藤本 茂

Objective Prehospital delay is the major cause of treatment delay in stroke. This study was conducted to clarify the contribution of specific stroke-related symptoms to prehospital delay. Methods A consecutive series of 469 patients hospitalized within 2 weeks of stroke onset was collected. In this study, prehospital delay was defined as the time interval from recognition of stroke-related symptoms to hospital arrival. The prevalence of each symptom or sign and prehospital delay were analyzed. Results Weakness of the lower limb (43.5%) was the most common symptom followed by weakness of the upper limb (37.1%) and dysarthria (31.6%). Conversely, the most common sign was weakness of the upper limb (54.1%) followed by dysarthria (53.3%), weakness of the lower limb (53.1%), and sensory disturbance (39.0%). The presence of confusion/decreased level of consciousness (p<0.001), aphasia (p<0.001), headache (p=0.017), and nausea/vomiting (p=0.035) were associated with earlier hospital visitation compared with the absence of these symptoms in univariate analyses. Conversely, the presence of sensory disturbance (p=0.0017) and vertigo/dizziness (p=0.044) were associated with a significant delay in hospital visitation compared with the absence of these symptoms. There was a discrepancy in the prevalence between symptoms recognized by the patients or bystanders and signs diagnosed by the physicians. Conclusion There was a significant overall correlation between prehospital delay and the National Institute of Health Stroke Scale scores. Public education is therefore necessary to encourage early hospital visitation even with the appearance of mild symptoms.

Case 1 involved a 77-year-old man admitted to our hospital after he lost almost all memory of work on the day before. Diffusion-weighted imaging (DWI) revealed spotty hyperintensities in the bilateral hippocampi, which were considered responsible for the amnesia. Transesophageal echocardiography (TEE) showed severe aortic arch atheroma, 6.8 mm in diameter with extension to the branch. Aortogenic embolism to the bilateral hippocampi was diagnosed. Case 2 involved a 66-year-old woman admitted to our hospital because she had lost all memory of the 5 hours she was out. She had transient anterograde amnesia. DWI revealed no ischemic lesions, but magnetic resonance angiography (MRA) revealed branch occlusion of the right posterior cerebral artery. MRA on hospital day 7 revealed partial recanalization. TEE showed aortic arch atheroma of 3.9 mm in diameter with extension to the branch. In both cases, aortogenic embolism to the hippocampus might have been causally related to transient memory disturbance. In patients with acute memory disturbance, the possibility of ischemic stroke should be considered.

Background: The incidence of intracranial bleeding during dabigatran treatment is lower than that during warfarin treatment. The characteristics of intracranial hemorrhage during dabigatran therapy, however, remain unclear. Methods and Results: The clinical data and treatment summaries of 9 intracranial bleeds that developed during dabigatran treatment in 8 patients with non-valvular atrial fibrillation were retrospectively reviewed. Five patients had small-moderate subdural hematomas, 2 had intracerebral hemorrhage and 1 had traumatic subarachnoid and parenchymal hemorrhage associated with cerebral contusion. Activated partial thromboplastin time upon admission ranged from 31.6 to 72.4 s. After admission, systolic blood pressure in the 2 patients with intracerebral hemorrhage was maintained below 140 mmHg, and the subdural hematomas in 4 patients were surgically treated. None of the hematomas became enlarged and outcome was good in most cases. Conclusions: Hematomas that arise due to acute intracranial bleeding during dabigatran treatment seem to remain small to moderate, hard to expand, and manageable.

