Tenth report of the Victorian Consultative Council on ...docs2.health.vic.gov.au/docs/doc... · tenth report of the victorian consultative council on anaesthetic mortality and morbidity

consultative council on anaesthetic mortality and morbidity

Tenth report of the Victorian Consultative Council on Anaesthetic Mortality and Morbidity

consultative council on anaesthetic mortality and morbidity

Tenth report of the Victorian Consultative Council on Anaesthetic Mortality and Morbidity

Cases which occurred during 2003–2005

Edited by Associate Professor Larry McNicol mbbs, frca, fanzcaMay 2011

Consultative Council on Anaesthetic Mortality and Morbiditygpo Box 4923, Melbourne, Victoria, Australia 3001Website: http://www.health.vic.gov.au/vccamm/ Email: [email protected]

Accessibility

If you would like to receive this publication in an accessible format, please phone 9096 1381 using the National Relay Service 13 36 77 if required, or email: [email protected]

This document is also available in PDF format on the internet at: http://www.health.vic.gov.au/vccamm/

Published by the Victorian Consultative Council on Anaesthetic Mortality and Morbidity, Victorian Government, Department of Health, Melbourne, Victoria

© Copyright, State of Victoria, Department of Health, 2011

This publication is copyright, no part may be reproduced by any process except in accordance with the provisions of the Copyright Act 1968.

December 2011 (1104019) Print managed by Finsbury Green.

Acknowledgements

As this Council enters its 35th year of operation, it is appropriate to identify the factors that have enabled it to meet the requirements of its terms of reference over such a long period. The most important element has undoubtedly been the longstanding support of all Victorian anaesthetists, who continue to maintain such a strong culture of reporting anaesthesia-related mortality and morbidity. The widespread recognition of the inherent value of adverse event reporting, analysis, integration and dissemination in enhancing the safety of anaesthesia, remains an example for other medical specialities to follow. Therefore, on behalf of the current and all previous members of this Council, but more importantly, all Victorian health care consumers, I would like to acknowledge all anaesthetists who have so willingly submitted cases for review. The Council is also strongly supported by the Directors of Anaesthesia and quality assurance (QA) coordinators in Victorian hospitals and our thanks to all of them.

One of the reasons for this trust between the anaesthesia community and the Council has been the culture within the Council itself. There is, among all Council members, a sense of privilege to be appointed to this role; recognising the importance of preservation of confidentiality and due diligence in the process of deliberation of cases. The ability to sustain this trust is also linked to the fact that so many members of the Council have been prepared to contribute over long periods of time. Dr Patricia Mackay (Chairman 1991−2005) continues to serve as Emeritus

Consultant. She is recognised world-wide for her contribution to patient safety and was awarded the medal of Order of Australia (OAM) in 2008.

I would also like to acknowledge members of Council who retired from the Council since publication of the last report in October 2007. Dr Margaret Griggs (2001−2011) is a highly respected anaesthetist who has given a decade of service to the Council and Dr Barbara Robertson (2001−2008) has been a leading advocate for the work of Council in rural Victoria. We also recognise that one of the great strengths of the Council is through the input of specialists from other disciplines. Many thanks to A/Prof David Francis (2001−2008) who represented the Royal Australasian College of Surgeons and Dr Helen Opdam (2007−2009) who represented the Australian and New Zealand Intensive Care Society (ANZICS) and at that time, The Joint Faculty of Intensive Care Medicine (JFICM), now the College of Intensive Care Medicine (CICM).

In addition to the clinician representatives on Council, we were fortunate to have the insightful and sage contribution to the work and operation of Council from Dr Michael Ackland (1994−2010) as a medical representative of the Department of Health. Michael has been a major advocate for the quality assurance role of Council and at all times was able to provide high level advice to Council from a public health perspective. I would like to thank Michael for his outstanding contribution over such a long period.

The VCCAMM enjoys a very good and collaborative working relationship with the other two consultative councils, the Consultative Council on Obstetric and Paediatric Mortality and Morbidity (CCOPMM) and the Victorian Surgical Consultative Council. I would like to acknowledge the current and most recent Chairs: Prof Jeremy Oats, A/Prof James King, (CCOPMM), and Mr Peter Field and Mr Jonathan Rush, AM (VSCC). I thank them for their help and support.

