Accumulation of allelic losses on chromosome 10 in human gliomas at recurrence

Aims-To elucidate the implications of allelic loss on chromosome 10 in the malignant progression of human gliomas.Methods-Eight microsatellite loci (D10S249, D10S191, D10S210, D10S219, D10S246, D10S222, D10S221, and D10S212) were analysed for chromosomal deletions in histologically benign and malignant, including recurrent, gliomas. Of the 16 original tumours studied (two astrocytomas, nine anaplastic astrocytomas and five glioblastomas), the histological diagnosis at recurrence was anaplastic astrocytoma in six cases and glioblastoma in 10. Genomic DNA was extracted from formalin fixed, paraffin wax embedded sections. Samples of original and recurrent tumours were paired and amplified using PCR. Samples of histologically normal brain served as controls.Results-Of the original tumours, all five glioblastomas, five (56%) of nine anaplastic astrocytomas and none of the astrocytomas demonstrated loss of heterozygosity (LOH) on chromosome 10. Additional LOH was detected in the five cases of anaplastic astrocytoma that progressed to glioblastoma at recurrence. Additional LOH was not detected in the two cases of astrocytoma that progressed to anaplastic astrocytoma at recurrence. With the exception of one case, additional LOH was observed in the recurrent glioblastomas.Conclusion-LOH was observed at the loci of two adjacent microsatellite markers, D10S222 and D10S221 (10q23-q25), suggesting that this region on chromosome 10 is closely related to progression from anaplastic astrocytoma to glioblastoma.

Chromosome 22q allelic losses at microsatellite loci in human astrocytic tumors

Common regions of deletion(s) on chromosome 22q and the correlations between loss of heterozygosity and patient survival were analysed in 18 deoxyribonucleic acid samples from astrocytic tumors (3 astrocytomas, 5 anaplastic astrocytomas, and 11 glioblastomas) and matched normal brain tissues. The polymerase chain reaction products using five microsatellite markers were electrophoresed on polyacrylamide gels and the ethidium bromide stained bands were photographed. Loss of heterozygosity was observed in 14 (74%) of 19 samples, with similar incidences in astrocytomas, anaplastic astrocytomas, and glioblastomas (67%, 80%, and 82%, respectively). The locus D22S300 (q12.1-q13.1) was most frequently involved, with loss of heterozygosity in eight (80%) of 10 informative glioblastomas at this locus. Increased loss of heterozygosity during tumor progression or recurrence was seen in two patients at the D22S300 (q12.1-q13.1) and TOP1P2 (q11.2-q13.1) loci. No correlation between loss of heterozygosity on chromosome 22 and the postoperative survival was found. These findings suggest that loss of heterozygosity on chromosome 22q probably occurs quite frequently in astrocytic tumors. The chromosome segment 22q12.1-q13.1, around the D22S300 locus, may be the common region of deletion in glioblastomas.

Chromosome 17 allelic loss in astrocytic tumors and its clinico-pathologic implications

To prognosticate the implications of various allelic losses on chromosome 17 in the morphology and biology of astrocytic tumors, we have examined loss of heterozygosity (LOH) at 14 microsatellite loci on chromosome 17 in a series of 19 astrocytic tumors (3 astrocytomas, 5 anaplastic astrocytomas, and 11 glioblastomas). The DNA samples extracted from tumor and matched normal brain tissue were amplified by polymerase chain reaction (PCR) followed by polyacrylamide gel electrophoresis and photography under UV transillumination. The molecular genetic data were compared with immunohistochemistry performed with antibodies to glial fibrillary acidic protein (GFAP), MIB-1 and p53 protein. LOH was observed in 11/19 (58%) instances with frequent involvement of TP53, NF1, and D17S795 loci, LOH at D17S578 and D17S520 occurred in recurrent tumors exclusively. Allelic status of D17S795 in all 12 informative instances were concordant with GFAP immunoreactivity (p < 0.01, Fisher's test). p53 immunopositivity (> 25% of tumor cell nuclei) was seen in 11 (58%) tumors, of which 6 were informative of TP53 locus with 2 (33%) demonstrating LOH. The MIB-1 staining indexes in astrocytomas, anaplastic astrocytomas, and glioblastomas were 1.9 +/- 0.9, 8.4 +/- 4.0, and 17.1 +/- 7.1% (mean +/- SD), respectively, and their differences were statistically significant (p < 0.05, Student's t test). A trend of inverse relationship between patient survival and the number of tumor cell nuclei with immunohistochemically detectable p53 protein was seen in glioblastoma cases: 20.5 +/- 12.7 versus 13.7 +/- 6.3 months (mean +/- SD) in instances with > or > or = 25% positive cells, respectively. We conclude, the intriguing correlation between allelic status of D17S795 microsatellite locus and GFAP immunoreactivity suggests the possible involvement of q21.2 segment of chromosome 17 in the morphology and biology of astrocytic tumors.

