Safety of Anaesthesia A review of anaesthesia- related mortality reporting in Australia and New Zealand 2009-2011 Report of the Mortality Sub-Committee convened under the auspices of the Australian and New Zealand College of Anaesthetists Editor: Associate Professor Larry McNicol, MBBS, FRCA, FANZCA CCoonntteennttss IInnddeexx ooff ttaabblleess aanndd fifigguurreess IInnddeexx ooff ttaabblleess aanndd fifigguurreess IIFFCC TTaabbllee 11.. SS yysstteemm ooff ccllaassssiifificcaattiioonn bbyy ssttaattee--bbaasseedd aannaaeesstthheessiiaa FFoorreewwoorrdd 11 mmoorrttaalliittyy ccoommmmiitttteeeess 77 MMoorrttaalliittyy SSuubb--CCoommmmiitttteeee mmeemmbbeerrss 22 TTaabbllee 22.. NN uummbbeerr ooff ddeeaatthhss ccllaassssiififieedd bbyy eeaacchh ccoommmmiitttteeee 88 EExxeeccuuttiivvee SSuummmmaarryy 33 TTaabbllee 33.. NN uummbbeerr ooff aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss dduurriinngg tthhee RReeccoommmmeennddaattiioonnss 55 22000099--1111 ttrriieennnniiuumm iinn rreellaattiioonn ttoo tthhee ppooppuullaattiioonn CClliinniiccaall aassppeeccttss ooff ccaatteeggoorryy oonnee aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss 66 ((ooff NNSSWW,, VViicc,, WWAA,, TTaass aanndd SSAA)) 88 MMeetthhooddss 77 TTaabbllee 44.. NN uummbbeerr ooff aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss iinn ccoommppaarriissoonn DDaattaa ccoolllleeccttiioonn 77 wwiitthh pprreevviioouuss rreeppoorrttss 88 SSyysstteemm ooff ccllaassssiifificcaattiioonn 77 TTaabbllee 55.. NN uummbbeerr ooff aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss iinn rreellaattiioonn ttoo FFiinnddiinnggss ffoorr NNSSWW,, VViicc,, WWAA,, TTaass aanndd SSAA 88 ppooppuullaattiioonn iinn ccoommppaarriissoonn ttoo pprreevviioouuss rreeppoorrttss 88 NNuummbbeerr ooff ddeeaatthhss ccllaassssiififieedd 88 TTaabbllee 66.. CC aauussaall oorr ccoonnttrriibbuuttoorryy ffaaccttoorrss iinn aannaaeesstthheessiiaa--rreellaatteedd NNuummbbeerr ooff aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss ((ccaatteeggoorryy oonnee ttoo tthhrreeee)) ddeeaatthhss 99 iinn rreellaattiioonn ttoo ppooppuullaattiioonn 88 CCaauussaall oorr ccoonnttrriibbuuttoorryy ffaaccttoorrss iinn aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss 99 TTaabbllee 77.. AA ggee ddiissttrriibbuuttiioonn iinn aannaaeesstthheessiiaa--rreellaatteedd ddeeaatthhss 1100 GGeennddeerr 1100 TTaabbllee 88.. LL eevveell ooff rriisskk ooff ppaattiieennttss bbyy AASSAA pphhyyssiiccaall ssttaattuuss 1100 AAggee 1100 TTaabbllee 99.. DD eeggrreeee ooff uurrggeennccyy ooff tthhee pprroocceedduurree wwiitthh aannaaeesstthheessiiaa-- LLeevveell ooff rriisskk 1100 rreellaatteedd ddeeaatthhss 1100 DDeeggrreeee ooff uurrggeennccyy 1111 TTyyppee ooff hhoossppiittaall 1111 TTaabbllee 1100.. TT yyppee ooff hhoossppiittaall 1111 LLooccaattiioonn ooff eevveenntt lleeaaddiinngg ttoo ddeeaatthh 1111 TTaabbllee 1111.. LL ooccaattiioonn ooff eevveenntt lleeaaddiinngg ttoo ddeeaatthh 1111 LLooccaattiioonn ooff ddeeaatthh 1111 TTaabbllee 1122.. LL ooccaattiioonn ooff ddeeaatthh 1111 GGrraaddee ooff aannaaeesstthheettiisstt 1122 TTyyppee ooff ssuurrggeerryy oorr pprroocceedduurree 1122 TTaabbllee 1133.. GG rraaddee ooff aannaaeesstthheettiisstt 1122 NNuummbbeerr ooff aannaaeesstthheettiiccss aaddmmiinniisstteerreedd 1133 TTaabbllee 1144.. TT yyppee ooff ssuurrggeerryy oorr pprroocceedduurree 1122 HHiieerraarrcchhyy uusseedd bbyy ‘‘ccooddeerrss’’ 1133 TTaabbllee 1155.. EE ssttiimmaatteedd nnuummbbeerr ooff aannaaeesstthheettiiccss aaddmmiinniisstteerreedd IInncciiddeennccee ooff ddeeaatthh rreellaatteedd ttoo aannaaeesstthheessiiaa 1144 iinn tthhee fifivvee ssttaatteess aanndd tthhee eessttiimmaatteedd aannaaeesstthheessiiaa IInncciiddeennccee ooff ddeeaatthh iinn ppaattiieennttss ccoonnssiiddeerreedd ttoo bbee ggoooodd oorr ffaaiirr rriisskk 1144 mmoorrttaalliittyy rraattee ppeerr nnuummbbeerr ooff pprroocceedduurreess 1133 SSttaattee,, TTeerrrriittoorryy aanndd NNaattiioonnaall IInnffoorrmmaattiioonn 1155 TTaabbllee 1166.. EE ssttiimmaatteedd aannaaeesstthheessiiaa--rreellaatteedd mmoorrttaalliittyy iinn NNeeww SSoouutthh WWaalleess 1155 rreellaattiioonn ttoo tthhee nnuummbbeerr ooff pprroocceedduurreess ccoommppaarreedd VViiccttoorriiaa 1166 ttoo pprreevviioouuss rreeppoorrttss 1144 WWeesstteerrnn AAuussttrraalliiaa 1188 TTaabbllee 1177.. II nncciiddeennccee ooff ddeeaatthh iinn ppaattiieennttss ccoonnssiiddeerreedd ttoo bbee ggoooodd TTaassmmaanniiaa 1199 oorr ffaaiirr rriisskk 1144 SSoouutthh AAuussttrraalliiaa 2200 QQuueeeennssllaanndd 2211 AAuussttrraalliiaann CCaappiittaall TTeerrrriittoorryy 2222 NNeeww