Extracranial-intracranial arterial bypass for middle cerebral artery stenosis and occlusion

Transcription

1 J Neurosurg 62:83-838, 985 Extracranial-intracranial arterial bypass for middle cerebral artery stenosis and occlusion Operative results in 65 cases BRIAN T. ANDREWS, M.D., NORMAN L. CHATER, M.D., AND PHILIP R. WEINSTEIN, M.D. Department of Neurological Surgery, School of Medicine, University of California, San Francisco, California u- Forty-seven patients with middle cerebral artery (MCA) stenosis and 8 patients with MCA occlusion underwent extracranial-intracranial arterial bypass procedures. Patients presented with a history of transient ischemic attacks (TIA's), reversible ischemic neurological deficits, TIA's after initial stroke, stroke-in-evolution, or completed stroke. Angiography revealed that the MCA stenosis ranged from 70% to over 95%. Two patients (4.3%) in the stenosis group had a perioperative stroke (within 30 days of operation). There was no perioperative mortality. In the occlusion group, no patient had a perioperative stroke, and one patient (5.5%) died from a non-neurological disease. The TIA's resolved completely in 90% of the patients with stenosis and in 9.6% of those with occlusion. No patient with MCA stenosis had a late ipsilateral stroke, although five had a contralateral or vertebrobasilar stroke. One patient with MCA occlusion had a late ipsilateral stroke. The bypass patency rate at late follow-up review was 00%. The results of intracranial-extracranial arterial bypass procedures appear to be similar for patients with either stenosis or occlusion of the MCA. Symptomatic relief of TIA's was excellent and, in two patients with progressive stroke-in-evolution, the deficit was stabilized. The incidence of postoperative ipsilateral stroke was low in patients with TIA's alone or with TIA's after an initial stroke, but among patients with completed stroke, improvement was confined to slight reduction in the neurological deficit. KEY WORDS " extracranial-intracranial bypass 9 middle cerebral artery 9 stenosis 9 occlusion 9 ischemia 9 stroke 9 anastomosis S INCE the extracranial-intracranial (EC-IC) arterial bypass procedure was introduced by Donaghy and Ya~argil in 967, 4 this technique has been used to treat inaccessible cerebrovascular occlusive lesions. 8,~4 The early results, although variable, indicate that this operation may increase blood flow to ischemic areas of the brain and relieve symptoms of vascular insufficiencyfl '9'~ Long-term follow-up review of these patients suggests that the incidence of stroke on the side of a functioning bypass appears to be low. 4'23 Endarterectomy and embolectomy for middle cerebral artery (MCA) stenosis and occlusion have also been attempted, but the results are not encoura~ng, z22'24 Patients with MCA stenosis often present with transient ischemic attacks (TIA's) that may be due to hemodynamic insufficiency, l,~ ~ providing an opportunity for therapeutic intervention before total occlusion and severe stroke occur. In some cases, a favorable prognosis with medical treatment alone has been observed with this lesion. In contrast, MCA occlusion appears to carry a worse prognosis. Up to 87% of patients suffer abrupt onset of moderate to severe stroke 6A3,7 with an early mortality rate as high as 20%. 6'3 In most patients, acute occlusion is thought to be the result of embolism. 6' In cases where atheromatous thrombosis is suspected, however, as many as 65% of patients have prodromal TIA's and up to 3% exhibit stepwise development of deficit. 6 This suggests that an opportunity for prophylactic intervention may exist in a significant number of patients at risk for stroke from MCA occlusion. The effect of medical treatment in such patients has not been clearly defined. We have reviewed the results of EC-IC bypass procedures in a series of 47 patients with MCA stenosis J. Neurosurg. / Volume 62/June,

