Six patients with transient or permanent visual loss associated with migraine are presented. In 3 patients with monocular and one patient with binocular episodes of transient visual loss subsequent visual episodes were relieved by prompt inhalation of isoproterenol. The authors review the possible mechanisms of action of isoproterenol in migraine and present evidence to support the prophylactic use of isoproterenol to prevent transient and possible persistent visual loss in patients with migraine.

In the two years since it has become available, computer assisted tomography of the head (CAT) has proven to be a revolutionary tool for the diagnosis of cerebral neurologic diseases. It has expedited diagnoses and has actually reduced the cost of patient investigation. Nowhere has CAT been a more accurate, convenient and reliable diagnostic tool than in the investigation of intracerebral hemorrhage. CAT has also been of considerable value in the evaluation of the head trauma patient. In this group of patients, while the results have been excellent, they have not been quite as reliable as in the diagnosis of intracerebral hematomas.

For 80% of 4,748 patients with cerebrovascular insufficiency, complete four-vessel examinations were made. Although the primary purpose was localization of surgically accessible extracranial lesions, intracranial studies were also obtained in a high percentage of cases. The overall grave complication rate was 1.2%. A test of interinstitutional comparability of x-ray interpretation revealed greatest agreement when lesions compromised more than 50% of the arterial lumen.

The diagnosis of acoustic neuroma usually does not present a major problem to the radiologist. The bony changes produced in the region of the internal acoustic meatus are well known and can usually be delineated in films of the petrous bone, either in the standard projections or with the aid of tomography (1, 3). Further information may be gained by fractional pneumoencephalography, which also helps to increase the number of cases in which a positive diagnosis can be made with certainty (6). There still remains, however, a small but significant group of patients in whom the above investigations fail to provide a clear-cut answer as to the presence or absence of a tumor, in most instances because of its small size. This report describes an additional method of examining the region of the internal acoustic meatus. It has been used in effective delineation of small acoustic neuromas and so thoroughly visualizes the contents of the cerebellopontine cistern and acoustic meatus that a negative study effectively excludes the presence of a mass. Method The examination is carried out on the Massiot-Philips tomographic table, using the hypocycloid movement. A spinal tap is performed in the usual way, and the patient is then placed prone on the table with a small support under each shoulder to allow for full flexion of the head. The head is flexed until the canthomeatal line is perpendicular to the table-top, with the forehead resting on the table, and then rotated 15–20 degrees toward the side to be examined in such a way that the corresponding cerebellopontine angle cistern is slightly elevated (Fig. 1). The optimal degree of rotation varies; in general we have found that the shorter the internal auditory canal, the less the required degree of rotation. Once the head is positioned correctly it must be fixed by a clamp or some other means so that the position remains the same throughout the study. The tomographic table is tilted head-up to bring the posterior fossa to a slightly higher level than the cervical and upper thoracic spine. Tomographic scout films are taken to establish the plane of the internal auditory canal. Seven milliliters of air is then injected into the subarachnoid space. The air will rise into the cerebellopontine cistern on the side being examined and will remain trapped there. A minute or two after the injection, tomographic films are taken to cover a depth of approximately 10 mm on either side of the internal auditory meatus. If the patient is correctly positioned the cerebellopontine angle cistern, internal auditory meatus and canal, and the lateral aspect of the belly of the pons should be clearly demonstrated (Fig. 2).

Advances in the application of computer technology to medicine have been proceeding at an ever-increasing rate (2–6, 12–18, 21). This report deals with a technic that the authors have recently applied to radiology. The method combines routine dictation and typing of patients' radiological reports utilizing a special, commercially available typewriter. This typewriter will simultaneously prepare the material in the form of hard copy for routine use, and punch paper tape which is in machine language for computer input, storage, and analysis. The procedure was adapted from a previously described program of computer processing of neurological and neurosurgical case summaries (10, 11). The technic is now being routinely applied to neuroradiology reports of special procedures such as pneumoencephalography, ventriculography, angiography, and myelography performed at University Hospital and Bellevue Hospital of the New York University Medical Center. The purpose of adopting this process is to take full advantage ...

In the course of reviewing the angiographic changes in suprasellar tumors the authors found several extradural lesions that seemed to be adequately demonstrated by cerebral angiography (ref. 1). It was also noted that those tumors without intradural components (with the exception oflarge chordomas) cause few, if any, changes in the encephalogram. It was, therefore, decided to review a series of epidural tumors and other lesions to see the frequency and the accuracy with which cerebral angiography is capable of demonstrating them.

It is a generally accepted fact that stereoscopic films may be helpful in the evaluation of complex anatomical details. There is also little doubt that stereoscopic angiographic studies may be of assistance particularly when investigating intracranial arterial aneurysms. Many techniques have been devised for obtaining stereoscopic angiographic films. These have usually utilized either two separate injections with a tube shift between injections, or they have involved a mechanism for shifting the roentgen tube between exposures, or involved double anode roentgen tubes. These methods have usually been rather complicated or the apparatus has been much too expensive for general use. Routine cerebral angiography at the New York Neurological Institute is done in simultaneous biplane with half a second intervals between exposures. The ideal stereoscopic system would enable any two successive films in either plane to be viewed stereoscopically. This may be accomplished by moving the a. p. and lateral tubes 10 em between exposures (using a 100 em anode film distance and 2 to 4 em distance from head of subject to film). A similar stereo-