All micrographs contain Movat pentachromic stain. A, A carotid plaque
from a patient affected by ipsilateral major stroke who underwent carotid
endarterectomy within 2 months of symptom onset. Fibrous cap plaque rupture
with intraluminal thrombus is evident (magnification ×2). B, Carotid
plaque from patient with transient ischemic attack, characterized by the presence
of an acute thrombus associated with a superficial erosion (arrowheads) (magnification
×10). C, Carotid plaque from an asymptomatic patient, characterized
by a large lipid core and a thin fibrous cap, without the presence of acute
or organized thrombus within the lumen (magnification ×2). D, An organized
thrombus characterized by stratified fibrous tissue, associated with typical
angiomatosis, with a network of large thin-walled vascular channels (arrowheads)
and a variable number of macrophagic cells loaded with hemosiderin, in a ruptured
carotid plaque of a patient affected by ipsilateral major stroke operated
within 12 months of symptom onset (magnification ×2).
All antibodies were used according to the avidin-biotin-peroxidase complex
methods with diaminobenzidine as final chromogen and hematoxylin as counterstain.
A, Fibrous cap adjacent to rupture site containing numerous macrophage-foam
cells positive to CD68 antibody (brown chromogen) (magnification ×10).
B, A high expression of interleukin 6 (brown chromogen) observed in the macrophage
cells present in the cap region of the carotid plaque (magnification ×10).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Spagnoli LG, Mauriello A, Sangiorgi G, et al. Extracranial Thrombotically Active Carotid Plaque as a Risk Factor for Ischemic Stroke. JAMA. 2004;292(15):1845–1852. doi:10.1001/jama.292.15.1845
Author Affiliations: Anatomic Pathology (Drs
Spagnoli, Mauriello, Fratoni, and Bonanno), Department of Cardiovascular Diseases
(Dr Sangiorgi), Division of Vascular Surgery (Drs Pistolese and Ippoliti),
University of Rome Tor Vergata, Rome, Italy; Minnesota Cardiovascular Research
Institute, Minneapolis (Dr Schwartz); Departments of Neurologic Surgery (Dr
Piepgras) and Internal Medicine and Cardiovascular Diseases (Dr Holmes), Mayo
Clinic and Mayo Foundation, Rochester, Minn.
Context Recent studies suggest that factors other than the degree of carotid
stenosis are involved in ischemic stroke pathogenesis, especially modifications
of plaque composition and related complications.
Objective To examine the role of carotid plaque rupture and thrombosis in ischemic
stroke pathogenesis in patients undergoing carotid endarterectomy, excluding
those with possible cardiac embolization or with severe stenosis of the circle
Design, Setting, and Patients A total of 269 carotid plaques selected from an Interinstitutional Carotid
Tissue Bank were studied by histology after surgical endarterectomy between
January 1995 and December 2002. A total of 96 plaques were from patients with
ipsilateral major stroke, 91 plaques from patients with transient ischemic
attack (TIA), and 82 plaques from patients without symptoms.
Main Outcome Measures Differences in the frequency of thrombosis, cap rupture, cap erosion,
inflammatory infiltrate, and major cardiovascular risk factors between study
Results A thrombotically active carotid plaque associated with high inflammatory
infiltrate was observed in 71 (74.0%) of 96 patients with ipsilateral major
stroke (and in all 32 plaques from patients operated within 2 months of symptom
onset) compared with 32 (35.2%) of 91 patients with TIA (P < .001) or 12 (14.6%) of 82 patients who were without
symptoms (P < .001). In addition, a
fresh thrombus was observed in 53.8% of patients with stroke operated 13 to
24 months after the cerebrovascular event. An acute thrombus was associated
with cap rupture in 64 (90.1%) of 71 thrombosed plaques from patients with
stroke and with cap erosion in the remaining 7 cases (9.9%). Ruptured plaques
of patients affected by stroke were characterized by the presence of a more
severe inflammatory infiltrate, constituted by monocytes, macrophages, and
T lymphocyte cells compared with that observed in the TIA and asymptomatic
groups (P = .001). There was no significant
difference between groups in major cardiovascular risk factors.
