Experimental cerebral infarctioneffects of pentobarbital in Mongolian gerbils.

Unilateral ligation of thecarotid artery was done in 110 gerbils anesthetized with either and 78 anesthetized with pentobarbital sodium. Thirty of the ether-anesthetized animals received no tratment. Starting one hour after ligation, 38 animals from each anesthesia group were given 3 mg/50 gm of body wight of pentobarbital intraperitoneally every eight hours for 72 hours. The remaining animals received normal saline on the same treatment schedule. Morbidity and mortality were recorded for the next six days. Eighty-five animals (45.2%) developed cerebral infarction. Only ten of the 38 animals (2693%) anesthetized and treated with pentobarbital developed infarction. The percentage of infarction in the remaining groups ranged from 45% to 53%. There were no significant differences among the groups in the number of deaths or in the location or extent of the lesion.

Anticardiolipin antibodies and the risk of recurrent thrombo-occlusive events and death. The Antiphospholipid Antibodies and Stroke Study Group (APASS).

We tested the hypothesis that an anticardiolipin antibody (aCL) titer of > or = 10 IgG phospholipid (GPL) at the time of an index ischemic stroke is associated with an increased risk of subsequent thrombo-occlusive events or death. First-time ischemic stroke patients from the Antiphospholipid Antibodies and Stroke Study Group's Prevalence Study were followed prospectively for a median time of 24 months for any thrombo-occlusive event or death. There was no significant difference for the endpoint of stroke, death, myocardial infarction, transient ischemic attack, deep venous thrombosis, pulmonary embolus, or arterial embolus between the aCL positive and negative patients. Although a single aCL value of > or = 10 GPL at the time of an initial ischemic stroke is a significant independent risk factor for stroke, when adjusted for other stroke risk factors in our study population, aCL positivity did not confer a significantly increased risk for subsequent thrombo-occlusive events or death.

Transient ischemic attacks: a prospective study of 225 patients.

Between the years 1964 and 1973, 225 patients with transient ischemic attacks (TIAs) due primarily to atherosclerosis were evaluated and treated. They have now been followed for from 3 to 14 years (average 5.5 years). As of 1976, 82 of the 225 patients were dead, 21 from cerebral infarction, 52 from heart disease and nine from other causes. Of the 56 untreated patients, 11 (19 percent) had cerebral infarctions, four (7 percent) of which were fatal; six (11 percent) were still having TIAs. Of the 45 patients medically treated, 10 (24 percent) had cerebral infarctions, three (7 percent) of which were fatal; 11(25 percent) still experienced TIAs. In the surgical group of 124, 27 (21 percent) had postoperative cerebral infarctions, seven (6 percent) of which were fatal; 23 (18 percent) had cerebral infarctions during follow-up, of which seven (6 percent) were fatal; and 15 (12 percent) were still having TIAs. No statistically significant differences (p less than 0.05) related to cerebral infarction or TIAs developed among the three groups. The majority (23 percent) eventually succumbed to myocardial infarction, leading us to conclude that great emphasis must be placed upon TIAs as a warning for cardiac as well as cerebrovascular disease.

Basilar artery embolism. Clinical syndrome and neuroradiologic patterns in patients without permanent occlusion of the basilar artery.

The objective of this study was to clarify the clinical and radiologic features, risk factors, and prognosis of basilar embolism without permanent basilar artery occlusion. Forty-five patients (mean age, 59 years) with basilar artery embolism participated in the study. Patients with basilar artery occlusion were excluded. The Glasgow Coma Scale (GCS) score on admission was < 7 in five patients, 7 to 12 in 11 patients, and > 12 in 29 patients. Etiologic factors were cardiac arrhythmia (17 patients), vertebral artery occlusion (12 patients), cervical spine trauma (4 patients), embolism following angiography (2 patients), and surgery (1 patient). MRI was performed in 17 patients and CT in 39 patients. Radiologic examinations were initially normal in 14 patients and remained normal in three patients. Final infarct localization was the thalamus (36 patients), cerebellum (20 patients), posterior cerebral artery territory (21 patients), midbrain (12 patients), and pons (8 patients). Eight to 12 weeks after stroke 12 patients were without clinical signs (Glasgow Outcome Scale [GOS] 1), 15 patients had minor neurologic deficits (GOS 2), 10 were severely disabled (GOS 3), and eight patients had died (GOS 5). Outcome correlated with GCS on admission (p < 0.0001) and with the number of ischemic lesions (p = 0.0001). The typical syndrome is an acute loss of consciousness followed by multiple brainstem symptoms. Usually, clinical symptoms improve rapidly and, in some patients, completely. Compared with basilar occlusion, basilar embolism has a relatively low mortality and outcome is frequently excellent.