It is important for the VCCAMM Chairman to be able to access cases from the Coroners Court of Victoria and over recent years, there have been some difficulties. I would therefore like to thank the State Coroner, Judge Jennifer Coate and her management team for their support and co-operation in facilitating a mechanism for timely access to the medical depositions referred to the coroner for all reportable deaths that have occurred in the setting of the administration of anaesthetic, sedative or analgesic drugs.

The Council has a very busy workload involving case collection, preparation and presentation to Council meetings, as well as database management and a wide range of other tasks. This is undertaken by the Confidential Project Officer and we are therefore especially indebted to Ms Pauline Berryman (2001−2006), Ms Zenaida Magtoto (2007−2010) and Ms Elena Alexandrova (2010), all of whom have also made a specific contribution to this report. Each of them has been very helpful to me personally and I thank them most sincerely for their support

and loyalty. I would also like to thank Mr Andrew Shelton, who was a seconded project officer (2006−2008) and provided specific expertise with data management.

Some years ago a model of secondment of Fellows to the Council was developed, pioneered by Dr Mathew Piercy (2004−2007) and more recently Dr Leona Leong (2010−2011), Dr Dana Pakrou (2010−2011) and Dr Louise Simpson (2010−2011). This is important in disseminating knowledge of the work of Council to new Fellows and we look forward to its further development. I thank all of them for their input to Council and specifically express my appreciation to Dr Leona Leong who also provided some editorial assistance with this report.

The Clinical Councils Unit is part of the Branch of Safety, Quality and Patient Experience, Department of Health, Victorian Government. I would like to personally thank the Director, Ms Alison McMillan and the Clinical Councils Unit Manager, Ms Anne-Maree Szauer for their ongoing support.

Finally, I express my profound appreciation to all the members of Council and in particular for their commitment to maintaining the highest standards of professionalism as a peer review group for Victorian anaesthetists.

Associate Professor Larry McNicol Chairman Consultative Council on Anaesthetic Mortality and Morbidity

Contents

Introduction 1

Terms of reference 1Definitions 1

Executive summary 3

Cases reviewed 3Anaesthesia mortality rate 3Causal or contributory factors in anaesthesia mortality and morbidity 3Value of sustainable morbidity reporting 5

Recommendations 7

Cases for 2003–2005 9

Reports of mortality and morbidity 9

Mortality 11

Total deaths reviewed 11Autopsy rate 11Anaesthesia mortality rate 12Causes of death 14Type of surgery 15

Anaesthesia-related deaths 17

Causal or contributory factors 17Age distribution of all cases of mortality 18General risk factors 19Specified medical risk factors 20Status of anaesthetist 21Location of event 21Location of death 22Type of hospital 22Type of anaesthesia 23

Morbidity 24

Reports of morbidity 2003–2005 24

Analysis of causes of anaesthesia-related mortality and morbidity 26

Preoperative assessment 28Airway-related events 28Aspiration 31Cardiac arrest 32Drug–related events 32Monitoring 35Myocardial infarction 37Neurological complications of central neural blockade 38Organisational problems 40Postoperative care and pain management 41Maternal mortality 43

References 45

Appendices 48

Appendix 1: Council membership 48Appendix 2: Operation of the council 50Appendix 3: Classification of cases reported to the council 55

List of tables and figures

Tables

Table 1: Total deaths reviewed and classified, 2003–2005 11Table 2: Primary cause of death, by classification, 2003–2005 14Table 3: Causal or contributory factors in anaesthesia-related mortality 2003–2005 17Table 4: Age distribution of all cases of mortality, 2003–2005 18Table 5: American Society of Anesthesiologists (ASA) physical status classification (P) 19Table 6: Risk in anaesthesia-related deaths, 2003–2005 19Table 7: Status of the anaesthetist in anaesthesia-related deaths, 2003–2005 21Table 8: Primary causes of morbidity, 2003-2005 24Table 9: Causes of anaesthesia-related mortality and morbidity, 2003-2005 26

Figures

Figure 1: Cases reported to the Council 2003–2005 9Figure 2: Sources of reports in 2003–2005 10Figure 3: Deaths by type of surgery, 2003–2005 15Figure 4: Medical co-morbidities and anaesthesia-related deaths, 2003–2005 20Figure 5: Location of event , 2003–2005 21Figure 6: Location of death, 2003–2005 22Figure 7: Type of hospital where the event occurred, 2003–2005 22Figure 8: Type of anaesthesia in anaesthesia-related deaths, 2003–2005 23

1

Introduction

Terms of reference

– To monitor, analyse and report on key areas of potentially preventable anaesthetic mortality and morbidity within the Victorian hospital system.

