A national survey (NAP5-Ireland baseline) to estimate an annual incidence of accidental awareness during general anaesthesia in Ireland
You can respond to this article at http://www.anaesthesiacorrespondence.com
Summary
As part of the 5th National Audit Project of the Royal College of Anaesthetists and the Association of Anaesthetists of Great Britain and Ireland concerning accidental awareness during general anaesthesia, we issued a questionnaire to every consultant anaesthetist in each of 46 public hospitals in Ireland, represented by 41 local co-ordinators. The survey ascertained the number of new cases of accidental awareness becoming known to them for patients under their care or supervision for a calendar year, as well as their career experience. Consultants from all hospitals responded, with an individual response rate of 87% (299 anaesthetists). There were eight new cases of accidental awareness that became known to consultants in 2011; an estimated incidence of 1:23 366. Two out of the eight cases (25%) occurred at or after induction of anaesthesia, but before surgery; four cases (50%) occurred during surgery; and two cases (25%) occurred after surgery was complete, but before full emergence. Four cases were associated with pain or distress (50%), one after an experience at induction and three after experiences during surgery. There were no formal complaints or legal actions that arose in 2011 related to awareness. Depth of anaesthesia monitoring was reported to be available in 33 (80%) departments, and was used by 184 consultants (62%), 18 (6%) routinely. None of the 46 hospitals had a policy to prevent or manage awareness. Similar to the results of a larger survey in the UK, the disparity between the incidence of awareness as known to anaesthetists and that reported in trials warrants explanation. Compared with UK practice, there appears to be greater use of depth of anaesthesia monitoring in Ireland, although this is still infrequent.
Introduction
Recently, the 5th National Audit Project (NAP5) conducted a survey of senior anaesthetists in the UK to ascertain the incidence of accidental awareness during general anaesthesia (AAGA) as became known to them in the course of their practice 1. The striking finding was that, using a denominator value from the previous NAP4 study of 2.9 million general anaesthetics administered annually 2, this estimated incidence was ~1:15 000. The confidence intervals of this incidence were insensitive to large changes in denominator value, and this estimate was considerably lower than the 1–2:1000 reported in prospective trials that employ direct questioning of patients (e.g. a Brice questionnaire 3, 4). The survey also established that there was very low usage of specific depth of anaesthesia monitors, with just 25% of UK senior anaesthetists ever using this technology (more than one third of UK centres did not possess any such monitoring) 1.
This National Audit Project is the fifth in a series conducted by the specialty focusing on important topics in service evaluation (see: http://www.nationalauditprojects.org.uk/NAP5_home). Through the involvement of the Association of Anaesthetists of Great Britain and Ireland (AAGBI), it also covers Ireland. Although there are some similarities with respect to content of training and a common language, the health service structure is very different in Ireland. We therefore wished to know if the incidence of AAGA reported to anaesthetists was as low in Ireland as it appeared to be in the UK. We also wished to ascertain if clinical practice in relation to depth of anaesthesia monitoring differed.
Methods
The NAP5 project in Ireland has received approval from the Department of Health and is endorsed by the Health Service Executive (HSE) National Quality and Patient Safety Directorate. The requirement for ethical approval was waived. The survey and project infrastructure in Ireland is as that for the UK, as described elsewhere 1, 5, 6. Local co-ordinators were established in each of the 41 anaesthetic departments that covered 46 public hospitals with surgical services, and they distributed the survey form 1. Compared with the UK survey, there were some differences in design and terminology. Irish consultants often conduct sessions in more than one hospital and hence were asked only to complete a form in the hospital/department where they had the majority of their sessions. The survey also determined the number of non-consultant hospital doctors in each anaesthetic department. ‘Non-consultant hospital doctors’ is a term used for all non-consultant doctors in the health system in Ireland (training and non-training posts) who require immediate, local or distant supervision by a consultant.
Briefly, the questionnaires asked about the number of consultant staff and their years of experience as seniors; about the number of new cases of AAGA that became known to them (under their direct care or the care of those they supervised) during 2011 (and some relevant case details) as well as during their career as a consultant in Ireland; about the availability and use of depth of anaesthesia monitoring; and whether the hospital had policies for prevention or management of AAGA. Local co-ordinators could contact the NAP5-Ireland Clinical Lead (EPOS) or the National Co-ordinator (WRJ) for further advice and, in turn, the Clinical Lead or National Co-ordinator could contact the local co-ordinator for clarification of data entries on the survey forms.
