No abstract

Background: Cerebral vasospasm (VS) is the most common cause of morbidity and mortality after aneurysmal subarachnoid hemorrhage (SAH). Reversal of VS by intra-arterial infusion of cyclic adenosine monophosphate (cAMP)-elevating agents has been reported; however, the preventive role in the development of VS is not fully understood. This study is designed to evaluate the possible efficacy of using cilostazol, a selective inhibitor of phosphodiesterase type 3 and a cAMP-elevating agent, in patients with SAH. Methods: In this prospective randomized study, we enrolled 100 SAH patients who met the following criteria: neck clipping within 72 h after onset, Hunt and Hess (HH) score ≤4, modified Rankin scale (mRS) score ≤2 prior to ictus, and no serious cardiovascular complications. Patients were divided into control and cilostazol groups; we focused on the effects of cilostazol on the decrease in the incidence of symptomatic VS, cerebral infarction, and the mRS score at discharge. Result: Patients’ age, male/female ratio, mRS score prior to ictus, HH grade, Fisher group, site of the aneurysm, drugs prescribed during the observation period, and length of hospital stay were not different between the groups. Cilostazol did not significantly decrease the incidence of symptomatic VS (37.3% in the control vs. 22.4% in the cilostazol group, p = 0.183) and cerebral infarction (27.5% in control vs. 10.2% in the cilostazol, p = 0.091). However, mRS score was significantly improved at discharge (2.6 in controls vs. 1.5 in the cilostazol group, p = 0.041). Patients’ age being ≤65 years (OR = 8.47, 95% CI = 2.45–29.32, p = 0.0007), Fisher group ≤3 (OR = 4.64, 95% CI = 1.00–21.45, p = 0.049), HH grade ≤2 (OR = 4.31, 95% CI = 1.27–14.59, p = 0.019), no hydrocephalus (OR = 8.55, 95% CI = 1.72–19.23, p = 0.0046), and cilostazol use (OR = 5.52, 95% CI = 1.61–18.90, p = 0.0065) were independent predictors of good outcomes (mRS score ≤2). Conclusion: Cilostazol may improve outcomes after SAH, but further double-blind, placebo-controlled studies are required for a definitive conclusion.

Twenty-six ependymal and 15 choroid plexus tumors were examined with monoclonal antibody against cytokeratin using the avidin-biotin-peroxidase complex (ABC) technique. Serial sections were examined with antisera to glial fibrillary acidic protein (GFAP). In five ependymal tumors (one ependymoma, two papillary ependymomas, and two primitive neuroectodermal tumors [PNEV with ependymal cells), a variable number of cytokeratin-positive cells were present. Most tumor cells (except two PNET) were positive with GFAP antisera. Many cytokeratin-positive cells were present in all choroid plexus tumors. GFAP-positive cells were present focally in six of 11 papillomas and in one of four carcinomas. Although their staining patterns and distribution were clearly different, focal coexistence of cytokeratin and GFAP was observed in six papillomas and two ependymal tumors. Thus, some ependymal tumors (especially papillary ependymomas and occasional PNET) and many choroid plexus tumors have de-monstrable positivity with antibody to cytokeratin, suggesting a transitional cell type with features of both ependyma and choroid plexus.

Twenty-four cases of classical medulloblastoma and one case of desmoplastic medulloblastoma were examined for glial fibrillary acidic protein (GFAP) using the immunoperoxidase method to assess astrocytic differentiation. In 16 cases of classical medulloblastoma GFAP-positive cells were present in variable numbers. These cells were classified as three different types according to size and shape. The type 1 cell was morphologically identical to the ordinary tumor cell, with a hyperchromatic nucleus and a scanty cytoplasm. The type 2 cell had a fairly rich cytoplasm with short cytoplasmic processes. The type 3 cell was characterized by a relatively large nucleus with sparse chromatin and well-developed cytoplasmic processes, and was considered a reactive astrocyte. The type 1 and some of the type 2 cells seemed to be neoplastic, displaying astrocytic differentiation. The remaining type 2 cells may have been reactive astrocytes. In one case of desmoplastic medulloblastoma, the majority of GFAP-positive cells were arranged in "islands", and had delicate fibrillated processes. GFAP-positive cells were also observed outside these "islands", though they were less numerous. Most of them were regarded as type 3 cells, but some were type 2. This may be interpreted as meaning that the glial character of the tumor was expressed more within than outside these "islands".

Microglia were demonstrated in paraffin-embedded human nervous tissues with an avidin-biotin peroxidase method and Ricinus communis agglutinin-1 (RCA-1). Specific staining was observed in cell bodies and processes of microglia. Although endothelial cells and blood cells reacted with RCA-1, they were easily distinguished morphologically from microglia. Astrocytes, oligodendrocytes, and neurons did not react with RCA-1. These results suggest that RCA-1 can be used as a new histochemical marker for microglia in normal human brain.