Since the first STA-MCA anastomosis was performed by Yasargil in 1967, extra-intracranial by-pass surgery is widely used for treatment of occlusive cerebrovascular lesions.
In our 171 STA-MCA anastomoses, there is no recurrence of TIA in 93.5% of TIA or RIND patients.
The indication, the technical aspect and the follow up study of this procedure have been reported.
The advantage and the limitation of EC-IC bypass have been discussed with the clinical recording of pre-, intra- and postoperative CBF measurement and computerized EEG monitoring.
The results of the OA-PICA anastomosis for vertebral occlusion and the STA-MCA anastomosis for Moya-Moya disease have been reported.

View full abstract

Investigations of head injury using monkeys were performed to elucidate the effect of linear as well as rotational acceleration impact to produce brain contusion and concussion.
The results of investigations using pure linear acceleration impact, which had already been reported, showed that concussion occurred in all monkeys, but that no visible lesion except subarachnoid hemorrhage was found, even though negative pressure of under minus 1 atmosphere was recorded in the contralateral side of the impact.
In this paper, the result of investigation using a newly designed and constructed impactor system to deliver rotational acceleration impact is reported. The impactor system was accelerated by compressed air, 1.5-3.5 kg/cm² in pressure, and velocity of the impact shaft was 20.2-36.4 m/sec. Under light anesthesia with ketamine hydrochloride, the impact was delivered to the occipital area of 18 monkeys, weighing 5.9 to 10.5 kg, through a restraint mask which was made of iron frame and plaster of Paris for each animal in order to obtain a broad impact area and to avoid fracture. Averaged frontal angular acceleration ranged from 29, 000 to 281, 000 rad/sec² and its duration was 0.64-4.0 msec. Averaged frontal acceleration was 480-1, 080 G's and its duration was 1.26-4.50 msec.
Skull fractures were found in 6 monkeys. Four of them had comminuted and depressed fractures that became fatal within 10 minutes after impact due to laceration of the medulla oblongata and/or of the main intracranial vessels. The remaining two had only multiple linear fractures without depression and these animals recovered from initial concussion. Among the 12 monkeys without any fracture, 11 recovered from the initial concussion within 10 minutes after the impact. One monkey died in the state of concussion. There was noted correlation between the severity of concussion and changes in systemic blood pressure immediately after the impact, which is classified into three groups (Fig. 3).
Among the 12 monkeys without fracture, 7 had cortical hemorrhage and contusion at the para-sagittal area, tip or base of the frontal and temporal lobe (viz. contre coup injury), or the brain stem and the upper cervical cord. Coup injury in the occipital lobe did not result in this series with the exception of the comminuted depressed fracture group. Intramedullary hemorrhage in the upper cervical cord was found in 5 monkeys; 2 with and 3 without fracture. Four among these 5 survived apnea followed by irregular respiration for several seconds after the impact.
As reported previously, pure linear acceleration impact did not produce any visible brain lesion. On the other hand, rotational acceleration impact caused brain damages. It was, therefore, supposed that rotational acceleration impact was a greater cause for producing blunt brain injury than translational impact. Furthermore, it was suggested that the contusion threshold of the brain was close to the concussion threshold.

View full abstract

Oxyhemoglobin (oxyHb) is postulated as the most potent trigger of prolonged vasoconstriction after subarachnoid hemorrhage, while methemoglobin (metHb) is much less potent. It is unknown whether or not oxyHb may directly act upon the vessel wall in subarachnoid hemorrhage. OxyHb may be responsible for vasoconstriction not directly but indirectly through superoxide anion radical (O·^-₂) derived from auto-oxidation of oxyHb.
Generation of O·^-₂ and activity of superoxide dismutase (SOD) was studied in the incubated whole blood or washed erythrocyte of human artery under the condition simulated to subarachnoid hemorrhage.
Results indicated that generation of O·^-₂ and activity of SOD were preserved during incubation for 8 days at 37°C. The preserved SOD activity might indicate that O·^-₂ does not react directly upon the vessel wall to bring about vasoconstriction. It was suggested that other noxious free radicals or active oxygen dismutated from O·^-₂ might participate in prolonged vasoconstriction.
The phasic changes of oxyHb on the course of auto-oxidation was analyzed with electron spin resonance (ESR). The characteristic changes of ESR signals of ferric protein compound from high to low spin corresponded to the changes from oxyHb to superoxide nietlIb, metllb and hemichronie during the incubation period. These changes were observed in the cerebrospinal fluid from patients who suffered from prolonged vasospasm after subarachnoid hemorrhage.
Oxyllb may participate in prolonged vasoconstriction indirectly.

