Reduced Cerebrovascular Reserve Capacity as a Biomarker of Microangiopathy in Alzheimer’s Disease and Mild Cognitive Impairment

Abstract

Background:

Cerebral microangiopathy in Alzheimer’s disease (AD) causes chronic hypoperfusion and probably accelerates neurodegenerative changes.

Objective:

We hypothesize microvascular impairment could be present already in mild cognitive impairment (MCI) and can be revealed using transcranial color-coded sonography (TCCS) and the breath-holding maneuver.

Methods:

Three groups of subjects (AD in the stage of dementia, MCI, and cognitively normal controls) with detailed neuropsychological testing and low cerebrovascular burden (no history of stroke, no intra- or extracranial artery stenoses, and no severe vascular lesions on brain MRI), underwent a TCCS assessment of peak systolic (PSV), mean flow (MFV), and end diastolic velocities (EDV) and resistance and pulsatility indices (RI, PI) in large intracranial vessels bilaterally. Cerebrovascular reserve capacity was assessed using the breath-holding index (BHI) in middle cerebral artery (MCA) bilaterally. The ultrasound parameters were compared between the groups, correlated with neuropsychological tests, and compared between amnestic and non-amnestic MCI subtypes.

Results:

Fourteen AD (3 males, 67.9±11.1 years, MMSE 18.0±4.6), 24 MCI (13 males, 71.9±7.3 years, MMSE 28.0±1.6), and 24 risk factor-matched controls (14 males, 67.8±6.4 years, MMSE 29.1±1.2) were enrolled. Significant differences were found between AD and controls in MFV, EDV, RI, PI in right MCA after breath holding, in PSV, MFV, EDV in left MCA after breath holding, and in BHI on the left side. The left BHI correlated positively with verbal memory test.

Conclusion:

Results show decreased cerebrovascular reserve capacity in AD as a sign of impaired cerebral hemodynamic status without severe underlying atherosclerosis. This can be identified using TCCS and BHI.

Keywords

Alzheimer’s disease breath-holding index cerebrovascular reserve capacity microangiopathy mild cognitive impairment transcranial color-coded sonography transcranial Doppler

INTRODUCTION

Cerebral atherosclerosis and cerebral microangiopathy have traditionally been associated with vascular dementia. However, there is epidemiological and clinico-pathological evidence that cerebrovascular impairment may play a role in the pathophysiology of cognitive decline in Alzheimer’s disease (AD) as well, alongside neurodegeneration [1].

There is a range of evidence for vascular impairment in AD. The classical vascular risk factors leading to atherosclerosis are more frequent in patients with AD than in the general population [2 –4], and patients with more prominent carotid atherosclerosis or greater carotid intima-media thicknesses are more likely to develop dementia of any etiology [5, 6]. Also, the manifestation of microangiopathy on imaging methods, such as white matter lesions or cerebral microbleeds, are more common in AD patients than those with healthy brains [7 –10]. In addition, cognitive decline progresses faster among those with impaired microvasculature [11].

Despite evidence of microangiopathy in AD, its cause and contribution to disease development is not fully elucidated. Very probably the vascular and neurodegenerative pathological processes are closely interconnected [12]. Both the structure and function of the whole neurovascular unit are impaired, and thus influencing its blood flow regulation function, blood brain barrier exchange, its immune and trophic function [13]. While atherosclerosis likely plays a great role in this process, there are several theories of the possible role of amyloid-β in microangiopathy and decreased brain perfusion in AD [14, 15]. Chronic tissue hypoperfusion leads to overproduction and decreased tissue clearance of amyloid-β and its accumulation in the wall of arterioles. The brain tissue is more prone to ischemic damage, which is then manifested as an increased incidence of white matter lesions and cerebral microbleeds [16].

To assess microangiopathy, the non-invasive examination with transcranial ultrasound can be used, and has been already employed in studies with AD patients in the stage of dementia. There are two modalities of transcranial ultrasound: transcranial Doppler (TCD), and transcranial color-coded sonography (TCCS). Both TCD and TCCS provide measurement of flow velocities, whereas TCCS integrates the Doppler flow signal with a duplex sonography offering the possibility of visualizing the insonated vessels. While flow velocities in the large intracranial arteries give information about relative brain perfusion, cerebrovascular reserve capacity is a parameter that describes the quality of cerebral perfusion and the capability of brain arterioles to regulate cerebral perfusion as a reaction to various oxygen-level changing stimuli, thanks to constriction or dilatation. Cerebrovascular reserve capacity measurements are used most importantly in chronic brain ischemia in chronic carotid occlusive disease before revascularization procedures. The most often used stimulus is a change of the arterial CO₂ level that can be induced using breath holding, CO₂ inhalation, or intravenous acetazolamide injection. The cerebrovascular reserve capacity is expressed as the ratio of mean flow velocity in basal conditions and mean flow velocity in the conditions of a higher CO₂ level. In a normal brain, there is an increase in flow velocities. When breath holding is the stimulus, the ratio can be multiplied by the duration of breath holding and expressed as breath-holding index (BHI). Cerebrovascular reserve capacity decreases with age [17]. All three transcranial ultrasound methods for assessment of cerebrovascular reserve capacity (acetazolamide injection, CO₂ inhalation, or breath holding) correlate very well to 133Xe SPECT [18 –20] with the breath-holding method being the less accurate but sufficient for first screening examination [20, 21]. Compared to scintigraphic techniques, the ultrasound examination is non-invasive and inexpensive. Decreased flow velocities and decreased cerebrovascular reserve capacity in AD has been suggested in some previous studies using TCD or TCCS [22 –27].

Because the possible effects of therapeutic or preventive measures in patients with dementia (and thus irreversible pathological changes) are low, the target population in AD research has generally moved from dementia patients to those with earlier clinical forms of the disease (i.e., mild cognitive impairment [MCI]), extending to preclinical forms as well. In MCI patients, there are efforts to identify those with greater risk of progression into dementia. It is known, that the amnestic subtype of MCI, which can be diagnosed clinically thanks to a prominent memory impairment, has a higher probability of conversion to AD than the non-amnestic subtype [28]. Similarly, using SPECT imaging, possible “convertors” can be differentiated from “non-convertors” based on regional cerebral perfusion [29 –31]. There are also efforts to recognize MCI patients who are at greater risk of progression into dementia using transcranial ultrasound in longitudinal projects [32, 33], but there were only few cross-sectional transcranial ultrasound trials mapping the character of vascular impairment in MCI patients [26, 34].

