Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 May 1.
Published in final edited form as: Neurobiol Aging. 2018 Jan 31;65:77–85. doi: 10.1016/j.neurobiolaging.2018.01.006

Quantitative cerebrovascular pathology in a community-based cohort of older adults

Swati Rane 1, Natalie Koh 1, Peter Boord 1, Tara Madhyastha 1, Mary K Askren 1, Suman Jayadev 1, Brenna Cholerton 2, Eric Larson 3, Thomas J Grabowski 1
PMCID: PMC5871567  NIHMSID: NIHMS936648  PMID: 29452984

Abstract

Cerebrovascular disease, especially small vessel pathology, is the leading comorbidity in degenerative disorders. We applied arterial spin labeling (ASL) and cerebrovascular reserve (CVR) imaging to quantify small vessel disease and study its effect on cognitive symptoms in non-demented older adults from a community based-cohort. We evaluated baseline cerebral blood flow (CBF) using ASL, and percent signal change as a marker of CVR using blood-oxygen level-dependent (BOLD) imaging following a breath-hold stimulus. Measurements were performed in and near white matter hyperintensities, which are currently the standard to assess severity of vascular pathology. We show that similar to other studies (i) CBF and CVR are markedly reduced in the hyperintensities as well as the tissue surrounding them, indicating susceptibility to infarction. (ii) Low CBF and CVR are significantly correlated with poor cognitive performance. (iii) Additionally, compared to a 58.4% reduction in CBF, larger exhaustion (79.3%) of CVR was observed in the hyperintensities with a faster, non-linear rate of decline. We conclude that CVR may be a more sensitive biomarker of small vessel disease than CBF.

Keywords: Aging, cerebral blood flow, cerebrovascular disease, ASL, white matter disease, BOLD contrast, CVR

1. INTRODUCTION

Small vessel cerebrovascular disease is a common occurrence in aging and related disorders such as Alzheimer’s Disease (Wardlaw et al., 2013; Pantoni L 2010; Al-Bachari S et al. 2014). It is a major contributor to mixed dementia pathology and has similar risk factors as those for cognitive decline (White L et al., 1996). Typical cerebrovascular pathology associated with dementia consists of cortical and subcortical infarcts, microbleeds, increased perivascular spaces, reduced perfusion, and tissue atrophy (Esiri et al., 1999; Bots et al., 1997). Histopathological studies show that individuals with Alzheimer’s disease show increased basal membrane thickening, collagenous deposits, and damaged perciytes in the microvasculature compared to age-matched older adults (Farkas and Luiten, 2001). The resulting vascular insufficiency causes neuronal damage and subsequently, cognitive decline, which progresses slowly. Therefore, sensitive methods are needed to monitor the advancing vascular pathology to subsequently arrest disease progression.

Recent advances in neuroimaging have provided tremendous insights into small vessel disease, especially by distinguishing normal appearing brain matter from pathological tissue. One important imaging measure is the volume of white matter hyperintensities i.e., regions of brain tissue, which appear bright on a T2-weighted MRI scan, in deep white matter tissue (deGroot JC et al., 200; Au R et al., 2006; Prins ND et al., 2015). White matter hyperintensities are thought to occur due to chronic hypoperfusion as a result of the small vessel disease. A higher volume of white matter hyperintensities is associated with poorer cognitive outcome in older adults (Prins ND et al., 2015; Brickman et al., 2009; Dalen et al., 2016; Bahrani et al., 2017). It is also associated with poor executive function, especially processing speed for executive tasks (Au R et al., 2006; Smith et al., 2011). While anti-hypertensive treatments manage to slow their development (Dufouil et al., 2015; Liao et al., 1996), these hyperintensities likely represent tissue that is damaged and is currently unresponsive to vascular therapies.

Cerebral blood flow (CBF) imaging is another imaging marker that can detect ischemic tissue (Detre et al., 1994; Wong et al., 1999). Similarly, cerebrovascular reserve (CVR) mapping using hypercapnia induced by acetazolamide, breath-hold or manipulating inhaled gas can identify tissue with low or exhausted vascular reserve (Gückel et al., 1996; Kastrup et al., 1998; Bright et al., 2013). Reduced CBF reflects presence of primary vasculopathy such as atherosclerosis of large vessels, microvascular injury (basal membrane thickening, pericytic degeneration, collagen deposits, mentioned earlier), or a reduced metabolic demand due to neuronal dysfunction/loss (Farkas and Luiten, 2001). A reduced CVR may indicate either microvascular injury or an exhausted reserve. We believe that a combinatorial approach may provide better understanding of the underlying vascular physiology. One study shows that CBF is reduced in the tissue surrounding the white matter hyperintensities and does indeed indicate tissue at risk of infarction as seen on a longitudinal scan 2 years later (Promjunyakul et al., 2015). In general, white matter hyperintensities has been extensively studied using CBF.

However, CVR measurements in white matter hyperintensities and direct comparisons of sensitivity of CBF and CVR to detect progression of small vessel disease are limited with variable results. Sam et al., (2016) measured white matter integrity, T2, and CBF in white matter hyperintensities and normal appearing white matter and found that reductions in CVR preceded hyperintensity development in normal white matter. Another contrast-based study showed reduced baseline CBF and blood volume as well reduced changes in CBF and blood volume in white matter hyperintensities during an acetazolamide challenge, compared to normal appearing white matter (Marstrand JR et al., 2002). The Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) study measured similar baseline CBF and CVR (with acetazolamide) and reduced CVR (not CBF) in white matter hyperintensities (Liem MK et al., 2009). Reduced CBF and CVR in the white matter hyperintensities, compared to normal appearing white matter has been reported in a wide variety of diseases such as type 2 Diabetes (Novak V et al., 2006), small vessel disease (Molina C et al., 1999), Alzheimer’s Disease (Makedonov I et al., 2013), etc. In this study, we compare CBF and CVR to assess which biomarker is more sensitive to small vessel disease pathology and cognitive function in a community-based cohort of older adults.

