Neurobiology of Aging 35 (2014) 2665e2670
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Brain volume and white matter hyperintensities as determinants of cerebral blood flow in Alzheimer’s disease Marije R. Benedictus a, *, Maja A.A. Binnewijzend b, Joost P.A. Kuijer c, Martijn D. Steenwijk b, Adriaan Versteeg b, Hugo Vrenken b, Philip Scheltens a, Frederik Barkhof b, Wiesje M. van der Flier a, d, Niels D. Prins a a
Alzheimer Center & Department of Neurology, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands Department of Radiology and Nuclear medicine, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands Department of Physics and Medical Technology, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands d Department of Epidemiology and Biostatistics, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 September 2013 Received in revised form 13 January 2014 Accepted 7 June 2014 Available online 13 June 2014
To better understand whether decreased cerebral blood flow (CBF) in patients with Alzheimer’s disease (AD) reflects neurodegeneration or cerebral small vessel disease, we investigated the associations of normalized brain volume (NBV) and white matter hyperintensity (WMH) volume with CBF. We included 129 patients with AD (66 7 years, 53% female) and 61 age-matched controls (64 5 years, 43% female). CBF was measured with pseudocontinuous arterial spin labeling at 3T in the whole brain and in partial volume corrected cortical maps. When NBV and WMH were simultaneously entered in age and sex adjusted models, smaller NBV was associated with lower whole brain (Stb: 0.29; p < 0.01) and cortical CBF (Stb: 0.28; p < 0.01) in patients with AD. Larger WMH volume was also associated with lower whole brain (Stb: 0.22; p < 0.05) and cortical CBF (Stb: 0.24; p < 0.05) in AD. Additional adjustments did not change these results. In controls, neither NBV nor WMH was associated with CBF. Our results indicate that in AD, lower CBF as measured using pseudocontinuous arterial spin labeling, reflects the combined disease burden of both neurodegeneration and small vessel disease. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Arterial spin labeling Cerebral blood flow Alzheimer’s disease Neurodegeneration Cerebral small vessel disease
1. Introduction Alzheimer’s disease (AD) is essentially regarded as a neurodegenerative disease, characterized by the accumulation of amyloid plaques and neurofibrillary tangles that eventually leads to brain atrophy (Jack et al., 2010). Increasing evidence indicates that AD patient not only have brain volume loss, but also often have an altered cerebral blood flow (CBF). Although some studies report relative regional increases in (early) AD (Alsop et al., 2008; Dai et al., 2009), the most consistent finding is a decrease in absolute CBF in patients with AD (Alexopoulos et al., 2012; Alsop et al., 2010; Binnewijzend et al., 2013). This decreased CBF is in general assumed to be a reflection of the neurodegenerative process (Wolk and Detre, 2012). Lower CBF in AD patients may not only relate to neurodegeneration, but may also be associated with small vessel disease * Corresponding author at: Department of Neurology and Alzheimer Center, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, the Netherlands. Tel.: þ31 204440183; fax: þ31 204448529. E-mail address: [email protected] (M.R. Benedictus). 0197-4580/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.06.001
(SVD). White matter hyperintensities (WMH) of presumed vascular origin are a commonly used magnetic resonance imaging (MRI) marker to indicate the presence of SVD (Wardlaw et al., 2013). WMH are assumed to result from ischemia and they are more prevalent in AD patients compared with the general elderly population (Scheltens et al., 1995). Previous studies have shown that WMH are associated with lower CBF as well (Bastos-Leite et al., 2008; Schuff et al., 2009; Vernooij et al., 2008). CBF can be measured by arterial spin labeling (ASL); a functional MRI technique that uses magnetically labeled arterial blood water as an endogenous tracer (Petersen et al., 2006). The pseudocontinuous variant of ASL (PCASL) uses a multitude of millisecondlong pulses to achieve a high labeling efficiency and effective compensation of magnetization transfer effects (Dai et al., 2008). Neurodegeneration and SVD are common in patients with AD, but to our knowledge, no previous studies have investigated how both processes relate to the generally described decreased CBF. Our aim was to explore whether independent relationships exist between normalized brain volume (NBV) or WMH on the one hand and CBF on the other hand. We hypothesized that the lower CBF in AD is not only reflective of the neurodegenerative process, but that
2666
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
CBF may be even further decreased when SVD is present. The well-characterized Amsterdam Dementia Cohort with PCASL measurement allowed us to investigate the determinants of CBF in AD patients and controls. 2. Methods 2.1. Subjects Subjects for this study were drawn from the memory clinic based Amsterdam Dementia Cohort. We included 129 AD patients and 61 age-matched controls who visited our memory clinic between October 2010 and June 2012. All subjects underwent an extensive dementia screening, including medical history, neurologic, and physical examination, cognitive assessment, and brain MRI. The diagnosis “probable AD” was made according to the NINCDS-ADRDA criteria, by consensus of a multidisciplinary team and all patients fulfilled the core clinical criteria of the NIA-AA (McKhann et al., 1984, 2011). The control group consisted of agematched control subjects, who presented with cognitive complaints, but for whom clinical investigations were normal and criteria for mild cognitive impairment (Petersen et al., 2001), dementia, or any other neurologic or psychiatric disorder were not met. As subjective complaints may represent preclinical AD in a subgroup (Sperling et al., 2011), we only included subjects with normal cerebrospinal fluid (CSF) Ab42 levels (see Mulder et al. (2010) for a detailed description of CSF analyses). For all subjects, the presence of hypertension, hypercholesteremia, and diabetes mellitus were determined based on self-reported medical history and medication use. Smoking status was defined as never, former, or current. Blood pressures were measured manually using a sphygmomanometer. Exclusion criteria were the presence of structural brain lesions and failure of preprocessing of the MRI scans. The ethical review board of the VU University Medical Center approved the study. We obtained informed consent from all patients to use their clinical data for research purposes.
(Jenkinson and Smith, 2001), and tissue segmentation (Zhang et al., 2001) yielding partial volume estimates. PCASL images were linearly registered to the brain-extracted T1-weighted images. Partial volume estimates were transformed to the ASL data space and used in a regression algorithm (Asllani et al., 2008) using a Gaussian kernel of 9.5 mm full width at half maximum, to create partial volume corrected (PVC) cortical CBF maps. Mean whole brain CBF was calculated using the segmented brain mask. Mean cortical CBF was calculated using the partial volume estimates as a weighting factor. CBF was defined in mL/100 g/min. 2.4. Normalized brain volumes NBV (mL) was estimated with the SIENAX software tool (Smith, 2002), part of FSL, using optimized brain extraction tool options as described previously (Popescu et al., 2012). To avoid lesionassociated segmentation biases, before segmentation lesions were filled with intensities of the normal appearing white matter using the automated lesion-filling technique LEAP (Chard et al., 2010). 2.5. White matter hyperintensities WMH were segmented using a locally developed k-Nearest Neighbor algorithm (Steenwijk et al., 2013) based on a previous work (Anbeek et al., 2008). In short, this algorithm uses fluidattenuated inversion-recovery and T1 tissue intensity, spatial information, and tissue priors to compare the brain voxels of a newly presented data set to a collection of manually labeled examples in a feature space. Based on the most similar examples, the probability of a voxel being a lesion is computed and thresholded to obtain a binary lesion segmentation. Importantly, the training set for automated lesion segmentation was generated on images acquired with the same scanner and pulse sequences as those in the present study. All segmentations were visually inspected. WMH volumes (in milliliter, mL) were normalized for head size by multiplying the volumes by a scaling factor, derived from the SIENAX estimation.
2.2. MRI protocol
2.6. Other MRI measures
MRI of the brain was acquired on a 3T whole body MR system (Signa, HDxt, General Electric Medical Systems, Milwaukee, WI, USA), using an 8-channel phased-array head coil. The MRI protocol included a sagittal 3D T1-weighted sequence (IR-FSPGR, repetition time [TR] ¼ 7.8 ms, echo time [TE] ¼ 3 ms, inversion time ¼ 450 ms, flip angle ¼ 12 , voxel sixe ¼ 1.0 0.9 0.9 mm), a sagittal 3D fluid-attenuated inversion-recovery (TR ¼ 8000 ms, TE ¼ 123.6 ms, inversion time ¼ 2350 ms, voxel size ¼ 1.0 1.0 1.0 mm) an axial 2D T2* gradientecho with an echo-planar read-out (EPI: TR ¼ 5300 ms, TE ¼ 25 ms, voxel size ¼ 1.0 0.5 0.5 mm), and an axial 2D proton density/T2weighted fast spin echo (PD-T2: TE ¼ 20/112 ms, TR ¼ 8680 ms, voxel size ¼ 1.0 0.5 0.5 mm). PCASL perfusion images (3D-FSE acquisition with background suppression, post-label delay ¼ 2.0 seconds, TR ¼ 4.8 seconds, TE ¼ 9 ms, spiral readout ¼ 8 arms 512 samples; voxel size ¼ 1.0 1.7 1.7 mm) were calculated using a single compartment model (Buxton et al., 1998) after the subtraction of labeled images from control images. Binnewijzend et al. (2013) provides a more detailed description of the ASL sequence.