Loss of heterozygosity of microsatellite loci on chromosome 9p in astrocytic tumors and its prognostic implications

We analyzed 19 samples of various astrocytic tumors (3 astrocytomas, 5 anaplastic astrocytomas, and 11 glioblastomas) for loss of heterozygosity (LOH) on chromosome 9p at 6 microsatellite loci (D9S54, IFNA, D9S171, D9S104, D9S165, and D9S166). Polymerase chain reaction was performed and the products were electrophoresed on polyacrylamide gel. As many as 16 of the 19 samples (84%) exhibited LOH. Three of the 7 informative loci (43%) showed LOH at D9S54, 7 of 17 (41%) at IFNA, 8 of 14 (57%) at D9S171, 7 of 14 (50%) at D9S104, 4 of 8 (50%) at D9S165, and 2 of 7 (29%) at D9S166. LOH was recognized in 57% of the informative loci in anaplastic astrocytomas and 54% in glioblastomas, while it was seen in only 8% of the astrocytomas. Accumulation of LOH with progression or recurrence of tumor was seen in 2 patients. Although, the survival period of the patients correlated well to the histological types of astrocytic tumors, we could not find any obvious correlations between the presence/absence of LOH and the survival period in these patients. In conclusion, we speculate that LOH on chromosome 9p is involved in malignant progression of astrocytomas, but has no significance in predicting survival period in these patients.

Ventriculolumbar perfusion chemotherapy with methotrexate and cytosine arabinoside for meningeal carcinomatosis: a pilot study in 13 patients

Thirteen patients with meningeal carcinomatosis were treated by ventriculolumbar perfusion using methotrexate (MTX) and cytosine arabinoside (Ara-C). MTX (10-30 mg) and Ara-C (40 mg) were infused at 8- to 12-hour intervals on six or nine occasions via an Ommaya reservoir placed in the lateral ventricle. Nine of thirteen patients had evaluable response (69% response rate with a mean survival of 8.8 months among responders) and ventriculolumbar perfusion therapy was effective in improving cerebral, cranial nerve, and spinal root signs and symptoms, especially sensorimotor disturbance in the lower limbs. Three of the six bedridden patients became ambulatory without assistance and two of the four patients who were walking with assistance became ambulatory without assistance. Urinary incontinence also markedly improved, except in one nonresponder. Lumbar cerebrospinal fluid parameters (cytological findings and tumor markers) also improved in association with the clinical improvement. Our pilot results were encouraging, especially the improvement of sensorimotor function in the lower limbs. However, the toxicity was unacceptable when compared with that of standard intrathecal chemotherapy. Thus, this therapy needs to be investigated further to establish the most appropriate drug doses and perfusate volume to reduce toxicity as well as determine its true efficacy in the treatment of meningeal carcinomatosis.

Difference in CDDP penetration into CSF between selective intraarterial chemotherapy in patients with malignant glioma and intravenous or intracarotid administration in patients with metastatic brain tumor