Conclusion These results demonstrate a major role of carotid thrombosis and inflammation
in ischemic stroke in patients affected by carotid atherosclerotic disease.
Patients with substantial carotid artery narrowing are at increased
risk for major stroke,1-6 but
the pathogenic mechanisms linking carotid atherosclerosis and ischemic brain
injury still need to be fully clarified.Clinical trials designed to evaluate
the beneficial effects of endarterectomy in symptomatic and asymptomatic patients
have focused on carotid stenosis severity and plaque ulceration as risk factors
for cerebrovascular events. The results of the European Carotid Surgery Trial2,3 and the North American Symptomatic
Carotid Endarterectomy Trial(NASCET)4,5 suggest
that surgery is indicated for stroke prevention when stenosis is more than
70%. TheNASCET study showed that among symptomatic patients with high-grade
carotid stenosis (>70%) not treated by surgery, the 2-year rate of any ipsilateral
stroke was 26.0% for major stroke and 13.1% for fatal stroke.5 The
Asymptomatic Carotid Atherosclerosis Study7 suggested
that indications for carotid endarterectomy could be broadened to include
asymptomatic patients with carotid stenosis of more than 60% and that the
aggregate risk over 5 years for ipsilateral stroke and any perioperative stroke
was estimated at 11.0% for asymptomatic patients treated medically.
The potential for accurately predicting ipsilateral stroke risk only
on the basis of carotid stenosis severity remains uncertain.1,8 Inzitari
et al8 found that patients with a total occlusion
of a carotid artery have a reduction in 5-year risk for ipsilateral stroke
in comparison with patients with severe stenosis. Moreover, the NASCET study
showed that the presence of an angiographically evident carotid plaque ulceration
in symptomatic patients with a high degree of stenosis increases the relative
risk for stroke.9 In addition, conclusive data
are still lacking regarding the possible benefit of carotid endarterectomy
in symptomatic patients with ulcerated lesions and a less severe degree of
The severity of carotid stenosis is no longer sufficient to identify
patients at high risk to develop an acute cerebrovascular event. Identification
of other factors may improve risk stratification and help to decrease the
likelihood of disabling stroke in the carotid territory of the brain. A major
future challenge is to identify high risk before the clinical event develops.
Plaque structure could be an independent risk factor for ischemic stroke.1,9,11-13 Atherosclerotic
plaque composition and related complications may represent an important determinant
of increased stroke risk as suggested by different pathologic studies, although
the reported results are contradictory.14-18 This
study examines the role of carotid plaque rupture and thrombosis in the pathogenesis
of acute cerebrovascular events.
The Interinstitutional Carotid Tissue Bank (ICTB) was created to collect
carotid specimens from patients affected by disabling ischemic stroke and
from patients with transient ischemic attack (TIA) and patients who were asymptomatic
who underwent carotid endarterectomy at the Mayo Clinic and Mayo Foundation,
Rochester, Minn, and the University of Rome Tor Vergata, Rome, Italy. The
ICTB is located at the University of Rome Tor Vergata, while the database
is available for both institutions. The electronic database was created to
collect all patients undergoing surgical endarterectomy at both institutions
and records all clinical variables, risk factors profile, pharmacological
treatment, assessment of cardiovascular pathology, and Duplex scan examination
of epiaortic vessels. This database also includes the results of brain computed
tomographic scan examination and the result of selective carotid and cerebral
The database included 351 patients who underwent carotid endarterectomy
between January 1995 and December 2002. A total of 4 plaques, removed by surgeons
in various fragments, were excluded to avoid gross artifacts at the histomorphometric
examination. Seventy-eight patients did not match the inclusion criteria for
the study and were excluded. Thus, a total of 269 cases formed the study population
(187 collected at Mayo Clinic and 82 at the University of Rome) (Table 1). Plaques were divided into 3 groups:
96 from patients with ipsilateral major stroke, 91 from patients with TIA,
and 82 from patients without symptoms (control), who underwent carotid endarterectomy
for asymptomatic high-grade stenosis. None of the patients enrolled underwent
bilateral endarterectomy. The data of 43 plaques from 91 patients from the
second group have been reported in a previous article.18
Major stroke was defined as a clinical syndrome characterized by rapidly
developing focal or at times global symptoms without significant clinical
improvement within 7 days in the distribution of symptomatic carotid artery,
not hemorrhagic and with no cause other than vascular origin, assessed by
brain computed tomographic study as a cortical or deep white matter or basal
ganglia lesion of more than 1 cm. Transient ischemic attack was defined as
recent (<120 days before surgery) occurrence of any sudden focal neurological
deficit that cleared completely within 24 hours, without previous stroke.
