Validation of a neuropsychological Wada procedure

by Eisenman, Daniel David.

The Intracarotid amobarbital procedure (IAP) is considered an essential part of the presurgical neurodiagnostic evaluation in most epilepsy centers throughout the country. Despite the IAP’s mainstream use, there is great variability in how the test is being performed across centers. The main purpose of this study was to validate a standardized IAP memory measure and explore its value in predicting lateralization as well as treatment outcome. The IAP memory measure had good overall reliability for all three Forms (Cronbach’s Alpha = .85 vi for Form I, .83 for Form II, and .69 for Form III). The majority of items on Form I and II had acceptable item difficulty values, item discrimination values, and item-total correlations. There were a few items on Forms I and II that may be candidates for revision, but most items were only slightly below predicted ranges for what defines a “good” measure, and the majority of items contributed to the reliability of the test. In terms of construct validity, correlations with other memory tests provided some evidence of convergent validity for the IAP memory measure, but were generally low. In terms of divergent validity, both Forms I and II had low or no correlations with the executive functioning measures, providing preliminary support for the construct validity of the measure. When utilizing a discrepancy score to predict lateralization in subjects with temporal lobe epilepsy, asymmetry scores from the IAP memory measure were able to classify 92% of subjects with either left or right TLE after applying a correction factor for left injection scores. There was limited data regarding post-operative seizure outcome. However, seizure-free subjects had a higher percentage of DS greater than 20% than those subjects who reported at least one seizure postoperatively. In addition, there was a significant difference on Form I of the IAP memory measure between subjects who reported they were seizure free following temporal resection as compared to subjects who continued to report symptoms. Overall, the IAP memory measure demonstrated good psychometric vii properties and this study represents one of the most thorough analyses of the IAP memory test to date. viii