2 B. T. Andrews, N. L. Chater and P. R. Weinstein Clinical Data TABLE Summary of clinical data* no. of patients 47 average age (yrs) age range (yrs) males 33 females 4 side of lesion right 20 left 27 * MCA = middle cerebral artery. MCA Stenosis MCA Occlusion No. Percent No. Percent TABLE 2 Risk factors in patients with MCA stenosis or occlusion* Risk Factors MCA Stenosis MCA Occlusion hypertension 60% 6% diabetes mellitus 20% 22% hyperlipidemia % % cigarette smoking 28% 44% miscellaneoust 6.6% 27% * MCA = middle cerebral artery. + Atrial fibrillation, rheumatic heart disease, and arteriosclerotic cardiovascular disease. TABLE 3 Presenting symptoms in patients with MCA stenosis or occlusion * Symptoms MCA Stenosis MCA Occlusion No. Percent No. Percent TIA's only RIND only stroke followed by onset of 3 I l TIA's or RIND stroke-in-evolution completed stroke * MCA = middle cerebral artery; TIA = transient ischemic attack; RIND = reversible incomplete neurological deficit. TABLE 4 Pre- and postoperative neurological status of patients with completed stroke* Preoperative Postoperative Case No. Grade Grade MCA stenosis 2 MCA occlusion * Grading system: 0 = normal; l = minimal or trace deficit; 2 = mild deficit, including dysphasia; 3 = moderate deficit, including severe dysphasia; 4 = severe deficit, including hemiplegia and aphasia. MCA = middle cerebral artery. and 8 patients with MCA occlusion, all of whom had symptoms of cerebral ischemia and were treated between 972 and 983. Patients were followed for 3 months to years. The outcome is evaluated in relation to the known natural history of MCA stenosis and occlusion, and compared with the results of medical management of these lesions with anticoagulant agents as reported in the literature. Summary of Cases Between May, 972, and June, 983, 65 patients underwent EC-IC bypass procedures for MCA stenosis (47 patients) or occlusion (8 patients). Preoperative arteriography was performed on all patients. The degree of stenosis was measured as the maximum percent reduction of luminal diameter in any angiographic projection. Superficial temporal artery (STA)-MCA bypass procedures or, in two cases, occipital artery to MCA end-to-side anastomoses were performed by the same surgeon (N.L.C.), using previously described techniques. 8'22 Patients received oral aspirin (325 rag/day) and sulfinpyrazone (00 mg three times daily) preoperatively; this regimen was continued postoperatively throughout the follow-up period. Clinical follow-up review was accomplished by hospital or office examination, by telephone contact with patients or their local physician, or by questionnaire returned by the patient. Recurrent TIA's, late stroke, and death were end points for the study. Graft patency was evaluated by auscultation with a Doppler flow probe,* following the STA up to the point at which the graft entered the craniectomy site. The clinical data are summarized in Table. Thirtythree men and 4 women had MCA stenosis; 5 men and three women had MCA occlusion. The average age in both groups was 60 years. The prevalence of risk factors was similar in both groups, and 80% of patients had at least one risk factor (Table 2). The presenting symptoms are listed in Table 3. In the stenosis group, 66% of patients had TIA's, compared with only 33% in the occlusion group. The frequency of TIA's was slightly greater in patients with occlusion. Only three patients (6.4%), all with stenosis, had a reversible incomplete neurological deficit lasting more than 24 hours. Fixed deficits with superimposed TA's or reversible incomplete neurological deficits were seen in 23% of patients with stenosis and 33% with occlusion. One patient in each group had strokein-evolution. One patient (2.%) with stenosis and five patients (28%) with occlusion had a completed stroke; the neurological status of patients with a completed stroke at the time of the bypass procedure is shown in Table 4. Ischemic symptoms had been present for week to 8 years (average 8.4 months) in patients with stenosis, * Doppler flow probe manufactured by Medsonics, Inc., 340 Pioneer Way, Mountain View, California. 832 J. Neurosurg. / Volume 62/June, 985

