Leslie, Kate MBBS, MD, MEpi, FANZCA*†; Chan, Matthew T. V. MBBS, FANZCA‡; Myles, Paul S. MBBS, MD, MPH, FANZCA, FCARSCI, FRCA§‖¶; Forbes, Andrew MSc, PhD#; McCulloch, Timothy J. MBBS, BSc (Med), FANZCA**††
The long-term consequences of an awareness episode vary. Some patients do not seem to experience any long-term disability,1 whereas others develop late psychological problems or posttraumatic stress disorder (PTSD).2 PTSD can develop after any traumatic event and is characterized by reexperiencing, avoidance, and physiological hyperarousal.3 The reported incidence of psychological sequelae of awareness varies widely (0%–78%),1,2,4–11 although many of these studies used retrospective study designs (which may introduce selection bias),4,6–10 lacked control groups,4–10 and did not formally assess patients using validated instruments.4,6–10 Two prospective studies did not diagnose severe late psychological sequelae in any of their aware patients,1,11 whereas Lennmarken et al.5 reported that the risk of severe late psychological sequelae in their prospective sample of noncardiac surgery patients was up to 50%. Further investigation of the consequences of awareness is therefore justified.12
The B-Aware Trial was a multicenter, randomized, controlled trial of bispectral index (BIS) monitoring to prevent awareness in 2463 patients presenting for cardiac and noncardiac surgery who were at high risk of awareness.13 The aim of this follow-up study was to compare the incidence of PTSD, diagnosed using a validated questionnaire, in B-Aware Trial patients with and without confirmed awareness.
We used a prospectively planned matched cohort study design. The B-Aware Trial was a multicenter, randomized, controlled trial of BIS monitoring to prevent awareness in adult patients at high risk for awareness presenting for surgery under general anesthesia with neuromuscular blockade.13 Patients were interviewed about awareness on 3 occasions postoperatively (2–6 h, 24–36 h, and 30 days) by a blinded observer using a standardized questionnaire.14 If patients reported awareness, they were asked whether any psychological consequences had ensued. Reports of awareness were adjudicated by a blinded independent committee. Psychological counseling (applied according to local practice) was offered to all patients who reported awareness. Thirteen patients had confirmed awareness (2 in the BIS group and 11 in the routine care group).
In this study, each surviving awareness case was matched with 4 controls for age (±10 yr), sex, cardiac or noncardiac surgery, date of surgery (closest possible), and hospital. The primary outcome was PTSD. Bias was minimized according to the guidelines of Horwitz and Feinstein.15 This included complete data collection for which identification of cases and controls was blinded, predetermined definition of cases and controls, clear specification of “exposure” (i.e., “awareness”), no reliance on patient recall at the time of follow-up (cases were identified in the original study), and equal diagnostic examination (i.e., all follow-up and interviews were analyzed equally, without knowledge of case/control identification). These features have been shown to protect against overestimation of the magnitude of an effect when compared with randomized trials on the same topic.16
Ethics Committee approval for this follow-up study was obtained in each center. In view of the patient consent obtained for inclusion in the B-Aware Trial, and the opportunity that existed for patients to refuse continuing participation in this study at the time of follow-up, waiving of a second consent process for this study was approved in all centers. After a review of the medical record, a face-to-face interview was conducted with each awareness patient and matched controls using the Clinician Administered PTSD Scale (CAPS).17 The interviews were conducted by clinical psychology trainees who were supervised by a registered clinical psychologist/board certified neuropsychologist. They attended a training workshop on the CAPS before interviewing began. The CAPS is a validated, structured clinical interview that meets Diagnostic and Statistical Manual of Mental Disorders (fourth edition) diagnostic criteria3 and is targeted to a specific traumatic event (i.e., the index surgery). This scale has been used in awareness research.2 The CAPS assesses the frequency and intensity of 17 core and 8 associated or hypothesized symptoms of PTSD, as well as the level of functional impairment. Frequency is rated on a 5-point scale (0 = never, 1 = once or twice, 2 = once or twice a week, 3 = several times a week, and 4 = daily). Intensity is rated on a 5-point scale (0 = no effort to avoid, 1 = minimal effort, 2 = moderate effort, 3 = considerable effort, and 4 = extreme effort). A symptom is defined as present if the frequency score is ≥1 and the intensity score is ≥2. The core symptoms are divided among 4 criteria (i.e., B, C, D, and F), and each criterion is deemed positive if a specific number of symptoms is present.18 Criterion A is deemed positive if the event was traumatic, and criterion E is deemed positive if the symptoms were present for ≥1 mo. A diagnosis of PTSD is made if all 6 criteria (A–F) are positive (Table 2). Finally, the interviewer estimated the validity of the set of symptom ratings obtained (0 = excellent to 4 = invalid) and the overall severity of the patient's illness (0 = symptomatic to 4 = extreme symptoms, pervasive impairment).