– To keep a register of anaesthetic mortality and morbidity within the Victorian hospital system.

– To liaise with other consultative councils on issues of common concern, including the development of appropriate systems for reporting of relevant cases by practitioners.

– To improve the practice of anaesthesia by publication and dissemination of relevant information and practical strategies identified during deliberations of the Council.

– To report as required to the Minister for Health and to the Victorian Quality Council.

– To respond to specific matters referred to the Council by the Minister for investigation and reporting, as required.

Definitions

While there is lack of uniformity in the definition of anaesthesia, the Council considers that modern anaesthesia encompasses the use of a sedative, analgesic, local or general anaesthetic drug, or any combination of these. Such agents may also be deployed in clinical settings outside the operating theatre or procedure room, and therefore Council also reviews events that may have occurred in the postoperative ward, intensive care and/or high dependency areas or during resuscitation in emergency departments or elsewhere.

Anaesthesia-related mortality

(1) A death that occurs during an operation or procedure, or within 24 hours of its completion, performed with the assistance of sedative, analgesic, local or general anaesthetic drugs, or any combination of these, or (2) a death that may be the result, either partly or totally, of an incident during or after such operation or procedure, even if more than 24 hours have elapsed.

The Victorian Consultative Council on Anaesthetic Mortality and Morbidity (the Council) was established in 1976 under section 13 of the Health Act 1958. The Council now operates pursuant to sections 33-43 of the Public Health and Wellbeing Act 2008.1

tenth report of the victorian consultative council on anaesthetic mortality and morbidity

2

Anaesthesia-related morbidity

Any event related to an anaesthetic procedure that causes a life-threatening incident, temporary or permanent disability, or significant distress. Morbidity is categorised as major or minor according to outcome. More information is available on the Council’s website with the suggested list for reporting to the Council.

Critical incident

Any incident that did or could, if not detected in time, affect patient safety. The Council does not specifically collect data on critical incidents, although a few such incidents are inevitably reported, and these are classified.

3

Executive summary

Cases reviewed

Council reviewed a total of 399 cases that occurred in 2003–2005, of which 157 were deaths, 219 were morbidity reports and 23 critical incidents. It is worth noting that the total number of cases reviewed in the previous triennium was 531, of which 238 were deaths. The difference in number of cases reviewed is due to limited access to coronial cases for deaths occurring during 2003–2005. Of the 157 deaths in 2003−2005, 40 were classified as anaesthesia-related, indicating once again that the Council undertakes review of many deaths in which anaesthesia has played no part. This workload however is important so that the Council is able to maintain confidence that anaesthesia factors are looked for in all deaths reviewed. It is of course possible that due to the reduced access to coronial cases, some anaesthesia-related deaths have not been identified.

Anaesthesia mortality rate

There were 40 anaesthesia–related deaths in Victoria in 2003–2005. Using the ICD-10 coding data from the Australian Institute of Health and Welfare, it is estimated that there were 2.35 million anaesthetics administered in Victoria in 2003–2005.2 Therefore the estimated anaesthesia related mortality is 1 in 58,664. This indicates a very high level of safety for anaesthesia in Victoria.

Causal or contributory

factors in anaesthesia

mortality and morbidity

Importantly, the most frequent causal or contributory factors associated with anaesthesia mortality were the patient factors of co-existing medical condition, and increasing age. This is consistent with recent literature and is unsurprising given the high proportion of older, sicker patients in the elective and emergency surgical population.3-7

However, there were 20 anaesthesia–related deaths in patients aged less than 70 years, and across all the mortality and morbidity cases there was a range of causal or contributory factors including inadequate and/or inappropriate preoperative assessment, as well as problems with anaesthesia drugs, techniques and/or management, including monitoring. There were also major problems identified with organisational issues and postoperative care, including pain management.