Specifically, we explore CBF and CVR measurements as markers of ischemic tissue in and around the white matter hyperintensities. We believe that these markers have the potential to detect normal appearing tissue that is susceptible to ischemic events and which has a high risk of developing into hyperintensities. We hypothesize that, similar to CBF, CVR will also be reduced in the regions of the white matter hyperintensities. It will also be lower in the tissue surrounding the hyperintensities. We used arterial spin labeling to measure CBF and breath-hold blood-oxygen level-dependent (BOLD) imaging to measure CVR in a community based cohort of older adults (Detre et al., 1994, Bright et al., 2013). We performed comparisons of CBF and CVR in gray matter, normal appearing white matter, and the hyperintensity regions. Correlation with overall cognitive and executive function was also assessed. The advantage of this study is that we applied a non-invasive approach to measure CBF and CVR. Furthermore, we used a simple breath-hold paradigm, which does not require any specialized equipment.

2. MATERIAL AND METHODS

Thirty older adults (76.4±7.1 year, 13M/15F) were recruited from the community-based Adult Changes in Thought (ACT) cohort (Montine et al., 2012). ACT is a population-based study of dementia risk designed to prospectively examine the incidence of AD and dementia, as well as risk factors of these diseases, in a cohort representative of the Group Health Cooperative in Seattle. Non-demented subjects, who are cognitively normal or mildly cognitively impaired with no effect on daily living are included in this cohort. All participants provided written informed consent to participate in the present study, which was conducted according to the Declaration of Helsinki and subsequent revisions. The study was approved by the Institutional Review Board at University of Washington, Seattle. Demographic data is shown in Table 1. All subjects underwent the National Alzheimer Coordinating Center Uniform Data Set cognitive battery including TRAILS A, TRAILS B, MMSE: MiniMental State Examination (MMSE), Logical Memory, etc. (Beekly et la., 2007, Reitan, 1992; Folstein M et al.; 1975; Mack et al. 1992; Wechsler, 2014; Wechsler, 1945). Cognitive diagnosis was adjudicated during a clinical case consensus diagnosis conference consisting of neurologists, psychiatrists, neuropsychologists, and other study clinicians. These subjects underwent both a pseudo continuous arterial spin labeling (pCASL) scan to measure CBF and a breath-hold BOLD MRI for measuring CVR on a Philips 3T Achieva scanner with a 32 channel SENSE coil reception. Of these, two subjects had low MMSE scores (20 and 22) and hence were excluded from this study.

Table 1.

Demographic information (n = 28) and summary statistics

DESCRIPTION MEAN STD. DEV
AGE (YEARS) 76.4 7.1
GENDER 13M/15F
APOE e4 CARRIERS 5
MEAN ARTERIAL PRESSURE (mm HG) 91 11
HEART RATE (BPM) 65 9
MMSE 28 1
TRAILS A 30 11
TRAILS B 80 31
LOGICAL MEMORY 15 3
white matter hyperintensities (% of Total ICV)* 0.21 0.21
Open in a new tab

white matter hyperintensities Burden Range: 0.01 – 0.8, includes both periventricular and deep white matter hyperintensities

2.1 Imaging

Each session included a structural T1-weighted image using a 3D- Turbo Field Echo (TFE) acquisition with TR/TE = 9.2/3.5 ms, resolution = 1 × 1 × 1 mm3 for image registration and segmentation purposes. FLAIR images were acquired to detect white matter hyperintensities with the following parameters: TR/TI/TE = 5000/1800/293 ms, resolution = 1 × 1 × 1 mm3. pCASL imaging parameters were: TR/TE = 5000/35 ms, 30 pairs of control and label images, resolution = 3.5 × 3.5 × 5 mm3, background suppression (BS1 = 1710 ms, BS2, 2860 ms), label plane 80 mm below the center of the imaging volume. Labeling duration, τ, = 1650 ms and post labeling delay, ω, = 2000 ms, total acquisition time = 5 minutes. For the 6-minute breath-hold BOLD MRI implementation, subjects were trained outside the scanner to perform 6 paced breaths followed by a breath-hold for 10–15s (as feasible for the subject) and then by free breathing. The breath-hold was initiated at the end of an exhale (Bright et al., 2013). Blocks of paced breathing, breath-hold, and free breathing were repeated a total of 6 times. Other imaging parameters were: resolution = 3.5 × 3.5 × 3.5 mm3, TR/TE = 2500/35 ms, dynamics = 122, SENSE factor = 2. The same stimulus paradigm was repeated inside the scanner. The breathing pattern of the subject was recorded using respiratory bellows and used as the time-course for analyses. The time-course was also used to determine if subjects followed the breathing instructions inside the scanner.

2.2 Analyses

In order to quantify CBF, the pCASL perfusion weighted images were motion- corrected and registered to the first dynamic image using FSL MCFLIRT (Alsop et al., 2015; Jenkinson et al., 2002). The M0 image was not acquired during the scan session. Therefore, M0 image was calculated using the first control image, Knowing the timing of the background suppression pulses, and excitation pulses, the M0 image was calculated from the first control image using Bloch equations. The equation for M0 calculation was

Mz=M0(1-2e-ΔBS2T1+2e-(ΔBS1+ΔBS2)T1+2e-(BS2+ΔBS2)T1) (1)

where Mz is the magnetization from the first control mage, BS1 and BS2 are the timings of the background suppression pulses after the labeling pulses, BS1 is the duration between the two background suppression pulses, and BS2 is the duration between the second background suppression pulse and the excitation pulse. Note that this equation neglects the effect of the saturation pulses applied before labeling begins. The T1 was chosen appropriately based on whether the tissue was predominantly gray matter (T1 = 1331 ms), white matter (T1 = 832 ms), or cerebrospinal fluid (T1 = 4100 ms) (Wanspura et al., 1999; Chen et al., 2001). Tissue classification was done based on T1 segmentation using FSL FAST (Zhang et al., 2001).

