[en] The correlation between the findings on MR images and CT findings, clinical symptoms, and the results of various specific investigations was determined. MR imaging was more sensitive in demonstrating the pathology of the periventricular tissue in hydrocephalus than was CT scanning, although the overall correlation between MR and CT findings was good. The authors classified the periventricular high-signal intensities (PVHI) of MR images into 4 types according to the extension and grade of the PVHI. Type 0 has no PVHI. Type I has a band of the PVHI around the wall of the lateral ventricles, including the corpus callosum and the septum pellucidum. In Type III, PVHI extended half of the width of the periventricular white matter, while in Type IV PVHI occupied the entire white matter. CT scans showed periventricular lucency (PVL) only in the periventricular white matter, especially surrounding the anterior horn, but the PVHI was visible not only in wider areas of the entire periventricular white matter, but also in the corpus callosum and the septum pellucidum. The PVHI seen in case of hydrocephalus due to posterior fossa tumors was characterized by PVHI of Type I; the PVHI disappeared after CSF shunting in all cases. There was no PVHI in the cases of marked hydrocephalus due to aqueductal stenosis, which appeared to be compensated for a long period. In the cases of normal-pressure hydrocephalus, there were various types of PVHI. There was no correlation between the types of PVHI and the grade of ventriculomegaly. There was, however, a close correlation between the types of PVHI on MR images and the clinical symptoms, and the cerebral blood flow in cases of normal-pressure hydrocephalus. In some patients with normal-pressure hydrocephalus, the PVHI improved and the cerebral blood flow increased after cerebrospinal fluid shunting. (J.P.N.)

[en] Xenon-enhanced brain CT has been used to determine the local cerebral blood flow (LCBF) noninvasively. By this method, a LCBF map can be generated, but it requires the calculation of numerous data and needs display equipment with a high resolution. Recently various personal computers have been developed for processing CT images. We ourselves have now developed a CT-image-processing system using a personal computer (NEC PC-9800, 8086cpu), and found it adequate for generating a LCBF map. This image-processing procedure consists of several steps; the conversion of the original CT image data into the data files of a personal computer, the shifting of the image data to correct head movement, the diminution of the CT noise through a filtering process, the calculation of the partition coefficient and the build-up rate constant in each pixel, and finally displaying the LCBF map image. Generated images have a sufficient quality and have been found satisfactory for a clinical use. (author)

[en] After the introduction of the high-resolution technique in CT scan, air CT cisternography (air CTC) has been one of the most sensitive neuroradiological modalities in the detection of small acoustic neurinomas. However, the number of false-positives is increasing, together with the addition of intracanalicular tumors visualized by air CTC. In this study, we examined the value of intracanalicular enhancement (IE) in the differential diagnosis between ''ear tumor'' and its false-positive. Eight cases (4 negative, 3 positive, and 1 false-positive) were analysed retrospectively. All but the four negative cases has been histologically verified on surgery. IE was judged to be positive when some diffuse enhancement was observed in the middle of the enlarged canal on serial thin-section scans under the display setting (window level and width) similar to the brain. Three intracanalicular tumors showed an apparent enhancement, while the three non-tumor cases and one false-positive case showed no enhancement. IE was inconclusive in the one remaining case without a tumor, possibly because of the narrow canal (5 mm). On the basis of these investigations, even though the number of cases has been limited, we can say that IE seems a very promising way to exclude false-positives in the diagnosis of intracanalicular tumors by air CTC. (author)

[en] A sequential MR scan was performed on 21 patients with intracranial hematoma, and simultaneously the T₁ values of the hematomas were calculated. The T₁ value of a hematoma was found to be longer than that of the white matter in the acute phase, but it soon becomes as short as that of the white matter (7 - 10 day after). After several days, the T₁ value again gradually becomes longer. In the experiment, 30 ml of fresh blood (15 samples) were stored at room temperature, and a sequential MR scan and the calculation of the T₁ were performed over a period of 20 days. In vitro, most of the T₁ values were long, but there was much variation on the first day. A shortening of the T₁ was observed as well in vivo, and after this shortening, no prolongation of the T₁ was observed. Perhaps the shortening of T₁ was caused by the denaturation of the hemoglobin to methemoglobin and by the coagulation of the blood. The lysis and absorption of the hematoma may, on the other hand, cause the prolongation of the T₁ in vitro. For the diagnosis of intracranial hematoma, CT was found to be a method superior to MRI, especially in the acute phase. However, MRI gives us more information about hematoma (concerning the denaturation of the hemoglobin to methemoglobin, the lysis and absorption of the hematoma, the range of hemorrhagic tissue and edema, etc.) than does CT. An IR (T₁-weighted) image shows a good contrast between the hematoma and the surrounding tissue (hemorrhagic tissue, edema) in the early phase. On the other hand, the SE (T₂-weighted) image informs us of the lesion when the hematoma is low ∼ isodense on the CT in the chronic phase. (author)

[en] The authors document their experience with the computerized tomographic (CT) scan technique for evaluating a model gunshot wound of the head. In the preoperative period, the CT scan is useful for identifying and localizing the missile track, any long, metallic fragments, intra and extraparenchymal hematomas, intracranial air, and brain edema. The limitations of the CT scan in evaluating a gunshot wound include an inability to define vascular lesions, such as traumatic aneurysm and post-traumatic spasm. Metallic scatter from missile fragments may also render certain CT cuts uninterpretable. The CT scan can, however, help to determine the nature of intracranial lesions. Thus, it is invaluable for the rational planning of surgical therapy. (author)

