0148-396 X/86/1803-0283$02.00/0
NEUROSURGERY
Copyright © 1986 by the Congress of Neurological Surgeons

Vol. 18, No. 3, 1986
Printed in U.S.A.

Potential Roles for Early Revascularization in Patients with
Acute Cerebral Ischemia

Hunt Batjer, M.D., Bruce Mickey, M.D., and Duke Samson, M.D.

Division of Neurosurgery, Department of Surgery, University of Texas Health Science Center, Dallas, Texas

The risk of carotid endarterectomy in acute cerebral ischemic situations is well documented. By using the superficial
temporal or occipital artery, it is possible to provide relatively low flow revascularization conduits, potentially avoiding
the risk of postoperative hemorrhage. Eighteen patients at The University of Texas Health Science Center at Dallas,
Texas, have been treated recently with extracranial to intracranial bypass in the setting of acute neurological deficit,
stroke in evolution, or recent completed infarction. Angiographic causes of these deficits included cervical carotid
occlusion in 5 patients, carotid siphon disease in 2 patients, middle cerebral stenosis or occlusion in 3 patients, and
delayed cerebral ischemia following subarachnoid hemorrhage in 8 patients. Neurological improvement was demon-
strated within 24 hours following revascularization in 15 cases (83%); 3 patients were unchanged following bypass, and
no patient's condition was worsened. The only case of postoperative intracerebral hemorrhage occurred | week following
a long saphenous vein graft from the subclavian to the middle cerebral artery. After an average follow-up of 19 months,
7 patients are neurologically normal, 8 patients have mild to moderate deficits, and 3 patients have died. Low flow
revascularization procedures appear to be safe in the setting of acute cerebral ischemia and may in selected patients
boost regional cerebral blood flow from levels of symptomatic ischemia into a range compatible with normal neuronal

function. (Neurosurgery 18:283-291, 1986)

Key words: Cerebral ischemia, Cerebral revascularization, Subarachnoid hemorrhage

The role of cerebral revascularization in the treatment of
acute cerebral ischemia has been incompletely investigated.
The major thrust in cerebral vascular reconstructive surgery
has been toward the prevention of future neurological events
ina variety of clinical settings. There is considerable literature
regarding carotid bifurcation endarterectomy suggesting that
this type of revascularization be avoided in patients with acute
neurological deficits and progressing stroke (stroke in evolu-
tion) due to the high risk of increasing the deficit and intra-
cranial hemorrhage (5. 20, 53, 57). Following the work of
Yasargil (58) and Donaghy (17) in 1967, it has become
possible through extracranial to intracranial (EC-IC) arterial
anastomoses to augment collateral flow to a variety of vascular
distributions. Many of the potential indications for this pro-
cedure remain controversial. Despite the fact that routine
superficial temporal to middle cerebral anastomosis provides
only approximately 35 cc/minute of flow in the acute phase
(2, 3. 44). it is probable that a successful anastomosis can
augment middle cerebral artery distribution flow by 15 to
20% (43). It is well known that these relatively modest early
increases in flow can be followed by dramatic augmentations
several weeks or months later (12, 25, 59). The technical
feasibility of this procedure. together with more sophisticated
knowledge of cerebral vascular hemodynamics, provides a
provocative foundation for its investigation in the setting of
acute cerebral ischemia. The use of low flow conduits, such
as the superficial temporal artery, should not subject patients
to the hemorrhagic complications accompanying such high
flow revascularization procedures as carotid endarterectomy.
middle cerebral embolectomy. or long saphenous vein bypass
grafts. Eighteen patients have been treated recently at The
University of Texas Health Science Center at Dallas with EC-
IC anastomoses in the setting of acute cerebral ischemia or
stroke in evolution due to a variety of anatomical lesions.

MATERIALS AND METHODS

For the purposes of this discussion the term “completed
stroke” refers to a fixed neurological deficit present for more

than 24 hours. “Acute deficit” is defined as a fixed neuro-
logical deficit observed less than 24 hours following its onset.
“Stroke in evolution” is defined as a focal hemispheral
neurological deficit that fluctuates or progresses over an inter-
val of more than 6 hours. Clinical data from this series are
presented in Table 1. Twelve patients demonstrated a clinical
pattern of stroke in evolution, 5 patients were seen after the
development of an acute neurological deficit, and | patient
was referred to our institution several days after the onset of
acompleted stroke. She was found to have bilateral enhancing
infarcts on her initial computed tomography (CT) scan. Pre-
operative CT scans were performed without enhancement in
most cases and demonstrated low density lesions in 7 patients,
enhancement in | patient, and normal findings in 10 cases.
In | patient, preoperative CT scanning was not done due to
the need for urgent operation for cervical carotid erosion.
Five patients were found to have cervical internal carotid
artery occlusion; two of these were iatrogenic. Internal carotid
artery siphon disease was present in 2 patients. Middle cere-
bral‘artery stenosis or occlusion was symptomatic in an
additional 3 patients. The final diagnostic group was com-
prised of 8 patients symptomatic from delayed cerebral ische-
mia due to vasospasm following subarachnoid hemorrhage.
Six of these cases of subarachnoid hemorrhage presented with
a clinical picture of stroke in evolution, and 2 patients had
an acute severe neurological deficit, producing coma in | case.
The common angiographic appearance of narrowing in the
proximal conducting vessels with poststenotic dilatation of
M-2 and M-3 branches, sluggish middle cerebral circulation,
and the consistently observed reversal of neurological deficits
with therapeutic hypervolemia and hypertension led us to
consider EC-IC bypass in an attempt to augment collateral
flow in this condition. Revascularization was reserved for
patients who failed medical therapy or had unsecured contra-
lateral aneurysms, in which case we were reluctant to induce
hypertension.