The aim of this study was to compare the neurosonological parameters (flow velocities and cerebrovascular reserve capacity expressed as breath-holding index) in AD patients, MCI patients, and cognitively normal controls, with very low cerebrovascular burden to ensure the results will not be influenced by severe underlying atherosclerosis. We expect lower cerebrovascular reserve capacity in patients with AD and to a lesser extent also in MCI patients compared to cognitively normal controls. We suppose the impairment may be more prominent in amnestic MCI subtype than in non-amnestic MCI subtype.

MATERIALS AND METHODS

Ethics statement

The Ethical Committee of Motol University Hospital approved the study protocol, and all participants or their legal designates gave written consent.

Study subjects

Patient selection and the examinations were carried out at the outpatient Memory Clinic, Department of Neurology, Motol University Hospital, Charles University in Prague, which serves both as a first-line diagnostic clinic as well as a specialized consultation center.

The diagnosis of AD was made according to DSM-IV-TR criteria and the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria [35]. The presence of dementia was determined based on the loss of independence in daily living activities. The diagnosis of MCI was made according to the revised Petersen’s criteria [28]. For the subanalysis, the MCI patients were divided in amnestic and non-amnestic subtypes according to the results of neuropsychological testing. Amnestic MCI patients had reduced scores in verbal and non-verbal memory tests, in non-amnestic MCI patients the memory was spared.

Exclusion criteria were a history of stroke or transient ischemic attack (TIA), ischemic lesion or signs of marked cerebrovascular disease on brain MRI (more than 1 point on the Fazekas white matter lesions scale [36]), any stenosis of more than 10% of the large extracranial vessels (internal carotid artery, common carotid artery, vertebral artery), a stenosis of large intracranial vessels (anterior, medial and posterior cerebral artery, basilar artery), atrial fibrillation or artificial heart valve as a possible source of cardioembolism, pacemaker (because the participant could not undergo brain MRI), any cause of dementia other than AD, any other neurological disease involving the central nervous system, any psychiatric disease, intellectual disability, history of alcoholism or substance abuse, fewer than 10 points on the Mini-Mental State Examination (MMSE), and the participant’s inability to undergo any of the examinations.

Controls were chosen among healthy volunteers at the Memory Clinic (staff family members, attendants of the University of the Third Age). Inclusion criteria were absence of subjective memory complaints and absence of cognitive impairment (defined as scoring less than 1.5 standard deviations from the mean in all neuropsychological tests normalized for age and level of education). Exclusion criteria were identical to those applied to the study groups.

Past medical history, physical and neurological examinations, blood sampling

All patients and controls underwent a first screening visit by a medical doctor from our Memory Clinic wherein a detailed medical history was collected and physical and neurological examinations were performed. Each participant then underwent a standard battery of examinations of the Memory Clinic (i.e., neuropsychological assessment, brain MRI, and blood sampling). Additionally, as part of the study protocol, we asked standardized questions about vascular risk factors and medications and performed blood pressure measurement, BMI calculation, and neurosonological examination.

Past medical history was obtained by a medical doctor from the participant, close family member, or caregiver as well as from medical records. Vascular risk questions probed, through direct questioning, hypertension, diabetes mellitus, dyslipidemia, stroke, carotid stenosis, myocardial infarction, chronic heart failure, atrial fibrillation, ischemic disease of the lower extremities, chronic renal failure, smoking, alcohol intake, operations under global anesthesia and/or in extracorporeal circulation, and current medications. Any other important health issues were recorded. Blood pressure was measured on the brachial artery in a sitting position after at least five minutes of rest using an automated device. Blood samples were taken to assess blood lipids, homocysteine, glycated hemoglobin, and renal parameters. All examinations requiring participants to concentrate were performed between 8 AM and 3 PM.

Neuropsychological examination

A comprehensive battery of standardized neuropsychological tests consisted of the MMSE, Rey Auditory Verbal Learning Test (AVLT), Trail Making Tests (A and B), Digit Symbol-Coding Test, digit span task (forward and backward), Enhanced Cued Recall test (16-item picture version of the Free and Cued Selective Reminding Test), Rey-Osterrieth Complex Figure Test (immediate recall and copy condition), Controlled Oral Word Association Test (Czech version, letters N, K, P), Boston Naming Test. Depression was assessed using the Geriatric Depression Scale (15-item version) and Beck’s Anxiety Inventory. An experienced neuropsychologist performed all testing.

MRI

Brain scans were performed using a 1.5 T device (Avanto; Siemens AG, Erlangen, Germany) with fluid attenuated inversion recovery (FLAIR) sequences in the axial plane with the following parameters: repetition time/echo time/inversion time = 12210/138/2200 ms, flip angle 150°, 25 contiguous partitions and slice thickness 5 mm. Images were visually assessed by an experienced neuroradiologist, blinded to the diagnosis and clinical or cognitive measures, for the severity of subcortical white matter lesions (WML) using the Fazekas rating scale [36], a 4-point visual scale (0–3 points) where 0 points signifies the absence of WML, 1 signifies sporadic WML, 2 signifies a confluence of WML, and 3 signifies severe WML. Those scoring ≥ 2 points (moderate to severe WML) were excluded.

