THERE are three fundamental goals of anesthesia: unconsciousness, amnesia, and immobility. In years past, most people would have assumed that anesthetics act in the brain to produce all three of these goals. In the past decade, data have emerged indicating that immobility is likely produced by anesthetic action in the spinal cord, prompting a reexamination of “macroscopic” sites of anesthetic action.1,2With regard to amnesia, the hippocampus is certainly involved in declarative memory, and hippocampal lesions can result in profound amnesia. However, the role of other brain areas, such as amygdala and entorhinal cortex, in memory formation is complex and apparently dependent on subtle aspects of the memory task.3What is the role, if any, of these structures in anesthetic-induced amnesia? The report by Alkire and Nathan4in this issue of Anesthesiology provides interesting data supporting the amygdala as a site at which inhaled anesthetics exert an amnestic effect on fear conditioning, one form of memory.

The amygdala is strongly implicated in learning under emotionally charged settings such as fear. An inhibitory avoidance paradigm was used by Alkire and Nathan in which rats were placed in a lighted chamber facing a dark tunnel, which they normally prefer to enter. However, entrance into the dark area was negatively reinforced by electrical shock; rats quickly learned to avoid entering the dark tunnel and remained in the nonpreferred but “safe” lighted environment. When retested the next day, the rats continued to avoid entering the dark tunnel; their memory retention latency (i.e. , time to enter the dark tunnel) was very long, indicating that they remember being shocked. If a low concentration of sevoflurane was administered during the initial training period, the animals quickly entered the dark tunnel the following day, i.e. , sevoflurane prevented avoidance learning. However, after bilateral lesion of the basolateral amygdala, rats exhibited equally long memory retention latencies regardless of whether low-dose sevoflurane was given. These data suggest that, at least for this type of learning, the basolateral amygdala is not the critical site for memory storage and that sevoflurane (and propfol5and diazepam6) acts in that structure to block this learning. It should be noted that the absence of amnesia in the amygdala-lesioned animals reported by Alkire and Nathan is at odds with the findings of other groups.7–9In any event, the data imply that in the intact animal, sevoflurane activates pathways in the basolateral amygdala that exert an inhibitory effect on avoidance learning at sites outside the lesioned area of the basolateral amygdala.

The data of diazepam, propofol, and sevoflurane with respect to inhibitory avoidance are remarkably similar. Figure 2 in Tomaz et al. ,6figure 4 of Alkire et al. ,5and figure 3 of Alkire and Nathan,4aside from slight differences in control latency, are virtually superimposable. Because diazepam and propofol act almost exclusively at the γ-aminobutyric acid receptor type A (GABAA) receptor and sevoflurane also enhances GABAAreceptor function, it seems logical to conclude that sevoflurane produces its effect on the amygdala at the GABAAreceptor. Ketamine, however, has virtually no action at the GABAAreceptor, but it produces amnesia in a dose range equipotent to that of volatile anesthetics.10Therefore, it would be folly to assume that, for amnesia, all anesthetics must act at the amygdala and/or via a GABAAreceptor effect. Furthermore, the nonimmobilizer 1,2-dichlorohexafluorocyclobutane abolishes fear conditioning, but low-dose isoflurane reverses, not potentiates, this action,11underscoring the complex nature of memory and anesthetic-induced amnesia. Clearly, more work is needed to identify the molecular and cellular mechanisms by which memory formation is prevented by inhaled and intravenous anesthetic agents.

The data of Alkire and Nathan indicate that the basolateral amygdala is quite sensitive to sevoflurane. The control animals in their study lost memory (or never learned) when the sevoflurane concentration was 0.3%—only 0.15 minimum alveolar concentration (MAC). It is unclear whether other structures associated with memory formation (e.g. , hippocampus) are more or less sensitive. Anesthetic effects on fear to context and tone have been extensively studied. Interestingly, fear to context has anesthetic sensitivity comparable to that of inhibitory avoidance, but ablation of fear to tone requires twice as much anesthesia.12Nonetheless, the results emphasize the importance of investigating the sites of anesthetic-induced amnesia. It is unknown, however, whether anesthetic action at one single anatomical site in humans prevents all intraoperative memories (implicit and explicit). If an anesthetic could be developed that had significant specificity at memory-formation sites, we would feel much better about giving a sub-MAC concentration of anesthesia. It is fortuitous that the anesthetic concentration needed to prevent memory is well below that needed to prevent movement. This fact, combined with the relative steepness of the population dose– response curves for amnesia and immobility, gives the anesthesiologist (and the patient) some reassurance that memories will be ablated even at anesthetic concentrations that otherwise would not prevent movement.

