In Akita Prefecture, there are nine hospitals established by the Akita Prefectural Federation of Agricultural Cooperatives for Health and Welfare (Koseiren). Half of the stroke patients in the prefecture were treated in these Koseiren hospitals, and half of the mass screening projects for the prevention of cardio vascular diseases were undertaken by these hospitals. A retrospective cohort study was done using mass-screening data (age, sex, past history of diabetes mellitus, blood pressure, body mass index, smoking and drinking habits) of 175,033 cases stored at these hospitals from 1988 to 1999, and the prefecture-wide stroke data of 2,520 initial stroke events registered from 1988 to 2003. The number of stroke cases was broken down into 1,428 cases of cerebral infarction (57%, CI), 693 cases of cerebral hemorrhage (27%, CH) and 399 cases of subarachnoid hemorrhage (16%, SAH). The subjects were also divided into five age groups:30-49, 50-59, 60-69, 70-79 and 80-89. Blood pressure (BP) was classified into six categories according to the JNC 6 criteria. Risk factors were determined using the Cox analysis. The hazard ratio for CI and CH was increasing with advancing age. CI showed a higher hazard ratio in men than women (hazard ratio for men was 1.8). The hazard ratio was increasing as BP became higher in any of three stroke subtypes, and especially CH showed the strongest correlation with BP. Uncontrollable risk factors were very closely associated with the attack of CI. On the other hand, BP (controllable risk) was closely linked with the attack of CH. Our results showed the prevention of CI was not easy. Controlling BP may be the most effective strategy for preventing hemorrhagic stroke (CH and SAH).
Cerebrovascular accident ; Cephalic index ; Blood pressure determination ; hazard ; Risk Factors

We describe a rare case of sacral epidural arteriovenous fistulas (edAVFs) with atypical clinical course of treatment. A 78-year-old man with a history of spinal surgery presented progressive gait disturbance and urinary incontinence. Spinal angiography demonstrated a sacral spinal AVF fed by bilateral lateral sacral arteries, draining to the venous pouch with subdural drainage. The first treatment by direct interruption of a subdural drainer was incompletely finished. Postoperative reassessment by 3D imaging analysis led to the diagnosis of sacral edAVF and 3D understanding of its angioarchitecture. The second treatment by transarterial embolization (TAE) resulted in complete occlusion of a sacral edAVF. However, spinal venous congestion didn’t improve, because the recruitment of occult edAVFs at the multiple lumbar levels and complex-shaped sacral ventral epidural venous plexus (VEP) were involved in the remnant of prior subdural drainage. The third treatment was performed by TAE for three occult edAVFs and the VEP compartment connecting between a patent edAVF and subdural drainage, which resulted in complete disappearance of spinal cord edema. Endovascular embolization of VEP compartment connecting to subdural drainage in addition to fistulous occlusion may be one of the treatment options for several edAVFs at the multiple spinal levels.