From the Medical Decision Making Unit (H.F., J.K.), and Department of
Surgery (H.F., J.K., J.M. v B., J.H. v B.), Leiden University Hospital,
Leiden, The Netherlands.
Correspondence to Jary M. Van Baalen, MD, PhD, Department of Surgery, Leiden University Hospital, PO Box 9600, 2300 RC Leiden, Netherlands.
Methods—A systematic review of the literature was performed using
standard meta-analytical techniques.
Results—Incidence of recurrent stenosis: The data
were very heterogeneous. The risk of recurrent
stenosis was 10% in the first year, 3% in the second, and 2%
in the third. Long-term risk of recurrent stenosis is about 1%
per year. Risk of stroke: The reported relative risks of stroke in
patients with recurrent stenosis compared with patients without
recurrent stenosis showed extreme heterogeneity
and ranged from 10 to 0.10. The random effects summary estimator of
relative risk was 1.88.
Conclusions—The data were very heterogeneous, and
much better data are needed to arrive at truly reliable estimates of
these important parameters of follow-up. It is clear,
though, that the risk of recurrent stenosis is highest in the
first few years after carotid endarterectomy and
very low in later years. By use of general decision-analytic arguments,
it can be argued that, given the test characteristics of carotid
ultrasound, a small number of tests can be done in the first few years
and that testing for restenosis should not be done after 4
years.
Articles were included if (1) they were published in 1985 or later, (2)
they reported on long-term follow-up of 100 patients or more, (3) the
operation carried out was carotid endarterectomy
with direct or patch closure, (4) restenosis was defined as a
50% stenosis of the operated artery, (5) patients were
followed up with noninvasive diagnostic techniques, (6)
follow-up was systematic and not just in case of symptoms, and (7) they
were written in English. Articles were excluded if (1) they reported on
asymptomatic patients only, (2) they reported on combined
coronary and carotid surgery only, (3) only external carotid
endarterectomies were performed, and (4) the carotid bifurcation was
completely resected and replaced by a graft.
Four variables were considered essential for determining the
incidence of restenosis: (1) the number of patients or arteries
at risk, (2) the number of patients or arteries with
restenosis, (3) the average follow-up time, and (4) the
definition of restenosis. Articles that met the inclusion
criteria, but in which one or more of these essential variables
were missing, were coded as "not evaluable." We carefully excluded
articles that reported on patients whose data were already used in
another article from the same institution, except when the methods
sections made it absolutely clear that the patients did not overlap. In
case of multiple publications, we chose the one most appropriate to
this review, preferably the most recent. We included only recent
articles because we wanted to make sure that modern noninvasive
techniques were used in a relatively uniform way for detecting
restenosis. In most hospitals these techniques became available
in the late 1970s and early 1980s. Allowing for a publication lead time
of 5 years, we arbitrarily set our cutoff year at 1985. The increasing
use of aspirin as standard postoperative treatment was another reason
not to include early reports. We excluded articles reporting on fewer
than 100 patients because restenosis is relatively rare and
symptomatic restenosis even rarer. Including small
studies would have meant sparse data tables with statistically
unmanageable "zero-cells."