Left and right hippocampal volumes (mL) were quantified using FSL FIRST (FMRIBs Integrated registration and segmentation tool) (Patenaude et al., 2011). All segmentations were visually inspected. Hippocampal volumes were normalized for head size by multiplying the volumes by the SIENAX derived scaling factor. For analytical purposes, left and right hippocampal volumes were summed. Cerebral microbleeds were visually assessed and defined as small round foci of hypointense signal, up to 10 mm in brain parenchyma on T2*-weighted images. Microbleed count was dichotomized as present or absent. Lacunes (of presumed vascular origin) were defined as deep lesions (3e15 mm), with CSF-like signal on all sequences; they were scored as present or absent.
2.3. PCASL cerebral blood flow measures After correcting T1-weighted and PCASL images for gradient nonlinearities in all the 3 directions, data-analyses were carried out using FSL (version 4.1.9; http://www.fmrib.ox.ac.uk/fsl). Preprocessing of T1-weighted images consisted of removal of nonbrain tissue (Smith, 2002), linear registration to standard space
2.7. Data analysis Statistical analyses were performed using SPSS (version 20; SPSS, Chicago, IL, USA). As WMH volumes were not normally distributed, we used log-transformed values. Differences in baseline characteristics between groups were investigated with Student t-test for continuous variables and c2 test for dichotomous variables. Differences in CBF between groups were analyzed using 1-way analysis of covariance, corrected for age and sex. Linear regression analysis was carried out to investigate the associations of NBV and WMH (independent) with CBF (dependent). All models were adjusted for age and sex. In model I, we investigated the univariate associations of NBV or WMH with CBF. In model II, NBV and WMH were
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
simultaneously entered. Model III consisted of model II, with additional adjustment for hippocampal volume, microbleed presence and lacune presence. Finally, we repeated the analyses with additional adjustment for hypertension, hypercholesteremia, diabetes mellitus, current smoking, and systolic and diastolic blood pressure. Linear regression analyses were stratified for diagnosis, to estimate the effects for controls and AD patients, separately. 3. Results Table 1 gives the patient characteristics by group, showing effective matching for age and sex. As expected, AD patients had a lower MMSE score than controls (p < 0.01). Vascular risk factors were comparable for AD patients and controls. AD patients had smaller NBVs, smaller hippocampal volumes, and larger WMH volumes compared with controls (all p < 0.01). There were no differences in the prevalence of microbleeds or lacunes. Whole brain CBF (mL/100gr/ min) was lower in AD patients compared with controls (27.3 5.8 vs. 31.5 5.3, p < 0.01). AD patients also had a lower PVC cortical CBF compared with controls (41.8 9.2 vs. 47.0 7.8, p < 0.01). Table 2 gives the associations of NBV and WMH with CBF by group. In AD patients, age and sex adjusted models (model I) showed that smaller NBV was associated with lower whole brain CBF (Stb: 0.28; p < 0.01) and PVC cortical CBF (Stb: 0.27; p < 0.01). In addition, in AD patients, larger WMH volume was associated with lower whole brain CBF (Stb: 0.21; p < 0.05) and PVC cortical CBF (Stb: 0.23; p < 0.05). These results remained essentially unchanged when NBV and WMH were simultaneously entered (model II) or when additional adjustment for MRI measures was performed (model III). Repeating the analyses with adjustment for vascular risk factors did not change these results (data not shown). Examples of whole brain CBF maps of 4 AD patients with different grades of atrophy and WMH are shown in Fig. 1. Table 1 Patient characteristics
Demographics Age, y Sex (% female) MMSE Vascular risk factors Hypertension, n (%) Hypercholesteremia, n (%) Diabetes mellitus, n (%) Smoking status Never, n (%) Former, n (%) Current, n (%) Systolic BP, mmHG Diastolic BP, mmHG MRI characteristics Normalized brain volume, mL Hippocampal volume (left and right), mL WMH volume, median (inter-quartile range), mL Microbleed presence, n (%) Lacune presence, n (%)
Controls (n ¼ 61)
AD patients (n ¼ 129)
64 5 26 (43) 28 2
66 7 69 (53) 21 5a
14 (23) 5 (8) 4 (7)
32 (25) 11 (8) 7 (5)
26 (44) 28 (48) 5 (8) 141 19 85 11
60 (48) 46 (37) 20 (16) 144 19 88 11
1424.3 81.3 9.9 1.1
1368.5 72.3a 8.7 1.2a
4.8 (3.7e7.5)
10.9 (6.7e19.8)a
13 (21) 3 (5)
40 (31) 7 (5)
Data are represented as mean SD, number of patients with variable present (%), or median (inter-quartile range). Group comparisons used Student t-test for continuous variables and c2-test for categorical variables. Availability for incomplete data in controls: MMSE: 59/61, BP: 59/61, smoking status: 59/61; and in AD patients: BP: 125/129, smoking status: 126/129, microbleeds: 127/129. Key: AD, Alzheimer’s disease; BP, blood pressure; MMSE, Mini mental state examination; NBV, normalized brain volume; SD, standard deviation; WMH, white matter hyperintensities. a p < 0.01.
2667
Table 2 Associations of normalized brain volumes and white matter hyperintensities with cerebral blood flow
Model I NBV WMH Model II NBV WMH Model III NBV WMH
Controls
AD patients
CBF (mL/100 g/min)
CBF (mL/100 g/min)
Uncorrected whole brain
PVC cortical
Uncorrected whole brain
PVC cortical
0.07 0.20
0.09 0.22
0.28a 0.21b
0.27a 0.23b
0.03 0.20
0.05 0.21
0.29a 0.22b
0.28a 0.24b
0.04 0.18
0.07 0.20
0.28a 0.25a
0.27a 0.27a
Standardized regression coefficients are displayed to allow for direct comparison of each variables’ contribution. Model I: NBV or WMH univariate; adjusted for age and sex. Model II: NBV and WMH simultaneously; adjusted for age and sex. Model III: additional adjustment for hippocampal volume, the presence of microbleeds and lacunes. Key: AD, Alzheimer’s disease; CBF, cerebral blood flow; NBV, normalized brain volume; PVC, partial volume corrected; WMH, white matter hyperintensities. a p < 0.01. b p < 0.05.
In controls, NBV was not associated with whole brain (Stb: 0.07; n.s.) or PVC cortical CBF (Stb: 0.09; n.s.) in age and sex adjusted models. These results remained essentially unchanged when NBV and WMH were simultaneously entered (model II), or after additional adjustment for MRI measures (model III) or vascular risk factors (data not shown). Similarly, we found no association between WMH volume and whole brain CBF (Stb: 0.20; n.s.) or PVC cortical CBF (Stb: 0.22; n.s.) in controls. These results did not change after additional adjustments in model II and model III. Repeating the analyses with adjustment for vascular risk factors did also not alter these results (data not shown). Fig. 2 shows the association of NBV with whole brain CBF and the association of WMH with whole brain CBF by group. 4. Discussion In the present paper we combined quantitative measurement of NBV and WMH volume with quantification of CBF as measured using PCASL. In AD patients, smaller NBVs and larger WMH volumes were both, independently, associated with a lower CBF. This indicates that CBF as measured using PCASL may be a final common pathway that reflects total disease burden in patients with AD. To our knowledge, we are the first to investigate associations of NBV and WMH with CBF as measured using PCASL in a well characterized set of AD patients and controls. In the present study, our control group consisted of subjects who presented with subjective complaints at our memory clinic. This may be considered a limitation, as it has previously been shown that the presence of subjective memory complaints may predict incident AD (Geerlings et al., 1999). However, we only included subjects with normal CSF Aß42, thereby limiting the change to have included subjects with preclinical AD (Sperling et al., 2011). Another limitation is that we could not check the reliability of our CBF measurement in white matter. The reliability of measuring white matter CBF is still being debated and the method that we used did not allow the performance of statistics to check measurement reliability (van Gelderen et al., 2008; van Osch et al., 2009). We were therefore not able to investigate white matter CBF. Moreover, a longitudinal design could have given more insight into the still largely unknown order in which neurodegeneration, SVD, and decreased CBF occur.
