Cerebral Neutrophil Recruitment, Histology, and Outcome
in Acute Ischemic Stroke
An Imaging-Based Study
C.J.S. Price, MRCP; D.K. Menon, PhD; A.M. Peters, MD; J.R. Ballinger, PhD; R.W. Barber, MSc;
K.K. Balan, MD; A. Lynch, PhD; J.H. Xuereb, MD, FRCPath; T. Fryer, PhD;
J.V. Guadagno, MRCP; E.A. Warburton, DM
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Background and Purpose—Evidence now exists for a pathogenic role for neutrophils in acute cerebral ischemia. We have
studied the patterns and temporal profile of cerebral neutrophil recruitment to areas of acute ischemic stroke (IS) and
have attempted to correlate this with neurological status and outcome.
Methods—Patients with cortical middle cerebral artery (MCA) IS were recruited within 24 hours of clinical onset.
Neutrophil recruitment was studied using indium-111 (111In) troponolate-labeled neutrophils, planar imaging, and
single-photon emission computed tomography (SPECT). Volume of brain infarction was calculated from concurrent
computed tomography (CT). Hematoxylin and eosin sections were obtained postmortem (n⫽2). Outcome was measured
using Barthel, Rankin, and National Institute of Health Stroke (NIHSS) scales.
Results—Fifteen patients were studied. Significant 111In–neutrophil recruitment to ipsilateral hemisphere, as measured by
asymmetry index (AI), was demonstrated within 24 hours of onset in 9 patients; this response was heterogenous between
patients and on repeated measurement attenuated over time. Histologically, recruitment was confirmed within
intravascular, intramural, and intraparenchymal compartments. Interindividual heterogeneity in neutrophil response did
not correlate with infarct volume or outcome. In an exploratory analysis, neutrophil accumulation appeared to correlate
significantly with infarct expansion (Spearman rho⫽0.66; P⫽0.03, n⫽12).
Conclusions—Neutrophils recruit to areas of ischemic brain within 24 hours of symptom onset. This recruitment attenuates
over time and is confirmed histologically. While neutrophil accumulation may be associated with either the magnitude
or the rate of infarct growth, these results require confirmation in future studies. (Stroke. 2004;35:000-000.)
Key Words: stroke 䡲 ischemia 䡲 neutrophils 䡲 SPECT 䡲 histology
D
espite decreases in mortality, ischemic stroke (IS) remains a leading cause of death and disability,1 and the
need for novel effective therapies remains imperative. Evidence from experimental studies suggests a direct role of
leukocytes in the pathogenesis of ischemic injury. This
evidence goes beyond histological accumulation2 to a more
direct role in ischemic pathophysiology.3 Neutrophils have
been implicated in the development of the “no-reflow”
phenomenon in a primate model,4 whereas selective experimental interventional studies have strengthened the case for a
causative role.5 Despite such studies, this remains a controversial area.6
In the clinical arena, evidence for such processes is less
robust. The failure of clinical trials7,8 has prompted a need for
precise definition of human pathophysiology and this has
partly been addressed by imaging studies. Single-photon
emission computed tomography (SPECT) studies have
localized mixed populations of leukocytes to ischemic
hemisphere using indium-111 (111In) and technetium-99m
(99mTc).9,10 While the use of mixed cells offers limited pathological insights, such studies are confounded because 99mTchexamethylpropylenamine oxime (HMPAO) elutes from leukocytes in hydrophilic-form that may undergo reuptake in the brain
as a result of blood-brain barrier (BBB) disruption.11 Furthermore, they provide little information about acute pathophysiology or outcome. Regions of cerebral infarction do attract
selective populations of 99mTc-HMPAO–labeled leukocytes, but
such studies may offer only limited insights into recovery
measured on a single scale.12 None of these studies has provided
histological evidence to support the imaging data.
Received November 24, 2003; final revision received March 9, 2004; accepted March 25, 2004.