3 Extracranial-intracranial arterial bypass and for month to 4 years (average 26 months) in those with occlusion. None of the six patients with completed stroke was treated less than 4 months after the onset of the ischemic deficit. Angiographically, the stenosis ranged from 70% to over 95%; 43 of the 47 patients had greater than 90% stenosis. Ischemic symptoms corresponded to the side of stenosis or occlusion in all cases. The side of disease was similar in both groups: the left hemisphere was affected in 58% of patients with stenosis and in 56% of those with occlusion. Three patients (6.4%) with stenosis and one patient (5.5%) with occlusion had a history of contralateral ischemic cerebrovascular disease. Three patients with stenosis and one patient with occlusion had undergone ipsilateral carotid endarterectomy for treatment of TIA's or stroke. In one patient, an MCA occlusion was diagnosed and treated with an EC-IC bypass procedure in 976 and stenosis of the contralateral MCA was diagnosed and treated similarly in 980 (Fig. ). Operative Results Perioperative Morbidity' and Mortality Perioperative morbidity and mortality (within 30 days of operation) are summarized in Table 5. Overall, the perioperative morbidity rate was 9.2%, the perioperative stroke rate was 3.%, and the mortality rate was.5 %. There were no perioperative deaths in the stenosis group. Three patients (6.4%) with stenosis had early postoperative transient neurological dysfunction on the side of the anastomosis: one had two brief TIA's, which were similar to her preoperative attacks, within 24 hours of the operation; one had a single TIA daily for 4 days after the bypass procedure; and one had postoperative dysphasia that resolved completely after 6 days. These patients have had no further clinical evidence of ischemia. Two patients (4.3 %) with stenosis had a perioperative stroke. One patient, who had a history of severe hypertension, hyperlipidemia, and recent onset of TIA's, developed mild postoperative hemiparesis that progressed to hemiplegia over the course of 4 weeks, despite a patent bypass as shown by Doppler ultrasound evaluation. Evidence of severe small-vessel arteriosclerotic disease involving the Sylvian branches of the MCA was found at surgery; pathologically, the arteriotomy specimen showed severe intimal hyperplasia and smoothmuscle proliferation. The second patient had chronic atrial fibrillation and a history of stroke with superimposed TIA's. Chronic anticoagulation therapy was discontinued preoperatively. After operation, the patient had no deficit until the sudden onset of aphasia 6 days later. Anticoagulation therapy was resumed, but gastrointestinal hemorrhage and anemia developed, followed by a progressive hemispheric stroke during the 3rd postoperative week. This patient's postoperative deficit was thought to be caused by embolism from a cardiac source. TABLE 5 Perioperative morbidity and mortality* MCA Stenosis MCA Occlusion Complications No. Percent No. Percent morbidity other than stroker 2. 0 transient neurological deficit stroke death * MCA = middle cerebral artery. 5- Posloperative angina pectoris. FIG.. Left and Center: Anteroposterior and lateral common carotid arteriograms showing occlusion of the fight middle cerebral artery (MCA) and retrograde flow to the MCA through a dilated bypass graft (arrow) 4 years after surgery. Right: Anteroposterior arteriogram showing a new symptomatic high-grade stenosis of the left MCA. J. Neurosurg. / Volume 62/June,

4 B. T. Andrews, N. L. Chater and P. R. Weinstein FIG. 2. Common carotid arteriograms, anteroposterior (left) and lateral (right) views, obtained 0 years after surgery when the patient began to have headaches on the side of the bypass graft. A high-grade stenosis (arrow) is visible at the origin of the internal carotid artery. A patent dilated superficial temporal artery graft supplies the middle cerebral artery branches. No perioperative morbidity or stroke occurred in the occlusion group. There was one early death (5.5%) in a patient operated on for stroke-in-evolution. This patient's neurological deficit stabilized postoperatively, but gastrointestinal hemorrhage and bowel infarction developed, and he died from peritonitis. In both patients with stroke-in-evolution, progressive neurological deficit was arrested after EC-IC anastomosis. In no patient with prior completed stroke was there early postoperative neurological improvement. TABLE 6 Late results after EC-IC arterial bypass procedure* Outcome MCA Stenosis MCA Occlusion No. Percent No. Percent complete resolution of TIA's 36 90t 9.7? ipsilateral TIA's contralateral or vertebro- 4 0t basilar TIA's stroke ipsilateral to bypass stroke contralateral to bypass vertebrobasilar stroke fatal stroke * EC-IC = extracranial-intracranial; MCA = middle cerebral artery; TIA = transient ischemic attack. t Percent of patients with preoperative TIA's. All other