Carotid Endarterectomy Improves Cerebrovascular
Reserve Capacity Preferentially in Patients with
Preoperative Impairment as Indicated by
Asymmetric BOLD Response to Hypercapnia
S.D. Goodea,b, N. Altafa,b, D.P. Auera,*, S.T.R. MacSweeneya,b
aDepartment of Academic Radiology, University of Nottingham, Queens Medical Centre, Nottingham NG7 2UH, UK
bDepartment of Vascular and Endovascular Surgery, Queens Medical Centre, Nottingham NG7 2UH, UK
Submitted 17 February 2009; accepted 14 June 2009
Available online 9 September 2009
Carotid artery disease;
mechanism causing ischaemic injury is considered to be thromboembolic, however compromise
of cerebral haemodynamics is considered to be a significant factor. Removal of the embolic
source is accepted as the major benefit from Carotid Endarterectomy (CEA), however improve-
ment in cerebral haemodynamics may be another beneficial outcome as suggested by Transcra-
nial doppler (TCD). Blood Oxygen Level-Dependent (BOLD) hypercapnia functional Magnetic
study was to assess the effects of carotid surgery on cerebral haemodynamics in patients with
Materials and methods: Seventeen patients with symptomatic internal carotid artery stenosis
were scanned using a clinical 1.5T MR scanner. Scanning was done immediately prior to and
between 4 and 8 weeks after CEA. 10% carbon dioxide was administered to achieve transient
episodes of hypercapnia. The data was analyzed using FMRIB Software Library (FSL) software
to derive percentage signal change (PSC) for the grey matter of the middle cerebral artery
(MCA-GM) territory for both hemispheres. MCA-GM PSC was furthermore normalized to the
contralateral hemisphere to derive an Hemispheric Asymmetry Index (hAI) for all patients pre-
Results: Ipsilateral GM CVR improved significantly following CEA (2.47% preoperatively vs. 2.73%
postoperatively, p Z 0.038). There was no change in CVR in the contralateral grey and white
matter MCA territories (p Z 0.27, p Z 0.1). Also, the hAI was significantly more shifted to the
ipsilateral hemisphere after CEA (preoperative hAI ?0.56, vs. ?3.90 postoperatively,
p Z 0.02). Patients with an impaired hAI preoperatively were found to show the greatest
improvement in PSC and hAI following CEA (p Z0.007).
Purpose: In patients with symptomatic carotid artery disease the predominant
* Corresponding author.
E-mail address: email@example.com (D.P. Auer).
1078-5884/$36 ª 2009 Published by Elsevier Ltd on behalf of European Society for Vascular Surgery.
Eur J Vasc Endovasc Surg (2009) 38, 546e551
Conclusions: CEA resulted in improved CVR in patients with carotid artery disease as shown by
cerebrovascular reserve. Moreover, hypercapnia BOLD fMRI may be a useful clinical tool in pre-
dicting this therapeutic potential in patients with severe carotid artery disease.
ª 2009 Published by Elsevier Ltd on behalf of European Society for Vascular Surgery.
Carotid Endarterectomy (CEA) is performed to prevent
stroke caused by a stenotic lesion in the ipsilateral internal
carotid artery (ICA). A series of randomized controlled trials
in the 1990s proved the benefits of CEA.1,2In patients with
symptomatic carotid artery disease the predominant mech-
anism causing ischaemic injury is considered to be throm-
boembolic, however carotid stenosis and occlusion can also
compromise cerebral haemodynamics. It is being increas-
ingly recognized that cerebral haemodynamic impairment
may convey an increased risk of stroke in these patients.3e5
Moreover, CVR (Cerebrovascular Reserve) impairment may
cerebral hypoperfusion secondary to carotid disease may
lead to impaired clearance of emboli and therefore
increasing the risk of developing a clinical stroke as a result
from carotid plaque embolization.5The identification of
patients with impairment of cerebral haemodynamics may
have a significant effect on their management.
The haemodynamic effect of ICA stenosis is not only
determined by the severity of stenosis, as the quality of the
collateral circulation is equally important. Collateral flow
capacity depends on anatomical characteristics and the
presence of additional steno-occlusive disease such as
contralateral ICA stenosis/occlusion. Therefore, individual
assessment of the haemodynamic down stream effects of
a carotid artery disease could help to identify high-risk
patients, in particular those patients with bilateral disease.