Increased platelet and red cell counts, blood viscosity, and plasma cholesterol levels during heat stress, and mortality from coronary and cerebral thrombosis.

Recorded deaths from coronary and cerebral thrombosis rise markedly in heat waves. In a British heat wave with little or no distortion due to air-conditioning, outside temperatures of 34.6 degrees C (maximum) and 20.8 degrees C (minimum) were followed by peak mortalities from coronary and cerebral thrombosis one to two days later. Experimental exposure of volunteers to moving air at 41 degrees C for six hours caused core temperature to rise 0.84 degree C, weight to fall 1.83 kg with sweating despite access to water, heart rate to increase 32 beats per minute, and arterial pressure to fall, particularly on standing. The red blood cell count increased 9 percent, and blood viscosity increased 24 percent, mostly after the first hour. The platelet count rose 18 percent, and the platelet volume fell, mostly in the first hour. The plasma cholesterol level increased 14 percent without a change in distribution among lipoprotein fractions. The changes seem able to explain the increased mortality from arterial thrombosis in hot weather.

Surgical resection in idiopathic hypertrophic subaortic stenosis with a combined approach through aorta and left ventricle.

The operative technique and the immediate and long-term results are described in 49 patients who have undergone myectomy for IHSS. There was a 4 percent early and 12 percent late mortality rate and surviving patients have been followed for up to 13 years. The operation has resulted in a striking symptomatic improvement in most patients in association with relief of outflow obstruction and mitral regurgitation. Chronic atrial fibrillation was poorly tolerated but has not occurred as a late complication in any of the patients operated upon.

Craniocerebral trauma induces hemorheological disturbances.

Several mechanisms are involved in the development of secondary ischemic brain damage, including microthrombi formation, which is thought to play a prominent role. Ninety-four autopsy cases were macro- and microscopically examined by specific staining for fibrin, 74 of which showed cortical contusion after a craniocerebral trauma. Twenty cases with no neurological pathology were used as controls. Traumatic cases comprised 52 males and 22 females, with a mean age of 48 years; most cases died in the first 48 h. The total number of fibrinous microthrombi in a slice of each hemisphere was determined. The mean number of microthrombi found in contused hemisphere was 152 (37-283), with 88 in the contralateral hemisphere (21-139) as compared to 13 (0-27) in control cases. Differences were statistically significant. Globular microthrombi or "shock bodies" (2-60 micro diameter) were present in five cases. Enhanced presence of microthrombi in contused brain areas, higher incidence in young people, an increase in the amount of microthrombi up to the 9th day after injury and involvement of the contralateral hemisphere free of contusion foci were all demonstrated. Microthrombi would therefore seem to be one of the central secondary events after brain trauma to bear in mind when designing treatment strategies.

Association of silent myocardial ischemia with new atherothrombotic brain infarction in older patients with extracranial internal or common carotid arterial disease with and without previous atherothrombotic brain infarction.

OBJECTIVE: To correlate silent myocardial ischemia with the incidence of new atherothrombotic brain infarction (ABI) in older patients with 40 to 100% extracranial carotid arterial disease (ECAD) with and without prior ABI. DESIGN: In a prospective study of 208 older patients with 40 to 100% ECAD diagnosed by carotid duplex ultrasonography, 24-hour ambulatory electrocardiograms were obtained to detect silent myocardial ischemia. At 42-month mean follow-up, silent myocardial ischemia was correlated with the incidence of new ABI in patients with and without prior ABI. SETTING: A large long-term health care facility where 208 older patients with 40 to 100% ECAD and technically adequate 24-hour ambulatory electrocardiograms for detecting silent myocardial ischemia were studied. PATIENTS: The 208 patients included 68 men and 140 women, mean age 81 +/- 8 years (range 60 to 100). One-hundred three (50%) of the patients had prior ABI. MEASUREMENTS AND MAIN RESULTS: Sixty-nine (33%) of the 208 patients had silent myocardial ischemia. Mean follow-up was 42 +/- 25 months (range 3 to 101 months). At follow-up, the incidence of new ABI was 64% in patients with prior ABI and 32% in patients with no prior ABI (P < .0001). At follow-up, the incidence of new ABI was 65% in patients with silent ischemia and 40% in patients with no silent ischemia (P = .0005). The multivariate Cox regression model showed that patients with prior ABI have a 2.5 times higher chance of developing new ABI than those without prior ABI after controlling other prognostic variables. Patients with silent ischemia have a 2.1 times higher probability of developing new ABI than those without silent ischemia after controlling other prognostic variables. CONCLUSIONS: Prior ABI and silent ischemia are independent risk factors for the development of new ABI in patients with 40 to 100% ECAD. This probably reflects that silent ischemia is a marker for more advanced or more significant atherosclerotic disease rather than a causal factor for ABI.

