Improvement in anaesthetic safety in the 1980s and 1990s
In all countries of Europe, anaesthesia has markedly progressed during the course of the last few decades. Over the last 40 years, there have been significant improvements in drugs, equipment and techniques. But the major change in our discipline has come from improvement in patient safety, and just how profoundly this has shaped our development will be set out below. The message for the next decade is that as anaesthesiologists, we must be alert to all the consequences that improvement in safety will inevitably have on our discipline. My concern for the next decade is that where there is a need to take decisions, the right decision will be taken. In making our decisions, we must be guided by our main objective as anaesthesiologists, to ensure the welfare of our patients and work for the best possible outcomes following anaesthesia and surgery.
This article will first address what has already been achieved in the field of safety, and the second part will identify the consequences of these achievements, on the demand for more anaesthetics, the pressures on staffing and finally on new safety targets for our discipline.
Safety in perspective
In the 1980s, anaesthetic-related catastrophes regularly featured in the headlines of the media, and a response was required. Several surveys were launched, and they all confirmed that anaesthesia mortality was at that time close to one death per 10 000 cases.1,2 The surveys also identified that two-third of deaths were occurring during the recovery phase.1 In Boston, Jeff Cooper et al.3 identified that accidents related to equipment failure or to disconnection were not rare, indicating that with better equipment, we might prevent some of these catastrophes. At that time, instrumental monitoring was used in less than 50% of anaesthetised patients, and at the end of anaesthesia, after monitoring for a while in the corridor of the operating room, 50% of patients were returned directly from the operating room to the ward.1 The information provided by the surveys, in most countries, led to proposals under three headings: training; monitoring equipment; and facilities, particularly recovery rooms. These proposals had the support of national bodies representing anaesthesia, who published recommendations, guidelines or standards. The measures were intended to standardise practice in these areas of concern.
It was understood that safety in anaesthesia had to be approached in the same manner that had been adopted in other high-risk activities. The approach was stepwise, starting with expertise and equipment, then addressing human errors and the different ways to reduce their occurrence, before looking at management problems and organisation.4 The first requirement was the need to change the mindset instilled in us by our education, the concept of the perfectible model, which had resulted in the belief that an error can only be the consequence of negligence or fallibility.5 It was necessary to realise that we all make errors, and that even with good training and the best intentions to respect the rules of good practice, errors still occur because of personal limitations or because of contributing factors embedded within the system. How factors from a variety of sources can interact is explained in the Swiss Cheese Model of James Reason.6 To reduce the incidence of one accident for more than 10 000 exposures, attention must be directed at the contributing factors that exist within the system, both at the level of the workplace and within the organisation.
The result was that, during the 1990s, in most of our departments, the problem of human error and its prevention was addressed. One example was drug administration errors in anaesthesia. First, it was important to admit that these errors were not so rare. In studies, anaesthesiologists and nurses were asked to prepare anaesthetic drugs as they commonly do in the operating room, and the number of errors was measured. Selection errors, taking the wrong vial from a drawer, occurred in 2% and the dilution error rate was 3.4%.7 Because these experiments required a departure from normal clinical procedures, some remained sceptical of their clinical relevance. Further studies were then conducted in the operating room, recovering the syringes that had been used from the trash. Concentration measurements conducted on these syringes found similar results: the differences between expected and measured concentrations were greater than 50% in 8% of the cases, and greater than 100% in 4% of the cases.8 The incidence of discrepancies was highest in the paediatric operating room. The measures adopted to improve practice involved changes in syringe labelling with the use of specific colours, and of new standards in drug labelling with recording of weight, volume and concentration and lastly with the introduction of prefilled syringes to remove the risk of dilution error altogether.
During the latter half of the 1990s and the first half of the new millennium, surveys on anaesthesia mortality were launched in Australia, The Netherlands, France and the U.S.9–12 They all confirmed that anaesthesia mortality had declined 10-fold, reaching a risk of one death directly related to anaesthesia alone per 200 000 anaesthetics. This was confirmed by data from other sources. During this same period, insurance premiums for anaesthesiologists were either stable or decreased.13 In addition, the number of claims for malpractice also decreased. When malpractice claims for anaesthesiology were compared with other disciplines, anaesthesiology appeared as a specialty with an intermediate or low risk of malpractice.14 Together, these different indicators confirmed that anaesthesia had become safer. Acknowledgement of this achievement by the media and our peers came in the famous report ‘To err is human’, from the American Institute of Medicine: ‘anesthesia is an area in which very impressive improvements in safety have been made’.15 Our profession can take pride in what has been accomplished.