Data Extraction
Data Analysis
Because most studies report that the rate of restenosis is not
constant over time, we tried to construct the summary incidence curve
by combining data from life table analyses of
restenosis rates. However, because only a few articles
presented life table data in tabular form, we chose a different
approach and plotted the proportion of patients or arteries who had
developed restenosis against the average follow-up time. If the
incidence rate is constant, these points should lie on a straight line
through the origin with a slope equal to the incidence per year in
accordance with the well-known formula incidence = (number of
restenoses)/(total person-years of follow-up). If the incidence
is not constant over time, the points should lie on a curved line that
is given (approximately) by the cumulative incidence (or hazard in more
statistical terms). A generalization of the constant incidence (or
hazard) with cumulative incidence proportional to time is given by the
Weibull model, in which the cumulative incidence is modeled as
Another 20 articles51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 were excluded because one or more
essential variables were missing: in 14
articles,1 the
definition of restenosis was either not given or different from
50%; in 5 articles,59 62 63 66 70 the average follow-up
time was not specified; and in 1 article,52 the number of
restenoses was not clear. The main points of the remaining 29
articles14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 that met all the requirements are summarized
in Table 1
Cumulative Incidence of Restenosis
We carefully studied each article for medical and epidemiological
factors that might explain heterogeneity, but
unfortunately only five variables were present in 20 or more of
the 29 studies: publication year, mean age, percentage men, percentage
asymptomatic patients, and percentage patch closure. Of
these, only mean age and percentage patch closure were statistically
significant (P<.001); increasing mean age was associated
with a higher risk of ipsilateral stroke, and increasing percentage
patch closure with a lower risk. Although they were statistically
significant, these two variables did not have an appreciable impact
on
The cumulative Weibull incidence curve (CI[t]) fitted to the data in
Fig 1
The curve rises steeply to a cumulative incidence of about 10% at 1
year and then gradually flattens to about 20% at 10 years. The yearly
risk of restenosis is by far the greatest in the first year:
about 10%. After the first year the risk is much lower, about 3% in
the second year, 2% in the third and fourth years, and then
gradually diminishing further until in the long run it
stabilizes at about 1% per year. When the five most extreme outliers,
Healy et al,24 Keagy et al,26 Rosenthal et
al,37 Kieny et al,27 and Donaldson et
al,20 were removed from the analysis, the resulting
curve did not change significantly.
Relative Risk of Ipsilateral Stroke in Patients With
Restenosis
It is clear from Table 3
The odds ratios vary widely, roughly ranging from 1:10 to 10:1, again
showing extreme heterogeneity. There are only two
studies, Healy et al24 and Kinney et al,28 in
which the 95% confidence interval does not include 1. All other
studies have wide confidence intervals including 1.
We calculated several commonly used summary estimators of relative
risk. The standard Mantel-Haenszel exact summary odds ratio is 0.95
(95% confidence interval, 0.54–1.61). The associated
In most Western countries, governments, insurance companies, and Health
Maintenance Organizations are expected to demand more
cost-effective health care. We believe that only two retrospective
studies on follow-up after carotid endarterectomy
will be insufficient evidence to meet these demands. For this reason,
we did a systematic review of the literature, hoping to obtain good
estimates of the two most important medical factors involved in
follow-up: the rate of restenosis and the risk of ipsilateral
stroke once restenosis has developed. Of some 500 articles
studied, only 29 contained evaluable data on the rate of
restenosis in 100 or more patients. Because the risk of
restenosis varies with time, we plotted the cumulative
incidence of restenosis against the average follow-up time.
Several authors, in comparing restenosis rates from various
studies, have failed to do this, thereby introducing an important bias
in their conclusions.
We preferred not to use the person-years of follow-up method because it
leaves valuable information in the data unused. The person-years method
is the ratio of the number of events and the total number of follow-up
years. The resulting number is a uniform value for the entire follow-up
period and so does not indicate whether the risk varies with time. If
we plot the number of events against the average follow-up time,
variations of the risk over time become immediately clear. The
resulting graph (Figure
Ten studies provided evaluable data on the relative risk of ipsilateral
stroke in patients with restenosis compared with patients
without restenosis. Because of the
heterogeneity between results of studies, it is
imperative that ipsilateral stroke rates for patients with and without
restenosis are compared only within studies. Several authors
have compared stroke rates between studies, but we believe this
introduces biases that make the resulting figures uninterpretable.