Complete thromboendarterectomy of the calcified ascending aorta and aortic arch.

BACKGROUND: Arteriosclerotic plaques of the ascending aorta and transverse arch increase the operative risk of cardiac operations and are strong predictors for late cerebrovascular events. METHODS: Twenty-two patients, mean age 68 +/- 6 years (range, 55 to 77 years), with grade IV + V plaques of the ascending aorta and transverse arch underwent coronary artery bypass grafting (n = 21) and aortic valve replacement (n = 8). Cerebrovascular emboli from unknown sources were found preoperatively in 8 patients (36%). All were in sinus rhythm. Complete thromboendarterectomy of the ascending aorta and transverse arch was performed during hypothermic circulatory arrest. After 21 +/- 12 months (range, 4 to 44 months), magnetic resonance imaging and transthoracic echocardiography of endarterectomized vessels was performed. RESULTS: There was one perioperative death (4.5%), one early (4.5%), and one late (4.7%) adverse neurologic event. Follow-up examinations revealed normal diameters of the endarterectomized aorta. CONCLUSIONS: For patients with grade IV + V plaques, thromboendarterectomy of the ascending aorta and transverse arch can be performed with an acceptable surgical risk and a low recurrence rate for cerebrovascular events. Dilatation of the endarterectomized aorta was not observed.

Valve replacement in patients with endocarditis and cerebral septic emboli.

Cerebral septic emboli complicate the cases in 20% to 40% of patients with left-sided endocarditis but the management of these patients who require a valvar operation remains unclear. From 1980 to 1988, the incidence of cerebral septic embolus was 42% (n = 45) among 106 patients with endocarditis who underwent valve replacement at the University of Illinois Hospital in Chicago. Of these 45 patients, 69% (n = 31) had symptomatic cerebral septic infarctions and 31% (n = 14) were asymptomatic. Findings on cerebral computed tomographic scans included ischemic infarcts (n = 36, 80%), hemorrhagic infarcts (n = 5, 11%), normal studies (n = 2, 4%), and unknown (n = 2, 4%). Neurological complications after valve replacement included postoperative strokes (n = 6, 6%), cerebral abscesses (n = 2, 2%), and seizure (n = 1, 1%). The presence of a hemorrhagic infarct preoperatively predisposed to a perioperative stroke (p less than 0.05). In conclusion, cerebral septic infarctions, both symptomatic and asymptomatic, are common among patients with endocarditis referred for valvar operation. In the absence of a hemorrhagic infarct, valve replacement can be performed with minimal risk of a perioperative stroke.

Cerebral intraarterial fibrinolysis at the crossroads: is a phase III trial advisable at this time?

PURPOSE: To describe the rationale for fibrinolysis, review the state of the art in cerebral fibrinolysis, and discuss whether it is time for phase III studies of cerebral intraarterial fibrinolysis. METHODS: Critical review of the literature with statistical reevaluation of significant clinical data. RESULTS: There are abundant phase III data supporting the use of thrombolysis in the cardiovascular system. However, there are no published phase III trials of intraarterial fibrinolysis in stroke. All reports of cerebral intraarterial fibrinolysis are case series. The studies are typically small with variable treatment protocols and designs that are susceptible to bias. The only analysis comparing cerebral intraarterial fibrinolysis with conventional therapy is based on nonconcurrent controls. CONCLUSIONS: Stroke is common and costly. Acute stroke intervention with fibrinolytic drugs is theoretically justified. Studies done to date have significant, inferential limitations. The data suggest an association between thrombolysis, recanalization, and prognosis. However, imprecision and inadequate control of systematic error preclude conclusions regarding clinical outcomes. Randomized, controlled trials are needed to establish the clinical value of cerebral local intraarterial fibrinolysis. However, cerebral local intraarterial fibrinolysis availability, the cerebral local intraarterial arterial fibrinolysis learning curve, anticipated technological advances, unresolved procedural controversies, and ethical and fiscal considerations make a large phase III trial impractical and ill-advised at the present time. Additional basic research is needed to set the stage for a successful clinical trial.