This study included 40 consecutive patients with athrothrombotic carotid occlusive disease (A-group) and 13 consecutive patients with moyamoya disease (M-group) who had an internal carotid artery occlusion and underwent EC-IC bypass. Flow velocity and diameter of the operated STA on duplex ultrasonography (STDU), as well as regional cerebral blood flow (rCBF) on single photon emission computed tomography (SPECT) were measured before, 14 days after, and 3 months after EC-IC bypass. The postsurgical changes in the ipsilateral STA mean flow velocity (MFV) were significantly higher (p = 0.0030) and those in the rCBF in the MCA territory were relatively higher (p = 0.0936) in the M-group than the A-group patients. On 14 days after EC-IC bypass, the STA MFV (76.0 ± 22.5 vs 55.2 ± 16.5 cm/s, p = 0.0006) and the rCBF (40.0 ± 8.0 vs 34.2 ± 5.9 ml/100 g/min, p = 0.0065) were significantly higher in the M-group than in the A-group. On 3 months after EC-IC bypass, these differences in the STA MFV or the rCBF disappeared. There were no significant differences in the postsurgical STA diameter and the ACZ reactivity between both groups. Changes in the STA MFV as well as the rCBF were higher in moyamoya disease than atherothrombotic carotid occlusive disease in the early phase after EC-IC bypass. The STA MFV is highly correlated with the rCBF after EC-IC bypass. © 2012 Elsevier B.V.

Background and Purpose: The purpose of the present study was to clarify the difference in the infarct topography on diffusion-weighted image (DWI) and cardiac and aortic findings on transesophageal echocardiography (TEE) in stroke patients with different embolic sources. Methods: We studied 270 consecutive patients with acute ischemic stroke who had DWI-documented cortical or subcortical infarcts without significant stenosis of the cerebral arteries. As embolic sources, cardiac diseases, right-to-left shunt diseases and an aortic arch atheroma >= 4.0 mm in thickness were identified using various diagnostic tools including TEE. Results: Seventy-eight (29%) patients had multiple embolic sources. Large infarcts were common in patients in whom cardiac disease was the only embolic source and uncommon in patients in whom aortic atheroma was the only embolic source (p < 0.0001). Vertebrobasilar infarcts were relatively common in patients only having aortic atheromas. Atrial septal aneurysms were more common in patients with a right-to-left shunt than in those with a shunt plus other embolic sources (p = 0.0036). Unique characteristics of the arch atheroma (mobile plaque, extension to branches, or ulcer formation; p < 0.0001) as well as small or moderate-sized infarcts (p = 0.0004) were more common in patients with arch atheromas as the only embolic source than in those with atheromas plus other embolic sources. Conclusions: Embolic stroke patients often have multiple embolic sources. The present study suggests the possibility that embolic stroke has unique clinical features depending on its source. DWI and TEE findings might be helpful in characterizing cardiogenic, paradoxical and aortogenic brain embolism. Copyright (C) 2011 S. Karger AG, Basel

Background and Purpose-The purpose of this study was to determine at which time points acute blood pressure (BP) was associated with neurological deterioration at 3 weeks in patients with ischemic stroke as a whole and in patients with different subtypes. Methods-BP was measured every 6 hours for the first 36 hours of emergent hospitalization in 565 consecutive patients (347 men, 70 +/- 11 years in age) presenting within 24 hours of an acute ischemic stroke. Neurological deterioration was defined as a >= 2-point increase in the National Institutes of Health stroke scale (NIHSS) score at 3 weeks compared to the admission score. Results-At 3 weeks, 64 patients (11.3%) had deteriorated neurologically. For the group as a whole, high systolic BP (SBP) values measured at 12, 18, 24, and 36 hours postadmission were independently related to neurological deterioration after adjustment for age, sex, and known predictors, including admission NIHSS score, admission blood glucose level, and large infarct size. At 24 hours, the odds of neurological deterioration increased by 20% per 10-mm Hg increase in SBP. For cardioembolic stroke patients, high SBP values measured at 12 through 36 hours were independently related to neurological deterioration after multivariate adjustment. For patients having stroke other than cardioembolism, no SBP values at any time point were related to neurological deterioration. Conclusions-Acute SBP values between 12 and 36 hours postadmission, but not those on admission or at 6 hours, were predictive of neurological deterioration within the initial 3 weeks of ischemic stroke, particularly for cardioembolic stroke patients. (Stroke. 2009; 40: 2585-2588.)