From the University of Cambridge Department of Medicine (C.J.S.P.) and Anaesthesia Section (D.K.M.), Addenbrookes Hospital, Cambridge, UK;
Department of Nuclear Medicine (A.M.P., J.R.B., R.W.B., K.K.B.), and Department of Clinical Neurosciences (E.A.W.), Cambridge University Teaching
Hospitals Trust, Cambridge, UK; Centre for Applied Medical Statistics (A.L.), Department of Pathology (J.H.X.), Wolfson Brain Imaging Centre (T.F.),
and Department of Neurology (J.V.G.), University of Cambridge, UK.
Correspondence to Dr Christopher J.S. Price, MRC Training Fellow in Stroke Medicine, Box 83, R3 Neurosciences, Addenbrookes Hospital,
Cambridge CB22QQ, UK. E-mail [email protected]
© 2004 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org
DOI: 10.1161/01.STR.0000130592.71028.92
1
2
Stroke
TABLE 1.
July 2004
Inclusion and Exclusion Criteria
Inclusion
● Cortical IS attributable to the MCA
● Male or females, aged 50 – 85 y
● Informed consent
● Patients are able to undergo neutrophil labeling and injection within 24
h of onset
● CT has excluded hemorrhage or other intracranial pathology
Exclusion
● Premorbid cognitive impairment
● Previous stroke
● Malignant disease
Figure 1. Scatter plot of asymmetry indices over time in 15
patients.
● Other ongoing organ failure
Imaging
● Myocardial infarction within 2 mo
Planar gamma camera and SPECT imaging were performed on an
Elscint dual-headed gamma camera fitted with medium energy
collimators. The targeted interval between injection of labeled cells
and onset of clinical symptoms for these 3 time points were 24 hours,
4 to 7 days, and 8 to 15 days. Planar images (anterior and posterior)
and SPECT were performed within 24 hours after injection. All
quantitative studies were performed on regions of interest (ROI)
defined manually from the planar hemispheric images (Figure 1C)
and outside the head as a background region. Counts were measured
from ipsilateral and contralateral hemispheres anteriorly and posteriorly, and decay corrected and adjusted for administered dose to
give counts 10 minutes⫺1MBq⫺1. Baseline images were acquired
before each follow-up study when decay-corrected counts were
subtracted. SPECT data were acquired to give a slice thickness of
⬇8 mm. Data were reconstructed into transaxial, coronal, and
sagittal planes using filtered back projection.
For coregistration, X-ray computed tomography (CT) and SPECT
images were downloaded in digital format and coregistered using
mutual information software.17 CT was performed within 24 hours of
follow-up nuclear imaging. Axial images were reconstructed with a
slice thickness of 5 mm. Volumetric analysis of CT determined
infarct volume was performed using manually defined regions of
interest using Analyze (Analyze 5.0; Mayo Clinic Biomedical
Imaging Resource). Differences in infarct volume between the initial
and subsequent studies were calculated from these volumes and
denoted, ⌬CT1–2 and ⌬CT1–3, for differences between the first and
second and first and third studies, respectively.
● Posterior circulation or noncortical stroke
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
● Systemic or intraarterial thrombolysis
In this article, we seek to study quantitatively the spatial
and temporal profiles of cerebral recruitment of 111In-labeled
neutrophils using planar imaging and SPECT. We have
attempted to underpin these imaging studies with postmortem
histology from a small subset of patients and explored the
relationship between neutrophil accumulation, infarct evolution, and clinical outcome.
Subjects and Methods
Patients
After informed consent, patients with a clear time of onset and
ischemic cortical middle cerebral artery (MCA) territory syndromes
fitting appropriate criteria (Table 1) were recruited. The Local
Regional Ethics Committee (LREC) and the Administration of
Radioactive Substances Advisory Committee (ARSAC) of the
United Kingdom approved the study.