Collateral circulation and outcome after basilar artery thrombolysis.

BACKGROUND AND PURPOSE: This study was undertaken to examine the relationship between collateral flow and outcome after local intraarterial thrombolytic treatment for basilar artery thrombosis. METHODS: Twenty-four patients with symptomatic basilar thrombosis were treated with intraarterial urokinase. Angiograms at the time of treatment were analyzed to characterize collateral flow. The number of posterior communicating arteries (PCoAs) and the degree of collateral filling of the basilar artery were then compared with symptom duration before treatment, with Glasgow Coma Scale (GCS) score at the time of treatment, with 90-day modified Rankin score, and with 90-day survival status. RESULTS: Of the 20 patients who had carotid artery injections at the time of the thrombolytic procedure, two had no PCoA, eight had one PCoA, and 10 had two PCoAs. Nine had no collateral opacification of the basilar artery, six had collateral opacification of the distal basilar artery, and five had collateral opacification of the distal and proximal basilar artery. Ninety-day survival was 38%; 25% of patients had good neurologic outcomes. No correlation was found between the number of PCoAs and symptom duration, pretreatment GCS score, survival, or neurologic outcome. Duration of symptoms before treatment was longer in patients with collateral flow to the basilar artery. Basilar artery collateral flow did not correlate with survival, but it did correlate with neurologic outcome for the 12 patients with middle or distal basilar artery thrombus in whom collateral flow to the basilar artery was assessed (83% with collateral flow had good neurologic outcomes, but only 17% without collateral flow had good outcomes). All six patients with proximal basilar artery thrombus in whom collateral flow was assessed died, independent of the collateral flow observed. CONCLUSION: In symptomatic acute basilar artery thrombosis, neurologic outcome was better after intraarterial thrombolysis in patients who had collateral filling of the basilar artery, except in cases of proximal basilar thrombosis. Patients with collateral filling of the basilar artery also tolerated longer symptom duration.

Relationship between clot location and outcome after basilar artery thrombolysis.

PURPOSE: To identify factors that predict survival and good neurologic outcome in patients undergoing basilar artery thrombolysis. METHODS: Over a 42-month period, 20 of 22 consecutive patients with angiographic proof of basilar artery thrombosis were treated with local intraarterial urokinase. Brain CT scans, neurologic examinations, symptom duration, clot location, and degree of recanalization were analyzed retrospectively. RESULTS: Overall survival was 35% at 3 months. Survival in patients with only distal basilar clot was 71%, while survival in patients with proximal or midbasilar clot was only 15%. At 3 months, 29% of patients with distal basilar clot and 15% of patients with proximal or midbasilar clot had good neurologic outcomes (modified Rankin score of 0 to 2 and Barthel index of 95 to 100). Complete recanalization was achieved in 50% of patients; 60% of those survived and 30% had good neurologic outcomes. Of patients with less than complete recanalization, only 10% survived. Neither duration of symptoms before treatment (range, 1 to 79 hours), age (range, 12 to 83 years), nor neurologic status at the initiation of treatment (Glasgow Coma Scale score range, 3 to 15) predicted outcome. Pretreatment CT findings (positive or negative for related ischemic changes) did not predict outcome or hemorrhagic transformation. CONCLUSION: The single best predictor of survival after basilar thrombosis and intraarterial thrombolysis was distal clot location. Complete recanalization favored survival. Radiologically evident related infarctions, advanced age, delayed diagnosis, and poor pretreatment neurologic status did not predict poor outcome and therefore should not be considered absolute contraindications for intraarterial thrombolysis in patients with basilar artery thrombosis.

Intraarterial thrombolysis in vertebrobasilar occlusion.