Similar calculations were performed in other studies (Donahue et al., 2006). The purpose of calculating the M0 was to quantify absolute CBF instead of using relative perfusion weighted values. To ensure that this approach did not alter interpretations, we implemented the same sequence with and without M0 acquisition in 9 young healthy subjects (post hoc). We then compared the CBF measurements using a true M0 image and a calculated M0 image and ensured that CBF values were tightly correlated within each subject (Average correlation across subjects, r = 0.92±0.03, Supplementary Figure A.1 for overall cortical CBF. CBF values with the estimated M0 map were only slightly lower than those estimated with an acquired M0 map by a factor of 0.97±0.06). CBF values were measured and compared in the frontal, temporal, parietal, and occipital lobes as well as in the caudate, putamen, thalamus, hippocampus, amygdala, normal appearing white matter, and white matter hyperintensities. All regional CBF values were adjusted for gray matter density to account for CBF reductions due to partial volume effects with white matter and cerebrospinal fluid. All comparisons were performed using paired t-tests.

For CVR quantification, the breath-hold time-course was first subsampled to match the TR of the BOLD acquisitions. The subsampled time-course was then convolved with the respiratory response function (Birn et al., 2008). This function has been shown to better reflect the cerebral vascular response than the conventional hemodynamic response function and accounts for the circulation delay in the cerebrovascular response due to breathing. The new convolved time-course was then used as a stimulus time-course for FSL FEAT (FSL v5.0, Woolrich et al., 2009). In FEAT, we performed motion correction using MCFLIRT, brain extraction, baseline drift correction, and smoothing (FWHM = 5mm). Standard gray and white matter masks from FSL in MNI space (2mm) were applied to the FEAT output using FEAT Query to obtain percent signal change in the two tissue types. We ensured that the amplitude of the time-course fluctuated between 0 and 1 for correct interpretation of the FEAT Query results.

For the white matter hyperintensities, the FleX automatic lesion detection algorithm was applied to the FLAIR images, and binary masks of the hyperintensities were obtained (Gibson et al., 2010). The number of voxels and subsequently volume of the binary masks were calculated and divided by the total intra-cranial volume to obtain the percent of brain occupied by the white matter hyperintensities i.e. hyperintensity burden. The total intra-cranial volume was calculated using FreeSurfer (v5.3, eTIV) (Dale AM et al., 1999). White matter hyperintensity burden represented the total burden due to both perivascular and deep white matter hyperintensities. Normal appearing white matter was all white matter excluding the hyperintense voxels considered as lesions and identified using the FLAIR image.

The standard space MNI atlas was resampled to 3 mm space. The FLAIR images were then downsampled to an isotropic resolution of 3mm3. The CBF and CVR images were then subsequently, registered and resampled to match 3 mm MNI space and the corresponding FLAIR images.

The white matter hyperintensity mask was then dilated by 1 voxel (3mm) in all directions to obtain the penumbra of tissue surrounding the hyperintensity that is likely at risk of tissue infarction. This dilation process was repeated 5 times to assess the spatial changes in CBF and CBF in the surround of the white matter hyperintensities. Percent decrease in CBF and CVR were calculated using the outermost layer of tissue as baseline (15 mm from the hyperintensity) with 5 spatial points. Curve fitting was performed in MATLAB R2016b® to determine the relationship between percent decrease in CBF or CVR and distance from the white matter hyperintensity. CBF and CVR values within the white matter hyperintensities, and 3, 6, 9, 12, 15 mm away from the hyperintensities were measured and compared using a paired t-test.

Correlation of gray matter CBF and CVR as well as white matter hyperintensity burden (normalized by total intracranial volume), with MMSE as a measure of overall cognitive function was evaluated. Since white matter hyperintensity burden is strongly associated with task processing speeds, we also evaluated the sensitivity of CBF and CVR to performance on the Trail Making Test, Part B as a measure of executive function using Spearman’s correlation (ρ).

3. RESULTS

Of the 28 individuals, 7 subjects were identified with mild cognitive impairment based on consensus diagnosis as outlined above. For the following results, all subjects were considered as a single group except while evaluating the association between cognitive performance, CBF, CVR, and hyperintensity burden. For these tests, the associations were adjusted for age and cognitive status.

3.1 Multi-modal vascular imaging markers

Figure 1 shows group (A) and single subject (B) maps for CVR (top row), CBF (middle row), and white matter hyperintensities (bottom row). The bottom row shows the regional distribution of white matter hyperintensities across all subjects in the group level. Red indicates higher frequency of finding white matter hyperintensities while blue indicates a lower frequency of white matter hyperintensities. In B, the green mask represents white matter hyperintensities segmentation in a single subject. The CVR maps show higher BOLD signal intensity change across all gray matter representing a vasodilatory response following a breath-hold stimulus. In the single subject map, a negative BOLD response is also observed near the location of the white matter hyperintensities. The negative response is likely due to steal phenomenon i.e., (Mandell, et al, 2008; Poublanc et al., 2013) diversion of blood flow from vasculature perfusing the white matter hyperintensities to surrounding brain regions with healthy tissue structure or due to the compression of cerebrospinal fluid in the ventricular compartment, also in the vicinity of the white matter hyperintensities. The average gray matter CBF was 57.9± 8.1 ml/100g/min and gray matter CVR was 1.0± 0.03%. Average white matter burden (including both periventricular and deep white matter) was 0.21± 0.21 % of the total intracranial volume (ICV).

Figure 1.

Figure 1

Open in a new tab

Group level and single subject maps of percent BOLD signal change (CVR) during a breath-hold paradigm (top row), CBF (middle row), frequency of white matter hyperintensities (bottom row). CVR maps represent the magnitude of the dilatatory response of the cerebral vasculature to a breath-hold. The group white matter map is a map of frequency of the occurrences of white matter hyperintensities. Red voxels indicate a higher frequency; more subjects had white matter hyperintensities in those voxel locations and blue indicates a lower frequency; fewer subjects had infarcted white matter in the corresponding voxel locations. The single subject CVR map shows negative BOLD signal near the white matter hyperintensities, likely due to steal phenomenon or due to a combined effect of partial voluming and CSF fluid movement in ventricles during breath-hold.