View full abstract

It is well known that oxyhemoglobin in its autoxidation to methemoglobin produces active species of oxygens (O·^-₂, H₂O₂, OH^·, ¹O₂), which in turn generate toxic free radical reactions, such as peroxidation of polyunsaturated fatty acids in the biomembrane. Since active species of oxygen and lipid hydroperoxides possess tissue toxicity, it is easy to speculate that they play some role in the production of chronic cerebral vasospasm. Based on this assumption, experimental as well as clinical studies were carried out.
Fresh arterial blood of a dog was incubated at 37°C for 14 days and the daily change of its peroxide value in each sample by thiobarbituric acid test was measured. To follow the conversion of oxyhemoglobin to methemoglobin, the spectrophotometric absorbance curve of each sample was traced. The peroxide value of the supernatant of each sample was initially low and gradually increased to reach a plateau in 3 to 5 days. This high level of the peroxide value was maintained thereafter. By the analysis of the spectrophotometric absorbance curves, it was confirmed that the conversion of oxyhemoglobin to methemoglobin occurred in parallel with the increase of peroxide value.
For the purpose of verifying the tissue toxicity of lipid hydroperoxide to the cerebral vessels, 15-hydroperoxy arachidonic acid was injected into the subarachnoid space of dogs and the histological change of the basilar artery was examined by electron microscope. Myonecrotic changes were observed in the canine basilar artery after intracisternal injection of 15-hydroperoxy arachidonic acid.
With 25 patients sustaining SAH due to rupture of an intracranial aneurysm, peroxide content of lumbar CSF was studied to examine the correlation between the peroxide formation and the occurrence of vasospasm. The peroxide values in patients with vasospasm were higher than those without vasospasm. As to the time course of peroxide value, a statistically significant difference (p<0.01) between the groups with and without vasospasm was revealed in I to 3rd and 7 to 9th days after the onset of SAH. These data strongly suggest that free radical reactions initiated by clot lysis play a significant role in the genesis of vasospasm.

View full abstract

The combination of scanning electron microscopy of microcorrosion casts with dark field microscopy of the cerebral vascular system has made it possible to clarify in detail the morphology of microcirculation in the rat. It was concluded by preliminary experiments using vasospasm provoked by direct administration of vasopressin to the basilar artery that the polyester casts could clearly represent the morphological change occurring just before commencement of perfusion fixation though slight artificial modification was observed during the succeeding procedures. The architecture of the blood vessels was observed in the parietal cortex and the ventrolateral nucleus of the thalamus with special reference to structures contributing to regulation of the microcirculation. By both of the two methods the circular foldings which are supposed to be a constrictive structure were frequently observed even in arterial vessels up to 10 microns constituting the capillary plexus. Similar structure was also seen in the medium to small sized venous vessels of the regions observed. The results obtained seem to need further studies.

View full abstract

A clinical survey of intracranial malignant lymphoma was conducted based on 10 cases experienced in our department during the past 15-year period. The following diagnostic clues considered characteristic to malignant lymphoma are presented :
1) The peak of incidence was patients in their thirties to fifties. Neurological deficits and increased intracranial pressure developed within one to 3 months.
2) Cerebrospinal fluid showed increased protein content of 100-250 mg/dl without pleocytosis and occasionally increased Ig-M, Ig-A and Ig-G appeared in immunoelectrophoresis.
3) Association of paraproteinemia was frequently found, which may indicate the generalization of this tumor.
4) Angiography showed mostly avascular masses and occasionally tumor stain. Cerebral deep medullary veins were shown in the capillary phase.
5) Radioisotope brain scan revealed hot spots in all cases.
6) Computed tomography was the most reliable method for definite diagnosis showing iso-density or slightly high density, nodular lesion in plain scan and homogenous enhancement by contrast medium.
7) The frequent sites of this tumor were subcortical white matter of the cerebrum, basal ganglia, corpus callusum, and cerebellum.

View full abstract

A modified extended middle cranial fossa approach for acoustic tumors as well as surgical techniques and surgical results are described in detail. This method is based upon a combination of Morrison and King's translabyrinthine-transtentorial approach and Bochenek's extended approach through the middle cranial fossa. It has the following advantages: 1) operation can be done in the supine position; 2) it is the shortest approach to the cerebellopontine angle; 3) it yields a more favorable operative field and enables easier operative procedure in the internal auditory canal; 4) even large tumors can be excised in one-stage operation; 5) it is easy to identify and preserve the facial nerve, the anterior inferior cerebellar artery, and its loop in the posterior cranial fossa; and 6) tumors can be orthoptically isolated and safely and certainly excised without compressing the cerebellum and brain stem.
There were no hospital deaths in 29 patients who were treated by this operative method. Total and radical removal were performed in 21 patients (78%) ; 13 of 19 (68%) large tumor cases (9 total removal and 4 radical removal), 2 (100%) medium-sized tumors, and 6 (100%) small tumors localized in the internal auditory canal. The facial nerve was anatomically preserved in the posterior cranial fossa and the internal auditory canal from 19 (70%) of the 29 cases; 14 (69%) of the 21 cases which received total removal or radical removal, and 8 (62%) of the large tumor 13 cases. Intracranial end to end anastomosis was conducted in 2 of 8 sacrificed facial nerves, and facial-hypoglossal anastomosis was conducted in the remaining 6 cases, one week after the operation. Long-term follow-up of facial nerve function revealed that 26 of the 27 cases were normal or satisfactory. As for postoperative complications, 3 cases of CSF leakage (one of them was surgically occluded, and the rest healed spontaneously), and one case of postoperative hemiparesis were observed. There was no postoperative temporal epilepsy or aphasia.
This method, that is, the neurosurgical-otological team approach, is compared with other methods, and advantages and superiorities of this method are emphasized.