Neurosonological examination

The transcranial ultrasound examination was performed by an experienced neurosonologist using the duplex ultrasound Toshiba Nemio 30 (Toshiba Healthcare Systems, Tokyo, Japan) with a 2 MHz ultrasound probe for transcranial examination and a 4–9 MHz linear vascular probe for extracranial vessels. The TCCS sonogram was obtained through the temporal bone window bilaterally and occipital bone window with the participant lying on his back and left side, respectively. Three flow velocities (peak systolic velocity [PSV], mean flow velocity [MFV], and end diastolic velocity [EDV]) were measured in each artery in this order: 1) through the temporal bone window, the anterior cerebral artery (ACA) A1 segment, posterior cerebral artery (PCA) P1 segment, and middle cerebral artery (MCA) M1 segment were evaluated first on the right side, then the left; 2) through the occipital bone window, intracranial segments of right and left vertebral artery, respectively, and basal artery were evaluated. The resistance index (RI) and pulsatility index (PI) were calculated for each artery on each side according to these standard formulas: RI = (PSV-EDV)/PSV, and PI = (PSV-EDV)/MFV [37]. These indices are measures of the blood flow curve, and they reflect the resistance of the microvascular bed distal to the site of measurement. Each measurement taken in the MCA comprised three measurements at basal conditions alternating with three measurements after breath holding. The measurement at basal conditions was made in a single cycle of the flow curve in a steady state when the flow curve was stable. The breath-holding maneuver was performed according to Markus [21], where the participant held his breath as long as possible, and this time was noted in seconds. The breath movements were checked visually and using a hand placed on the patient’s chest to assure that the patient is not breathing during the breath-holding phase. In case of uncertainty the measurement was repeated. The velocity measurement was taken in the last single cycle of the flow curve immediately before the end of breath holding. The end-tidal CO₂ was not monitored. After each breath holding a minimum two-minute recovery period was given before another rest measurement was taken, allowing flow velocities to normalize. The BHI was calculated for both the left and right MCA using the formula BHI = (MFV_basal – MFV_BH)/MFV_basal]*100/t_BH [21]. All final parameters for the MCA on each side were arithmetically averaged from the three measurements. When the temporal bone window was insufficient, the ultrasound contrast agent (sulfur hexafluoride microbubbles) was applied intravenously. In such cases, a separate informed consent was obtained. To exclude those with internal carotid artery (ICA) stenoses or occlusions of at least 10%, the morphology of the vessel wall and the flow velocities in the ICA on both sides were assessed.

The examination scheme and exclusion criteria can be seen in flowchart in Fig. 3.

Statistical analysis

Comparisons were made between predefined groups: the AD patients, the MCI patients, and cognitively normal controls. Statistical significance for intergroup differences was assessed using Pearson’s chi-square or the Fisher exact test for categorical variables and the Student t test for continuous variables. Correlations between numerical variables were assessed using Pearson’s correlation coefficient, as appropriate. All tests were two-tailed and were conducted at the 5% significance level. Due to exploratory nature of our study we did not correct for multiple comparisons. Results are presented as means and standard deviations (SDs). Statistical analyses were performed using IBM SPSS Statistics 23.0.0 (IBM, USA) software.

RESULTS

Of 187 potential participants attending the Memory Clinic, 45 did not meet subgroup criteria, primarily because of normal neuropsychological examination in spite of subjective memory complaints. 142 were examined with the full diagnostic battery (medical history collection, neurological examination, MRI, neuropsychiatric examination, blood sampling, and neurosonological examination). Additional participants were excluded because of focal post-ischemic lesions seen on brain MRI (6), history of TIA (3), more than one point on the Fazekas rating scale for white matter lesions (31) [36], occlusion of the internal carotid artery (1), arrhythmia (14), dementia disorder other than AD (11 [7 with fronto-temporal lobe degeneration, 3 with Parkinson’s disease, and 1 with dementia with Lewy bodies]), other neurological disease involving the central nervous system (4 [1 with progressive supranuclear palsy and 3 with tumors]), psychiatric disease (1 with bipolar affective disorder), and insufficient temporal bone window and refusal of ultrasound contrast agent application (9).

Of the 62 included participants, 14 were in the AD group, 24 in the MCI group, and 24 were controls. Of those in the MCI group, 15 (62.5%) had amnestic MCI and 9 (37.5%), non-amnestic MCI.

There were more men in the AD group and the participants in MCI group were slightly older than AD and controls. As expected, there was a significant difference in MMSE, BMI, and education levels. The characteristics of the study groups are listed in Table 1.

Table 1

Groups characteristics and clinical data

Parameter	AD patients (n = 14)	MCI patients (n = 24)	Controls (n = 24)	Groups AD and MCI differed significantly (p < 0.05)^*	Groups AD and Controls differed significantly (p < 0.05)^*	Groups MCI and Controls differed significantly (p < 0.05)^*
Group characteristics
Age [y]	67.9 (11.1)	71.9 (7.3)	67.8 (6.4)	0.189	0.977	0.045^*
Men, n (%)	3 (21.4)	13 (54.2)	14 (58.3)	0.043^*	0.044^*	0.771
Years of education, [years]	12.8 (2.8)	15.5 (3.1)	16.0 (2.6)	0.015^*	0.002^**	0.613
Right-handedness, n (%)	12 (85.7)	22 (91.7)	23 (95.8)	0.542	0.965	0.256
BMI [kg/m²]	25.3 (3.0)	25.6 (3.2)	27.1 (2.1)	0.794	0.046^*	0.071
Fazekas score – 1 point, n (%)	11 (78.6)	21 (87.5)	16 (66.7)	0.650	0.488	0.086
Neuropsychological tests
MMSE	18.0 (4.6)	28.0 (1.6)	29.1 (1.2)	<0.001^***	<0.001^***	0.008^**
GDS	4.57 (2.76)	3.35 (3.12)	1.86 (1.91)	0.383	0.010^*	0.076
AVLT 1–5	22.71 (4.92)	37.60 (12.41)	50.71 (8.39)	<0.001^***	<0.001^***	<0.001^***
TMT A [time/s]	73.79 (22.90)	50.35 (24.96)	38.62 (11.63)	<0.001^***	0.007^**	0.018^*
TMT B [time/s]	269.17 (73.14)	150.13 (80.30)	85.32 (27.64)	<0.001^***	<0.001^***	0.001^**
COWAT Total, [points]	22.71 (8.28)	34.75 (13.13)	48.33 (9.56)	0.003^**	<0.001^***	<0.001^***
ECR-FR	1.57 (2.23)	6.40 (3.38)	10.33 (1.77)	0.013^*	<0.001^***	<0.001^***
ECR-TR	9.73 (5.39)	13.73 (3.67)	15.96 (0.21)	0.043^*	0.003^**	0.010^*
Personal history
Smoking [pack/y]	6.0 (10.2)	3.8 (8.7)	7.4 (11.6)	0.478	0.714	0.228
Alcohol [units/day]	0.54 (0.26)	0.60 (0.55)	0.98 (0.75)	0.713	0.017^*	0.056
Hypertension, n (%)	8 (57.1)	12 (50.0)	11 (45.8)	0.670	0.501	0.773
Diabetes mellitus, n (%)	0	4 (16.7)	3 (12.5)	0.276	0.283	1.0
Dyslipidemia, n (%)	2 (14.3)	10 (41.7)	9 (37.5)	0.147	0.160	0.768
Ischemic heart disease, n (%)	0	3 (12.5)	0	0.283	n.a.	1.0
Peripheral artery disease, n (%)	0	1 (4.2)	0	1.0	n.a.	0.234
Chronic renal disease, n (%)	1 (7.1)	3 (12.5)	0	1.0	0.368	0.234
Hypothyreosis, n (%)	3 (21.4)	2 (8.3)	4 (16.7)	0.337	1.0	0.666
Operations in general anesthesia [number]	1.42 (1.38)	1.71 (1.78)	1.92 (1.87)	0.623	0.422	0.696
Diastolic blood pressure [mmHg]	82.2 (10.9)	76.9 (10.7)	77.9 (9.6)	0.299	0.394	0.788
Systolic blood pressure [mmHg]	139.5 (24.1)	134.3 (19.0)	133.0 (18.9)	0.584	0.530	0.849