2668
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
Fig. 1. Examples of FLAIR scans and uncorrected whole brain cerebral blood flow maps of 4 patients with Alzheimer’s disease with different degrees of atrophy and white matter hyperintensities. Abbreviations: CBF, cerebral blood flow; FLAIR, fluid-attenuated inversion-recovery; MMSE, Mini mental state examination; WMH, white matter hyperintensities.
In AD patients, we found that smaller NBVs and larger WMH volumes were both associated with lower CBF. We did not only find an association for NBV with whole brain CBF, but also with PVC cortical CBF, in which errors that have been induced by atrophy have been accounted for (Asllani et al., 2008). To our knowledge, relatively little research has been performed on the determinants of decreased CBF in AD patients. In a previous perfusion weighted imaging report, regional brain volume changes and regional CBF decreases appeared to be dissociated in early AD, suggesting different underlying pathogenetic mechanisms (Luckhaus et al.,
2010). Findings regarding an association between WMH volume and decreased CBF in AD are not straightforward. Previous ASL (Zhang et al., 2012) and positron emission tomography (Schuff et al., 2009) studies did not find an association between WMH and CBF in AD, whereas a single photon emission computed tomography study did find a decreased regional CBF in AD patients with WMH (Kimura et al., 2012). The use of various methods limits comparison with previous reports, but our results suggest that NBV and WMH are both, independently, associated with PCASL-measured whole brain and cortical CBF in AD patients.
Fig. 2. (A) Association of normalized brain volume (NBV) with whole brain cerebral blood flow (CBF) for controls (dotted line) and patients with Alzheimer’s disease (AD) (black line). (B) Association of log-transformed normalized white matter hyperintensity (log WMH) volume with whole brain cerebral blood flow (CBF) for controls (dotted line) and patients with AD (black line).
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
The independency of the associations that we found in AD patients indicates that in the presence of severe neurodegeneration, CBF is even lower when additional SVD is present. As we previously showed that CBF was associated with cognition in patients with AD (Binnewijzend et al., 2013), our findings have clinical relevance and may have several implications. In the first place, this study again underlines the importance of the prevention and treatment of modifiable risk factors for vascular disease in AD patients. In addition, efforts to improve CBF, for instance by means of exercise (Barnes et al., 2013), may have beneficial effects. Most importantly, however, PCASL may provide a new measure for total disease burden in AD. Accumulating evidence suggests that cerebrovascular pathology interacts with AD pathology, not only affecting the risk of AD (van der Flier et al., 2004), but also its course and cognitive symptoms (Brickman et al., 2008). The exact mechanisms are, however, still not well understood. CBF measured with PCASL may be a final common pathway that reflects the cumulative burden of neurodegeneration and SVD in patients with AD. Contrary to our findings in AD patients, we found no associations between NBV or WMH with CBF in controls. To our knowledge, no previous literature exists on the association of brain volume with CBF in healthy elderly individuals. In patients that suffer from vascular pathology, however, total brain volume or atrophy was not found to be associated with whole brain CBF (Appelman et al., 2008; van Es et al., 2010). Previous reports on the association between WMH and CBF in healthy elderly individuals are not straightforward. Vernooij et al. (2008) did report an association between lower perfusion and larger WMH volume in the general elderly population using phase-contrast MRI. Using ASL, however, Schuff et al. (2009) did not find an association within their (small) group of controls. Our results suggest that in healthy elderly individuals, variability in PCASL-measured CBF reflects normal variation that is not determined by brain volume or the burden of SVD. Overall our results indicate that independent processes contribute to a decreased CBF. We conclude that CBF as measured using PCASL may provide a bridge between neurodegeneration and SVD and offers opportunities for future research regarding both pathologic processes in AD. Disclosure statement Prof. Dr Philip Scheltens serves/has served on the advisory boards of Genentech, Novartis, Roche, Danone, Nutricia, Lilly, and Lundbeck. He has been a speaker at symposia organized by Lundbeck, Merz, Danone, Novartis, Roche, GE, and Genentech. For all his activities he receives no personal compensation. Prof. Dr Frederik Barkhof serves/has served on the advisory boards of Bayer-Schering Pharma, Sanofi-Aventis, Biogen Idec, UCB, MerckSerono, Novartis, and Roche. He received funding from the Dutch MS Society and has been a speaker at symposia organized by the Serono Symposia Foundation. All other authors report no conflicts of interest. Acknowledgements The authors thank Ajit Shankaranarayanan of GE Healthcare for providing the 3D pseudocontinuous arterial spin labeling sequence that was used to obtain data for this article. M.R. Benedictus is supported by Stichting Dioraphte. Research of the VUmc Alzheimer center is part of the neurodegeneration research program of the Neuroscience Campus Amsterdam. The VUmc Alzheimer center is supported by Alzheimer Nederland and Stichting VUmc fonds. The clinical database structure was developed with funding from Stichting Dioraphte (VSM 12 01 02 00).
2669
References Alexopoulos, P., Sorg, C., Forschler, A., Grimmer, T., Skokou, M., Wohlschlager, A., Perneczky, R., Zimmer, C., Kurz, A., Preibisch, C., 2012. Perfusion abnormalities in mild cognitive impairment and mild dementia in Alzheimer’s disease measured by pulsed arterial spin labeling MRI. Eur. Arch. Psychiatry Clin. Neurosci. 1, 69e77. Alsop, D.C., Casement, M., de, Bazelaire C., Fong, T., Press, D.Z., 2008. Hippocampal hyperperfusion in Alzheimer’s disease. Neuroimage 4, 1267e1274. Alsop, D.C., Dai, W., Grossman, M., Detre, J.A., 2010. Arterial spin labeling blood flow MRI: its role in the early characterization of Alzheimer’s disease. J. Alzheimers. Dis. 3, 871e880. Anbeek, P., Vincken, K.L., Viergever, M.A., 2008. Automated MS-lesion segmentation by k-nearest neighbor classification. MIDAS J., 1e8. Appelman, A.P., van der Graaf, Y., Vincken, K.L., Tiehuis, A.M., Witkamp, T.D., Mali, W.P., Geerlings, M.I., 2008. Total cerebral blood flow, white matter lesions and brain atrophy: the SMART-MR study. J. Cereb. Blood Flow Metab. 3, 633e639. Asllani, I., Borogovac, A., Brown, T.R., 2008. Regression algorithm correcting for partial volume effects in arterial spin labeling MRI. Magn. Reson. Med. 6, 1362e1371. Barnes, J.N., Taylor, J.L., Kluck, B.N., Johnson, C.P., Joyner, M.J., 2013. Cerebrovascular reactivity is associated with maximal aerobic capacity in healthy older adults. J. Appl. Physiol. (1985) 10, 1383e1387. Bastos-Leite, A.J., Kuijer, J.P., Rombouts, S.A., Sanz-Arigita, E., van Straaten, E.C., Gouw, A.A., van der Flier, W.M., Scheltens, P., Barkhof, F., 2008. Cerebral blood flow by using pulsed arterial spin-labeling in elderly subjects with white matter hyperintensities. AJNR Am. J. Neuroradiol. 