There was one case of malignant hyperthermia in a young patient who received suxamethonium and died despite very appropriate crisis management. Anaphylaxis to suxamethonium was responsible for three deaths and fatal outcomes are more likely if there is any delay in instituting aggressive resuscitation and/or there is co-existing coronary artery disease. There were also eight morbidity cases confirmed as anaphylaxis to muscle relaxants, of which four each were due to


4

suxamethonium and rocuronium. Cephazolin (five cases) and gelofusine (four cases) were the next most frequently implicated agents in anaphylaxis. Good outcomes were achieved due to early recognition and good crisis management, including the rapid escalation of adrenaline dosage in order to restore adequate arterial blood pressure and cardiac output. Drug errors (23 reports) persist as the single most frequently reported anaesthesia-related morbidity, despite the widespread promotion of prevention strategies.8

Aspiration was responsible for three deaths despite appropriate precautions being deployed in high-risk patients, but several morbidity cases were deemed preventable.

The prevalence of ischaemic heart disease is reflected in the surgical population in that there were eight anaesthesia-related deaths and six morbidity cases attributed to perioperative myocardial infarction. Importantly intra-operative hypotension was implicated in four mortality cases and inadequate preoperative assessment in three cases of non-fatal acute myocardial infarction (AMI). Failure to adequately account for pre-existing cardiac disease was also identified in two of the four mortality cases involving intra-operative cardiac arrest.

Of major concern were four deaths directly related to anaesthesia monitoring. Three patients died from pulmonary artery rupture associated with the use of pulmonary artery catheters (PAC) and there was one death from oesophageal perforation due to the

use of TOE. There were also eight morbidity cases involving oesophageal injury (four perforations and four tears) secondary to the use of TOE during cardiac surgery. It is recognised that these forms of sophisticated cardiovascular monitoring provide very valuable information to guide perioperative management, but it is imperative to deploy them based on an individual patient risk versus benefit analysis.

An increasingly implicated factor in both anaesthesia–related mortality (43%) and morbidity (17%) is organisational failure. This was due to one or more of: inadequate supervision, poor service provision, or failure of interdisciplinary perioperative care planning. Importantly, there were two airway-related deaths in ICU patients both of which involved organisational problems. There were also three airway-related morbidity cases involving inappropriate anaesthetic planning and management.

Problems with postoperative care continue to emerge as one of the major concerns, and these often also involve organisational issues. There were two deaths attributed to hypoxia in which there were deficiencies in the provision of appropriate postoperative care in the general ward. The delivery of safe postoperative pain management also continues to present challenges. There were eight major neurological complications involving actual or potential spinal cord injury from either epidural haematoma or abscess secondary to central neural blockade. In most cases, the outcome was contributed to by organisational failure

5

involving one or more of: (i) inadequate neurological observation of the patient, (ii) poor co-ordination of anticoagulant and/or anti-platelet therapy, and (iii) failure to urgently perform appropriate diagnostic imaging (MRI or CT myelography) to either diagnose or exclude spinal cord compression. Finally, of equal concern was the number of reports (almost certainly reflecting a more widespread problem) of postoperative respiratory depression, sometimes to the point of respiratory arrest or unresponsiveness, associated with postoperative opioid analgesia, in particular, patient controlled analgesia (PCA) with morphine. This was often also associated with organisational failure involving inadequacies in the provision of postoperative orders and/or documentation of patient observations, including sedation level, respiratory status and oxygenation monitoring.

Value of sustainable

morbidity reporting

This report reconfirms the important contribution to anaesthesia safety by the collection, analysis and dissemination of information obtained from morbidity reports. There were a total of 219 morbidity reports, of which 179 (82%) were anaesthesia–related, compared with 157 mortality reports of which 40 (25%) were anaesthesia-related. There is of course some reporting bias in that anaesthetists are more likely to report a morbidity

case deemed to be anaesthesia-related. However, and particularly in light of the extremely low anaesthesia mortality rate, the recognition that there is substantially more knowledge to be gained from detailed review of morbidity is widely appreciated by the Victorian anaesthesia community.