3.2 Correlation of vascular markers with cognitive performance

CBF and CVR were significantly correlated with each other (Figure 2a, r = 0.34, p <0.05, n = 28). Linear regression model, after adjusting for age and diagnosis, showed a trend towards correlation between MMSE and CBF (p = 0.1; Pearson’s correlation between CBF and MMSE, r = 0.40, and a significant correlation between MMSE and CVR (p = 0.02; r = 0.42). Note: The p-values reflect the regression output showing association of CBF and CVR with MMSE after accounting for the effect of age and diagnosis. After adjusting for age and diagnosis, TRAILS B scores were not related significantly with CBF (p = 0.72; r = 0.05) or CVR (p = 0.45; r = 0.10). High white matter hyperintensity burden was associated with poorer MMSE scores and with longer processing times on the TRAILS B test (Figure 2b, p = 0.03; r = −0.40 for MMSE and 0.006; r = 0.52 for TRAILS B). No correlation was observed with Logical Memory scores for CBF (p = 0.22; r = −0.12), CVR (p = 0.63; r = 0.05), or white matter hyperintensities (p = 0.24; r = −0.13).

Figure 2.

Figure 2

Open in a new tab

Overall cortical CBF and CVR values are directly proportional to cognitive status of subjects. Subjects with low CBF and low CVR had poor MMSE scores. Note that although the correlation between CBF and CVR is statistically significant, only 12.25%(R2) of the variance in the data is explained by this relationship. High white matter hyperintensities burden (Number of voxels with white matter hyperintensities divided by total intracranial volume) was a strong indicator of poor performance on MMSE and TRAILS. High CVR was related to high MMSE scores (significant) and quicker processing times on TRAILS (marginally significant). Relationship between CBF and cognitive performance was the weakest of the three vascular pathology markers in healthy older adults. Logical memory did not appear to be correlated with any marker. r = Pearson’s correlation, p significance of association between cognitive test and vascular marker after adjusting for age and diagnosis.

Additionally, although not significant, the APOE-ε4 carriers had lower CBF (n = 5, 57.5±4.23 ml/100g/min), CVR (0.99±0.08), and higher white matter hyperintensity burden (0.33±0.19) compared to the non-carriers (CBF: n = 23, 57.9±9.07 ml/100g/min; CVR: 1.00±0.37; hyperintensity burden: 0.17±21). Of the 5 APOE-ε4 carriers 4 were cognitively normal older adults and 1 individual had mild cognitive impairment.

3.3 Regional variations in CBF and CVR

Figure 3 depicts regional CBF and CVR values in all gray matter as well as white matter and white matter hyperintensities. Table 2 lists the CBF and CVR in the different brain regions. No significant difference was observed between the CBF values in the different gray matter regions. CBF in the white matter and white matter hyperintensities was significantly different from all gray matter regions. CBF and CVR in the white matter hyperintensities were both significantly lower than all other brain regions. CBF and CVR in the white matter were also significantly (p<0.05) lower than that in the gray matter regions.

Figure 3.

Figure 3

Open in a new tab

Plots 3a and 3b show CBF and CVR in the gray matter (GM), white matter (WM), and white matter hyperintensities (WMH). White matter includes all brain white matter but not the white matter hyperintensities. CBF is significantly different between gray matter and white matter, as well as between the white matter and white matter hyperintensities. CVR also, is significantly different between gray matter, white matter, and white matter hyperintensities. (* p<0.05).

Table 2.

Regional variations in CBF and CVR in older adults.

N=28 CBF (ml/100g/min) CVR (% signal)
REGION MEAN STD. DEV MEAN STD. DEV
Frontal cortex 57.2 8.3 1.00 0.40
Temporal cortex 63.1 9.0 0.94 0.32
Parietal cortex 65.8 9.6 1.18 0.42
Occipital cortex 42.2 16.3 1.35 0.54
Insula 55.4 6.7 0.61 0.25
Caudate 52.9 18.2 0.55 0.32
Putamen 45.7 9.1 1.50 0.84
Thalamus 39.2 12.7 2.41 0.91
Hippocampus 41.7 8.6 0.87 0.48
Amygdala 70.1 25.9 0.98 0.57
Total gray matter 57.9 8.1 1.00 0.03
Total white matter 33.1 5.3 0.32 0.11
White matter hyperintensities* 16.9 5.9 0.12 0.15
Open in a new tab
*

includes both periventricular and deep white matter hyperintensities

The size of white matter hyperintensities was inversely proportional to the gray matter CBF (n = 28, r = −0.10, not significant) as well as CVR (n = 28, r = −0.16, not significant).

3.4 Spatial patterns of CBF and CVR in and around white matter hyperintensities

Figure 4 shows how CBF and CVR vary in the tissue surrounding the white matter hyperintensities. Average CBF values within the hyperintensities were 13.9±5.9 ml/100g/min. CBF was 17.6±5.2, 22.8±5.59, 29.5±5.3, 33.0±4.9, and 33.1±5.0 ml/100g/min at 3, 6, 9, 12, and 15 mm from the hyperintensity. Average CVR values within the hyperintensities were 0.12±0.15 %. It was 0.16±0.10, 0.23±0.10, 0.32±0.11, 0.38±0.13, and 0.43±0.14 % at 3, 6, 9, 12, and 15 mm from the hyperintensity and significantly different (p<0.005) from CVR in the hyperintensity. CBF values between 0, 3, 6, and 9 mm distance from the white matter hyperintensities were significantly different from each other. CBF values in the white matter hyperintensities decreased by 58.4±13.9% compared to that at 15 mm. There was no difference between CBF values at 9 and 12 mm from the white matter hyperintensities. However, CVR values were significantly different at 0, 3, 6, 9, as well 12 mm distance from the white matter hyperintensities. CVR values in the white matter hyperintensities decreased by 79.3±48.9% (significantly greater than CBF decreases, p <0.05) compared to that at 15 mm.

Figure 4.

Figure 4

Open in a new tab

Top row: CBF (closed black circles) and CVR (open black circles) changes within white matter hyperintensities (0 mm - red) and 3 (orange), 6 (yellow), 9 (green), 12 (blue), 15 (red) mm away from the hyperintensity as shown in the inset image. Both CBF and CVR decrease as we approach the white matter hyperintensity. CVR decreases (79.3±48.9%) are significantly higher than CBF decreases (58.4±13.9%). Bottom 2 rows indicate residual for the linear and non-linear fits of CBF and CVR respectively.