As there was no hypothesis test, there were no statistical comparisons, but continuous data were described as median (IQR [range]) and categorical data with 95% CI for binomial or Poisson distributions 7, 8. Where illustrative, the goodness of fit of the data to a Poisson distribution was estimated by the least squares regression of actual vs modelled data.
Results
Table 1 shows the estimated numbers of consultant anaesthetic staff in public hospitals in Ireland, and the generally high proportions responding. It also shows the estimated number of non-consultant hospital doctors in Ireland. Median responses per centre were 100%, but in two hospitals where three consultants held the majority of their sessions, only one replied, and in one hospital where seven consultants held the majority of their sessions, only one replied, yielding a rather wide range.
| Totals | Consultants | NHCDs | |
|---|---|---|---|
| Total (n = 342) | Responding (n = 299; 87%) | Total (n = 430) | |
| Numbers/centre | 6 (4–13 [1–23]) | 5 (3–9 [1–23]) | 7 (5–12 [0–33]) |
| Response rate/centre | 100% (84–100% [14–100%]) | ||
Figure 1 shows the demography of staffing across anaesthesia departments in the public hospitals in Ireland, many of which are relatively small. The majority of hospitals (29; 70%) consisted of < 10 consultants with the majority of their sessions in that hospital. No hospital had > 30 consultant anaesthetists with the majority of their sessions in that hospital. In three-quarters of the hospitals (31, 75%), the number of non-consultant hospital doctors equalled or exceeded the number of consultants with the majority of their sessions in that hospital (Fig. 1b).
Figure 2 shows the distribution of mean years of experience of consultants across the hospitals, showing a bimodal distribution with one peak at a mean of ~15 years’ of experience, and a second peak at mean ~25 years’ of experience. Notwithstanding less than full-time individuals and details of job plans, the crude sum of years’ experience as a consultant of those responding to the survey was 3685 years.
There were eight new cases of AAGA reported to anaesthetists for the year 2011 (Table 2). Half were young or middle-aged adults (25–44 years) and half > 45 years. Five cases were volunteered by patients and three established through direct questioning by staff. Two cases related to experiences of AAGA at or soon after anaesthetic induction, but before surgery commenced, four during surgery and two were after completion of surgery, but before full emergence. Thus, the combined total for experiences during induction and emergence (i.e. the ‘dynamic phases’ of anaesthesia) was equal to those experienced during surgery (the ‘steady state phase’). Half the patients suffered pain or distress as part of their experience, three during surgery and one for an experience at or soon after induction. The consultants who responded to the survey did not know of any formal complaints or legal proceedings taken during 2011.
| Age range; years | How ascertained | Phase of anaesthesia/surgery in which awareness occurred | Pain or distress? |
|---|---|---|---|
| 24–44 | Volunteered | Surgery | Yes |
| 25–44 | Volunteered | Surgery | Yes |
| 25–44 | Questioning | After surgery, before full recovery | No |
| 25–44 | Volunteered | Surgery | No |
| 45–64 | Volunteered | Surgery | Yes |
| 45–64 | Volunteered | Induction | Yes |
| 45–64 | Questioning | After surgery, before full recovery | No |
| > 65 | Questioning | Induction | No |
Using a denominator for the number of general anaesthetics administered in public hospitals in Ireland in one year of 187 000 (rounded to the nearest 100 and obtained from a contemporaneous Anaesthetic Activity Survey 9), we can estimate an incidence of AAGA as becomes known to anaesthetists for the year 2011: one case for every 23 366 general anaesthetics (Table 2). Even if our method of estimating the denominator is inaccurate, Fig. 3 shows how that the calculated incidence will vary little across a wide range of denominator values.