7, 1296e1301. Binnewijzend, M.A., Kuijer, J.P., Benedictus, M.R., van der Flier, W.M., Wink, A.M., Wattjes, M.P., van Berckel, B.N., Scheltens, P., Barkhof, F., 2013. Cerebral blood flow measured with 3D pseudocontinuous arterial spin-labeling MR imaging in Alzheimer disease and mild cognitive impairment: a marker for disease severity. Radiology 1, 221e230. Brickman, A.M., Honig, L.S., Scarmeas, N., Tatarina, O., Sanders, L., Albert, M.S., Brandt, J., Blacker, D., Stern, Y., 2008. Measuring cerebral atrophy and white matter hyperintensity burden to predict the rate of cognitive decline in Alzheimer disease. Arch. Neurol. 9, 1202e1208. Buxton, R.B., Frank, L.R., Wong, E.C., Siewert, B., Warach, S., Edelman, R.R., 1998. A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn. Reson. Med. 3, 383e396. Chard, D.T., Jackson, J.S., Miller, D.H., Wheeler-Kingshott, C.A., 2010. Reducing the impact of white matter lesions on automated measures of brain gray and white matter volumes. J. Magn Reson. Imaging 1, 223e228. Dai, W., Garcia, D., de Bazelaire, C., Alsop, D.C., 2008. Continuous flow-driven inversion for arterial spin labeling using pulsed radio frequency and gradient fields. Magn. Reson. Med. 6, 1488e1497. Dai, W., Lopez, O.L., Carmichael, O.T., Becker, J.T., Kuller, L.H., Gach, H.M., 2009. Mild cognitive impairment and Alzheimer disease: patterns of altered cerebral blood flow at MR imaging. Radiology 3, 856e866. Geerlings, M.I., Jonker, C., Bouter, L.M., Ader, H.J., Schmand, B., 1999. Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. Am. J. Psychiatry 4, 531e537. Jack, C.R., Knopman, D.S., Jagust, W.J., Shaw, L.M., Aisen, P.S., Weiner, M.W., Petersen, R.C., Trojanowski, J.Q., 2010. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol. 1, 119e128. Jenkinson, M., Smith, S., 2001. A global optimisation method for robust affine registration of brain images. Med. Image Anal. 2, 143e156. Kimura, N., Nakama, H., Nakamura, K., Aso, Y., Kumamoto, T., 2012. Effect of white matter lesions on brain perfusion in Alzheimer’s disease. Dement. Geriatr. Cogn. Disord. 3-4, 256e261. Luckhaus, C., Cohnen, M., Fluss, M.O., Janner, M., Grass-Kapanke, B., Teipel, S.J., Grothe, M., Hampel, H., Peters, O., Kornhuber, J., Maier, W., Supprian, T., Gaebel, W., Modder, U., Wittsack, H.J., 2010. The relation of regional cerebral perfusion and atrophy in mild cognitive impairment (MCI) and early Alzheimer’s dementia. Psychiatry Res. 1, 44e51. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., Stadlan, E.M., 1984. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 7, 939e944. McKhann, G.M., Knopman, D.S., Chertkow, H., Hyman, B.T., Jack Jr., C.R., Kawas, C.H., Klunk, W.E., Koroshetz, W.J., Manly, J.J., Mayeux, R., Mohs, R.C., Morris, J.C., Rossor, M.N., Scheltens, P., Carrillo, M.C., Thies, B., Weintraub, S., Phelps, C.H., 2011. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers. Dement. 3, 263e269. Mulder, C., Verwey, N.A., van der Flier, W.M., Bouwman, F.H., Kok, A., van Elk, E.J., Scheltens, P., Blankenstein, M.A., 2010. Amyloid-beta(1-42), total tau, and phosphorylated tau as cerebrospinal fluid biomarkers for the diagnosis of Alzheimer disease. Clin. Chem. 2, 248e253. Petersen, E.T., Zimine, I., Ho, Y.C., Golay, X., 2006. Non-invasive measurement of perfusion: a critical review of arterial spin labelling techniques. Br. J. Radiol. 944, 688e701. Patenaude, B., Smith, S.M., Kennedy, D.N., Jenkinson, M., 2011. A Bayesian model of shape and appearance for subcortical brain segmentation. Neuroimage 3, 907e922.
2670
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
Petersen, R.C., Stevens, J.C., Ganguli, M., Tangalos, E.G., Cummings, J.L., DeKosky, S.T., 2001. Practice parameter: early detection of dementia: mild cognitive impairment (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 9, 1133e1142. Popescu, V., Battaglini, M., Hoogstrate, W.S., Verfaillie, S.C., Sluimer, I.C., van Schijndel, R.A., van Dijk, B.W., Cover, K.S., Knol, D.L., Jenkinson, M., Barkhof, F., de, Stefano N., Vrenken, H., 2012. Optimizing parameter choice for FSL-Brain Extraction Tool (BET) on 3D T1 images in multiple sclerosis. Neuroimage 4, 1484e1494. Scheltens, P., Barkhof, F., Leys, D., Wolters, E.C., Ravid, R., Kamphorst, W., 1995. Histopathologic correlates of white matter changes on MRI in Alzheimer’s disease and normal aging. Neurology 5, 883e888. Schuff, N., Matsumoto, S., Kmiecik, J., Studholme, C., Du, A., Ezekiel, F., Miller, B.L., Kramer, J.H., Jagust, W.J., Chui, H.C., Weiner, M.W., 2009. Cerebral blood flow in ischemic vascular dementia and Alzheimer’s disease, measured by arterial spin-labeling magnetic resonance imaging. Alzheimers. Dement. 6, 454e462. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 3, 143e155. Sperling, R.A., Aisen, P.S., Beckett, L.A., Bennett, D.A., Craft, S., Fagan, A.M., Iwatsubo, T., Jack Jr., C.R., Kaye, J., Montine, T.J., Park, D.C., Reiman, E.M., Rowe, C.C., Siemers, E., Stern, Y., Yaffe, K., Carrillo, M.C., Thies, B., MorrisonBogorad, M., Wagster, M.V., Phelps, C.H., 2011. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers. Dement. 3, 280e292. Steenwijk, M.D., Pouwels, P.J., Daams, M., van Dalen, J.W., Caan, M.W., Richard, E., Barkhof, F., Vrenken, H., 2013. Accurate white matter lesion segmentation by k nearest neighbor classification with tissue type priors (kNN-TTPs). Neuroimage Clin. 3, 462e469. van der Flier, W.M., Middelkoop, H.A., Weverling-Rijnsburger, A.W., AdmiraalBehloul, F., Spilt, A., Bollen, E.L., Westendorp, R.G., van Buchem, M.A., 2004.
Interaction of medial temporal lobe atrophy and white matter hyperintensities in AD. Neurology 10, 1862e1864. van Es, A.C., van der Grond, J., ten Dam, V., de Craen, A.J., Blauw, G.J., Westendorp, R.G., Admiraal-Behloul, F., van Buchem, M.A., 2010. Associations between total cerebral blood flow and age related changes of the brain. PLoS One 3, e9825. van Gelderen, P., de Zwart, J.A., Duyn, J.H., 2008. Pittfalls of MRI measurement of white matter perfusion based on arterial spin labeling. Magn. Reson. Med. 4, 788e795. van Osch, M.J., Teeuwisse, W.M., van Walderveen, M.A., Hendrikse, J., Kies, D.A., van Buchem, M.A., 2009. Can arterial spin labeling detect white matter perfusion signal? Magn. Reson. Med. 1, 165e173. Vernooij, M.W., van der, Lugt A., Ikram, M.A., Wielopolski, P.A., Vrooman, H.A., Hofman, A., Krestin, G.P., Breteler, M.M., 2008. Total cerebral blood flow and total brain perfusion in the general population: the Rotterdam Scan Study. J. Cereb. Blood Flow Metab. 2, 412e419. Wardlaw, J.M., Smith, E.E., Biessels, G.J., Cordonnier, C., Fazekas, F., Frayne, R., Lindley, R.I., O’Brien, J.T., Barkhof, F., Benavente, O.R., Black, S.E., Brayne, C., Breteler, M., Chabriat, H., Decarli, C., de Leeuw, F.E., Doubal, F., Duering, M., Fox, N.C., Greenberg, S., Hachinski, V., Kilimann, I., Mok, V., Oostenbrugge, Rv, Pantoni, L., Speck, O., Stephan, B.C., Teipel, S., Viswanathan, A., Werring, D., Chen, C., Smith, C., van Buchem, M., Norrving, B., Gorelick, P.B., Dichgans, M., 2013. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 8, 822e838. Wolk, D.A., Detre, J.A., 2012. Arterial spin labeling MRI: an emerging biomarker for Alzheimer’s disease and other neurodegenerative conditions. Curr. Opin. Neurol. 4, 421e428. Zhang, Y., Brady, M., Smith, S., 2001. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Trans. Med. Imaging 1, 45e57. Zhang, Q., Stafford, R.B., Wang, Z., Arnold, S.E., Wolk, D.A., Detre, J.A., 2012. Microvascular perfusion based on arterial spin labeled perfusion MRI as a measure of vascular risk in Alzheimer’s disease. J. Alzheimers. Dis. 3, 677e687.
