Awareness 1960 – 2001; incidence, consequence, prevention
Rolf H Sandin, M.D., PhD
Chairman
Department of Anaesthesia and Intensive Care
Länssjukhuset
S-391 85 Kalmar
Sweden
In 1960 Ruth Hutchinson published the first study on the incidence of awareness during general anaesthesia (GA) (1). She found that 1.2% among 656 surgical patients had been aware. The year after, in 1961, Meyer and Blacher illustrated the mental consequences after awareness (2). Patients who had awakened paralysed during cardiac surgery tended to suffer from repetitive nightmares, anxiety, irritability and preoccupation with death. This was before the posttraumatic stress disorder (PTSD) had been identified as a syndrome (3,4).
Four decades have elapsed since the pioneering papers by Hutchinson, and Meyer and Blacher were published. We have witnessed the potent inhaled agents and the MAC-concept being introduced. Focus was shifted from vaporiser setting to a measure more relevant for the brain when continuous measurement of end-tidal anaesthetic gas concentrations (ETAGC) became possible. Non-depolarising neuro-muscular blocking agents with moderate duration of action and the possibility to monitor a limited degree of neuromuscular blockade are other leaps forward. Total intravenous anaesthesia (TIVA) became a real possibility with the introduction of propofol, although some feared that this technique would be associated with an increased incidence of awareness. Spinal anaesthesia has reduced the number of general anaesthetics for caesarian section – a procedure with increased risk for awareness. The laryngeal mask airway made relaxation for intubation unnecessary in a large number of procedures. The knowledge about pharmacokinetics and pharmacodynamics has increased tremendously leading to new concepts as “context-sensitive half time” and “t½ke0”, while leaving old parameters like t½b out in the cold. Cognitive function during anaesthesia is addressed not only in terms of explicit memory, much interest is currently devoted to implicit memory.
However, to what extent has progress in anaesthesia actually affected the risk for wakening in the middle of surgery since 1960, and if this should happen, what do we know about the general severity of this complication?
Incidence of awareness
Five prospective incidence studies based on reasonable numbers of patients, representing various types of anaesthesia and surgery, have been published since 1991 (5-9). However, most studies have been terminated soon after anaesthesia. It was recently found that memory for intraoperative wakefulness was delayed by several days in 50% of awareness cases (9), and previous studies may have failed to identify the true incidence of awareness. Apart from that and despite the fact that some relevant data are missing in most available studies, the average incidence of awareness in relaxant anaesthesia seems to be about 0.2%. In comparison with the period 1960 - 1986 it seems that the incidence of awareness with explicit recall has been reduced by approximately 80% during the last decade, albeit the specific reason for this remains unknown (see 6 for older references). Awareness is not confined to the use of neuro-muscular blockade, as it may occur in non-relaxant anaesthesia as well. Four patients among 4032 non-relaxed cases (0.1%) recalled intraoperative events (9). Two of those patients denied any attempt to move despite that they realised their situation during wakefulness. No randomised controlled trial (RCT) has compared the incidence of awareness in TIVA with other types of GA. The only available cohort study of awareness in TIVA with muscle relaxation reported an incidence of 0.2% (6). No RCT has evaluated the effectiveness of ETAGC in order to prevent awareness. The only cohort study differentiating between relaxant anaesthesia with and without ETAGC found a similar incidence of awareness - 0.2% - whether ETAGC was used or not (9). However, in that study at least 5 of the 14 awareness episodes in relaxant anaesthesia occurred during laryngoscopy or intubation, before any inhaled agent had been administered.
Suffering during and after awareness
Suffering due to awareness can be immediate in terms of pain, mental distress or both. Despite sufficient cognitive capacity to experience pain and anxiety, victims of awareness may not always be able to understand what is going on. In addition to immediate suffering, neurotic symptoms may follow (4). Most information about suffering due to awareness is published as case reports. However, all patients do not suffer during wakefulness, and it seems less likely that a case of awareness will be reported if the patient did not find any reason to tell anyone about intraoperative experiences and did not care about it (6,9). Thus, selection bias makes it impossible to assess the average severity of suffering due to awareness from case reports.
It is very laborious to collect a sufficiently large prospective cohort to draw any conclusions about the general severity of suffering among awareness cases. A significant number of non-consecutive cases have been enrolled in 3 studies by the use of other methods, i.e. by advertising (10), referral from colleagues of known cases (11), and both these methods (12) (Table 1). Another way to provide data from non-consecutive awareness cases, analysis of “closed-claims”, was used in a recent publication (13) (Table 1). However, these methods carry a risk for recruiting an unproportionate fraction of “complainers”, patients seeking economic compensation or patients with more severe suffering (11,13). Avoidance is one of the symptoms of PTSD (4). While some patients with PTSD due to violent crime or accidents avoid situations that remind them of the corresponding eliciting event, victims of awareness may avoid health care providers and be reluctant to discuss previous awareness#. Thus, selection bias cannot be ruled out with any of these methods. There is one reasonably large prospectively identified cohort of awareness cases with published details about late mental symptoms (9) (Table 3). In that study, 4 of 19 patients with awareness (n=18) or inadvertent paralysis (n=1) experienced neurotic symptoms, but all these patients claimed to have recovered within 3 weeks. However, this seemingly lower incidence of late symptoms in a prospectively identified cohort compared to studies using other methods for inclusion should await a more definite interpretation, since this cohort has been investigated two years after awareness, albeit not yet published# .
