The Bispectral Index (BIS) has become widely accepted as a measurement of hypnosis under anesthesia. It is derived from a processed electroencephalogram and computer algorithm that assigns a numerical value based on the probability of consciousness. Values <60 correlate with a low probability of consciousness (1). We report a case wherein a patient experienced explicit recall of intraoperative events, despite a BIS of 47.
A 28-yr-old, 85-kg male with a history of smoking, hypercholesterolemia, and hypertension presented to the operating room for three coronary artery bypass grafts, secondary to severe coronary artery disease with normal left ventricular function. There was no family history significant for heart disease. The patient was employed as a corrections officer and had no previous medical training or experience with cardiac surgery. He had no history of recreational drug use or alcohol abuse. His surgical history was previously negative. Monitors applied included an Aspect A-2000 BIS monitor (Aspect Medical Systems, Newton, MA). Anesthesia was induced with thiopental, 4 mg/kg, nitrous oxide, 70%, in oxygen and sevoflurane, 2–3%, with paralysis induced by vecuronium. After endotracheal intubation, he was turned laterally, and preservative-free morphine, 5 μg/kg, administered via a 22-gauge spinal needle at the L3-4 interspace. The patient was then returned to a supine position for the duration of surgery. During placement of the intrathecal morphine, the BIS sensor had become dislodged and was replaced after returning to the supine position. As the monitor was turned on, electrode impedances were automatically checked and within acceptable limits. Anesthesia was maintained with sevoflurane and nitrous oxide until the start of cardiopulmonary bypass (CPB), when it was changed to isoflurane in oxygen. At the conclusion of the case, the neuromuscular block from vecuronium was reversed, and the patient was awakened, tracheally extubated, and transferred to the cardiovascular intensive care unit in stable condition. He had an uneventful recovery and was discharged home on the fourth postoperative day.
During his postanesthesia interview, 48 h after surgery, the patient reported that he could recall voices in the operating room, the sound of the sternal saw being tested, and could feel his chest being opened. The patient denied any pain, and did not recall anything after sternotomy. He clearly described the sound of the nitrogen-powered sternal saw to the anesthesiology resident conducting the postoperative evaluation. This was elicited spontaneously by the patient and not as a result of a specific inquiry about recall. The patient had no prior history of intraoperative recall. Review of the medical record revealed a BIS of 47 at the time of sternotomy, 35 min after induction of anesthesia. At the time of sternotomy, Fio2 was 0.26, Fin2o was 0.67, and inspired sevoflurane was 2%. The sevoflurane initially was started at 4%, and had been at 3% for most of the time before sternotomy. The patient’s heart rate and blood pressure were slowly decreasing after the induction, and the sevoflurane concentration was also slowly decreased to 2%, as noted. The vaporizer in use on the anesthesia machine had previously functioned well, and end-tidal sevoflurane was appropriately detected by a RASCAL gas-monitoring device (Ohmeda, Tewksbury, MA). Despite this, the patient was quite satisfied with his anesthetic and reported no psychological difficulties resulting from the awareness event.
Awareness during anesthesia is a significant complication with potentially devastating psychological consequences. Up to 70% of patients who suffer intraoperative awareness may experience after-effects, similar to a postraumatic stress disorder (2). This has become a particularly infamous and emotional issue during the last decade, as several cases became publicized through television and radio talk shows. Awareness during cardiac surgery has been reported in as many as 23% of patients (3). Most cases occur during or after rewarming from CPB and are attributed to alterations in drug pharmacokinetics and dynamics resulting from the bypass process. Other cases are associated with painful procedures such as sternotomy and intubation (4). This case is particularly unusual, as the patient had not yet undergone sternotomy and was stable with only a minimal level of surgical stimulation. Additionally, the patient’s anesthetic essentially consisted of nitrous oxide and volatile anesthetic at a total concentration of more than 1 MAC, which had been maintained for at least 30 minutes before the recall occurred. Volatile anesthetics are especially protective of intraoperative awareness and have a well-documented amnestic property at concentrations <1 MAC (5).
The patient in this case was administered an anesthetic that we frequently use if early or immediate postoperative extubation is desired. No previous instances of recall have been discovered in any other patient using the same anesthetic either before or since (approximately 40 patients), although this small sample size lacks the statistical power to draw conclusions about a relatively uncommon event. Indeed, most patients are significantly sedated on emergence, despite the avoidance of systemic narcotics or benzodiazepines. Although the etiology for this sedation is unclear, we speculate that it may be related to cerebral changes secondary to CPB. The first several patients who received this anesthetic had intrathecal morphine administered at a dose of 10 μg/kg. This was decreased to 5 μg/kg as a result of excessive sedation associated with the larger dose. It should be noted that this is the youngest patient in our series and it is unclear if his age may have influenced the propensity for awareness.
Because of our avoidance of any additional narcotics or benzodiazepines, we depended heavily on the BIS as a monitor of adequate hypnosis. The BIS frequently increases during rewarming from CPB and isoflurane is administered via the oxygenator for much of the pump run. The monitor has performed well under the conditions encountered during cardiac surgery and the BIS usually trends as expected, increasing to more than 95 as the patients are awakened and tracheally extubated at the conclusion of the case. Although cardiac surgery and, specifically, CPB cause alterations in cerebral blood flow, the resultant electroencephalographic changes should be accurately interpreted by the BIS, but this has not been studied.
In conclusion, we report a case of explicit recall at a BIS of 47 in a patient during cardiac surgery using a significant dose of volatile anesthesia. Consciousness and memory formation during anesthesia are not completely understood phenomena. It is apparent that there may be an incidence of intraoperative awareness that persists despite the best preventative efforts. Additionally, although the BIS is a valuable tool in the monitoring of hypnosis, it is a probability function, and it can be expected that recall may occur despite a relatively low displayed value. Although the chance of recall at a BIS of <60 may be small, it is obviously not guaranteed, consistent with a statistical outlier in a population. Given a sufficiently large number of cases, this phenomenon is not unexpected. Fortunately, our patient suffered no adverse sequelae. In the preanesthetic evaluation, it is our practice to discuss the possibility of recall with all patients, and again discuss any possible awareness during the postanesthesia follow-up. It is unknown whether these discussions may alleviate the psychological impact of intraoperative awareness.
1. Glass PS, Bloom M, Kearse L, et al. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane and alfentanil in healthy volunteers. Anesthesiology 1997; 86: 836–47.
2. Moerman N, Bonke B, Oosting J. Awareness and recall during general anesthesia. Anesthesiology 1993; 79: 454–64.
3. Ghoneim MM, Block RI. Learning and consciousness during general anesthesia. Anesthesiology 1992; 76: 279–305.
4. Tempe DK, Siddiquie RA. Awareness during cardiac surgery. J Cardiothorac Vasc Anesth 1999; 13: 214–19.
5. Quasha AL, Eger EI II, Tinker JH. Determination and applications of MAC. Anesthesiology 1980; 53: 315–34.