вход по аккаунту


Does carotid endarterectomy decemberrease stroke and death in patients with transient ischemic attacks.

код для вставкиСкачать
Does Carotid Endarterectomy
Decrease Stroke and Death in Patients
with Transient Ischemic Attacks?
Committee on Health Care Issues, American Neurological Association"
Our committee has critically evaluated the evidence
regarding the value of carotid endarterectomy for preventing death and stroke in patients with carotid territory transient ischemic attacks (TIAs), who have carotid
stenosis on the side corresponding to the symptoms.
We conclude that the procedure may be of value if the
surgical complication rate is very low, but that the net
effect in the United States as a whole may be unfavorable. In this report we supply the evidence on
which this judgment was based.
Since the first report of carotid endarterectomy in
1954 1101, the procedure has been done with increasing frequency in the U.S.: in 1984, 103,000 such procedures were done in civilian hospitals {8]. The procedure has not been as popular in the United IGngdom
where there were an estimated 1,000 endarterectomies done in 1980 per 30 X lo6 population, compared to 85,000 per 130 x 10' population in the U.S.
This represents an annual rate for the population of
the United Kingdom of less than 5% of that in the rate
for the U.S. C291. These figures do not allow us to
determine whether there are too few endarterectomies
in the United Kingdom or too many in the U.S.
It has been estimated that there are about 35,000
new patients each year with carotid TIA or recovered
stroke with stenosis who would be candidates for
carotid endarterectomy 1311. Assuming the estimate is
correct, the number of carotid endarterectomies each
year far exceeds the number of likely new patients.
Some of this excess may be due to surgery for asymptomatic carotid stenosis. In at least one U.S. city, about
half of the carotid endarterectomies were done in patients without cerebral ischemic symptoms { 31.
There have been two published randomized clinical
trials that assessed the effect of carotid endmerectomy
in patients with TIAs. 'The first was a cooperative study
in the U.S. 1131. The investigators reported a favorable
outcome following carotid endarterectomy. However,
when the surgical mortality and stroke morbidity were
added to the stroke probability in follow-up, there was
no significant difference between the patients who had
surgery and those who did not. Since nearly 45% of
the patients who were randomized were noted to have
had vertebrobasilar T1,4s, it is not possible to reach
any conclusion concerning benefit from carotid endarterectomy in patients with carotid TIAs.
Another randomized controlled trial was reported
from Newcastle-upon-Tyne, England 122). After 20
patients underwent endarterectomy, it was noted that
7 suffered stroke or death at the time of surgery or
within the first 30 days thereafter. Because of this high
mortalitytmorbidity the study was abandoned.
At least two pathophysiological mechanisms have
been evoked as the primary causes for TIAs: (1) embolization of thrombus or atheromatous material from
an ulcerated atherosclerotic plaque; and (2) reduced
blood flow from high-grade stenosis of a carotid artery
Hemorrhage into an athLerosclerotic plaque has been
noted to be commonly present in plaques that were
associated with ischemic symptoms and correlated well
with severity of stenosis 1161. The relative importance
of these related phenomena is uncertain and there may
be other mechanisms that cause TIAs.
The central question is whether there is a favorable
risk-to-benefit ratio for carotid endarterectomy in patients who have carotid stenosis and focal TIAs in the
'This report was authored by Jack P. Whisnanr, MD, Department of
Neurology, Mayo Clinic and Mayo Medical School, Rochester, MN;
Lloyd Fisher, PhD, Department of Biostatistics, University of Washington, Seattle, WA; James T. Robertson, MD, Department of
Neurosurgery, University of Tennessee College of Medicine, Memphis, T N - and Peritz Scheinherg, MD, Department of Neurology,
University of Miami School of Medicine, Miami, FL. It was developed for the American Neurological Association Committee on
Health Care Issues. Members of the ANA Committee on Health
Care Issues are: Peter J. Dyck, MD (Chairman), Roger N. Rosenberg, MD, Nicholas Vick, MD, and Joseph J. Volpe, MD.
Received Dec 2, 1986, and in revised form Jan 9, 1987. Accepted
for publication "Tan 9.. 1987.
Address correspondence to Dr Whisnant, Mayo Clinic, Rochester,
M N 55905.
distribution of the artery where the lesion is present.
Several interrelated considerations must be weighed to
address this central question:
1. What is the probability of survival over time in patients with carotid TIAs and specifically in those
with carotid stenosis?
2. What is the probability of stroke in patients with
carotid stenosis and TIAs when left untreated or
when treated medically?
3. Does the degree of carotid stenosis in patients with
carotid TIAs affect outcome?