The values represent mean (SD) unless indicated otherwise. ^* p≤0.05; ^** p≤0.01; ^*** p≤0.001. AD, Alzheimer’s disease; AVLT 1–5, Auditory Verbal Learning Test, sum of trials 1 to 5; BMI, body mass index; COWAT, Controlled Oral Word Association Test, total score; ECR-FR, Enhanced Cued Recall, free recall; ECR-TR, Enhance Cued Recall, total recall after cueing; GDS, Geriatric Depression Scale; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; TMT A, Trail Making Test A; TMT B, Trail Making Test B.

There were no significant differences in vascular risk factors or atherosclerotic disease (smoking, hypertension, dyslipidemia, diabetes mellitus, myocardial infarction, or ischemic disease of lower extremities), chronic renal disease, hypothyreosis, procedures under general anesthesia in personal history, and in laboratory findings. The alcohol intake was higher in controls than in AD patients (Tables 1 and 2). There were no differences in medications between the groups, except gingko biloba and acetylcholinesterase inhibitors were not taken by controls (Table 3).

Table 2

Laboratory findings

Parameter	AD patients (n = 14)	MCI patients (n = 24)	Controls (n = 24)	Groups AD and MCI differed significantly (p < 0.05)^*	Groups AD and Controls differed significantly (p < 0.05)^*	Groups MCI and Controls differed significantly (p < 0.05)^*
Creatinine [μmol/l]	72.77 (19.88)	78.29 (28.61)	71.87 (13.41)	0.541	0.872	0.333
Triacylglycerides [mmol/l]	1.28 (0.61)	1.47 (0.77)	1.67 (1.24)	0.489	0.335	0.498
Total cholesterol [mmol/l]	5.15 (1.04)	5.29 (0.99)	5.31 (0.92)	0.695	0.645	0.952
HDL [mmol/l]	1.37 (0.37)	1.53 (0.42)	1.36 (0.35)	0.292	0.941	0.140
LDL [mmol/l]	3.12 (0.88)	3.06 (0.88)	3.16 (0.72)	0.856	0.884	0.670
Apolipoprotein B [g/l]	0.93 (0.18)	0.97 (0.18)	0.96 (0.15)	0.724	0.795	0.890
Lipoprotein A [mg/ml]	307.00 (502.74)	462.46 (652.73)	484.32 (510.31)	0.670	0.573	0.934
Homocysteine [μmol/l]	18.63 (12.17)	13.29 (3.68)	13.26 (7.42)	0.058	0.137	0.989
Glycated hemoglobin [mmol/l]	40.43 (4.54)	40.86 (6.12)	41.00 (5.66)	0.867	0.812	0.939

The values represent mean (SD) unless indicated otherwise. AD, Alzheimer’s disease; HDL, high density lipoproteins; MCI, mild cognitive impairment; LDL, low density lipoproteins.

Table 3

Medication

Parameter	AD patients (n = 14)	MCI patients (n = 24)	Controls (n = 24)	Groups AD and MCI differed significantly (p < 0.05)^*	Groups AD and Controls differed significantly (p < 0.05)^*	Groups MCI and Controls differed significantly (p < 0.05)^*
Statine	4 (28.6)	8 (33.3)	6 (25.0)	1.0	1.0	0.525
Fibrate	1 (7.1)	0	1 (4.2)	0.368	1.0	1.0
Antiplatelet drug	1 (7.1)	5 (20.8)	2 (8.3)	0.383	1.0	0.416
Anticoagulants	0	2 (8.3)	0	0.522	1.0	0.489
Diuretic	2 (14.3)	4 (16.7)	5 (20.8)	1.0	1.0	1.0
Calcium channel blocker	2 (14.3)	3 (12.5)	5 (20.8)	1.0	1.0	0.701
Beta blocker	2 (14.3)	5 (20.8)	4 (16.7)	1.0	1.0	1.0
ACEI/ARA	3 (21.4)	10 (41.7)	9 (37.5)	0.294	0.472	0.768
Central antihypertensive	0	3 (12.5)	2 (8.3)	0.283	0.522	1.0
Ginkgo	4 (28.6)	9 (37.5)	1 (4.2)	0.728	0.052	0.004^**
AChEI	7 (50.0)	5 (20.8)	0	0.081	<0.001^***	0.05
Memantine	1 (7.1)	0	0	0.368	0.368	1.0
SSRI	4 (28.6)	3 (12.5)	1 (4.2)	0.387	0.052	0.609
Thyroidal hormone replacement therapy	0	5 (20.8)	4 (16.7)	0.137	0.276	1.0
OAD	0	2 (8.3)	2 (8.3)	0.522	0.522	1.0

The values represent n (%) unless indicated otherwise. ^* p≤0.05; ^** p≤0.01; ^*** p≤0.001. AD, Alzheimer’s disease; ACEI/ARA, angiotensin converting enzyme inhibitor/Angiotensin receptor II antagonist; AChEI, acetylcholine esterase inhibitor; MCI, mild cognitive impairment; OAD, oral antidiabetic; SSRI, selective serotonine reuptake inhibitor.

The ultrasound parameters were compared between the respective groups. The flow velocities in the main intracranial arteries, both at basal conditions and at breath holding (in case of MCA), consistently decreased with the severity of cognitive decline (Table 4 – only MCA results). The differences were significant between AD and controls for MFV and EDV in the right MCA after breath holding (p = 0.004 and p = 0.001, respectively) and for all velocities (PSV, MFV, and EDV) in the left MCA after breath holding (p = 0.003, p = 0.002, and p = 0.001, respectively). Both RI and PI increased with the severity of cognitive decline, suggesting higher resistance in the cerebral vessels. These differences were significant between AD and controls in the right MCA after breath holding (p = 0.048 and p = 0.028 respectively).