ZZeeaallaanndd 2222 RReeffeerreenncceess 2255 AAppppeennddiicceess 2266 AAppppeennddiixx 11:: GGlloossssaarryy ooff tteerrmmss –– ccaassee ccllaassssiifificcaattiioonn 2266 AAcckknnoowwlleeddggeemmeennttss 2288 ©© CCooppyyrriigghhtt –– AAuussttrraalliiaann aanndd NNeeww ZZeeaallaanndd CCoolllleeggee ooff AAnnaaeesstthheettiissttss 22001144.. TThhiiss wwoorrkk iiss ccooppyyrriigghhtt.. NNoo ppaarrtt mmaayy bbee rreepprroodduucceedd wwiitthhoouutt pprriioorr wwrriitttteenn ppeerrmmiissssiioonn ffrroomm AANNZZCCAA,, ootthheerr tthhaann ffoorr bboonnaa fifiddee ppeerrssoonnaall uussee oorr rreesseeaarrcchh oonn aa nnoonn--ccoommmmeerrcciiaall bbaassiiss.. TThhee mmaatteerriiaall ccoonnttaaiinneedd iinn tthhiiss ddooccuummeenntt hhaass bbeeeenn pprreeppaarreedd bbaasseedd oonn iinnffoorrmmaattiioonn ssuupppplliieedd ffrroomm ootthheerr ppeerrssoonnss aanndd oorrggaanniissaattiioonnss.. WWhhiillee eevveerryy eeffffoorrtt hhaass bbeeeenn ttaakkeenn ttoo vveerriiffyy tthhee aaccccuurraaccyy ooff tthhee iinnffoorrmmaattiioonn,, nnoo rreessppoonnssiibbiilliittyy iiss ttaakkeenn bbyy tthhee ppuubblliisshheerr ffoorr aannyy lloossss aarriissiinngg ffrroomm rreelliiaannccee oorr uussee.. IISSBBNN 997788--00--99887733223366--22--00 Foreword Dr Genevieve Goulding Ensuring our patients enjoy the highest in healthcare standards To ensure we have a more complete picture of the Australian and outcomes is a critical role for ANZCA and the reason reports situation, the reporting is set to expand, with ANZCA’s ACT such as Safety of Anaesthesia: A review of anaesthesia-related Regional Committee working with the ACT Audit of Surgical mortality reporting in Australia and New Zealand 2009-2011 play Mortality (Royal Australasian College of Surgeons) to develop an important role in what we do. anaesthesia mortality reporting. The South Australian committee could receive data from the Northern Territory as has occurred Reporting this important information also complies with ANZCA’s previously. New Zealand has a multi-disciplinary perioperative first strategic priority to “advance standards through training, mortality committee, and although this committee is unable to education, accreditation and research”. specifically provide the information needed for our reports, its This year, Safety of Anaesthesia contains data from five Australian report is very informative. states (NSW, SA, Tasmania, Victoria and WA). This represents The efforts of all involved in compiling this report, in particular 17.3 million or 70 per cent of the Australian population and its editor, Associate Professor Larry McNicol, the mortality can thus be considered a reasonable estimate of anaesthetic committees and the reporting anaesthetists are gratefully mortality for this period. It equates to an anaesthesia mortality acknowledged along with the co-operating coroners. rate of three deaths per million population per annum, which is very similar to the figure (2.79) in the previous triennium (2006-08) for the four states covered (NSW, Tasmania, Vic, Dr Genevieve Goulding and WA). President, ANZCA Importantly, this report shows a continued reduction in the percentage of category one deaths (where it was considered “reasonably certain” that death was caused by anaesthesia factors alone). In 2003-05, category one deaths made up 21 per cent of the total anaesthesia-related deaths, and this reduced to 15 per cent in 2006-08 and 14 per cent in 2009-11. For the first time, a brief clinical summary of the category one causes of death is included. Interestingly, of the 22 category one deaths, seven were due to anaphylaxis, five involved management of the airway, five involved pulmonary aspiration and three deaths involved cardiac arrest. This report continues a long tradition in reporting anaesthesia- related deaths, which started in 1960 in NSW. More recently, the South Australian Mortality Committee was re-established (in 2010) and has been able to provide mortality data for this ninth triennial report. The next report (2012-14) will contain Queensland data thanks to the re-establishment in 2012 of the Queensland Perioperative and Peri-procedural Anaesthetic Mortality Review Committee (QPPAMRC). A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 1 Mortality Subcommittee The Mortality Subcommittee members who produced this report include the president of the Australian and New Zealand College of Anaesthetists (ANZCA), the chairs or co-ordinators of functioning state mortality committees, and other interested parties as listed: ANZCA President Dr Genevieve Goulding Chairs or co-ordinators of functioning (2014) Australian state/territory and New Zealand anaesthesia mortality committees Chair/Co-ordinator Representing Dr David Pickford New South Wales Dr Jennifer (Jay) Bruce Western Australia Associate Professor Larry McNicol Victoria Dr Margaret Walker Tasmania Dr Simon Jenkins South Australia Dr James Troup Queensland Dr Carmel McInerney Australian Capital Territory Dr Leona Wilson New Zealand Other Interested Parties Dr Brian Spain Northern Territory Dr Phillipa Hore ANZCA Dr Peter Roessler ANZCA ANZCA Safety and Quality Co-ordinator Ms Karen Gordon-Clark Details on each jurisdiction, including (where available) terms of reference, legislative protection and information regarding coronial legislation, can be found in the State, Territory and National Information section, starting on page 15. 2 A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 Executive Summary Larry McNicol 1. This is the ninth triennial report of anaesthesia-related mortality 5. During the triennium, the number of anaesthesia-related in Australia (the first being for the triennium 1985-871-8). The deaths (categories one, two and three) reported from the format is similar to previous reports, and contains data from five states was 156. However, in only 22 cases were the five states (New South Wales, South Australia, Tasmania, deaths classified as category one (where it was it considered Victoria and Western Australia). The ANZCA Mortality Sub- ‘reasonably certain’ that death was caused by anaesthesia Committee has supported these states in their collection of factors alone). In 15 cases there was ‘some doubt’ (category data and encouraged the establishment or re-establishment two), and in the remaining 119 cases, ‘medical, surgical of anaesthetic mortality reporting in other Australian states and and anaesthetic’ factors were implicated (category three). territories and in New Zealand. The South Australian Mortality This demonstrates a continued reduction in the percentage Committee was re-established in 2010 and has been able to of category one deaths in recent triennial reports. In provide mortality data for 2009-11. 2003-05, category one deaths were 21 per cent of the total anaesthesia-related deaths, and this reduced to 15 per cent 2. While this report contains data from only five states, these in 2006-08 and 14 per cent in 2009-11. five states include more than 70 per cent of the population of Australia. The report is therefore likely to provide a 6. During the triennium, the combined population for the reasonable estimate of anaesthetic mortality across five states was about 17.3 million (Australian population Australia for this period. statistics)9. Using this figure, the anaesthesia-related mortality rate for these five states was 3.01 deaths per 3. The Australian Capital Territory (ACT), the Northern Territory million population per annum. This is slightly higher than the and Queensland, did not provide anaesthetic mortality data figure (2.79) for the four states (NSW, Tasmania, Vic and for this report because they did not have functioning WA) in the previous triennium (2006-08)7. It is, however, anaesthetic mortality committees during the 2009-11 very similar to the anaesthesia mortality rate per million triennium. However, the Queensland Perioperative and population per annum in all triennial reports since 1997-99. Peri-procedural Anaesthetic Mortality Review Committee (QPPAMRC) was re-established in 2012 and will be able to 7. During the triennium there were about 9.05 million individual provide mortality data for the next triennial report (2012-14). episodes of anaesthesia care in the five states. This figure The ACT Regional Committee of the Australian and New was obtained from the Australian Institute of Health and Zealand College of Anaesthetists (ANZCA) is working with the Welfare (AIHW)10. The AIHW data were obtained from coders ACT Audit of Surgical Mortality (Royal Australasian College of at all public and private hospitals. A coding hierarchy was Surgeons) to develop anaesthesia mortality reporting similar used to ensure only one anaesthesia item number was to the model established in Tasmania. It is possible the counted per episode of anaesthesia care10, 11. Using this South Australia committee could receive data from the denominator, the anaesthesia-related mortality rate was Northern Territory as has occurred previously. New Zealand 1:58,039 for the five states included in this report. This is has established a multi-disciplinary perioperative mortality similar to figure for the four states (NSW, Tasmania, Vic, WA) committee, and