Objective To investigate the best surgical approach to remove of trigeminal neurinomas (TNs).　Methods 75 patients with TNs in Huashan Hospital were reviewed retrospectively. Results In early group (1978～1984) conventional intradural approaches were used, and in late group (1985～1995) an epidural approach via the skull-base craniotomy was utilized. Total removal of was achieved in 58%(20/35) in the early group and 80% (32/40)(P<0.05) in the late group. emporary or permanent cranial nerves morbidities were 62.7%, 37% in the early group and 28.1%, 10% (P<0.01) in the late group, respectively. Conclusions 　The best surgical approach with microsurgical technique for removal of TNs is epidural approach or epiduro-trans-duro-transtentorial approach via a skull-base craniotomy except the tumor only confined to the posterior fossa

目的 探讨三叉神经鞘瘤最佳外科手术入路。方法 回顾性分析75例三叉神经鞘瘤的临床资料和随访结果。按时间先后分为早期组和后期组进行对比。结果 早期组（1978～1984年）应用常规硬脑膜内入路；后期组（1985～1995年）除颅后窝型应用枕下硬膜内入路外，均用颅底开颅硬脑膜外入路。早期和后期组肿瘤全切率分别为58（20/35）和80%（32/40）（P＜0.05），暂时和永久颅神经障碍分别为63%，37%和28%，10%（P＜0.01）。结论 除非肿瘤位于颅后窝，经颅底开颅硬膜外入路显微外科手术是切除三叉神经鞘瘤最好的方法。

diagnosis rate was 13/22. Total tumor removal was achievedin 20 patients. Two of the 22 patients died after surgery, one of brain-stem injury and the other from an intracranial hemorrhage. Twenty patients were followed up for 2 months to 8 yeas (average, 2 years). Karnofsky scale months to 8 yeas (average, 2 years). Karnofsky scale was ≥80 in 15 patients, 60-70 in 1 patient and not measured in 4 patients who died during follow-up. The causes of death were pneumonia (2 patients) and purulent meningitis (2). Conclusion MRI and DSA (digital subtraction angiography) are major preoperative diagnostic modalities for PFSH, but PFSH remains a challenging benign neoplasm. Special microsurgical techniques and mproved operative manipulation can improve surgical efficacy.

rospectively, and a review of relevant literature was conducted. RESULTS These 20 cases accounted for 16.3% of all intracranial cavernous hemangiomas surgically treated in the same period at Huashan Hospital. The patients were aged from 32～69 years with an average age of 47 years. There were 17 females and 3 males. The common clinical manifestations were visual loss, diplopia, headache, facial numbness and extraocular muscle palsy. Computed tomography (CT) and magnetic resonance imaging (MRI) were major preoperative diagnostic modalities, but demonstrated nonspecific features. The preoperative misdiagnostic rate was 38.9%. Of the 20 patients undergoing craniotomy via modified pterional approach with or without orbitozygomatic osteotomy, tumors were removed through epidural approach (EDA) in 13 cases, intradural approach (IDA) in 7 cases. Hypotension was induced during the operation in 2 cases. Total tumor removal was achieved in 12 cases (92.3%) in the EDA group and nil in the IDA group. Incomplete tumor removal was 1 case (7.7%) in the EDA group and 7 cases (100%) in the IDA group. One patient in the IDA group died of intracranial hemorrhage postoperatively. Compared with preoperative manifestations, cranial neuropathies at discharge were worsened in 76.9% of patients in the EDA group and 83.3% of patients in the IDA group, improved in 15.4% of patients in the EDA group and 16.7% of patients in the IDA group, unchanged in 7.7% of patients in the EDA group. Follow-up study (range, 16 years; mean, 3 years) was available in 17 patients (89%). All patients in the EDA group improved without tumor recurrence. Nonimprovement (2 cases) and continued worsening (3 cases) occurred in the IDA group. In patients with tumor incompletely removed, the tumor enlarged in 2 cases, and unchanged or decreased in size in 3 cases, in which 2 cases underwent postoperative radiosurgery or radiotherapy. According to pathologic and MRI characteristics, the CSHs can be divided into two types, sponge-like type and mulberry-like type. CONCLUSION Cavernous hemangioma should be included in differential diagnosis for middle aged females with cavernous sinus tumors. Two types of the CSHs, sponge-like type and mulberry-like type, can be identified. The best microsurgical approach for the removal of the CSHs is epidural approach via the skull-base craniotomy. Radiosurgery should be considered for patients with incomplete tumor removal.

From 1978 to 1988, 14 giant intracranial aneurysms (more than 2.4 cm in diameter) and one large aneurysm (1.5 cm in diameter) were treated by extracranial/intracranial (EC/IC) bypass or cerebral artery reconstruction. Of the aneurysms, 10 were located at the intracavernous carotid artery (CCA). One of the 10 anourysms was posttraumatic and located at both the carotid-ophthalmic artery segment and the bifurcation of the internal carotid artery (1CA). Three were seen at the middle cerebral artery (MCA) trunk. The aneurysms were demonstrated by angiography and CT scanning. They were treated with trapping of the aneurysm and superficial temporal artery (STA)/middle cerebral artery (STA-MCA) bypass with/without a graft (6 cases), cervical ICA !igation and STA-MCA bypass with/without a graft (6) aneurysm excision with an end-to-end anastomosis of the MCA and a STA-MCA bypass with a graft (1), proximal MCA occlusion and STA MCA bypass with a graft (1), and aneurysm neck clipping following a STA-MCA bypass with a graft (1). The patients showed marked improvement after operation except one whose neurological deficit was aggravated temporarily. Postoperative angiography revealed that the anastomoses were patent in all cases. No surgical mortality or any delayed ischemic complications were noted after follow-up for 5.6 years. We believe that cerebral artery reconstruction or EC/IC bypass is still effective in the treatment of large and giant intracranial aneurysms.