Most patients in this series were treated preoperatively with
volume expansion and induced hypertension, but none were

284 BATJER et al.

Neurosurgery, Vol. 18, No. 3

TABLE |
Clinical Data*

Preop CT

; Al TT -
Neuro/Response (24 Final outcome

inical:Presenta ,
Clinical Presentation Cause ischemic hr postop) ral
Case rape a “ery Si Interval - a cee
Acute SYK Compl jg) Low Enhance. Cr Siphon MCA SAH/ “(a)” Hypervolemia/ pace Normal Mild/ severe Dead UP “™
deficit stroke O"™" dens ment sten/ sten/ symp hypertension PASS: NOTMAL mod PEvEre ea
evol occl ocel occl vasosp
1 et + + 7 ty + + 14
2 oe = Eo 0.5 « + + I
3 te oH +63 fe 9 is 2 + I
4 + + + 0.42 No No + int
5 + + + 1 7 + + 63
6 + + ot 1.5 No + + Died
7 + oa LS No + + Died
8 + + + 1 + + + 8
9 + Not done aH 0.17 a + + Died
10 + + + 0.25 a + 75
ia Bd * id 0.5 i + + 11
12 + + + 2 ? + + 63
13 ty aS + 0.35 + + Sf 4
14 + + + 1.0 @ +: aa 3
15 oe + + 0.75 4 + + 3
16 + + + 1.0 + = + 2
17 + + + 0.33 ? + + 2
18 + + + 5 + + + 2

— * Abbreviations: cerv, cervical: compl, completed; CT, computed tomography; dens, density; evol, evolutiot

; ICA, internal carotid artery; mod, moderate; neurol,

neurological: occl, occlusion: postop, postoperative: preop, preoperative; sten, stenosis; symp, symptomatic; vasosp, vasospasm.

anticoagulated. Fourteen patients underwent a total of sev-
enteen superficial temporal to middle cerebral artery bypass
procedures. Three patients were treated with short saphenous
interposition grafts from the occipital artery or proximal
superficial temporal artery to a cortical middle cerebral artery
branch due to inadequate extracranial donor vessels. One
patient underwent a long saphenous interposition graft from
her subclavian artery to a branch of her middle cerebral artery.

RESULTS

In order to evaluate the results of medical as well as surgical
therapy in this series, patients have been graded according to
the severity of their neurological deficits. Mild to moderate
hemispheral deficit has been defined as the presence of at least
antigravity strength in the involved extremities or a speech
deficit that does not result in either total loss of speech output
or comprehension. Severe hemispheral deficit is defined as
less than antigravity strength, hemiplegia, or, in the case of
dominant hemisphere involvement, either absent speech out-
put or complete lack of comprehension. Altered level of
consciousness refers to lethargy or obtundation. After the
institution of medical therapy (hypervolemia and hyperten-
sion), patients were regarded as neurologically improved if
there was either resolution of their prior deficit or improve-
ment in a fixed deficit. Surgical patients were evaluated 24
hours following the reversal of their anesthetic and were
considered neurologically improved if their deficit had re-
solved, they showed improvement in their prior fixed deficit,
or their neurological status could be maintained despite low-
ering their mean arterial blood flow pressure to previously
nontolerated levels while weaning pressor agents (Table 2).

For the series as a whole, neurological improvement was
demonstrated within 24 hours after bypass in 15 patients
(83%). Three patients were unchanged after their bypass
procedure, and no patient’s condition was worsened by op-
eration. Five patients at least transiently showed neurological
improvement following the preoperative institution of thera-
peutic hypertension and volume expansion. Four of the 5
patients who showed some response to medical therapy also
responded to bypass. Two interesting patients (Cases 6 and 7)

clearly did not respond to aggressive medical therapy, but did
respond quickly to their bypass procedure. Case 6 was a
patient who became symptomatic following subarachnoid
hemorrhage before his aneurysm had been secured; we were
reluctant to elevate his systolic pressure higher than 170 mm
Hg. Case 7 also had suffered a subarachnoid hemorrhage and
at Day 7 very rapidly became decorticate due to severe
bilateral vasospasm. The mean duration of neurological defi-
cit prior to bypass procedure in these 18 patients was 44
hours; the range was from 4 hours to 9 days. Eleven of the 18
patients underwent microanastomosis within 24 hours follow-
ing the onset of symptoms. As shown in Table 1, with an
average follow-up of 19 months, 7 patients are neurologically
normal, 8 patients have mild to moderate neurological defi-
cits, and 3 patients have died.

The clinical presentation of stroke in evolution was associ-
ated with the highest rate of favorable response to the bypass
procedure, as well as favorable final outcome. Eleven of the
12 patients with this presentation were judged neurologically
improved within 24 hours following bypass (Table 3). Seven
of these 12 ultimately made a normal recovery. Four of the 5
patients with acute neurological deficits responded to the
bypass procedure, but at final follow-up 3 of the patients were
left with mild to moderate neurological deficits, and 2 had
died. Table 4 illustrates the relationship between angiographic
diagnosis, response to bypass, and eventual outcome. All 5
patients with cervical internal carotid artery occlusion dem-
onstrated neurological response to their bypass procedure,
and 2 eventually made a normal recovery. Similarly, 7 of the
8 patients with subarachnoid hemorrhage responded at 24
hours to bypass. As is shown in Table 5, in this operative
series by far the most favorable outcomes after bypass proce-
dure are those of patients with a preoperatively normal CT
scan. All 10 of these patients were neurologically improved
within 24 hours following bypass. Six of these patients re-
covered normally, 2 were left with mild to moderate neuro-
logical deficits, and 2 ultimately died. Of the 7 patients who
preoperatively had low density lesions on their CT scan, only
4 responded to the bypass procedure initially, and 6 of these
7 were left with mild to moderate neurological deficits. The
patient who was treated following a completed stroke who

March 1986

TABLE 2
Neurological Improvement

No. Patients Improved

Maintenance of

Therapy Resolution Improvement __ neurological
Achat of fixed deficit status on lower
MABP
Medical (hypervo- 2 3
lemia, hyperten-
sion)
Surgical (bypass)? 7 7 L
“Result at 24 hours.
TABLE 3
Clinical Presentation
No. Patients
Response
to
Presentation Bypass Outcome
at 24
hr

+ — Normal Mild/moderate Dead
Acute deficit 41 0 3 2
Stroke in evolution 11 1 7 4 1
Completed stroke | 0 i] 0

TABLE 4

Relationship between Angiographic Diagnosis, Response to Bypass,
and Outcome

No. Patients

Response

Angiographic
Poe aphic to Outcome
Diagnosis .

bypass

+ — Normal Mild/moderate Dead
Cervical ICA occlusion 5 0 2 2 1
ICA siphon stenosis/oc- 1 1 1 1 0

clusion

MCA stenosis/occlusion 2 1 2 1 0
SAH 71 2 4 2

had bilaterally enhancing CT lesions did not respond to bypass
and was left with a mild to moderate deficit. The interval
between the onset of neurological symptoms and the estab-
lishment of a bypass conduit did not appear to be related to
ultimate outcome. For the 7 patients who ultimately re-
covered normally, the mean ischemic interval was 39.8 hours.
For the 8 patients who recovered with mild to moderate
neurological deficits, the mean ischemic interval was 55 hours.
For those 3 patients who died. the mean period of ischemia
was 25.4 hours. The severity of the preoperative neurological
deficit was tightly linked to ultimate outcome (Table 6). Of
the 10 patients with preoperative mild to moderate deficits, 7
ultimately made a normal recovery. All five patients who
preoperatively demonstrated severe hemispheric deficits were
left with mild to moderate disabilities. Two of the 3 patients
who preoperatively were found to have altered levels of con-
sciousness died.