Objective. Restenosis of the carotid artery after carotid endarterectomy (CEA) is a major complication. The frequency, time of occurrence, and tissue characteristics of carotid restenosis were assessed with sonography. Methods. Two hundred sixteen patients who had CEA for carotid stenosis were studied; follow-up sonography and magnetic resonance angiography were done 2 weeks, 3 months, and then every year after CEA. On sonography, restenosis was defined as an internal carotid artery (ICA) with a peak systolic velocity of 170 cm/s or greater or a maximum area of stenosis of 90% or greater. Results. During 605 artery-years of follow-up, 18 patients (7.5%) were found to have restenosis on sonography: 4 at 3 months, 11 at I year, and 3 at 2 years after CEA. At the time that restenosis was detected, in all 18 ICAs the peak systolic velocity exceeded 200 cm/s and had more than doubled since the last measurement (mean +/- SD, 103 +/- 27 to 321 +/- 107 cm/s), whereas the area of stenosis exceeded 90% in 6 patients, and magnetic resonance angiography revealed stenosis of 60% or greater in 8 patients. On sonography, all of the restenotic plaques were isoechoic and concentric. The restenosis was asymptomatic in 17 patients. Vascular risk factors or the severity of initial carotid stenosis before CEA were not associated with development of restenosis. Eleven patients had successful endovascular therapy, and the others received medical treatment. Conclusions. A marked increase in the flow velocity through an operated ICA is a good indication of restenosis. The isoechogenicity and concentricity of the restenotic plaques suggest that the restenosis is primarily the result of intimal hyperplasia.

Objectives: To explore the cerebral hemodynamics in subclavian steal syndrome, we examined the cerebral perfusion of seven patients with subclavian steal (one symptomatic and six asymptomatic) using single-photon emission computed tomography (SPECT) during resting, arm exercise, and acetazolamide-activated conditions. Methods: The regional CBF (rCBF) was measured with SPECT under all conditions, and region of interest (ROI) analysis was performed using a three-dimensional stereotaxic ROI template (3DSRT). We evaluated the relationship between arm exercise-induced rCBF change and (1) presence of subclavian artery stenosis, (2) vertebral reverse flow severity, (3) presence of vertebro-basilar insufficiency (VBI) symptoms, and (4) cerebrovascular reactivity (CVR) to acetazolamide. Results: Overall, no arm exercise-induced rCBF reduction was observed on either the affected or the unaffected side, even in patients with severe vertebral reverse flow. One patient with VBI symptoms showed an arm exercise-induced global rCBF reduction in the cerebrum and cerebellum, whereas the other asymptomatic patients did not. The %rCBF changes in segments with severely impaired CVR (-8.6%+/- 10.7%, mean +/- SD) were significantly lower than those in other segments with less impaired CVR (P < 0.01). Conclusions: Our results suggest that subclavian steal is a benign condition in asymptomatic patients. On the other hand, arm exercise-induced rCBF reduction can occur in the cerebrum and cerebellum in patients with VBI symptoms possibly related to low CVR.