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Brain volume and white matter hyperintensities as determinants of cerebral blood flow in Alzheimer’s disease Marije R. Benedictus a, *, Maja A.A. Binnewijzend b, Joost P.A. Kuijer c, Martijn D. Steenwijk b, Adriaan Versteeg b, Hugo Vrenken b, Philip Scheltens a, Frederik Barkhof b, Wiesje M. van der Flier a, d, Niels D. Prins a a
Alzheimer Center & Department of Neurology, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands Department of Radiology and Nuclear medicine, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands Department of Physics and Medical Technology, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands d Department of Epidemiology and Biostatistics, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, the Netherlands b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 September 2013 Received in revised form 13 January 2014 Accepted 7 June 2014 Available online 13 June 2014
To better understand whether decreased cerebral blood flow (CBF) in patients with Alzheimer’s disease (AD) reflects neurodegeneration or cerebral small vessel disease, we investigated the associations of normalized brain volume (NBV) and white matter hyperintensity (WMH) volume with CBF. We included 129 patients with AD (66 7 years, 53% female) and 61 age-matched controls (64 5 years, 43% female). CBF was measured with pseudocontinuous arterial spin labeling at 3T in the whole brain and in partial volume corrected cortical maps. When NBV and WMH were simultaneously entered in age and sex adjusted models, smaller NBV was associated with lower whole brain (Stb: 0.29; p < 0.01) and cortical CBF (Stb: 0.28; p < 0.01) in patients with AD. Larger WMH volume was also associated with lower whole brain (Stb: 0.22; p < 0.05) and cortical CBF (Stb: 0.24; p < 0.05) in AD. Additional adjustments did not change these results. In controls, neither NBV nor WMH was associated with CBF. Our results indicate that in AD, lower CBF as measured using pseudocontinuous arterial spin labeling, reflects the combined disease burden of both neurodegeneration and small vessel disease. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Arterial spin labeling Cerebral blood flow Alzheimer’s disease Neurodegeneration Cerebral small vessel disease
1. Introduction Alzheimer’s disease (AD) is essentially regarded as a neurodegenerative disease, characterized by the accumulation of amyloid plaques and neurofibrillary tangles that eventually leads to brain atrophy (Jack et al., 2010). Increasing evidence indicates that AD patient not only have brain volume loss, but also often have an altered cerebral blood flow (CBF). Although some studies report relative regional increases in (early) AD (Alsop et al., 2008; Dai et al., 2009), the most consistent finding is a decrease in absolute CBF in patients with AD (Alexopoulos et al., 2012; Alsop et al., 2010; Binnewijzend et al., 2013). This decreased CBF is in general assumed to be a reflection of the neurodegenerative process (Wolk and Detre, 2012). Lower CBF in AD patients may not only relate to neurodegeneration, but may also be associated with small vessel disease * Corresponding author at: Department of Neurology and Alzheimer Center, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, the Netherlands. Tel.: þ31 204440183; fax: þ31 204448529. E-mail address: [email protected] (M.R. Benedictus). 0197-4580/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.06.001
(SVD). White matter hyperintensities (WMH) of presumed vascular origin are a commonly used magnetic resonance imaging (MRI) marker to indicate the presence of SVD (Wardlaw et al., 2013). WMH are assumed to result from ischemia and they are more prevalent in AD patients compared with the general elderly population (Scheltens et al., 1995). Previous studies have shown that WMH are associated with lower CBF as well (Bastos-Leite et al., 2008; Schuff et al., 2009; Vernooij et al., 2008). CBF can be measured by arterial spin labeling (ASL); a functional MRI technique that uses magnetically labeled arterial blood water as an endogenous tracer (Petersen et al., 2006). The pseudocontinuous variant of ASL (PCASL) uses a multitude of millisecondlong pulses to achieve a high labeling efficiency and effective compensation of magnetization transfer effects (Dai et al., 2008). Neurodegeneration and SVD are common in patients with AD, but to our knowledge, no previous studies have investigated how both processes relate to the generally described decreased CBF. Our aim was to explore whether independent relationships exist between normalized brain volume (NBV) or WMH on the one hand and CBF on the other hand. We hypothesized that the lower CBF in AD is not only reflective of the neurodegenerative process, but that
2666
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
CBF may be even further decreased when SVD is present. The well-characterized Amsterdam Dementia Cohort with PCASL measurement allowed us to investigate the determinants of CBF in AD patients and controls. 2. Methods 2.1. Subjects Subjects for this study were drawn from the memory clinic based Amsterdam Dementia Cohort. We included 129 AD patients and 61 age-matched controls who visited our memory clinic between October 2010 and June 2012. All subjects underwent an extensive dementia screening, including medical history, neurologic, and physical examination, cognitive assessment, and brain MRI. The diagnosis “probable AD” was made according to the NINCDS-ADRDA criteria, by consensus of a multidisciplinary team and all patients fulfilled the core clinical criteria of the NIA-AA (McKhann et al., 1984, 2011). The control group consisted of agematched control subjects, who presented with cognitive complaints, but for whom clinical investigations were normal and criteria for mild cognitive impairment (Petersen et al., 2001), dementia, or any other neurologic or psychiatric disorder were not met. As subjective complaints may represent preclinical AD in a subgroup (Sperling et al., 2011), we only included subjects with normal cerebrospinal fluid (CSF) Ab42 levels (see Mulder et al. (2010) for a detailed description of CSF analyses). For all subjects, the presence of hypertension, hypercholesteremia, and diabetes mellitus were determined based on self-reported medical history and medication use. Smoking status was defined as never, former, or current. Blood pressures were measured manually using a sphygmomanometer. Exclusion criteria were the presence of structural brain lesions and failure of preprocessing of the MRI scans. The ethical review board of the VU University Medical Center approved the study. We obtained informed consent from all patients to use their clinical data for research purposes.
(Jenkinson and Smith, 2001), and tissue segmentation (Zhang et al., 2001) yielding partial volume estimates. PCASL images were linearly registered to the brain-extracted T1-weighted images. Partial volume estimates were transformed to the ASL data space and used in a regression algorithm (Asllani et al., 2008) using a Gaussian kernel of 9.5 mm full width at half maximum, to create partial volume corrected (PVC) cortical CBF maps. Mean whole brain CBF was calculated using the segmented brain mask. Mean cortical CBF was calculated using the partial volume estimates as a weighting factor. CBF was defined in mL/100 g/min. 2.4. Normalized brain volumes NBV (mL) was estimated with the SIENAX software tool (Smith, 2002), part of FSL, using optimized brain extraction tool options as described previously (Popescu et al., 2012). To avoid lesionassociated segmentation biases, before segmentation lesions were filled with intensities of the normal appearing white matter using the automated lesion-filling technique LEAP (Chard et al., 2010). 2.5. White matter hyperintensities WMH were segmented using a locally developed k-Nearest Neighbor algorithm (Steenwijk et al., 2013) based on a previous work (Anbeek et al., 2008). In short, this algorithm uses fluidattenuated inversion-recovery and T1 tissue intensity, spatial information, and tissue priors to compare the brain voxels of a newly presented data set to a collection of manually labeled examples in a feature space. Based on the most similar examples, the probability of a voxel being a lesion is computed and thresholded to obtain a binary lesion segmentation. Importantly, the training set for automated lesion segmentation was generated on images acquired with the same scanner and pulse sequences as those in the present study. All segmentations were visually inspected. WMH volumes (in milliliter, mL) were normalized for head size by multiplying the volumes by a scaling factor, derived from the SIENAX estimation.
2.2. MRI protocol
2.6. Other MRI measures
MRI of the brain was acquired on a 3T whole body MR system (Signa, HDxt, General Electric Medical Systems, Milwaukee, WI, USA), using an 8-channel phased-array head coil. The MRI protocol included a sagittal 3D T1-weighted sequence (IR-FSPGR, repetition time [TR] ¼ 7.8 ms, echo time [TE] ¼ 3 ms, inversion time ¼ 450 ms, flip angle ¼ 12 , voxel sixe ¼ 1.0 0.9 0.9 mm), a sagittal 3D fluid-attenuated inversion-recovery (TR ¼ 8000 ms, TE ¼ 123.6 ms, inversion time ¼ 2350 ms, voxel size ¼ 1.0 1.0 1.0 mm) an axial 2D T2* gradientecho with an echo-planar read-out (EPI: TR ¼ 5300 ms, TE ¼ 25 ms, voxel size ¼ 1.0 0.5 0.5 mm), and an axial 2D proton density/T2weighted fast spin echo (PD-T2: TE ¼ 20/112 ms, TR ¼ 8680 ms, voxel size ¼ 1.0 0.5 0.5 mm). PCASL perfusion images (3D-FSE acquisition with background suppression, post-label delay ¼ 2.0 seconds, TR ¼ 4.8 seconds, TE ¼ 9 ms, spiral readout ¼ 8 arms 512 samples; voxel size ¼ 1.0 1.7 1.7 mm) were calculated using a single compartment model (Buxton et al., 1998) after the subtraction of labeled images from control images. Binnewijzend et al. (2013) provides a more detailed description of the ASL sequence.
