.68; MCA: p = 0.52: PCA: p = 0.23).

DF from proximal to distal arteries

DF for cerebral pulsatile flow is shown in Table 3. A progressive dampening of pulsations was observed along all cerebral arterial branches of ICA (MCA, ACA, MCA_dist, and ACA_dist) and VA (BA, PCA). The dampening of pulsations was significantly higher in HY than in HE for all cerebral arteries. Figure 3 illustrates in percentage how much of the proximal pulsation that are left while traveling along the cerebral arterial tree (i.e., PIdistal/PIproximal). In subjects with fetal PCA (n = 17), the DF of ACA, MCA, and PCA was similar between the ipsilateral and the contralateral side (ACA: p = 0.98; MCA: p = 0.23: PCA: p = 0.08).

Table 3.

Dampening factor for cerebral and extra-cerebral arteries in healthy elderly and healthy young.

Arteries	HE (n = 49) DF mean ± SD	HY (n = 45) DF mean ± SD	p-Value
Branches of ICA
MCA	1.08 ± 0.12	1.18 ± 0.15	=0.001
MCAdist	1.16 ± 0.20	1.31 ± 0.20	<0.001
ACA	1.10 ± 0.15	1.27 ± 0.18	<0.001
ACAdist	1.26 ± 0.22	1.48 ± 0.25	<0.001
OA	0.82 ± 0.15	0.80 ± 0.17	0.54
Branches of VA
BA	1.27 ± 0.18	1.53 ± 0.27	<0.001
PCA	1.32 ± 0.20	1.69 ± 0.38	<0.001
Branches of ECA
MMA	1.07 ± 0.25	1.43 ± 0.31	<0.001

Note: Dampening factor for an artery was calculated by dividing the PI of the proximal feeding artery (ICA, VA, ECA) by the distal arteries. PI: pulsatility index; ICA: internal carotid artery; VA: vertebral artery; MCA: middle cerebral artery; MCA_dist: distal branch of middle cerebral artery; ACA: anterior cerebral artery; ACA_dist: distal branch of anterior cerebral artery; BA: basilar artery; PCA: posterior cerebral artery; ECA: external carotid artery; MMA: middle meningeal artery; OA: ophthalmic artery.

Figure 3.

Alteration of pulsatile flow in percentage for cerebral arteries in young (red) and elderly (black) healthy subjects. To visualize the remaining BFR pulsatility along the arterial tree with respect to the feeding artery, the pulsatility of a distal artery was divided by the feeding proximal artery (ICA and VA) and expressed as a percentage (DF_dist/DF_prox) × 100. Branches of ICA: MCA, MCA_dist, ACA, ACA_dist. Branches of VA: BA and PCA. ICA: internal carotid artery; VA: vertebral artery; MCA: middle cerebral artery; ACA: anterior cerebral artery; distal branch of middle cerebral artery; ACA_dist: distal branch of anterior cerebral artery; BA: basilar artery; PCA: posterior cerebral artery; OA: ophthalmic artery.

Aortic PWV and cerebral PI

Table 4 shows the partial correlation analysis between PWV and cerebral arterial PI. After correction using a false discovery rate procedure, the correlation between PWV and PI was not significant for any cerebral artery.

Table 4.

Partial correlation analysis between cerebral pulsatility index and pulse wave velocity.

Arteries	β-Value	p-Value
ICA	−0.036	0.734
MCA	0.129	0.228
MCAdist	0.187	0.076
ACA	0.281	0.007
ACAdist	0.135	0.208
VA	−0.093	0.382
BA	0.091	0.396
PCA	0.230	0.029

Note: Partial correlation between cerebral arterial PIs and aortic PWV. Age and sex was set as control variables. After correction using a false discovery rate procedure, the correlation between PWV and PI was not significant for any cerebral artery. PI: pulsatility index; PWV: pulse wave velocity; ICA: internal carotid artery; VA: vertebral artery; MCA: middle cerebral artery; MCA_dist: distal branch of middle cerebral artery; ACA: anterior cerebral artery; ACA_dist: distal branch of anterior cerebral artery; BA: basilar artery; PCA: posterior cerebral artery.