Background: To determine the underlying conditions that affect the degree of calcification of carotid arterial plaques, measured quantitatively using multidetector row computed tomography (MDCT), and to study the association of carotid calcification with clinical symptomatology. Methods: We measured the calcification volume of stenotic lesions at the carotid bifurcation using MDCT in 84 consecutive patients who were scheduled to undergo carotid revascularization. These results were compared with the clinical and radiological characteristics of the patients. Results: On MDCT, calcification in the carotid plaques was present in 78 patients (93%). Compared to the other patients, patients in the highest quartile of calcification volume (quartile 4) had higher serum creatinine levels (p < 0.001) and tended to have fewer symptomatic ischemic events in the territory of the affected carotid artery in the preceding 6 months (29 vs. 49%, p = 0.099); in particular, there were fewer transient symptoms ( 5 vs. 27%, p = 0.032) and symptoms possibly occurring due to local embolism (14 vs. 37%, p = 0.045). On ultrasound, plaque ulceration was less prevalent in patients in quartile 4 than in the remaining patients (5 vs. 29%, p = 0.026), although the severity of carotid stenosis was similar among all the quartiles. Conclusions: Renal dysfunction was associated with enhanced carotid plaque calcification. Patients with severe carotid calcification were found to have a low risk of recent ischemic stroke, presumably due, in part, to a lower prevalence of emboligenic carotid ulceration. MDCT was valuable for the quantitative evaluation of carotid calcification. Copyright (c) 2007 S. Karger AG, Basel.

Background and Purpose-The purpose of this study was to clarify the differences in the acute blood pressure course among different ischemic stroke subtypes. Methods-We divided 588 consecutive patients with acute brain infarction into four clinical subgroups to study the blood pressure levels during the initial 6 hospital days. Results-During the 6 days. systolic blood pressure of lacunar and atherothrombotic patients was higher (P=0.0001) and diastolic blood pressure of lacunar patients was higher (P=0.0371) than of patients with the other subtypes. Preexisting hypertension was associated with elevated acute systolic blood pressure in all patients and in each subtype and with elevated acute diastolic blood pressure in all patients, cardioembolic patients, and patients with stroke of other etiology. After adjustment by preexisting hypertension, diabetes mellitus with a hemoglobin A1c > 7.0% was associated with elevated systolic blood pressure in all, lacunar, and cardioembolic patients and with diastolic blood pressure in all patients. Conclusions-Blood pressure course of patients sustaining acute stroke varied widely according to stroke subtypes. Poorly controlled diabetes mellitus, as well as preexisting hypertension, appeared to influence blood pressure during the initial week of stroke.

Using transcranial color-coded sonography (TCCS), we evaluated the acute changes in the hemodynamics of cerebral hyperperfusion in two cases. The mean flow velocity of the cerebral arteries increased at the onset of clinical symptoms, together with an increase in the regional cerebral blood flow (rCBF). In serial follow-up studies, the flow velocity gradually returned to normal in parallel with the normalization of the rCBF values. TCCS can be useful for evaluation of acute cerebral hyperperfusion. (E-mail: fujimoto.s@ns.yawatamhp.or.jp) (c) 2006 World Federation for Ultrasound in Medicine & Biology.

Background: To investigate the accuracy of conventional carotid ultrasonography ( CCU) combined with transoral carotid ultrasonography ( TOCU) for distinguishing pseudo-occlusion from total occlusion of the internal carotid artery ( ICA). Methods: This study included 95 patients who were suspected of having an occlusion of the ICA on magnetic resonance angiography ( MRA) and underwent both CCU and conventional digital subtraction angiography ( DSA) in order to confirm the diagnosis. TOCU was also performed to observe the cervical portion of the ICA distal to the stenosis. We compared the ultrasonographic findings with the DSA findings. Results: Twelve of the 95 patients were defined as having an ICA pseudo-occlusion on DSA. On B-mode images with CCU color Doppler, slight residual flow signals in the ICA lumen were shown in 20 patients. Among them, 2 patients had a pulsed Doppler waveform of the distal ICA occlusion pattern. Among the remaining 18 patients, 4 had a pulsed Doppler waveform of the to and fro flow pattern, and 14 had a weak antegrade flow pattern in the ICA lumen. The conventional ultrasonographic method showed 100% sensitivity with 93% specificity for diagnosing an ICA pseudo-occlusion. The addition of TOCU findings increased the specificity to 98%. In 2 patients, who were overdiagnosed as having an ICA pseudo- occlusion even using TOCU, DSA revealed an occlusion of the ICA distal to the ophthalmic artery with a severe stenosis of the proximal ICA. Conclusions: Using conventional and transoral carotid ultrasonography, an ICA pseudo- occlusion can be diagnosed with higher accuracy. Copyright (c) 2006 S. Karger AG, Basel