Removal of the embolic source is accepted as the major
benefit from CEA, however improvement in cerebral
haemodynamics may be another beneficial outcome in
selected patients. In fact previous studies using transcranial
tomography demonstrated improved CVR following CEA.6e9
A parameter used to test the haemodynamic status of
the cerebral circulation is the CVR capacity. CVR refers to
the capacity of cerebral blood flow to increase in response
to regional metabolic demand. Measuring CVR can be done
by various techniques; flow velocity measurements using
TCD, Cerebral Blood Flow (CBF) measurements using Posi-
tron emission tomography (PET) or Arterial Spin Labeling
(ASL) or more recently Blood oxygen level-dependent
(BOLD) functional magnetic resonance imaging (fMRI). To
actually assess the dynamic reserve status a vasodilatory
substance is also needed such as CO2or acetazolamide.
Under normal physiological conditions, these substances
induce vasodilatation, resulting in an increase in CBF. Any
preexisting vasodilation will interfere with the ability of the
cerebral vessels to dilate further in response to a vaso-
dilatory stimulus. Thus, the degree of CO2 reactivity
provides indirect information about the extent of the
cerebral blood vessel dilatation, directly reflecting the
remaining reserve capacity of the cerebral circulation.
BOLD fMRI combines the advantages of availability and
high spatial resolution, without the use of radiation. This
technique relies on endogenous contrast being created
from changing levels in haemoglobin oxygenation satura-
tion. BOLD fMRI can be used to produce high spatial reso-
lution CVR maps, semiquantitative data on brain reactivity
and also information on the cerebral vascular response.
This technique can be used to assess CVR over any chosen
region of interest (e.g. MCA territory), and furthermore
limiting to tissue specific areas such as grey matter or white
matter. BOLD fMRI has been used to study the CVR in
healthy volunteers and patients with carotid stenosis and
occlusion and has proved as good as TCD in detecting
impaired CVR in addition to the above mentioned advan-
The aim of this study was to assess the effects of carotid
surgery on haemodynamics in patients with carotid artery
disease using hypercapnia BOLD fMRI as assessed by abso-
lute signal change and the hemispheric asymmetry index
(hAI). Secondly, we aimed to search for potential predictors
of who may benefit most from the heamodynamic
according to a preoperative hAI of CVR.
Recruitment and patients
Patients considered for CEA were prospectively identified
between March 2006 and March 2008 from the hospital
transient ischemic attack (TIA) clinic. Eligibility criteria
using established ultrasound criteria15as used in the
Carotid and Vertebral Artery Transluminal Angioplasty
Study.16The patients had experienced stroke, TIAs or
amaurosis fugax in the previous 12 months. All partici-
pants gave written informed consent, and the study was
approved by the hospital research and development
Exclusion criteria were contraindication for MRI or when
waiting for MRI would have delayed CEA. Diagnostic work-
up and therapeutic management of the patients and its
timing was not affected by this study.
We studied 17 patients with a mean age of 68.2 years
(range, 42e81 years); there were 16 men and 1 woman, this
gender distribution reflects the patients available for
were taking statin medications. One patient had a post-
operative stroke and was therefore excluded from the final
analysis. 8 patients had contralateral carotid artery disease;
Carotid Endarterectomy Improves Cerebrovascular Reserve Capacity547
4 patients with occlusion, 3 patients with 80e89% and 1
patient with 70e79% stenosis. There was no significant
vertebral disease amongst these patients. All subjects were
asked not to consume alcohol or caffeine for 4 h before the
Magnetic resonance imaging (MRI) studies were performed
Netherlands). A standard 8 channel head coil was used for
imaging, deploying a standard gradient echo echoplanar
sequence (Repetition time (TR) ms/echo time ms 3500/60,
flip angle 90 degree, matrix size 64? 64, field of view
192 mm with 33 slices, 3 mm thick and no gap). A total of
160 volumes (approx 9 min acquisition time) were acquired
for each experiment. Preoperative MRI scanning was per-
formed within 30 days of the procedure (Median 6.5 days),
apart from 1 patient who had been scanned 5 months prior
to CEA. Post operative MRI scanning was performed within 2
months of the carotid endarterectomy. Additionally at
postoperative MRI scanning patients had DWI to assess for
postoperative evidence of ischaemia.