Discussion

Using a noninvasive technique, blood flow pulsations were measured in multiple cerebral arteries in healthy subjects. We found that with increasing age, the brain is exposed to higher pulsatile flow, which reaches further and further out into the cerebral vascular tree. In healthy subjects, aortic stiffness (i.e., PWV) did not reflect cerebral pulsatile flow. Our results could have implications for explaining the development of small vessel disease and vascular dementia.

PI increases with age in the cerebral arteries

Changes in the arterial system can be due to aging, pathological processes, or both. Knowledge about age-related arterial changes is therefore needed to understand pathological processes. The increase in PI with age observed in this study supports the theory that age is a main factor for increased arterial pulsatility.¹¹ PI was measured using flow (mL/min), in contrast to using Doppler ultrasonography, which is based on measured velocity (cm/s). In the present study, PI calculation based on flow or velocity should not differ since the cross-sectional area of the measured artery was kept constant during the cardiac cycle. However, lower temporal sampling rate associated with PCMRI can give a reduction of estimated systolic peak velocity leading to lower PI values using PCMRI, compared to Doppler ultrasonography.²¹

We found that cerebral arterial PI increases and PI dampening is reduced along cerebral arteries due to aging (Tables 2 and 3). In spite of the fact that PI is related to cerebral vascular resistance, studies have shown that PI is not only a measure of vascular resistance.^22,23 In addition, it has been shown that cerebral vascular resistance is not solely a function of arterioles but also a contribution of large cerebral arteries.^24,25 An interpretation of the physiological mechanisms behind our observations requires an understanding of the parameters influencing PI. PI is defined as the ratio between the pulsatile and the mean BFR. By looking at the mean BFR part of the equation, an increase of PI can be interpreted as an increase of the cerebral vascular resistance. However, we also have to consider the pulsatile component of BFR that should depend on pulse pressure. Therefore, an increase in PI can also be interpreted as increased aortic stiffness forcing a larger pulse pressure to the cerebral vasculature, thus producing increased cardiac related variations in BFR.²⁶ Cerebral arteries, which may become stiffer with age, will be exposed to a higher pulsatile flow, which would yield a higher PI further out in the vascular tree, that is, a reduced dampening function. Furthermore, mean cerebral blood flow is known to decrease with age¹⁶ and the systolic blood flow velocity in the cervical arteries increases while the diastolic decreases in a curve-linear pattern due to aging.²⁷ The increased PI with aging found in our study was thus an expected finding based on the known physiological changes in the arterial system due to aging and the definition of the PI.

In the present study, the PI of the arteries supplying the brain was lower than in the extra-cerebral arteries. Extra-cerebral arteries supply a high-resistance vascular bed (mainly the skin and muscles of the face). Thus, the flow profile is different with a lower diastolic flow leading to a higher PI in ECA compared to ICA.²⁸ In addition, we found that the PI in ECA was higher in young than in elderly, which is explained by a higher pulsatile flow in young subjects since the mean BFR in ECA is not affected by age.¹⁶

In this study, no correlation was found between PWV and cerebral arterial PI. Previous studies have found a strong correlation between PWV and PI in MCA using Doppler ultrasonography. Possible explanations could be that previous studies did not adjust for the effect of aging,²⁷ or that the participants had cerebrovascular disease.²³ This is in contrast to our study where all subjects were healthy and without severe leukoariosis. Our finding needs to be confirmed by others, but we consider that PWV should not alone be used to describe cerebral arterial function. In addition, the effects of aging on the arteries are heterogeneously distributed, and the properties of the arteries vary depending on the type of artery that is studied, for example, elastic or muscular arteries. Therefore, investigations at only one site, for example, the carotids, will not tell us the overall impact of age or disease.²⁹