Consequences of the improvement in anaesthesia safety
Growth in anaesthesia activity
The consequence of safer anaesthesia was that, by the end of the 1990s, the demand had increased dramatically, putting pressure on the staff and requiring renewal of the targets for our discipline. A French evaluation of the national workload confirmed that in comparison with 1980, the number of anaesthetics given had doubled.16 In terms of the population, it meant that the rate of anaesthetics had increased from 6.6 per 100 individuals in 1980, to 13.5 in 1996. Between the different age groups, there was a large variability in the rate of anaesthetics, with the major increase in the elderly population. The figures showed that 30% of men and 25% of women of 75 years or older had an anaesthetic every year, and it was evident that 50% of patients older than 85 years were still ASA physical status 1 or 2.
With regard to the type of procedures for which anaesthesia was required, it was clear that globally, most surgical activities had nearly doubled in 15 years, but some procedures, such as endoscopies, had exploded. This was seen in the French survey, in which anaesthesia for gastrointestinal endoscopies increased by 50-fold between 1980 and 1996. The same phenomenon was observed, progressively, in all countries. In Geneva University Hospitals, in the last 10 years, the number of procedures taking place outside the operating room has increased three times. Its growth now means that it represents 10 to 15% of total anaesthesia activity. In many hospitals, this figure approximates to the commitment to emergency anaesthesia.
Recent reports point out that, in ambulatory operating rooms, sedation or monitored anaesthesia care (MAC) represents one-third of operating room time for the anaesthesia team.17 Although there is little doubt that this is a huge development, it remains unclear whether this development is one we want, one we follow or one we undergo. How should we control the growth of this activity? This point will be considered later.
Turning now to the present situation, one of the challenges in the years to come will be how to manage the explosion of demand and the persistent growth in the activity of anaesthesia. It is unfortunate that the evolution of the development of the activity of anaesthesia is not something that is followed by the European Society of Anesthesiology. There are data available from France where workload has been monitored, and the figures are probably representative of most Western European countries. In France, over the last 30 years, population growth was around 16%. Over the same period, the growth of anaesthesia cases reached 322%.1,16,18 This was equivalent to a mean annual increase of 4% over 30 years. The most impressive increases were observed for endoscopies, caesarean sections and cataract surgery. The rate of anaesthetics, which was 13.5 per 100 in 1996, had reached 17.8 anaesthetics per 100 individuals. On average, each of us requires an anaesthetic every 6 years.
The next question for the future is to ask just how much our increasing workload will grow. An additional challenge for the next decade comes from the perspective of growth of the elderly population. In most European countries, those older than 65 years currently represent 15 to 20% of our population, and with this number expected to increase by 50% over the next 15 years. As the elderly represent 27% of current anaesthetic activity, we should anticipate that, if there is no major change in the indications for surgery, the growth in the elderly population over the next 15 years should be associated with a 13% increase in the number of anaesthetics. This is equivalent to approximately an annual 1% increase in our activity for the next 15 years. But, in comparison with the 4% increase in development that we have experienced during the last three decades, it will represent an important qualitative change, rather than a major quantitative change.
Staffing perspectives for our departments
Another challenge for the next decade is related to anticipated changes in the structure of anaesthetic departments. A real concern in the next years is the retirement of the ‘baby-boomers’, who will move from the care provider group to the patient group. This may well create a further problem in countries that already have trouble keeping up with the increasing activity of anaesthesia.
Data from several European studies over the last 10 years19–21 remind us that in the 1980s, the density of anaesthesiologists in Western Europe was around 5 per 100 000 population. By the year 2000 (Fig. 1), in spite of great national variability, the density had increased to around 15:100 000 in Western Europe, 10:100 000 in Central Europe and 6:100 000 in Eastern Europe. But a mean density can hide great variability within a single country. This was shown in France, with more than a three-fold difference between the south of France and a rural northern province, such as Picardie.20 However, we should understand that if we have a shortage of anaesthesiologists in Picardie, the response is a rapid movement of anaesthesiologists on lower salaries from European countries where usually the density is even lower. Staffing difficulties in high-revenue European countries have great importance for the lower revenue countries of Europe.