4. What is the mortality and stroke morbidity associated with carotid endarterectomy in patients with
carotid TIAs?
5. What is the probability of stroke in patients following a successful carotid endarterectomy ?
Survival after TIA
The mortality ratio (observedexpected mortality) after
the first TIA is nearly 10: 1 at 1 month but drops very
sharply, so that by 2 years it is about 1.5 where it
remains for the next 8 years of follow-up 1301. This
simply means that the actual mortality corrected for
age is stable after the first 2 years, with relatively small
mortality in excess of that for a normal population.
Patients with TIAs who have survived free of stroke
for 30 days still have about a 50% chance of surviving 7 years [J. P. Whisnant, unpublished data from
Rochester, MN, population, 19731.
Mortality following the onset of TIAs is due primarily to cardiac disease [23, 27, 28, 321, so there is little
reason to expect that carotid endarterectomy or other
treatment aimed at preventing stroke would have
much effect on survival. Patients in the carotid endarterectomy cooperative study in the U.S. 111 experienced survival comparable to that of the medically
treated patients when the surgical mortality was discounted. It is not likely that any treatment for TIAs,
unless it has a beneficial effect on cardiac disease, will
have much effect on survival.
Probability of Stroke after TIA
There is a large body of literature on the frequency of
stroke following TIA based on observations from clinical series. The differing referral patterns, patient characteristics, and time after TIA make it difficult to generalize from these data. The most useful information
has come from population-based studies. In the
Rochester, MN, population, patients were excluded if
their first TIA was identified after a stroke had already
occurred, to avoid a bias in follow-up. The resulting
probability of stroke (on an actuarial basis) was 8% in
the first month and 5% per year for the next 3 years
and about 3% per year subsequently 1341. This corresponds well with the findings from the clinical trial of
aspirin and sulfinpyrazone, in which patients who received placebo had strokes after entry at the rate of
about 5% per year for a 2Y2-year period 151. These
latter data do not confirm the relatively higher probability of stroke in the first month, but patients were
rarely entered into that trial within the first month.
These rates do not seem to be age-dependent once
TIA has occurred EJ. P. Whisnant, unpublished data,
These observations do not consider the angiographic
characteristics of the patients. Patients with TIAs who
have normal carotid angiograms still have an increased
frequency of stroke when compared to the expected
rate 1171, but whether such patients have a different
frequency of stroke than those with various degrees of
carotid stenosis has not been adequately tested.
In the randomized U.S. cooperative study, 79 patients with carotid TIAs and angiographically defined
surgical lesions did not undergo surgery. These patients had stroke after entry at a rate of slightly less
than 6% per year for 3 years, after which the rate was
less than 2% per year [l21, rates that are similar to
those observed in the population-based study, following the first month after the TIA 1341. These patients
all had angiographically defined lesions and were considered good surgical candidates, but the degree of
stenosis that allowed randomization to surgical or medical treatment was as little as 30% or more 112).
These probabilities of stroke have to be assessed
after considering the risk of performing arteriography
to define the surgical lesion. In a study of TIAs occurring in the hospital, with most of the arteriograms
involving transfemoral catheterization, there was less
than 1% risk of stroke related to arteriography 1251.
Degree of Carotid Stenosis and Outcome
It is widely believed that patients with TIAs have an
increasingly greater probability of stroke as the carotid
stenosis or ulceration worsens. Because angiography
has been used to search for carotid stenosis and ulceration in screening patients for carotid endarterectomy,
a patient with severe stenosis has usually had surgery
following the arteriogram, if the patient was a reasonable surgical risk. Therefore, there is limited information about the natural history of patients with TIAs
who have severe carotid stenosis and ulceration.
In a study of TIAs in a group of hospitals 1251,
about 80% of the patients with definite carotid TIAs
had carotid atherosclerotic lesions on the appropriate
side, with about half exhibiting more than 50% stenosis and about 20% being occlusions; the others had
less than 50% stenosis. These data do not allow any
conclusions to be drawn concerning outcome with or
without surgery, based on the severity of the carotid
occlusive disease. It also is not possible to estimate a
differential stroke probability on the basis of a varying
Health Care Issues: Carotid Endarterectomy in Patients with TIAs
degree of carotid stenosis. However, higher degrees of
carotid stenosis are associated with more advanced
pathological changes in the carotid arteries, such as
ulceration and hemorrhage into a plaque 116).