Asymptomatic patients never developed neurological symptoms or cerebral lesions
assessed by computed tomography. Angiographic carotid stenosis was measured
in both institutions using the method from theNASCET trial5 by
2 independent physicians; interobserver and intraobserver reliability was
more than 95%.
Patients undergoing carotid endarterectomy at both institutions were
excluded from the ICTB if they had a probable cardiac embolization source
(rhythm disorders, mitral valve stenosis, prolapse or calcification, mechanical
cardiac valves, recent myocardial infarction, left ventricular thrombus, atrial
myxoma, endocarditis, dilated cardiomyopathy, patent foramen ovale), because
the cerebrovascular event might then be caused from cardiac and not carotid
emboli; symptoms that could be attributed to nonatherosclerotic disease (aneurysm,
fibromuscular dysplasia); or stenosis of more than 50% of the circle of Willis.
The study was approved by the institutional review boards of the Mayo
Clinic and Mayo Foundation and the University of Rome Tor Vergata; all patients
gave oral consent to be entered in the ICTB database.
To rule out the influence of major cardiovascular risk factors on plaque
histomorphological characteristics, the risk factor profile of patients was
assessed. Hypertension was determined by a case history of antihypertensive
drug treatment. Patients affected by either insulin-dependent diabetes mellitus
or treated with diet, oral hypoglycemic agents, or both were included in the
diabetic group. Patients who smoked at least 20 cigarettes per day during
the past 2 years were classified as current smokers. Patients with a total
serum cholesterol level of more than 200 mg/dL (>5.18mmol/L)19 or
who were treated in the past 12 months with a lipid-lowering drug were considered
hypercholesterolemic. Hypertriglyceridemia was documented either on the basis
of the case history or when the serum triglyceride level exceeded 150 mg/dL
(>1.70 mmol/L).19 History of coronary artery
disease and obstructive peripheral vascular disease, previous coronary artery
bypass graft surgery, and preoperative use of aspirin and statins were assessed
from the clinical history.
Intraoperative carotid plaques were removed en bloc to preserve plaque
structure. Samples were fixed immediately on removal in 10% buffered formalin
for 24 hours. The sampling method has been previously reported.18,20 Briefly,
after decalcification, if necessary, specimens were cut transversely every
5 mm, embedded in paraffin, and stained with hematoxylin-eosin and Movat pentachrome
stains. Each segment was removed and numbered sequentially to reconstruct
the entire plaque length by sequential slices. For each plaque, 3 to 10 sections
were examined according to the extension of the plaque (mean, 5 sections per
artery). The entire plaque was evaluated for the presence of an acute or organized
thrombosis, plaque rupture or erosion, extension of necrotic core, calcification,
and intraplaque hemorrhage.