These ten studies, too, showed great heterogeneity,
their odds ratios ranging from roughly 1:10 to 10:1. The small numbers
of events complicate matters further because they destabilize the odds
ratios—a few events more or less may cause dramatic changes—and are
associated with wide confidence intervals and low statistical power. In
addition, they make heterogeneity much harder to deal
with, because they may be the single cause of it or there may be other
causes involved. Unfortunately, the other causes cannot be dealt with
until more events are available. Under these circumstances, the results
of any statistical analysis of these data must be approached
with caution, and we felt uncomfortable in presenting overall
summary estimators. We did so only for the benefit of the reader who
wishes to know them. We do not encourage that they be used in patient
care. In theory, the Mantel-Haenszel and inverse variance weighted
estimators should not be too far apart. The abnormal difference found
here, 0.95 versus 1.61, should be attributed to extreme
heterogeneity. The DerSimonian-Laird random effects
summary estimator, which in recent years has become a popular method to
deal with "unexplained" heterogeneity, showed a
moderately increased overall odds ratio of 1.88. But considering the
extreme heterogeneity and low event rates, we think it
is better to accept the simple fact that, as yet, there is no
scientific evidence for any conclusion about the relative risk. In the
absence of harder evidence, the literature shows that many vascular
surgeons reason by analogy. They assume that the secondary lesions of
intimal hyperplasia and recurrent atherosclerosis carry
the same increased risk of obstructive and thromboembolic
ischemia as primary obstructive and atherosclerotic lesions.
Until more reliable data are available, this reasoning by analogy is
probably a safe way to go.
If this review has shown one thing, it is that to develop better
follow-up schedules, better data are needed. First of all, a clear and
uniform definition of restenosis is necessary, both for
angiographic and ultrasound examination. Second, if restenosis
rates are to be compared between studies, identical follow-up schedules
should be used to prevent timing biases caused by benign hyperplastic
lesions that quickly develop and regress. Third, it is absolutely
necessary that studies use identical definitions and detection methods
for ipsilateral stroke. Fourth, more ipsilateral stroke events are
needed for a reliable estimate of the relative risk of stroke in
patients with restenosis. For example, assuming a relative risk
of 2 and using the event rates from Table 3
In the first 2 years, then, approximately 10% of patients develop
restenosis. If one ultrasound test is done at the end of this
period, 9% of all patients will be true-positive and 12% will be
false-positive. If more tests are done in between, the prevalence of
restenosis and the number of true-positives will be lower and
the number of false-positives will be higher. In short, more frequent
testing will cause larger numbers of false-positive patients and not
necessarily increase the number of true-positives. Given the benign
nature of the stenotic lesion in this period and the
possibility of regression, we have decided to test only twice at our
institution: after 1 year and after 2 years. In later years, recurrent
atherosclerosis will occur in 1% to 2% of patients
per year. Because this lesion is more dangerous, it may be more
important to detect it. Unfortunately, the low prevalence makes for
even higher numbers of false-positive tests than in the first 2 years.
Without a full cost-effectiveness study, it is hard to find a balance
between the harm done by missing dangerous restenoses and the
discomfort caused by false-positive tests. Until such a study is
available, we recommend that ultrasound testing for restenosis
be done sparingly.
2 References 17, 18, 22, 24, 28, 29, 32, 33, 37, 42, 49, 50.
Received July 23, 1997;
revision received September 18, 1997;
accepted October 24, 1997.
© 1998 American Heart Association, Inc.
Comments, Opinions, and Reviews
Carotid Recurrent Stenosis and Risk of Ipsilateral Stroke
A Systematic Review of the Literature
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—The main goal of
follow-up after carotid endarterectomy is to
prevent new strokes caused by recurrent stenosis. To determine
the most cost-effective follow-up schedule, it is necessary to know the
incidence of recurrent stenosis and the risk of stroke it
carries.