If summarised data can hide problems within a country, they are capable of revealing international staffing difference in Europe.19 The density of anaesthesiologists varies by more than two-fold between Sweden, Germany and Switzerland and other countries such as The Netherlands, Greece, Ireland and Poland (Fig. 2). These data are interesting, but they only offer a static view that does not predict the future and does not tell us whether anaesthesiologists are young and eager or ready to retire. Another predictor that we have is the density of trainees, which is also very variable between countries; there are countries that have made a great effort to increase the number of trainees, such as the UK and Ireland, whilst there are others with low densities of trainees, as in Italy, France and Greece.19
Another way to anticipate staffing difficulties is to make a demographic analysis of anaesthesiologists. Pontone et al.20,21 at the French Society made a number of assessments. In 1989, the situation was reassuring; there were plenty of young anaesthesiologists, who could cover all the night shifts, and a small number of older colleagues. Ten years later, the situation was still good, but now the baby-boomers were 45 years old, but they were still required to cover at night, with a limited number of young trainees. But by 2005, the situation had become difficult; the same generation of baby-boomers were now approaching their 60s, but still had to work at night. Approximately one half of French anaesthesiologists were more than 55 years old. At the current time, the situation is on course for correction, as the density of anaesthesiologists in France is still around 15:100 000, but the number of trainees has greatly increased. But this process is not finished, since between 2012 and 2016, it is anticipated that 2000 French anaesthesiologists will retire. This need will not be wholly addressed from within France, and inward migration can be expected to have a negative impact on the donor countries. This is in keeping with the pattern observed by Pontone et al.20,21 over the last 5 years. During this time, they analysed where the new anaesthesiologists were coming from: 62% came from French teaching centres; 25% from the European Union; and 13% were from countries outside the European Union.21
Another factor that may serve to protect departments with staffing difficulties is the composition of the anaesthesia team. There are different models for the anaesthesia team in Europe: some countries have nurse-anaesthetists and others have ‘circulation nurses’, nurses who can help, but who cannot be left alone with the patient in the operating room.22 If our assessment combines the density of anaesthesiologists, of nurse-anaesthetists and of trainees, it may be possible to identify countries that should be preserved from major staffing difficulties, and these include Sweden, Denmark, Norway and Switzerland. Those countries with low densities of anaesthetists, of trainees and of nurse-anaesthetists can expect shortages unless they have already embarked on corrective measures, as the problems were highlighted in 2007.19
If the current situation is summarised, using the French data, it appears that in the last 30 years (Fig. 3):
- anaesthesia activity has increased by a factor of 3;
- the number of anaesthetics per 100 000 population has increased by a factor of 2.7 (Table 1);
- to permit the above, the number of anaesthesiologists has increased by a factor of 3.4.
Instead of reviewing the last 30 years, it might be more relevant to consider the trends of the last 15 years, because it was during this time that staff numbers struggled to keep pace with increasing activity. Between 1996 and 2010, activity increased more than manpower, and new recruits could be found only by importing physicians from other countries that may have been experiencing the same or greater difficulties themselves. We cannot ascertain the exact position of the donor countries because this information, if it exists, is not in the public domain.
Finally, crossing anaesthesia activity with the density of anaesthesiologists gives an activity index for each individual, the mean number of cases per anaesthesiologist per year (Table 2). This index declined by 27% between 1980 and 1996, but increased again in the 1996 to 2010 period, reaching values 6% below the 1980 level.1,16,18 What is currently ignored is whether the increase in the number of anaesthetics between 1996 and 2010 was associated with a change in the level of safety. This defines an important challenge for anaesthesia within the next years: how to accommodate growth in activity without compromising on safety. It will be difficult to maintain current standards unless there is sufficient recruitment to accommodate the projected growth in activity. Assuming that both activity and staffing continue to grow at the same level experienced during the past 30 years, an annual rate of 4%, the projection would be that staffing should increase by 50% in 10 years and should double in 20 years.
Maintaining standards requires that the equation is balanced. Different solutions exist, but none of them is easy. One would be to reduce the demand for more anaesthetics, but there is no clear way to achieve this, and in any case, it is unclear what consequences this might have. Another is to recruit more anaesthesiologists. The easy solution has been to use higher salaries to tempt anaesthesiologists from public hospitals to the private sector or, from a lower revenue country to a higher one. Yet, another solution would be to increase the overall number of medical graduates, or alternatively to increase the proportion of anaesthesiologists at the expense of other specialists. Given that only 5 to 8% of graduates choose anaesthesia, the latter will not be an easy option. The last solution is to imagine a different pattern of working, with another method of specialty organisation.