Cell Labeling
Separation and labeling of neutrophils was performed with the
assistance of the Nuclear Medicine Department using a standard
protocol as previously described.13 Briefly, venous blood is mixed
with acid citrate dextrose (ACD) (NIH Formula A) before centrifugation and resuspension of cells in cell free plasma (CFP). Neutrophils were separated using an iso-osmotic discontinuous gradient of
Percoll (Pharmacia) and labeled with 111In troponolate. Autologouslabeled neutrophils were returned to the patient within 24 hours and
thereafter at follow-up studies at 4 to 7 and 8 to 15 days. Administered radiation dose aimed for a target of 16 MBq equating to 7.2
mSv effective dose equivalent (ED). Cell recovery was calculated
from a 40-minute venous sample and extrapolated to a blood volume
of 5 L to allow comparison with previous studies. Recovery provides
an indication of the percentage of labeled cells that remain within the
circulation. In follow-up studies, recovery figures were corrected for
residual radioactivity in a presample. Low recovery figure (⬍ 10%)
may indicate impaired cell viability or removal of cells into the
reticuloendothelial system as a result of ex vivo activation. Previous
studies for 111In troponolate human granulocyte cell recovery are
⬇40%.14,15 Labeling efficiency was calculated by measuring radioactivity in labeled cells versus that in washes, expressed as a
percentage of radioactivity incorporated into cells. Labeling efficiency may vary with cell number, and incubation time and figure
from previous studies are ⬇72% for 111In troponolate.16
Histology
Consent was obtained for brain postmortem. Brains were fixed in
buffered 10% formal saline before sectioning. Sections were taken
from peri-infarct areas in transaxial or coronal planes, stained with
hematoxylin and eosin (H&E), and examined by an independent
neuropathologist blinded to the imaging studies.
Statistical Methods
Asymmetry indices (AI) were calculated for each SPECT study as
previously described,12
AI⫽(A⫺B)/(A⫹B),
where A⫽ipsilateral and B⫽contralateral summed count rates. AI values
from the 3 imaging studies were referred to as AI1, AI2, and AI3.
A and B were modeled as being Poisson variables with a common
parameter, Lambda (␭), and the posterior distribution (having observed A and B) of (A⫺B)/(A⫹B) was used to generate a 95%
confidence interval for the AI. A and B were observed for all
patients. If the observation of AI was not within the confidence
interval, then this was deemed significant evidence that the assumption of a common parameter, ie, the assumption of symmetry, did not
hold. This analysis was conducted with WinBUGS version 1.4
(MRC Biostatistics Unit).18 Because of the simplicity of the model,
achieving convergence was not apparently problematic, nor did
varying the previous distribution of ␭ (within reasonable choices)
Price et al
TABLE 2.
Patient N
Imaging Neutrophils in Ischemic Brain
3
Asymmetry Indices, Cell Recovery, and Labeling Efficiency for Each Patient
AI1
Acute
Cell
Recovery
%
Acute
Labeling
Efficiency
%
AI2
0.29*
Day 4 –7
Cell
Recovery
%
Day 4 –7
Labeling
Efficiency
%
Day 8 –15
Cell
Recovery
%
Day 8 –15
Labeling
Efficiency
%
55.5
80
39
71
0.25*
85
88
0.04
33
75
75
0.01
25.5
59
AI3
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
1
0.69*
47
87
2
0.00
29.5
64
3
0.12*
9.5
76
0.13*
27
4
0.10*
32.5
79
0.01
41.5
83
0.03
n/a
92
5
0.17*
24
84
—
—
—
—
—
—
6
0.50*
31
61
0.17*
7
0.26*
50.5
89
—
8
0.25*
68
87
9
0.07
51.5
84
10
0.72*
59
74
—
11
0.07
41.5
76
0.02
12
0.03
45.5
67
⫺0.04
13
0.04
50.5
82
0.05
⫺0.01
41
95
0.01
35.5
67
—
—
—
—
—
0.28*
44.5
75
⫺0.08
0.00
76
76
0.06
—
38.5
70
68
84
—
—
—
—
80
0.01
51
77
41
88
⫺0.02
63.5
78
53
0.11*
33.5
37
39
68
14
0.11
39
86
0.10
33
86
⫺0.02
56
68
15
0.07*
48
85
0.03
42.5
88
0.01
N/A
88
*Significant AIs, P⬍0.05.