Key Words: carotid artery • carotid endarterectomy • carotid stenosis • follow-up studies
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
Many vascular
surgeons and neurologists prefer to follow up their patients after
carotid endarterectomy because they assume that (1)
a considerable number of patients will develop restenosis and
(2) restenosis increases the risk of ipsilateral stroke. There
is, however, little consensus on the best follow-up schedule, mainly
because reported restenosis rates and associated ipsilateral
stroke risks vary widely. Although there is little consensus as yet, we
believe that important trends will require a more uniform follow-up
policy in the future. First, the results of the ECST,1 the
NASCET,2 and the ACAS3 will generate a steady
inflow of patients into vascular follow-up programs, especially in the
aging Western populations. Second, insurance companies and governments
will demand more cost-effective health care and will take measures to
cut spending while maintaining quality of care. These trends will
sooner or later spur vascular surgeons and neurologists to develop
cost-effective follow-up schedules that offer the greatest medical
benefit at a reasonable cost. The most pivotal medical factors in
determining a cost-effective follow-up schedule are the
restenosis rate after carotid
endarterectomy and its associated risk of
ipsilateral stroke. In this article we present a systematic review
of restenosis and ipsilateral stroke rates published since
noninvasive diagnostic techniques became widely
available.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Literature Search and Inclusion Criteria
We searched for articles by running the query "(carotid
surgery or carotid endarterectom1) and
(restenosis or recurrent adj5 stenosis)" on the
MEDLINE database.4 We purposely chose this highly specific
query to minimize the risk of false-positive hits. To find any articles
we might have missed, we ran the highly sensitive but less specific
query "carotid surgery endarterectomy
restenosis recurrent stenosis" on a MEDLINE
implementation that uses fuzzy search algorithms.5 The
abstract of each article was carefully studied, and if there was any
suggestion of the data we looked for, the full text was retrieved. If
an article contained data on restenosis and ipsilateral stroke
risk, its reference section was checked for further leads.
Data were extracted by use of a predefined data sheet. If
present, the following variables were extracted: (1)
general—year of publication, number of patients, institution; (2)
patient characteristics—sex and mean age; (3) risk
indicators—hypertension, smoking, diabetes,
hyperlipidemia, coronary artery disease; (4)
indication for operation—asymptomatic, transient
ischemic attack/amaurosis fugax stroke; (5) ipsilateral stroke
events—fatal and nonfatal, perioperative and late; (6)
restenosis events—number of patients or arteries at risk,
number of patients or arteries with restenosis, average
follow-up time, definition of restenosis; (7) risk of
ipsilateral stroke in patients with restenosis—2x2 table of
ipsilateral stroke by restenosis; (8) screening
characteristics—diagnostic technique used; and (9)
surgical technique—use of patch angioplasty. The data were extracted
by the first author and checked by one of the other authors.
Data analysis had two goals: (1) to construct a summary
curve of the cumulative incidence of restenosis after carotid
endarterectomy, and (2) to derive a summary value
of the relative risk of ipsilateral stroke in patients with
restenosis compared with patients without restenosis. 
tß. Despite the great heterogeneity,
we fitted such a Weibull model to get an impression of the decreasing
incidence rate over time. We used the nonlinear regression module of
SPSS-7, taking the sample sizes at risk as weights. The summary value
for the relative risk of ipsilateral stroke in patients with
restenosis was determined by calculating the weighted sum of
the relative risks of the individual studies using
Mantel-Haenszel–type weights and maximum likelihood inverse variance
weights.7 8 We used the odds ratio as an estimate of the
relative risk because of computational advantages and low event rates.
All results were tested for
heterogeneity.7 8 If
heterogeneity was present, we followed Greenland's
recommendations9 and carefully looked for variables
that could "explain" the interstudy variation in a clinically
plausible way. If we succeeded in constructing a
homogeneous model, we used its resulting summary estimate.