Control the growing demand: example of anaesthesia outside the operating room
Procedures requiring sedation represent one example of how we can control the continuously growing demand for anaesthetics. At the Massachusetts General Hospital in Boston,23 Cravero et al.24 quantified anaesthesia activity and the potential for further increase. In 2007, they counted 37 000 anaesthesia cases, among which 12.4% were outside of the operating room. They also measured how many sedation cases were performed in this hospital by nonanaesthesia personnel, some 26 000 cases. If we view this as a specialty, can we afford to increase our activity by 70%? Alternatively, can we refuse to do it, allowing nonanaesthetists to manage these procedures? It is not easy to restrict the procedures to a number that we can cope with, nor can we decline if we consider ourselves to be the real experts on safety. But if we devote all our human resources to this activity, we risk finding ourselves unable to support other developments that might be essential for our future.
Two completely different answers to the sedation problem have been found. The first one has come from U.S. gastroenterologists. As anaesthesia was unable or unwilling to supervise endoscopy procedures, gastroenterologists decided to train their own staff to administer sedation with propofol. They conducted several studies on the use of propofol, generating heated debate in our journals. But their analysis indicates that they have done it quite safely with 650 000 procedures under sedation with acceptable results.25 This should not come as a surprise, as for low-risk procedures, even personnel with a low level of expertise, can be sufficient provided they have received adequate training.
The second approach came from paediatricians, who chose the opposite solution. In 2000, the U.S. Foods and Drugs Administration (FDA) published a serious alert concerning 95 severe accidents in children, including 51 deaths, which occurred during sedation procedures.26 Analysis of these cases showed that in a majority, sedation had been administered by a radiologist or a gastroenterologist without prior training, and the procedures had taken place with no monitoring in a remote part of the hospital. Nearly 10 years later, American paediatricians published a series of nearly 50 000 procedures under sedation with no mortality.24 Their solution was to give responsibility for managing sedation to a professional who was neither the radiologist nor the gastroenterologist, but was mostly a physician, whose exclusive task was to administer sedation. This physician was an intensivist, an emergency physician or a paediatric anaesthesiologist. However, an analysis of adverse events, and the interventions necessary to correct the situation, found that it was mainly jaw thrust, repositioning of the head and mask ventilation that were required. The conclusion was that safety needed no more than someone who was qualified to recognise the problem and intervene with the appropriate manoeuvre.
For anaesthesiologists, these are interesting developments. The suggestion is that in assuming the position of exclusive guardians of sedated patients we may be overreacting. In fact, by adopting such an attitude, we could push struggling anaesthesia departments to refuse to cover sedation, or we might create the risk of overqualification, with highly trained individuals deployed to low-risk areas, when they might be better used for example, dealing with the severely ill. Opinions on the different solutions may differ. What might be more important is to be certain we are asking the right questions. Two that come at the top of the list are: do we have the staffing for the solution we propose? Where in the clinical environment can the skills of anaesthesiologists be best used?
New targets for anaesthesia within the next decade?
The improvement in anaesthesia safety has been a significant achievement, and one way to put it in perspective is to consider perioperative risk from all causes. Two recent surveys have measured postoperative mortality; one from the Netherlands collected data from 3.7 million surgical procedures in 102 hospitals27 and found that postoperative mortality was 1.8%. The other surveyed 7-day cohorts of noncardiac surgical in-patients in 28 European countries28 and found postoperative mortality to be 4%, with a wide variation between countries. If we go on to compare surgical mortality with the data from anaesthesia mortality alone, it is clear that anaesthesia mortality is now 1000 times lower than surgical mortality. It should not be a surprise if authorities are little concerned with the needs of anaesthesia.
There is little advantage to the patient in pursuing greater safety for anaesthesia if surgical mortality remains unchanged. Improving the safety of anaesthesia has changed our priority for the future. The crux of the issue is whether we, as anaesthesiologists, should be concerned by the deaths of patients if they die, as most of them do, after they have left the postanaesthesia care unit?
As the practice of anaesthesia has improved, the global risk of surgery has moved temporally, from the intraoperative period to the first postoperative week. An evaluation of the complications that occur postoperatively reveals that most of them are not strictly surgical, but are the consequences of the surgical trauma. A surgical complication that would require a return to the operating room represents less than 5% of all complications. Again, should this concern us? Who is the best placed to manage these complications? The anaesthesiologists certainly have the expertise, but others have too.