Patients 5, 7, and 10 died before follow-up imaging studies. A blood volume of 5 L was assumed for all patients to allow for recovery calculation.
N/A indicates not available.
seriously affect the results. Correlation coefficients presented are
Spearman rank rho (␳) with P values for the 2-tailed test. Comparison of groups was conducted using the Mann–Whitney U test. The
Friedman test was applied as a measure of decrement of AI over time
when AI was measured repeatedly.
Outcome
Clinical assessment on admission, at 4 to 7 days, 8 to 15 days, and
3 months included functional rating on Barthel and Rankin scales
and neurological rating on the National Institute of Health Stroke
Scale (NIHSS). These scales have been validated in large-scale trials
of thrombolysis.19 Scales were measured throughout the study by a
single physician (C.J.S.P.). When subjects died before follow-up
assessments, they were allocated maximal deficit Barthel, Rankin,
and NIHSS scores of 0, 5, and 36, respectively. For the purposes of
correlation, corresponding indices were plotted against each scale at
each time point. Patients were divided into groups using 3 stratification approaches to address separate questions. In the first instance,
patients were classified into 2 groups (group 1 and group 2), based
on whether the AI was significantly abnormal, to assess the relationship of hemispheric neutrophil localization to clinical presentation.
In the second instance, patients were classified into 2 groups (group
A and group B), based on their stroke severity as either mild/
moderate (NIHSS ⱕ15) or severe (NIHSS ⬎15).20 This stratification
allowed us to assess the relationship of clinical stroke severity to
hemispheric neutrophil recruitment. Finally, patients were stratified
into 2 groups (group X and group Y) depending on clinical
progression of deficit: those with a reduction in NIHSS scores over
the first 3 months (improving) or those with an increase in NIHSS
over the first 3 months (deteriorating). This stratification allowed us
to analyze whether patients who exhibited clinical deterioration
showed greater hemispheric neutrophil localization. Finally, we
undertook an exploratory analysis to determine whether the degree of
neutrophil recruitment to the damaged hemisphere was related to the
extent and speed of infarct expansion (as defined by ⌬CT1–2 and
⌬CT1–3).
Results
Imaging
Fifteen patients were entered into the study. Administered
activity for each study ranged from 7.5 to 18.5 MBq with a
calculated radiation burden range of 3.4 to 8.3 mSv ED. All
patients were imaged with CT within 24 hours of symptom
onset, except patient 10 who was imaged with magnetic
resonance imaging and died after the first 111In SPECT study.
Mean time from clinical onset (and ranges) to respective
SPECT scanning sessions were 38.7 hours (23 to 49), 5.9
days (4 to 7), and 12.1 days (10 to 14), and corresponding
intervals for CT were 10.9 hours (2 to 24), 4.9 days (3 to 7),
and 11 days (8 to 12). Primary cerebral hemorrhage was
excluded in all subjects. Quality of images varied according
to movement and the clinical state of the patient.
Neutrophils were separated, labeled with 111In, and injected
into all patients initially. Cell recovery, labeling efficiency
data, and AIs derived from a total of 40 studies are given in
Table 2. Assuming a blood volume of 5 L, our figures are
comparable to those in established literature.14 AIs showed a
trend toward decrement over the time period (Figure 1); the
Friedman test for the 12 patients with repeated measures
gives P⫽0.038, with the mean rank monotonically decreasing
with time point. Significant AI1s were seen in 9 patients
(range: 0.07 to 0.72) (Figures 2 and 3 and Table 2). Volumetric CT data did not correlate with AI across all time points
(rho⫽0.11, P⫽0.53, n⫽35).
