Printer Friendly

Are you aware of awareness?

Despite its relatively infrequent occurrence, awareness remains a significant and highly unpleasant adverse outcome following general anaesthesia which is of considerable concern to patients and has been associated with onset of post traumatic stress disorder (Heath et al 1996, Ghoneim 2000). Estimates of the incidence of awareness during anaesthesia are generally obtained through postoperative reporting and interview. However, accurate estimates are difficult to ascertain, due to the relatively small number of occurrences and the need for repeated interviews (Brice et al 1970, Liu et al 1991).

The term 'awareness' is commonly used to describe conscious or explicit recall of events during general anaesthesia. However, this term is confusing as it implies knowledge or cognisance on the part of the patient who is 'aware'. More accurate descriptions of this phenomenon focus on the presence of explicit or implicit memory (Schacter 1987, Ghoneim 2000, Sneyd & Matthews 2008). Explicit memory is best described as remembering; a patient with explicit recall may remember hearing conversations or sounds during the operation which can be recalled later. Implicit memory is less clear and may lead to the formation of memories that change performance or behaviour without the ability to recall specific events that took place leading to those changes.

Ghoneim (2000) gives the example of patients exposed to a list of words containing the word 'pension' during anaesthesia. Postoperatively, when presented with the stem 'PEN ...' and asked to supply the first word that came into their mind beginning with those letters, the formation of an implicit memory would mean they gave the word 'pension' more often than 'pencil' or 'peninsula' or others. In general, intraoperative awareness is regarded as awareness with explicit recall during anaesthesia (AER)(Cork 2006) and numerous risk factors have been suggested as influencing its incidence.. These include: age, ASA (American Society of Anesthesiologists) grade (Saklad 1941), anaesthetic technique, hypovolaemia, difficult intubation, previous history of awareness, obesity, chronic alcohol or drug use, and type of surgery--in particular trauma, obstetrics and cardiac surgery (Cork 2006, Ghoneim 2007).

The prevention of AER depends upon accurate delivery of anaesthetic agents and the accurate monitoring of clinical parameters to aid in the assessment of depth of anaesthesia. Use of pupil size, lacrimation, sweating and cardiovascular parameters alone are poor predictors of the depth of anaesthesia (Hadzidiakos et al 2006) and measurement of end tidal volatile anaesthetic agent concentration (ETAA) has become standard anaesthetic practice in the United Kingdom. ETAA is generally accepted as reflecting the concentration of volatile anaesthetic agent in arterial blood and brain tissue (effect site), provided there is an appropriate period of equilibrium of 15-30 minutes (Ropcke 2001). ETAA reflects the anaesthetic partial pressure at the site of action and is recorded as a proportion of the minimum alveolar concentration of anaesthetic (MAC) that prevents a sympathetic nervous response to a supramaximal pain stimulus in 50% of subjects.

Although ETAA measurements provide acceptably reliable estimates of brain anaesthetic concentration, these do not reflect the hypnotic effect of anaesthetics on brain activity. The need to guarantee prevention of AER has led to the development of processed electroencephalogram (EEG) monitors which reflect brain activity and may provide accurate data on the depth of anaesthesia. Several brain function monitors based on processed EEG or evoked potentials have been developed to assess anaesthetic depth, the most widely used of which in UK anaesthetic practice is the bispectral index (BIS). The aim of this article is to examine some of the current evidence to identify the risk of AER in contemporary anaesthetic practice and to ask whether this risk is reduced through the use of BIS.

Literature review

A literature search was undertaken to establish the extent of knowledge relating to the topic under investigation. The medical databases Medline, British Nursing Index, CINAHL, EMBASE, Cochrane Database of Systematic Reviews, ACP, DARE and The Cochrane Central Register of Controlled Trials were searched using the key words:

* anaesthesia/anesthesia/general anaesthesia/general anesthesia/GA

* awareness/recall

* anaesthesia/anesthesia/general anaesthesia/general anesthesia/ GA and awareness/recall.

