A Retrospective Study of Intraoperative Awareness with Methodological Implications

Mashour, George A. MD, PhD; Wang, Luke Y.-J. MD; Turner, Christopher R. MD, PhD, MBA; Vandervest, John C. BS; Shanks, Amy MS; Tremper, Kevin K. PhD, MD

Section Editor(s): Brull, Sorin J.

Anesthesia & Analgesia:
doi: 10.1213/ane.0b013e3181732b0c
Patient Safety: Research Reports

BACKGROUND: Awareness during general anesthesia is a problem receiving increased attention from physicians and patients. Large multicentered studies have established an accepted incidence of awareness during general anesthesia as approximately 1–2 per 1000 cases or 0.15%. More recent retrospective data, however, suggest that the actual incidence may be as low as 0.0068%.

METHODS: To assess the incidence of awareness at our institution, we conducted a review of adult patients undergoing surgical procedures over a 3-year period. Information on awareness came from entries of “Intraoperative Awareness” captured during our standard evaluations on postoperative day one in our perioperative information system. Patients were not questioned specifically about awareness.

RESULTS: We reviewed 116,478 charts; 65,061 patients received general anesthesia and 51,417 received other types of anesthesia. Of the patients receiving general anesthesia, 44,006 had complete postoperative documentation. The reported incidence of undesired intraoperative awareness in this population was 10/44,006 (1/4401 or 0.023%). Of the patients who received other anesthetic modalities, 22,885 had complete postoperative documentation. Undesired intraoperative awareness was reported in 7/22,885 patients who did not receive general anesthesia (1/3269 or 0.03%). The reported incidence of intraoperative awareness was not statistically different between the two groups (P = 0.54). Relative risk of intraoperative awareness during a general anesthetic compared with a nongeneral anesthetic was 0.74, with 95% confidence interval [0.28, 2.0].

CONCLUSION: Using a retrospective methodology, reports of intraoperative awareness are not statistically different in patients who received general anesthesia compared with those who did not. These results suggest that, despite success with other rare perioperative events, the resolution of retrospective database analyses may be too low to study intraoperative awareness.

In Brief

IMPLICATIONS: This retrospective study found that the incidence of self-reports of intraoperative awareness was not statistically different in patients receiving general anesthesia (0.023%, n = 44,006) compared with those who did not (0.03%, n = 22,885). This finding has methodological implications for the retrospective study of intraoperative awareness.

Author Information

From the Department of Anesthesiology, University of Michigan Medical School, Michigan.

Accepted for publication March 3, 2008.

Supported solely through institutional and departmental funds. The authors have no conflicts of interest to declare.

Address correspondence and reprint requests to 1H247 UH/Box 0048, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0048. Address e-mail to gmashour@umich.edu.

Article Outline

Awareness during general anesthesia, which denotes both awareness and subsequent explicit recall of intraoperative events, is a problem receiving increased attention by both patients and clinicians. A proportion of patients experiencing awareness may subsequently develop serious psychological sequelae, including posttraumatic stress disorder (PTSD).1,2 In 2004, the Joint Commission on Accreditation of Hospital Organizations issued a Sentinel Alert to promote greater attention to the problem.3 Despite recent attention by the medical community and the lay press, the incidence of intraoperative awareness—and hence the magnitude of the problem—remains uncertain. A multicenter study in the United States by Sebel et al. estimated an incidence of awareness with explicit recall of approximately 0.13%, a rate consistent with large European studies demonstrating awareness in 1–2/1000 cases.4,5 In contrast, a recent study of awareness in a regional medical system by Pollard et al.6 reported a much lower incidence of 1 episode of awareness/14,560 cases, or 0.0068%.

Establishing the validity of the retrospective study of awareness is necessary if we are to adjudicate between these disparate reports. Retrospective analyses of data derived from an electronic perioperative information system have been successfully used at our institution to study rare events such as impossible mask ventilation7 and postoperative renal failure.8 To compare the incidence of undesired awareness at our institution with the disparate rates in the literature, as well as explore the use of electronic databases for the study of intraoperative awareness, we conducted a review of more than 100,000 cases over a 3-year period.