Our estimated sample size calculation was based on a study power of 80% and an expected incidence of PTSD in awareness patients and controls of 60% and 10%, respectively; this required a study size of 10 awareness patients and 40 controls.
Continuous data were summarized using median (range) and compared using Wilcoxon's ranked sum test. Categorical data were summarized using number (percent) and were compared using Fisher's exact test. Exact conditional logistic regression (adjusting for the matching factors of age, sex, cardiac or noncardiac surgery, date of surgery, and hospital) was used to compute adjusted odds ratios and 95% confidence intervals (CIs) relating awareness with PTSD. P < 0.05 was considered statistically significant.
Data collection for this study occurred between June 2006 and March 2007, with a median follow-up time of 5.3 yr (range, 4.3–5.7 yr).
Six of the 13 confirmed awareness patients had died. These patients were younger at the time of surgery (50 [29–64] yr vs 67 [64–81] yr) than surviving patients, but similar in terms of gender (3 females in each group) and ASA physical status. The causes of death were myocardial infarction (1), stroke (1), cancer (1), liver failure (1), and respiratory failure (2).
Four of the 7 surviving awareness patients were matched with 4 controls each, and 3 awareness patients could only be matched with 3 controls each. Patients with and without confirmed awareness had comparable demographic and surgical characteristics at baseline, except that control patients were more likely to be BIS-monitored than awareness patients (Table 1).
Five of the 7 confirmed awareness patients (71%) and 3 of the 25 controls (12%) fulfilled the criteria for PTSD (adjusted odds ratios = 13.3 [95% CI: 1.4–650]; P = 0.02) (Table 2). All these patients reported symptoms at the time of the interview. The median onset time of symptoms was 14 days (range, 7–243 days) after surgery, and the median duration of symptoms was 4.7 yr (range, 4.4–5.6 yr). The onset of symptoms was delayed ≥3 wk in only 1 patient (a control patient). The interviewers estimated the validity of the set of symptom ratings as excellent and the severity of the patient's illness as moderate. The incidences of associated or hypothesized features of PTSD were generally higher in the confirmed awareness patients compared with controls, but these were not statistically significant: guilt over acts of commission or omission (3 [12%] vs 3 [43%]; P = 0.10), reduction in awareness or surroundings (4 [16%] vs 2 [29%]; P = 0.59), derealization (2 [8%] vs 3 [43%]; P = 0.057), or depersonalization (2 [8%] vs 0 [0%]; P = 1.0), no survivor guilt in either group.
The awareness reports of the 7 patients are presented in Table 3. Four of the 7 aware patients reported the awareness at all the interviews that were conducted (3 developed PTSD). Two patients only reported awareness at the 30-day interview (both developed PTSD). One patient reported awareness at the 24- to 36-h interview but not at the 30-day interview (2- to 4-h interview not completed because patient was still sedated; no PTSD). Four of the 5 aware patients who developed PTSD reported pain in their description of the awareness episode, 4 reported being unable to move, and 1 reported being terrified. One of the aware patients who did not develop PTSD only recollected hearing talking. The other patient who did not develop PTSD reported being unable to move but “not too much pain.” Two of the 5 aware patients who developed PTSD and both of the aware patients who did not develop PTSD reported psychological consequences within 30 days of surgery.
The 3 control patients who developed PTSD underwent sternotomy for removal of slow-growing mediastinal tumor, redo aortic valve replacement for rheumatic heart disease, and repair of iliac artery aneurysm. All 3 patients required mechanical lung ventilation in the intensive care unit after surgery.
Severe late psychological sequelae were common and persistent in the confirmed awareness patients of the B-Aware Trial. Symptoms commenced several weeks after surgery and were present at the time of the interview approximately 5 yr later in all aware patients fulfilling the criteria of PTSD on the CAPS. We believe, therefore, that strategies to prevent awareness in patients under general anesthesia are justified.12,19
The incidence of PTSD in our patients (71%) was similar to the incidence of severe late psychological symptoms in the study by Lennmarken et al.5 They attempted to interview 18 prospectively identified awareness patients about 2 yr after surgery. Six declined to participate, 2 were lost to follow-up, and 1 had died. Of the remaining 9 patients, 4 (44%) had severe late psychological sequelae and 3 (33%) had milder psychological symptoms (4 of 9 patients versus 5 of 7 patients in our study; P = 0.35). Our aware patients were older, sicker, and undergoing more extensive surgery than the patients in the study by Lennmarken et al.5,20 In addition, the follow-up time of our patients was about 5 yr compared with 2 yr in the study by Lennmarken et al. However, as the number of patients in both these studies was small, it is difficult to draw statistically relevant comparisons. Samuelsson et al.4 reported a lower incidence of severe sequelae (17%), but they used a retrospective design and a telephone interview. A pooled estimate of the risk of severe sequelae from these studies is 26% (95% CI: 15%–37%).