although this committee is unable to provide for the previous triennium (2006-08; 1:55,4907). specific anaesthesia mortality data using the Australian 8. The accuracy of the number of episodes of anaesthesia classification system, the report is very informative. care (the denominator) obtained from the AIHW is supported (See State, Territory and National Information, page 15.) by the relatively constant ratio between the number of 4. As with all anaesthesia mortality reporting, it should be episodes of anaesthesia care identified for each state and appreciated that classification of anaesthesia-related deaths the population of each of the five states. The ratio was relies on expert opinion or consensus, and therefore remains consistent across all five states (NSW 0.16, WA 0.17, subjective to some extent. It is also possible the state SA 0.18, Tas. 0.18, Vic. 0.18). mortality committees may vary in their interpretation of the categorisation of anaesthesia-related deaths. The methodology for identifying potential anaesthesia-related deaths is variable across the jurisdictions and therefore it must be recognised that some anaesthesia-related deaths may be missed despite the efforts made at individual, state and national levels. Nevertheless, due to the comprehensive processes in place in all five states reporting, it is unlikely that many cases were missed or classified incorrectly. A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 3 Executive Summary continued 9. The majority of anaesthesia-related deaths (84 per cent) 13. For the first time, this triennial report includes a brief clinical occurred in older patients (age over 60 years). Fifty-one summary of the causes of death in those classified as per cent of cases were female. It is of some interest category one (where it is ‘reasonably certain’ death was that 70 per cent of anaesthesia-related deaths occurred caused the anaesthesia or other factors under the control when surgery was either urgent or emergent. This is a of the anaesthetist). Of the 22 category one deaths, seven significant change from the previous report (2006-08) when were due to anaphylaxis, five involved management of the approximately one third were urgent or emergent. Only a very airway, five involved pulmonary aspiration, three deaths small proportion (7 per cent) occurred in patients considered involved cardiac arrest attributed to inappropriate choice low risk (ASA-P 1-2)12. Hence 93 per cent of anaesthesia- or application of anaesthesia technique and there were two related deaths occurred in patients assessed as higher risk fatal outcomes resulting from invasive cardiovascular (ASA-P 3-5). The types of surgery most frequently associated procedures. Of note; (i) anaphylaxis remains one of the less with anaesthesia-related death were orthopaedics (48 per preventable causes of anaesthesia-related deaths, but early cent), cardiothoracic (14 per cent), vascular (10 per cent) diagnosis and appropriate crisis management with escalating and abdominal surgery (10 per cent). An emerging trend is doses of adrenaline and aggressive fluid replacement are the increased frequency of anaesthesia-related deaths in paramount; (ii) in more than one of the airway-related gastro-intestinal endoscopy and interventional procedures deaths there was an inappropriate choice or application of in cardiology/radiology (10 per cent). Of note, some of these anaesthesia technique and inadequate monitoring; and one did not involve an anaesthetist at all. case involved a non-anaesthesia trained practitioner; (iii) in four of the five aspiration-related deaths, aspiration risk was 10. For the first time, information has been included regarding high and no airway protection was provided; in two of these the location of the event leading to death as well as the cases, no anaesthetist was involved; (iv) the three deaths location of death. The vast majority of fatal events, due to cardiac arrest all involved inadequate preoperative 96 per cent, occurred in the operating or procedure room assessment or management and inappropriate choice or (139 cases out of 145, excluding WA from which data was application of anaesthesia technique and were deemed to unavailable). The most common location of death was the have been preventable; and (v) both the deaths due to intensive care unit (39 per cent), followed by the operating or invasive procedures involved uncertainty about the procedure room (27 per cent), the general ward 19 per cent) anatomical position of the vascular access device. and the post-anaesthesia care unit (10 per cent). 