Back to Top | Article Outline


With IRB approval, a retrospective electronic chart review was conducted on adult patients receiving anesthesia at the University of Michigan Health System between January 1, 2004 and February 20, 2007. Information regarding awareness was obtained from patient interview on postoperative day 1. Inpatients were interviewed directly by residents. Nurses called outpatients by phone. All patients were asked if they experienced any problems related to anesthesia. If they discussed intraoperative awareness, these data were entered into a perioperative clinical information system (Centricity™ from General Electric Healthcare, Waukesha, WI) by selecting the category of “Intraoperative Awareness” in the postoperative documentation window. Patients were not interviewed using a Brice et al9 or modified Brice interview, i.e., they were not asked specifically about awareness.

The electronic charts were queried for postoperative documentation of “Intraoperative Awareness.” After the initial query, all charts with reports of “Intraoperative Awareness” were reviewed and correlated with existing Quality Assurance data regarding awareness. The intraoperative record was analyzed for anesthetic technique and anesthetic drugs, as well as the use of benzodiazepines, opioids, and neuromuscular blocking drugs. No electroencephalographic devices were used for the detection of intraoperative awareness during this time period examined. Statistical comparison was performed using a χ2 test, as well as assessing relative odds ratios. A P value of <0.05 was considered significant.

Back to Top | Article Outline


We reviewed 116,478 charts between January 1, 2004 and February 20, 2007; 65,061 patients received general anesthesia and 51,417 received other anesthetic modalities. Of the patients receiving general anesthesia, 44,006 had complete postoperative documentation, for a compliance rate of 67%. Of the 44,006 patients who received general anesthesia and had complete postoperative documentation during the time period of study, 10 complained of some degree of awareness (Table 1), an incidence of 1/4401 or 0.023%. Demographic data for this patient cohort are shown in Table 2. Analysis of the general anesthetics for the time period of study indicates that 90% of cases were performed using inhaled anesthetics, whereas 10% used total IV anesthesia (TIVA).

Of the 10 patients who complained of awareness in this group, five were men and five were women. There were no consistent findings regarding anesthetic choice, use of benzodiazepines, or use of opioids. All 10 patients had received neuromuscular blocking drugs at some time. Of the 10 cases of awareness, two patients were in the high-risk category: one was undergoing an emergent cesarean delivery, and the other was undergoing a heart transplant. One patient had a confirmed awareness event during TIVA after the discontinuation of nitrous oxide (patient 10). Figures 1A and B depict the anesthetic regimen for each case. Several cases documented insufficient anesthesia on the electronic record that correlated with complaints of awareness. For patient 1, awareness at the end of the procedure was likely due to low levels of isoflurane documented at minute 210 of the case. Patient 2 had low sevoflurane concentrations and insufficient IV anesthesia at the start of surgery (“S”), when he reported awareness. Patient 4 reported awareness at the beginning of the case, around the time a vaporizer leak was noted in the chart. Finally, patient 10 experienced awareness after nitrous oxide was discontinued and an IV line infiltration was recorded in the record. It was not possible to identify the temporal location of awareness in the remaining cases.

An additional 51,427 patients underwent procedures during the time period of study, but did not receive general anesthesia. Of the patients who received other anesthetic modalities, 22,885 had complete postoperative documentation for a compliance of 45%. Of this cohort, seven patients complained of being aware during the case, an incidence of 1/3269 or 0.03%. These patients had been managed with a variety of anesthetic modalities (Table 3). Of the seven patients complaining of awareness in this population, six were women. In the case of the one male patient, the report of awareness was given by his daughter. This was the only patient in the study who did not report intraoperative awareness independently.

There was no statistically significant difference between the incidence of awareness in the general anesthesia (0.023%) and nongeneral anesthesia (0.03%) populations at our institution (P = 0.54). Relative risk of undesired intraoperative awareness during a general anesthetic compared with a nongeneral anesthetic was 0.74, with 95% confidence interval [0.28, 2.0].