Some of our results support the notion that early posttraumatic dissociation is a risk factor for long-term psychological disability.21 Dissociative symptoms, such as depersonalization or emotional numbing, were not formally assessed in our study. However, 2 of the 5 aware patients who developed PTSD in our study reported their awareness experience only at the 30-day interview. Furthermore, only 2 of the 5 awareness patients who developed PTSD admitted to psychological consequences of their awareness episode within 30 days of the operation. Yet, 7 of the 8 patients with PTSD subsequently reported that their symptoms had appeared within 3 wk of their surgery. An alternative explanation is that patients did not report their awareness episodes and their distress to the investigators.8 In either case, our results may support the view that long-term psychological follow-up should be offered to patients who report awareness regardless of their early postoperative psychological state.5
We chose not to include features of the awareness episodes (such as pain or inability to move) as potential predictors of PTSD. The descriptions of the awareness episodes were generally short and such post hoc classification by observers may be prone to bias. Ghoneim et al.22 collected 271 awareness cases reported in the literature since 1950. Twenty-two percent of these patients complained of postoperative psychological sequelae, although few of them were formally assessed for PTSD. The inability to move and feelings of helplessness and panic were associated with more psychological sequelae than other complaints including pain. Osterman et al.2 also emphasized the importance of “inescapable stress” rather than pain in their study. Qualitatively, the awareness reports of our 5 aware patients who did develop PTSD seemed more overwhelming than the reports of patients who did not develop PTSD.
An important feature of our study was the inclusion of a matched control group. The case-cohort study design is efficient because the follow-up interview only needs to be conducted in the patients selected to be controls, not the whole of the original sample. We included recommended steps to minimize bias,15 and this has been shown to provide estimates of effect that are comparable to randomized trials (although it should be noted that all the factors that determine awareness and PTSD cannot be measured and a trial randomizing patients to awareness cannot be conducted).16 By using a control group, we hoped to account for the underlying rate of severe psychological sequelae in the B-Aware Trial patients and avoid overestimating the effect of awareness. We did not match patients on psychological, sociological, or anesthetic factors because these data were not collected and/or have not been proven to be risk factors for PTSD, and/or we wanted to assess their predictive effect. Instead, we matched patients on core demographic criteria.
Three of the 25 control patients developed PTSD. These patients were undergoing highly invasive surgery with postoperative intensive care treatment. Previous studies confirm that major cardiac23 and noncardiac surgery24 and intensive care treatment25 represent significant psychological trauma for some patients, although the actual risk factors for development of PTSD are debated, and the risk of PTSD after routine surgery is not known. Basic screening for psychological consequences of surgery therefore could be justified in patients after major surgery, but we must emphasize that generalizations should not be made from our data to routine surgical patients.
A potential limitation of our study, as of others in this field, is the small sample size. The small sample size limited our ability to draw firm conclusions about the exact incidence of severe psychological sequelae in aware or control patients, and resulted in wide CIs. Furthermore, only 7 of the 13 confirmed awareness patients in our study were alive at long-term follow-up. The mortality rate in the B-Aware Trial patients was high (4% within 30 days of surgery),13 because the entry criteria included high-risk cardiac and noncardiac surgery, acute cardiovascular instability, end-stage lung disease, and airway endoscopy for investigation of cancer; the median age of the patients was 61 yr (interquartile range, 46–71 yr), and 75% of patients were ASA physical status 3–5. The alive and dead patients were similar at baseline, and all 7 surviving patients consented to participation. We chose a long follow-up time because we believed that demonstrating long-term psychological disability would be important.
In conclusion, severe psychological sequelae were common and persistent in awareness patients of the B-Aware Trial. Strategies to prevent awareness in patients having general anesthesia and psychological follow-up therefore are justified.
From the *Department of Anaesthesia and Pain Management, Royal Melbourne Hospital; †Department of Pharmacology, University of Melbourne, Melbourne, Australia; ‡Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Hong Kong, People's Republic of China; §Department of Anaesthesia and Perioperative Medicine, Alfred Hospital; ‖Academic Board of Anaesthesia and Perioperative Medicine, Monash University, Melbourne; ¶National Health and Medical Research Council, Canberra; #Department of Epidemiology and Preventive Medicine, Monash University, Melbourne; **Department of Anaesthetics, University of Sydney; and ††Department of Anaesthetics, Royal Prince Alfred Hospital, Sydney, Australia.