Left and right hippocampal volumes (mL) were quantified using FSL FIRST (FMRIBs Integrated registration and segmentation tool) (Patenaude et al., 2011). All segmentations were visually inspected. Hippocampal volumes were normalized for head size by multiplying the volumes by the SIENAX derived scaling factor. For analytical purposes, left and right hippocampal volumes were summed. Cerebral microbleeds were visually assessed and defined as small round foci of hypointense signal, up to 10 mm in brain parenchyma on T2*-weighted images. Microbleed count was dichotomized as present or absent. Lacunes (of presumed vascular origin) were defined as deep lesions (3e15 mm), with CSF-like signal on all sequences; they were scored as present or absent.
2.3. PCASL cerebral blood flow measures After correcting T1-weighted and PCASL images for gradient nonlinearities in all the 3 directions, data-analyses were carried out using FSL (version 4.1.9; http://www.fmrib.ox.ac.uk/fsl). Preprocessing of T1-weighted images consisted of removal of nonbrain tissue (Smith, 2002), linear registration to standard space
2.7. Data analysis Statistical analyses were performed using SPSS (version 20; SPSS, Chicago, IL, USA). As WMH volumes were not normally distributed, we used log-transformed values. Differences in baseline characteristics between groups were investigated with Student t-test for continuous variables and c2 test for dichotomous variables. Differences in CBF between groups were analyzed using 1-way analysis of covariance, corrected for age and sex. Linear regression analysis was carried out to investigate the associations of NBV and WMH (independent) with CBF (dependent). All models were adjusted for age and sex. In model I, we investigated the univariate associations of NBV or WMH with CBF. In model II, NBV and WMH were
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
simultaneously entered. Model III consisted of model II, with additional adjustment for hippocampal volume, microbleed presence and lacune presence. Finally, we repeated the analyses with additional adjustment for hypertension, hypercholesteremia, diabetes mellitus, current smoking, and systolic and diastolic blood pressure. Linear regression analyses were stratified for diagnosis, to estimate the effects for controls and AD patients, separately. 3. Results Table 1 gives the patient characteristics by group, showing effective matching for age and sex. As expected, AD patients had a lower MMSE score than controls (p < 0.01). Vascular risk factors were comparable for AD patients and controls. AD patients had smaller NBVs, smaller hippocampal volumes, and larger WMH volumes compared with controls (all p < 0.01). There were no differences in the prevalence of microbleeds or lacunes. Whole brain CBF (mL/100gr/ min) was lower in AD patients compared with controls (27.3 5.8 vs. 31.5 5.3, p < 0.01). AD patients also had a lower PVC cortical CBF compared with controls (41.8 9.2 vs. 47.0 7.8, p < 0.01). Table 2 gives the associations of NBV and WMH with CBF by group. In AD patients, age and sex adjusted models (model I) showed that smaller NBV was associated with lower whole brain CBF (Stb: 0.28; p < 0.01) and PVC cortical CBF (Stb: 0.27; p < 0.01). In addition, in AD patients, larger WMH volume was associated with lower whole brain CBF (Stb: 0.21; p < 0.05) and PVC cortical CBF (Stb: 0.23; p < 0.05). These results remained essentially unchanged when NBV and WMH were simultaneously entered (model II) or when additional adjustment for MRI measures was performed (model III). Repeating the analyses with adjustment for vascular risk factors did not change these results (data not shown). Examples of whole brain CBF maps of 4 AD patients with different grades of atrophy and WMH are shown in Fig. 1. Table 1 Patient characteristics
Demographics Age, y Sex (% female) MMSE Vascular risk factors Hypertension, n (%) Hypercholesteremia, n (%) Diabetes mellitus, n (%) Smoking status Never, n (%) Former, n (%) Current, n (%) Systolic BP, mmHG Diastolic BP, mmHG MRI characteristics Normalized brain volume, mL Hippocampal volume (left and right), mL WMH volume, median (inter-quartile range), mL Microbleed presence, n (%) Lacune presence, n (%)
Controls (n ¼ 61)
AD patients (n ¼ 129)
64 5 26 (43) 28 2
66 7 69 (53) 21 5a
14 (23) 5 (8) 4 (7)
32 (25) 11 (8) 7 (5)
26 (44) 28 (48) 5 (8) 141 19 85 11
60 (48) 46 (37) 20 (16) 144 19 88 11
1424.3 81.3 9.9 1.1
1368.5 72.3a 8.7 1.2a
4.8 (3.7e7.5)
10.9 (6.7e19.8)a
13 (21) 3 (5)
40 (31) 7 (5)
Data are represented as mean SD, number of patients with variable present (%), or median (inter-quartile range). Group comparisons used Student t-test for continuous variables and c2-test for categorical variables. Availability for incomplete data in controls: MMSE: 59/61, BP: 59/61, smoking status: 59/61; and in AD patients: BP: 125/129, smoking status: 126/129, microbleeds: 127/129. Key: AD, Alzheimer’s disease; BP, blood pressure; MMSE, Mini mental state examination; NBV, normalized brain volume; SD, standard deviation; WMH, white matter hyperintensities. a p < 0.01.
2667
Table 2 Associations of normalized brain volumes and white matter hyperintensities with cerebral blood flow
Model I NBV WMH Model II NBV WMH Model III NBV WMH
Controls
AD patients
CBF (mL/100 g/min)
CBF (mL/100 g/min)
Uncorrected whole brain
PVC cortical
Uncorrected whole brain
PVC cortical
0.07 0.20
0.09 0.22
0.28a 0.21b
0.27a 0.23b
0.03 0.20
0.05 0.21
0.29a 0.22b
0.28a 0.24b
0.04 0.18
0.07 0.20
0.28a 0.25a
0.27a 0.27a
Standardized regression coefficients are displayed to allow for direct comparison of each variables’ contribution. Model I: NBV or WMH univariate; adjusted for age and sex. Model II: NBV and WMH simultaneously; adjusted for age and sex. Model III: additional adjustment for hippocampal volume, the presence of microbleeds and lacunes. Key: AD, Alzheimer’s disease; CBF, cerebral blood flow; NBV, normalized brain volume; PVC, partial volume corrected; WMH, white matter hyperintensities. a p < 0.01. b p < 0.05.
In controls, NBV was not associated with whole brain (Stb: 0.07; n.s.) or PVC cortical CBF (Stb: 0.09; n.s.) in age and sex adjusted models. These results remained essentially unchanged when NBV and WMH were simultaneously entered (model II), or after additional adjustment for MRI measures (model III) or vascular risk factors (data not shown). Similarly, we found no association between WMH volume and whole brain CBF (Stb: 0.20; n.s.) or PVC cortical CBF (Stb: 0.22; n.s.) in controls. These results did not change after additional adjustments in model II and model III. Repeating the analyses with adjustment for vascular risk factors did also not alter these results (data not shown). Fig. 2 shows the association of NBV with whole brain CBF and the association of WMH with whole brain CBF by group. 4. Discussion In the present paper we combined quantitative measurement of NBV and WMH volume with quantification of CBF as measured using PCASL. In AD patients, smaller NBVs and larger WMH volumes were both, independently, associated with a lower CBF. This indicates that CBF as measured using PCASL may be a final common pathway that reflects total disease burden in patients with AD. To our knowledge, we are the first to investigate associations of NBV and WMH with CBF as measured using PCASL in a well characterized set of AD patients and controls. In the present study, our control group consisted of subjects who presented with subjective complaints at our memory clinic. This may be considered a limitation, as it has previously been shown that the presence of subjective memory complaints may predict incident AD (Geerlings et al., 1999). However, we only included subjects with normal CSF Aß42, thereby limiting the change to have included subjects with preclinical AD (Sperling et al., 2011). Another limitation is that we could not check the reliability of our CBF measurement in white matter. The reliability of measuring white matter CBF is still being debated and the method that we used did not allow the performance of statistics to check measurement reliability (van Gelderen et al., 2008; van Osch et al., 2009). We were therefore not able to investigate white matter CBF. Moreover, a longitudinal design could have given more insight into the still largely unknown order in which neurodegeneration, SVD, and decreased CBF occur.