If no homogeneous model could be obtained, we calculated a
random-effects summary estimator as described by DerSimonian and
Laird.10 Calculations were done on a personal computer
using Excel spreadsheet software,11 and SPSS,12
and Egret13 statistical software.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
The literature search resulted in 57 articles that met the
inclusion criteria.14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 Of these, 8
articles43 44 45 46 47 48 49 50 were excluded because they reported on
patients already used in other publications from the same institution:
Bandyk et al43 was excluded in favor of Kinney et
al28 ; Callow et al44 and O'Donnell et
al50 were excluded in favor of Mackey et al29 ;
De Letter et al45 and Eikelboom et al46 were
excluded in favor of Ten Holter et al41 ; and Healy et
al47 was excluded in favor of Healy et
al.24 
.
View this table:
[in a new window]
Table 1. Main Points of 29 Studies Reporting Incidence of
Restenosis After Carotid Endarterectomy
The Figure shows a plot of the
cumulative incidence of restenosis against average follow-up
time. On inspection, the data show enormous
heterogeneity, which is confirmed by formal statistical
tests that show 
2 values in excess of 300 at 28 degrees
of freedom.
View larger version (15K):
[in a new window]
Figure 1. Plot of cumulative incidence of carotid restenosis
against average follow-up period. 2 scores for heterogeneity. Outside
of the five variables present in 20 or more studies, no other
variables were tested because there would have been too many
missing cases. 
is given by CI(t)=0.0864xt0.28. The corresponding
risk of restenosis in a certain year for somebody who has not
experienced restenosis before is given by
1-e-
CI, with CI equal to the increase in
CI during that year. Numerical results are given in Table 2.
View this table:
[in a new window]
Table 2. Life Table of Cumulative Incidence of Carotid
Restenosis Associated With the Fitted Weibull Curve in the
Figure
Of the 57 articles, 12 provided ipsilateral stroke rates for
both patients with and without restenosis, allowing calculation
of within-study relative risks.* The
studies by Nicholls et al49 and O'Donnell et
al50 were excluded because their data were already used in
Healy et al24 and Mackey et al,29 respectively.
The main points of the remaining 10 articles are summarized in Table 3. The study by Cuming et
al18 was excluded from calculations because the number of
ipsilateral strokes is zero in both the restenosis and no
restenosis groups.
View this table:
[in a new window]
Table 3. Summary of 10 Studies Reporting Incidence of
Ipsilateral and Contralateral Stroke in Relation to Restenosis
After Carotid Endarterectomy1 that there are only very few ipsilateral
strokes. In the restenosis group, only one study has more than
five events, and no less than three studies have no events at all. In
the no restenosis group, only three studies have more than 10
events, and five studied have fewer than 5 events. Only the study by
Hansen et al22 has a reasonable number of events in both
groups. 2
test for heterogeneity is 37.2 (8 df,
P<.001), which confirms the impression of extreme
heterogeneity gained from the table. We also calculated
a summary estimator using maximum likelihood estimates of variance. Its
value is 1.61 (95% confidence interval, 0.95–2.72), and the
associated 2 test for heterogeneity is
23.8 (8 df, P<.01). The DerSimonian-Laird random
effects estimator is 1.88 (95% confidence interval, 0.71–4.98).
2 tests for heterogeneity are not done
when this method is used because heterogeneity is
already accounted for in the summary estimate and its confidence
interval.
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
Three state-of-the-art, large-scale clinical trials have now
clearly established the value of carotid
endarterectomy in both
symptomatic1 2 and
asymptomatic3 patients. Although some debate on
the indications for operation will continue, especially in
asymptomatic patients,71 it is certain that
carotid endarterectomy will take its place as a
widely accepted surgical procedure for a number of indications. It is
equally certain that the number of carotid endarterectomies carried out
will be considerable, although it is hard to make exact predictions.
