A series of 14 patients with intracerebral hemorrhage after carotid endarterectomy is reviewed. This complication occurred in 0.6% of 2362 consecutive carotid endarterectomies performed at the Mayo Clinic from 1972 through 1986. All hemorrhages occurred within the first 2 weeks after operation and were ipsilateral to the side of the operation. Eight patients died, and only two made a good recovery. Significant risk factors are hypertension and chronic hemispheric hypoperfusion with impaired autoregulation. The "normal pressure-hyperperfusion breakthrough" syndrome was considered to be operative in 12 of the 14 patients. Nine patients had documented hyperperfusion (at least 100% increase of baseline cerebral blood flow) at the time of surgery. In an additional three patients, normal perfusion-pressure breakthrough was inferred by the clinical course and radiological findings, as well as by the absence of alternative explanations. Patients at risk for postendarterectomy intracerebral hemorrhage include those who have a clinical history suggestive of hemodynamic cerebral ischemia, severe carotid stenosis with limited hemispheric collateral flow, and postendarterectomy hyperperfusion, as measured by intraoperative cerebral blood flow. To minimize the risk of hemorrhage in these patients, strict maintenance of blood pressure at normotensive or even relatively hypotensive levels during the intraoperative and early postoperative periods is advised.

The bifurcation beyond the circle of Willis (that is, the MCA) closely approximated optimality principles, whereas the bifurcations within the circle of Willis (that is, the distal ICA and BA) did not. This indicates that the confluence of hemodynamic forces plays an important role in the distribution of work at bifurcations within the circle of Willis. In addition, the observed branch angles were predictors for the presence of aneurysms.

In this study, 24 aneurysms occurring in 23 patients under the age of 18 years (mean 12 years) are analyzed. The male:female ratio was 2.8:1, and the youngest patient was 3 months old. Mycotic lesions and those associated with other vascular malformations were excluded. Forty-two percent of the aneurysms were located in the posterior circulation, and 54% were giant aneurysms. Presenting symptoms included subarachnoid hemorrhage in 13 and mass effect in 11. Several of these aneurysms were documented to rapidly increase in size over a 3-month to 2-year period of observation. All aneurysms were surgically treated: direct clipping was performed in 14; trapping with bypass in four; trapping alone in four; and direct excision with end-to-end anastomosis in two. The postoperative results were excellent in 21 aneurysms (87%), good in two (8%), and poor in one. The pathogenesis of cerebral aneurysms is reviewed.

The vein of Galen malformation is a midline arteriovenous fistula with aneurysmal dilatation of the vein of Galen. The clinical details of diagnosis and treatment in 13 patients with such lesions together with a review of 232 cases collected from the literature are presented in this report. There were 132 males, 77 females, and 36 cases in which the sex was not stated. Eighty patients presented as neonates, 82 were 1 to 12 months old, 39 were 1 to 5 years old, 22 were 6 to 20 years old, and 22 were over the age of 20. The most common presenting symptoms were congestive cardiac failure (110 cases), raised intracranial pressure secondary to hydrocephalus (94 cases), cranial bruit (57 cases), focal neurological deficit (37), seizures (26 cases), and hemorrhage (25 cases). The most characteristic vascular supply to the midline fistula involved multiple bilateral vessels, although bilateral posterior cerebral and unilateral posterior cerebral supply was relatively common. The overall figures for treatment and outcome showed that 91 patients (37.1%) were treated by direct operation and 29 patients (11.3%) were treated by other forms of operation, predominantly shunting or remote vessel ligation. Forty-six patients (18.8%) were treated by medical means (digoxin, diuretics, and ventilatory support). In 79 patients (22.2%), there was no treatment or no details of treatment were available. There was an overall series mortality of 55.6% (no details were available in 33 cases) and a 37.4% mortality for surgically treated cases. After operation, there was a 46.3% incidence of significant morbidity in surviving patients. Neonatal patients fared worst, with an overall mortality of 64 of 70 cases (91.4%) where details were available. The outcome was equally bad for surgically and conservatively treated cases. Operation in the 1- to 12-month age group was more successful, but still carried a mortality of 31.7%, with a significant morbidity in approximately half of the surviving patients. Over the age of 1 year, the surgically treated patients had a 25.6% mortality and a 42.3% major morbidity in survivors. Consideration is given to some of the ways in which these figures may be improved, in particular a staged approach during the neonatal period, with the use of selective embolization or occlusion of vessels to reduce the volume of the arteriovenous shunt until the patient is older and better able to tolerate major operation.