ETCO2 was controlled using a standard non-rebreathing
anaesthetic circuit using two one way valves to prevent
rebreathing. Subjects breathed through a comfortably
secure standard anaesthetic mask to ensure a closed circuit.
ETCO2was continuously monitored via a sampling tube at the
mouthpiece level, with recordings being made during the
whole experiment. The two periods of hypercapnia were
intercalated with three periods of normocapnia. The %CO2
was varied around 10% aiming for an increase in ETCO2
between 7 and 8 mmHg change. Control of CO2gas flow was
via an external source outside of the MRI scanner. We
continuously monitored the subjects blood pressure, pulse
(Invivo, Siemens, Malvern, US).
All CEAs were performed as per the local standard of care in
conscious patients with locoregional anesthetic techniques
in a tertiary referral center. Three consultant vascular
surgeons performed the procedure and the center specific
perioperative stroke rate is <3%17A shunt was used if there
was any loss of consciousness or focal neurological deficit in
the patient. Dacron patch repairs were performed in all
patients. The surgeons were not made aware of the
preoperative MRI findings.
The first level fMRI data analysis was carried out using FEAT
v5.63 software (FMRI Expert Analysis Tool (FEAT), version
5.63, Oxford Centre for Functional Magnetic Resonance
part of the FMRIB software library v3.3 (FSL).18FEAT
deployed standard preprocessing with high-pass filtering,
smoothing and motion correction using MCFLIRT.19During
the FSL analysis theETCO2data acquired during the scanning
session was incorporated into the design matrix of the
General Linear Model (GLM). A Gaussian model was used for
modeling the haemodynamic response (HDR) function. The
FSL analysis method compensated for any lags in our setup
(e.g. sampling delay from patient to capnometer) by allow-
ing for a temporal derivative of the design matrix waveform,
which essentially shifts the waveform in time during the
analysis to enable a better fit of the data and model. FSLs
Brain Extraction Tool (BET) was used to remove non brain
Following this the T1 images were segmented into grey and
white matter using FMRIB Automated Segmentation Tool
(FAST) from FSL.20A manually drawn middle cerebral artery
was then combined with each patients grey matter mask
resulting in individualized grey matter MCA territory masks.
Absolute percent signal change (PSC) was derived from the
GM MCA mask using FSL software FEAT query following
conversion from standard space into functional space. This
method has been previously published.21To normalize the
PSC we regressed the effect of ETCO2 change over the
experiment data using a linear correction method derived
from our normative set.21
To address potential problems from normalization, we
additionally normalized ipsilateral PSC to the contralateral
hemisphere expressed as hemispheric asymmetry index
(hAI). We defined hemispheric asymmetry according to
hAI Z200 ? ((PSC Contralateral hemisphere? PSC Ipsilat-
eral operated hemisphere)/(PSC sum both hemispheres)) as
reported.22e24For the second level subgroup analysis the
group was divided into those patients with CVR lateralized
preserved CVR and those lateralized to the contraleteral
hemisphere (>0) defined as impaired CVR.
Statistical calculations were performed using the Statistical
Package for Social Sciences (SPSS version 15.0; SPSS,
Chicago, Ill) software. Statistical significance was set as
p < 0.05. Data is expressed as mean percentage signal
change ? standard deviation (SD). T tests and U tests were
performed for parametric and nonparametric variable
respectively. A univariate analysis of covariance (ANCOVA)
was performed to look at the effect CEA on CVR (as
assessed by PSC) whilst controlling for age, MAP, contra-
lateral carotid disease, time from symptoms to CEA and
from CEA to post op scan.