percentages are of the entire stenosis or occlusion group. Late Follow-Up Review The results after the first 30 postoperative days are summarized in Table 6. The TIA's resolved completely in 36 (90%) of 40 patients with stenosis and in (9.7%) of 2 patients with occlusion. Four (0%) of these 40 patients with stenosis had one or two late ipsilateral TIA's to 26 months after the operation. In three of these patients, arteriography showed patent bypass grafts. One of the three had a new ulcerative lesion at the carotid bifurcation; this was treated with carotid endarterectomy and the symptoms disappeared. The fourth patient had two poorly characterized TIA's 4 weeks after bypass. She did not undergo arteriography and died from an occlusion of the basilar artery 4 months later. One patient with occlusion had recurrent ipsilateral TIA's 88 months after operation. Angiography revealed a patent bypass and stable MCA occlusion. Chronic anticoagulation was prescribed and he has had no further symptoms. One patient in each group developed late symptoms consistent with vertebrobasilar TIA's. Arteriography failed to show a significant lesion and both patients had patent bypasses. No patient with stenosis had a late ipsilateral stroke. One patient with occlusion had stepwise progression, 36 and 4 months after bypass, to hemiplegia. Angiographic evaluation of the graft was not possible in this case. Two patients with stenosis (4.3%) had a contra- 834 J. Neurosurg. / Volume 62/June, 985

5 Extracranial-intracranial arterial bypass FIG. 3. Lateral common carotid arteriograms showing an occipital-to-middle cerebral artery anastomosis (arrow) 2 weeks after surgery (left) and at follow-up study 0 years later (right). lateral stroke 6 and 39 months after bypass; one of these strokes was caused by hypertensive hemorrhage. Late death in three patients (6.4%) with stenosis was attributed to vertebrobasilar stroke, but autopsies were not performed. No patient in the occlusion group had a contralateral or vertebrobasilar stroke during the follow-up period. The postoperative neurological status in the six patients who had a completed stroke before their bypass procedure is shown in Table 4. Five of these patients improved slowly and had minimal to moderate residual deficit at the time of follow-up examination. One patient remained stable with a moderate hemiparesis and dense aphasia, and one suffered a recurrent ipsilateral stroke. Late mortality is summarized in Table 7. Ten patients with stenosis (2.3%) and two with occlusion (.%) died more than 30 days after the bypass procedure. In the stenosis group, three (30%) of the deaths were from contralateral or vertebrobasilar stroke. No late deaths from stroke occurred in the occlusion group. Cardiac disease was the most common cause of death. Late Vascular Operations One patient with right MCA stenosis developed severe ipsilateral migrainous headaches and a tender, palpable STA 0 years after the bypass procedure. An arteriogram revealed a 90% stenosis of the proximal right internal carotid artery and a markedly enlarged bypass graft (Fig. 2), which supplied much of the hemisphere. No neurological deficit was noted. After carotid endarterectomy, this patient's headaches resolved. Two other patients with stenosis had late ipsilateral carotid endarterectomy. One, as noted above, began to have TIA's 6 months after the bypass procedure. The other was treated for asymptomatic stenosis of the internal carotid artery month after anastomosis. None of the occlusion patients had a late ipsilateral extracranial vascular procedure. One patient from each group underwent contralateral carotid endarterectomy for symptomatic cerebrovascular disease. Evaluation of Graft and MCA Patency Perioperative Doppler ultrasound evaluation demonstrated graft patency in all 65 patients. Late postoperative testing at a mean of 42 months (range 6 months to years) confirmed graft patency in all survivors. Seven patients underwent postoperative arteriography, which confirmed graft patency in each case; in one patient, angiography 0 years after operation showed a widely patent graft (Fig. 3). Three patients had progression of stenosis after the bypass procedure, two to complete occlusion (Fig. 4) and one at a more proximal location. None had associated neurological symptoms. TABLE 7 Causes of death after EC-IC arterial bypass procedure* Cause of MCA Stenosis MCA Occlusion Death No. Percent No. Percent all causes stroke ipsilateral stroke 0 0 cardiac renal failure 2. 0 unknown * EC-IC = extracranial-intracranial; MCA = middle cerebral artery. J. Neurosurg. / Volume 62/June,