Memories differ in their emotional content. If you are a surgical patient, remembering what you had for dinner the night before surgery does not carry the emotional content that an intraoperative memory might have. Hence, the amygdala has a potential key role in ablation of those memories that we do not want patients to have. Patients who report intraoperative awareness sometimes do not describe these memories as distressful. Is this because the anesthetic sensitivity of the amygdala blocks the emotional aspects of the experience but not the experience itself? Nonetheless, memory during anesthesia and surgery can be distressing and is associated with posttraumatic stress disorder. Intraoperative awareness has received much recent attention and has prompted the recent sentinel event issued by the Joint Commission on Health Care Organizations.†Reports in the lay press have increased public awareness of this issue. Therefore, increased funding of research into anesthetic mechanisms and subsequent development of newer and safer anesthetics are likely to gain wider support, especially among the estimated 20,000–40,000 Americans who experience intraoperative awareness every year.

We began by stating the basic goals of anesthesia, but some have argued that amnesia, along with immobility, is all that is needed13; intraoperative awareness is immaterial if it cannot be remembered. This approach, at the very least, minimizes the role of implicit memories. In any case, as anesthesiologists, we would like for our patients to have a pleasant perioperative experience. Is it any wonder that we are intensely interested in how anesthetics affect brain sites associated with unpleasant events and memories? If we could affect those areas, would patients otherwise have any problems with their perioperative experience, including the possibility of intraoperative awareness? After all, if we can ask our patients if their experience was pleasant, and they always say yes, what more could we ask for?

* University of California, Davis, California.

Antognini JF, Schwartz K: Exaggerated anesthetic requirements in the preferentially anesthetized brain. Anesthesiology 1993; 79:1244–9
Rampil IJ, Mason P, Singh H: Anesthetic potency (MAC) is independent of forebrain structures in the rat. Anesthesiology 1993; 78:707–12
McDonald RJ, White NM: A triple dissociation of memory systems: Hippocampus, amygdala, and dorsal striatum. Behav Neurosci 1993; 107:3–22
Alkire MT, Nathan SV: Does the amygdala mediate anesthetic-induced amnesia? Basolateral amygdala lesions block sevoflurane-induced amnesia. Anesthesiology 2005; 102:754–60
Alkire MT, Vazdarjanova A, Dickinson-Anson H, White NS, Cahill L: Lesions of the basolateral amygdala complex block propofol-induced amnesia for inhibitory avoidance learning in rats. Anesthesiology 2001; 95:708–15
Tomaz C, Dickinson-Anson H, McGaugh JL: Basolateral amygdala lesions block diazepam-induced anterograde amnesia in an inhibitory avoidance task. Proc Natl Acad Sci U S A 1992; 89:3615–9
Fanselow MS, LeDoux JE: Why we think plasticity underlying Pavlovian fear conditioning occurs in the basolateral amygdala. Neuron 1999; 23:229–32
Vazdarjanova A: Does the basolateral amygdala store memories for emotional events? Trends Neurosci 2000; 23:345–6
Maren S: Reply to Vazdarjanova. Trends Neurosci 2000; 23:345–6
Hetem LA, Danion JM, Diemunsch P, Brandt C: Effect of a subanesthetic dose of ketamine on memory and conscious awareness in healthy volunteers. Psychopharmacology (Berl) 2000; 152:283–8
Eger EI, Xing Y, Pearce R, Shafer S, Laster MJ, Zhang Y, Fanselow MS, Sonner JM: Isoflurane antagonizes the capacity of flurothyl or 1,2-dichlorohexafluorocyclobutane to impair fear conditioning to context and tone. Anesth. Analg 2003; 96:1010–8
Dutton RC, Maurer AJ, Sonner JM, Fanselow MS, Laster MJ, Eger EI: The concentration of isoflurane required to suppress learning depends on the type of learning. Anesthesiology 2001; 94:514–9
Eger EI: Anesthesia produced by inhaled anesthetics is an immediately reversible state mediated by the central nervous system that consists of amnesia and immobility in the face of noxious stimulation, Molecular and Basic Mechanisms of Anesthesia. Edited by Urban BW, Barann M. Berlin, Pabst, 2004, pp 530–1Urban BW, Barann M