EEG RECORDINGS IN THE COURSE OF RECOVERY FROM STROKE

Background and Purpose This study was designed to assess the value of quantified electroencephalography (EEG) in the follow-up of stroke and its possible correlation with other measures of recovery. Methods EEG s were recorded over a period of 6 months in 34 patients with ischemic stroke in the middle cerebral artery territory who underwent spectral analysis. Two groups were formed: (1) 28 patients with a Barthel Inde x score of less than 60 at admission and (2) 6 patients with a Barthel score of over 60 and therefore with a much better prognosis. Ten reco rdings were made in each patient in the first group and at least three in the second group. Motor functions and activities of daily living ( ADL) were assessed three times, on admission to the study and after 3 and 6 months. Results Quantified EEG was significantly abnormal in the affected hemisphere of the first group. Side and site were not releva nt. Longitudinal recordings showed a significant improvement of the po wer spectrum in the first 3 months. Alpha mean weighted frequency of t he injured hemisphere was always slower than that of the contralateral side. All patients improved in motor performance and ADL, with the gr eatest gain in the first 3 months. However, no correlation between qua ntified EEG and clinical testing was found. By looking at individual p rofiles, a frequent and unpredictable instability was seen in the ''un affected'' hemisphere. In those who made the best recovery, the EEG sp ectrum became more symmetrical over the left and right hemispheres, an aspect that continued beyond the point of a good motor and ADL recove ry. Conclusions Quantified EEG undergoes early and subtle changes in t he follow-up of stroke that can outlast clinical recovery. Routine eva luation for prognosis is not recommended because motor and ADL scales provide earlier and more accurate indicators. In contrast, quantified EEG can be useful for individual patients undergoing rehabilitation to monitor mechanisms of local repair and also to detect changes in the so-called normal hemisphere. It may reveal focal abnormalities that ar e undetected by coarse clinical evaluation and that call for appropria te neuropsychological testing.