14. Notwithstanding the effect of jurisdictional differences in 11. As in previous reports, the majority of deaths occurred in methodology for case reporting and classification, this report metropolitan teaching hospitals and larger regional teaching indicates that anaesthesia mortality rates in modern Australia hospitals (55 per cent), as would be expected with the acuity are low, whether assessed by the number of anaesthesia of the cases in these hospitals. By far the majority of deaths deaths per million population per annum (3.01) or by the (83 per cent) involved specialist anaesthetists (121/145 number of anaesthesia-related deaths per number of as data for WA was not available). Twelve cases involved non- anaesthesia procedures per annum (one in 58,021). The specialist/GPs, seven were anaesthesia trainees, and in at emerging pattern is that anaesthesia risk is now extremely least four cases there was no anaesthetist in attendance. low in patients who are basically fit and well (ASA-P 1 – 2). 12. An interesting trend over the past decade has been a However, most anaesthesia-related deaths occur in older, progressive reduction in the ratio of the number of anaesthetic sicker patients having non-elective surgery. Further causal or contributory factors per death. This was 2.42 in reductions in mortality may be achieved by reviewing the 2000-02, 1.58 in 2003-05, 1.30 in 2006-08 and 1.01 in timing of surgery to allow better optimisation of such 2009-11. Over the same period, there has been a progressive patients. Of course it is important to maintain the very high increase in the percentage of deaths in which the patient’s standards of anaesthesia training, enhanced by continuous chronic medical condition (H) was deemed to have contributed professional development using interactive workshops and to the death. This was 28 per cent in 2000-02, 58 per cent in simulation training in airway management, resuscitation and 2005-05, 72 per cent in 2006-08 and 81 per cent in other crisis management scenarios. However, the fact that 2009-11. These data are consistent with the likelihood that some deaths, such as those due to drug anaphylaxis, are there has been a progressive reduction in preventable currently deemed unpreventable re-enforces the ongoing anaesthesia-related mortality over this period, and that the need for research to develop better, safer alternatives. most important factor is the severity of the patient’s underlying medical condition (H). It is also noteworthy that the number Larry McNicol, FRCA FANZCA of deaths in which no correctable factor could be identified Editor (G) has also progressively increased, from 20 per cent in Chair, ANZCA Mortality Sub-Committee 2000-02, 33 per cent in 2003-05, 49 per cent in 2006-08 to 58 per cent in 2009-11. As in the previous report, these figures were heavily influenced by a large number of cases from NSW classified 3GH. This classification typically describes extremely high-risk patients, in which the stress of surgery and anaesthesia most likely contributed to or hastened death, but in which the death was assessed as non-preventable, other than by withholding the surgery and anaesthesia. 4 A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 Executive Summary continued Recommendations The ANZCA Mortality Sub-Committee makes the following recommendations: 1. The broader community should be informed that modern anaesthesia care is very safe as indicated by the very low anaesthetic mortality rates in Australia and the ongoing aim to avoid all anaesthesia-related deaths. 2. Patients, health authorities, anaesthetists, other medical specialists and healthcare workers should recognise the role of current anaesthesia training, research, accreditation, continuing professional development and education in achieving and maintaining the highest standards of safety and quality in the practice of anaesthesia in Australia and New Zealand. 3. The ANZCA Mortality Sub-Committee and the anaesthesia mortality committees in all states and territories of Australia should continue to work collaboratively to establish and maintain robust methodologies to obtain accurate anaesthesia mortality data. 4. Healthcare authorities should recognise that anaesthesia mortality is higher in older, sicker patients having major or urgent surgery and appropriate perioperative resources are required, including the provision of appropriate levels of specialist anaesthetist care, supervision and high dependency facilities (ANZCA professional document PS59: Statement on Roles in Anaesthesia and Perioperative Care). 