The majority of patients in this series demonstrated a
clinical pattern of stroke in evolution. Due to the fact that

EARLY REVASCULARIZATION 285

this group of 12 patients was weighted strongly in favor of
subarachnoid hemorrhage with vasospasm as the etiology of
deficit, we are reluctant to draw any general conclusions
regarding their management outcome. However, 8 of these
patients were found preoperatively to have normal CT scans.
All 8 responded initially to their bypass procedure, and 6
made normal recoveries. Four patients with clinical pictures
of stroke in evolution were left with mild to moderate neu-
rological deficit.

Of the eight patients who underwent bypass procedures
after the development of delayed cerebral ischemia due to
vasospasm related to subarachnoid hemorrhage, three are
normal, two are mildly aphasic, one has a fixed hemiparesis,
and two have died. Neither of these deaths appeared to be
related to the bypass procedure. Bypass was offered in this
interesting group of patients due to the occurrence of contra-
lateral unsecured aneurysms in five patients, failure of medical
management in two patients, and inability to elevate blood
pressure despite toxic doses of pressors in one patient. For
patients at high risk for delayed cerebral ischemia, we now
routinely plan the opening of their original craniotomy so
that the anterior branch of the superficial temporal artery is
preserved. This requires that the scalp incision be carried
slightly posterior and slightly more medial than in a classically
described pterional craniotomy. For the aneurysm procedure,
we reflect the scalp in a single layer. If the patient returns to
the operating room for bypass, the original scalp incision is
opened and reflected in the subgaleal plane. Following this,
the anterior branch of the superficial temporal artery is dis-
sected out to a length that is measured to reach the Sylvian
fissure. A small amount of the temporal squama is then
removed, enlarging the original bony exposure. The distal
Sylvian fissure is dissected until an acceptable recipient middle
cerebral branch is isolated. The amount of brain fullness

TABLE 5
Relationship between Preoperative CT, Response to Bypass, and
Outcome
No. Patients
Response
Preoperative “a
CT bypass Outcome
at
24 hr
+ — Normal Mild/moderate Dead
Normal 10 0 6 2 2
Low density 4 3 1 6 0
Enhancing oO ol 0 1 0
TABLE 6
Severity of Preoperative Deficit Related to Outcome
No. Patients
Severity of —= — — —
Preoperative Deficit Normal Mild/moderate Dead
outcome deficit
Mild/moderate hemispheral 7 2 1
deficit
Severe hemispheral deficit 0 5 0
Altered level of conscious- 0 1 2

ness

286 BATJER et al.

encountered has not posed much of a technical disadvantage,
and postoperative intracerebral hemorrhage has not occurred.

Grafts were demonstrated to be patent by postoperative
angiography in 15 cases and by Doppler ultrasound in 2 cases.
Postoperative angiography was almost always performed
within the Ist week to 10 days postoperatively, before full
graft maturity: this prevents significant observations regarding
quality and quantity of flow. A superficial temporal graft in
Case 16 that was performed following subarachnoid hemor-
rhage with diffuse distal middle cerebral territory spasm was
demonstrated to be nonpatent on follow-up angiography, The
only instance of postoperative hemorrhage occurred following
the use of a long saphenous vein graft.

Three patients in this series died. The first patient (Case 6)
had undergone a right subtemporal approach to a very low
lying ruptured basilar bifurcation aneurysm (middle one-third
of clivus) at another institution. His aneurysm was not secured
because of difficulty in visualizing the contralateral P-1. On
the day of his arrival in our institution, he developed a dense
right hemiparesis. Due to severe internal carotid and middle
cerebral spasm on the left and the presence of an unsecured
aneurysm, he underwent emergency left superficial temporal
to middle cerebral artery bypass. His neurological deficit
improved, and 2 days later he underwent reoperation through
a right subtemporal approach. An aperture Sugita clip was
applied without good demonstration of the contralateral P-1.
Angiography under the same anesthetic revealed a very small
amount of residual neck remaining patent. It was decided to
operate again and apply a second distal clip for definitive
aneurysm closure. The patient awakened well, but 12 hours
later suddenly deteriorated. Reexploration revealed a swollen
and necrotic temporal lobe. Despite lobectomy, the patient
was left with a dense left hemiparesis. He died 3 months later
from a pulmonary embolus. The second mortality (Case 7)
also presented in Hunt-Hess Grade III condition after sub-
arachnoid hemorrhage. His anterior communicating artery
aneurysm was clipped on Day | postsubarachnoid hemor-
rhage. On Day 7, the patient suddenly became bilaterally
decorticate. Angiography demonstrated severe bilateral vaso-

subclavian to the middle cerebral artery.

1. Case 9, A 50-year-old-woman with erosion of left carotid bifurcation after a radical neck operation and radiation therapy.
osterior (1) and lateral (B) angiograms disclose severe stenosis at the anastomotic site after long saphenous vein grafting from the

Neurosurgery, Vol. 18, No. 3

spasm with extremely sluggish filling of his carotid circulation
bilaterally, After failure to respond to volume expansion or
hypertension, he underwent bilateral EC-IC bypass. Within
24 hours, he responded purposefully bilaterally, but he never
improved beyond that point and later died of pulmonary
complications. The final case (Case 9) suffered erosion of her
left common carotid artery following radical neck surgery and
radiation therapy. She presented to the emergency room with
a mild right hemiparesis. She underwent emergency subcla-
vian to middle cerebral artery long saphenous interposition
grafting and awakened from the procedure neurologically
normal. Angiography on the 7th postoperative day revealed
excellent filling of her entire middle cerebral distribution
through the graft but severe stenosis at the anastomotic site
(Fig. 1). Regrettably, exploration was done in an attempt to
prevent graft closure. She was found to harbor an intrapar-
enchymal temporal hematoma, and graft sacrifice was re-
quired in order to obtain hemostasis. She developed a dense
hemiplegia and later died of gram-negative meningitis.

