Memory assessment during the intracarotid amobarbital procedure: influence of injection order.

Abstract

To analyze, in patients with unilateral regional epilepsy, the influence of intracarotid amobarbital injection order on the level of recognition memory, with both injections performed within approximately half an hour, when the order was varied by a first injection into either the epileptogenic or the nonepileptogenic hemisphere. Of a series of 74 consecutive intracarotid amobarbital procedures (IAP), 61 valid bilateral tests were extracted for analysis. The protocol was changed from a congruent sequence (epileptogenic hemisphere injected first; n = 32) to an incongruent sequence (nonepileptogenic hemisphere injected first; n = 29). Memory was quantified by the percentage of presented items recognized before the recovery of full motor power (corrected for guessing). The congruent group showed a poorer recognition percentage after injection into the nonepileptogenic hemisphere than after injection into the epileptogenic hemisphere. The incongruent group had a similar recognition percentage after both injections. This was especially apparent in patients with a right epileptogenic hemisphere. The commonly used congruent injection order leads to a poorer recognition by the epileptogenic hemisphere than the nonepileptogenic hemisphere. This relation cannot be found with an incongruent sequence. The results raise doubts about the internal validity of the IAP in describing isolated hemispheric memory functions.