Back to Top | Article Outline


Awareness during general anesthesia is a problem that has captured the attention of clinicians, patients, and the general public. Although awareness is a significant source of fear for many patients undergoing surgery, the actual incidence and sequelae of awareness remain a matter of controversy. This is highlighted by recent studies reporting rates of awareness and subsequent PTSD that were lower than previously thought.2,6

In this retrospective study, we found the incidence of complaints of intraoperative awareness during general anesthesia to be 1/4401 or 0.023%. We acknowledge that these data likely represent an under-estimate of the actual incidence of awareness in the population studied. As Sebel et al.4 noted: “A single short postoperative visit by an anesthesiologist without use of a structured interview is unlikely to elicit many cases of awareness,” an effect also noted by Moerman et al.10 Sandin et al. and Sebel et al. found considerably increased reports of awareness during the second interview at 1 wk postoperatively.4,5 Indeed, our data were obtained retrospectively, our patients did not receive a Brice interview or other technique of specifically assessing awareness, and our patients were interviewed on postoperative day 1, all of which might result in an under-estimation of the true incidence.

Even with suboptimal conditions for detection of awareness, our rate of undesired awareness was still more than three times that of 0.0068% reported by Pollard et al.,6 in which a structured interview was used. It is important to note that the structured interview in the Pollard et al. study omitted a question specifically assessing recall that is present in the standard Brice interview. Our demographic data appeared comparable, with an average ASA classification of 2.27 (vs 2.37 in. Pollard et al.), an average age of 49 ± 18 years (vs 46 ± 16), and a male:female ratio of 1:1.1 (vs 1:1.3). Although data reported in the Pollard et al. study were gathered, in part, at an academic medical center, no resident or nurse trainees were identified as being involved in patient care. Since resident physicians are routinely involved in patient care at our institution, this may account for some disparity in outcome.

Another possible difference relates to the use of TIVA at our institution. The centers in the Pollard et al. study “rarely used” IV drugs as the sole anesthetic; in our study, 1 of the 10 patients who complained of awareness during general anesthesia experienced the event during a failed TIVA. Given the large number of cases analyzed, it is difficult to establish the precise number of anesthetics in which TIVA was used at some point. We have established, however, that approximately 9/10 cases used an inhaled anesthetic. Thus, the rate of awareness during known TIVA (1/10) is comparable to the overall use of TIVA in the study population (1/10).

Although our incidence of 0.023% was considerably less than that reported by Myles et al.11 and Sebel et al.4 (which ranged from approximately 0.10% to 0.20%), this disparity is likely mitigated by our lack of a structured interview and a 1 wk postoperative interview. Although not in perfect agreement with either study, it is easier to reconcile our data with that of Sebel et al.3 than Pollard et al.6

All patients who have postoperative complaints receive follow-up phone calls. Those reporting intraoperative awareness are offered psychiatric counseling. Of the 17 patients reporting undesired intraoperative awareness, only one requested psychiatric care (patient 10 receiving general anesthesia, described in Table 1). Although formal postawareness psychiatric evaluation was not systematically performed on all patients, it would appear that the occurrence of sustained psychiatric sequelae in our population was likely closer to that reported by Samuelsson et al.2 rather than Osterman et al.1 It must be noted, of course, that patients afflicted with PTSD often avoid health care professionals and clinical settings because they can serve as triggers that evoke traumatic memories.12

The most surprising finding of the present study is that the incidence of intraoperative awareness in patients who did not undergo general anesthesia (0.03%, n = 22,885) was not statistically different compared with those who did (0.023%, n = 44,006) (P = 0.54). There are several possible interpretations of this finding. We could postulate that, since sedation also suppresses consciousness and memory, perhaps the incidence was truly the same in anesthetized and sedated patients. This is, however, an absurd conclusion. Sedated patients are often very aware of their surroundings, as well as talking with the anesthesiologist and surgical team during the procedure. We therefore reject this interpretation. The more likely interpretation is that the resolution of this retrospective study of more than 100,000 anesthetics at a single institution was too low to capture the incidence with accuracy. Thus, the current study suggests critical methodological limitations to retrospective analyses, despite large sample sizes, and supports prospective approaches to assessing intraoperative awareness.