Histology
Consent for histopathological study was obtained for patients
10 and 7 who died at 3 and 11 days after ictus from brain stem
herniation and cardiac failure, respectively.
4
Stroke
July 2004
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Figure 2. Patient 1, session 1. Asymmetry index⫽0.69. A and B,
Paired radiological format noncontrast CT (16 hours) and corresponding transaxial SPECT (42 hours). C, Manually defined
hemispheric region of interest (ROI), planar image. D and E,
Anterior and posterior planar images at 46 hours after clinical
onset. F, Coregistered image of noncontrast CT (16 hours after
clinical onset) and transaxial SPECT (42 hours after clinical
onset) of patient 1 viewed caudo-rostrally from rostral (top) to
caudal (bottom).
For patient 10 (Figure 4A through 4C), a macroscopic
examination of a coronal slice revealed left hemisphere
infarction throughout the entire territory of the MCA cerebral
artery with involvement of subcortical structures. Microscopically, a brisk inflammatory infiltrate was noted in both
meninges and superficial cortex in peri-infarct zones with
Figure 4. A to C, Patient 10, who died on day 3. AI⫽0.72; coronal peri-infarct hematoxylin and eosin sections. A, ⫻10, dense
meningeal inflammatory infiltrate adjacent to blood vessel. B,
⫻40, infiltrate in Virchow–Robin spaces. C, ⫻40, intraparenchymal inflammatory infiltrate. D to F, Patient 7, who died on day
11. AI⫽0.26; transaxial peri-infarct hematoxylin and eosin sections. D, ⫻40, intramural infiltrate in blood vessel in peri-infarct
zone showing intravascular red cells. E, Dilated vessel with
intraluminal accumulation of leukocytes. F, ⫻40, marginating
granulocytes within blood vessel adjacent to infarct.
evidence of perivascular neutrophil aggregation in underlying
white matter. Macrophages were also present in this cellular
reaction.
For patient 7 (Figure 4D through 4F), macroscopic examination of coronal sections revealed infarction of the right
MCA cortex with involvement of subcortical structures.
Microscopic examination revealed cellular infiltrate, consisting of neutrophils and macrophages in the cerebral cortex
extending into the Virchow–Robin spaces.
Clinical Progression and Outcome
AI did not correlate with any outcome measure at any time
point. Although high AIs were associated with worse neurological scores, these were not predictive. No significant
differences in NIHSS were detected between groups 1 and 2
(P⫽0.5) or in AI between groups A and B or X and Y,
(P⫽0.5 and 0.4, respectively). The ⌬CT1–2 correlated significantly with AI1 (rho⫽0.66; P⫽0.03) but not with AI2 or AI3.
There were no significant relationships detected between
⌬CT1–3 and AI at any time point. The ⌬CT1–2 in groups 1 and
2 did not differ significantly (P⫽0.26).
Discussion
Figure 3. Patient 6. Planar anterior and posterior images at 44
hours and 6 and 13 days after symptom onset (A, B, and C,
AI⫽0.5, 0.17, and 0.01, respectively).
Using robust, quality-controlled methods, we have demonstrated that within the first 24 hours of IS, autologous
neutrophils may be separated, labeled with 111In troponolate,
and used for quantitative and sequential studies of ipsilateral
cerebral hemispheric recruitment. This is the first study to our
knowledge to use histological confirmation in this context.
The labeling process does not activate neutrophils and hence
provides a measure of in vivo recruitment, presumably
triggered by ischemia. Neutrophils appear early in regions of
ischemic cerebral hemisphere defined structurally by coregistration of SPECT images and CT, and may be present from
as early as 19 hours. Consistent with previous studies,12
recruitment shows a trend toward ipsilateral attenuation over
time. Late or persisting neutrophil recruitment could not be
Price et al
attributed to hemorrhagic transformation. AI1 values appear
to be associated with the rate at which stroke volume expands
as defined by CT. However, the extent of eventual (presumably penumbral) brain recruitment into the stroke was unrelated to AI in our study. Further AI values did not discriminate between patients with different clinical progression
patterns or outcome.