The search was limited to English language and human. The time period covered was 1999-2010 to ensure that sufficient research was retrieved to enable a balanced perspective. Some backward chaining from the articles obtained was used to identify any relevant research that was missed. In total, six research articles were identified, those examining the incidence of AER following general anaesthesia (Sebel et al 2004, Pollard et al 2007, Errando et al 2008) and those that explored treatment comparisons comparing the incidence of AER with or without use of processed EEG (Myles et al 2004, Punjasawadwong et al 2007, Avidan et al 2008).

Discussion of reviewed literature

Three studies examining the incidence of AER were identified (Sebel et al 2004, Pollard et al 2007 and Errando et al 2008). Pollard et al (2007) conducted a multi site retrospective descriptive cohort study, to test the hypotheses that the incidence of AER in an academic centre would be similar to that reported elsewhere in the literature and that, due to the more complex nature of the cases being undertaken, the incidence would be higher than that seen in the other sites (community hospitals) taking part in the study. The level of AER detected by Pollard et al (2007) was 1:14,560 (0.0068%). Patients with AER or possible AER were found to be slightly older (55.5 vs 46 years) with a higher mean ASA classification (3.67 vs 2.37) and a longer duration of anaesthetic (340.7 mins vs 126 mins). However, Pollard et al (2007) found that, despite an increased average age and ASA grade in the academic centre, there was no statistical difference in incidence of AER between sites.

These results suggest a far lower level of AER than either of the contemporaneous studies by Sebel et al (2004) and Errando et al (2008). Sebel et al (2004) carried out a multi centre prospective descriptive cohort study to determine the incidence of AER in patients undergoing a general anaesthetic in the USA and found similar levels in each institution at approximately 0.13%-0.23%. These authors found no correlation between age or gender and AER, although a higher ASA grade was found to increase AER (odds ratio 2.41 when comparing ASA 1-2 with ASA 3-4; (95% CI 1.04-5.60)). Errando et al (2008) examined the incidence of AER in patients undergoing a general anaesthetic in Spain and found an even higher incidence of AER at 0.6% rising to 0.8% in high risk patients, such as caesarean section patients or patients operated on at night (p=0.019). A lower risk of AER was identified in patients receiving a benzodiazepine premedication (p<0.001). These three studies all suggest different rates of AER and so the methodologies need to be examined in more detail to determine the reasons for this variance.

The sample size varied in each of these studies, with none justifying their sample size with a power calculation. Sufficient numbers of participants are required to represent the population being studied; ultimately, the larger the sample size, the less likely a sampling error will occur (Parahoo 2006). Errando et al (2008) was the smallest study (n=4001), whereas Pollard et al (2007) and Sebel et al (2004) managed sample sizes of 87,361 and 19,575 respectively, which may reasonably be expected to be adequate sample sizes to test their hypotheses.

Each of these studies utilised a descriptive study design. Both Sebel et al (2004) and Errando et al (2008) utilised prospective cohort study designs, while Pollard et al (2007) conducted a non concurrent (retrospective) cohort study. Cohort studies cannot show an association between cause and effect, as no comparison is made to a control group, but they can determine the incidence of disease and so may be regarded as appropriate to determine the incidence of AER. Levels of attrition were accounted for in the studies by Sebel et al (2004) and Errando et al (2008), although Pollard et al (2007) may be exposed to some bias as the retrospective design meant that some patient records were incomplete or difficult to decipher and no explanation was given for missing data.

The Brice questionnaire (Brice et al 1970) modified by Liu et al (1991) is generally regarded as the gold standard to provide reliable and efficient data regarding AER and asks:

1. What is the last thing you remember before anaesthesia?

2. What is the first thing you remember after waking up?

3. Do you remember anything between going to sleep and waking up?

4. Did you dream?

5. What was the worst thing about the experience?

Although this questionnaire can be criticised due to its low diagnostic potential, it also carries a low risk of pseudo memory generation. However, the timing and frequency of interviews remain important considerations. AER may not be reported spontaneously by patients unless specifically asked, whether due to fear of reprisals, fear of not being believed or not wanting to appear ungrateful to the healthcare providers (Ghoneim 2007). Recall may be delayed by several days (Macleod & Maycock 1992, Russell & Wang 1997, Sandin et al 2000), as in the immediate postoperative period patients tend to focus on common symptoms such as pain and post operative nausea and vomiting, rather than recalling memory of intraoperative events. It has also been suggested that the trauma of awareness during surgery might lead to dissociation of memory of the event into sensory fragments and emotion states which interfere with the ability to develop a narrative of the events and that these memories may gradually return over time (Ghonheim 2007).