Percent decrease in CBF was non-linearly proportional to the distance from the white matter hyperintensity i.e., closer the tissue voxel to the hyperintensity, greater the decrease in CBF (R2 = 0.98, adjusted R2 = 0.98 for a cubic fit). Percent decrease in CVR was also non-linearly proportional to the distance from the white matter hyperintensity (R2 = 0.95, adjusted R2 = 0.95 for a cubic fit). Residual plots for both fits are shown below. We therefore, conclude that CVR is a more sensitive measure of cerebrovascular morbidity compared to CBF.

4. DISCUSSION

In this work, we show that (1) compared to CBF, CVR is a more sensitive measure of compromised vasculature. (2) Both CBF and CVR correlate with MMSE status. Only CVR was correlated with TRAILS B. White matter hyperintensity burden was correlated with MMSE as well as processing speeds in TRAILS B. Based on our finding we believe that CVR can be an early marker of tissue ischemia and can be applied to evaluate risk of developing white matter hyperintensities.

4.1 CBF and CVR as imaging markers of vascular pathology in small vessel disease

In Figure 4, we show that CVR decreases much more rapidly than CBF. Although CBF and CVR are strongly positively correlated, CVR decreases in the tissue surrounding the white matter hyperintensities are larger than the corresponding CBF decreases. This is an important finding. CBF reductions are in part due to neurodegeneration and likely also due to loss of vascular reserve. This is in accordance with recent studies showing slower reductions in CBF in AD compared to reductions in CVR (Liu et al., 2014; Gao et al., 2013). It is therefore likely that CVR is an earlier predictor of vascular failure and tissue infarction compared to CBF.

4.2 Paradigm selection for CVR measurement

In this study, we used a breath-hold paradigm for CVR calculations. Several issues need to be discussed. There are multiple ways to measure CVR. BOLD is a semi-quantitative approach compared to ASL-derived CVR (Faraco et al., 2015, Donahue et al., 2014). Since BOLD has a significantly higher SNR and has greater temporal resolution than ASL, we believe that BOLD signal response is more sensitive than the ASL signal for a breath-hold challenge. The change in EtCO2 in a breath-hold task was found to be approximately 3 mm Hg compared to typical hypercapnic gas challenges, where the change in EtCO2 is about 15mm Hg in young adults (< 40 years of age) (Murphy et al., 2011). It is likely that this EtCO2 change is even smaller in magnitude in older adults. Our cortical GM CVR was 1.00± 0.03% for 15s breath-hold. This result is similar to the CVR obtained by other studies using a paced breath-hold paradigm (<1% Chang et. al (2009), 0.43% for 9s breath-hold Magon et. al (2009)). Although the magnitude of CVR response with a hypercapnic gas challenge is an order of magnitude higher (2–5%) than the breath-hold challenge, our preference for breath-hold was due to the ease of implementation. No additional equipment was needed. We believe that the absence of the breathing apparatus increases subject co-operation. The drawback is that the study relies heavily on accurate placement of the respiratory bellows. The subjects in this study were trained outside the scanner to perform the breath-hold paradigm until they are able to follow the written instructions well.

4.3 Perfusion measurements using ASL

CBF measurements using ASL is now a routine measurement that is readily available in the clinic. We used this method to measure perfusion in older adults and found correlation between cortical CBF and MMSE scores. This correlation between regional perfusion and cognitive tests has been observed in many other studies. It is likely that reduced CBF can serve as a functional MRI marker of cerebrovascular disease and poor cognitive performance. We also show some regional variations in CBF however, they were not significant. Note that most studies use a label duration around 1500 ms and a post labeling delay of 1650 ms, which is insufficient to accurately quantify white matter CBF and is likely unsuitable for quantitative comparison across groups (van Osch et al., 2009). However, arterial spin labeling signal in the white matter hyperintensities and white matter will still be proportional to tissue perfusion and hence can provide information regarding slowed perfusion in abnormal tissue compared to normal tissues within a subject and as shown in all previous studies of perfusion in white matter hyperintensities (Brickman et al., 2009; Dalen et al., 2016; Promjunyakul et al., 2015).

4.4 Location of white matter hyperintensities

We calculated a total white matter hyperintensity burden without separating periventricular and deep white matter contributions. Our current analyses did not support the separation of the hyperintensities based on location. The pathologic origins of the two hyperintensities are believed to be distinct; Periventricular hyperintensities occur due to venous collagenosis and deep white matter hyperintensities due to arterial insufficiency (Pantoni and Garcia, 1997). It is likely that the reduced CBF and CVR may be more strongly associated with deep white matter hyperintensities. Future work will involve separating the white matter hyperintensities based on location and evaluating correlations with CBF, CVR, and cognitive performance.

5. CONCLUSION

We show feasibility of ASL and breath-hold BOLD to evaluate CBF and CVR as markers of cerebrovascular disease in older adults. While both CBF and CVR are indicators of declining cognitive performance in this cohort, our study showed that CVR is a more sensitive marker of tissue infarction and cerebrovascular disease compared to CBF. Future studies will investigate whether reduced CVR around the white matter hyperintensities can be measured longitudinally to predict tissue voxels at risk of infarction.

Supplementary Material

supplement. Supplementary Figure A.1.

Strong correlation between CBF values calculated with an acquired M0 image and an estimated M0 image. Quantified CBF values using the estimated M0 image were very similar to (0.97±0.06) times the CBF values using a true/acquired M0 image.

Highlights.

  • Cerebral blood flow (CBF) and cerebrovascular reserve (CVR) is measured in and around white matter hyperintensities as markers of small vessel disease and cognitive function.

  • Association of CVR and cognitive performance is stronger than the association of CBF with cognitive performance.

  • CVR decreases faster than CBF in tissue at risk of developing white matter hyperintensities and is likely a more sensitive vascular pathology marker than CBF.

Acknowledgments

Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Institutional funds were provided by University of Washington School of Medicine to TG. Analysis effort for SRL were partly supported by K01AG055669 (NIH/NIA).

Footnotes

Author Contributions: SRL, and NK are responsible for processing and analyses of data and manuscript preparation. MA and PB are responsible for experimental setup and data acquisition, TM assisted with software tools for data analysis. TG, BC, SJ, and EL coordinated the MRI and subject recruitment as well as supervised the manuscript preparation.