Monitoring and prevention
A large variety of methods aimed to avoid awareness have been reviewed elsewhere (14,15). Currently much attention is drawn to neuro-physiologic techniques based on either EEG or middle latency auditory evoked response (15). However, awareness is a rare phenomenon in terms of statistics, and studies advocating the merits of these various techniques have almost invariably aimed at other primary end-points than awareness per se. In fact, no single study has demonstrated a reduced incidence of awareness by the use of any proposed method. For some measures such as avoiding muscle relaxants, or the use volatile anaesthetics for caesarean section, this lack of evidence is probably due only to the fact that no relevant study has been conducted, while for other proposals the true benefit remains more obscure. Still, the absence of conclusive studies makes it futile to recommend how the current incidence of awareness should be reduced. This dilemma is enhanced by the availability of sophisticated, costly equipment. Peer-reviewed studies are needed to aid our decisions as to whether we should incorporate devices like BISÔ, A-lineÔ, NarcotrendÔ or PSA 4000Ô in clinical routine, or if limited health care resources in this era of financial constrains are better used in other ways. Unfortunately, sufficiently large RCTs will be very difficult to conduct (9), and we may have to accept also the second-best protocols. Even less than optimal studies, however, have to include very large numbers of patients, and they should focus on explicit recall rather than surrogate measures. Those studies are urgently needed and awaited.
References
1. Hutchinson R. Awareness during surgery. Br J Anaesth 1960; 33: 463-9
2. Meyer BC, Blacher RS. A traumatic neurotic reactioninduced by succinylcholine chloride. NY State J Med 1961; 61: 1255-61
3. Osterman JE, van der Kolk BA. Awareness during anaesthesia and posttraumatic stress disorder. General Hospital Psychiatry 1998; 20: 274-81
4. American Psychiatric Association: Diagnostic and statistical manual of mental disorders. 4th edition; Washington DC, American Psychiatric Association, 1994
5. Liu WH, Thorp TA, Graham SG, Aitkenhead AR. Perception and memory during general anaesthesia. Anaesthesia 1991; 46: 435-7
6. Nordström O, Engström AM, Persson S, Sandin R.Incidence of awareness in total i.v. anaesthesia based on propofol, alfentanil and neuromuscular blockade.
Acta Anaesthesiol Scand. 1997; 4: 978-84.
7. Ranta SOV, Laurila R, Saario J, Ali-Melkkilä T, Hynynen M. Awareness with recall during general anesthesia: incidence and risk factors. Anesth Analg 1998; 86: 1084
8. 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 Jan; 84: 6-10
9. Sandin RH, Enlund G, Samuelsson P, Lennmarken C. Awareness during anaesthesia: A prospective case study. The Lancet 2000; 355:707-11.
10. Evans JM. Patients´experiences of awareness during general anaesthesia. Consciousness, awareness and pain in general anaesthesia. Edited by Rosen M, Lunn JN. London, Butterworths, 1987, pp 184-92
11. Moerman N, Bonke B, Oosting J. Awareness and recall during general anesthesia. Anesthesiology 1993; 79; 454-64
12. Schwender D, Kunze-Kronawitter H, Dietrich P, Klasing S, Forst H, Madler C. Conscious awareness during general anaesthesia: patients´perceptions, emotions cognition and reactions. Br J Anaesth 1998; 80: 133-9
13. Domino KB, Posner KL, Caplan RA, Cheney FW. Awareness during anesthesia. Anesthesiology 1999; 90: 1053-61
14. Heier T, Steen PA. Assessment of anaesthesia depth. Acta Anesthesiol Scand 1996; 40: 1087-1100
15. Drummond JC. Monitoring depth of anesthesia. Anesthesiology 2000; 93: 876-82
# C. Lennmarken pers. comm.
Table 1.
Author Method Mental Pain Late
Distress symptoms
 |
Evans 1987 (10) retrospective 78% 41% NA
n=27 advertising
Moerman 1993 (11) retrospective 92% 39% 69%
n=26 referral
Schwender 1998 (12) retrospective 49%§ 24% 49%
n=45 advertising and referral
Domino 1999 (13) retrospective 11% 21% 84%#
n=79 closed claims
Sandin 2000 (9) prospective 47% 37% 21%*
n=19 repeated interview
§ 60% described helplessness
# 10% classified as PTSD
*The incidence of late symptoms in (9) is confined to statements within 3 weeks after surgery. The cases in this study have been subjected to a not yet published follow up study (C. Lennmarken, pers.comm.)