ILLUSTRATIVE CASES
Case 11

This patient is a 76-year-old, right-handed man who pre-
sented with a several-week history of multiple left hemispheric
TIAs. He initially was treated in an outlying hospital where
bilateral carotid arteriograms were obtained. The patient had
mild cervical carotid disease bilaterally, right greater than left,
and left middle cerebral artery stenosis (Fig. 2). He underwent
a right carotid endarterectomy from which he made an un-
eventful recovery. Over the ensuing 2 days, the patient expe-
rienced episodic hypotension that produced in each case the
onset of aphasia and a right hemiparesis. These findings
resolved with reinstitution of normotension. On the day of
his ultimate transfer to our hospital, he had a similar episode
that did not resolve with normotension. Immediately after his
transfer, an unenhanced CT scan that was negative was ob-

March 1986

tained. He was taken to the operating room emergently; left
superficial temporal to middle cerebral artery bypass was
completed within 12 hours of the onset of this fresh neurolog-
ical deficit. Within 24 hours postoperatively, the patient’s

FiG. 2. Case 11. A 76-year-old, right-handed man with left hem-
ispheral stroke in evolution due to middle cerebral artery stenosis.

EARLY REVASCULARIZATION 287

hemiparesis had resolved; his subtle speech deficit resolved
over the next several days. He is neurologically normal at this
time and having no neurological symptoms.

Case 13

This 36-year-old, right-handed woman presented on Day |
after subarachnoid hemorrhage with a Hunt-Hess Grade II
neurological status. CT scan demonstrated diffuse and thick
subarachnoid blood. Angiographically, she was found to have
a right posterior carotid wall aneurysm (Fig. 34). She was
taken to the operating room on Day 2 postsubarachnoid
hemorrhage, and her aneurysm was uneventfully clipped. She
remained neurologically normal and underwent a routine
postoperative arteriogram on Day 5 that disclosed moderately
severe vasospasm in her right internal carotid and M-! distri-
bution (Fig. 3B). On Day 6, the patient suddenly developed
a dense left hemiparesis. The patient was immediately vol-
ume-expanded, and hypertension was induced with pressors.
At a blood pressure of 180 systolic her fixed neurological
deficit persisted, but it was partially reversible at a systolic
pressure of 200. She was taken to the operating room at that
time and underwent anastomosis between the anterior branch
of the superficial temporal artery, which had been preserved
at her original craniotomy, and her middle cerebral artery.
The anastomosis was completed about 8'2 hours following
the onset of her deficit. Within 12 hours postoperatively, it
was apparent that the patient’s arm drift had disappeared, and
her leg weakness steadily improved over the next 48 hours.
She was quickly weaned from the pressors that had been used
postoperatively to maintain her blood pressure at about 150
to 160 systolic. She was discharged 7 days following her bypass
procedure neurologically intact and has remained so.

Fic. 3, Case 13. A 36-year-old. right-handed woman with subarachnoid hemorrhage. .4, lateral right carotid angiogram demonstrates a

posterior carotid wall aneury
middle cerebral artery spasm. which became symptomatic on Day 6.

sm on Day | after subarachnoid hemorrhage. B. Day 5 postsubarachnoid hemorrhage. Patient now has significant

288 BATJER et al.
DISCUSSION

Evaluation of the efficacy of acute surgical intervention in
the treatment of syptomatic cerebral ischemia is rendered
difficult by the inherent variability of the natural history of
transient ischemic episodes and stroke in evolution, Despite
Fisher’s belief that TIA’s resolve within 10 minutes in 70% of
cases (21), the definition of TIA in clinical usage currently
includes resolution of neurological deficit within 24 hours of
onset (37). Stroke in evolution is the preferred clinical defi-
nition if progressive worsening in neurological status occurs
over 6 to 72 hours following the onset of neurological dys-
function (54, 55). Therefore, if evaluation and therapy begin
within the first 24 hours following the onset of ischemic
neurological deficit, it becomes impossible to determine if
treatment has altered the natural history of the ischemic
process. This issue becomes even more complicated when
patients are placed under anesthesia acutely for surgical pro-
cedures, preventing evaluation of the evolution of the clinical
course over the next several hours. For this reason our diag-
nostic groupings are obviously suspect, because we have not
permitted the ischemic process in this group of patients to
remain untreated for time periods sufficient to allocate pa-
tients to the temporally defined categories of transient is-
chemic episodes, stroke in evolution, or completed stroke.

Millikan in 1973 detailed the natural history of stroke in
evolution in a series of 204 cases (36). Of these, 74% of the
patients developed permanent neurological deficits, 14% died,
and only 12% returned to normal at 2 weeks follow-up.
Millikan, as a result, recommended aggressive diagnostic and
therapeutic procedures in an effort to counter this unfavorable
outlook. Carter (10) and Baker et al. (4) have both reported
randomized trials of anticoagulation in patients with stroke
in evolution, documenting that the long term mortality in
this condition was not altered by anticoagulant therapy, but
that residual neurological deficits appeared to be somewhat
less in the treated groups of patients. In all of these studies
only a small number of patients were submitted to angiogra-
phy, reflecting a reluctance to undertake angiographic evalu-
ation of the cerebral circulation in neurologically unstable
patients for fear of exacerbating preexisting deficits. Careful
examination of the literature, however, suggests that this
concern is exaggerated. Both the Joint Study of Extracranial
Arterial Occlusion (28) and a more recent Mayo Clinic series
(18) report permanent neurological morbidity rates of 1%
and 0.63%, respectively, in patients studied with symptomatic
cerebral vascular disease. It is only after appropriate radiolog-
ical studies, including angiography and often CT scanning,
that rational decisions about therapeutic options are possible.
The plethora of potentially causative vascular lesions, as well
as the frequent occurrence of multiple or incidental vascular
anomalies, mandates early angiographic diagnosis to deter-
mine the natural history of and the therapeutic alternatives
available in ischemic cerebral vascular disease.