Each of these studies conducted an interview immediately postoperatively in the post anaesthetic care unit and followed this up post recovery. Errando et al (2008) interviewed subjects three times, whereas Sebel et al (2004) and Pollard et al (2007) interviewed subjects only twice. Pollard et al (2007) conducted their second interview 12 days post recovery, whereas this was extended to 1-2 weeks in the study by Sebel et al (2004) and Errando et al (2008), with Errando et al (2008) carrying out a third interview 30 days post recovery for those patients earlier identified as possible cases of AER. There is no definitive time span regarded as the best time to interview patients, although the 1-2 days allowed by Pollard et al (2007) may have been too short to detect all cases of AER and may account in part for the much lower incidence of AER detected in this study. One further point regarding the interviews conducted by Pollard et al (2007) is the modification to the Brice questionnaire used by these authors. Question 3 'do you remember anything between going to sleep and waking up?' was replaced with 'did you have any dreams while you were asleep for surgery?' thereby introducing confusion between dreaming and awareness. This, coupled with the problems of missing data, reduces the validity of these results meaning that they should be viewed with some caution.

From the literature identified through the search strategy, one systematic review (Punjasawadwong et al 2007) and two randomised controlled trials (Myles et al 2004, Avidan et al 2008) were identified that examined the incidence of awareness with explicit recall (AER) with BIS guided anaesthesia. Punjasawadwong et al (2007) carried out a systematic review to determine whether the use of BIS reduced anaesthetic use, recovery times, AER and cost. The review found that anaesthesia within the guided BIS range 40-60 could reduce incidence of AER in patients at high risk (OR 0.20 (95% CI 0.05-0.79)), could reduce time to eye opening by 2.43 min (95% CI--3.60-1.27) and response to verbal command by 2.28 min (95% CI -3.47-1.09).

This systematic review utilised two reviewers to assess studies to be included, which helped increase the reliability of the findings through reducing the possibility of researcher bias. However, although methodological quality was assessed according to a standardised quality checklist, only randomisation was considered and, of the 20 trials included, only four were deemed to have adequate allocation concealment. Heterogeneity in some analyses suggests wide differences in the design of included studies and, as the study by Myles et al (2004) accounted for 2463 of the total 4056 subjects, this may have influenced the findings, meaning that statistical significance may not equate to clinical significance.

In the first of the retrieved randomised controlled trials, Myles et al (2004) examined whether BIS guided anaesthesia reduced the incidence of AER in an adult surgical population. These authors found that up to day 30, reported AER was lower in the BIS guided group (0.17% vs 0.91%) with a number needed to treat (NNT) to prevent AER of 138 (95% CI 77-641). These results contrast with those found by Avidan et al (2008) in a randomised controlled trial to determine whether a BIS guided protocol is better than an end tidal volatile anaesthetic agent concentration (ETAA) guided protocol for decreasing AER in adult patients at high risk. Avidan et al (2008) found that incidence rates of AER or possible AER were higher in the BIS guided group (0.62%) than the ETAA guided group (0.31%) with a NNT to demonstrate benefit of an ETAA guided protocol of 179 and a number needed to harm (NNH) using a BIS guided protocol of 175. Differences in the study design of these randomised controlled trials may provide some explanation for these conflicting results.

Myles et al (2004) based their sample size (n=2463) on a power calculation to reduce AER from 1% to 0.1% which reduces the possibility of type two errors. Although Avidan et al (2008) did not state that a power calculation was used, their sample size was similar (n=2000) and so the chance of a type two error can be regarded as fairly similar. Both studies utilised a computer generated randomisation method which is regarded as reliable. However, despite this, each study displayed potentially significant differences in the baseline characteristics of comparison arms. Although individually each of these factors can be regarded as increasing the risk of AER, whether these factors each increase the risk to the same degree is uncertain. The control (standard anaesthesia) group in the study by Myles et al (2004) had more heavy alcohol users and ASA grade 5 subjects than the intervention (BIS guided) group which may have favoured the results of the intervention (BIS) group. Avidan et al (2008) included more patients with pre-existing neurological conditions and moderate exercise tolerance in the ETAA guided group, which also may have influenced the results.