PURPOSE: To report our experience using intraarterial thrombolysis in the treatment of vertebrobasilar occlusion. METHODS: Twelve patients with 13 angiographically proved thromboses of the vertebrobasilar system underwent local intraarterial thrombolysis with urokinase. Angiographic and clinical outcomes were analyzed with respect to clinical examination at presentation, arterial occlusion patterns, and time to recanalization. RESULTS: The overall mortality was 75%. Recanalization could not be achieved in 3 of 13 treatments; all patients in whom recanalization failed died. The mortality rate was 60% in those patients in whom recanalization was successful. Coma or quadriparesis at the time of therapy uniformly predicted death. There were two cases each of bilateral proximal vertebral occlusions and midbasilar occlusions and nine cases of bilateral distal vertebral occlusions. There were three cases of fatal rethrombosis after initial successful thrombolysis. The mortality rate in the recanalized group before rethrombosis was 30%. There were two fatal hemorrhages of the central nervous system. CONCLUSION: Recanalization of the vertebrobasilar system is necessary but not sufficient for effective treatment of vertebrobasilar occlusive disease. The site of occlusion may help predict angiographic and clinical outcome. Time to initiation of thrombolysis is not an invariable correlate of survival, although clinical condition at presentation may be. Rethrombosis and hemorrhage are significant problems affecting mortality after successful thrombolysis.

Thrombolytic therapy in acute occlusion of the intracranial internal carotid artery bifurcation.

PURPOSE: To evaluate efficacy and clinical benefit of early thrombolytic therapy in intracranial internal carotid artery occlusion. METHODS: Thirty-two patients (mean age, 56 years) with acute intracranial internal carotid artery occlusion were studied clinically and with CT and angiography before and after thrombolytic therapy with intravenous alteplase (n = 16), superselective intraarterial alteplase (n = 8), and superselective intraarterial urokinase (n = 8). RESULTS: Initial CT showed a large parenchymal hypodensity in 11 (34%) patients, a small hypodensity in 15 (47%) patients, and no hypodensity in 6 (19%) patients. Recanalization after thrombolytic therapy was observed in 4 patients (12.5% in each treatment group). Follow-up CT showed six hemorrhagic infarcts and four parenchymal hematomas unrelated to recanalization, alteplase, or urokinase administration, but commonly associated with intraarterial treatment. Clinical outcome was fatal in 53%, poor in 31%, and moderate or good in 16% of the patients. Outcome was equal in different treatment groups and closely linked to both the quality of leptomeningeal collaterals and the extent of parenchymal hypodensity on the first CT. CONCLUSION: Because intravenous or intraarterial treatment with alteplase or urokinase fails to recanalize the vascular obstruction, it does not improve the prognosis of intracranial internal carotid artery occlusion over that of the natural course. Improved results may be possible with novel recanalization techniques.

Emergency embolectomy for acute occlusion of the middle cerebral artery.

Twenty cases treated with emergency embolectomy for acute occlusion of the middle cerebral artery were reviewed. There were 10 males and 10 females, with an average age of 55 years. The left middle cerebral artery was involved in 17 patients and the right in three. Flow was restored in 16 patients (75%). The embolus originated in the heart in seven, the carotid artery in seven, the aorta in three, an aneurysm in one, and an indeterminate source in two. It was technically most difficult to achieve patency with atheromatous emboli from the aorta. Two patients (10%) had an excellent result with no neurological deficit, five (25%) were left with a minimal deficit but were employable, seven (35%) had a fair result but were still independent and employable, four (20%) did poorly, and two (10%) died. Patients with an associated ipsilateral carotid artery occlusion did poorly. Collateral flow, as judged from preoperative angiograms, was the best predictor of outcome.

The heart in the primary antiphospholipid syndrome.