Conflicts of Interest: None noted

Verification:
  1. The authors do not have any actual or potential conflicts of interest including any financial, personal or other relationships with other people or organizations within three years of beginning the work submitted that could inappropriately influence (bias) their work. Examples of potential conflicts of interest which should be disclosed include employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/registrations, and grants or other funding.
  2. No author's institution has contracts relating to this research through which it or any other organization may stand to gain financially now or in the future.
  3. There are no other agreements of authors or their institutions that could be seen as involving a financial interest in this work.
  4. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Institutional funds were provided by University of Washington School of Medicine to Thomas Grabowski.
  5. Parts of the work in this manuscript was presented at AAIC 2016 and 2017 as conference abstracts.
  6. All participants provided written informed consent to participate in the present study, which was conducted according to the Declaration of Helsinki and subsequent revisions. The study was approved by the Institutional Review Board at University of Washington, Seattle.
  7. All authors have reviewed the contents of the manuscript being submitted, approve of its contents and validate the accuracy of the data.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Al-Bachari S, Parkes LM, Vidyasagar R, Hanby MF, Tharaken V, Leroi I, Emsley HC. Arterial spin labelling reveals prolonged arterial arrival time in idiopathic Parkinson's disease. NeuroImage: Clinical. 2014;6:1–8. doi: 10.1016/j.nicl.2014.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alsop DC, Detre JA, Golay X, Günther M, Hendrikse J, Hernandez-Garcia L, Lu H, MacIntosh BJ, Parkes LM, Smits M, Osch MJ. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magnetic resonance in medicine. 2015;73(1):102–16. doi: 10.1002/mrm.25197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Au R, Massaro JM, Wolf PA, Young ME, Beiser A, Seshadri S, D’Agostino RB, DeCarli C. Association of white matter hyperintensity volume with decreased cognitive functioning: the Framingham Heart Study. Archives of neurology. 2006;63(2):246–50. doi: 10.1001/archneur.63.2.246. [DOI] [PubMed] [Google Scholar]
  4. Bahrani AA, Powell DK, Yu G, Johnson ES, Jicha GA, Smith CD. White Matter Hyperintensity Associations with Cerebral Blood Flow in Elderly Subjects Stratified by Cerebrovascular Risk. Journal of Stroke and Cerebrovascular Diseases. 2017 doi: 10.1016/j.jstrokecerebrovasdis.2016.10.017. (available online) [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beekly DL, Ramos EM, Lee WW, Deitrich WD, Jacka ME, Wu J, Hubbard JL, Koepsell TD, Morris JC, Kukull WA. The National Alzheimer's Coordinating Center (NACC) database: the uniform data set. Alzheimer Disease & Associated Disorders. 2007 Jul 1;21(3):249–58. doi: 10.1097/WAD.0b013e318142774e. [DOI] [PubMed] [Google Scholar]
  6. Birn RM, Smith MA, Jones TB, Bandettini PA. The respiration response function: the temporal dynamics of fMRI signal fluctuations related to changes in respiration. Neuroimage. 2008;40(2):644–54. doi: 10.1016/j.neuroimage.2007.11.059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarction the Rotterdam Study. Circulation. 1997;96(5):1432–7. doi: 10.1161/01.cir.96.5.1432. [DOI] [PubMed] [Google Scholar]
  8. Brickman AM, Zahra A, Muraskin J, Steffener J, Holland CM, Habeck C, Borogovac A, Ramos MA, Brown TR, Asllani I, Stern Y. Reduction in cerebral blood flow in areas appearing as white matter hyperintensities on magnetic resonance imaging. Psychiatry Research: Neuroimaging. 2009;172(2):117–20. doi: 10.1016/j.pscychresns.2008.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bright MG, Murphy K. Reliable quantification of BOLD fMRI cerebrovascular reactivity despite poor breath-hold performance. NeuroImage. 2013;83:559–68. doi: 10.1016/j.neuroimage.2013.07.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. de Groot JC, Oudkerk M, Gijn JV, Hofman A, Jolles J, Breteler MM. Cerebral white matter lesions and cognitive function: the Rotterdam Scan Study. Annals of neurology. 2000;47(2):145–51. doi: 10.1002/1531-8249(200002)47:2<145::aid-ana3>3.3.co;2-g. [DOI] [PubMed] [Google Scholar]
  11. Chang C, Glover GH. Relationship between respiration, end-tidal CO 2, and BOLD signals in resting-state fMRI. Neuroimage. 2009;47(4):1381–93. doi: 10.1016/j.neuroimage.2009.04.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chen L, Bernstein M, Huston J, Fain S. Measurements of T1 relaxation times at 3.0 T: implications for clinical MRA. Proceedings of the 9th Annual Meeting of ISMRM; Glasgow, Scotland. 2001. [Google Scholar]
  13. Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis: I. Segmentation and surface reconstruction. Neuroimage. 1999;9(2):179–94. doi: 10.1006/nimg.1998.0395. [DOI] [PubMed] [Google Scholar]
  14. Detre JA, Zhang W, Roberts DA, Silva AC, Williams DS, Grandis DJ, Kortesky AP, Leigh JS. Tissue specific perfusion imaging using arterial spin labeling. NMR in Biomedicine. 1994;7(1–2):75–82. doi: 10.1002/nbm.1940070112. [DOI] [PubMed] [Google Scholar]