Several studies have shown that cerebral ischemic
symptoms are more likely in patients with higher degrees of carotid stenosis, as determined by ultrasound
imaging and Doppler flow studies of the carotid arteries [ l S , 21). These findings are difficult to interpret
because many of the patients with stenosis had undergone endarterectomy, so they were not appropriate for
follow-up and there was a low incidence of new ischemic events from which to make a judgment. A study
of patients with asymptomatic cervical bruits showed
an increasing incidence of cerebral ischemic symptoms
with increasing degrees of carotid stenosis [6J
Mortality and Stroke Morbidity Related
to Endarterectomy
Many publications have considered the operative risk
of carotid endarterectomy. There are several examples
of clinical series in which the mortahty at surgery or
within the first 30 days thereafter was 1%) or less and
the additional stroke morbidity during that time was
3% or less, considering stroke as a persistent focal
neurological deficit of more than 24 hours' duration [4,
19, 20, 26, 331. It is not possible to determine whether
these are representative results from the best centers
or whether they represent reporting bias-that is, only
the institutions with the best results reporting their
By the same token, there are examples of community-wide studies in which combined surgical mortality
and stroke morbidity range from 10 to 21% 12, 111. A
recent audit of 46 institutions with over 3,300 endarterectomies indicated a 6% combined risk of stroke
and death within the first 30 days of surgery 114). It
was noted in a National Hospital Discharge S w e y [7]
that 2.8% of patients died following carotid endarterectomy before hospital discharge. Since stroke morbidity after endarterectomy is usually about three
times the surgical mortality, this may represent a combined mortality/stroke morbidity of greater than 10%.
Several factors appear to increase the mortality and
stroke morbidity related to carotid endarterectomy in
selected patients. The frequency of stroke is higher
among patients who have severe contralateral occlusive
carotid disease and in those who have severe stenosis
in the carotid siphon { l 5 , 2 4 ] .Stroke also occurs more
frequently when the neurological deficit is unstable
[ l S , 24). Mortality is higher among patients who have
symptomatic coronary artery disease, severe hypertension, and chronic obstructive pulmonary disease 124).
Therefore, patient selection for endarterectomy plays
an important role in the immediate surgical results.
The type of repair at the time of endarterectomy
74 Annals of Neurology Vol 22 No 1 July 1987
may influence the operative and early postoperative
mortality/stroke morbidity. In the multicenter audit,
patients who had vein patch grafting did better than
did those with fabric repair or those with primary closure [141. Also, patiients who had electroencephalographic (EEG) monitoring had less operative and postoperative mortality/stroke morbidity than did those
who did not have EBG monitoring [ l 3 ) . The skill of
the surgeon may be the overriding consideration, but
this can be quantified only by the surgeon's individual
mortality/stroke morbidity after endarterectomy.
The risk from carotid endarterectomy varies greatly
in the United States, but the overall combined mortality and stroke morbidity is probably between 6 and
Long-term Probability of Stroke
after Endarterectomy
Most reports do not allow one to assess the long-term
probability of stroke on an actuarial basis after carotid
endarterectomy for 'HA. An incidence of stroke of
2% per year has been noted following successful carotid endarterectomy in one clinical series [33), which
is not necessarily representative. About two-thirds of
the strokes in follow-up occurred on the side of
operation. Some of the factors that may influence the
long-term stroke rate after endarterectomy are related
to surgical technique and therefore the likelihood of
re-stenosis. The stroke rate after recovery from endarterectomy is low enough so that it is difficult to judge
the effect of techniques such as vein patch angioplasty
in preventing subsequent cerebral ischemic symptoms.
Re-stenosis is said to occur at a rate of about 3% per
year, but only a small fraction of those who develop restenosis have associated symptoms, and others who
develop symptoms do not have re-stenosis 17).
Strokes that occur during follow-up after endarterectomy generally haw the same variation in degree of
severity as strokes thatt occur within the first 30 days of
surgery and also the same variation of severity as those
in patients who do not have surgery [33].
Assessment of the value of endarterectomy largely depends on whether the surgical mortality and stroke
morbidity is low enough to be balanced relatively early
in follow-up by the (cumulative rate of stroke occurrence in patients with unoperated carotid stenosis.
There appears to be great variation in surgical mortality and stroke morbidity with endarterectomy; therefore, the effect of the procedure must be judged in
relation to the risk of the surgery in the hospital where
the operation is performed. Patient selection for endarterectomy plays an important role in the results following surgery.
If we presume a stroke rate of 5% per year after the
the stroke rate is probably higher in the first month
after onset. However, this higher stroke rate in the
......10% surgical mortalitylstroke morbidity
first month has been noted in only one population-- 6%surgical monalityistroke morbidity
based study of fewer than 200 patients 1341. Further-.- 4% surgical monalityistroke morbidity
more, the risks of endarterectomy could be higher in
the first month after the onset of TIA, but there are no
data upon which to judge this consideration. There
.... ....._..._...........
may be other patients with a very high probability of
1 0 1.....,............ ........... . .............