Background and Methods: We studied changes in brain volume on magnetic resonance imaging in 10 patients with and without extracranial-intracranial ( EC- IC) bypass surgery who met the inclusion criteria for the Japanese EC- IC Bypass Trial, a multicenter, randomized, prospective study of patients with hemodynamic brain ischemia due to cerebral artery occlusive disease. We also examined the association of cerebral hemodynamics on single-photon emission computed tomography with the changes in brain volume. Results: The affected/unaffected ratio of the percent brain volume declined in patients without EC- IC bypass surgery ( p < 0.02, n = 4), and the affected/ unaffected percent regional cerebral blood flow ratio increased in patients with the surgery ( p < 0.03, n = 6). Acetazolamide reactivity increased in the affected hemisphere of patients with surgery ( p < 0.01). Two- year increase ( decrease) in acetazolamide reactivity of the affected hemisphere showed a significant positive correlation with 2- year changes in the affected/ unaffected percent brain volume ratio ( R-2= 0.737, p = 0.0007). Conclusions: Change in acetazolamide reactivity might be a good predictor for brain atrophy in cerebral artery occlusive disease. Copyright (c) 2006 S. Karger AG, Basel

Background: The Acute Candesartan Cilexetil Therapy in Stroke Survivors (ACCESS) study indicated that early treatment with an angiotensin type 1 receptor blocker in acute stroke patients who had relatively high blood pressure improved cardiovascular morbidity and mortality in the chronic stage. To better interpret the findings of this study, we determined whether stroke patients with high acute blood pressure had specific underlying conditions. Methods: We divided 712 consecutive patients who were hospitalized within 48 h after the onset of brain infarction into two groups: 77 patients with high acute blood pressure that met the criteria of the ACCESS study and the 635 remaining patients. Underlying risk factors and comorbidities, stroke characteristics, as well as mortality, vascular events, and disability at 3 weeks were compared between the two groups. Results: Patients with high acute blood pressure more frequently had diabetes mellitus (p < 0.01), intracranial arterial stenosis (p < 0.02), higher levels of hemoglobin A1c (p < 0.005), higher creatinine levels (p < 0.005), and tended to more frequently have ischemic heart disease (p < 0.09) and infarcts < 1.5 cm in diameter (p < 0.09) than the other patients. On multivariate analysis, high levels of hemoglobin A1c, high creatinine levels, and intracranial arterial stenosis were independently predictive of high acute blood pressure. At 3 weeks after the stroke onset, patients with high acute blood pressure were more dependent in their daily living activities (p < 0.02) and more frequently developed vascular events or death (p < 0.005) than the other patients. Conclusions: Poorly controlled diabetes mellitus and advanced renal damage appeared to correlate with acute hypertension after stroke. Since intracranial arterial stenosis also seemed to contribute to high acute blood pressure, one should be careful not to induce cerebral hypoperfusion by the early use of antihypertensives. Copyright (c) 2006 S. Karger AG, Basel.