  15. Donahue MJ, Faraco CC, Strother MK, Chappell MA, Rane S, Dethrage LM, Hendrikse J, Siero JC. Bolus arrival time and cerebral blood flow responses to hypercarbia. Journal of Cerebral Blood Flow & Metabolism. 2014;34(7):1243–52. doi: 10.1038/jcbfm.2014.81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dufouil C, Chalmers J, Coskun O, Besancon V, Bousser MG, Guillon P, Macmahon S, Mazoyer B, Neal B, Woodward M, Tzourio-Mazoyer N. Effects of blood pressure lowering on cerebral white matter hyperintensities in patients with stroke. Circulation. 2005;112(11):1644–50. doi: 10.1161/CIRCULATIONAHA.104.501163. [DOI] [PubMed] [Google Scholar]
  17. Esiri MM, Nagy Z, Smith MZ, Barnetson L, Smith AD. Cerebrovascular disease and threshold for dementia in the early stages of Alzheimer's disease. The Lancet. 1999;354(9182):919–20. doi: 10.1016/S0140-6736(99)02355-7. [DOI] [PubMed] [Google Scholar]
  18. Farkas E, Luiten PG. Cerebral microvascular pathology in aging and Alzheimer's disease. Progress in neurobiology. 2001;64(6):575–611. doi: 10.1016/s0301-0082(00)00068-x. [DOI] [PubMed] [Google Scholar]
  19. Faraco CC, Strother MK, Dethrage LM, Jordan L, Singer R, Clemmons PF, Donahue MJ. Dual echo vessel-encoded ASL for simultaneous BOLD and CBF reactivity assessment in patients with ischemic cerebrovascular disease. Magnetic resonance in medicine. 2015;73(4):1579–92. doi: 10.1002/mrm.25268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. Journal of psychiatric research. 1975;12(3):189–98. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  21. Gao YZ, Zhang JJ, Liu H, Wu GY, Xiong L, Shu M. Regional cerebral blood flow and cerebrovascular reactivity in Alzheimer's disease and vascular dementia assessed by arterial spin labeling magnetic resonance imaging. Current neurovascular research. 2013;10(1):49–53. doi: 10.2174/156720213804806016. [DOI] [PubMed] [Google Scholar]
  22. Gibson E, Gao F, Black SE, Lobaugh NJ. Automatic segmentation of white matter hyperintensities in the elderly using FLAIR images at 3T. Journal of Magnetic Resonance Imaging. 2010;31(6):1311–22. doi: 10.1002/jmri.22004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gückel FJ, Brix G, Schmiedek P, Piepgras Z, Becker G, Köpke J, Gross H, Georgi M. Cerebrovascular reserve capacity in patients with occlusive cerebrovascular disease: assessment with dynamic susceptibility contrast-enhanced MR imaging and the acetazolamide stimulation test. Radiology. 1996;201(2):405–12. doi: 10.1148/radiology.201.2.8888232. [DOI] [PubMed] [Google Scholar]
  24. Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage. 2002;17(2):825–41. doi: 10.1016/s1053-8119(02)91132-8. [DOI] [PubMed] [Google Scholar]
  25. Kastrup A, Li TQ, Takahashi A, Glover GH, Moseley ME. Functional magnetic resonance imaging of regional cerebral blood oxygenation changes during breath holding. Stroke. 1998;29(12):2641–5. doi: 10.1161/01.str.29.12.2641. [DOI] [PubMed] [Google Scholar]
  26. Liao D, Cooper L, Cai J, Toole JF, Bryan NR, Hutchinson RG, Tyroler HA. Presence and severity of cerebral white matter lesions and hypertension, its treatment, and its control. Stroke. 1996;27(12):2262–70. doi: 10.1161/01.str.27.12.2262. [DOI] [PubMed] [Google Scholar]
  27. Liem MK, Oberstein SL, Haan J, Boom R, Ferrari MD, Buchem M, Grond J. Cerebrovascular reactivity is a main determinant of white matter hyperintensity progression in CADASIL. American Journal of Neuroradiology. 2009 Jun 1;30(6):1244–7. doi: 10.3174/ajnr.A1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liu P, Lu H, Rodrigue K, Kennedy K, Peng SL, Devous MD, Park DC. Impact of beta-Amyloid burden on brain perfusion and vascular reactivity in normal aging. Alzheimer's & Dementia: The Journal of the Alzheimer's Association. 2014;10(4):P143–4. [Google Scholar]
  29. Mack WJ, Freed DM, Williams BW, Henderson VW. Boston Naming Test: shortened versions for use in Alzheimer's disease. Journal of gerontology. 1992;47(3):P154–8. doi: 10.1093/geronj/47.3.p154. [DOI] [PubMed] [Google Scholar]
  30. Magon S, Basso G, Farace P, Ricciardi GK, Beltramello A, Sbarbati A. Reproducibility of BOLD signal change induced by breath holding. Neuroimage. 2009;45(3):702–12. doi: 10.1016/j.neuroimage.2008.12.059. [DOI] [PubMed] [Google Scholar]
  31. Makedonov I, Black SE, MacIntosh BJ. Cerebral small vessel disease in aging and Alzheimer's disease: a comparative study using MRI and SPECT. European journal of neurology. 2013;20(2):243–250. doi: 10.1111/j.1468-1331.2012.03785.x. [DOI] [PubMed] [Google Scholar]
  32. Mandell DM, Han JS, Poublanc J, Crawley AP, Kassner A, Fisher JA, Mikulis DJ. Selective reduction of blood flow to white matter during hypercapnia corresponds with leukoaraiosis. Stroke. 2008;39(7):1993–8. doi: 10.1161/STROKEAHA.107.501692. [DOI] [PubMed] [Google Scholar]
  33. Marstrand JR, Garde E, Rostrup E, Ring P, Rosenbaum S, Mortensen EL, Larsson HB. Cerebral perfusion and cerebrovascular reactivity are reduced in white matter hyperintensities. Stroke. 2002;33(4):972–6. doi: 10.1161/01.str.0000012808.81667.4b. [DOI] [PubMed] [Google Scholar]
  34. Molina C, Sabín JA, Montaner J, Rovira A, Abilleira S, Codina A. Impaired cerebrovascular reactivity as a risk marker for first-ever lacunar infarction. Stroke. 1999;30(11):2296–2301. doi: 10.1161/01.str.30.11.2296. [DOI] [PubMed] [Google Scholar]
  35. Montine T, Sonnen AJ, Montine SK, Crane KP, Larson BE. Adult Changes in Thought study: dementia is an individually varying convergent syndrome with prevalent clinically silent diseases that may be modified by some commonly used therapeutics. Current Alzheimer Research. 2012;9(6):718–23. doi: 10.2174/156720512801322555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Murphy K, Harris AD, Wise RG. Robustly measuring vascular reactivity differences with breath-hold: normalising stimulus-evoked and resting state BOLD fMRI data. NeuroImage. 2011;54(1):369–79. doi: 10.1016/j.neuroimage.2010.07.059. [DOI] [PubMed] [Google Scholar]