Table 4

Ultrasound parameters in MCA

Parameter	AD patients (n = 14)	MCI patients (n = 24)	Controls (n = 24)	Groups AD and MCI differed significantly (p < 0.05)^*	Groups AD and Controls differed significantly (p < 0.05)^*	Groups MCI and Controls differed significantly (p < 0.05)^*
Right MCA
Basal conditions
PSV, [cm/s]	65.35 (18.83)	74.07 (26.47)	71.44 (15.39)	0.309	0.296	0.681
MFV, [cm/s]	39.61 (11.26)	44.45 (15.63)	45.69 (9.67)	0.340	0.094	0.747
EDV, [cm/s]	25.05 (7.26)	28.96 (10.02)	29.85 (6.71)	0.232	0.051	0.724
RI	0.61 (0.07)	0.60 (0.08)	0.58 (0.05)	0.713	0.187	0.265
PI	1.03 (0.22)	1.03 (0.24)	0.91 (0.11)	0.951	0.095	0.061
Breath holding
PSV, [cm/s]	76.20 (22.34)	82.53 (28.36)	89.22 (16.81)	0.499	0.053	0.334
MFV, [cm/s]	47.81 (13.24)	54.17 (20.63)	59.92 (10.35)	0.330	0.004^**	0.235
EDV, [cm/s]	31.92 (8.34)	37.63 (17.61)	41.57 (7.25)	0.283	0.001^**	0.321
t, [s]	48.0 (29.5)	50.1 (18.0)	58.5 (27.1)	0.801	0.284	0.234
RI	0.57 (0.08)	0.55 (0.08)	0.53 (0.05)	0.364	0.048^*	0.420
PI	0.93 (0.21)	0.87 (0.24)	0.80 (0.13)	0.485	0.028^*	0.225
BHI	0.53 (0.21)	0.64 (0.36)	0.67 (0.24)	0.271	0.097	0.794
Left MCA
Basal conditions
PSV, [cm/s]	64.23 (15.41)	73.32 (22.43)	75.17 (17.51)	0.221	0.075	0.755
MFV, [cm/s]	40.20 (9.91)	45.43 (13.91)	47.71 (12.15)	0.259	0.073	0.555
EDV, [cm/s]	26.21 (7.76)	29.81 (9.28)	31.43 (7.85)	0.263	0.067	0.523
RI	0.60 (0.09)	0.59 (0.07)	0.58 (0.05)	0.782	0.386	0.505
PI	1.00 (0.29)	0.98 (0.21)	0.93 (0.12)	0.801	0.306	0.349
Breath holding
PSV, [cm/s]	66.90 (17.03)	83.12 (29.38)	94.69 (26.95)	0.091	0.003^**	0.175
MFV, [cm/s]	44.79 (11:93)	55.44 (21.98)	64.40 (18.50)	0.132	0.002^**	0.145
EDV, [cm/s]	31.48 (8.90)	39.52 (16.73)	45.63 (13.45)	0.081	0.001^**	0.182
t, [s]	44.5 (19.4)	56.7 (23.9)	67.6 (35.1)	0.145	0.016^**	0.235
RI	0.54 (0.08)	0.53 (0.06)	0.51 (0.07)	0.765	0.377	0.420
PI	0.84 (0.23)	0.82 (0.17)	0.77 (0.16)	0.793	0.344	0.360
BHI	0.34 (0.16)	0.49 (0.31)	0.57 (0.23)	0.124	0.005^**	0.365

The values represent mean (SD) unless indicated otherwise. ^* p≤0.05; ^** p≤0.01; ^*** p≤0.001. AD, Alzheimer’s disease; BHI, breath-holding index; EDV, end diastolic velocity; MCA, middle cerebral artery; MCI, mild cognitive impairment; MFV, mean flow velocity; PI, pulsatility index; PSV, peak systolic velocity; RI, resistance index; t, time of breath-holding.

Differences were found in the BHI in the left MCA. It decreased as the severity of cognitive decline increased suggesting decreasing cerebrovascular reserve capacity (0.57±0.23 for controls, 0.49±0.31 for those with MCI, and 0.34±0.16 for those with AD). The difference was significant between AD and controls (p = 0.005; Fig. 1). The same effect was seen on the right side (0.67±0.24 for controls, 0.64±0.36 for those with MCI, and 0.53±0.21 for those with AD; Fig. 2), but the differences were not statistically significant. Homogeneity of variances, as assessed by Levene’s test for equality of variances, was found on the left side (p = 0.222) but not on the right side (p = 0.036). Finally, BHIs were different between groups only on the left side (F[2, 54] = 3.217, p = 0.048).

Fig.1

BHI in left MCA. Bold horizontal line represents median, box represents the upper and lower quartile, the upper and lower whiskers represent confidence interval 95%. AD, Alzheimer’s disease; BHI, breath-holding index; MCA, middle cerebral artery; MCI, mild cognitive impairment.

Fig.2

BHI in right MCA. Bold horizontal line represents median, box represents the upper and lower quartile, the upper and lower whiskers represent confidence interval 95%. AD, Alzheimer’s disease; BHI, breath-holding index; MCA, middle cerebral artery; MCI, mild cognitive impairment.

Fig.3

Examination scheme and exclusion criteria. AD, Alzheimer’s disease; BMI, body mass index; CNS, central nervous system; MMSE, Mini-Mental State Examination; MRI, magnetic resonance imaging; TIA, transient ischemic attack.

The average time of breath holding is stated in the Table 4. In four AD subjects, two MCI subjects and two healthy controls the time of breath holding was lower than usually accepted 30 s (in all of them >26 s). However, according to Markus et al., the reliability of BHI does not differ whether the time of breath holding is short (<27 s) or long (>27 s) [21].

When all ultrasound parameters were compared separately for those with amnestic versus non-amnestic MCI, no significant differences were found.