Twenty consecutive female patients (mean age 30.1 years) with Primary Antiphospholipid Syndrome (PAPS) were studied cardiologically through noninvasive methods and compared with 20 age-and sex-matched healthy subjects. On physical examination 13/20 patients (65%) with PAPS had a valvular abnormality. In 12/20 (60%) patients with PAPS the ECG was abnormal, mainly due to sinus tachycardia in 5 (25%) and acute myocardial infarction in 3 (15%). In 7/20 patients with PAPS (35%) abnormal pulmonary findings were detected by X-ray and in 6 (30%) they were related to dilated pulmonary arch and pulmonary hypertension. In 14/20 cases (70%) with PAPS, abnormal echocardiographic findings were present; 13/20 patients (65%) had valvular complications attributable to PAPS: mitral insufficiency in six cases; mitral valve prolapse in three and aortic insufficiency in three. Two had pulmonary artery hypertension and two, tricuspid regurgitation (one attributable to PAPS). All valve diseases were regurgitant with mild to moderate hemodynamic repercussion. No stenotic lesions were detected. The mean mitral thickness in patients with mitral valve involvement was 7.0 +/- 1.6 mm, compared to 2.8 +/- 0.7 mm in patients with normal valves and 3.1 +/- 0.9 mm in the control group (p < 0.001). The mean aortic valve thickness in patients with aortic valve involvement was 3.6 +/- 0.5 mm compared to 1.5 +/- 0.3 mm in patients with normal valves and 1.4 +/- 0.4 in the control group (p < 0.001). None of the patients from the control group had valve disease (p < 0.0001). Three cases (15%) had pericardial effusion diagnosed by echo. Two patients with PAPS died during the 4.7 +/- 1.2 years of cardiological follow up, due to acute myocardial infarction and embolic cerebrovascular accident, respectively. In conclusion, cardiologic complications are common in PAPS, including left side regurgitant lesions that might be hemodynamically significant, acute myocardial infarction, pericardial effusion and pulmonary hypertension.

Clinical predictors of outcome in ischemic stroke.

Stroke is the most common neurologic disorder of adults and one of the three leading causes of death in North America. Results of a study suggest that inexpensive and easily obtained clinical variables can be used to predict which patients with stroke will do well and which will not. This capability is invaluable to physicians, patients with stroke and their families, and future investigators.

Predictive factors for postoperative cerebral complications in patients with thoracic aortic aneurysm.

OBJECTIVE: Perioperative factors affecting the outcomes of postoperative brain function in patients with thoracic aortic aneurysm were demonstrated. PATIENTS AND METHODS: From December 1977 to September 1994, 745 patients with thoracic aortic aneurysm underwent 846 operations. The mean age at surgery was 57.1 +/- 14.2 years old. Four hundred seventy-four patients had true aneurysm and 372 had aortic dissection. Two hundred forty-four patients underwent repair in the ascending aorta, 189 arch repair, 242 repair in the descending aorta. 79 replacement of the thoracoabdominal aorta, and 92 extra-anatomical bypass or thrombo-exclusion of the aorta. Conventional cardiopulmonary bypass was used in 297 patients, partial cardiopulmonary bypass through femoral access in 167, selective cerebral perfusion in 253, deep hypothermic circulatory arrest and retrograde cerebral perfusion in 50, temporary shunt in 29, and no circulatory support was applied in 50. Postoperative cerebral complications were divided into permanent cerebral dysfunction. RESULTS: The early mortality rate was 15.5% (131 patients). Incremental risk factors for hospital mortality were non-preexisting cardiac lesions, ruptured aneurysm, postoperative cerebral complications, sepsis, bleeding, low output syndrome and renal failure. Cerebral complications occurred in 81 patients (9.6%), involving 47 permanent and 34 transient sequelae. The early mortality rate in patients with postoperative brain damage was 42.0%. The etiologies of the brain damage diagnosed by computed tomography were embolism in 41 patients, cerebral hypoperfusion in 16 and unknown in 24. Incremental risk factors for postoperative cerebral complications were: operation early in the series advanced age at surgery, preoperative renal failure, aortic arch lesions, atherosclerotic aneurysm, aortic arch procedures and clamping of the aortic arch. CONCLUSIONS: Although there was an increased incidence of advanced age and complex lesions in patients with aortic aneurysm, an improvement in surgical results has recently been achieved using advanced diagnostic and surgical techniques.

Is it possible to reduce the number of cerebral complications in carotid endarterectomy?

An effort was made to reduce the number of cerebral complications in carotid endarterectomy by focusing attention on the perioperative prevention of embolism and thrombosis in a prospective series of carotid endarterectomies (group I, n = 62) performed during a 24 months' period 1982-1984 in Tampere University Hospital. Controls were two other series of carotid endarterectomies (group II, n = 60, and group III, n = 60) comprising all 120 carotid endarterectomies carried out by other trained vascular surgeons in the same hospital during the years 1976-1984. During the first 30 postoperative days there were no deaths, one stroke and one TIA in group I. On the other hand there were three deaths, seven strokes and five TIAs in group II and four deaths, two strokes and three TIAs in group III. The number of cerebral complications in carotid endarterectomy was significantly less in the study group than in the control groups, and thus it was possible to reduce the perioperative complications in carotid endarterectomy possibly by pre- and postoperative drug treatment and routine intraoperative heparinization.