Platinum (Pt) levels in plasma and cerebrospinal fluid (CSF) in patients with malignant glioma were determined after initiation of selective intraarterial chemotherapy with a combination of VP-16 (etoposide) and CDDP (cisplatin), and were compared with the CSF Pt levels in patients with metastatic brain tumors after intravenous or intracarotid administration of VP-16 and CDDP. CSF Pt levels were also compared for various administration routes, doses, CSF sampling routes and blood-CSF barriers in metastatic brain tumor. Changes in the blood-CSF barrier to CDDP during treatment in a patient with meningeal lymphoma and in a patient recovering from surgical removal of a metastatic brain tumor were also examined by periodic administration of CDDP. All CSF samples were taken through Ommaya reserviors placed in the anterior horn of the lateral ventricle or the postoperative cavity. The mean peak CSF/plasma total Pt ratio (T/T ratio) and the mean CSF total Pt/plasma ultrafiltrable Pt ratio (T/U ratio) were highest (15.0% and 24.4%, respectively) following selective intraarterial infusion of CDDP in patients with malignant glioma, followed by intravenous infusion in meningeal carcinomatosis (11.5% and 18.9%), intracarotid administration (5.4% and 8.7%) and intravenous infusion (60 mg/m2 2.5% and 100 mg/m2 2.9%; and 60 mg/m2 3.5% and 100 mg/m2 7.7%) in patients with the solid type of metastatic brain tumor. In CSF obtained from the postoperative cavity in cases of metastatic brain tumor, T/T and T/U ratios were extremely high (40.9% and 62.4%). However, the CSF Pt level even after selective intraarterial administration of CDDP in malignant glioma was 0.51-1.64 micrograms/ml total Pt and 0.43-1.08 micrograms/ml ultrafiltrable Pt. Even the CSF level obtained from the postoperative cavity was 1.0-4.7 micrograms/ml total Pt. These low levels of total and ultrafiltrable Pt are considered not to be cytotoxic to disseminated cells in the CSF space and to normal brain cells. As for changes in the blood-CSF barrier, repeated administration of CDDP showed that the rate of entry of Pt into the CSF decreased in parallel with improvements apparent on CT scans in the patient with meningeal lymphoma, and also showed that the blood-CSF barrier to Pt was gradually repaired after the metastatic brain tumor had been removed.

Surgical treatment of brain metastases of lung cancer: retrospective analysis of 89 cases

The records of 89 patients who underwent surgery for solitary or multiple parenchymal brain metastases of lung cancer at the Osaka Center for Adult Diseases between 1978 and 1990 were reviewed with follow up until March 1992. The aim of this retrospective analysis was to identify prognostic features that were associated with a favourable outcome. The benefits of brain tumour surgery were evaluated in terms of the cause of death (brain metastasis, tumour in another organ, or treatment related) as well as the postoperative changes in functional state indicated by the Karnofsky scale. The overall mean survival time was 11.6 months, and the one and two year survival rates were 24% and 8%. The brain lesion itself was the cause of death in only 19% of the patients; the other 81% died of systemic disease. Functional state improved after surgical excision of the brain tumour in 36%, remained unchanged in 53%, and worsened in 11%. These data suggest that surgical intervention is beneficial for patients with parenchymal brain metastases. Variables significantly associated with a favourable prognosis included surgical excision of the primary lesion, adenocarcinoma as the histological diagnosis, the use of adjuvant treatment, especially chemotherapy, a preoperative score of over 80% on the Karnofsky scale, and metastasis confined to the brain with no extracranial metastatic foci or residual primary tumour. Additional but non-significant contributors to a good prognosis included age under 65 or 70 years, early tumour stage (stage 1), curative lung cancer surgery, a single metastatic brain tumour (v multiple lesions), a solid tumour (v cystic), and a supratentorial location of the brain metastasis. The disease free interval and the cerebrospinal fluid cytology were not significant prognostic factors. On the basis of these findings, it is concluded that the surgical removal of brain metastases of lung cancer should be undertaken if the primary tumour has already been removed whether or not there are extracranial metastases, and that postoperative chemotherapy should generally be given.

[Chemotherapy for metastatic brain tumors with CDDP and other agents: correlation between chemotherapeutic effects and the results of in vitro chemosensitivity tests using collagen gel-embedded culture combined with computerized image analysis in metastatic brain tumors]

Chemotherapy with CDDP and/or other agents was performed in 15 patients after removal of metastatic brain tumors. A chemosensitivity test using a system of collagen gel-embedded culture and computerized image analysis was performed on the tumors from these patients. The clinical usefulness of the chemosensitivity test was evaluated by comparing chemotherapeutic effects with the results of the test. The rates of correlation of the chemosensitivity test with clinical response on brain MRI was 80%, and that of the chemosensitivity test with clinical response in tumor markers or on primary tumors was 75%. This observation suggests that the chemosensitivity test using collagen gel-embedded culture and computerized image analysis is useful in determining optimal chemotherapy for metastatic brain tumors. In ten multiple metastatic brain tumors, three complete responses, two partial responses, one minor response and four non-responses were observed on MRI. Only one case showed a false negative result on the chemosensitivity test and showed partial response. This result also indicates the effectiveness of chemotherapy based on chemosensitivity testing.