5. There should be no option for any single operator proceduralist to administer sedation or anaesthesia (ANZCA professional document PS09: Guidelines on Sedation and/or Analgesia for Diagnostic and Interventional Medical, Dental or Surgical Procedures). 6. Many anaesthesia-related deaths are potentially preventable and these could be further reduced by continuous improvement in training, medical education and increased allocation of specialist anaesthesia resources. However, there remains a component of anaesthesia- related mortality deemed not preventable in the current state of knowledge. The pursuit of “no deaths attributable to anaesthesia” therefore requires further research into safer drugs and techniques. A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 5 Executive Summary continued Clinical aspects of category one anaesthesia-related procedure. Another death was attributed to loss of the airway deaths during anaesthesia with spontaneous respiration with a supraglottic airway, resulting in hypoxic cardiac arrest prior to For the first time, in this the ninth triennial anaesthesia mortality successful endotracheal intubation. report, we have included clinical information from the 22 deaths (category one) where it is reasonably certain that the death was Note: In more than one of these cases there was an caused by anaesthesia or other factors under the control of the inappropriate choice or application of anaesthesia technique and anaesthetist. The inclusion of this information was deemed inadequate monitoring; and one case involved a non- appropriate in order to highlight the major clinical issues involved anaesthesia trained practitioner. in the deaths directly related to anaesthesia and it is anticipated this has been achieved without compromise to confidentiality. Aspiration (five) There were seven deaths due to anaphylaxis, five involving There were five deaths due to pulmonary aspiration, four of management of the airway, five involving pulmonary aspiration, which occurred in the setting of endoscopy with an unprotected three deaths involving cardiac arrest attributed to inappropriate airway. One case involved a patient who 12 hours previously had choice or application of anaesthesia technique and two fatal been administered anaesthesia for gastroscopy, which was outcomes resulting from invasive cardiovascular procedures. abandoned due to limited mouth opening and failure to intubate. In the setting of ongoing bleeding, and in the absence of any Anaphylaxis (seven) anaesthetist, the endoscopist administered sedation for the There were seven deaths from anaphylaxis due to drugs repeat attempt gastroscopy. The patient had a cardiac arrest, administered by the anaesthetist. Five of them involved profound which was attributed to aspiration of blood, hypoxia, hypotension and cardiac arrest and in the other two the major hypovolaemia and underlying cardiac disease. There were three initial presentation was severe bronchoconstriction and other cases of aspiration in which high-risk upper gastrointestinal hypoxaemia, with subsequent cardiac arrest. In four of the endoscopy was performed under anaesthesia without protection cases, the trigger agent was a neuromuscular blocker of the airway. In an elderly frail patient with an incarcerated (atracurium, rocuronium x two, suxamethonium). In another umbilical hernia, an emergency physician trainee attempted to case, although atracurium had been given earlier, the likely reduce the hernia under intravenous sedation, but abandoned trigger was cephazolin. Of the remaining two cases, one involved the procedure due to apnoea and aspiration. administration of vecuronium as well as both cephazolin and Note: In four of the five cases, aspiration risk was high and gentamicin, and the other was due to ampicillin. In at least one no airway protection was provided. In two of these cases, of the cases, initial crisis management did not take account of no anaesthetist was involved. possible anaphylaxis and in five other cases, there were significant co-morbidities, which were likely to have contributed Cardiac arrest (three) to the failure of resuscitation. There were three deaths involving cardiac arrest resulting from Note: Anaphylaxis remains one of the less preventable causes of inappropriate choice or application of anaesthesia technique. anaesthesia-related deaths, but early diagnosis and appropriate There were two patients with multiple co-morbidities who crisis management with escalating doses of adrenaline and suffered cardiac arrest after induction of anaesthesia, both aggressive fluid replacement are paramount. of