PREDICTORS OF NEUROPSYCHOLOGICAL CHANGE FOLLOWING ANTERIOR TEMPORAL LOBECTOMY - ROLE OF REGRESSION TOWARD THE MEAN

Citation
Bp. Hermann et al., PREDICTORS OF NEUROPSYCHOLOGICAL CHANGE FOLLOWING ANTERIOR TEMPORAL LOBECTOMY - ROLE OF REGRESSION TOWARD THE MEAN, Journal of epilepsy, 4(3), 1991, pp. 139-148
Citations number
32
Journal title
ISSN journal
08966974
Volume
4
Issue
3
Year of publication
1991
Pages
139 - 148
Database
ISI
SICI code
0896-6974(1991)4:3<139:PONCFA>2.0.ZU;2-3
Abstract
The purpose of this investigation was to identify the determinants of changes in visual-spatial, language, and memory function following ant erior temporal lobectomy (ATL). Demographic (age, education, and IQ), epilepsy (age at onset), and surgical variables (laterality of resecti on and surgical outcome) were examined for their ability to predict po stoperative changes in neuropsychological function. Particular interes t was focused on the ability of an additional variable, initial level of performance, to predict postoperative cognitive change. This latter variable reflected the effects of a statistical artifact (regression toward the mean). The predictive ability of these variables was examin ed in three data sets involving over-lapping groups of patients who un derwent dominant or nondominant ATL. Results indicated that initial le vel of performance was the most powerful predictor of postoperative co gnitive change on measures of cortical ability (visual-spatial and lan guage function), whereas laterality of resection was not a significant predictor of postoperative cognitive decline. With regard to postoper ative changes in memory function, initial level of performance remaine d a significant predictor. However, laterality of resection was also a consistent and significant predictor of postoperative impairments in memory ability. These results suggest that regression toward the mean exerts a significant effect on pre- to postoperative neuropsychologica l evaluations and needs to be taken into consideration in research wit h groups of patients as well as when interpreting the results of test findings from individual patients.