6

7

Recommendations

1. The Council should seek to reduce the delay in publication of annual and triennial reports, such that (i) each annual report should be published within twelve months of the end of the calendar year to which the report pertains, and, (ii) each triennial report should be published within three years of the end of the triennium to which the report pertains.

2. The Council should continue to seek acquisition of case reports of preventable anaesthesia–related adverse events (the numerator) through the following mechanisms: (i) direct voluntary reporting from individual practitioners through a web-based system, (ii) co-operation and support of hospital department of anaesthesia QA coordinators to promote reporting to Council of suitable cases discussed at departmental QA meetings, and (iii) improved access to coroner’s cases whereby the Council chairman is able to review all medical depositions which have involved deaths occurring in association with the administration of any sedative, analgesic, local or general anaesthetic agents.

3. The Council should maintain its membership of the ANZCA National Mortality Working Group in order to continue to share mortality data, obtain from the Australian Institute of Health and Welfare (AIHW) the most accurate information available on the number of anaesthetic administrations performed annually (the denominator) and work collaboratively in the promotion of anaesthesia safety through mortality and morbidity reporting.

4. The Council should continue to take a leadership role in the specific promotion of morbidity reporting in anaesthesia9 and work collaboratively with other agencies such as the Australian and New Zealand Tripartite Anaesthesia Data Committee (ANZTADC), the ANZCA Quality and Safety Committee, the Anaesthesia Safety Project (ASP) and the Victorian Health Incident Management System (VHIMS).

5. The Council should continue to pursue the collection, review and analysis of all anaesthesia–related adverse events (the numerator) and seek to obtain information on the number of anaesthetics (or particular drugs or techniques) performed (the denominator), in order to provide the most accurate assessment of the risk of anaesthesia.


8

6. The Council should continue to promote the importance of adverse event reporting in the improvement in safety and quality of clinical anaesthesia practice.

7. The Council recommends that all hospitals in Victoria should seek to implement the recently launched patient safety initiatives, namely:

(a) The World Health Organisation (WHO) Safe Surgery Check List,10

(b) The Postoperative Orders Project developed by the Victorian Surgical Consultative Council (VSCC),11 and

(c) The Pain Management Measurement Toolkit developed by the Victorian Quality Council (VQC).12

8. The Council recommends that all hospitals in Victoria should seek to improve the resource allocation and coordination of postoperative care, including higher numbers of high-dependency beds, and improved systems for patient surveillance and escalation of care on the general wards, particularly out of hours.

9

Cases for 2003–2005

Reports of mortality and morbidity

The Council reviewed 399 cases that comprised 157 deaths, 219 cases of morbidity and 23 critical incidents.

There was a 25 per cent reduction in the number of cases reviewed and classified during 2003−2005 compared with 2000−2002 triennium.13 The decrease is attributed mainly to the lower numbers of cases sourced from the State Coroner’s Office. The number of both direct and

quality assurance reports received by Council remained stable. This is a good indication of continued support from the Victorian anaesthesia community and QA coordinators from individual hospital departments.

figure 1: cases reported to the council 2003–2005

0

50

100

150

200

250

Critical incident

23

Mortality

157

Morbidity

219


10

The Council has three sources from which it obtains cases for review: voluntary reporting from medical practitioners (156 direct), referral of cases from QA coordinators of individual hospital anaesthesia departments (187 QA) and through liaison with the Coroners Court of Victoria (56 coronial). The significant decline in the number of coronial cases reviewed by Council could be attributed to reduced access to cases of potential interest from the State Coroners Office. System changes at the Coroners Court of Victoria resulted in the suspension of case lists referred to Council by the Clinical Liaison Service of the Victorian Institute of Forensic Medicine (VIFM) in 2005.

However, improved definitions of reportable deaths in the recent legislation of the Coroner’s Act 200814 and the direct support and co-operation of the State Coroner, Judge Jennifer Coate are enhancing access to cases.

figure 2: sources of reports in 2003–2005

0

25

50

75

100

125

150

175

200

QADirectCoronial

187

156

56

11

Mortality

Total deaths reviewed

The Council reviewed 157 deaths from 2003–2005. The number of deaths reviewed each year within the three-year period is listed in Table 1. Of the 157 deaths, 40 (25%) were identified as categories 1, 2 or 3 (wholly or partly related to anaesthesia). It is important to note that the remainder (41 surgical, 72 inevitable, four either fortuitous or un-assessable) comprise 75% of the mortality cases and therefore a considerable workload of case reviews is undertaken by the Council to ensure that any contribution by anaesthesia to the outcome is identified. The large number of inevitable deaths indicates that often anaesthesia is required in emergency situations of extremely high risk where it is frequently impossible to assess inevitability until after the event.