Methodological Issues
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Although previous studies have addressed leukocyte recruitment after stroke, these results are confounded by the use of
mixed cells or 99mTc-based techniques, because label elution
is a significant problem. In contrast to earlier studies, we can
be confident that ipsilateral cerebral signal relates not to
eluted complexes, but specifically to the neutrophil component of the inflammatory response. 99mTc-HMPAO decomposition products are hydrophilic and are unable to cross an
intact BBB. Where BBB is disrupted, as is likely in IS
patients, 99mTc-HMPAO–labeled leukocyte studies remain
uninterpretable. The imaging data do not discriminate between intravascular and intraparenchymal neutrophils when
histology remains key. Cell recovery and labeling efficiency
were comparable to optimal figures in the literature, suggesting that variations in AI were unlikely to be caused by
deficiencies in labeling methodology.
This study used validated, neurological, and functional
stroke scales, shown to be reproducible in larger trials; this
contrasts with the Mathew scale used in one study.12 The
Mathew scale is dominated by consciousness, combines
impairments and disabilities, and its validity and reliability
have not been proven.21
Pathophysiological Implications
In our small sample of patients, a statistically significant
relationship between AI and outcome could not be established. In a larger sample, AI may predict outcome but this
remains speculative. Given the available data, the contribution of neutrophils to pathophysiology also remains speculative. Instead, neutrophils may act as biological markers of
disease. A pathophysiological role of neutrophils in clinical
stroke may only be put beyond doubt within the context of
clinical interventional studies or by satisfying more stringent
etiological criteria.6 Such criteria, as set out in this article, are
not met in our study, and the results of such interventions
have, to date, been disappointing.22,23 Significant AI values
were observed at later time points, but only in those patients
with higher values of AI initially. Our data are consistent with
animal models in which polymorphonuclear leukocytes
(PMNL) were observed histologically 24 hours after MCA
occlusion in rats.24 The data presented further serve to
illustrate the heterogeneity of cellular inflammatory pathophysiological responses to IS.
Insignificant AI values could not be attributed to variations in cell recovery and labeling efficiency. However, a
number of possibilities could account for these negative
results. First, we did not know if patients spontaneously
recanalized. Complete and persistent MCA occlusion,
accompanied by a concurrent lack of collateral circulation,
may have prohibited neutrophil recruitment to densely
Imaging Neutrophils in Ischemic Brain
5
ischemic areas. Hence, it is possible that significant
neutrophil recruitment may be not only dependent on the
ischemic process per se but also a marker of reperfusion
injury. This is consistent with experimental models in
which temporary MCA occlusion results in more a rapid
and dense pattern of cerebral leukocyte recruitment.25
Opportunities for assessment of recanalization are provided by modern magnetic resonance angiography techniques or transcranial Doppler. This type of data was not
available in this study. Second, an early time window of
neutrophil recruitment to ischemic brain may be being
missed. Granulocytes have been documented as early as 6
hours after experimental ischemia,26 whereas clinical studies suggest 15 hours as the earliest time point at which
cerebral neutrophils appear.27 Signal derived from contralateral hemisphere may itself represent pathological
invasion and hence may not provide an adequate control
ROI for comparison. In contrast to other studies, in which
many ROIs from ipsilateral hemisphere were selected (and
in which the corresponding control ROIs are not explicitly
defined), we have tested activity on a hemispheric basis as
defined on planar images, with individual AI values tested
statistically for significance. Given the spatial resolution
of SPECT, it is currently not possible to precisely coregister areas of maximal neutrophil density with areas of
infarction as defined by CT. The use of appropriate
fiducial markers would improve anatomical localization
within this context. Furthermore, we can only speculate as
to whether such signal relates to ischemic penumbral areas,
ie, areas that may represent a neuroprotective target or to
core infarction. The limited histological evidence that we
have would suggest that the former was indeed the case.