On the left side the BHI correlated with AVLT score (AVLT 1–5) (r = 0.299, p = 0.031), Controlled Oral Word Association Test (letter K, Czech version) (r = 0.289, p = 0.034), and Rey-Osterrieth Complex Figure Test (recall) (r = 0.300, p = 0.027). Additionally, BHI for the left side negatively correlated with number of mistakes in the Boston Naming Test (r = –0.337, p = 0.035).

DISCUSSION

Our main finding was the decreased cerebrovascular reserve capacity in AD patients in the stage of dementia, with very low cerebrovascular burden. The cerebrovascular reserve capacity seems to be decreasing with the increasing severity of cognitive decline, although in MCI patients, the results did not reach the statistical significance. Our finding of impaired cerebrovascular reserve capacity in AD patients is comparable with other cross-sectional studies using TCD or TCCS [22 –26]. Two of them employed the BHI for the assessment of the cerebrovascular reserve capacity [25, 26]. Lower BHI also correlates with a faster progression of cognitive decline in AD patients in a longitudinal study [27]. Despite the trend, the differences between our MCI group and control group were not significant. This is in accordance with two cross-sectional transcranial ultrasound trials assessing the cerebrovascular reserve capacity in MCI (one of them using BHI) [26, 38]. A substantial strength of our study is its very strict exclusion criteria with respect to cerebrovascular disease. All participants had essentially no ICA stenoses and very low extents of WML (Fazekas scores of 0 or 1 [36]). Because our cohort had very low cerebrovascular burden and because we found microvessel impairment, we can assume that the pathophysiological mechanism leading to microangiopathy in AD has at least a combined etiology and that the atherosclerosis is only one of the contributing factors.

The assessment of cerebrovascular reserve capacity seems to be more promising than the assessment of flow velocities only. We found that flow velocities in cerebral arteries were lower in impaired participants than in controls, but these findings were not significant at basal conditions. Other studies revealed significant differences in flow velocities at basal conditions between AD patients and controls, but the results of these studies varied (significant differences were found in various vessels and various flow velocities) [24 , 39–42]. However, we found significant differences in flow velocities in the MCA after the CO₂ challenge, which is, again, associated with pathological cerebrovascular reserve capacity. It is also possible that our study, which included patients with very well defined and less-expressed cerebrovascular disease, could not detect smaller changes in velocities at basal conditions.

The reduced or exhausted cerebrovascular reserve capacity is an expression of chronic hypoperfusion and the inability of brain arterioles to further dilate or to dilate at all, when oxygen demand is increasing. This reduced capacity cannot, unlike in plain hypoperfusion, be explained by lower demand by degenerating neurons, and it expresses impairment in microvessel function in AD. However, the exact cause remains unclear. The role of amyloid in the pathophysiology of microangiopathy in AD can be supported by compromised cerebrovascular reserve capacity in cerebral amyloid angiopathy [43]. Another hypothesis suggests that acetylcholine production is insufficient for the vasodilatation needed. Therapeutic tests with acetylcholine inhibitors (galantamine or donepezil) demonstrated an increase in flow velocities and improvements in cerebrovascular reserve capacity both in vascular disease and AD after therapy [22 , 45].

The finding that BHI was significantly lower only in the left, and not the right MCA, was surprising. Such asymmetry in cerebrovascular reserve capacity was, until now, not noted in previous Doppler studies in AD patients [22 –25]. This is most probably related to technique. All other researchers used classical TCD, simultaneously measuring MCA on both sides using a special headframe and averaging them. We used TCCS, which does not provide simultaneous measurement and gives separate values for each side. A natural asymmetry in brain perfusion was found in several SPECT and PET studies of healthy individuals. This asymmetry increases with age and shows a lower perfusion in the left hemisphere [46, 47]. One SPECT study also showed asymmetry in AD patients with reduced regional cerebral blood flow in the left hippocampus and parahippocampal gyrus [48]. Our finding suggests more prominent decreases in cerebrovascular reserve capacity on the left side, and this should be taken into account when assessing patients with MCI because averaging values could mask the severity of impairment.

Our study did not prove statistical differences between those with amnestic versus non-amnestic MCI. This could be a focus in future studies because it is known that amnestic MCI most often progresses to AD [28]. The non-amnestic MCI seems to be more inhomogeneous group concerning etiologies and prognoses of cognitive decline than the amnestic MCI group. In addition, amnestic MCI patients with pathological values of BHI have greater risks of converting to dementia than those with normal values as shown in longitudinal projects [32, 33]. This association also supports our finding that BHI significantly correlates with AVLT, which is a memory test.

The mean BHI values found among cognitively normal controls (0.67 on the right, 0.57 on the left) were lower than usually quoted normal values. The literature does not provide much guidance for norms, but Zavoreo et al. [49] gives them as 1.29±0.31 for those aged 60 to 69 years and 1.13±0.33 for those aged more than 70 years. Vernieri et al. [50] examined cerebrovascular reserve capacity in patients with internal carotid artery occlusion and found that BHI in the MCA on the contralateral side (with no or only mild [<30% ] stenosis of the ICA) was 1.07±0.3 in an age range 67.8±5.5 years. The same author postulated a cut-off value of 0.69 to differentiate normal BHI and pathological BHI in patients with internal carotid stenoses (BHI <0.69 was found to be pathological and indicative of a greater risk for stroke or TIA ipsilateral to the side of internal carotid artery stenosis). An explanation for this discrepancy is that a different ultrasound technique was used. All older studies, including those by Markus [21], Zavoreo et al. [49], and Vernieri et al. [50] used the classical TCD technique without the B-mode, whereas we used TCCS with the B-mode. The insonation angle correction could have influenced the measurements and values for flow velocities, from which BHI is calculated.

We found the mean BMI to be significantly lower among those with AD than in the other two groups, consistent with the literature, but the cause is not clear. Both hormonal changes and cognitive decline with worsening independence likely play roles [51, 52].

The average amount of education was higher among our controls and MCI patients than in AD patients. This is likely a result of our recruitment of volunteers; those with higher levels of education may be more motivated to participate in an experimental project. Also, patients with higher levels of education may be motivated to come to assessment in an earlier stage of the disease.

The study groups were not matched in terms of gender (fewer men in the AD group than in the MCI and control group). This is probably a result of the small sample size.

A major limitation of our study is its sample size. Recruiting subjects was complicated because of our strict exclusion criteria.

Another limitation was that some participants used antihypertensives, acetylcholinesterase inhibitors, and certain other drugs that can influence cerebral perfusion.