The finding of awareness complaints in patients not receiving general anesthesia is provocative. It is important to note that it was not simply awareness of pain, but awareness itself that was a source of distress. Several patients reported that they heard conversations during their procedure, indicating that this level of consciousness was inconsistent with their expectations. Furthermore, although 5/7 patients in this group reported pain, it was not the sole complaint. For example, patient 3 (Table 3) had a functioning spinal anesthetic but was distraught at hearing conversations, seeing bright lights, and feeling as if she had died. Although the significance is unclear, complaints of intraoperative awareness in patients receiving general anesthesia had a 1:1 male:female ratio, whereas this ratio was 1:6 in patients who did not have general anesthesia. Undesired intraoperative awareness in patients not receiving general anesthesia indicates the potential for disparity between the expectations of the patient and those of the anesthesiologist. We must also recognize that prior patient conversations with our surgical colleagues may establish expectations (e.g., complete unconsciousness) that are not met during the procedure itself. Unmet expectations, rather than events in themselves, may contribute to patient distress.

In conclusion, the incidence of undesired awareness during general anesthesia at our institution was more than three times as high as that recently reported by Pollard et al.,6 despite the fact that no formal awareness interview was used. The self-reported incidence of intraoperative awareness was not statistically different in patients receiving general anesthesia and those who did not. These results suggest that large retrospective analyses are probably inadequate to study intraoperative awareness. Furthermore, the dissatisfaction with awareness during nongeneral anesthetics suggests that prospective studies should evaluate the relationship between patient’s pre-procedure expectations and post-procedure perceptions of anesthetic adequacy.

Back to Top | Article Outline


1. Osterman JE, Hopper J, Heran WJ, Keane TM, van der Kolk BA. Awareness under anesthesia and the development of posttraumatic stress disorder. Gen Hosp Psychiatry 2001;23:198–204
2. Samuelsson P, Brudin L, Sandin RH. Late psychological symptoms after awareness among consecutively included surgical patients. Anesthesiology 2007;106:26–32
3. JCAHO. Joint Commission on Accreditation of Hospital Organizations Sentinel Event Alert, Report No. 32, 2004
4. Sebel PS, Bowdle TA, Ghoneim MM, Rampil IJ, Padilla RE, Gan TJ, Domino KB. The incidence of awareness during anesthesia: A multicenter United States study. Anesth Analg 2004;99:833–9
5. Sandin RH, Enlund G, Samuelsson P, Lennmarken C. Awareness during anaesthesia: A prospective case study. Lancet 2000;355:707–11
6. Pollard RJ, Coyle JP, Gilbert RL, Beck JE. Intraoperative awareness in a regional medical system: a review of 3 years’ data. Anesthesiology 2007;106:269–74
7. Kheterpal S, Han R, Tremper KK, Shanks A, Tait AR, O’Reilly M, Ludwig TA. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology 2006;105:885–91
8. Kheterpal S, Tremper KK, Englesbe MJ, O’Reilly M, Shanks AM, Fetterman DM, Rosenberg AL, Swartz RD. Predictors of postoperative acute renal failure after noncardiac surgery in patients with previously normal renal function. Anesthesiology 2007; 107:89
9. Brice DD, Hetherington RR, Utting JE. A simple study of awareness and dreaming during anaesthesia. Br J Anaesth 1970;42:535–42
10. Moerman N, Bonke B, Oosting J. Awareness and recall during general anesthesia. Facts and feelings. Anesthesiology 1993;179: 454–64
11. Myles PS, Williams DL, Hendrata M, Anderson H, Weeks AM. Patient satisfaction after anaesthesia and surgery: results of a prospective survey of 10,811 patients. Br J Anaesth 2000;84:6–10
12. Mashour GA, Jiang YD, Osterman J. Perioperative treatment of patients with a history of intraoperative awareness and post-traumatic stress disorder. Anesthesiology 2006;104:893–4
© 2009 International Anesthesia Research Society