A thrombotically active plaque (TAP) was defined by the presence of
an acute thrombus constituted of platelets or fibrin on the plaque surface
and characterized by lamination with or without interspersed red and white
blood cells. Thrombosis was divided into 2 categories: thrombosis associated
with plaque rupture or superficial erosion. Plaque rupture was defined as
a complete disruption of the fibrous cap over a lipid core with contact of
an acute thrombus with the lipid pool.14 Superficial
erosion was defined as plaque de-endothelialization, associated with the presence
of an acute thrombus in direct contact with the subepithelial tissue of the
cap without any contact with the lipid pool demonstrated in serial sections.
Organized thrombus was characterized by fibrous tissue, sometimes stratified,
associated with a typical angiomatosis, with a network of large thin-walled
vascular channels and a variable number of macrophagic cells loaded with hemosiderin,
visible as scattered brown refractive pigments. Two pathologists (A.M., S.F.),
who were blinded to the patients’ clinical findings, evaluated all histocytological
components of the plaques; intraobserver and interobserver reliability was
more than 98%.
The immunohistochemical study was used to characterize and quantitate
the inflammatory cells present in the cap of ruptured plaques, using the CD68
(antihuman macrophages, Dakopatts, Denmark) and CD3 (antihuman T cell, Dakopatts)
monoclonal antibodies. Cell counting was performed at a magnification of ×400
using a test grid with an area of 0.22 mm2. An average of 10 fields
per section (until the SEM was <5%) were counted. In 30 randomly selected
plaques, 10 for each group, we evaluated the expression of interleukin 6 (IL-6)
antibody (RD System, Minneapolis, Minn).
Data were analyzed using SPSS version 11.0 (SPSS Inc, Chicago, Ill)
software. The Pearson χ2 test was used to assess the differences
in the frequency of thrombosis, cap rupture, and cap erosion, and risk factors
between the groups. t Test for unpaired samples was
used to evaluate differences in age, grade of angiographic stenosis, and degree
of inflammatory infiltrate between the groups. Partial correlation coefficients
describing the relationship between stroke and acute mural thrombosis were
calculated, adjusting for age and major cardiovascular risk factors. P<.05 was considered statistically significant.
Asymptomatic patients and patients affected by major stroke were older
than those affected by TIA (Table 1).
There was no significant difference in the degree of angiographic stenosis
observed between groups either in the ipsilateral or contralateral carotid.
In particular, the mean (SEM) ipsilateral carotid stenosis values were 86.1%
(1.0%) in patients with stroke, 79.5% (3.0%) in patients with TIA, and 84.6%
(1.1%) in asymptomatic patients, whereas the mean (SEM) contralateral carotid
stenosis values were 60.9% (2.7%), 64.2% (5.6%), and 57.5% (3.4%), respectively.
Forty-three percent of patients were treated with statins with no statistically
significant differences between the 3 groups. All patients with symptoms (both
stroke and TIA) were taking aspirin at the time of intervention. The time
interval between symptom onset and carotid endarterectomy was 1 to 30 months
for patients with stroke and 3 to 22 months for patients with TIA.
A TAP was observed in 71 (74.0%) of 96 plaques from patients with ipsilateral
major stroke. This prevalence was significantly higher compared with plaques
from patients with TIA (32 [35.2%] of 91 cases, P < .001)
or asymptomatic patients (12 [14.6%] of 82 cases, P < .001)
(Table 2). An acute thrombus was associated
with cap rupture in 64 (90.1%) of 71 thrombosed plaques from patients with
stroke (Figure 1A) and with cap erosion
in the remaining 7 cases (9.9%). A thrombotic occlusion of the vascular lumen,
documented by angiographic stenosis of more than 95%, was observed in 29 cases
(40.8% of cases with a TAP). In the TIA group, the prevalence of erosion was
approximatively twice that of patients with stroke, being present in 11 of
32 plaques with TAP (Figure 1B). In
asymptomatic patients (Figure 1C), cap
erosion was uncommon (1.2%); in this group, the presence of a TAP was also
less common than in the other 2 groups, although 14.6% of these patients had
pathological evidence of a TAP.