These data mean that just one consultant anaesthetist out of approximately 37 will know of a new case each year (Table 3 and Fig. 4). This seems broadly consistent with the experience for 2011, with the likelihood that an individual consultant anaesthetist will have personal experience of an AAGA event just once every 46 years of his/her career (i.e. possibly never in his/her working life; Table 4). The vast majority have never had direct experience of a case for which they were responsible, but one respondent reported having experience of five cases (Fig. 4).
| Descriptor | Incidence |
|---|---|
| Cases of AAGA | 8 (3–16) |
| Incidence per general anaesthetic | 0.0043% (0.0016–0.0086%) |
| Cases: anaesthetic | 1:23 375 (1:11 628 – 1:62 500) |
| Cases per consultant per year | 1:37 (1:19 – 1:86) |
| Descriptor | Incidence |
|---|---|
| Cases; n | 82 (65–102) |
| Incidence; cases/consultant per year | 0.022 (0.018–0.028) |
| Cases: years of consultant practice | 1:45.5 (1:35.7–1:55.6) |
The majority of Irish hospitals possessed depth of anaesthesia monitoring, and almost two thirds of anaesthetists used it either routinely or in selected cases (Table 5), with almost 7% using it routinely. The Bispectral Index appears by far the most frequently used, with about two thirds of those who used any depth of anaesthesia monitoring employing this technique.
| Centres with depth of anaesthesia monitoring; n = 41 | Anaesthetists using depth of anaesthesia monitoring in selected cases and routinely; n = 299 | Anaesthetists using depth of anaesthesia monitoring in selected cases; n = 299 | Anaesthetists using depth of anaesthesia monitoring routinely; n = 299 | Type of depth of anaesthesia monitor used; n = 184 | |||||
|---|---|---|---|---|---|---|---|---|---|
| BIS | Entropy | EP | Narcotrend | IFT | Other | ||||
| 33 (80.0%) | 184 (61.5%) | 164 (54.8%) | 20 (6.7%) | 126 (68.5%) | 42 (22.8%) | 15 (8.2%) | 1 (0.5%) | 0 | 0 |
- BIS, bispectral index; EP, evoked potential monitoring; IFT, isolated forearm technique.
No public hospital in Ireland reported a policy to prevent or manage awareness.
Discussion
The results of this Irish survey require interpreting in the light of our recent, similar survey in the UK 1. The incidence of new cases of AAGA that becomes known to Irish consultant anaesthetists of ~1:23 000 seems comparable, even a little lower, to that in the UK (~1:15 000), but the wider confidence intervals for the Irish estimate are due to the smaller denominator (Fig. 5) and encompass the UK estimates. In other words, the incidence of AAGA known to anaesthetists in Ireland is unlikely to be more common than 1:10 000, and is at least as rare as that in the UK, if not rarer. The estimated career incidence per year of consultant practice is almost identical to the UK estimate of one case every 36–47 years, underlining the similarity of the data 1.
It is important to emphasise that the Irish estimate applies only to public hospitals. The accompanying paper indicates that approximately 40% of the caseload in Ireland occurs in the independent sector 9. If the seniority of staff or the relative fitness of patients (and lack of obstetric cases in independent hospitals) is influential upon the AAGA rate, then our statistic of 1:23 000 does not apply to Ireland as a nation, but to the public sector specifically.
There were too few cases of AAGA to examine detailed sub-correlations with age, phase of anaesthesia, etc (Table 2). Nonetheless, two national surveys from different countries now consistently show that estimates using anaesthetists’ knowledge of cases are very much lower than estimates obtained from direct questioning in prospective trials (1–2:1000). We, and others 10-15, have previously discussed possible reasons for the disparity, which broadly relate to potential patient, organisational or methodological factors. Patient factors include such severe psychological trauma that there is a reluctance even to discuss, let alone report, the experience. Conversely, it might be the case that the experience is felt by patients to be so trivial that they omit or forget to report it 16. Organisational factors include deficiencies in hospital reporting systems, or the fact that anaesthetists rarely see patients in outpatient clinics postoperatively, where a report of AAGA might be made. Methodological factors include the different nature of the studies undertaken to produce an ‘incidence’.
The strengths and limitations of this survey are similar to those of our parallel survey in the UK and have been extensively discussed 1, 15. One notable feature is the very large difference in size of the anaesthetic communities to which the surveys were directed. A total of just 342 consultant anaesthetists in public hospitals in Ireland (population 4 588 252 in the 2011 census 17) are dwarfed by 8672 consultants and staff grade/associate specialist anaesthetists in the NHS hospitals in the UK (population 63 200 000 in the 2011 census 18). The number of senior anaesthetists in Irish public hospitals per head of population is half that for UK (1:13 415 vs 1:7287). Expressed differently, the ratio of public hospital senior anaesthetists per 100 000 in Ireland and the UK is 7.5 and 13.7, respectively. This is, in part, due to the inclusion of the staff and associate specialist and career-grade anaesthetists as seniors, working in UK NHS hospitals as well as the small independent hospital sector in the UK.