The classification system used by the Council is described in detail in Appendix 3. Categories 1, 2 and 3 are anaesthesia-related, 4 is surgical, 5 is inevitable, and 6–8 include fortuitous and unassessable cases.

Autopsy rate

It is important to have access to coronial autopsy or investigation reports to assist the Council in the classification of mortality cases. It is worth noting that such information was available in only 95 (60%) of the 157 mortality cases reviewed by Council in this 2003−2005 period. Of the 40 anaesthesia-related deaths, 27 cases (68%) were provided with either an autopsy report or an investigation report. An investigation report is provided in circumstances when no autopsy is performed. It is anticipated that subsequent to the review of the definition of ‘Reportable Deaths’ within the Coroner’s Act, there may be an opportunity to improve the rate of provision of autopsy and/or investigation reports made available to the Council.

table 1: total deaths reviewed and classified, 2003–2005

category 1 2 3 4 5 6–8 total

2003 8 3 8 16 34 1 70

2004 3 2 10 17 25 3 60

2005 1 1 4 8 13 0 27

Total 12 6 22 41 72 4 157


12

Anaesthesia mortality rate

The accuracy of data regarding the number of anaesthesia-related deaths (the numerator) is dependent on the ability of the Council to obtain information on all potentially anaesthesia-related deaths in Victoria. Under the current arrangements in which there is voluntary reporting of cases directly referred to Council, as well as some deficiencies in the automatic capture of potential anaesthesia-related deaths from the Coroners Court, it is possible that the number of anaesthesia-related deaths is more than 40 for this triennium (2003−2005). However, given the longstanding tradition by Victorian anaesthetists to report such cases, it is reasonable to assume that most deaths have been notified. Importantly also, there is now provision under sections 39−40 of the Public Health and Wellbeing Act 20081, for health services to provide requested information from the Chairman of the Council. It is also worth noting that the Coroners Court of Victoria has recently provided a link to the Council for cases in which the reportable death has occurred in association with the administration of a sedative, analgesic, local or general anaesthetic medication. These mechanisms are likely to improve the accuracy of the number of anaesthesia-related deaths. It is also important to note that the Council remains committed to review deaths (associated with the administration of an anaesthetic, sedative or analgesic drug) that may have occurred outside the domain of the anaesthetist in the operating theatre or procedure room, and this may

lead to an over estimate of ‘anaesthesia related death’, particularly if such cases involved intervention by other (non-anaesthetist) clinicians.

The issue of the frequency of anaesthesia-related deaths, or the anaesthesia mortality rate has historically involved debate about how it is best measured. This is also reviewed in the ANZCA publication Safety of anaesthesia in Australia: a review of anaesthesia mortality in Australia 2003–2005, which includes the 40 anaesthesia-related deaths from Victoria.15 The debate involves the source of data used to measure the denominator. The preferred denominator is the total number of anaesthesia administrations. This is now available using the ICD-10 coding from the Australian Institute of Health and Welfare2, which indicate that in Victoria there were estimated to be 2.35 million anaesthetics administered during 2003−2005. This provides an estimated anaesthesia-related mortality of one in 58,664. The collated data for New South Wales, Victoria and Western Australia combined yields a mortality of one in 53,426.

An alternative denominator is the total population. In the report Safety of anaesthesia in Australia: a review of anaesthesia mortality in Australia 2003–2005, the data for anaesthesia-related deaths and population for NSW, Victoria and WA are combined. There were a total of 112 anaesthesia-related deaths in an estimated total population of 13.68 million. This results

13

in a frequency of anaesthesia-related death as 2.73 per million population per annum.15 It is recognised that from an epidemiological perspective, the population denominator is appropriate. However, provided there is reasonable confidence in the estimation of the number of anaesthetics administered, this is a more useful denominator in the assessment of anaesthesia-related risk.