The original B-Aware Trial received sensors and some unrestricted funding from Aspect Medical Systems. Loaned BIS monitors were returned at the end of the original trial. Aspect Medical Systems had no role in the concept, design, data collection, data analysis, data interpretation, or writing of reports of either the original trial or the follow-up study.
1.Lopez U, Habre W, Van der Linden M, Iselin-Chaves I. Intra-operative awareness in children and post-traumatic stress disorder. Anaesthesia 2008;63:474–81
2.Osterman J, Hopper J, Heran W, Keane T, Van der Kolk B. Awareness during anesthesia and the development of post-traumatic stress disorder. Gen Hosp Psychiatry 2001;23:198–204
3.American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association, 1994
4.Samuelsson P, Brudin L, Sandin R. Late psychological symptoms after awareness among consecutively included surgical patients. Anesthesiology 2007;106:26–32
5.Lennmarken C, Bildfors K, Enlund G, Samuelsson P, Sandin R. Victims of awareness. Acta Anaesthesiol Scand 2002;46:229–31
6.Schwender D, Kunze-Kronawitter H, Dietrich P, Klasing S, Forst H, Madler C. Conscious awareness during anaesthesia patients' perceptions, emotions, cognition and reactions. Br J Anaesth 1998;80:133–9
7.Ranta S, Ranta V, Aromaa U. The claims for compensation for awareness with recall during general anaesthesia in Finland. Acta Anaesthesiol Scand 1997;41:356–9
8.Moerman N, Bonke B, Oosting J. Awareness and recall during general anesthesia. Facts and feelings. Anesthesiology 1993; 79:454–64
9.Evans J. Patients' experiences of awareness during general anaesthesia. In: Rosen M, Lunn J, eds. Consciousness, awareness and pain in general anaesthesia. London: Butterworths, 1987:184–92
10.Domino K, Posner K, Caplan R, Cheney F. Awareness during anesthesia: a closed claims analysis. Anesthesiology 1999;90: 1053–61
11.Ranta S, Laurila R, Saario J, Ali-Melkkila T, Hynynen M. Awareness with recall during general anesthesia: incidence and risk factors. Anesth Analg 1998;86:1084–9
12.Deem S, Souter M. B-Aware: recall of intraoperative events (letter). Lancet 2004;364:840
13.Myles P, Leslie K, McNeil J, Forbes A, Chan M. A randomised controlled trial of BIS monitoring to prevent awareness during anaesthesia: the B-Aware Trial. Lancet 2004;363: 1757–63
14.Brice D, Hetherington R, Utting J. A simple study of awareness and dreaming during anaesthesia. Br J Anaesth 1970;42:535–42
15.Horwitz R, Feinstein A. Methodologic standards and contradictory results in case-control research. Am J Med 1979;66:556–62
16.Concato J, Shah N, Horwitz R. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000;342:1887–92
17.Blake D, Weathers F, Nagy L, Kaloupek D, Gusman F, Charney D, Keane T. The development of a clinician-administered PTSD scale. J Trauma Stress 1995;8:75–90
18.Weathers F, Ruscio A, Keane T. Psychometric properties of nine scoring rules for the Clinician-Administered Posttraumatic stress disorder scale. Psychol Assesss 1999;11:124–33
19.American Society of Anesthesiologists Task Force on Intraoperative Awareness. Practice advisory for intraoperative awareness and brain function monitoring: a report by the American Society of Anesthesiologists Task Force on Intraoperative Awareness. Anesthesiology 2006;104:847–64
20.Sandin R, Enlund G, Samuelsson P, Lennmarken C. Awareness during anaesthesia: a prospective case study. Lancet 2000; 355:707–11
21.van der Kolk B, Fisler R. Dissociation and the fragmentary nature of traumatic memories: overview and exploratory study. J Trauma Stress 1995;9:505–25
22.Ghoneim M, Block R, Haffarnan M, Mathews M. Awareness during anesthesia: risk factors, causes and sequelae: a review of reported cases in the literature. Anesth Analg 2009;108:527–35
23.Rothenhäusler H, Grieser B, Nollert G, Reichart B, Schelling G, Kapfhammer H. Psychiatric and psychosocial outcome of cardiac surgery with cardiopulmonary bypass: a prospective 12-month follow-up study. Gen Hosp Psychiatry 2005;27:18–28
24.Liberzon I, Abelson J, Amdur R, King A, Cardneau J, Henke P, Graham L. Increased psychiatric morbidity after abdominal aortic surgery: risk factors for stress-related disorders. J Vasc Surg 2006;43:929–34
25.Griffiths J, Fortune G, Barber V, Young J. The prevalence of post traumatic stress disorder in survivors of ICU treatment: a systematic review. Intensive Care Med 2007;33:1506–18