whom received excessive doses of induction agents. One of them was also scheduled for emergency surgery and was Airway related deaths (five) hypovolaemic. Another patient with severe cardiac disease There was one death in which the airway was lost during died during intravenous sedation/anaesthesia for a very minor maxillofacial surgery performed with the use of a submentally procedure that was either not required at all or could have been placed endotracheal tube. Airway obstruction from tube performed under local anaesthesia alone. malpositioning resulted in hypoxic cardiac arrest prior to the Note: These three deaths all involved inadequate preoperative difficult replacement with an oral endotracheal tube. Another assessment or management and inappropriate choice or death was attributed to airway obstruction, which occurred application of anaesthesia technique and were deemed to have immediately after extubation in a patient who had undergone been preventable. prolonged emergency surgery. The initial laryngscopy had been rated as grade 3 and endotracheal intubation involved the use Invasive procedure related deaths (two) of a bougie. Emergent repeat direct laryngoscopy noted oedema There was a death associated with the use of a pulmonary artery and bleeding and there were two unsuccessful attempts at catheter (PAC) used for monitoring during cardiac surgery. re-intubation, prior to hypoxic cardiac arrest. During CPR, Pulmonary artery rupture was attributed to uncertainty regarding endotracheal re-intubation was established via a blind technique the position of the PAC and inappropriate advancement. Another through a Fastrach LMA, but hypoxic encephalopathy ensued. death resulted from inadvertent misplacement of a central Another patient had a cardiac arrest (presumed to be due to venous device inserted to provide access for parenteral nutrition. hypoxia from airway obstruction) during spontaneous respiration There were issues associated with the type of catheter used and under intravenous anaesthesia for a minor procedure the monitoring of its position after insertion. administered by a medical practitioner without anaesthesia expertise. A patient with obstructing malignant pathology of the Note: Both these cases involved uncertainty about the upper airway died from the sequelae of barotrauma complicating anatomical position of the vascular access device. jet ventilation used during anaesthesia for the endoscopic 6 A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 Methods Data collection Confidentiality of information, an absolute requirement for all committees, was ensured by no primary data being examined in the compiling of the report. 1. State coroners’ acts, other public health legislation and the collection of data Information relating to the various coroners’ acts and jurisdictional public health legislation and state mortality committee terms of reference can be found in the State, Territory and National Information section, page 15. 2. Uniformity in analysing reports To uphold uniformity between the states in analysing reports, the chairs of the state-based mortality committees have continued to use the agreed Glossary of Terms – Case Classification form wherever possible. The use of this classification system was developed in March 2000, and has been in use in all states since 2006. – see Appendix 1 to view the form in its entirety. System of classification The system of classification and the term ‘death attributable to anaesthesia’ is defined in Table 1 and the report focuses on deaths in which anaesthesia played a part, that is, categories one, two and three. For the most part, the term ‘anaesthesia-attributable’ has been replaced with ‘anaesthesia-related’ in this and other reports. It should be noted that this classification system is also used to classify morbidity by the Victorian Consultative Council on Anaesthetic Mortality and Morbidity (VCCAMM). VCCAMM is the only state committee to collect data on morbidity as well as mortality. Table 1: System of classification by state-based anaesthesia mortality committees Death attributable to anaesthesia Category 1 Where it is reasonably certain that death was caused by the anaesthesia or other factors under the control of the anaesthetist. Category 2 Where there is some doubt whether death was entirely attributable to the anaesthesia or other factors under the control of the anaesthetist. Category 3 Where it is reasonably certain death was caused by both medical/surgical and anaesthesia factors. Explanatory notes: • T he intention of the classification is not to