This acceptance will raise the question of what to do with the
patient after the operation: intensive follow-up, limited follow-up, or
no follow-up at all? So far, there seems to be a tacit consensus that
some form of follow-up is necessary, motivated by the assumptions that
patients may develop restenosis and that restenosis
increases the risk of ipsilateral stroke. There were, however, only
very few articles that directly study both assumptions, most notably
Healy et al,24 Mackey et al,29 and Mattos et
al.32 Of these three, only Mackey and Mattos study the
effectiveness of follow-up, both raising doubts about its value. Most
other authors in this review were primarily interested in the
occurrence of restenosis and its relation to surgical
technique, intraoperative detection of residual disease,
cardiovascular risk indicators, and medical
treatment. ) shows discouraging
heterogeneity: for studies with an average follow-up
time between 2 and 3 years, for example, the reported cumulative
incidence ranges between 1.9%27 and 22.1%.26
Formal statistical tests showed 2 values in excess of
300 at 28 df. In general, most data analysts will consider 1
to 2 2 points per df a good indication of
homogeneity. Ten points per df is extreme and usually means
that it will not be possible to deal with heterogeneity
in a satisfactory way. When we tried to explain
heterogeneity, we were greatly impeded by the scarcity
of relevant study characteristics. Only five characteristics were
present in 20 or more studies. Moreover, important characteristics
such as average number of follow-up investigations per patient,
definition of stroke, stroke ascertainment, exact technical definition
of 50% stenosis, etc, were often absent altogether.
Nevertheless, we believe the Weibull summary curve we constructed is
reasonably accurate. First, it agrees with the common clinical
observation that the rate of restenosis is much higher in the
first years and decreases over time. Second, the shape of the curve
coincides remarkably with five of the largest studies in which
graphical curves were published.19 24 25 32 41 Third, the
curve is unaffected by outlier removal. Fourth, the
heterogeneity in the first few years may well be caused
by commonly observed regression of intimal hyperplasia. Because this
lesion may quickly develop and regress, the exact timing of noninvasive
testing may to a large extent determine the observed rate of
restenosis. We therefore see the summary curve as a useful
instrument in thinking about follow-up schedules. However, we are well
aware that the incidence of stroke after carotid
endarterectomy is low and that restenosis
in itself is only one of the factors that influence the odds of stroke
as patients may suffer stroke from other causes than
restenosis. , a series of 1500 patients
or more is needed to obtain numerically stable and statistically
significant estimates of relative risk. Fifth, and last, uniform
standards for the pathological classification of restenosis are
necessary to assess whether early hyperplasia and late
atherosclerosis have different risks of ipsilateral
stroke. Despite the weaknesses in the data, it is possible to make a
few important points on follow-up using general decision analytic
principles. In doing this we have made five assumptions: (1) the
incidence curve constructed in this review is reasonably accurate; (2)
the relative risk of ipsilateral stroke in patients with
restenosis is moderately increased, say 2.0; (3) in the first 2
years restenosis is caused by largely benign, obstructive
intimal hyperplasia that may quickly develop and regress; (4) in later
years restenosis is caused by recurrent
atherosclerosis that may give rise to thromboembolic
ischemia; and (5) the sensitivity and specificity of ultrasound
examination for a 50% stenosis of the carotid artery are 91%
and 87%, respectively.72
Selected Abbreviations and Acronyms
ACAS
=
Asymptomatic Carotid Atherosclerosis Study
ECST
=
European Carotid Surgery Trial
NASCET
=
North American Symptomatic Carotid
Endarterectomy Trial
Acknowledgments
The authors thank Prof Hans van Houwelingen and Dr Koos
Zwinderman for statistical guidance.