Therefore, this estimated anaesthesia-related mortality in Victoria of one death per 58,664 anaesthetics represents a very high level of anaesthesia safety. The challenge remains to pursue not only the most accurate data for the number of anaesthetics administered, but also to sustain and improve the mechanisms for reporting, analysing and classifying anaesthesia-related deaths.

This estimated anaesthesia-related mortality in Victoria of one death per 58,664 anaesthetics represents a very high level of anaesthesia safety.


14

Causes of death

The Council identified 27 primary causes of death among the 157 cases that were classified.

There were four cases classified as Categories 6−8.

table 2: primary cause of death, by classification, 2003–2005

primary cause of death cat 1 cat 2 cat 3 cat 4 cat 5

anaesthesia-related surgical inevitable

anaphylaxis 3 2

airway 1 1 1 1

aspiration 2 1

bowel infarction 2 4

cardiac arrest /cardiomyopathy 1 1 1

cardiac arrest/hypovolaemia 1

cardiac arrest/sepsis 1

cardiac failure 2 18

cardiac tamponade 1 2

cardiogenic shock 1 1 4

embolism (air) 2

embolism (amniotic fluid) 1 1

embolism (cement/fat) 4

embolism (pulmonary) 1 2

equipment 1

haemorrhage 2 16 14

hypotension 1 1

hypovolaemia 2

hypoxia 1 1 1 1 1

malignant hyperthermia 1

monitoring (pa catheter) 2 1

monitoring (toe) 1

myocardial infarction 1 3 4 2 12

myocardial ischaemia 1

neurological 1 1

pneumothorax 1

pulmonary hypertension 1

pulmonary oedema 1 1

respiratory failure 1 1

septic shock 2 2

stroke 4

trauma 3

Total 12 6 22 41 72

15

Type of surgery

Comments on mortality

and surgical procedures:

Unsurprisingly, the most common causes of inevitable death were cardiac failure, haemorrhage and myocardial infarction. It is recognised that the differentiation between inevitable and surgical-related deaths is often difficult, particularly in the setting of extremely high-risk emergency cardiac or major aortic surgery. Uncontrollable haemorrhage was the most common cause of surgical deaths and this occurred across the range of cardiothoracic, aortic and general surgery.

It is important to note that there were four surgical deaths due to cement and/or fat embolism during or after orthopaedic surgery.

It is imperative that anaesthetists remain aware of this risk during bone reaming and cement implantation, and respond rapidly to any haemodynamic or gas exchange problems which may indicate acute right ventricular (RV) dysfunction. Of the two surgical deaths due to air embolism, one occurred after an injury to the superior vena cava during surgical placement of a vascular access device and the other resulted from a prolonged and difficult gastroscopy in a patient with oesophageal ulceration and varices. In the latter case paradoxical embolism was demonstrated with TOE, and was attributed to large volume right to left intrapulmonary shunting due to end stage liver disease.

figure 3: deaths by type of surgery, 2003–2005

0 5 10 15 20 25 30 35

Other Procedural

Endoscopy

Vascular

Neurosurgery

Other Specialties

Thoracic

Aortic grafting

Cardiac surgery

Orthopaedic

General surgery

6

6

8

9

13

14

16

25

30

31


16

17

Anaesthesia-related deaths

Causal or contributory factors

There are often multiple factors that can be implicated in deaths attributable to anaesthesia. There were 93 contributory factors identified in the 40 cases of anaesthesia-related mortality, which represents an average of 2.3 per case.

table 3: causal or contributory factors in anaesthesia-related mortality 2003–2005

medical co-morbidities 27

organisational issues 17

drugs (including adverse drug reactions) 13

anaesthesia technique 10

anaesthesia management (crisis management, monitoring, resuscitation) 10

preoperative assessment and/or management (inadequate or inappropriate) 8

postoperative care (inadequate or inappropriate) 8


18

Age distribution of all cases of mortality

Notwithstanding a number of inevitable deaths in very young children with life threatening congenital disease, these data demonstrate the important association of increasing age and mortality. In all mortality categories, 108 (68%) were greater than 60 years of age. In the anaesthesia-related

deaths, 55% were aged over 60, and 25% were over 80 years of age. These data continue to support the published evidence that old age is a major risk factor for perioperative mortality.3-7

table 4: age distribution of all cases of mortality, 2003–2005

age category

1 2 3 1–3 total(%) 4 5 6–8 total (%)