View full abstract

Clinical analysis and surgical therapy of cerebellar arteriovenous malformation (AVM) were studied in 13 patients, consisting of 6 males and 7 females. Ages of the patients ranged from 6 to 64 years. The AVM manifested itself as subarachnoid hemorrhage in all cases, and consciousness was impaired in 9 patients. Feeding arteries were the superior cerebellar artery (SCA) in 3, the posterior inferior cerebellar artery (PICA) in 6, the anterior inferior cerebellar artery (AICA) and PICA in 2, the SCA and AICA in one the SCA and PICA in one patient. The nidus was located at the cerebellar hemisphere in 5, at the vermis in 4, at the cerebellopontine (CP) angle in 2 and over the cerebellar hemisphere extending to the vermis in 2 patients.
The different surgical approaches were applied depending on the location of the AVM. When in was located over the superior surface of the cerebellar hemisphere or at the CP angle, operations were carried out through the subtemporal transtentorial approach with the patient in the lateral position. The tentorium was incised posterior to the vein of Labbe in cases of AVM at the superior surface of cerebellar hemisphere, and anterior to the vein of Labbe in cases of AVM at the CP angle. To avoid ischemic changes of the pons a temporary clip was applied on the cortical branch of the SCA and the main trunk of the SCA around the pons was spared. Difficulty in surgical management depended upon the relative location of the feeding artery to the trochlear nerve. In the cases of AVM at the CP angle, management of the trigeminal and facial nerves became a problem, but, since the feeding artery was easily managed, hardly any detectable cranial nerve palsy was present postoperatively.
For the AVM located at the inferior surface of the cerebellar hemisphere and vermis the AVM was approached through the suboccipital route with the patient in sitting position.
Postoperatively mild trochlear nerve palsy was observed in 2 patients, facial nerve palsy in one and cerebellar signs in 5 patients. In a few months. however, these symptoms became almost undetectable. As the result of operation 10 out of 13 patients recovered to the state before the onset of the disease. One patient remained unable to work, but was able to care for herself. Of the total, 12 patients underwent total removal of the AVM. The last patient in whom a small portion of the AVM was left undisected died of postoperative bleeding.

View full abstract

The estimation of flow rate in cerebrospinal fluid shunt was evaluated in vitro using radionuclide and a gamma camera.
Approximately 50-100 μCi of ^99mTc-pertechnetate in a volume of 0.05 ml was injected into a Pudenz 12 mm or a standard Rickham reservoir through a 26-gauge needle, and a time-activity curve at the region of the reservoir was recorded for a duration of 10 minutes with a gamma camera and data processor (Picker Dyna Camera 2C).
The flow rate of the perfusion fluid in the shunt system was controlled by a hydrostatic pressure with a siphon or an infusion pump. The flow rates were changed from 0.01 ml/min to 1.0 ml/min. This range covered the rate of cerebrospinal fluid production. The perfusion fluid was distilled water or diluted alubumin solution.
The time-activity curves represented the radioactivity clearance from the reservoir, and they showed two exponentials (the first and the second compartments) on the experiments using an infusion pump. The flow rates (F ml/min) determined from the radioactivity clearance half-time (T_1/2 min) of the first compartment were identical with those calculated from the classical clearance formula, F = 0.693 V/T_1/2 where V is the physical volume (V ml) of the reservoir. On the other hand, the relationships between F and T_1/2 of the second compartment were -1.3 log₁₀(T_1/2) - 0.016 ≤ log₁₀(F) ≤ -1.3 log₁₀(T_1/2) + 0.16 in the Pudenz reservoir and -1.9 log₁₀(T_1/2) 0.12 ≤ log₁₀(F) %le; -1.8 log₁₀(T_1/2) + 0.076 in the Rickham reservoir, which coincided with the results of the experiments using a siphon in each reservoir. The regression equation was log₁₀(F) = -1.3 log₁₀(T_1/2) + 0.087 or log₁₀(F) = -1.8 log₁₀(T_1/2) + 0.019 in the Pudenz reservoir or the Rickham one, respectively.
The concentration of the albumin in the perfusion fluid, ranging between 25 mg/dl and 5, 000 mg/dl, showed no significant effect on the T_1/2 of the second compartment. In the Rickham reservoir, however, it was difficult to record the T_1/2 of the second compartment when the albumin was more than 1, 000 mg/dl, because the radioactivity in the reservoir was quickly cleared during the first compartment in these conditions.
The duration of the first compartment observed on the experiment using an infusion pump was within only a few minutes at each flow rate, which is too short to record in clinical practice. Thus, the cerebrospinal fluid flow rate in the shunt system can be determined by using the above equations derived from the second compartment rather than the classical clearance formula.

View full abstract