Table 3 shows the prevalence of
TAP in relation to the time interval between the acute cerebral event and
carotid endarterectomy in patients with stroke. In these cases, all 32 plaques
from patients who had endarterectomy within 2 months of symptom onset showed
a TAP associated with rupture of the fibrous cap (Figure 1A). The presence of TAP decreased in parallel with the increase
of the time interval from symptom onset to surgery, although it was still
present in 53.8% of the plaques from patients who had endarterectomy 13 to
24 months after symptom onset. Along with a reduced presence of TAP, organization
of the thrombus was observed in patients who had endarterectomy later than
2 months after symptom onset (Figure 1D).
In a small percentage of cases, fibrin was present inside the organized thrombus
mass. In addition to TAP, an associated organized thrombus was also observed
in 3 of 13 plaques of patients with stroke who underwent carotid endarterectomy
3 to 6 months after symptom onset. A similar finding was observed in 18 plaques
of patients who had endarterectomy at least 7 months after symptoms. Only
2 of 96 cases of stroke had neither acute thrombus nor organized thrombus
present at the histological examination.
The immunohistochemical study demonstrated that the cap of ruptured
plaques in patients affected by stroke was characterized by many inflammatory
cells, principally monocytes and macrophages (CD68) and T lymphocytes (CD3)
(Figure 2A) compared with those cells
observed in ruptured plaques from patients with TIA and patients without symptoms.
Ruptured plaques in patients with stroke had inflammation in the cap almost
twice as dense as that in patients with TIA (P=.001)
and patients without symptoms (P = .001)
(mean [SEM], 49.2 [3.5] vs 29.9 [4.4] vs 29.1 [3.8] inflammatory cells ×
mm2, respectively). No significant differences were observed in
inflammation between patients with TIA and patients without symptoms (P = .89). A high expression of IL-6 was observed
in the macrophagic cells present in the plaques cap of patients with stroke
By univariate analysis, there was no significant difference in the distribution
of the various risk factors between the 3 study groups (Table 1). This may be due to the fact that only patients who underwent
carotid endarterectomy and who presented with a high degree of carotid stenosis
and a large atherosclerotic plaque burden were selected. The correlation between
stroke and mural thrombosis remained highly significant (P = .001), after adjustment for age and major cardiovascular
Our study is the first to our knowledge to describe a large clinicopathologic
series of carotid endarterectomy specimens demonstrating that thrombosis associated
with plaque rupture is one of the major determinants of ischemic stroke in
patients affected by carotid atherosclerotic disease. In addition, our study
showed the presence of a fresh thrombus several months after the first cerebrovascular
event, suggesting that a continuous vulnerable status of the plaque, if not
removed, may trigger continuous release of embolic material, which in turn
may be related to subsequent cerebrovascular events.
Plaque histopathologic examination showed the presence of a fresh endoluminal
thrombus, localized at the rupture site of the fibrous cap in all cases submitted
to carotid endarterectomy within 2 months of symptom onset. This finding suggests
that thrombosis plays a crucial role in the pathogenesis of stroke and supports
the notion that neurological symptoms are probably caused by emboli arising
from thrombotic lesions, as demonstrated by 24-hour transcranial Doppler examinations.11,12
Considering the different types of circulation, it can be assumed that
major stroke has a pathogenetic mechanism due to rupture and thrombosis of
a moderately stenotic plaque, similar to that observed in myocardial infarction.21 Nevertheless, it has been suggested that patients
with high-grade stenosis and poor collateral flow in the circle of Willis,
and therefore secondary decreased intracranial flow, are probably more susceptible
to embolic events than patients with emboli into normally perfused cerebral
The fundamental role of cap rupture and thrombosis in major cerebrovascular
events is confirmed by the fact that we have excluded cases in which the cerebrovascular
event could have been related either to cardiac embolic disease or to stenosis
or malformation in the circle of Willis. Furthermore, the absence of high-grade
stenosis or imaging suggestive of thrombosis at the angiographic examination
of the contralateral carotid artery suggests that the ischemic stroke was
related to the lesion in the ipsilateral carotid artery.