[en] We reported three cases of neuro-Behcet's syndrome which showed brainstem lesions on MRI compatible with the clinical symptoms. In Case 1, MRI showed a large, abnormal signal-intensity area in the pons and small, abnormal signal-intensity areas at the right cerebral peduncle, the bilateral basal ganglia, and the left thalamus. These lesions disappeared on MRI, in accordance with the remission of clinical symptoms. On the other hand, CT showed no positive findings. In Case 2, an abnormal signal-intensity area was disclosed at the left cerebral peduncle on MRI. This lesion was also identified on the CT scan. In Case 3, an abnormal signal-intensity area was present in the pons on MRI. In this case, CT showed no positive findings. In Cases 2 and 3, these lesions seemed to represent inflammatory or necrotic areas attributable to vasculitis; however, the extensive brainstem lesion seen on the MRI of Case 1 was a quite unique finding, for which no exact pathophysiological explanation is possible at the present time. (author)

[en] A case of the persistence of the primitive hypoglossal artery is reported, with a roentgenographic demonstration of the enlarged hypoglossal canal. A 63-year-old man was admitted to this hospital as a result of a malfunction of the ventriculo-peritoneal shunt. At the age of 51, the patient had been operated on in this hospital for an aneurysm of the right middle cerebral artery; at that time, the right primitive hypoglossal artery was observed on right carotid angiograms. On the day following admission, bilateral retrograde vertebral angiography was performed and the right persistent primitive hypoglossal artery was recognized again. Stenvers views of the skull demonstrated an enlargement of the hypoglossal canal, with a smooth sclerotic rim. High-resolution computed tomography with a contrast infusion delineated the right primitive hypoglossal artery through the enlarged hypoglossal canal. The diameter of the enlarged right hypoglossal canal and that of the left one were found to be 8 mm and 4 mm respectively on the CT. When an enlargement of the hypoglossal canal with a sclerotic rim is observed, the persistence of the primitive hypoglossal artery should be considered in the differential diagnosis. (author)

[en] Sequential changes in magnetic resonance imaging (MRI) were investigated in comparison with computed tomography (CT) in 31 cases of head trauma. Twenty-one of them were of acute head trauma; the first MRI study was performed within 48 hours after the accident. Forty-two intracranial lesions were observed in these cases on MRI. The other 10 cases were of chronic subdural hematoma, two cases of which had bilateral lesions. Fourteen lesions of acute head trauma and two lesions of chronic subdural hematoma were detected only by MRI. MRI was superior to CT for the detection of small contusions and thin extra-axial collections, especially those which were located near the bony structures. The abnormal lesions were visualized in MRI during a longer period than in CT. Because the signal intensity of a hematoma changed sequentially, the detection of brain edema was easier than that of a subarachnoid and parenchimal hemorrhage. Judging from this experience, it seems that careful attention should be taken in the diagnosis of hemorrhagic lesions. However, MRI was poor in tissue characterization because of the too-high tissue sensitivity. T₂-weight SE imaging was essentially sensitive and useful in the early stage. (author)

[en] An enlarged superior ophthalmic vein (SOV) on computed tomography has been considered by several authors to be a pathognomonic sign of a carotid cavernous sinus fistula (CCF). However, according to some other investigators, SOV has been observed in various orbital and cavernous sinus diseases, and even in normal persons. We reviewed the ordinal axial head and orbital CT results (third generation) of 1293 patients with miscellaneous neurologic diseases in an attempt to ascertain the diagnostic significance of SOV. SOV was observed in 88 patients, unilaterally in 29 and bilaterally in 59. SOV was detected in 65 patients by means of 5-mm-thick slice scanning and in 23, by 10-mm. SOV was observed in 76 cases on plain CT and in 12 on enhanced. The maximal diameter of SOV was 3.3 mm of more in all patients except one with orbital or cavernous sinus desease (a skullbase tumor extending into the cavernous sinus, a case of orbital cellulitis, and a case with CCF) in the present study. On the other hand, the maximal diameter of SOV with diseases other than orbital of cavernous sinus pathology was less than 3.0 mm. In conclusion, a SOV is not pathognomonic on CCF and is observed under various conditions. However, when the diameter of the SOV is 3.3 mm or more one should consider the possibility of orbital or cavernous sinue disease. (author)

[en] Many reports have been concerned with the effectiveness of CT scan and cerebral angiography in predicting the histological grading of intracranial astrocytomas, since those with a contrast-enhancing effect with perifocal edema on CT are likely to be malignant and those with tumor vascularity and early venous drainage on angiography must be malignant. On reviewing 50 cases of intracranial astrocytomas, we encountered two groups showing some discrepancy between the presurgical radiological findings and the postsurgical (or autopsied) histological diagnosis. In the first group of 7 cases, CT showed an apparent contrast enhancement within the tumor, whereas angiography showed no tumor vascularity. Four of these were cases of glioblastoma multiforme, and only two were low-grade astrocytomas. In the second group (8 cases), no contrast-enhancing effect of the tumor was observed on CT, while an unequivocal tumor blush was demonstrated by angiography. Of these, the four with angiographically early venous filling were Grade III astrocytoma, while the other four, lacking it, were Grade II. This result is another suggestion that the contrast enhancement of astrocytomas depends upon not only the vasoproliferation but also the extravasation of the contrast medium through a disrupted blood-brain barrier. A lack of tumor vascularity on angiography does not always reflect benign histology in cases of contrast enhancement on CT. Grade II or III astrocytomas may not have any contrast enhancement on CT when their bloodbrain barrier is mature or not disrupted, even if there is hypervascularity on angiography. (author)