[Successful reconstruction of a skull base fracture with frontal lobe contusion by omental transplantation for recurrent posttraumatic tension pneumocephalus: a case report]

In July 1983, 26 year-old male was admitted to our neurosurgical clinic after severe head injury caused by a car accident. Recovery of consciousness was delayed for months due to bilateral frontal lobe contusions with anterior skull base fractures on both sides. He was first discharged 6 months after surgery for ventriculo-peritoneal shunting. He was readmitted to our department due to an episode of urinary incontinence with gait disturbance 11 months after the accident. CT film of the head revealed the presence of an air shadow at the left frontal base. Utilizing lyophyllized dura mater, the first cranial surgery for closure of cerebrospinal fluid leakage was carried out in July, 1984. Pneumocephalus with meningitis recurred again five years later. Repeated conventional surgery failed including a transsphenoidal and transfrontal sinus approach to treat the recurrent pneumocephalus. In order to close the defect in the skull base, and to obliterate the dead space in the left frontal lobe, vital tissue transplantation was planned. On May 23, 1990, seven years after the accident, the skull base defect was repaired by suturing fascia taken from the temporal muscle. Then the patient's vascularized omentum was utilized as an autograft by micro-surgical technique. A superficial temporal artery and vein, and superficial sylvian vein were used to vascularize the omentum. The patient has been totally free of pneumocephalus for more than three and half years following the radical surgery. Clinical omental transplantation using microsurgical technique for vascularization of cerebral circulation as well as plastic surgery has been reported.(ABSTRACT TRUNCATED AT 250 WORDS)

Myelin basic protein in the cerebrospinal fluid of patients with brain tumors

We measured the level of myelin basic protein (MBP) in the cerebrospinal fluid (CSF) of patients with various kinds of tumors, including malignant tumors, using radioimmunoassay. The CSF had been obtained by lumbar puncture through an Ommaya reservoir or a shunt device placed in the lateral ventricle. The level of MBP was high (> 4 ng/ml) in the patients with meningeal dissemination of malignant tumors, but in those who showed a good response to chemotherapy and/or radiation, it decreased or returned to the normal level, with improvement on the computed tomography and magnetic resonance imaging, cytological, general CSF, and neurological findings. Of seven malignant gliomas without CSF dissemination, six showed an elevated level of MBP before selective intra-arterial chemotherapy with a combination of etoposide and cisplatin administered via a microcatheter placed at A1, M1, P1-P2, and the basilar top. All CSF specimens obtained during the period of the intra-arterial chemotherapy showed an abnormally high (> 4 ng/ml) level of MBP that exceeded the prechemotherapy level. The MBP level decreased or returned to normal in the patients with a good response to chemotherapy after intra-arterial chemotherapy. In some patients with multiple metastatic brain tumors, the MBP level was elevated before treatment and returned to normal after treatment (surgical removal, chemotherapy, and/or irradiation) in all except one. Thus, there was a clear correlation between the timing of treatment and changes in imaging studies and MBP levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Spinal canal stenosis at the level of the atlas: case report

We describe a rare case of cervical canal stenosis at the level of the atlas, presenting with cervical myelopathy as an initial symptom. A 55-year-old man was admitted to our hospital with a 4-month history of gait disturbance and clumsiness in both hands. He had no history of trauma, and showed left-sided hemihyperesthesia and mild quadriplegia with exaggerated deep tendon reflexes. Plain cervical tomography revealed marked spinal canal stenosis, with the diameter of 8.0 mm at the level of the atlas. Magnetic resonance imaging demonstrated severe compression of the spinal cord at the atlas, with an extensive high intensity area indicating edema on T2-weighted images. The patient showed an excellent recovery after laminectomy of the atlas combined with decompression of the lower posterior fossa of the foramen magnum and dural plasty.

[Myelin basic protein in the cerebrospinal fluid of patients with neurological disease: especially with malignant brain tumors]

Myelin basic protein (MBP) in the cerebrospinal fluid (CSF) of patients with brain tumors and other neurological diseases was measured before, during and after various treatments such as surgery, chemotherapy and irradiation. We assessed the significance of changes in the MBP levels during the course of treatment, and speculate on what the elevated level of MBP in brain tumor patients indicates. In meningeal dissemination of malignant tumors, meningeal carcinomatosis from cancer of the systemic organ showed the highest level of MBP followed by meningeal gliomatosis and meningeal lymphoma. Meningeal carcinomatosis and meningeal lymphoma, which have responded to chemotherapy, showed normal levels of MBP after chemotherapy. Six of eight patients with newly diagnosed malignant glioma showed moderate to high levels of MBP (range 4.6-35.5ng/ml) just before intraarterial chemotherapy with VP-16 and CDDP. The level increased in five patients during the course of chemotherapy and then decreased in relation to the degree of tumor reduction by chemotherapy. In the solid type of metastatic brain tumor, five of seven patients with multiple tumors showed high levels of MBP and these levels also returned to normal after treatment in four patients. As for the influence of irradiation, levels of MBP did not increase after irradiation except in three patients who developed radiation necrosis, local extensive edema or atrophic change. In other brain tumors, levels of MBP were high in a patient with a large meningioma with very extensive edema and during an unstable postoperative condition after total removal of a large craniopharyngioma.(ABSTRACT TRUNCATED AT 250 WORDS)