The 111In neutrophil SPECT provides evidence of acute
neutrophil localization to infarcted hemispheres after IS.
Extent of hemispheric neutrophil recruitment appeared
unrelated to stroke severity at onset, clinical progression,
or outcome, but may be related to early infarct expansion.
Whether such findings represent a causal relationship
remains to be established. The 111In neutrophil SPECT may
be a useful tool for the initial assessment of the efficacy of
antiinflammatory interventions aimed at reducing
neutrophil-mediated injury in stroke. Failure to improve
clinical outcomes with such interventions may be because
of the fact that the agent is not effective in preventing
neutrophil recruitment at the time points or dosages used.
However, the agent may be efficacious in reducing neutrophil recruitment but fail to improve outcome because
these cells may not cause pathology in IS. Labeled
leukocyte studies may be useful in making the important
distinction between these two scenarios.
Acknowledgments
C.J.S.P. and this work is funded by the Medical Research Council
and the Sackler Fund. A.G.L. is supported by a grant from the NHS
(RCC57117). D.K.M. is supported by a grant from the MRC (grant
G9439390 ID 56833). E.A.W. is supported by a PPP Mid Career
Fellowship and by the Stroke Association.
References
1. Warlow CP. Epidemiology of stroke. Lancet. 1998;352:SIII1—SIII4.
6
Stroke
July 2004
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
2. Garcia JH, Kamijyo Y. Cerebral infarction. Evolution of histopathological changes after occlusion of a middle cerebral artery in primates.
J Neuropathol Exp Neurol. 1974;33:408 – 421.
3. Hallenbeck JM. Polymorphonuclear leukocyte accumulation in brain
regions with low blood flow during the early postischemic period. Stroke.
1986;17:246 –253.
4. del Zoppo GJ, Schmid-Schonbein GW, Mori E, Copeland BR, Chang
CM. Polymorphonuclear leukocytes occlude capillaries following middle
cerebral artery occlusion and reperfusion in baboons. Stroke. 1991;22:
1276 –1283.
5. Dawson DA, Ruetzler CA, Carlos TM, Kochanek PM, Hallenbeck JM.
Polymorphonuclear leukocytes and microcirculatory perfusion in acute
stroke in the SHR. Keio J Med. 1996;45:248 –253.
6. Emerich DF, Dean RL 3rd, Bartus RT. The role of leukocytes following
cerebral ischemia: pathogenic variable or bystander reaction to emerging
infarct? Exp Neurol. 2002;173:168 –181.
7. Investigators EAST. Enlimomab Acute Stroke Trial: final results. Neurology. 1997;48:A270.
8. Directory ST. ASTIN: acute stroke therapy by inhibition of neutrophils.
Available at: www.strokecenter.org/trials and www.corvas.com/
press_releases. Accessed 2003.
9. Pozzilli C, Lenzi GL, Argentino C, Carolei A, Rasura M, Signore A,
Bozzao L, Pozzilli P. Imaging of leukocytic infiltration in human cerebral
infarcts. Stroke. 1985;16:251–255.
10. Wang PY, Kao CH, Mui MY, Wang SJ. Leukocyte infiltration in acute
hemispheric ischemic stroke [see comments]. Stroke. 1993;24:236 –240.
11. Peters AM, Lavender JP. Imaging inflammation with radiolabeled white
cells: 99Tcm-HMPAO or 111In? [editorial]. Nucl Med Commun. 1991;
12:923–925.
12. Akopov SE, Simonian NA, Grigorian GS. Dynamics of polymorphonuclear leukocyte accumulation in acute cerebral infarction and
their correlation with brain tissue damage. Stroke. 1996;27:1739 –1743.
13. Danpure HJ, Osman S. Optimum conditions for radiolabeling human
granulocytes and mixed leucocytes with 111In-tropolonate. Eur J Nucl
Med. 1988;13:537–542.