6 B. T. Andrews, N. L. Chater and P. R. Weinstein FIG. 4. Upper Left: Anteroposterior common carotid arteriogram showing stenosis of the right middle cerebral artery (arrow). Lower. Follow-up arteriogram year later showing a patent bypass graft on the lateral projection. Upper Right: Angiogram, anteroposterior projection, obtained at the same time as the lateral projection (lower) showing progression of stenosis to a complete occlusion of the middle cerebral artery. The occlusion was asymptomatic. Discussion Stenosis and occlusion of the MCA account for up to 33% of patients considered for EC-IC bypass procedures because of inaccessible symptomatic cerebrovascular disease.~.",s'~4,~s Stenosis is probably a less common cause of ischemia than is occlusion of the MCA or its branchesj '~ Patients with stenosis frequently have TIA's, presumably caused by hemodynamic insufficiency, although embolization distal to the stenosis has been reported (Weinstein, et al., unpublished data, 982). Stenosis patients may have a benign course, z but progression to complete occlusion and cerebral infarction does occur, ~ and some reports suggest that the risk of stroke may be as high as 5% per yearj 3 Occlusion, however, causes abrupt or rapidly progressive stroke in up to 87% of cases, 6'3'7 and TIA's precede stroke in only 2% to 38%. 6'5'7'2~ In patients whose occlusion is considered to be embolic, the onset of stroke is generally abrupt. In contrast, up to 80% of patients whose occlu- sion is thrombotic have prodromal TIA's or stepwise development of a neurological deficit. 6 Even though up to 72% of occlusion patients with strokes recover spontaneously to an improved neurological status, 2'~2 4.5% per year will die of recurrent stroke53 In stenosis patients, the reduction in the frequency of TIA's and in the incidence of subsequent stroke achieved by EC-IC bypass procedures appears to be similar to the reduction after medical therapy with anticoagulant agents. Hinton, et al.,~l reviewed 6 stenosis patients who were treated medically. During a follow-up period of up to 6 years, a benign course was observed in 4 patients and only one had persistent TIA's. Two patients (2%) suffered a stroke before therapy was started. In our series, two patients (4.3%) in the stenosis group had a perioperative stroke. One of these strokes was probably caused by embolism related to chronic atrial fibrillation, and the other was related to severe arteriosclerotic disease of small intracerebral vessels. Perioperative morbidity was minimal, and there were no deaths. No patient had a late stroke on the side of the bypass procedure. The problems and possible complications of chronic therapy with anticoagulant agents were avoided in our series, although one patient is taking Coumadin (sodium warfarin) for vertebrobasilar TIA's. The reduction in the frequency of TIA's after bypass grafting in occlusion patients was similar to that in stenosis patients. The effect of medical therapy on patients with occlusion is not known, as there are no studies that provide data on the results of anticoagulation therapy in these patients. All of our patients received antiplatelet agents postoperatively. This treatment may have influenced the incidence of postoperative ischemic events. Although some randomized trials have shown that such agents have no effect, 3'2~ others have indicated the benefit of aspirin and sulfinpyrazone therapy on the incidence of stroke and TIA's. 7'6 Our results are in general agreement with the results from other series of patients with bypass grafting, most of whom had lesions other than stenosis or occlusion of the MCA. Perioperative stroke rates of % to 4.2% and perioperative mortality rates of up to 7% have been reported, 4's'ls'9,23 Of patients with TIA's, symptoms resolve completely in 82% to 92%. 4"s'~s'25 Yonekura, et al., 25 noted resolution of symptoms in fewer patients with TIA's after an initial stroke. We found no difference in the rate of TIA resolution between patients with preoperative TIA's only and those with mild stroke and superimposed TIA's. The frequency of preoperative TIA's did not appear to influence the development of postoperative neurological dysfunction in our series. Patients with stroke-in-evolution have reportedly fared less well with bypass procedures. Gratzl, et al., s found that four of five patients with MCA occlusion or stenosis and acutely progressing strokes at the time of EC-IC anastomosis died with brain edema soon afterward or had progressive deterioration and died within 836 J. Neurosurg. / Volume 62/June, 985