In deciding whether or not to participate in postoperative care, we must bear in mind that, because it is a frequent cause of death, postoperative mortality is becoming a public health issue.29 Because reports have shown that 30 to 50% of these deaths are preventable, policy makers are now interested in postoperative deaths. It is why the WHO has launched programmes intended to reduce surgical mortality such as ‘safe surgery saves lives’.30 On our planet, every year, one inhabitant out of 25 will have surgery. If the decision for the future is to participate in postoperative care, the challenge is how to achieve this. Would it be a scandal to deploy overqualified staff to administer sedation and be unavailable to help improve surgical mortality in intermediate care units admitting severely ill patients after major surgery?
To give greater depth to these issues, there are two studies to consider. de Vries et al.31 decided to implement a 60-point checklist for surgical patients (SURPASS), starting with the patients’ admission, including the intraoperative and postoperative periods, and finishing with discharge from hospital. SURPASS was associated with a 46% reduction in postoperative mortality in the six hospitals where the checklist was used, although no change was observed in the five control hospitals. Clearly, if the whole team, anaesthesiologists, surgeons and nursing staff, work more closely together, there is the potential for a profound reduction in postoperative mortality. There is no drug that can influence postoperative outcome to this degree.
The second study addresses the question of how best to organise postoperative care. We know that complications are the main determinant of postoperative mortality. We know that often it is the first complication that initiates the cascade of events that ends in the death of a patient. For example, when a patient develops a major postoperative cardiac complication, the risk of postoperative death increases by a factor of 92.32 As a result, we target prevention of complications by improving intraoperative management and better postoperative care. With fewer complications, postoperative mortality should fall.
Unfortunately, this strategy is not as sound as we might like. Ghaferi et al.33 investigated the large database of postoperative complications and mortality developed by the American College of Surgeons. They measured the postoperative mortality rate for each hospital, adjusting it for the type of surgery and the severity of illness. Then, they ranked the different hospitals according to the mortality rate, grading them into five groups, from higher to lower mortality. Next, they looked at the rate of postoperative complications. We might expect the hospitals with a high mortality rate to have a high complication rate. In fact, all five groups had similar complication rates. So, what creates the different mortality rates between the hospitals is not the complication rate but probably the way the complications were managed. If this work is confirmed, then it seems that both concepts are valid. Although prevention of complications remains a worthwhile goal, the organisation of postoperative care must be given equal priority. But how is the latter to be achieved?
These two studies raise fascinating questions regarding the reduction of postoperative complications and postoperative mortality. We as anaesthesiologists should be involved in this adventure. If the coming generation accepts this challenge, I am convinced that postoperative mortality will reduce by 20%.
In conclusion, we can salute the generation of the mid-19th century who brought anaesthesia into mainstream hospital practice. They were followed by my generation, the builders, who developed hospital departments of anaesthesia, gave anaesthesia credibility as a clinical and academic discipline, and improved safety and the patient experience in general. With the coming generation, anaesthesia moves into adulthood. The issues debated here need to be addressed. The demand for evermore anaesthetics needs to be met for the sake of the patients. There are decisions to be taken with regard to the particular mix of skills in the staff that must be recruited to meet the demand. The other challenge is political; we must decide the scope of our practice.
Acknowledgements relating to this article
Assistance with the article: none.
Financial support and sponsorship: none.
Conflicts of interest: none.
Comment from the editor: this special article is based on the text of the Sir Robert Macintosh Lecture, held at the Euroanaesthesia meeting in Stockholm on 31 May 2014. The article was checked and accepted by the editors but was not sent for external peer review.
1. Hatton F, Tiret L, Maujol L, et al. Enquête épidémiologique sur les accidents d’anesthésie. Premiers resultants (Epidemiological survey of accidents in anaesthesia. Initial results). Ann Fr Anesth Réanim
2. Hovi-Viander M. Death associated with anaesthesia in Finland. Br J Anaesth
3. Cooper JB, Newbower RS, Kitz RJ. An analysis of major errors and equipment failures in anesthesia management. Considerations for prevention and detection. Anesthesiology
4. Visser JP. Developments in HSE management in oil and gas exploration and production. In: Hale AMB, editor. Safety management: the challenge of change. Oxford: Pergamon; 1998. pp. 43–65.