Conclusion

Our study shows that AD is associated with decreased cerebrovascular reserve capacity. This suggests chronic hypoperfusion and underlying microangiopathy. The etiology of microangiopathy cannot be assigned only to atherosclerosis. Microvascular impairment can be seen also in those with MCI. Cerebrovascular reserve capacity expressed as BHI seems to be the more promising neurosonological parameter in patients with cognitive decline than flow velocities only, particularly because the range of normal values of flow velocities is broad and because the cerebrovascular reserve capacity probably corresponds better to the underlying pathology. Attention should also be paid to left-right asymmetry in decreased cerebrovascular reserve capacity, expecting more prominent impairment on the left side. Further studies to confirm this finding are needed.

Footnotes

ACKNOWLEDGMENTS

Supported by IGA MZ CR grant (No. NT/13319). Supported by GAUK grant (No 347711). Supported by the project FNUSA- ICRC (no. CZ.1.05/1.1.00/02.0123) from the European Regional Development Fund.

Authors’ disclosures available online ().

References

Kalaria

(2002) Similarities between Alzheimer’s disease and vascular dementia. J Neurol Sci 203-204, 29–34.

Skoog

, Lernfelt

, Landahl

, Palmertz

, Andreasson

, Nilsson

, Persson

, Odén

, Svanborg

(1996) 15-year longitudinal study of blood pressure and dementia. Lancet 347, 1141–1145.

Ott

, Stolk

, van Harskamp

, Pols

, Hofman

, Breteler

(1999) Diabetes mellitus and the risk of dementia: The Rotterdam Study. Neurology 53, 1937–1942.

Notkola

, Sulkava

, Pekkanen

, Erkinjuntti

, Ehnholm

, Kivinen

, Tuomilehto

, Nissinen

(1998) Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer’s disease. Neuroepidemiology 17, 14–20.

Hofman

, Ott

, Breteler

, Bots

, Slooter

, van Harskamp

, van Duijn

, Van Broeckhoven

, Grobbee

(1997) Atherosclerosis, apolipoprotein E, and prevalence of dementia and Alzheimer’s disease in the Rotterdam Study. Lancet 349, 151–154.

van Oijen

, de Jong

, Witteman

, Hofman

, Koudstaal

, Breteler

(2007) Atherosclerosis and risk for dementia. Ann Neurol 61, 403–410.

Breteler

, van Swieten

, Bots

, Grobbee

, Claus

, van den Hout

, van Harskamp

, Tanghe

, de Jong

, van Gijn

(1994) Cerebral white matter lesions, vascular risk factors, and cognitive function in a population-based study: The Rotterdam Study. Neurology 44, 1246–1252.

DeCarli

, Miller

, Swan

, Reed

, Wolf

, Carmelli

(2001) Cerebrovascular and brain morphologic correlates of mild cognitive impairment in the National Heart, Lung, and Blood Institute Twin Study. Arch Neurol 58, 643–647.

, Mungas

, Petkov

, Eberling

, Zrelak

, Buonocore

, Brunberg

, Haan

, Jagust

(2002) Brain structure and cognition in a community sample of elderly Latinos. Neurology 59, 383–391.

10.

Cordonnier

, van der Flier

, Sluimer

, Leys

, Barkhof

, Scheltens

(2006) Prevalence and severity of microbleeds in a memory clinic setting. Neurology 66, 1356–1360.

11.

Wolf

, Ecke

, Bettin

, Dietrich

, Gertz

(2000) Do white matter changes contribute to the subsequent development of dementia in patients with mild cognitive impairment? A longitudinal study. Int J Geriatr Psychiatry 15, 803–812.

12.

de la Torre

(2004) Alzheimer’s disease is a vasocognopathy: A new term to describe its nature. Neurol Res 26, 517–524.

13.

Iadecola

(2010) The overlap between neurodegenerative and vascular factors in the pathogenesis of dementia. Acta Neuropathol 120, 287–296.

14.

Pimentel-Coelho

, Rivest

(2012) The early contribution of cerebrovascular factors to the pathogenesis of Alzheimer’s disease. Eur J Neurosci 35, 1917–1937.

15.

Weller

, Subash

, Preston

, Mazanti

, Carare

(2008) Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer’s disease. Brain Pathol 18, 253–266.

16.

Kokjohn

, Maarouf

, Roher

(2012) Is Alzheimer’s disease amyloidosis the result of a repair mechanism gone astray? Alzheimers Dement 8, 574–583.

17.

Peisker

, Bartoš

, Skoda

, Ibrahim

, Kalvach

(2010) Impact of aging on cerebral vasoregulation and parenchymal integrity. J Neurol Sci 299, 112–115.

18.

Bishop

, Powell

, Rutt

, Browse

(1986) Transcranial Doppler measurement of middle cerebral artery blood flow velocity: A validation study. Stroke 17, 913–915.

19.

Dahl

, Lindegaard

, Russell

, Nyberg-Hansen

, Rootwelt

, Sorteberg

, Nornes

(1992) A comparison of transcranial Doppler and cerebral blood flow studies to assess cerebral vasoreactivity. Stroke 23, 15–19.

20.

Müller

, Voges

, Piepgras

, Schimrigk

(1995) Assessment of cerebral vasomotor reactivity by transcranial Doppler ultrasound and breath-holding. A comparison with acetazolamide as vasodilatory stimulus. Stroke 26, 96–100.

21.

Markus

, Harrison

(1992) Estimation of cerebrovascular reactivity using transcranial Doppler, including the use of breath-holding as the vasodilatory stimulus. Stroke 23, 668–673.

22.

Bar

, Boettger

, Seidler

, Mentzel

, Terborg

, Sauer

(2007) Influence of galantamine on vasomotor reactivity in Alzheimer’s disease and vascular dementia due to cerebral microangiopathy. Stroke 38, 3186–3192.

23.

Lee

, Jung

, Lee

(2007) Decreased vasomotor reactivity in Alzheimer’s disease. J Clin Neurol 3, 18–23.

24.

Vicenzini

, Ricciardi

, Altieri

, Puccinelli

, Bonaffini

, Di Piero

, Lenzi

(2007) Cerebrovascular reactivity in degenerative and vascular dementia: A transcranial Doppler study. Eur Neurol 58, 84–89.

25.