All seventeen subjects tolerated the experiment well
without adverse reaction during the course of study to
inhalation of CO2. CVR maps were produced for each
patient, see Fig. 1 for example of pre and post CEA CVR
map. Patient 11 had to be excluded from the group analysis
due to a postoperative stroke. The observed motion as
548S.D. Goode et al.
estimated by FSL was within acceptable limits in all cases,
i.e. less than 3 mm in x, y, z axis. The mean preoperative
ETCO2 change was 7.44 ? 3.7 mmHg and the mean post-
p Z0.667). The mean preoperative MAP was 105.8 ? 18.8
and postop MAP was 105.9? 19.7 (T test p Z 0.984).
8.58 ? 5 mmHg(U
Grey matter CVR
Ipsilateral GM CVR improved significantly following CEA
(ipsilateral preoperative PSC was 2.75 ?0.41 and the mean
postoperative PSC was 3.05 ?0.44, p Z 0.025). The mean
contralateral preoperative PSC was 2.75 ?0.49 and the
mean postoperative PSC was 2.93 ?0.39, this was not
significant (p Z0.14), (See Table 1).
The mean preoperative hAI was ?0.44 ?6.31 which means
that the overall haemodynamic impairment was towards
the operated MCA territory. Following CEA mean post-
operative hAI changed to ?3.9 ? 6.18(p Z0.025), indi-
cating an increase in the ipsilateral CVR relative to the
contralateral side (See Table 2).
Effect of preoperative impaired hAI
To try to assess the effect of preoperative ipsilateral haemo-
dynamic impairment our subgroup analysis looked at the
effect of preoperative hAI. Our patients were split into two
groups defined by their hAI e preserved CVR (hAI<0, nZ7)
with an impaired CVR preoperatively showed a greater
improvement following CEA (pZ0.007, see Table 3).
Effect of high grade ipsilateral stenosis
We divided our patients into two groups; those with ipsilat-
eral low grade stenosis (60e80%) and those patients with
a higher degree of stenosis (>80%). We found that those
patients with a higher degree of stenosis seemed to have
For many years, CVR has been clinically assessed using TCD,
which measures blood flow velocity in the middle cerebral
artery (MCA). Problems with CO2reactivity testing using
TCD include: measuring blood flow velocities rather than
the more physiologically important tissue perfusion, no
information on other vascular territories, no assessment of
CVR due to collateral flow increase, and only 85e90% of
people have a sonable temporal bone window. Currently
established methods of assessing CVR include PET and
SPECT, however these techniques are expensive, time
consuming and due to the nonnegligable radiation exposure
cannot be considered completely non-invasive. More
recently MRI techniques have been used to estimate CVR
non-invasively with and without vasodilatory stimulus.12,14
Hypercapnia BOLD fMRI is a promising tool for the haemo-
dynamic assessment of patients with carotid artery disease,
due to it being non-invasive and because there is no radiation
exposure. It enables not only visual qualitative assessment of
CVR, like previous studies using SPECT,25but also semi-
quantitative assessment of CVR. Our main findings are that
patients with carotid artery disease have a significant
improvement in their ipsilateral CVR as shown by increase in
it has been published pre and post carotid artery stenting
hemispheric asymmetry pre and post CAS and did not find any
significant changes in cerebral perfusion patterns. Their hAI
is improvement in CVR on the operated side as indicated by increased BOLD signal change in the left hemisphere.
Examples of CVR maps for pre and post CEA for patient with left sided 80e95% stenosis. Following recanalisation there
CEA for MCA grey matter territory.
Table 1 shows results for all patients undergoing
P Value Contralateral
N Z 16
P values are given for differences between pre and post PSC.
* indicates significant difference to a level of 0.05.
Results for all patients undergoing CEA.
?0.44 ? 6.31
?3.90 ? 6.19
P values are given for differences between pre and post hAI.
* indicates significant difference to a level of 0.05.