7 Extracranial-intracranial arterial bypass month after surgery. All of their patients deteriorated neurologically after surgery. In other reports, however, up to 86% of such patients improved or stabilized neurologically after bypass grafting. 4'~s'2~ Both of our patients with stroke-in-evolution stabilized postoperatively, although late follow-up data are not available in either case. Neurological improvement after a bypass procedure has been reported in up to 70% of patients with completed stroke. 4'8'7'8'25 In all series, the patients with mild preoperative deficits made the best recovery. However, it has not been shown that EC-IC anastomosis improves the outcome over that expected in untreated patients. Kaste and Waltimo I2 noted that 56 (72%) of 78 patients with stroke caused by documented occlusion of the MCA attained a level of complete independence over 30 months without treatment. In our series, 83% improved neurologically. Perhaps because none of our patients with completed stroke was operated on less than 4 months after onset, no patient had a permanently worsened deficit or a hemorrhagic infarction after EC-IC anastomosis. The rate of recurrent stroke on the side of the bypass graft in this group was 3.4% per year, which is similar to the incidence of 4.5% per year for untreated patients with MCA occlusion reported by Lascelles and Burrows.~3 Moreover, Allcock 2 found that only two (5%) of 40 patients with MCA occlusion or stenosis and completed stroke had recurrent strokes in a follow-up period of up to 2 years. These results suggest that although EC- IC anastomosis may slightly improve neurological recovery in patients with completed strokes compared with the results in untreated patients, it may not lower the risk of recurrent stroke. Doppler evaluation in our patients at a mean of 42 months after surgery showed that all grafts remained patent, and that location of the bypass graft closer to the lesion than in cases of carotid occlusion does not appear to alter the patency rate. Postoperative angiography in this and previous series 4'8'8'23 has confirmed long-term patency and enlargement of bypass grafts, as well as progression of stenosis or occurrence of complete occlusion without clinical symptoms, as documented in two of our patients. Neither of our two stenosis patients who had a perioperative stroke underwent angiography after the onset of the deficit to determine whether progression to occlusion had occurred. Although stenosis and occlusion of the MCA have a significantly different initial presentation, natural history, and outcome, EC-IC anastomosis followed by antiplatelet therapy was associated with resolution of symptoms in most of our patients with these lesions who presented with TIA's or with TIA's after initial stroke. The subsequent stroke rate was low in both groups. This treatment yielded results that equaled or improved upon those obtained with anticoagulant therapy in one small series of patients with MCA stenosis, 'l while avoiding the risk of complications from longterm anticoagulation. The bypass procedure halted the progression of neurological deficits in both patients treated for stroke-in-evolution but, among patients with completed stroke, improvement was confined to slight reduction of the neurological deficit. A prospective randomized study is needed to assess the effectiveness of EC-IC bypass procedures in reducing symptoms, preventing stroke, and increasing patient survival. Nonetheless, our study describes the results of bypass procedures in patients with a single site of cerebrovascular occlusive disease treated in a single neurosurgical department. These data may help to clarify differences in the results of this operation among patients with specific vascular occlusive lesions treated under standardized