apportion blame in individual cases but to establish the contribution of the anaesthesia factors to the death. • T he above classification is applied regardless of the patient’s condition before the procedure. However, if it is considered that the medical condition makes a substantial contribution to the anaesthesia-related death, subcategory H should also be applied. • I f no factor under the control of the anaesthetist is identified which could or should have been done better, subcategory G should also be applied. Death in which anaesthesia played no part Category 4 Death where the administration of the anaesthesia is not contributory and surgical or other factors are implicated. Category 5 Inevitable death, which would have occurred irrespective of anaesthesia or surgical procedures. Category 6 Incidental death, which could not reasonably be expected to have been foreseen by those looking after the patient, was not related to the indication for surgery and was not due to factors under the control of the anaesthetist or surgeon. Unassessable death Category 7 Those that cannot be assessed despite considerable data, but where the information is conflicting or key data are missing. Category 8 Cases that cannot be assessed because of inadequate data. A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011 7 Findings for NSW, VIC, WA, Tas and SA Number of deaths classified The total number of deaths reviewed by the five states for the triennium was 1052, of which 156 were considered to be wholly or partly related to anaesthetic factors (categories one, two and three Table 2). Of the 1052 cases reviewed, 30 were classified ‘unassessable’ due to inadequate or conflicting data (category seven or eight, Table 1). Table 2: Number of deaths classified by each committee Total Category 1 Category 2 Category 3 Total classified anaesthesia-related NSW 647 11 8 106 125 Vic 61 7 0 11 18 WA* 271 4 6 1 11 Tas 55 0 0 0 0 SA 18 0 1 1 2 Total 1052 22 15 119 156 The disparity in total cases classified reflects both population differences and different requirements and methodology for reporting in different states (see State, Territory and National Information, page 15). The differences between states in relation to the various categories may represent some subjectivity in classification. There is less subjectivity, however, in relation to category one cases. *The WA data are based on deaths reported during the triennium. The data from the other states are based on deaths that occurred during the triennium. Number of anaesthesia-related deaths in relation to population Table 3: N umber of anaesthesia-related deaths during the 2009-11 triennium, in relation to the population* of NSW, Vic, WA, Tas and SA No. of deaths considered anaesthesia-related 156 Population of NSW, Vic, WA, Tas and SA. (17.3 million) No. of anaesthesia-related deaths per million population, 2009-11 9.02 No. of anaesthesia-related deaths per million population per annum 3.01 * E stimated resident population for 2010 (Australian Bureau of Statistics)9 [NSW 7.25, Vic 5.56, WA 2.30, Tas 0.51, SA 1.65 (x million)]. The ‘estimated resident population’ is considered more accurate than the Census figure. The number of anaesthesia-related deaths per million population at 3.01 was slightly greater than in the previous reports (2003-05)7, in which there were approximately 2.73 anaesthesia-related deaths per million population per annum and (2006-08)8 in which there were approximately 2.79 anaesthesia-related deaths per million population per annum. Table 4: Number of anaesthesia-related deaths in comparison with previous reports 1997-995 2000-026 2003-057 2006-088 2009-11 NSW 56 67 53 92 125 Vic 32 39 40 21 18 SA and NT 11 3 - - - SA - - - - 2 WA 11 16 19 7 11 Qld 20 12 - - - Tas - - - 4 0 Total 130 137 112 124 156 Table 5: Number of anaesthesia-related deaths in relation to population in comparison to previous reports 1997-99* 2000-02* 2003-05* 2006-08** 2009-11*** Population (x million) 13.40 13.75 13.68 14.80 17.30 Number of anaesthesia-related deaths 130 137 112 124 156 Anaesthesia-related death rate per million population per triennium 9.70 9.96 8.19 8.37 9.02 Anaesthesia-related death rate per million population per annum 3.23 3.32 2.73 2.79 3.01 Population source – Australian Bureau of Statistics *NSW, Vic, WA **NSW, Vic, WA, Tas ***NSW, Vic, WA, Tas, SA It is noteworthy that the anaesthesia-related death rate per million population per annum is very similar in all triennial reports since 1997-99. 8 A review of anaesthesia-related mortality reporting in Australia and New Zealand 2009-2011