Carotid Endarterectomy Improves Cerebrovascular Reserve Capacity 549
values were similar to our data, however they had a greater
ofacquiringdatawherebythe values arerelying onverylarge
numbers of technicium gamma rays being emitted from the
patient. As opposed to our more physiological of looking at
changing oxyhaemoglobin levels which may be a more accu-
Our findings confirm previously work.6e8,28,29In partic-
ular our results correspond very closely with Haller et al.,29
who looked at CVR pre and postoperatively using a hyper-
capnia BOLD fMRI technique. However, they did not
measureETCO2change following administration of a fixed
amount of CO2. They were therefore unable to comment on
the direct change pre and post op, instead utilising
improvement in ipsilateral to contralateral carotid disease
ratios. Their results showed a significant interhemispheric
difference in pre procedure %SC in ipsilateral vs. contra-
lateral whole MCA territories and no significant difference
in those territories post procedure, thereby concluding that
there had been improvement in the ipsilateral CVR. Our
study differed in a number of ways; we used a more phys-
iologically relevant grey matter specific MCA territory
masks, we collated ETCO2data following CO2stimulation
thus enabling us to directly quantify the degree of CO2
stimulation administered and our statistical analysis uti-
lised further relevant demographic and physiological
parameters. Using theETCO2data from our experiment we
were able to include this into our analysis using a univariate
design and thereby making a direct comparison between
pre and post op PSC. Also of note in our univariate analysis
is that we included age,30gender,31,32and BP33which have
been shown previously to have a profound effect on CVR.
In our group of patients we had only one major complica-
tion with one patient having a postoperative stroke. This
happened on the table immediately following the operation
and the patient was reexplored, his symptoms had resolved
the following day except for a slight numbness in his hand.
Haller et al. also noted patients with new DWI lesions had
worsening of CVR. Our data is in agreement with this.
An important question that we tried to address relates to
patient selection based on individual prediction of haemo-
dynamic improvement. When we looked at those patients
with an ipsilateral impaired CVR towards the operated side
it was this group that improved most following CEA. In
contrast, those patients with a preserved hAI towards the
ipsilateral operated MCA territory preop did not have an
improvementfollowing CEA.This suggeststhat patients with
a preoperative ipsilateral CVR impairment may gain
a greater improvement in their overall haemodynamic
status. These findings confirm previous work by Nielsen
et al.34and Hosodo et al.25who showed that patients with
preoperative hypoperfusion benefit from carotid revascu-
larisation, whereas the haemodynamic benefit in patients
with good preoperative perfusion is only limited.34Further
work is needed to assess whether this functional improve-
ment translates into clinical benefit for patients either by
future stroke reduction or improvement in cognition as
suggested by previous work by Ghogawala et al.35
The main limitation of our study is the inclusion of only
a relatively small number of subjects. The study size was
however sufficient to demonstrate a significant improve-
ment in CVR following CEA. With larger numbers of patients
we may have been able to perform more subgroup analyses.
The second limitation is the variable timings of the pre and
postoperative scans, although by using a univariate analysis
and including time from symptom to preoperative scan and
time for CEA to postoperative scan we removed this con-
founding variable from the analysis. Lastly, there may have
been a variable effect on CVR mediated by the use statins
in our patients. Preliminary studies have shown an effect of
statin treatment on CVR in healthy volunteers and
patients,36,37whereby CVR is improved following initiation
of treatment. In our study patients may have only just
started their statins which may not have been acting at
time of preoperative scanning session, whereas others may
have been on long term treatment.
The need to be able to identify the high-risk patient
undergoing carotid intervention, be it either endarterec-
tomy or angioplasty and stenting is an important topic.
Indeed in patients with high grade carotid artery disease can
be further stratified into those with a high-risk carotid pla-
que38or those with an obvious haemodynamic impairment.
in those patients with a greater ipsilateral haemodynamic
impairment. This is probably mediated by improvement in
patients ability to vasodilate suggesting a return of cerebral
autoregulation and vasodilatory reserve capacity. We would
advocate the inclusion of this simple haemodynamic assess-
ment for patients in the diagnostic work-up for those
undergoing revascularisation procedures. However, more
studies are needed to establish the true clinical relevance of
haemodynamic risk assessment afforded by this technique.
Conflict of Interest/Funding
There are no conflicts of interest.
Special Trustees for Nottingham University Hospitals, Peel
Medical Research Trust and Royal College for Surgeons for
providing 1 year Surgical Research Fellowship.
1 Randomised trial of endarterectomy for recently symptomatic
carotid stenosis: final results of the MRC European Carotid
Surgery Trial (ECST). Lancet 1998;351(9113):1379e87.
preop hAI on CVR change following CEA.
