Search

Cortical stimulation of the supramarginal and angular gyri elicited dysphasia. Delays in verbal response, misnaming, and difficulties in speech production were also elicited from this region. In two patients speech arrest occurred when stimulation was extended into the occipital cortex and the cortex medial to the supramarginal gyrus, respectively. In a left-handed patient, with speech representation presumed to be in the right hemisphere, neither stimulation nor ablation of the angular gyrus resulted in dysphasia. After anterior temporal resections two patients exhibited dysphasia in association with cortical edema. In one of these an additional resection in the inferior temporo-occipital region produced a marked but transient dyslexia. Observations of the disintegration of speech function during stimulation suggest that such stimulation interferes with a search mechanism by which the nonverbal concept of a visual stimulus is linked to a specific word in memory that is then withdrawn for use. A review of autopsy specimens demonstrates how close the cortex bearing indispensable speech representation lies to the occipital pole and the parietal midline. Because the gyral pattern is obscured by the meninges and is subject to anomalies, elective resection in these areas should be preceded by cortical stimulation and functional mapping.

Although the earlier supposition was that the n. dorsalils superficialis (n. lateralis dorsalis) of the thalamus projected to the parietal region, more recent evidence has linked it to the posterior cingulate gyrus and possibly adjacent regions near the splenium of the corpus callosum. An afferent supply from lower levels was in more doubt, although some report had been made of cell and fibre degeneration in the n. dorsalis superficialis after extensive temporal resections and section of the fornix in lower primates. The five human hemispheres of the present study all had lesions of long duration below the level of the splenium of the corpus callosum in the posteromedial temporal region. All showed marked degeneration in the fornix and n. dorsalis superficialis. In favourably stained cases, gliotic fascicles could be followed from the descending column of the fornix to the n. dorsalis superficialis via the region lateral to the stria medullaris thalami. The cell loss in the nucleus thus appeared to be an instance of anterograde transynptic degeneration. These cases provided an interesting instance in which human infarctions provided natural lesions that would have been hard to duplicate in experimental animals.

The authors describe the use of temporal lobectomy following careful and repeated electroencephalogram (EEG) evaluation (with implanted electrodes in otherwise unresolvable cases) in the epileptic group characterized by automatisms (psychomotor seizures) with temporal epileptiform activity complicated by EEG foci in the opposite temporal lobe or by extratemporal activity. They found that this can render a significant number of patients (between 25% and 50%) either seizure-free or with significant and useful reduction in their seizure frequency. The cure and improvement rates of cases followed up after temporal resection with or without prior study with implanted electrodes were approximately equal. However, the implanted electrodes permitted surgical treatment of certain cases which would have been rejected on the basis of evidence derived from the scalp recordings alone. Of 28 of these 34 patients with persisting EEG epileptiform activity in the postoperative period, only one had such activity in a different location in a follow-up period of 6 years. No evidence of spreading epileptic activity or appearance of "mirror foci" was seen during a follow-up period averaging 8.2 years. Seizure remission up to 15 years with eventual recurrence of the original seizure type may occur following surgical therapy. Follow-up studies of surgical epileptic treatment of less than 3 to 5 years are of doubtful value.