5. Leape LL. Error in medicine. JAMA
6. Reason J. Understanding adverse events: human factors. Qual Healthcare
7. Garnerin P, Pellet-Meier B, Chopard P, et al. Measuring human-error probabilities in drug preparation: a pilot simulation study. Eur J Clin Pharmacol
8. Stucki C, Sautter AM, Wolff A, et al. Accuracy of preparation of i.v. medication syringes for anesthesiology. Am J Health Syst Pharm
9. Arbous MS, Grobbee DE, van Kleef JW, et al. Mortality associated with anaesthesia: a qualitative analysis to identify risk factors. Anaesthesia
10. Gibbs N, Borton C. Safety of anaesthesia in Australia. A review of anaesthesia related mortality 2000-2002. Australian and New Zealand College of anaesthetists. Anaesthesia
11. Lienhart A, Auroy Y, Pequignot D, et al. Survey of anesthesia-related mortality in France. Anesthesiology
12. Li G, Warner M, Lang BH, et al. Epidemiology of anesthesia-related mortality in the United States, 1999-2005. Anesthesiology
13. Cook TM, Bland L, Mihai R, Scott S. Litigation related to anaesthesia: an analysis of claims against the NHS in England 1995–2007. Anaesthesia
14. Jena AB, Seabury S, Lakdawalla D, Chandra A. Malpractice risk according to physician specialty. N Engl J Med
15. Institute of medicine. To err is human. Washington, DC: National Academy Press; 2000.
16. Clergue F, Auroy Y, Pequignot F, et al. French survey of anesthesia in 1996. Anesthesiology
17. Bayman EO, Dexter F, Laur JJ, Wachtel RE. National incidence of use of monitored anesthesia care. Anesth Analg
18. Dadure C. ‘Photographie’ des actes d’anesthésie réalisés en France en 2010 (Photography of anaesthetic procedures taken in France in 2010). SFAR; 2012. http://www.sfar.org/accueil/article/940/laquo-photographie-raquo-des-actes-d-rsquo-anesthesie-realises-en-france-en-2010
. Accessed 29/10/2012
19. Egger Halbeis CB, Cvachovec K, Scherpereel P, et al. Anaesthesia workforce in Europe. Eur J Anaesthesiol
20. Pontone S, Brouard N, Scherpereel P, et al. Cfar-Sfar-Ined Working Group. Demography of French anaesthesiologists. Results of a national survey by the French College of Anaesthesiologists (CFAR) and the French National Society of Anaesthesia and Intensive Care (SFAR). Eur J Anaesthesiol
21. Pontone S, Brouard N. La démographie médicale des anesthésistes réanimateurs est-elle encore compromise en France à l’horizon 2020? (Despite corrective measures, will there still be a lack of anaesthetists and intensive care physicians in France by 2020?). Ann Fr Anesth Réanim
22. Meeusen V, van Zundert A, Hoekman J, et al. Composition of the anaesthesia team: a European survey. Eur J Anesthesiol
23. Pino RM. The nature of anesthesia and procedural sedation outside of the operating room. Curr Opin Anaesthesiol
24. Cravero JP, Beach ML, Blike GT, et al. Pediatric Sedation Research Consortium. The incidence and nature of adverse events during pediatric sedation/anesthesia with propofol for procedures outside the operating room: a report from the Pediatric Sedation Research Consortium. Anesth Analg
25. Rex DK, Deenadayalu VP, Eid E, et al. Endoscopist-directed administration of propofol: a worldwide safety experience. Gastroenterology
26. Cote CJ, Notterman DA, Karl HW, et al. Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics
27. Noordzij PG, Poldermans D, Schouten O, et al. Boersmae. Postoperative mortality in The Netherlands: a population-based analysis of surgery-specific risk in adults. Anesthesiology
28. Pearse RM, Moreno RP, Bauer P, et al. European Surgical Outcomes Study (EuSOS) group for the Trials groups of the European Society of Intensive Care Medicine and the European Society of Anaesthesiology. Mortality after surgery in Europe: a 7 day cohort study. Lancet
29. Bartels K, Karhausen J, Clambey ET, et al. Perioperative organ injury. Anesthesiology
30. World Health Organization. World alliance for patient safety. The second global. Patient safety challenge. Safe surgery saves lives. http://www.who.int/patientsafety/safesurgery/knowledge_base/SSSL_Brochure_finalJun08.pdf
. Geneva: WHO Press, World Health Organization; 2008.
31. de Vries EN, Prins HA, Crolla RM, et al. Effect of a comprehensive surgical safety system on patient outcomes. N Engl J Med
32. Silber JH, Rosenbaum PR, Trudeau ME, et al. Changes in prognosis after the first postoperative complication. Med Care
33. Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in hospital mortality associated with inpatient surgery. N Engl J Med