Traditionally, surgical revascularization procedures have
been thought to be contraindicated in the presence of acute
cerebral ischemia. Data from the Joint Study of Extracranial
Arterial Occlusive Disease documented an operative mortality
rate of slightly less than 50% in patients with acute severe
neurological dysfunction who underwent carotid endarterec-
tomy (5), and several subsequent reports have suggested that
operative mortality and morbidity figures are four to five
times greater in neurologically unstable patients than in pa-
tients without neurological deficit (7, 8, 27, 46, 47, 52, 57).
Most authors have attributed these increased mortality and
morbidity figures associated with endarterectomy in the face
of acute neurological deficit to a high incidence of postoper-
ative intraparenchymal hemorrhage. Although recently sev-

Neurosurgery, Vol. 18, No. 3

eral investigators have advocated a more aggressive approach
to the acute unstable patient with neurological deficit (exclud-
ing patients with severe neurological deficit or altered level of
consciousness) (22-24), current recommendations suggest
that emergency endarterectomy is indicated only in patients
with progressive but nondebilitating ischemia or in patients
with evidence of an acute thrombosis who can be operated
upon within several hours following the onset of deficit (35).
The etiology of intracerebral hemorrhage following endarter-
ectomy remains controversial. Established risk factors include
previous cerebral infarction, postoperative hypertension (6, 7,
34, 40, 51), frequent transient ischemic episodes prior to
surgery (6, 7), and postoperative anticoagulation (45). Caplan
et al. described two cases of delayed intracerebral hemorrhage
at 4 and 5 weeks following endarterectomy that were attrib-
uted to the long standing hypertension of the patients (9).

Normal cervical internal carotid artery flow is in the range
of 330 cc/minute (30), of which some 150 to 170 cc/minute
routinely enters the middle cerebral distribution (38, 39). This
arterial inflow is responsible for a normal regional cerebral
blood flow that approximates 50 cc/100 gm of brain tissue/
minute. However, studies have demonstrated repeatedly that
symptomatic neuronal dysfunction does not occur until re-
gional cerebral blood flow falls below 20 cc/100 gm/minute,
and neuronal death does not occur until flow is reduced to
levels of 10 to 11 cc/100 gm/minute (49, 50). Theoretically,
it should be possible through relatively modest augmentations
of middle cerebral distribution flow acutely to boost regional
perfusion from levels of symptomatic ischemia into the range
of asymptomatic oligemia, potentially salvaging marginally
perfused tissue. By performance of a superficial temporal-
middle cerebral cortical branch anastomosis, between 25 and
70 cc of blood/minute may be delivered into the middle
cerebral distribution (44). Using a thermal diffusion flow
probe, intraoperative measurements of cortical blood flow
have been obtained immediately prior to and following EC-
IC bypass and demonstrate focal increases in regional cerebral
flow of 50 to 75% (11). By contrast, long saphenous vein
grafts have been found to carry a volume as high as 120 cc/
minute (42). It would appear that the relatively small volume
of blood from routine superficial temporal anastomoses can
assume major significance in the setting of acute cerebral
ischemia without subjecting the chronically vasodilated mi-
crocirculation to the massive increase in perfusion pressure
associated with acute carotid endarterectomy.

In addition to the finding of Hayashi et al. that induced
hypertension improved outcome in a monkey experimental
stroke model (29), a number of investigators have reported
the use of EC-IC bypass in animal models of acute cerebral
ischemia. Using a transcranial approach in the canine and
macaque monkey, Crowell occluded the middle cerebral ar-
tery proximally and performed an EC-IC bypass 2 hours after
middle cerebral ligation (13). He achieved a 50% patency rate
in this model and found that some animals improved follow-
ing microanastomosis. In a similar animal model, Fein and
Molinari achieved a patency rate of 71% (19). Of 10 animals
with patent anastomoses, 5 recovered with no neurological
deficits. They concluded that in selected cases microanasto-
mosis might be beneficial. Diaz and coworkers in a canine
model of middle cerebral artery infarction performed revas-
cularization at 4 and 24 hours following middle cerebral
occlusion. Although as a group the animals with anastomoses
did worse than the controls, 3 of the revascularized animals
were significantly better than the control animals (14, 15).
These authors implied that acute revascularization appeared
to be harmful in cases of large infarction but were of potential
benefit in animals with smaller infarcts.

Numerous electrophysiological and imaging modalities

March 1986

have been employed in the clinical setting of human cerebral
ischemia to aid in selection of patients who might benefit
from acute revascularization. Yonekawa and Yasargil in 1976
ascribed little importance to electroencephalograph (EEG)
monitoring for localization of focal ischemia or for prediction
of reversibility (60). Holbach et al. suggested that improve-
ment in EEG patterns under the influence of hyperbaric
oxygenation might reflect some evidence of reversibility (31).
Ito et al. have suggested the use of evoked potentials following
median nerve stimulation in hemiparetic limbs for the detec-
tion of functional reversibility following complete stroke (32).
In his monograph Kletter discusses the use of cerebral scintig-
raphy with Tc to determine blood-brain barrier integrity as
a measure of safety of revascularization (33). Perhaps the
most provocative modality to suggest reversible ischemia is
the measurement of regional cerebral blood flow (rCBF) with
xenon-133 inhalation and single photon emission tomogra-
phy. Vorstrup investigated a patient who had suffered a minor
infarction who subsequently developed orthostatic TIAs (56).
Angiographically, this patient had a critical stenosis of his
symptomatic internal carotid artery. Two weeks following his
clinical infarction, his CT scan and brain scintigraphy were
both normal. Resting cerebral blood flow (CBF) measurement
demonstrated an area of hypoperfusion corresponding to the
involved middle cerebral territory. This area of low perfusion
was felt to represent either irreversible ischemic tissue damage
or potentially reversible hemodynamic insufficiency and
pharmacologial manipulations were employed to distinguish
between these two possibilities. Following the administration
of acetazolamide, known to increase CBF in a fashion similar
to hypercapnia, rCBF increased in the asymptomatic hemi-
sphere coincident with a relative decrease in the symptomatic
hemisphere, suggesting a steal phenomenon. By contrast,
theophylline, a cerebral vasoconstrictor, caused a reduction
in CBF in the unaffected hemisphere and a relative increase
in flow in the previously hypoperfused area, consistent with
an inverse steal. These paradoxical responses were interpreted
as indicating failure of cerebral vasoreactivity and maximal
vasodilation in a region subjected to chronic marginal perfu-
sion. Following carotid endarterectomy, the paradoxical flow
responses to these pharmacological agents normalized and
the patient’s symptoms disappeared. Although considerable
investigation needs to be done, this powerful imaging tool
offers great potential to demonstrate areas of reversible cere-
bral ischemia noninvasively.