One significant difference between these studies is the use of a protocol for the non BIS group. Myles et al (2004) stipulated that the control group should receive standard anaesthesia care (not protocol driven), whereas Avidan et al (2008) utilised an ETAA guided protocol for anaesthesia of the non BIS comparison arm. This difference may mean that the results of the study by Avidan et al (2008) are indicative of a lower AER when following a protocol rather than whether the protocol is based on BIS or ETAA.


This article has examined some of the available evidence in an attempt to identify the incidence of AER following general anaesthesia and whether this incidence is reduced through monitoring of ETAA or processed EEG (BIS). Three cohort studies, each attempting to identify the incidence of AER provided widely different results. Of the three, Pollard et al (2007) was the weakest in study design and this may be reflected in the lower incidence of AER detected in their study. The other studies by Sebel et al (2004) and Errando et al (2008) were both fairly rigorous in design, but results still varied from 1:1000 to 1:100 respectively. This difference may be accounted for in part by the longer follow up to final interview of 30 days by Errando et al (2008). Repeated interviews may detect a higher incidence of AER, either as memory formation and explicit recall return, or through the subconscious development of false memories. Although the reality may be somewhere between these two differing sets of results, it would seem prudent to err on the side of caution and assume the higher level of AER in clinical practice.

Whether the incidence of AER can be reduced through the use of processed EEG monitors to guide depth of anaesthesia has also provided conflicting results. The systematic review by Punjasawadwong et al (2007) suggested that BIS guided anaesthesia could reduce incidence of AER in patients considered at high risk. However, this meta-analysis was largely influenced by the randomised controlled trial by Myles et al (2004) who found a reduction in AER through use of a BIS guided protocol. Were the systematic review by Punjasawadwong et al (2007) to include data from the later randomised controlled trial by Avidan et al (2008), who found no such reduction when compared to an ETAA guided protocol, the results would clearly have been different.

What can be concluded from the results of these studies is that the use of a protocol to guide anaesthesia delivery can help reduce incidence of AER. However, whether this protocol should be guided by ETAA, BIS or some other form of processed EEG cannot yet be concluded upon. Two further randomised controlled trials are currently being undertaken which may go some way to answering these questions. Although the findings of these trials are not yet known, the study protocols were both published in 2009. The Michigan Awareness Control Study (Mashour et al 2009) is the largest prospective randomised controlled trial of awareness prevention yet undertaken and is designed to assess the incidence of AER at 28-30 days post operatively. The Bag-Recall clinical trial protocol (Avidan et al 2009) is powered to examine whether a BIS guided anaesthetic technique decreases AER when compared to an ETAA guided technique. The results of these concurrent trials are eagerly anticipated, as they should go some way to addressing the controversies and answering questions related to the incidence of AER and whether a BIS or ETAA guided protocol can reduce the incidence of this highly unpleasant clinical outcome.

No competing interests declared

Provenance and Peer review: Unsolicited contribution; Peer reviewed; Accepted for publication December 2010.


Avidan MS, Zhang L, Burnside BA et al 2008 Anaesthesia awareness and the Bispectral index The New England journal of medicine 358 1097-1108

Avidan MS, Palanca BJ, Glick D et al 2009 Protocol for the BAG--RECALL clinical trial: a prospective, multicenter, randomized, controlled trial to determine whether a bispectral index- guided protocol is superior to an anesthesia gas--guided protocol in reducing intraoperative awareness with explicit recall in high risk surgical patients BMC Anesthesiology 9 8

Brice DD, Hetherington RR, Utting JE 1970 A simple study of awareness and dreaming during anaesthesia. British Journal of Anaesthesia 42 535-542

Cork RC 2006 Awareness under anaesthesia Journal of Perianesthesia Nursing 21 288-290

Errando CL, Sigl JC, Robles M et al 2008 Awareness with recall during anaesthesia: a prospective observational evaluation of 4001 patients British Journal of Anaesthesia 101 1781-85