Table showing effect of preserved/ impaired
P values are given for differences between pre and post PSC.
* indicates significant difference to a level of 0.05.
550S.D. Goode et al.
2 Ferguson GG, Eliasziw M, Barr HW, Clagett GP. The North
American Symptomatic Carotid Endarterectomy Trial: surgical
results in 1415 patients. Stroke 1999;30(9):1751e8.
3 Markus H, Cullinane M. Severely impaired cerebrovascular
reactivity predicts stroke and TIA risk in patients with carotid
artery stenosis and occlusion. Brain 2001;124(Pt 3):457e67.
4 Nemoto EM, Yonas H, Kuwabara H, Pindzola RR, Sashin D,
Meltzer TC, et al. Identification of hemodynamic compromise
by cerebrovascular reserve and oxygen extraction fraction in
occlusive vascular disease. Journal of Cerebral Blood Flow and
5 Sedlaczek O, Caplan L, Hennerici M. Impaired washout-embo-
lism and ischemic stroke: further examples and proof of
concept. Cerebrovascular Diseases 2005;19(6):396e401.
reserve capacity following carotid endarterectomy. European
Journal of Vascular Endovascular Surgery 1996;11(1):83e9.
7 Cikrit DF, Dalsing MC, Harting PS, Burt RW, Lalka SG,
Sawchuk AP, et al. Cerebral vascular reactivity assessed with
acetazolamide single photon emission computer tomography
scans before and after carotid endarterectomy. American
Journal of Surgery 1997;174(2):193e7.
8 D’Angelo V, Catapano G, Bozzini V, Catapano D, De Vivo P,
Ciritella P, et al. Cerebrovascular reactivity before and after
carotid endarterectomy. Surgical Neurology 1999;51(3):321e6.
9 Kawaguchi S, Sakaki T, Uranishi R. Effects of bypass on CO2
diseases Based on the intra-operative LCBF and CO2 cerebro-
vascular reactivity studies. Acta Neurochirurgica 1999;141(4):
10 Lythgoe DJ, Williams SC, Cullinane M, Markus HS. Mapping of
cerebrovascular reactivity using BOLD magnetic resonance
imaging. Magnetic Resonance Imaging 1999;17(4):495e502.
11 Moonen CT, editor. Functional MRI. Berlin: Springer; 2000.
12 Van der Zande FH, Hofman PA, Backes WH. Mapping hyper-
capnia-induced cerebrovascular reactivity using BOLD MRI.
13 Vesely A, Sasano H, Volgyesi G, Somogyi R, Tesler J, Fedorko L,
et al. MRI mapping of cerebrovascular reactivity using square
wave changes in end-tidal PCO2. Magnetic Resonance in Medi-
14 Ziyeh S, Rick J, Reinhard M, Hetzel A, Mader I, Speck O. Blood
oxygen level-dependent MRI of cerebral CO2 reactivity in severe
carotid stenosis and occlusion. Stroke 2005;36(4):751e6.
15 Sidhu PS, Allan PL. Ultrasound assessment of internal carotid
artery stenosis. Clincal Radiology 1997;52(9):654e8.
16 Endovascular versus surgical treatment in patients with carotid
stenosis in the Carotid and Vertebral Artery Transluminal
Angioplasty Study (CAVATAS): a randomised trial. Lancet 2001;
17 Wadhwa K, Oluwale A, Altaf N, Goode S, MacSweeney S,
Tennant W. Local anaesthesia versus general anaesthesia for
carotid endarterectomy: the Nottingham experience. Cerebro-
vascular Diseases 2006;21(S4):151.
18 SmithSM,Jenkinson M,
Behrens TEJ, Johansen-Berg H. Advances in functional and
structural MR image analysis and implementation as FSL. In:
Conference on mathematics in brain imaging. Los Angeles, CA:
Academic Press Inc. Elsevier Science; 2004.
19 Jenkinson M, Bannister P, Brady M, Smith S. Improved optimiza-
tion for the robust and accurate linear registration and motion
correction of brain images. NeuroImage 2002;17(2):825e41.