Background: To investigate the utility of superficial temporal artery (STA) duplex ultrasonography (STDU) for evaluating the improvement of the cerebral hemodynamics after extracranial-intracranial (EC-IC) bypass. Methods: This study included 40 consecutive patients who underwent EC-IC bypass for occlusive disease of cerebral arteries. STDU was performed to measure the flow velocity, pulsatility index, and diameter of the operated STA before and 14 days after EC-IC bypass. Regional cerebral blood flow (rCBF) and acetazolamide (ACZ) reactivity of the ipsilateral middle cerebral artery (MCA) territory were evaluated by quantitative single-photon emission computed tomography with the ACZ challenge test. We investigated the correlation between STA flow velocity/diameter and rCBF/ACZ reactivity in the ipsilateral MCA territory. Results: Mean flow velocity (MFV; 26.3 +/- 8.8 to 55.3 +/- 16.3 cm/s, p < 0.0001) and diameter and rCBF (29.1 +/- 3.1 to 35.0 +/- 6.4 ml/100 g/min, p < 0.0001) and ACZ reactivity (-0.02 +/- 0.10 to 0.28 +/- 0.21, p < 0.0001) of the MCA territory increased after EC-IC bypass compared with the baseline values. STA MFV was significantly correlated with the rCBF 14 days after EC-IC bypass (R = 0.70, p < 0.0001). A cutoff value of postsurgical STA MFV greater than 48.5 cm/s yielded the highest diagnostic accuracy (sensitivity 86%; specificity, 82%) for rCBF 6 32 ml/100 g/min after EC-IC bypass. Conclusions: STDU was available for evaluating postsurgical patency of the bypass flow and the rCBF of the ipsilateral MCA territory. The mean blood flow velocity of the operated STA is a highly sensitive parameter for predicting rCBF in the ipsilateral MCA territory after EC-IC bypass. Copyright (C) 2005 S. Karger AG, Basel.

Background and Purpose-The purpose of the present study was to evaluate availability of transcranial color-coded real-time sonography (TCCS) to detect hyperperfusion after carotid endarterectomy (CEA). Methods-This prospective study included 105 consecutive patients who underwent CEA for severe carotid stenosis. TCCS with echo contrast agents was performed serially to evaluate flow velocity of the middle cerebral artery (MCA). Regional cerebral blood flow (rCBF) and vasodilatory capacity of the MCA territory were evaluated using single-photon emission computed tomography. We compared the changes in MCA flow velocity with rCBF. Results-Using echo contrast agents, we could evaluate the MCA flow in 95 (90%) of 105 patients. Twelve patients showed hyperperfusion syndrome. Changes in the MCA mean flow velocity (MFV) before and 4 days after CEA were significantly correlated with those in rCBF (r = 0.48; P < 0.0001). An increase of >50% in MCA MFV was observed within 4 days after CEA in all 12 patients with hyperperfusion syndrome. Multivariate analysis revealed that reduced vasodilatory capacity ( odds ratio, 0.14; 95% CI, 0.04 to 0.46) was an independent risk factor for a 1.5-fold increase in the MFV of MCA ipsilateral to CEA. Conclusions-Findings of a 1.5-fold increase in the MCA MFV can accurately identify those patients with high risk of developing post-CEA hyperperfusion syndrome. TCCS with echo contrast agents is available for the evaluation of hyperperfusion after CEA.

The goal of this study was to determine whether there was difference in behavioral situation at onset of stroke between the patients with right-to-left shunt whose stroke was diagnosed as paradoxical brain embolism according to the new criteria proposed by Strategies against Stroke Study for Young Adults in Japan (SASSY-Japan) and whose stroke was not. Among 365 consecutive patients with possible acute ischemic stroke who underwent transesophageal echocardiography, we compared clinical profiles of the following three patient groups: patients with patent foramen ovale whose stroke was diagnosed as paradoxical brain embolism (group A), patients with patent foramen ovale whose stroke was not diagnosed as paradoxical brain embolism (group B), and patients without patent foramen ovale whose stroke was diagnosed as cardiogenic brain embolism (group C). Patent foramen ovale was present on echocardiogram in 76 of 365 patients (21%). Among them, 19 patients were classified into group A and 34 into group B. Group C was composed of 69 patients. At stroke onset, 3 patients in group A took behavior with positive Valsalva maneuver, 2 in group B, and 1 in group C. Similarly at stroke onset, 4 patients in group A just stood up from long-time sitting position, 1 in group B, and 1 in group C. In all, stroke onset following Valsalva maneuver or long-time sitting position was positive in 37% of group A, 9% of group B, and 3% of group C, and was most frequent in group A (p&lt 0.0001). In group A, the frequency of Valsalva maneuver or long-time sitting position at stroke onset was higher than the frequency of detection of venous thrombus (21%). The result indicates that Valsalva maneuver and long-time sitting position contribute to occurrence of paradoxical brain embolism. We think that these behavioral situations are appropriate diagnostic criteria for paradoxical brain embolism.