Our estimate of staff numbers correlates well with other estimates. The HSE estimated a figure of 336 consultants (unpublished data), and the College of Anaesthetists in Ireland independently identified 379 consultants practising in public hospitals, a further 64 solely in private practice and 464 non-consultant hospital doctors 19. The public hospital anaesthetic consultant to anaesthetic non-consultant hospital doctor ratio of 1.26 and the consultant to 100 000 population of 7.5, estimated by our survey, fall well short of the recommendations set in the Report of the National Taskforce on Medical Staffing 2003 20 of 0.61 and 11, respectively. This raises concerns and challenges for anaesthetic training and service delivery in Ireland.
Another striking contrast between the UK and Irish data sets is the adoption of depth of anaesthesia monitoring. Whereas in the UK, more than one third (39%) of hospitals possessed no depth of anaesthesia monitoring and only a quarter of anaesthetists ever used this technology, in Ireland 80% of hospitals had access to depth of anaesthesia monitoring and the majority of anaesthetists (62%) used it at some time as part of their practice (Table 5). These data are broadly consistent with the findings of the accompanying paper 9. Over three times as many Irish anaesthetists (albeit a low absolute proportion; 6%) used it routinely compared with UK anaesthetists (< 2%). The fact that only 9% of patients were monitored with depth of anaesthesia suggests that, perhaps, it is these routine users who constitute the bulk of depth of anaesthesia use, whereas other anaesthetists employed it infrequently. The relative proportions using Bispectral Index and Entropy are broadly similar (75% and 69% for Bispectral Index vs 17% and 23% for Entropy in UK and Ireland, respectively), and perhaps the only minor but intriguing difference is that whereas 14 (0.7%) practitioners used the isolated forearm technique in the UK 1, it was not used at all in Ireland. We can speculate on the causes of these differences. The smaller size of each Irish anaesthetic department (a median of just 6 (Table 1) vs 27 for the UK 1) may lead to greater standardisation such that practice is more homogenous. Fewer monitors are therefore also needed in each hospital for usage to be high as a proportion of operating theatres or anaesthetists. It is possible that Irish anaesthetists regard AAGA as a more serious problem, to be tackled with greater use of depth of anaesthesia monitoring. However, this interpretation is not consistent with the finding that no Irish hospital reported having a policy to prevent or manage AAGA. It would be interesting to see if this pattern of depth of anaesthesia monitoring use changes in the light of the recent guidance from the National Institute for Health and Care Excellence (see http://guidance.nice.org.uk/DT/7), or as a result of criticism of that guidance 13, 21. There is no clear evidence provided by this survey that the apparently greater use of depth of anaesthesia monitoring by Irish anaesthetists has led to a lower annual or career incidence of AAGA compared with the UK.
In summary, this survey provides important information on staffing and demography of public hospital anaesthetic departments in Ireland. The annual and career incidence of AAGA that becomes known to anaesthetists is very similar to the incidence calculated using similar methodology in the UK 1. This incidence is, however, much lower than reported in prospective trials that use direct patient questioning, and this large disparity warrants further research and explanation.
Acknowledgements
This article is written on behalf of the 5th National Audit Project of the Royal College of Anaesthetists and AAGBI concerning ‘Accidental Awareness during General Anaesthesia’, of which JJP is Clinical Lead and TMC is Advisor. Other acknowledgements are given in Appendix S1.
This project was funded by the AAGBI, the Royal College of Anaesthetists, and the College of Anaesthetists of Ireland. Members of the NAP5 Steering Panel have registered their interests with the project, and there are no material interests to declare other than those related to the affiliations listed above.
Competing interests
JJP is Scientific Officer of the Difficult Airway Society and Editor of Anaesthesia (this paper has undergone additional external review as a result) and TMC and EPOS are on the editorial board of the British Journal of Anaesthesia. EPOS is President of the College of Anaesthetists of Ireland. No other competing interests declared.