20 Zhang Y, Brady M, Smith S. Segmentation of brain MR images
through a hidden Markov random field model and the expecta-
tion-maximization algorithm. IEEE Transactions on Medical
in ischaemic cerebrovascular
21 Goode SD, Krishan S, Alexakis C, Mahajan R, Auer DP. Precision
of cerebrovascular reactivity assessment with use of different
quantification methods for hypercapnia functional MR imaging.
American Journal of Neuroradiology 2009;30(5):972e7.
22 Baciu M, Juphard A, Cousin E, Bas JF. Evaluating fMRI methods
for assessing hemispheric language dominance in healthy
subjects. European Journal of Radiology 2005;55(2):209e18.
23 De Boorder MJ, van der Grond J, van Dongen AJ, Klijn CJ,
Jaap Kapelle L, Van Rijk PP, et al. Spect measurements of
regional cerebral perfusion and carbondioxide reactivity:
artery occlusive disease. Journal of Neurology 2006;253(10):
24 Kaminogo M, Ochi M, Onizuka M, Takahata H, Shibata S. An
additional monitoring of regional cerebral oxygen saturation to
HMPAO SPECT study during balloon test occlusion. Stroke 1999;
25 Hosoda K, Fujita S, Kawaguchi T, Shose Y, Shibata Y, Tamaki N.
Influence of degree of carotid artery stenosis and collateral
pathways and effect of carotid endarterectomy on cerebral
vasoreactivity. Neurosurgery 1998;42(5):988e94.
26 Sfyroeras GS, Arsos G, Karkos CD, Liasidis C, Spyridis C,
Boundas D, et al. Interhemispheric asymmetry in brain per-
fusion before and after carotid stenting: a 99mTc-HMPAO
SPECT study. Journal of Endovascular Therapy 2006;13(6):
28 Bishop CC, Butler L, Hunt T, Burnand KG. Effect of carotid
endarterectomy on cerebral blood flow and its response to
hypercapnia. British Journal of Surgery 1987;74(11):994e6.
29 Haller S, Bonati LH, Rick J, Klarhofer M. Reduced cerebrovas-
cular reserve at CO2 BOLD MR imaging is associated with
increased risk of periinterventional ischemic lesions during
results. Radiology 2008;249(1):251e8.
30 Kastrup A, Dichgans J, Niemeier M, Schabet M. Changes of
cerebrovascular CO2 reactivity during normal aging. Stroke
31 Kastrup A, Thomas C, Hartmann C, Schabet M. Sex dependency
of cerebrovascular CO2 reactivity in normal subjects. Stroke
32 Matteis MTE, Monaldo BC, Caltagirone C, Silvestrini M. Age and
sex differences in cerebral hemodynamics: a transcranial
Doppler study. Stroke 1998;29:963e7.
33 Kimura Y, Oku N, Kajimoto K, Katoh H, Tanaka MR, Takasawa M,
et al. Diastolic blood pressure influences cerebrovascular reac-
photon emission computed tomography in medically treated
Annals of Nuclear Medicine 2006;20(3):209e15.
34 Nielsen MY, Sillesen HH, Jorgensen LG, Schroeder TV. The
haemodynamic effect of carotid endarterectomy. European
Journal of Vascular Endovascular Surgery 2002;24(1):53e8.
35 Ghogawala Z, Westerveld
outcomes after carotid revascularization: the role of cerebral
emboli and hypoperfusion. Neurosurgery 2008;62(2):385e95.
36 Sander K, Hof U, Poppert H, Conrad B, Sander D. Improved
cerebral vasoreactivity after statin administration in healthy
adults. Journal of Neuroimaging 2005;15(3):266e70.
37 Sterzer P, Meintzschel F, Rosler A, Lanfermann H, Steinmetz H,
Sitzer M. Pravastatin improves cerebral vasomotor reactivity in
patients with subcortical small-vessel disease. Stroke 2001;
38 Moody AR, Murphy RE, Morgan PS, Martel AL. Characterization
of complicated carotid plaque with magnetic resonance direct
thrombus imaging in patients with cerebral ischemia. Circula-
Carotid Endarterectomy Improves Cerebrovascular Reserve Capacity551