Selective intra-arterial chemotherapy with a combination of etoposide and cisplatin for malignant gliomas: preliminary report

We administered selective intra-arterial chemotherapy consisting of a combination of etoposide and cisplatin to 20 patients with malignant glioma (seven with recurrent and six with enlarged tumors after initial treatment, and seven newly diagnosed patients). Evaluation of efficacy was based on computed tomographic and magnetic resonance imaging findings. In the process of establishing a safe technique for superselective intra-arterial chemotherapy, we encountered cerebrovascular accidents in two patients (after etoposide in one and after etoposide plus cisplatin in the other). In these two cases, 100 mg/m2 of etoposide and 100 mg/m2 of cisplatin were delivered via the horizontal segment of the middle cerebral artery (M1) or the tip of the basilar artery, with the infusion time reduced to 20 minutes. Thereafter, the etoposide was diluted, and the doses of both drugs were reduced to 80 or 50 mg/m2, and finally to 60 mg/m2, and both were infused over 60 minutes. In addition, for prevention of local spasm, papaverine hydrochloride and nicardipine were given via the same catheter at 5-minute intervals during administration of etoposide and cisplatin. No complications developed in the later cases. Thereafter, selective intra-arterial infusion of etoposide and cisplatin into the anterior cerebral artery, middle cerebral artery, posterior cerebral artery, or the basilar artery for malignant gliomas in the basal ganglia, internal capsule, and brainstem--a procedure generally considered risky in terms of potential complications--was performed safely, with tolerable side effects. Computed tomography and magnetic resonance imaging indicated improvement in 13 patients, including four whose tumors completely disappeared. This method of intra-arterial chemotherapy may be useful as an adjuvant treatment for malignant glioma.

[Pharmacokinetics of plasma and cerebrospinal fluid cisplatin in patients with malignant glioma and metastatic brain tumor after selective intraarterial or intravenous and intracarotid administration of etoposide and cisplatin]

CSF and plasma platinum levels were examined in patients with malignant glioma after administration of etoposide and cisplatin each at doses of 60 mg/m2 by 60-minute selective intraarterial infusion. These same factors were also examined in patients with metastatic brain tumors after administration of cisplatin at a dose of 60 or 100mg/m2 by 60-minute intracarotid or intravenous infusion. Plasma and CSF samples taken through an Ommaya reservoir placed in the lateral ventricle or postoperative cavity were analyzed for platinum content by atomic absorption spectroscopy. Plasma and CSF platinum levels were dose dependent. The overall plasma platinum curves were biphasic, with mean half-lives of 35 minutes and 56 hrs. The mean peak total CSF concentration was 10.0% of the peak total plasma platinum and 20.2% of the peak free plasma platinum in patients with malignant glioma. In patients with a solid metastatic brain tumor, the mean peak total CSF concentration was 1.9% of the peak total plasma platinum and 4.0% of the peak free plasma platinum after i.v. infusion. After intracarotid infusion, the mean peak total CSF concentration was 3.4% of the peak total plasma platinum and 7.0% for the peak free plasma platinum. In patients with meningeal carcinomatosis, the mean peak CSF concentrations were 7.7% of the peak total plasma platinum and 13.7% of the peak free plasma platinum. The free to total platinum ratio in plasma decreased quickly and that in CSF increased and was maintained at the high levels of 80% for two hours or more.(ABSTRACT TRUNCATED AT 250 WORDS)

[Penetration of potassium clavulanate/ticarcillin sodium into cerebrospinal fluid in neurosurgical patients]