In addition, our study demonstrated that the carotid plaque remains
thrombotically active after the initial clinical event, thus possibly predisposing
patients to a continuous release of emboli in the intracranial vascular bed.
These observations suggest that after an acute cerebrovascular event the carotid
plaque remains chronically unstable if triggering factors, such as a high
inflammatory infiltration of the plaque and an increase in shear stress due
to luminal narrowing progression, are not removed. This vulnerable state may
cause new cap rupture, erosion, or both, together with thrombus formation,
and may be responsible for cases of “stroke in evolution” or “delayed
stroke” as demonstrated by observations showing release of embolic debris
from carotid occlusion up to 22 months after a major stroke.23 The
presence of TAP observed in plaques removed several months after symptom onset
may help to explain the recurrence of a second ipsilateral stroke, which affected
12 of 96 patients in the stroke group in our study. Therefore, these observations
favor the consideration of early percutaneous/surgical intervention in association
with aggressive medical treatment to reduce the risk of subsequent events.
The role of carotid ulceration and thrombosis has been previously evaluated
in various clinicopathological studies with contradictory results.14-17 These
discordant findings may be related to different patient groups enrolled in
some studies (ie, symptomatic patients included only a limited number of cases
of stroke), and thus published data on patients with symptoms are representative
of TIA rather than stroke. In particular, a significant difference in the
prevalence of carotid thrombosis in patients with and without symptoms (29%
vs 12%, respectively) was observed by Van Damme and Vivario.17 In
that series, the low prevalence of thrombosis in patients with symptoms was
most likely because only 20 of 121 cases included were affected by stroke.
A higher prevalence of plaque thrombosis has been also demonstrated in a morphologic
study by Sitzer et al11 (74% of the symptomatic
population), although in this study the total number of patients affected
by stroke was not reported. Conversely, no significant differences were observed
in rates of carotid thrombosis in patients with and without symptoms in 2
different clinical studies,14,15 although
the number of patients affected by stroke was low in both studies.
In addition to plaque rupture and thrombosis, other factors may increase
the risk of stroke. Our results showed that the severity of clinical event
is significantly correlated with cap inflammation in ruptured plaques, suggesting
that diffuse plaque inflammation may be related to the severity of cerebrovascular
events influencing embolic size and composition as well as brain vessels and/or
tissue responsiveness.24,25 Increasing
evidence suggests that postischemic inflammation contributes to the extension
of ischemic brain injury.25,26 Cerebral
ischemia is accompanied by a marked inflammatory reaction that is initiated
by ischemia-induced expression of cytokines, adhesion molecules, and other
inflammatory mediators.27-29 Furthermore,
therapeutic strategies aimed at reducing inflammation have decreased progression
of brain damage.26,30 These studies
together with our data suggest the hypothesis that inflammatory cells infiltrating
the ruptured carotid plaques may release vasoactive substances promoting severe
cerebral ischemia. Moreover, carotid inflammatory cells could release some
cytokines, such as IL-6, a key regulator of inflammatory mechanism in stroke
pathophysiology,31 directly in the intracerebral
The major limitation of our study is that we did not include all patients
who had a carotid stroke, but limited our analysis only to patients who underwent
carotid endarterectomy at different time intervals from the acute event. In
the future, sophisticated imaging technique with tissue characterization may
give similar morphologic information without the need of tissue sample analysis.
Another limitation is that C-reactive protein and cytokine levels were not
available in the database for all patients enrolled in our study.