The reported results of EC-IC arterial bypass in the clinical
setting of stroke in evolution have been mixed; the procedure,
although effective in reversal of ischemic neurological deficit
in 50 to 60% of cases, has in some series been associated with
progression of stroke and significant operative mortality (12,
25, 41, 43, 48, 59). Kletter described two subpopulations of
patients with clinical patterns of stroke in evolution (33). He
defined a group with rapid progressive stroke in whom a
severe deficit reached its maximal density within 6 to 24
hours. Retarded progressive stroke reached maximal density
over a period longer than 24 hours. He concluded as a result
of his experience and review of the literature that revascular-
ization was contraindicated in patients with rapid progressive
stroke but that indications for the procedure might exist in
those patients with deficits evolving over more than 24 hours.
Our current study, coupled with a recent similar small series
(1), suggests that a population of patients with acute sympto-
matic cerebral ischemia that can benefit from early augmen-
tation of middle cerebral artery collateral flow can be identi-
fied and that such therapeutic intervention is not fraught with
significant perioperative mortality and morbidity. The safety
of routine superficial temporal to middle cerebral grafts in
this setting is demonstrated in our study by the fact that, of a

EARLY REVASCULARIZATION 289

total of 20 bypass procedures in 17 patients, none developed
an intracerebral hemorrhage. The sole hemorrhagic compli-
cation in this series involved the use of a long saphenous vein
interposition graft carrying significantly higher minute flows.
Such hemorrhagic complications with these high volume con-
duits have been reported to occur in some 10 to 15% of cases
in our series (42) and in the experience of Diaz et al. (16).

SUMMARY

Based on the results of our experience and other available
data, we think that patients with stroke in evolution and mild
to moderate neurological impairment due to atherosclerotic
disease producing internal carotid occlusion or internal ca-
rotid siphon stenosis in whom CT scan demonstrates no areas
of low density should be offered a revascularization procedure
regardless of the duration of symptoms. Because of concern
about provoking a disabling occlusion in patients with stroke
in evolution due to middle cerebral artery stenosis, we usually
recommend anticoagulation as the initial therapeutic modal-
ity (26). If the neurological deficit remains unstable or pro-
gresses despite anticoagulation, revascularization is per-
formed. In patients with symptomatic delayed cerebral ische-
mia following subarachnoid hemorrhage, we continue to in-
vestigate the potential role of EC-IC bypass. Based on this
preliminary experience, we feel optimistic that management
morbidity may be improved by performing bypass in the
setting of severe proximal spasm of the internal carotid or
middle cerebral artery with poststenotic dilatation of the distal
middle cerebral tree. Patients whose medical management has
failed or in whom aggressive induced hypervolemia and hy-
pertension are contraindicated due to the presence of unse-
cured aneurysms or underlying cardiovascular disease appear
to be the best candidates.

The actual efficacy of this procedure in altering the natural
history of these admittedly diverse disease processes is impos-
sible to verify. However, its relative safety will, we hope,
prompt increasing willingness on the part of neurologists and
neurosurgeons to evaluate the patient with potentially revers-
ible cerebral ischemia and to consider the use of the bypass
procedure in those with the appropriate angiographic and
clinical profiles. We hope that rCBF measurements with
pharmacological manipulations using '*Xe inhalation and
single photon emission tomography may significantly im-
prove our ability to select those patients most likely to harbor
regions of potentially salvageable ischemic cerebral tissue.

ACKNOWLEDGMENT

The authors extend special thanks to Mrs. Vicki Rankin
for assistance in the preparation of this manuscript.

Received for publication, March 25, 1985; accepted, November 2,
1985.

Presented in part at the 57th Annual Meeting of the American
Heart Association, Miami Beach, Florida, November | 1-15, 1984.

Reprint requests: Hunt Batjer, M.D.. Department of Surgery.
Division of Neurosurgery, The University of Texas Health Science
Center at Dallas, 5323 Harry Hines Boulevard, Dallas, Texas 75235.

REFERENCES

1. Andrews BT, Weinstein PR, Chater NL: Emergency extracranial-
intracranial arterial bypass for stroke in evolution, in Spetzler
RF, Selman WR, Carter LP, Martin NA (eds): Cerebral Revas-
cularization for Stroke. New York, Thieme-Stratton, Inc, 1985,
pp 542-547.

2. Austin G, Laffin D, Hayward W: Physiologic factors in the
selection of patients for superficial temporal artery-to-middle
cerebral artery anastomosis. Surgery 75:86 1-868, 1974.

290

3:

wn

6.

cast

BATJER et al.

Austin G, Laffin D, Hayward W: Cerebral blood flow in patients
undergoing microanastomosis for modification or prevention of
stroke. Ann Clin Lab Sci 5:229-235. 1975.

. Baker RN. Broward JA. Rang HC, Fisher CM. Groch SN,

Heyman A. Karp HR. McDevitt E. Scheinberg P. Schwartz W.
Toole JF: Anticoagulant therapy in cerebral infarction: Report
on Cooperative Study. Neurology (NY) 12:823-835, 1962.

. Blaisdell WF. Clauss RH. Galbraith JG. Imparato AM, Wylie

EJ: Joint study of extracranial arterial occlusion: IV. Review of
surgical considerations. JAMA 209:1889-1895, 1969.

Bland JE. Chapman RD, Wylie EJ: Neurological complications
of carotid artery surgery. Ann Surg 171:459-464, 1970.

. Breutman MF. Fields WS. Crawford ES, Debakey ME: Cerebral

hemorrhage in carotid artery surgery. Arch Neurol 9:458-467,
1963.

. Brown TR. Poskanzer DC: Treatment of strokes (first of two

parts). N Engl J Med 281:594-602. 1969.

. Caplan LR, Skillman J, Ojemann R, Fields WS: Intracerebral

hemorrhage following carotid endarterectomy: A hypertensive
complication? Stroke 9:457-460, 1978.

. Carter AB: Anticoagulant treatment in progressing stroke. Br

Med J 2:70-73, 1961.

. Carter LP, Crowell RM. Sonntag VKH, Spetzler RF: Cortical

blood flow during extracranial-intracranial bypass surgery. Stroke
15:836-839, 1984.

. Chater N, Popp J: Microsurgical vascular bypass for occlusive

cerebrovascular disease: Review of 100 cases. Surg Neurol 6:1 15-
118, 1976.

. Crowell RM: STA-MCA bypass for acute focal cerebral ischemia,

in Schmiedek P (ed): Microsurgery for Stroke. New York, Sprin-
ger-Verlag. 1977, pp 244-250.

. Diaz FG. Ausman JI: Experimental cerebral ischemia. Neuro-

surgery 6:436-445, 1980.

. Diaz FG. Mastri AR, Ausman JI, Chou SN: Acute cerebral

revascularization after regional cerebral ischemia in the dog: Part
2. Clinicopathological correlation. J Neurosurg 51:644-653,
1979.