Background and Purpose-Aortic arch atherosclerotic lesions are often associated with embolic brain infarction. We investigated the relationship between stroke recurrence and the characteristics of aortic arch atherosclerotic lesions. Methods-Among 487 stroke patients who underwent transesophageal echocardiography, 283 patients with brain embolism diagnosed without significant occlusive lesions (greater than or equal to50%) in their cerebral arteries were included in this study. We measured the intima-media thickness (IMT) and evaluated the extension and mobility of the aortic arch atherosclerotic lesions. During a mean follow-up period of 3.4 years, we investigated the relationship between stroke recurrence and the various characteristics of the aortic arch atherosclerotic lesions. Results-An IMTgreater than or equal to4.0 mm was found in 67 patients (25.3%). In 51 of these patients, the aortic lesions extended to the origin of the branches of the arch. Recurrences of cerebral ischemic events were found in 32 patients (recurrence group) and not in the other 251 (nonrecurrence group). Aortic atheroma greater than or equal to4.0 mm (41% versus 22%), aortic atheroma extending to the branches (63% versus 39%), and both (38% versus 16%) were more frequently seen in the recurrence group than in the nonrecurrence group (P<0.05, P<0.1, P<0.01, respectively). After adjustment for age and the presence of hypertension, an aortic atheroma that was ≥4.0 mm as well as extending to the branches was found to be an independent predictor of ischemic stroke recurrence (hazard ratio=2.42, P<0.05). Conclusions-Stroke recurrence is associated with the severity of the atheroma (IMTgreater than or equal to4.0 mm) and plaque extension to the branches.

The first purpose of this study was to detect clinical and radiological factors on admission which predict early outcome of patients in brain hemorrhage with medical treatment. For 50 consecutive patients in our Cerebrovascular Center, NIH Stroke Scale score was a useful indicator for prediction of independent daily life, discharge to home, and death in the acute stage the score ≦ 4, ≦ 7, and ≧ 23 were the most appropriate cut-off values for the above events, respectively. The second purpose of this study was to produce critical paths of medical management for brain hemorrhage based on the above results. We prepared three courses of paths according to clinical severity. As inclusion criteria for each course, we used the above three cut-off values and hematoma volume. Duration of hospitalization of the three courses was 16, 20, and 28 days. When we applied the paths to 200 patients with brain hemorrhage who were enrolled in the research grant supported by the Japanese ministry of health, labor and welfare (12C-10), duration of hospitalization for the majority of the patients were 5 days or more than the planned duration in the paths. It is indispensable to manage acute stroke patients according to critical paths, because standard and efficient strategies of clinical medicine have been stressed these years. We will immediately apply the new paths in this study to patients in our center, and renew them at short intervals. We think that we can contribute to new evidences for standard medical management of brain hemorrhage by our approach to the critical paths.

Dural arteriovenous fistulas (AVFs) cause several types of intracranial hemorrhage, but rarely cause primary intraventricular hemorrhage (IVH). We report a 67-year-old man with sudden headache and a long history of a pulsatile bruit who developed intraventricular hemorrhage without any parenchymal hemorrhage. Cerebral angiogram revealed dural arteriovenous fistulas in transverse and sigmoid sinuses. Severe retrograde venous drainage seemed to have caused backward flow into the subependymal veins with their consequential rupture. Transvenous embolization was successful. (C) 2003 Elsevier B.V. All rights reserved.