Ghoneim MM 2000 Awareness during anaesthesia Anesthesiology 92 597-602

Ghoneim MM 2007 Incidence of and risk factors for awareness during anaesthesia Best Practice and Research Clinical Anaesthesiology 21 327-343

Hadzidiakos D, Nowak A, Laudahn N, Baars J, Herold K, Rehberg B 2006 Subjective assessment of depth of anaesthesia by experienced and unexperienced anaesthetists European Journal of Anaesthesiology 23 292-299

Heath KJ, Sadler P, Winn JH, McFadzean WA 1996 N2O reduces the cost of intravenous anaesthesia European Academy of Anaesthesiology 13 369-372

Liu WH, Thorp TA, Graham SG, Aitkenhead AR 1991 Incidence of awareness with recall during general anaesthesia Anaesthesia 46 435-437

Macleod AD, Maycock E 1992 Awareness during anaesthesia and post traumatic stress disorder Anaesthesia and Intensive Care 20 378-382

Mashour GA, Tremper KK, Avidan MS 2009 Protocol for the Michigan Awareness Control Study: a prospective, randomized, controlled trial comparing electronic alerts based on bispectral index monitoring or minimum alveolar concentration for the prevention of intraoperative BMC Anesthesiology 9:7

Myles PS, Leslie K, McNeil J, Forbes A, Chan MTV 2004 Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomized controlled trial The Lancet 363 1757-1763

Parahoo K 2006 Nursing Research: Principles, process and issues 2nd Edition London, MacMillan Press

Pollard RJ, Coyle JP, Gilbert RL and Beck JE 2007 Intraoperative awareness in a regional medical system Anesthesiology 106 269-274

Punjasawadwong Y, Phongchiewboon A, Bunchungmongkol N 2007 Bispectral index for improving anaesthetic delivery and postoperative recovery Cochrane Database of Systematic Reviews 4 1-65

Ropcke H 2001 Effects of N2O on MAC Best Practice and Research Clinical Anaesthesiology 15 (3) 409-416

Russell IF, Wang M 1997 Absence of memory for intraoperative information during surgery under adequate general anaesthesia British Journal of Anaesthesia 78 3-9

Saklad M 1941 Grading of patients for surgical procedures Anesthesiology 2 281-284

Sandin RH, Enlund G, Samuelsson P, Lennmarken C 2000 Awareness during anaesthesia: a prospective case study The Lancet 355 707-711

Schacter DL 1987 Implicit memory: History and current status Journal of Experimental Psychology: Learning Memory and Cognition 134 501-518

Sebel PS, Bowdle TA, Ghoneim MM et al 2004 The incidence of awareness during anaesthesia: A multicenter United States study Anaesthesia and Analgesia 99 833-839

Sneyd JR, Matthews DM 2008 Memory and awareness during anaesthesia British Journal of Anaesthesia 100 (6) 742-744


Awareness--to have a conscious knowledge

Bispectral index monitoring--processed electroencephalogram signals used to measure sedation depth on a unitless scale from 0 to 100 (40 to 60=general anaesthesia; 60 to 90=sedation; 100=awake)

Explicit memory--ability to directly recall specific events

Implicit memory--the formation of memories that have changed performance or behaviour, but without the ability to recall specific events that took place leading to those changes

Intraoperative awareness--generally regarded as awareness with explicit recall

Memory--the stored information produced by learning

Minimum alveolar concentration concentration of anaesthetic at the alveolar level that prevents a sympathetic nervous response to a supramaximal pain stimulus in 50% of subjects

Recall--ability to remember specific events

Correspondence address: Luke Ewart, Canterbury Christ Church University, Rowan Williams Court, 30 Pembroke Court, Chatham Maritime, Kent, ME4 4UF. Email:

About the author

Luke Ewart

MSc, BSc (hons), PGCE, RODP

Senior Lecturer and ODP Pathway Director, Canterbury Christ Church University.
COPYRIGHT 2011 Association for Perioperative Practice
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ewart, Luke
Publication:Journal of Perioperative Practice
Article Type:Report
Geographic Code:1USA
Date:Mar 1, 2011
Previous Article:Reducing the risk of surgical site infection: a case controlled study of contamination of theatre clothing.
Next Article:Cyanoacrylate as a microbial sealant: examining the evidence.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters