Transient Global Amnesia

To the Editor: I read with interest the case report by Koht and Moss [1]. It contradicts all the literature on the benzodiazepines and memory in that it suggests that midazolam may cause retrograde amnesia and may impair autobiographical memory; flumazenil may reverse the retrograde but not the anterograde amnesia. Benzodiazepines impair acquisition or encoding of new information but have no effect on either retention or retrieval of previously stored information. As such, they produce anterograde amnesia (i.e., reduced learning of new information after treatment), but not retrograde amnesia, (i.e., loss of information learned before treatment) [2]. There are no drugs that can produce retrograde amnesia and amnesia for autobiographical memories. Flumazenil adequately reverses benzodiazepine-induced anterograde amnesia [3]. A plausible explanation for the patient's condition is his development of transient global amnesia (TGA). TGA is a distinct neurological disorder that was described in 1956 [4,5]. The patient with TGA has sudden onset of severe memory impairment, including both anterograde and retrograde amnesia, which lasts 2-12 h. Clinical examination during TGA demonstrates a relatively isolated amnesic syndrome with an otherwise normal neurologic examination. TGA generally occurs in persons >50 yr and resolves spontaneously after several hours. Approximately one third of TGA attacks are precipitated by physical or psychological stress. It has been reported after general anesthesia [6]. After the attacks, patients remain unable to recall the period of TGA, and they occasionally exhibit a period of permanent retrograde amnesia before the onset of TGA. Kritchevsky et al. [7] recently studied 11 patients with TGA. During the episode, the patients had severe anterograde amnesia for verbal and nonverbal material and retrograde amnesia that typically covered at least two decades. The similarities between the case report by Koht et al. [1] and the characteristics of TGA are obvious-namely the age, the sudden onset of profound amnesia, both anterograde and retrograde, and the complete mental recovery except for the memory of the episode. However, the role of flumazenil in Koht et al. report is not readily apparent. It is possible that the administration of the drug and spontaneous recovery were coincidental or that flumazenil's action at the benzodiazepine receptors may play a role in terminating TGA episodes. A clinical trial of flumazenil in patients with TGA may be warranted. M. M. Ghoneim, MD Department of Anesthesia; University of Iowa; Iowa City, IA 52242-1079