An 86-year-old woman developed cardioembolic stroke three times. In the last one, she fell into sudden coma and fatal outcome due to acute occlusion of bilateral internal carotid arteries (ICAs) and the basilar artery. Diffusion-weighted magnetic resonance imaging (MRI) delineated brain ischemia in the whole bilateral cerebral hemisphere soon after the stroke onset. Signal intensity of the brain parenchyma increased in whole the hemisphere. Especially, all the cortical rims glittered. This is the first report of the cardioembolic stroke due to simultaneous occlusion of the three major arteries to the brain. (C) 2003 Elsevier Science B.V. All rights reserved.

Extracorporeal double filtration plasmapheresis (EDFP) can quickly lower plasma viscosity and fibrinogen concentration. EDFP has the potential to improve cerebral microcirculation in acute ischemic stroke and ultimately to salvage penumbral tissue. However, no evidence is available to show that EDFP can increase cerebra blood flow (CBF). Therefore, we investigated whether EDFP could increase CBF by quantitative CBF measurements and documented the clinical effects of EDFP in acute ischemic stroke. EDFP was performed ten times in seven patients diagnosed as having acute atherothrombotic brain infarction caused by major artery occlusive lesion. They also fulfilled one of the following entry criteria: 1) diffusion/perfusion mismatch demonstrated by MRI on admission; 2) a hemispheric syndrome, but only a small lesion on diffusion weighted MRI (< 25% of MCA territory); or 3) progressing stroke. Exclusion criteria were 1) contraindication of heparin or 2) spontaneous improvement of symptoms. Time from stroke onset to EDFP varied from 5 hr to 7 days. Plasma viscosity was quickly lowered by EDFP without affecting RBC counts, Hb, or Hct in all patients. Positron emission tomography (PET) with 15-O labeled H2O measurements revealed a significant CBF increase from 36.4 +/- 8.3 ml/100 g/min to 40.7 +/- 6.8 ml/100 g/min in the affected hemisphere (P = 0.048). Definite CBF improvement was also demonstrated by single photon emission computed tomography (SPECT) in one of two patients who had severe stenosis of the middle cerebral artery. Furthermore, this patient showed remarkable improvement of hemiplegia immediately following EDFP (NIHSS score: 18 to 13). In conclusion, EDFP can increase CBF in ischemic brain tissue in acute atherothrombotic brain infarction. Further clinical studies should focus on the efficacy of EDFP on outcome of patients with this stroke subtype. J. Clin. Apheresis 18:167-174, 2003. (C) 2003 Wiley-Liss, Inc.

The accuracy of the acetazolamide (ACZ) challenge test using semiquantitative SPECT in detecting Stage II hemodynamic failure, i.e. cerebral misery perfusion, in patients so diagnosed with PET has yet to be determined. This study was carried out in 53 patients who had a unilateral occlusion or severe stenosis of their cerebral artery. Asymmetry index (AI) was used to determine relative CBF distribution on each SPECT image. DeltaAI (regional vasodilatory capacity) values were compared with the values of several PET parameters. We also repeated SPECT and PET studies in 15 patients after a mean interval of 2.6 years from entry. The correlation between changes in DeltaAI and PET data were examined. The DeltaAI value closely correlated with the ipsilateral OEF value (r=-0.64, p<0.001). Sensitivity-specificity curve analysis revealed that the optimal cut-off value for detecting Stage II hemodynamic failure was -13.5% of &UDelta;AI, at which point the diagnostic accuracy was 82%. In follow-up studies, changes in &UDelta;AI correlated significantly with changes in OEF values (r=-0.68, p<0.01). Semi quantitative SPECT examination with ACZ challenge detects Stage 11 hemodynamic failure with a diagnostic accuracy of 82%. An improvement in reduced vasodilatory capacity as determined by SPECT coincides with a reduction in OEF values. (C) 2002 Elsevier Science B.V All rights reserved.