_____.-.--stroke who would benefit from surgery, more so than
noted herein, but such patients cannot be detected
based on published information.
Carotid endarterectomy may be of value, provided
Years after carotid endarterectorny
the procedures are performed with a very low surgical
complication rate. No clinical trial has addressed adeEstimated surgical mortality and cumulatizie probability of stroke
quately the benefit or lack of benefit of the procedure.
following carotid endarterectomy, comparing expected rates to
It is possible that the net effect of carotid endarterecthose associated with various percentages of surgical risk. The
tomy in patients with carotid TIAs in the United States
following assumptions were made: (1) expected rate starts more
is unfavorable. Even the best surgical results that have
than 1 month after onset of the transient ischemic attack (TIA);
been published probably have produced about a 33%
(2)long-term suwiiial in patients with TIA is not affected by
treatment; (31 postendzrterectomy stroke rate is 2% per year; and
reduction in stroke in 5 years.
(4)outcome at 1 month includes mortality and stroke morbidity
on the aLy of surgery or within 30 days thereafter.
- Expected probability of stroke
first month in untreated or medically treated patients
in the first 3 years and 3% per year thereafter, a combined surgical mortalitylstroke morbidity of 4%, and a
stroke rate of 2% per year following the first 30 days
after endarterectomy, then the negative effect of the
surgery is balanced after a little more than 1 year,
yielding a 33% reduction in stroke at the end of 5
years (Figure). If the surgical mortality/stroke morbidity is lo%, there would be little difference in the outcome at 5 years, but with the negative effect of the
early mortalitylmorbidity (see Figure). At a 6% surgical mortality/stroke morbidity, the stroke probability is
balanced at 2 years, but this would result in a less than
25% reduction in stroke at the end of 5 years. The
surgical mortaliqdstroke morbidity would probably
have to be less than 1% to effect a 50% reduction in
stroke at 5 years of follow-up.
The Figure and the interpretation of the rates must
be taken as a very crude guide for the following reasons: (1) reports of patients with TIAs who were untreated or who had medical or surgical treatment are
variable, indicating that there may be substantial variability in outcome among different patient populations
and among different treatment centers; (2) the individual estimates are subject to statistical fluctuations associated with their sample sizes; and ( 3 ) the results of
surgical and medical treatment came from different
populations, without adequate data for multivariate adjustment of likely different baseline characteristics.
If endarterectomy is performed within the first
month after the first TIA, it may provide a greater
differential benefit than it would at a later time because
1. Bauer IU3, Meyer JS, Fields WS, et al: Joint study of extracranial
arterial occlusion. 111. Progress report of long-term survival in
patients with and without operation. JAMA 2081509-5 18,
2. Brott T, Thalinger K The practice of carotid endarterectomy in
a large metropolitan area. Stroke 15950-955, 1984
3. Brott TG, Labutta RJ, Kempczinski RF. Changing patterns in
the practice of carotid endmerectomy in a large metropolitan
area. JAMA 255:2609-2612, 1986
4. Browse NL, Ross-Russell R: Carotid endarterectomy and the
Javid shunt: the early results of 215 consecutive operations for
transient ischaemic attacks. Br J Surg 71:53-57, 1984
5. Canadian Cooperative Study Group: A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med,
299:53-59, 1978
6. Chambers BR, Norris JW: Outcome in patients with asymptomatic neck bruits. N Engl J Med 315:860-865, 1986
7. Colgan MP, Kingston V, Shanik DG: Stenosis following carotid
endarterectomy. J Cardiovasc Surg 261300-302, 1985
8. Dyken ML: Carotid endarterectomy studies: a glimmering of
science. Stroke 17:355-358, 1986
9. Dyken ML, Pokras R The performance of endarterectomy for
disease of the extracranial arteries of the head. Stroke 151948950, 1984
10. Eastcotr HHG, Pickering GW, Rob CG: Reconstruction of internal carotid artery in a patient with intermittent attacks of
hemiplegia. Lancet 2994-996, 1954
11. Easton JD, Sherman DG: Stroke and mortality rate in carotid
endmerectomy: 228 consecutive operations. Stroke 8:565568, 1977
12. Fields WS, Lemak NA: Joint study of extracranial arterial occlusion. IX. Transient ischemic attacks in the carotid territory.