Dampening of pulsatile flow in the cerebral arteries

When investigating PI with a proximal to distal approach, information about of the pulsatile blood flow attenuation can be achieved. Recent studies found that pulsatile flow is affected by physiological and pathological vascular geometry when investigating in a proximal-distal approach.^13,30,31 An important finding in this study was that the pulsatile flow that reaches distal arterial segments was augmented with age. Firstly, the PIs in large proximal arteries were increased in elderly (Table 2) and secondly the dampening capacity of the cerebral arteries was more efficient in the young group for all cerebral arterial segments (Table 3). It is well known that the smallest arteries in the brain are exposed to higher pulsatile pressure than the arteries in other organs, due to the low resistance in the cerebral vascular bed.³² Experimental studies have shown that myogenic constriction in MCA is impaired and less responsive to pulsatile pressure due to aging.³³ In addition, studies have shown that resistance arteries in hypertensive aged mice are less functioning leading to microvascular injury including increased blood brain barrier disruption³⁴ and increased incidence of cerebral microhemorrhages.³⁵ The findings from these studies together with the present study provide important clues and suggestions for future studies in the field of age-related cerebrovascular alteration.

Bateman coined the term “pulse-wave encephalopathy” (PWE) to hypothesize a joint pathophysiology leading to leukoariosis and several diseases that occur with increasing age.⁴ The theory behind PWE is that excessive blood flow pulsatility is thought to cause cerebral microvascular damage. Not all elderly subjects are affected by PWE and we are unaware of the intensity and duration of the excessive pulses, which can cause the development of PWE.

Traditionally, Doppler ultrasonography has been the main method for describing pulsatility in cerebral arteries. This method is restricted to only the proximal segments of the cerebral vascular system, which limits its applicability for a full assessment of pulsatility dampening. By using 2D PCMRI, arteries can be investigated without anatomical restrictions, although not simultaneously, which is the main limitation of this method. We used the maximum cross-section area during the cardiac cycle by manual delineation of the artery. Hence, flow measurements is a valid measure because all of the artery cross-section is enclosed by the ROI, and potential transient underestimations in velocity is adequately compensated by an overestimation in area for the volumetric flow rate calculation (ml/min). In addition, previous studies have shown no systematic error between scan re-scan for PI assessment³⁶ and a good agreement between PCMRI flow measurements and flow in vessel-phantoms.¹⁴ To estimate measurement uncertainty in the DF, we used a reference value of 10%³⁷ for the standard deviation of pulsatility measures (repeated within-subject measurements). Further, we applied the Gaussian error propagation formula on the DF equation and estimated the standard deviation of repeated dampening measurements to be about 14%. We also note that although this measurement uncertainty will limit the ability to detect differences in DF, it will not contribute to false positive results regarding age effects in DF. Future studies should use 4D PCMRI, which makes it a promising method for describing simultaneously the flow pulsatility in the entire arterial tree. The dampening capacity of cerebral arteries, as measured with PCMRI, is a potential biomarker that should be evaluated in subjects at risk of developing diseases linked to increased pulsatile flow.

Conclusion

Cerebral arterial pulsatile flow was found to dampen in a progressive manner moving from proximal to distal parts of the cerebral arterial system. Dampening was more pronounced in younger than older subjects. There was no correlation between aortic PWV and cerebral pulsatile flow in healthy subjects. This study supports the PWE theory by showing a higher pulsatile stress reaching distal arterial segments in older subjects.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by: The Swedish Research Council Grant 621-2011-5216; The European Union Objective 2 Norra Norrland (Project: 148273 CMTF); The County Council of Västerbotten; and Swedish Heart and Lung Foundation Grants 20110383 and 20140592. The present study was sponsored by the Swedish Brain Foundation.

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

Zarrinkoob and Ambarki – substantial contribution to conception and design. Acquisition of data and analysis and interpretation of data. Revising the article for important intellectual content. Final approval of the version to be published. Wåhlin, Eklund, and Malm – substantial contribution to conception and design. Interpretation of data. Revising the article for important intellectual content. Final approval of the version to be published. Birgander and Carlberg – substantial contribution to conception and design. Revising the article for important intellectual content. Final approval of the version to be published.

Supplementary material

Supplementary material for this paper can be found at http://jcbfm.sagepub.com/content/by/supplemental-data