. Diaz FG, Pearce J, Ausman JI: Complications of cerebral revas-

cularization with autogenous vein grafts. Neurosurgery 17:271-
276, 1985.

. Donaghy RMP: Neurologic surgery. Surg Gynecol Obstet

134:269-271. 1972.

. Earnest F, Forbes G. Sandok BA, Piepgras DG, Faust RJ, Ilstrup

DM. Arndt LJ: Complications of cerebral angiography: Prospec-
tive assessment of risk. AJR 142:247-253, Feb 1984.

. Fein JM, Molinari G: Hemodynamic evaluation of superficial

temporal cortical artery microanastomosis in the dog, in Austin
GM (ed): Microneurosurgical Anastomosis for Cerebral Ischemia.
Springfield IL, Charles C Thomas, 1976, pp 5-14.

. Fields WS: Selection of patients with ischemic cerebrovascular

disease for arterial surgery. World J Surg 3:147-154, 1979.

. Fisher CM: Discussion: Transient ischemic attack—an update,

in Scheinberg (ed): Cerebrovascular Disease: Tenth Princeton
Conference. New York, Raven Press, 1976, p 50.

. Goldstone J. Effeney DJ: The role of carotid endarterectomy in

the treatment of acute neurologic deficits. Prog Cardiovasc Dis
22:415-422. 1980.

23. Goldstone J. Moore WS: Emergency carotid artery surgery in

284-1291,

neurologically unstable patients. Arch Surg I1
1976.

. Goldstone J. Moore WS: A new look at emergency carotid artery

operations for the treatment of cerebrovascular insufficiency.
Stroke 9:599-602. 1978.

. Gratz] O. Schmiedek P. Spetzler R. Steinhoff H. Marguth F:

Clinical experience with extra-intracranial arterial anastomosis
in 65 cases. J Neurosurg 44:313-324, 1976.

. Gumerlock MK. Ono H, Neuwelt EA: Can a patent extracranial-

intracranial bypass provoke the conversion of an intracranial
arterial stenosis to a symptomatic occlusion? Neurosurgery
12:391-400. 1983.

ss WK. Clauss RH. Goldberg AF. Johnson AL. Imparato AM.
Ransohoff J: Special problems associated with surgical and
thrombolytic treatment of strokes. Arch Surg 92:27-31. 1966.

. Hass WK. Fields WS. North RR, Kricheff IJ. Chase NE, Bauer

30.

al.

33,
34.
35:.

36.

3.

38.

40.

41.
42.

43.

44,

45.

46.

47.

48.

49,

Neurosurgery, Vol. 18, No. 3

RB: Joint study of extracranial arterial occlusion: Arteriography,
techniques. sites, and complications. JAMA 203:961-968, 1968.

29. Hayashi S, Nehls DG, Kieck CF, Vielma J, DeGirolami U,

Crowell RM: Beneficial effects of induced hypertension on ex-
perimental stroke in awake monkeys. J Neurosurg 60:151-157,
1984.

Hilal SK, Resch JA, Anaplatz: Determination of the carotid and
regional cerebral blood flow by a radiographic technique. Acta
Radiol (Diagn) 5:232-240, 1966.

Holbach KH, Wassman H, Bodosi M, Bonatelli AP: Superficial
temporal-middle cerebral artery anastomosis for internal carotid
occlusion. Acta Neurochir (Wien) 37:201-217, 1977.

. Ito Z, Hen R, Nakajima K, Suzuki A, Moriyama T, Nemura K:

Evaluation of functional reversibility of ischemic brain—to select
appropriate patients with completed stroke for STA-MCA bypass
surgery. Neurol Med Chir 16:121-129, 1976.

Kletter G: The Extra-intracranial Bypass Operation for Preven-
tion and Treatment of Stroke. New York, Springer-Verlag, 1979.
Lehv MS, Salzman EW, Silen W: Hypertension complicating
carotid endarterectomy. Stroke 1:307-313, 1970.

Loftus CM, Quest DO: Current status of carotid endarterectomy
for atheromatous disease. Neurosurgery 13:718-723, 1983.
Millikan CH: Clinical management of cerebral ischemia, in
McDowell FH, Brennan RW (eds): Cerebral Vascular Disease,
Eighth Conference. New York, Grune and Stratton, 1973, pp
209-218.

Millikan CH, Bauer RB, Goldschmidt J, et al: A classification
and outline of cerebrovascular disease: II. Ad hoc committee,
Advisory Council, National Institute of Neurological and Com-
munication Disorders and Stroke, National Institutes of Health,
Bethesda, Maryland. Stroke 6:565-616, 1975.

Nornes H: Pulsatile blood flow pattern in cerebral circulation, in
Langfitt TW, McHenry LC Jr, Reivich M, Wollman H (eds):
Cerebral Circulation and Metabolism. New York, Springer-Ver-
lag, 1975, pp 152-154.

. Nornes H, Wikeby P: Cerebral arterial blood flow and aneurysm

surgery: Part I. Local arterial flow dynamics. J Neurosurg 47:810-
818, 1977.

Ojemann RG, Crowell RM, Roberson GH, Fisher CM: Surgical
treatment of extracranial carotid occlusive disease. Clin Neuro-
surg 22:214-263, 1975.

Reichman OH: Complications of cerebral revascularization. Clin
Neurosurg 23:318-335, 1976.

Samson DS, Batjer HH, Hodosh RM: Re-evaluation of saphen-
ous vein interposition grafis in the treatment of cerebral ischemia:
A five year experience. Presented at the 33rd Annual Meeting of
the Congress of Neurological Surgeons, Chicago, Illinois, October
28-November 2 1983.

Samson DS, Boone S; Extracranial-intracranial (EC-IC) arterial
bypass: Past performance and current concepts. Neurosurgery
3:79-86, 1978.

Spetzler, Chater N: Microvascular bypass surgery: Part 2: Physi-
ologic studies. J Neurosurg 45:508-513, 1976.

Sundt TM Jr: The ischemic tolerance of neural tissue and the
need for monitoring and selective shunting during carotid end-
arterectomy. Stroke 14:93-98, 1983.

Sundt TM, Houser OW, Sharbrough FW: Carotid endarterec-
tomy: Results, complications, and monitoring techniques. Adv
Neurol 16:97-119, 1977.

Sundt TM, Sandok BA, Whisnant JP: Carotid endarterectomy:
Complications and preoperative assessment of risk. Mayo Clin
Proc 50:301-306, 1975.

