The terms posterior reversible leukoencephalopathy, reversible posterior cerebral edema syndrome, and posterior reversible encephalopathy syndrome (PRES) all refer to a clinicoradiologic entity characterized by headaches, confusion, visual disturbances, seizures, and posterior transient changes on neuroimaging. Clinical findings are not sufficiently specific to readily establish the diagnosis; in contrast, magnetic resonance imaging pattern is often characteristic and represents an essential component of the diagnosis of PRES. Typical lesions predominate in the posterior white matter, with some involvement of the overlying cortex; are hyperintense on T2-weighted images; and are usually hypointense or isointense on diffusion-weighted images, with an increase of the apparent diffusion coefficient, indicating vasogenic edema. The pathogenesis is incompletely understood, although it seems to be related to the breakthrough of autoregulation and endothelial dysfunction. Since its initial description, this syndrome has been subsequently described in an increasing number of medical conditions, including hypertensive encephalopathy, eclampsia, and the use of cytotoxic and immunosuppressive drugs. The diagnosis has important therapeutic and prognostic implications because the reversibility of the clinical and radiologic abnormalities is contingent on the prompt control of blood pressure and/or discontinuing the offending drug. On the contrary, when unrecognized, conversion to irreversible cytotoxic edema may occur.

The overall haemorrhagic risk of a cerebral arteriovenous malformation (cAVM) is 2-4% per year. However, the individual risk of haemorrhage has never been determined. This study was undertaken to assess the haemorrhage risk of an individual cAVM. Neuroangiographic findings of 160 cAVM were analysed retrospectively, looking at 30 angiographic features. A statistical model was established by logistic regression to evaluate the risk of an individual cAVM. We statistically correlated 15 parameters with the haemorrhage risk. The statistical model includes five independent parameters. Four are unfavourable: exclusively deep drainage, venous stenoses, venous reflux and the radio of afferent to efferent systems; one is favourable: venous recruitment. This model quantifies the individual risk of haemorrhage. When this model is applied to the population studied, the error rate is 5%. This model can contribute to therapeutic strategy, and to a better understanding of the natural history of cAVM.

We report five cases of intracranial dural arteriovenous fistula (DAVF) with perimedullary venous drainage. All the patients presented with rapidly progressive myelopathy and three had autonomic disorders. The DAVF were on the tentorium cerebelli (two cases), sigmoid (one), superior petrosal (one), and cavernous sinus (one). Slow venous drainage was directed through dilated perimedullary cervical veins. The transverse sinus was occluded in two cases. MRI, performed in four cases, demonstrated high signal on T2-weighted spin-echo sequences in the medulla oblongata and upper cervical spinal cord consistent with oedema, which signal resolved after complete cure of the DAVF in three cases. Embolisation was performed in all cases. It was followed by clinical deterioration in two cases and in the dramatic improvement in the other three, with complete clinical cure in two. Extensive venous thrombosis may explain the deterioration observed in one case.

Our preliminary observations prompt to perform a subsequent controlled study to examine if rTMS may constitute an alternative to electroconvulsive therapy.

Lateral sinus thrombosis may be difficult to differentiate angiographically from lateral sinus hypoplasia, which mainly affects its proximal transverse portion. Using magnetic resonance imaging, we evaluated six patients who demonstrated poor filling or lack of filling of one or both lateral sinuses at angiography. In each patient, magnetic resonance imaging unambiguously demonstrated either lateral sinus thrombosis or lateral sinus hypoplasia. The latter was characterized by a frank asymmetry in size (surface of section) of the transverse portion of the lateral sinuses on parasagittal images without any abnormal signal in the course of the sinus. Lateral sinus thrombosis was indicated by increased intraluminal signal on all planes and with all pulse sequences. By virtue of its freedom from bone-related artifact, its multiplanar imaging capability, and its sensitivity to both blood flow and thrombus formation, magnetic resonance imaging is an excellent tool for the evaluation of lateral sinus thrombosis or hypoplasia. (Stroke 1990;21:1350-1356) D ural sinus thrombosis is often difficult to diagnose because it presents with a wide spectrum of nonspecific clinical manifestations such as intracranial hypertension, focal signs, altered level of consciousness, and mental disturbances.12 One of the most frequent patterns, particularly when sinus thrombosis does not extend to the cerebral veins, is "benign intracranial hypertension," of which cerebral venous thrombosis is a wellestablished cause.Lateral sinus thrombosis is almost as common as superior sagittal sinus thrombosis, 1 but diagnosis of the former is more difficult. First, direct signs of lateral sinus thrombosis are difficult to assess on computed tomograms (CT scans). "5 Second, it may be impossible at angiography to differentiate lateral sinus thrombosis from congenital hypoplasia, which mainly affects its proximal transverse portion. Recent reports 6 -13 have stressed the usefulness of magnetic resonance imaging (MRI) in the diagnosis of dural sinus thrombosis. We describe MRI findings in patients with lateral sinus hypoplasia or lateral sinus thrombosis that illustrate the value of MRI in differ- Received March 14, 1990; accepted May 15, 1990. entiating these two conditions, a distinction that can have major therapeutic implications. Subjects and MethodsInclusion in this study was based solely on one criterion: the nonvisualization or poor visualization of one or both lateral sinuses at angiography. Each patient had selective opacification of both carotid arteries and a good-quality posterior fossa angiogram with opacification of the dominant vertebral artery or both vertebral arteries. In all patients MRI was performed on a Philips 0.5-T Gyroscan S15 (Eindhoven, The Netherlands) with a standard 30-cm head coil, using Tl-and T2-weighted spin-echo pulse sequences with a slice width of 8 mm and an interslice spacing of 0.8 mm. Parameters for Tl-weighted imaging included a repetition time (TR) of 450 msec and an echo time (TE) of 14 msec, whereas T2-weig...