JAMA 235:2608-2610, 1976
13. Fields WS, Maslenikov V, Meyer JS, et al: Joint study of extracranial arterial occlusion. V. Progress report of prognosis following surgery or nonsurgical treatment for transient cerebral
ischemic attacks and cervical carotid artery lesions. JAMA
211~1993-2003, 1970
14. Fode NC, Sundt TM, Robertson JT, er al: Multicenter retrospective review of results and complications of carotid endarterectomy in 1981. Stroke 17:370-376, 1986
Health Care Issues: Carotid Endarterectomy in Patients with TIAs
15. Imparato AM, Ramirez A, Riles T, Mintzer R: Cerebral protection in carotid surgery. Arch Surg 117:1073-1078, 1982
16. Imparato AM, Riles TS, Mintzer R, Bauman FG: The importance of hemorrhage in the relationship between gross morphologic characteristics and cerebral symptoms in 375 carotid
artery plaques. Ann Surg 197:195-203, 1983
17. Marshall J. Wilkinson IMS: The prognosis of carotid transient
ischaemic attacks in patients with normal angiograms. Brain
94~395-402, 1971
18. Moore DJ, Sheehan MP, Kolm P, et ak Are strokes predictable
with non-invasive methods: a five-year follow-up of 303 unoperated patients. J Vasc Surg 2:654-660, 1985
19. Nunn DB: Carotid endmerectomy: an analysis of 234 operative
cases. Ann Surg 182:733-738, 1975
20. Ojemann RG, Crowell RM, Roberson GH, Fisher CM: Surgical
treatment of extracranial carotid occlusive disease. Clin NeuroSurg 22:214-263, 1975
21. Roederer GO, Ianglois YE, Jager KA, et al: The natural history
of carotid arterial disease in asymptomatic patients with cervical
bruits. Stroke 15:605-613, 1984
22. Shaw DA, Venables GS, Cartlidge NEF, et al: Carotid endarterectomy in patients with transient cerebral ischaemia. J Neurol
Sci 64:45-53, 1984
23. Simonsen N, Christiansen HD, Heltberg A, et al: Long-term
prognosis after transient ischemic attacks. Acta Neurol Scand
63:156-168, 1981
24. Sundt TM, Jr, Sandok BA, Whisnant J P Carotid endmerectomy: complications and preoperative assessment of risk. Mayo
Clin Proc 50:301-306, 1975
25. Swanson PD, Calanchini PR, Dyken ML, et al: A cooperative
study of hospital frequency and characrer of transient ischemic
76 Annals of Neurology Vol 22 No 1 July 1987
attacks. 11. Performance of angiography among six centers.
JAMA 237: 2202-2206, 1977
26. Thompson JE, Austin DJ, Patman RD. Carotid endmerectomy
for cerebrovascular insufficiency: long-term results in 592 patients followed up to thirteen years. Ann Surg 172:663-679,
27. Toole JF, Janeway R, Choi K, et al: Transient ischemic attacks
due to atherosclerosis: A prospective study of 160 patients.
Arch Neurol 32:5-12, 1975
28. Toole JF, Yuson CP, Janeway R: Transient ischemic attacks: a
prospective study of 225 patients. Neurology 28:746-753,
29. Warlow C , Whisnant JP, Dyken ML Carotid endarterectomy.
In Plum F, Pulsinelli 'XI (eds): Cerebrovascular Diseases. New
York, Raven, 1985
30. Whisnant JP: Morbidirf and mortality in patients with stroke. In
Transactions of the Association of k f e Insurance Medical Directors of America. 63:86-95, 1979
31. Whisnant J P The rob: of the neurologist in the decline of
stroke. Ann Neurol 14:l-7, 1983
32. Whisnant JP, Matsumcito N, Elveback LR.Transient cerebral
ischemic attacks in a community: Rochester, Minnesota, 1955
through 1969. Mayo Clin Proc 48:194-198, 1973
33. Whisnant JP, Sandok BA, Sundt TM: Carotid endarterectomy
for unilateral carotid system transient cerebral ischemia. Mayo
Clin Proc 58:171-175, 1983
34. Whisnant JP, Wiebers IDO: Clinical epidemiology of transient
cerebral ischemic attacks (TIA) in the anterior and posterior
cerebral circulation. In Sundt TM Jr: Occlusive Cerebrovascular
Disease: Diagnosis anal Surgical Management. Philadelphia,
Saunders, 1987, pp 60-65
Без категории
Размер файла
511 Кб
decemberrease, stroki, doesn, death, carotid, patients, transiente, ischemia, attack, endarterectomy
Пожаловаться на содержимое документа