  37. Novak V, Last D, Alsop DC, Abduljalil AM, Hu K, Lepicovsky L, Cavallerano J, Lipsitz LA. Cerebral blood flow velocity and periventricular white matter hyperintensities in type 2 diabetes. Diabetes care. 2006;29(7):1529–1534. doi: 10.2337/dc06-0261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pantoni L. Cerebral small vessel disease: from pathogenesis and clinical characteristics to therapeutic challenges. The Lancet Neurology. 2010;9(7):689–701. doi: 10.1016/S1474-4422(10)70104-6. [DOI] [PubMed] [Google Scholar]
  39. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis. Stroke. 1997;28(3):652–659. doi: 10.1161/01.str.28.3.652. [DOI] [PubMed] [Google Scholar]
  40. Prins ND, Scheltens P. White matter hyperintensities, cognitive impairment and dementia: an update. Nature Reviews Neurology. 2015;11(3):157–65. doi: 10.1038/nrneurol.2015.10. [DOI] [PubMed] [Google Scholar]
  41. Poublanc J, Han JS, Mandell DM, Conklin J, Stainsby JA, Fisher JA, Mikulis DJ, Crawley AP. Vascular steal explains early paradoxical blood oxygen level-dependent cerebrovascular response in brain regions with delayed arterial transit times. Cerebrovascular diseases extra. 2013 Apr 12;3(1):55–64. doi: 10.1159/000348841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Promjunyakul N, Lahna D, Kaye JA, Dodge HH, Erten-Lyons D, Rooney WD, Silbert LC. Characterizing the white matter hyperintensity penumbra with cerebral blood flow measures. NeuroImage: Clinical. 2015;8:224–9. doi: 10.1016/j.nicl.2015.04.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Reitan RM. Trail Making Test: Manual for administration and scoring. Reitan Neuropsychology Laboratory; 1992. [Google Scholar]
  44. Sam K, Crawley AP, Conklin J, Poublanc J, Sobczyk O, Mandell DM, Venkatraghavan L, Duffin J, Fisher JA, Black SE, Mikulis DJ. Development of White Matter Hyperintensity is Preceded by Reduced Cerebrovascular Reactivity. Ann Neurol. 2016 Aug;80(2):277–85. doi: 10.1002/ana.24712. [DOI] [PubMed] [Google Scholar]
  45. Silva AC, Kim SG. Pseudo-continuous arterial spin labeling technique for measuring CBF dynamics with high temporal resolution. Magnetic resonance in medicine. 1999;42(3):425–9. doi: 10.1002/(sici)1522-2594(199909)42:3<425::aid-mrm3>3.0.co;2-s. [DOI] [PubMed] [Google Scholar]
  46. Smith EE, Salat DH, Jeng J, McCreary CR, Fischl B, Schmahmann JD, Dickerson BC, Viswanathan A, Albert MS, Blacker D, Greenberg SM. Correlations between MRI white matter lesion location and executive function and episodic memory. Neurology. 2011;76(17):1492–1499. doi: 10.1212/WNL.0b013e318217e7c8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. van Dalen JW, Mutsaerts HJ, Nederveen AJ, Vrenken H, Steenwijk MD, Caan MW, Majoie CB, van Gool WA, Richard E. White Matter Hyperintensity Volume and Cerebral Perfusion in Older Individuals with Hypertension Using Arterial Spin-Labeling. American Journal of Neuroradiology. 2016;37(10):1824–30. doi: 10.3174/ajnr.A4828. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. van Osch MJ, Teeuwisse WM, van Walderveen MA, Hendrikse J, Kies DA, van Buchem MA. Can arterial spin labeling detect white matter perfusion signal? Magnetic resonance in medicine. 2009;62(1):165–73. doi: 10.1002/mrm.22002. [DOI] [PubMed] [Google Scholar]
  49. Wansapura JP, Holland SK, Dunn RS, Ball WS. NMR relaxation times in the human brain at 3.0 tesla. Journal of magnetic resonance imaging. 1999;9(4):531–8. doi: 10.1002/(sici)1522-2586(199904)9:4<531::aid-jmri4>3.0.co;2-l. [DOI] [PubMed] [Google Scholar]
  50. Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O'Brien TJ, Barkhof F, Benavente OR, Black SE. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. The Lancet Neurology. 2013;12(8):822–38. doi: 10.1016/S1474-4422(13)70124-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wechsler D. Wechsler Adult Intelligence Scale–Fourth Edition (WAIS–IV) 2014. [Google Scholar]
  52. Wechsler D. A standardized memory scale for clinical use. The Journal of Psychology. 1945;19(1):87–95. [Google Scholar]
  53. White L, Petrovitch H, Ross GW, Masaki KH, Abbott RD, Teng EL, Rodriguez BL, Blanchette PL, Havlik RJ, Wergowske G, Chiu D. Prevalence of dementia in older Japanese-American men in Hawaii: the Honolulu-Asia aging study. JAMA. 1996;276(12):955–60. [PubMed] [Google Scholar]
  54. Wong EC, Buxton RB, Frank LR. Quantitative perfusion imaging using arterial spin labeling. Neuroimaging Clinics of North America. 1999;9(2):333–42. [PubMed] [Google Scholar]
  55. Woolrich MW, Jbabdi S, Patenaude B, Chappell M, Makni S, Behrens T, Beckmann C, Jenkinson M, Smith SM. Bayesian analysis of neuroimaging data in FSL. Neuroimage. 2009;45(1):S173–86. doi: 10.1016/j.neuroimage.2008.10.055. [DOI] [PubMed] [Google Scholar]
  56. Zhang Y, Brady M, Smith S. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE transactions on medical imaging. 2001;20(1):45–57. doi: 10.1109/42.906424. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement. Supplementary Figure A.1.

Strong correlation between CBF values calculated with an acquired M0 image and an estimated M0 image. Quantified CBF values using the estimated M0 image were very similar to (0.97±0.06) times the CBF values using a true/acquired M0 image.

NIHMS936648-supplement.eps (7KB, eps)

RESOURCES