Axial vs sagittal T2-weighted brain MR images in the evaluation of multiple sclerosis

Axial and sagittal proton density and T2-weighted MR images (TR 2,500-3,000 ms, TE 15-22 and 85-90 ms) were performed in 50 patients with multiple sclerosis (MS) on a 1.5 T superconductive system. The number of plaques on the axial and sagittal images in the periventricular white matter, the corpus callosum, the brain stem, the cerebellum, and the basal ganglia were counted separately by two independent observers. A total of 858 lesions (mean 17.40 +/- 21.57) were seen on the axial series and 1,196 (mean 24.32 +/- 26.22) on the sagittal scans. More lesions were visualized on sagittal images in the periventricular region (mean 18.79 +/- 21.69 versus 13.34 +/- 16.45; p less than 0.001) and the corpus callosum (mean 3.00 +/- 2.72 versus 0.57 +/- 1.19; p less than 0.001). In the brain stem more lesions were visualized on the axial images (mean 1.55 +/- 2.55 versus 0.87 +/- 1.20; p less than 0.05). In the cerebellum and basal ganglia, scans in the two planes were equivalent (p greater than 0.5). In three patients lesions were seen on the sagittal series, while the axial scans were normal. Sagittal T2-weighted images appear to demonstrate significantly more MS plaques than transverse images, especially in the periventricular region and the corpus callosum. This is explained by partial volume averaging, by the orientation of some cerebral structures (e.g., corpus callosum) with regard to the section plane, and by the longer diameter of the lesions in the axial plane.     

MR features of intracranial hemangioblastomas.

The contribution of MR in 10 patients with surgically proven intracranial hemangioblastomas was retrospectively evaluated and correlated with the computed tomographic findings. Angiography was obtained in eight cases. Multiple lesions were shown in only two patients, but none proved to have von Hippel-Lindau disease. In six patients the tumor typically appeared as a cystic lesion with an intensely enhancing mural nodule at the pial surface of the cyst; three patients presented with a solid lesion with a central cyst and one with a solid lesion only. Thirteen tumors were located in the posterior fossa, with one in the brain stem extending towards the spinal cord. A supratentorial localization was found in two patients. Abnormal serpiginous vessels supplying or draining the mural nodule or solid lesion were not visualized on contrast-enhanced computed tomography, but were easily identified as flow voids on MR in five patients. Calcification or hemorrhage were not seen in the entire series. Because of the multiplanar imaging capability and the lack of streak artifacts from the petrous bone, MR provides better localization than CT. MR is more accurate in predicting the number, extension and vascularity of hemangioblastomas, and provides a sound basis for further diagnostic and therapeutic procedures.     

Ventricular enlargement, clinical correlates and treatment outcome in chronic schizophrenic inpatients.

The ventricle-brain ratio (VBR) of 42 chronic schizophrenic patients was compared with that of 42 age-matched medical controls. For the schizophrenics, the relationship of various clinical parameters to the VBR was assessed, and the outcome of 12 weeks of double-blind treatment with either risperidone or haloperidol. The results confirm that schizophrenic patients have slightly enlarged lateral ventricles compared with medical controls. Only for schizophrenics, an effect of age, but not of duration of illness, was noticed. This study does not support the validity of a clinical subdivision of chronic schizophrenic patients on the basis of the VBR. Neither negative, positive nor general psychopathological symptoms, as measured by the Positive and Negative Syndrome Scale for Schizophrenia (PANSS), were related to the VBR, nor were abnormal involuntary movements or extrapyramidal symptoms. No association between season of birth or a family history of major mental disorder and VBR could be demonstrated. Treatment response was predicted by the total PANSS score and the PANSS general psychopathology subscale score at baseline. There was a trend for patients with higher VBR to have a more or haloperidol). or haloperidol).