2668
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
Fig. 1. Examples of FLAIR scans and uncorrected whole brain cerebral blood flow maps of 4 patients with Alzheimer’s disease with different degrees of atrophy and white matter hyperintensities. Abbreviations: CBF, cerebral blood flow; FLAIR, fluid-attenuated inversion-recovery; MMSE, Mini mental state examination; WMH, white matter hyperintensities.
In AD patients, we found that smaller NBVs and larger WMH volumes were both associated with lower CBF. We did not only find an association for NBV with whole brain CBF, but also with PVC cortical CBF, in which errors that have been induced by atrophy have been accounted for (Asllani et al., 2008). To our knowledge, relatively little research has been performed on the determinants of decreased CBF in AD patients. In a previous perfusion weighted imaging report, regional brain volume changes and regional CBF decreases appeared to be dissociated in early AD, suggesting different underlying pathogenetic mechanisms (Luckhaus et al.,
2010). Findings regarding an association between WMH volume and decreased CBF in AD are not straightforward. Previous ASL (Zhang et al., 2012) and positron emission tomography (Schuff et al., 2009) studies did not find an association between WMH and CBF in AD, whereas a single photon emission computed tomography study did find a decreased regional CBF in AD patients with WMH (Kimura et al., 2012). The use of various methods limits comparison with previous reports, but our results suggest that NBV and WMH are both, independently, associated with PCASL-measured whole brain and cortical CBF in AD patients.
Fig. 2. (A) Association of normalized brain volume (NBV) with whole brain cerebral blood flow (CBF) for controls (dotted line) and patients with Alzheimer’s disease (AD) (black line). (B) Association of log-transformed normalized white matter hyperintensity (log WMH) volume with whole brain cerebral blood flow (CBF) for controls (dotted line) and patients with AD (black line).
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
The independency of the associations that we found in AD patients indicates that in the presence of severe neurodegeneration, CBF is even lower when additional SVD is present. As we previously showed that CBF was associated with cognition in patients with AD (Binnewijzend et al., 2013), our findings have clinical relevance and may have several implications. In the first place, this study again underlines the importance of the prevention and treatment of modifiable risk factors for vascular disease in AD patients. In addition, efforts to improve CBF, for instance by means of exercise (Barnes et al., 2013), may have beneficial effects. Most importantly, however, PCASL may provide a new measure for total disease burden in AD. Accumulating evidence suggests that cerebrovascular pathology interacts with AD pathology, not only affecting the risk of AD (van der Flier et al., 2004), but also its course and cognitive symptoms (Brickman et al., 2008). The exact mechanisms are, however, still not well understood. CBF measured with PCASL may be a final common pathway that reflects the cumulative burden of neurodegeneration and SVD in patients with AD. Contrary to our findings in AD patients, we found no associations between NBV or WMH with CBF in controls. To our knowledge, no previous literature exists on the association of brain volume with CBF in healthy elderly individuals. In patients that suffer from vascular pathology, however, total brain volume or atrophy was not found to be associated with whole brain CBF (Appelman et al., 2008; van Es et al., 2010). Previous reports on the association between WMH and CBF in healthy elderly individuals are not straightforward. Vernooij et al. (2008) did report an association between lower perfusion and larger WMH volume in the general elderly population using phase-contrast MRI. Using ASL, however, Schuff et al. (2009) did not find an association within their (small) group of controls. Our results suggest that in healthy elderly individuals, variability in PCASL-measured CBF reflects normal variation that is not determined by brain volume or the burden of SVD. Overall our results indicate that independent processes contribute to a decreased CBF. We conclude that CBF as measured using PCASL may provide a bridge between neurodegeneration and SVD and offers opportunities for future research regarding both pathologic processes in AD. Disclosure statement Prof. Dr Philip Scheltens serves/has served on the advisory boards of Genentech, Novartis, Roche, Danone, Nutricia, Lilly, and Lundbeck. He has been a speaker at symposia organized by Lundbeck, Merz, Danone, Novartis, Roche, GE, and Genentech. For all his activities he receives no personal compensation. Prof. Dr Frederik Barkhof serves/has served on the advisory boards of Bayer-Schering Pharma, Sanofi-Aventis, Biogen Idec, UCB, MerckSerono, Novartis, and Roche. He received funding from the Dutch MS Society and has been a speaker at symposia organized by the Serono Symposia Foundation. All other authors report no conflicts of interest. Acknowledgements The authors thank Ajit Shankaranarayanan of GE Healthcare for providing the 3D pseudocontinuous arterial spin labeling sequence that was used to obtain data for this article. M.R. Benedictus is supported by Stichting Dioraphte. Research of the VUmc Alzheimer center is part of the neurodegeneration research program of the Neuroscience Campus Amsterdam. The VUmc Alzheimer center is supported by Alzheimer Nederland and Stichting VUmc fonds. The clinical database structure was developed with funding from Stichting Dioraphte (VSM 12 01 02 00).
2669
References Alexopoulos, P., Sorg, C., Forschler, A., Grimmer, T., Skokou, M., Wohlschlager, A., Perneczky, R., Zimmer, C., Kurz, A., Preibisch, C., 2012. Perfusion abnormalities in mild cognitive impairment and mild dementia in Alzheimer’s disease measured by pulsed arterial spin labeling MRI. Eur. Arch. Psychiatry Clin. Neurosci. 1, 69e77. Alsop, D.C., Casement, M., de, Bazelaire C., Fong, T., Press, D.Z., 2008. Hippocampal hyperperfusion in Alzheimer’s disease. Neuroimage 4, 1267e1274. Alsop, D.C., Dai, W., Grossman, M., Detre, J.A., 2010. Arterial spin labeling blood flow MRI: its role in the early characterization of Alzheimer’s disease. J. Alzheimers. Dis. 3, 871e880. Anbeek, P., Vincken, K.L., Viergever, M.A., 2008. Automated MS-lesion segmentation by k-nearest neighbor classification. MIDAS J., 1e8. Appelman, A.P., van der Graaf, Y., Vincken, K.L., Tiehuis, A.M., Witkamp, T.D., Mali, W.P., Geerlings, M.I., 2008. Total cerebral blood flow, white matter lesions and brain atrophy: the SMART-MR study. J. Cereb. Blood Flow Metab. 3, 633e639. Asllani, I., Borogovac, A., Brown, T.R., 2008. Regression algorithm correcting for partial volume effects in arterial spin labeling MRI. Magn. Reson. Med. 6, 1362e1371. Barnes, J.N., Taylor, J.L., Kluck, B.N., Johnson, C.P., Joyner, M.J., 2013. Cerebrovascular reactivity is associated with maximal aerobic capacity in healthy older adults. J. Appl. Physiol. (1985) 10, 1383e1387. Bastos-Leite, A.J., Kuijer, J.P., Rombouts, S.A., Sanz-Arigita, E., van Straaten, E.C., Gouw, A.A., van der Flier, W.M., Scheltens, P., Barkhof, F., 2008. Cerebral blood flow by using pulsed arterial spin-labeling in elderly subjects with white matter hyperintensities. AJNR Am. J. Neuroradiol. 