The use of neurometrics in the study of patients with cerebral ischaemia.

The 'neurometric' method, as developed by John, was used for the study of patients with one-sided supratentorial ischaemia. The results of a reference group of 64 normal volunteers were compared with those of a group of 94 patients. This patient group included 54 patients with permanent neurological deficit (completed stroke or partial non-progressive stroke) and 40 patients with transient neurological symptoms (reversible ischaemic neurological deficit or transient ischaemic attacks). In 90% of all patients the neurometric method indicated that the EEG had to be considered as abnormal (compared with 3% of false positives in the reference group). A significant overall asymmetry was found in 82% of all patients compared with 2% in the reference group. During the follow-up period of 3 months there was a significant improvement of the EEG (as indicated by 3 neurometric parameters) as well as an improvement in clinical condition. In 57.5% of the EEGs, both the neurometric score and the visual assessment of the EEG indicated that some abnormality was present. In 6.6% of the cases both methods failed to indicate any abnormality. In 34.0% the neurometrics method revealed an abnormality in the EEG which was not found by visual assessment. In only 1.9% of the patients did visual assessment suggest abnormalities which could not be confirmed by neurometrics. In the group of EEGs which were considered as normal by visual assessment, the neurometrics method revealed abnormalities in 84%. The neurometric method is at the moment the most sensitive approach for detecting abnormalities in the EEG of patients with unilateral cerebral ischaemia. However, the method seems not very accurate for lateralization or localization of the lesions.