The intracarotid amobarbital procedure: when is it worth repeating?

Despite the reported diagnostic value of the intracarotid amobarbital procedure (IAP) or "Wada test" for determining hemispheric lateralization and memory functioning, it has never undergone formal reliability testing because a prospective test-retest study design is neither feasible nor ethical. However, some patients require repeat testing for clinical purposes, a circumstance that allows for exploration of issues related to reliability. The current investigation sought to: (1) evaluate the frequency of and reasons for repeated IAPs and (2) describe the test-retest reliability of repeated IAPs in a large tertiary epilepsy center.

A 10-year review (2001-2011) of the New York University Langone Medical Center Comprehensive Epilepsy Center patient registry revealed 630 IAPs. Review of medical records identified 20 individuals who underwent two or more IAPs on separate days. Because IAPs repeated due to technical problems should be considered separate from IAPs repeated for other reasons because these IAPs likely included a change in the procedure (e.g., lower medication dose) in an attempt to ameliorate the complication, patients were grouped accordingly. Six patients underwent repeated IAPs due to technical complication and 14 patients underwent a repeated IAP due to other reasons (e.g., unexpected memory outcome, reconsideration of surgery years after a previous surgical work-up in which no surgery was performed, and/or consideration of a second surgery). Given that data obtained from injections ipsilateral to a seizure focus are sometimes considered in a manner clinically different from data obtained from injections contralateral to the seizure focus, memory outcome was classified relative to the side of identified seizure focus. The degree to which language and memory data were consistent across repeated IAPs was examined.

Language functioning was consistently lateralized across IAPs in all but one case. Among the six patients who experienced technical problems in the first IAP, three were fully participatory in the second procedure such that valid data were obtained. For the other three, the technical problem recurred with no change in outcome across procedures. Among the 14 patients with repeated IAPs due to other reasons, 79% of the available ipsilateral and 73% of the contralateral pass/fail outcomes were consistent across procedures. No difference between ipsilateral or contralateral injections was observed for the likelihood of a change in results (p = 0.57).

Our data identified overall high reliability for both the ipsilateral and contralateral sides with repeated IAP testing. Results indicated that although patients for whom a correctable technical problem was identified during the IAP may benefit from a repeat study, there is little benefit to repeating the IAP in patients with discordant or unexpected results (i.e., results are not likely to change). These data support the overall reliability of both the language and memory data obtained from the IAP.