7, 1296e1301. Binnewijzend, M.A., Kuijer, J.P., Benedictus, M.R., van der Flier, W.M., Wink, A.M., Wattjes, M.P., van Berckel, B.N., Scheltens, P., Barkhof, F., 2013. Cerebral blood flow measured with 3D pseudocontinuous arterial spin-labeling MR imaging in Alzheimer disease and mild cognitive impairment: a marker for disease severity. Radiology 1, 221e230. Brickman, A.M., Honig, L.S., Scarmeas, N., Tatarina, O., Sanders, L., Albert, M.S., Brandt, J., Blacker, D., Stern, Y., 2008. Measuring cerebral atrophy and white matter hyperintensity burden to predict the rate of cognitive decline in Alzheimer disease. Arch. Neurol. 9, 1202e1208. Buxton, R.B., Frank, L.R., Wong, E.C., Siewert, B., Warach, S., Edelman, R.R., 1998. A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn. Reson. Med. 3, 383e396. Chard, D.T., Jackson, J.S., Miller, D.H., Wheeler-Kingshott, C.A., 2010. Reducing the impact of white matter lesions on automated measures of brain gray and white matter volumes. J. Magn Reson. Imaging 1, 223e228. Dai, W., Garcia, D., de Bazelaire, C., Alsop, D.C., 2008. Continuous flow-driven inversion for arterial spin labeling using pulsed radio frequency and gradient fields. Magn. Reson. Med. 6, 1488e1497. Dai, W., Lopez, O.L., Carmichael, O.T., Becker, J.T., Kuller, L.H., Gach, H.M., 2009. Mild cognitive impairment and Alzheimer disease: patterns of altered cerebral blood flow at MR imaging. Radiology 3, 856e866. Geerlings, M.I., Jonker, C., Bouter, L.M., Ader, H.J., Schmand, B., 1999. Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. Am. J. Psychiatry 4, 531e537. Jack, C.R., Knopman, D.S., Jagust, W.J., Shaw, L.M., Aisen, P.S., Weiner, M.W., Petersen, R.C., Trojanowski, J.Q., 2010. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol. 1, 119e128. Jenkinson, M., Smith, S., 2001. A global optimisation method for robust affine registration of brain images. Med. Image Anal. 2, 143e156. Kimura, N., Nakama, H., Nakamura, K., Aso, Y., Kumamoto, T., 2012. Effect of white matter lesions on brain perfusion in Alzheimer’s disease. Dement. Geriatr. Cogn. Disord. 3-4, 256e261. Luckhaus, C., Cohnen, M., Fluss, M.O., Janner, M., Grass-Kapanke, B., Teipel, S.J., Grothe, M., Hampel, H., Peters, O., Kornhuber, J., Maier, W., Supprian, T., Gaebel, W., Modder, U., Wittsack, H.J., 2010. The relation of regional cerebral perfusion and atrophy in mild cognitive impairment (MCI) and early Alzheimer’s dementia. Psychiatry Res. 1, 44e51. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., Stadlan, E.M., 1984. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 7, 939e944. McKhann, G.M., Knopman, D.S., Chertkow, H., Hyman, B.T., Jack Jr., C.R., Kawas, C.H., Klunk, W.E., Koroshetz, W.J., Manly, J.J., Mayeux, R., Mohs, R.C., Morris, J.C., Rossor, M.N., Scheltens, P., Carrillo, M.C., Thies, B., Weintraub, S., Phelps, C.H., 2011. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers. Dement. 3, 263e269. Mulder, C., Verwey, N.A., van der Flier, W.M., Bouwman, F.H., Kok, A., van Elk, E.J., Scheltens, P., Blankenstein, M.A., 2010. Amyloid-beta(1-42), total tau, and phosphorylated tau as cerebrospinal fluid biomarkers for the diagnosis of Alzheimer disease. Clin. Chem. 2, 248e253. Petersen, E.T., Zimine, I., Ho, Y.C., Golay, X., 2006. Non-invasive measurement of perfusion: a critical review of arterial spin labelling techniques. Br. J. Radiol. 944, 688e701. Patenaude, B., Smith, S.M., Kennedy, D.N., Jenkinson, M., 2011. A Bayesian model of shape and appearance for subcortical brain segmentation. Neuroimage 3, 907e922.
2670
M.R. Benedictus et al. / Neurobiology of Aging 35 (2014) 2665e2670
Petersen, R.C., Stevens, J.C., Ganguli, M., Tangalos, E.G., Cummings, J.L., DeKosky, S.T., 2001. Practice parameter: early detection of dementia: mild cognitive impairment (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 9, 1133e1142. Popescu, V., Battaglini, M., Hoogstrate, W.S., Verfaillie, S.C., Sluimer, I.C., van Schijndel, R.A., van Dijk, B.W., Cover, K.S., Knol, D.L., Jenkinson, M., Barkhof, F., de, Stefano N., Vrenken, H., 2012. Optimizing parameter choice for FSL-Brain Extraction Tool (BET) on 3D T1 images in multiple sclerosis. Neuroimage 4, 1484e1494. Scheltens, P., Barkhof, F., Leys, D., Wolters, E.C., Ravid, R., Kamphorst, W., 1995. Histopathologic correlates of white matter changes on MRI in Alzheimer’s disease and normal aging. Neurology 5, 883e888. Schuff, N., Matsumoto, S., Kmiecik, J., Studholme, C., Du, A., Ezekiel, F., Miller, B.L., Kramer, J.H., Jagust, W.J., Chui, H.C., Weiner, M.W., 2009. Cerebral blood flow in ischemic vascular dementia and Alzheimer’s disease, measured by arterial spin-labeling magnetic resonance imaging. Alzheimers. Dement. 6, 454e462. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 3, 143e155. Sperling, R.A., Aisen, P.S., Beckett, L.A., Bennett, D.A., Craft, S., Fagan, A.M., Iwatsubo, T., Jack Jr., C.R., Kaye, J., Montine, T.J., Park, D.C., Reiman, E.M., Rowe, C.C., Siemers, E., Stern, Y., Yaffe, K., Carrillo, M.C., Thies, B., MorrisonBogorad, M., Wagster, M.V., Phelps, C.H., 2011. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers. Dement. 3, 280e292. Steenwijk, M.D., Pouwels, P.J., Daams, M., van Dalen, J.W., Caan, M.W., Richard, E., Barkhof, F., Vrenken, H., 2013. Accurate white matter lesion segmentation by k nearest neighbor classification with tissue type priors (kNN-TTPs). Neuroimage Clin. 3, 462e469. van der Flier, W.M., Middelkoop, H.A., Weverling-Rijnsburger, A.W., AdmiraalBehloul, F., Spilt, A., Bollen, E.L., Westendorp, R.G., van Buchem, M.A., 2004.
Interaction of medial temporal lobe atrophy and white matter hyperintensities in AD. Neurology 10, 1862e1864. van Es, A.C., van der Grond, J., ten Dam, V., de Craen, A.J., Blauw, G.J., Westendorp, R.G., Admiraal-Behloul, F., van Buchem, M.A., 2010. Associations between total cerebral blood flow and age related changes of the brain. PLoS One 3, e9825. van Gelderen, P., de Zwart, J.A., Duyn, J.H., 2008. Pittfalls of MRI measurement of white matter perfusion based on arterial spin labeling. Magn. Reson. Med. 4, 788e795. van Osch, M.J., Teeuwisse, W.M., van Walderveen, M.A., Hendrikse, J., Kies, D.A., van Buchem, M.A., 2009. Can arterial spin labeling detect white matter perfusion signal? Magn. Reson. Med. 1, 165e173. Vernooij, M.W., van der, Lugt A., Ikram, M.A., Wielopolski, P.A., Vrooman, H.A., Hofman, A., Krestin, G.P., Breteler, M.M., 2008. Total cerebral blood flow and total brain perfusion in the general population: the Rotterdam Scan Study. J. Cereb. Blood Flow Metab. 2, 412e419. Wardlaw, J.M., Smith, E.E., Biessels, G.J., Cordonnier, C., Fazekas, F., Frayne, R., Lindley, R.I., O’Brien, J.T., Barkhof, F., Benavente, O.R., Black, S.E., Brayne, C., Breteler, M., Chabriat, H., Decarli, C., de Leeuw, F.E., Doubal, F., Duering, M., Fox, N.C., Greenberg, S., Hachinski, V., Kilimann, I., Mok, V., Oostenbrugge, Rv, Pantoni, L., Speck, O., Stephan, B.C., Teipel, S., Viswanathan, A., Werring, D., Chen, C., Smith, C., van Buchem, M., Norrving, B., Gorelick, P.B., Dichgans, M., 2013. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 8, 822e838. Wolk, D.A., Detre, J.A., 2012. Arterial spin labeling MRI: an emerging biomarker for Alzheimer’s disease and other neurodegenerative conditions. Curr. Opin. Neurol. 4, 421e428. Zhang, Y., Brady, M., Smith, S., 2001. Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Trans. Med. Imaging 1, 45e57. Zhang, Q., Stafford, R.B., Wang, Z., Arnold, S.E., Wolk, D.A., Detre, J.A., 2012. Microvascular perfusion based on arterial spin labeled perfusion MRI as a measure of vascular risk in Alzheimer’s disease. J. Alzheimers. Dis. 3, 677e687.