Concentrations of potassium clavulanate (CVA) and ticarcillin sodium (TIPC) in the plasma and cerebrospinal fluid (CSF) of patients after neurosurgical intervention were determined at various times after a 1-hour drip infusion (3.2-g dose). Patients whose blood-brain barriers were supposed to be maintained in almost a normal condition were selected. CSF was obtained through a catheter placed in the anterior horn of the lateral ventricle in all the patients. Maximum plasma levels (micrograms/ml) of 57.6 to 384.0 with an average of 169.7 (TIPC) and 0.41 to 26.2 with an average of 6.1 (CVA) were achieved at the termination of infusion. The maximum CSF levels (micrograms/ml) were 0.61 to 18.8 (TIPC) and 0.1 to 6.81 (CVA) with mean values of 4.5 and 1.2, respectively. Plasma half lives (T1/2) (minute) were 24 to 93 (TIPC) and 32 to 227 with mean values of 58 and 127, respectively. The mean values of the CSF half lives (minute) were 237 (TIPC) and 113 (CVA). The ratios (%) of CSF levels to plasma levels in maximum concentration (Cmax), AUC (area under concentration curve) and half life (T1/2) were calculated. Cmax ratios were 0.2 to 29.2 (TIPC) and 1.4 to 69.8 (CVA) with mean values of 4.4 and 22.8, respectively. AUC ratios were 0.3 to 23.5 (TIPC) and 1.1 to 70.2 (CVA) with mean values of 4.3 and 22.4, respectively. T1/2 ratios were 1.3 to 18 (TIPC) and 1.1 to 4.3 (CVA) with mean values of 5.5 and 2.3, respectively. These values indicate that CVA/TIPC may be classified into a group of antibiotics with good penetration into the CSF.

[The analysis of nuclear organizer regions of astrocytomas with various histologic malignancies]

Nucleolar organizer regions (NORs) correspond to the loops of DNA which encode the ribosomal RNA. Acid proteins related to NORs can be stained by the silver colloidal technique (AgNORs). Since the configurations of AgNORs may be related to the protein metabolism or the proliferative activity of the cell, we tried to evaluate the corelationship between the morphology of AgNOR and the histologic malignancy in astrocytic tumors. For the quantitative evaluation the histographic pattern of AgNORs was analysed. Twenty-seven surgical specimens of astrocytomas (astrocytoma; 7 Cases, anaplastic astrocytoma; 9 cases, glioblastoma; 11 cases) were examined. The average of the means of AgNOR count in astrocytoma, anaplastic astrocytoma, glioblastoma were 1.68, 1.85 and 2.76 respectively. The averages of standard deviations (S. D.) of AgNOR count were 0.87, 1.03 and 1.26, respectively. In those tumors, the AgNOR histograms were flattered and the means and S. D. increased significantly as the malignancy increased. We speculate that the increased number and variations of AgNOR count could be a reflection of phenotypic alterations of astrocytoma cells such as cellular anaplasia and pleomorphism.

Efficacy and toxicity of thromboxane synthetase inhibitor for cerebral vasospasm after subarachnoid hemorrhage

The efficacy and possible side effects of thromboxane A2 (TXA2) synthetase inhibitor in the treatment of cerebral vasospasm after subarachnoid hemorrhage (SAH) were assessed for 24 patients who presented with grades I to IV of the Hunt and Hess classification. All patients underwent aneurysmal clipping within 48 hours after SAH. Postoperatively, TXA2 synthetase inhibitor, Cataclot [sodium (E)-3-[p-(1H-imidazol-1-ylmethyl)phenyl]-2-propenoate] was administered to 13 patients by continuous drip infusion at a dose of 1 microgram/kg/min for 8 to 14 days (group A). The remaining 11 patients did not receive this drug (group B). Of the 13 patients in group A, seven patients (54%) showed no symptomatic vasospasm after SAH. Four patients (31%) developed a transient deterioration of consciousness and/or motor disturbance. Three of these patients fully recovered, while one of them showed a mild neurological deficit on discharge. One patient (8%) developed permanent dysphasia and hemiparesis as a result of ischemic brain damage due to vasospasm. One patient (8%) died of the side effect. On the other hand, of the 11 patients in group B, only three (27%) showed no symptomatic vasospasm. One (9%) patient presented a transient neurological deficit but fully recovered upon discharge. Four patients (36%) showed permanent neurological deficits, although they all could lead an independent life after discharge. The three remaining patients developed a severe disturbance of consciousness caused by ischemia due to vasospasm, and two of them died within 1 month after the onset of SAH. In the group treated with Cataclot, two patients developed an epidural hematoma late during the administration of the drug. Of these two, one patient died of increased intracranial pressure that was accelerated by the complication. These results indicate that TXA2 synthetase inhibitor is effective in not only decreasing the occurrence of symptomatic vasospasm but also reducing the neurological deterioration due to vasospasm after SAH. However, this drug has a hazardous side effect in that it may promote a tendency to bleed, which caused death in one of our patients.