14. Peters AM, Roddie ME, Danpure HJ, Osman S, Zacharopoulos GP,
George P, Stuttle AW, Lavender JP. 99Tcm-HMPAO labeled leucocytes:
comparison with 111In-tropolonate labeled granulocytes. Nucl Med
Commun. 1988;9:449 – 463.
15. Saverymuttu SH, Peters AM, Keshavarzian A, Reavy HJ, Lavender JP.
The kinetics of 111indium distribution following injection of 111indium
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
labeled autologous granulocytes in man. Br J Haematol. 1985;61:
675– 685.
Schauwecker DS, Burt RW, Park HM, Mock BH, Tobolski MM, Yu PL,
Wellman HN. Comparison of purified indium-111 granulocytes and
indium-111 mixed leukocytes for imaging of infections. J Nucl Med.
1988;29:23–25.
Studholme C, Hill DL, Hawkes DJ. Automated 3-D registration of MR
and CT images of the head. Med Image Anal. 1996;1:163–175.
Spiegelhalter DJ. WinBUGS Version 1.2 User Manual. MRC Biostatistics
Unit, 1999.
Group NINDS. Tissue plasminogen activator for acute ischemic stroke.
The National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. N Engl J Med. 1995;333:1581–1587.
Allder SJ, Moody AR, Martel AL, Morgan PS, Delay GS, Gladman JR,
Lennox GE. Differences in the diagnostic accuracy of acute stroke
clinical subtypes defined by multimodal magnetic resonance imaging.
J Neurol Neurosurg Psychiatry. 2003;74:886 – 888.
Wade DT. Measurement in neurological rehabilitation. Curr. Oxford
Medical Publications. 1992; 292–293.
Investigators EAST. Use of anti-ICAM-1 therapy in ischemic stroke:
results of the Enlimomab Acute Stroke Trial. Neurology. 2001;57:
1428 –1434.
Lees KR, Diener HC, Asplund K, Krams M, UK-279,276 –301 Study
Investigators. UK-279,276, a neutrophil inhibitory glycoprotein, in acute
stroke. Tolerability and pharmacokinetics. Stroke. 2003;34:1704 –1709.
Barone FC, Hillegass LM, Price WJ, White RF, Lee EV, Feuerstein GZ,
Sarau HM, Clark RK, Griswold DE. Polymorphonuclear leukocyte infiltration into cerebral focal ischemic tissue: myeloperoxidase activity assay
and histologic verification. J Neurosci Res. 1991;29:336 –345.
Clark RK, Lee EV, White RF, Jonak ZL, Feuerstein GZ, Barone FC.
Reperfusion following focal stroke hastens inflammation and resolution
of ischemic injured tissue. Brain Res Bull. 1994;35:387–392.
Barone FC, Hillegass LM, Tzimas MN, Schmidt DB, Foley JJ, White RF,
Price WJ, Feuerstein GZ, Clark RK, Griswold DE. Time-related changes
in myeloperoxidase activity and leukotriene B4 receptor binding reflect
leukocyte influx in cerebral focal stroke. Mol Chem Neuropathol. 1995;
24:13–30.
Lindsberg PJ, Carpen O, Paetau A, Karjalainen-Lindsberg ML, Kaste M.
Endothelial ICAM-1 expression associated with inflammatory cell
response in human ischemic stroke. Circulation. 1996;94:939 –945.
Cerebral Neutrophil Recruitment, Histology, and Outcome in Acute Ischemic Stroke. An
Imaging-Based Study
C. J.S. Price, D. K. Menon, A. M. Peters, J. R. Ballinger, R. W. Barber, K. K. Balan, A. Lynch,
J. H. Xuereb, T. Fryer, J. V. Guadagno and E. A. Warburton
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Stroke. published online May 20, 2004;
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2004 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/early/2004/05/20/01.STR.0000130592.71028.92.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/