Impact of aphasia on verbal memory: insights from the Selective Anesthesia for Functional Evaluation.

Objective: The Wada test, an intracarotid amobarbital procedure, is the gold standard for the preoperative evaluation of postoperative memory decline in patients with drug-resistant epilepsy. However, because the anesthetic perfuses the language areas, infusion into the language-dominant hemisphere also causes aphasia. Although it has been noted that aphasia may affect verbal memory scores in the Wada test, the extent of the effects has not been verified. This study aimed to examine the extent to which verbal memory is affected by aphasia induced by selective anesthetization of the proximal middle cerebral artery (M1) area (which does not affect the hippocampus) in patients with drug-resistant epilepsy.

Methods: This single-center retrospective analysis included patients with drug-resistant epilepsy who underwent the Selective Anesthesia for Functional Evaluation (SAFE) as a preoperative evaluation between May 2018 and December 2023. SAFE includes cognitive tasks of equal difficulty administered before and immediately after anesthetic infusion. The preinfusion assessments confirmed the baseline, whereas the postinfusion assessments evaluated brain function under selective anesthesia. The patients were required to memorize 8 words and 8 pictures. After the anesthetic effects subsided, recognition tasks were performed for both words and pictures. The number of words and pictures successfully recognized were counted as verbal and visual memory scores, respectively. Memory scores were compared between the left and right M1 infusions.

Results: Of the 180 SAFE trials conducted during the study period, 38 trials (25 left and 13 right M1 infusions) in 29 patients with confirmed left language dominance and propofol infusion into the M1 were included. All left M1 infusions induced aphasia without causing significant disturbances in consciousness that could have interfered with task completion. The results indicated significantly lower verbal memory scores during left M1 infusions than during right M1 infusions, whereas the visual memory scores were comparable.

Conclusions: The verbal memory score was significantly lower after infusion into the M1 of the language-dominant hemisphere than into the nondominant hemisphere, suggesting that aphasia during stimulus encoding may impair verbal memory. Thus, the Wada test with intracarotid anesthetic infusion may not accurately assess memory function due to aphasia. This study highlights the utility of SAFE in reducing aphasic interference in verbal memory, leading to a more accurate evaluation of surgical candidacy in patients with epilepsy.