conditions. Acknowledgment The authors thank Stephen Ordw~ay for editorial assistance. References. Adams HP Jr, Gross CE: Embolism distal to stenosis of the middle cerebral artery. Stroke 2: , Allcock JM: Occlusion of the middle cerebral artery: serial angiography as a guide to conservative therapy. J Neurosurg 27: , Bousser MG, Eschwege E, Haguenau M, et al: "ACLA" controlled trial of aspirin and dipyridamole in the secondary prevention of athero-thrombotic cerebral ischemia. Stroke 4:5-4, Chater N: Results of neurosurgical microvascular extracranial-intracranial bypass for stroke: a decade of experience. West J Med 38:53-533, de Weerd AW, Veering MM, Mosmans PCM, et al: Effect of the extra-inlracranial (STA-MCA) arterial anastomosis on EEG and cerebral blood flow. A controlled study of patients with unilateral cerebral ischemia. Stroke 3: , Fisher CM: The natural history of middle cerebral artery trunk occlusion, in Austin GM (ed): Mieroneurosurgical Anastomosis for Cerebral Ischemia. Springfield, Ill: Charles C Thomas, 976, pp Genton E, Barnett HJM, Fields WS, et al: Cerebral ischemia: the role of thrombosis and of antithrombotic therapy. Study Group on Antithrombotic Therapy. Stroke 8:50-75, Gratzl O, Schmiedek P, Spetzler R, et al: Clinical experience with extra-intracranial arterial anastomosis in 65 cases. J Neurosurg 44:33-324, Gratzl O, Schmiedek P, Steinhoff H, et al: Microneurosurgical anastomoses for cerebral ischemia in 39 patients. Clinical results, angiography, and regional blood flow, in Austin GM (ed): Microneurosurgicai Anastomosis for Cerebral Ischemia. Springfield, Ill: Charles C Thomas, 976, pp Halsey JH Jr, Morawetz RB, Blauenstein UW: The hemodynamic effect of STA-MCA bypass. Stroke 3: 63-67, 982 l. Hinton RC, Mohr JP, Ackerman RH, et al: Symptomatic middle cerebral artery stenosis. Ann Neurol 5:52-57, Kaste M, Waltimo O: Prognosis of patients with middle cerebral artery occlusion. Stroke 7: , Lascelles RG, Burrows EH: Occlusion of the middle cerebral artery. Brain 88:85-96, Little JR, Yamamoto YL, Feindel W, et al: Superficial J. Neurosurg. / Volume 62/June,

8 B. T. Andrews, N. L. Chater and P. R. Weinstein temporal artery to middle cerebral artery anastomosis. Intraoperative evaluation by fluorescein angiography and xenon-3 clearance. J Neurosurg 50: , Marti-Vilalta JL, Lopez-Pousa S, Grau JM, et al: Transient ischemic attacks. Retrospective study of 50 cases of ischemic infarct in the territory of the middle cerebral artery. Stroke 0: , Millikan CH, McDowell FH: Treatment of transient ischemic attacks. Stroke 9: , Mouren P, Bonnal J, Massad L: Etude clinique et art6riographique des thromboses sylviennes (fi propos de 25 observations). Presse Med 63: , 955 t8. Samson DS, Boone S: Extracranial-intracranial (EC-IC) arterial bypass: past performance and current concepts. Neurosurgery 3:79-86, Samson D, Watts C, Clark K: Cerebral revascularization for transient ischemic attacks. Neurology 27:767-77, Silverstein A, Hollin S: Internal carotid vs middle cerebral artery occlusions. Clinical differences. Arch Neurol 2:468-47, Sorensen PS, Pedersen H, Marquardsen J, et al: Acetylsalicylic acid in the prevention of stroke in patients with reversible cerebral ischemic attacks. A Danish Cooperative Study. Stroke 4:5-22, Weinstein PR, Chater NL, Ya~argil MG: Microsurgical treatment of intracranial cerebrovascular occlusive disease, in Goldsmith H (ed): Lewis' Practice of Surgery. Hagerstown, Md: Harper & Row, 978, pp Weinstein PR, Rodriguez y Baena R, Chater NL: Results of extracranial-intracranial arterial bypass for intracranial internal carotid artery stenosis: review of 05 cases. Neurosurgery 5: , Yonekawa Y, Handa H, Mitsumsa T, et al: Revascularization in the acute stage of middle cerebral artery occlusion, in Peerless S J, McCormick CW (eds): Microsurgery for Cerebral Ischemia. Berlin/Heidelberg]New York: Springer-Verlag, 980, pp Yonekura M, Austin G, Hayward W: Long-term evaluation of cerebral blood fow, transient ischemic attacks, and stroke after STA-MCA anastomosis. Surg Neurol 8:23-30, 982 Manuscript received February 6, 984. Accepted in final form December, 984. Address reprint requests to: Brian T. Andrews, M.D., c/o The Editorial Office, Department of Neurological Surgery, 360 Ninth Avenue, Suite 20, San Francisco, California J. Neurosurg. / Volume 62/June, 985