Sundt TM Jr, Seikert RG, Piepgras DG, Sharbrough FW, Houser
OW: Bypass surgery for vascular disease of the carotid system.
Mayo Clin Proc 51:677-692, 1976.

Sundt TM Jr, Sharbrough FW, Anderson RE, Michenfelder JD:
Cerebral blood flow measurements and electroencephalograms
during carotid endarterectomy. J Neurosurg 41:310-320, 1974.

. Symon L. Paztor E, Branston NM: The distribution and density

of reduced cerebral blood flow following acute middle cerebral
artery occlusion: An experimental study by the technique of
hydrogen clearance in baboons. Stroke 5:355-364, 1974.

. Thompson JE: Complications of carotid endarterectomy and

their prevention. World J Surg 3:155-165, 1979.

March 1986

52. Thompson JE. Talkington CM: Carotid endarterectomy. Ann
Surg 184:1-15, 1976.

53. Toole JF: Management of TIA’s and acute cerebral infarction.
Adv Neurol 16:71-80, 1977.

54. Toole JF, Cole M: Ischemic cerebrovascular disease, in Baker
AB. Baker LH: Clinical Neurology. New York, Harper and Row,
1983. vol 2, ch 15, p 1-51.

55. Torvik A, Fisher CM: Progressing stroke: Pathogenesi
RG, Whisnant JP (eds): Cerebral Vascular Diseases,
ference. New Y ork, Grune and Stratton, 1961, pp 133-150.

56. Vorstrup S, Engell HC, Lindewald H, Lassen NA: Hemody-
namically significant stenosis of the internal carotid artery treated
with endarterectomy: Case report. J Neurosurg 60:1070-1075,

1984.
57. Wylie EJ, Hein MF. Adams JE: Intracranial hemorrhage follow-
ing surgical cularization for treatment of acute strokes. J

Neurosurg 21:21 , 1964.

58. Yasargil MG: Anastomosis between the superficial temporal
artery and a branch of the middle cerebral artery, in Yasargil
MG (ed): Microsurgery Applied to Neurosurgery. Stuttgart, Georg
Thieme Verlag, 1969, pp 105-115.

59. Yasargil MG. Yonekawa Y: Results of microsurgical extra-intra-
cranial arterial bypass in the treatment of cerebral ischemia.
Neurosurgery . 1977.

60. Yonekawa Y, Yasargil MG: Extra-intracranial arterial anasto-
mosis: Clinical and technical aspects. Results, in Krayenbuhl H
(ed): Advances and Technical Standards in Neurosurgery. Wien,
Springer-Verlag, 1976, vol 3, pp 47-78.

COMMENTS

More is not necessarily always better. Batjer et al. have
demonstrated that in the treatment of focal cerebral ischemia
the increase of cerebral blood flow by a small but significant
amount through a relatively low flow bypass conduit may be
more effective and safer than providing a high flow of blood
by opening occluded or highly stenotic vessels. This small
augmentation in perfusion may deliver the critical amount of
flow that is necessary to prevent ischemic damage. This is an
attractive idea that coincides with the current theories of acute
cerebral ischemia. The neurons in the ischemic penumbra,
which receive enough blood flow to prevent ischemic death
but not enough to support electrical activity, have been shown
to have a survival time that is related to the depth and duration
of ischemia. As Batjer et al. have shown, small increases in
cerebral blood flow can have a significant impact on these
neurons. As bypass grafts mature and deliver more blood
flow, these neurons can move from the oligemic susceptible
range to the ischemic nonfunctional range to the nonischemic
range. Additionally, a slow progression may be more advan-
tageous from a theoretical standpoint. Abrupt reperfusion has
been associated not only with increased hemorrhagic infarc-
tion and hemorrhage, but with increased cerebral edema as
well. In acute cerebral ischemia. to get the most blood flow
to the ischemic hemisphere may not be the optimal goal. To
increase the blood flow about a critical level to prevent cell
death may be the best treatment.

The authors report promising results with the use of early
revascularization in the face of acute cerebral ischemia by
using the EC-IC anastomosis in several interesing clinical

EARLY REVASCULARIZATION 291

situations. Although it is true that the authors cannot be
certain of the diagnostic categories into which their patients
fall because they intervened early in the acute phase of the
ischemia, their results for treatment of stroke in evolution are
impressive in comparison to the natural history of this con-
dition or the results of carotid endarterectomy in such pa-
tients. The use of this procedure in treating vasospasm after
subarachnoid hemorrhage is a more novel application, and
the benefits are not as well established. This technique may
be another option for treating symptomatic spasm when
volume expansion, hemodilution, and induced hypertension
have failed.

We agree with the authors that it is essential to evaluate
patients with ischemic neurological deficits as early as possible
with angiography. Dangerous lesions such as carotid thrombi
can be treated by endarterectomy and, based upon the results
of the present study, patients without correctable lesions can
be considered for EC-IC anastomosis.

Our own clinical experience in revascularization for acute
stroke some years ago was poor, and we abandoned it. It
paralleled the experience by others who noted increased
edema and elevated intracranial pressure. Caution is manda-
tory in this group of patients; experiments with early reper-
fusion in animals also raise questions about its benefit. Timing
is essential. The authors’ results suggest that another look at
revascularization is warranted; nonetheless, clear-cut results
in an appropriate animal model are desirable before wide-
spread clinical adoption of revascularization for acute stroke
can be recommended.

Daniel G. Nehls, M.D.
Robert F. Spetzler, M.D.
Phoenix, Arizona

At first blush, the reader may question the publication of
another paper extolling the virtues of microsurgical revascu-
larization for ischemic intra-cranial vascular disease in light
of recently reported evidence that the procedure is apparently
no better than the best medical management for patients with
transient cerebral ischemia. It is important to remember that
the latter group of patients constitute only one segment of the
population of patients with ischemic cerebral vascular disease,
and the reported study, although large and significant, is only
one study. It is unlikely that there will ever be a final answer
regarding the role of surgery in the management of intra-
cranial ischemic cerebral vascular disease, because our under-
standing of the problems continues to change. Additionally,
statistics providing tests of significance, are useful in analyzing
populations, not individuals. Here is where the experience of
the surgeon and his understanding of the pathophysiology of
the disease addressed is most valuable. Thus, surgeons must
continue to attack the multi-faceted problem of ischemic
cerebral vascular disease with their skills and continue to
report their experiences.

Clark Watts, M.D.
Columbia, Missouri