Footnotes
1 References 51, 53–58, 60, 61, 64, 65, 67–69. 
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  W. E. Hellings, F. L. Moll, J.-P. P.M. de Vries, P. de Bruin, D. P.V. de Kleijn, and G. Pasterkamp Histological Characterization of Restenotic Carotid Plaques in Relation to Recurrence Interval and Clinical Presentation: A Cohort Study Stroke, March 1, 2008; 39(3): 1029 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. Hellings, F. L. Moll, J.-P. P. M. De Vries, R. G. A. Ackerstaff, K. A. Seldenrijk, R. Met, E. Velema, W. J. M. Derksen, D. P. V. De Kleijn, and G. Pasterkamp Atherosclerotic Plaque Composition and Occurrence of Restenosis After Carotid Endarterectomy JAMA, February 6, 2008; 299(5): 547 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Szeplaki, L. Varga, J. Laki, E. Dosa, S. Rugonfalvi-Kiss, H. O. Madsen, Z. Prohaszka, A. Kocsis, P. Gal, A. Szabo, et al. Low C1-Inhibitor Levels Predict Early Restenosis After Eversion Carotid Endarterectomy Arterioscler. Thromb. Vasc. Biol., December 1, 2007; 27(12): 2756 - 2762. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Schanzer, A. Hoel, C. D. Owens, N. Wake, L. L. Nguyen, M. S. Conte, and M. Belkin Restenosis After Carotid Endarterectomy Performed With Routine Intraoperative Duplex Ultrasonography and Arterial Patch Closure: A Contemporary Series Vascular and Endovascular Surgery, July 1, 2007; 41(3): 200 - 205. [Abstract] [PDF]
|
|
|
|
|
|
W. S. Lesley and C. J. Buckley Stent-Angioplasty for Internal-to-External Carotid Artery Anastomotic Stenosis: A Case Report Vascular and Endovascular Surgery, July 1, 2007; 41(3): 262 - 264. [Abstract] [PDF]
|
|
|
|
|
|
S. Rugonfalvi-Kiss, E. Dosa, H. O. Madsen, V. Endresz, Z. Prohaszka, J. Laki, I. Karadi, E. Gonczol, L. Selmeci, L. Romics, et al. High Rate of Early Restenosis After Carotid Eversion Endarterectomy in Homozygous Carriers of the Normal Mannose-Binding Lectin Genotype Stroke, May 1, 2005; 36(5): 944 - 948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Groschel, A. Riecker, J. B. Schulz, U. Ernemann, and A. Kastrup Systematic Review of Early Recurrent Stenosis After Carotid Angioplasty and Stenting Stroke, February 1, 2005; 36(2): 367 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Higashida, P. M. Meyers, C. C. Phatouros, J. J. Connors III, J. D. Barr, D. Sacks, and for the Technology Assessment Committees of the Am Reporting Standards for Carotid Artery Angioplasty and Stent Placement Stroke, May 1, 2004; 35(5): e112 - e134. [Full Text] [PDF]
|
|
|
|
|
|
R. Bond, K. Rerkasem, and P.M. Rothwell Systematic Review of the Risks of Carotid Endarterectomy in Relation to the Clinical Indication for and Timing of Surgery Stroke, September 1, 2003; 34(9): 2290 - 2301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Kapral, H. Wang, P. C. Austin, J. Fang, D. Kucey, B. Bowyer, and J. V. Tu Sex Differences in Carotid Endarterectomy Outcomes: Results From the Ontario Carotid Endarterectomy Registry Stroke, May 1, 2003; 34(5): 1120 - 1124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E.J. Cunningham, R. Bond, Z. Mehta, M.R. Mayberg, C.P. Warlow, and P.M. Rothwell Long-Term Durability of Carotid Endarterectomy for Symptomatic Stenosis and Risk Factors for Late Postoperative Stroke Stroke, November 1, 2002; 33(11): 2658 - 2663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. N. Post, J. Kievit, J. M. van Baalen, W. B. van den Hout, and J. H. van Bockel Routine Duplex Surveillance Does Not Improve the Outcome After Carotid Endarterectomy: A Decision and Cost Utility Analysis Stroke, March 1, 2002; 33(3): 749 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Phatouros, R. T. Higashida, A. M. Malek, P. M. Meyers, T. E. Lempert, C. F. Dowd, and V. V. Halbach Carotid Artery Stent Placement for Atherosclerotic Disease: Rationale, Technique, and Current Status Radiology, October 1, 2000; 217(1): 26 - 41. [Abstract] [Full Text]
|