Stefani

, Sancesario

, Pierantozzi

, Leone

, Galati

, Hainsworth

, Diomedi

(2009) CSF biomarkers, impairment of cerebral hemodynamics and degree of cognitive decline in Alzheimer’s and mixed dementia. J Neurol Sci 283, 109–115.

26.

Shim

, Yoon

, Shim

, Kim

, An

, Yang

(2015) Cognitive correlates of cerebral vasoreactivity on transcranial Doppler in older adults. J Stroke Cerebrovasc Dis 24, 1262–1269.

27.

Silvestrini

, Pasqualetti

, Baruffaldi

, Bartolini

, Handouk

, Matteis

, Moffa

, Provinciali

, Vernieri

(2006) Cerebrovascular reactivity and cognitive decline in patients with Alzheimer disease. Stroke 37, 1010–1015.

28.

Petersen

(2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256, 183–194.

29.

Caroli

, Testa

, Geroldi

, Nobili

, Barnden

, Guerra

, Bonetti

, Frisoni

(2007) Cerebral perfusion correlates of conversion to Alzheimer’s disease in amnestic mild cognitive impairment. J Neurol 254, 1698–1707.

30.

Encinas

, De Juan

, Marcos

, Gil

, Barabash

, Fernández

, De Ugarte

, Cabranes

(2003) Regional cerebral blood flow assessed with 99mTc-ECD SPET as a marker of progression of mild cognitive impairment to Alzheimer’s disease. Eur J Nucl Med Mol Imaging 30, 1473–1480.

31.

Huang

, Wahlund

, Svensson

, Winblad

, Julin

(2002) Cingulate cortex hypoperfusion predicts Alzheimer’s disease in mild cognitive impairment. BMC Neurol 2, 9.

32.

Viticchi

, Falsetti

, Vernieri

, Altamura

, Bartolini

, Luzzi

, Provinciali

, Silvestrini

(2012) Vascular predictors of cognitive decline in patients with mild cognitive impairment. Neurobiol Aging 33, 1127.e1–1127.e9.

33.

Buratti

, Balestrini

, Altamura

, Viticchi

, Falsetti

, Luzzi

, Provinciali

, Vernieri

, Silvestrini

(2015) Markers for the risk of progression from mild cognitive impairment to Alzheimer’s disease. J Alzheimers Dis 45, 883–890.

34.

Roher

, Garami

, Tyas

, Maarouf

, Kokjohn

, Belohlavek

, Vedders

, Connor

, Sabbagh

, Beach

, Emmerling

(2011) Transcranial doppler ultrasound blood flow velocity and pulsatility index as systemic indicators for Alzheimer’s disease. Alzheimers Dement 7, 445–455.

35.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(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 7, 263–269.

36.

Wahlund

, Barkhof

, Fazekas

, Bronge

, Augustin

, Sjögren

, Wallin

, Ader

, Leys

, Pantoni

, Pasquier

, Erkinjuntti

, Scheltens

, European Task Force on Age-Related White Matter Changes (2001) A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 32, 1318–1322.

37.

David Školoudík

OŠ

, Michal

Bar

, Miroslav

Brozman

, Daniel

Václavík

(2003) Galén. Neurosonologie.

38.

Anzola

, Galluzzi

, Mazzucco

, Frisoni

(2011) Autonomic dysfunction in mild cognitive impairment: A transcranial Doppler study. Acta Neurol Scand 124, 403–409.

39.

Caamaño

, Gómez

, Cacabelos

(1993) Transcranial Doppler ultrasonography in senile dementia: Neuropsychological correlations. Methods Find Exp Clin Pharmacol 15, 193–199.

40.

Sun

, Zhu

, Liu

, Zhu

, Zhou

, Liu

(2007) Decreased cerebral blood flow velocity in apolipoprotein E epsilon4 allele carriers with mild cognitive impairment. Eur J Neurol 14, 150–155.

41.

Claassen

, Diaz-Arrastia

, Martin-Cook

, Levine

, Zhang

(2009) Altered cerebral hemodynamics in early Alzheimer disease: A pilot study using transcranial Doppler. J Alzheimers Dis 17, 621–629.

42.

Gucuyener

, Yenilmez

, Ayranci

, Ozdemir

, Uzuner

, Ozkan

, Kaptanoglu

, Ozdemir

(2010) An analysis of changes in cerebral blood flood velocities in depressive pseudo-dementia and Alzheimer disease patients. Neurologist 16, 358–363.

43.

Menendez-Gonzalez

, Garcia-Garcia

, Calleja

, Rojo

, Ribacoba

(2011) Vasomotor reactivity is similarly impaired in patients with Alzheimer’s disease and patients with amyloid hemorrhage. J Neuroimaging 21, e83–e85.

44.

Rosengarten

, Paulsen

, Molnar

, Kaschel

, Gallhofer

, Kaps

(2007) Activation-flow coupling differentiates between vascular and Alzheimer type of dementia. J Neurol Sci 257, 149–154.

45.

Ghorbani

, Chitsaz

, Shishegar

, Akbari

(2010) Evaluation of the effect of donepezil on cerebral blood flow velocity in Alzheimer’s disease. Neurosciences (Riyadh) 15, 172–176.

46.

Bekier

(1976) Asymmetric brain perfusion in the aged? Nuklearmedizin 15, 217–222.

47.

Van Laere

, Dierckx

(2001) Brain perfusion SPECT: Age- and sex-related effects correlated with voxel-based morphometric findings in healthy adults. Radiology 221, 810–817.

48.

Kogure

, Matsuda

, Ohnishi

, Asada

, Uno

, Kunihiro

, Nakano

, Takasaki

(2000) Longitudinal evaluation of early Alzheimer’s disease using brain perfusion SPECT. J Nucl Med 41, 1155–1162.

49.

Zavoreo

, Demarin

(2004) Breath Holding Index in the evaluation of cerebral vasoreactivity. Acta Clin Croat 43, 15–19.

50.

Vernieri

, Pasqualetti

, Passarelli

, Rossini

, Silvestrini

(1999) Outcome of carotid artery occlusion is predicted by cerebrovascular reactivity. Stroke 30, 593–598.

51.

Buchman

, Wilson

, Bienias

, Shah

, Evans

, Bennett

(2005) Change in body mass index and risk of incident Alzheimer disease. Neurology 65, 892–897.

52.

Berlinger

, Potter

(1991) Low Body Mass Index in demented outpatients. J Am Geriatr Soc 39, 973–978.