Our study may have several important clinical implications. The findings
suggest that major stroke is significantly associated with an acute thrombosis
which in turn complicates a vulnerable plaque characterized by a diffuse inflammatory
infiltrate. Conversely, patients affected by TIA and patients without symptoms
were characterized by more stable plaques at the histologic examination. These
data, therefore, suggest 2 types of carotid artery disease: one form that
is stable and unlikely to produce symptomatic embolization or carotid occlusion
and another form that is not necessarily stenotic but unstable and at high
risk of producing symptomatic embolization or complete occlusion of a carotid
The present study demonstrates that the grade of stenosis is not sufficient
to identify patients at high risk to develop an acute cerebrovascular event,
and it is of great importance to identify other factors for correct risk stratification
of ischemic disabling events in patients affected by atherosclerotic carotid
disease. Besides imaging techniques, such as pixel analysis at ultrasonography,
magnetic resonance, and local temperature probes that could help in identifying
vulnerable plaques,32,33 high
sensitive inflammatory circulating markers, such as C-reactive protein, pregnancy-associated
plasma protein A, cytokines (eg, IL-6), are possible candidates for active
plaque detection.31,34-37 In
the Physicians’ Health Study,38 high
baseline C-reactive protein levels were associated with increased risk of
stroke, independent of smoking and other risk factors. In addition, we previously
demonstrated that patients with hyperfibrinogenemia were characterized by
a greater inflammatory infiltrate and thinner atherosclerotic plaque cap and
also had increased risk of thrombosis and rupture compared with patients with
lower fibrinogen levels, independent of other risk factors.18 Moreover,
individual major histocompatibility complex haplotypes may determine the specific
inflammatory patterns, the type of immune response to exogenous antigens,
and the induction of autoimmune reactions in the plaque. Therefore, polymorphisms
in genes coding for metabolic proteins (cytokines, proteolytic enzymes) involved
in the processes related to plaque destabilization might represent useful
markers to stratify the population at high risk for vascular atherosclerotic
The present study has implications for the natural history and management
of acute cerebrovascular disease. Our results showed the presence of TAP up
to 2 years after the onset of a cerebrovascular event. Only 46.2% of cases
had stabilized plaques associated with an organized thrombus. Further studies
are necessary to establish whether, immediately after an ischemic event, initiating
aggressive medical therapy designed to stabilize plaque and decrease the inflammatory
infiltrate will reduce the risk of stroke progression and whether early carotid
endarterectomy procedure that completely removes the atherosclerotic plaque
will decrease the risk of subsequent ischemic events.
Corresponding Author: Luigi Giusto Spagnoli,
MD, Cattedra di Anatomia ed Istologia Patologica, Dipartimento di Biopatologia
e Diagnostica per Immagini, Universita' di Roma Tor Vergata, Via Montpellier
1, 00133 Rome, Italy (firstname.lastname@example.org).
Author Contributions: Dr Spagnoli had full
access to all of the data in the study and takes responsibility for the integrity
of the data and the accuracy of the data analysis.
Study concept and design: Spagnoli, Mauriello,Sangiorgi,
Fratoni, Bonanno, Schwartz, Pistolese,Ippoliti, Holmes.
Acquisition of data: Mauriello, Fratoni, Piepgras,Pistolese,
Analysis and interpretation of data: Spagnoli,Mauriello,
Sangiorgi, Fratoni, Bonanno, Schwartz, Holmes.
Drafting of the manuscript: Mauriello, Sangiorgi,Fratoni,
Bonanno, Schwartz, Pistolese, Ippoliti, Holmes.
Critical revision of the manuscript for important
intellectual content: Spagnoli, Mauriello, Sangiorgi,Fratoni, Schwartz,
Statistical analysis: Mauriello.
Obtained funding: Spagnoli.
Administrative, technical, or material support:Fratoni,
Study supervision: Spagnoli, Mauriello, Schwartz,Holmes.
Funding/Support: This study was supported by
grant COFIN 2001 from MURST and the Mayo Foundation, Rochester, Minn.
Role of the Sponsors: MURST and the Mayo Foundation
did not participate in the design and conduct of the study, in the collection,
analysis, and interpretation of the data, or in the preparation, review, or
approval of the manuscript.
Acknowledgment: We thank William D. Edwards,
MD, for providing access to the Mayo specimens and Alfredo Colantoni, for
his helpful technical assistance.