Abstracts

Most versions of the language and memory component of the intra-carotid amytal (ICA) test employ a cutoff score denoting the point at which test failure is judged to occur. There may be patients with low enough baseline cognitive functioning who might fail such tests under any circumstances. For such patients the ICA procedure for testing language and memory functioning would be invalid. This study attempted to identify such patients and the variables that would predict them. The study examined a group of adult candidates (n=54) for epilepsy surgery. Three equivalent versions of the language and memory portion of the ICA test were developed. The version of the language and memory test used during the pre-injection, baseline phase was randomly selected. The study analyzed the relationship of clinical variables and neuropsychological test variables to performance on the practice version of the test. Average scores on the practice version of the test were uniformly quite high (mean language score=14.93, mean memory score=9.81). No subjects scored below the cutoff for failure. No relationship was found between clinical variables and performance on the baseline test. A moderate correlation was found between the baseline memory test and one pre-surgical neuropsychological test of memory. Otherwise, scores on the baseline test were unrelated to any of the neuropsychological test scores. In some cases, scores on the pre-surgical neuropsychological tests were quite low (WAIS-III full scale IQ=50). In none of these cases were there failures on the baseline test. No relationship was found between scores on the baseline test and the ability of patients to cooperate with the actual ICA procedure. Thirty-one subjects had surgery for seizure control. Seventy per cent of these had good surgical outcomes. Several had post-surgical subjective complaints of language or memory problems. None had amnestic disorder. No relationship was found between scores on the baseline test and surgical outcome or post-surgical subjective complaints. The validity of the ICA test depends on the use of test procedures that any patient can perform at baseline. The results of this study indicate that procedures can be designed whose validity is resistant to the effects of extremely low pre-surgical cognitive functioning. With ICA test procedures where this presumption may not automatically be present, the validity of test participation can be confirmed with an equivalent baseline version of the test prior to the actual clinical procedure.