0–9 - - - - 1 5 – 6 (3.8)

10–19 2 - - 2 (5) - 2 - 4 (2.5)

20–29 - 2 - 2 (5) 1 3 1 7 (4.4)

30–39 - - 3 3 (7.5) - 3 1 7 (4.4)

40–49 2 - 1 3 (7.5) 2 4 - 9 (5.7)

50–59 2 2 4 8 (20) 5 13 (8.2)

60–69 - - 2 2 (5) 5 10 17 (10.8)

70–79 4 - 6 10 (25) 14 17 - 41 (26.1)

80–89 2 2 6 10 (25) 12 17 2 41 (26.1)

90–100 – – – - 1 7 - 8 (5.1)

Total 12 6 22 40 (100) 41 72 4 157* (100)

19

General risk factors

Level of risk was assigned according to the physical status classification (P) of the American Society of Anesthesiologists (ASA).17

In categories 1, 2 and 3 (anaesthesia-related deaths), eleven cases were classified as ASA (P) 1 or 2. The remaining 29 cases had severe systemic disease with multiple

risk factors. Nineteen of the forty anaesthesia-related deaths were undergoing emergency procedures.

table 5: american society of anesthesiologists (asa) physical status classification (p)

asa or p-1 a normal healthy patient

asa or p-2 a patient with mild systemic disease

asa or p-3 a patient with severe systemic disease

asa or p-4 a patient with severe systemic disease that is a constant threat to life

asa or p-5 a moribund patient who is not expected to survive without the operation

asa or p-6 a brain dead patient whose organs are being removed for donor purposes

e patient requires emergency procedure

table 6: risk in anaesthesia-related deaths, 2003–2005

category 1 2 3 total %

asa (p) 1 1 1 1 3 7.5

asa (p) 2 3 1 4 8 20

asa (p) 3 6 1 11 18 45

asa (p) 4 2 3 6 11 27.5

asa (p) 5 – - - - -

Total 12 6 22 40 100

emergency 4 4 11 19 47.5


20

Specified medical risk factors

A total of 83 medical risk factors were identified in the 40 cases of anaesthesia mortality, yielding an average of approximately two co-morbidities per case. Cardiac disease remains the most important risk factor.

figure 4: medical co-morbidities and anaesthesia-related deaths, 2003–2005

0 5 10 15 20 25

Other

Obesity

Renal

Diabetes

Vascular

Gastrointestinal

Respiratory

Hypertension

Cardiac

11

3

4

5

6

10

11

13

20

Where other includes endocrine, musculoskeletal, and neurological conditions.

21

Status of anaesthetist

It is recognised that specialists are most likely to be involved in the anaesthesia care of more complex, high risk cases.

table 7: status of the anaesthetist in

anaesthesia-related deaths, 2003–2005

specialist 30

non-specialist/Gp 2

trainee 8

Location of event

Of the 40 deaths related to anaesthesia, the event leading to death occurred in the operating room during induction of anaesthesia in 15 cases, in the operating room during the surgical procedure in 13 cases, in the PACU in five cases, in the postoperative general ward in five cases, and in the ICU in two cases.

figure 5: location of event , 2003–2005

15 ORI = operating (or procedure) room during induction of anaesthesia

13 ORS = operating (or procedure) room during surgery

5 PACU = post anaesthesia care unit

5 POW = postoperative general ward

2 HDU/ICU = high dependency or intensive care unit


22

Location of death

Of the 40 deaths related to anaesthesia, 12 occurred within the operating room, one in the PACU, 20 within the ICU or high-dependency areas, and seven occurred in the postoperative general ward.

figure 6: location of death, 2003–2005

Type of hospital

The interpretation of the type of hospital data for anaesthesia-related deaths is difficult in the absence of information about the caseload denominator. More complex high-risk cases will be undertaken in tertiary teaching hospitals, but more detailed analysis regarding supervision level and type of surgery linked to type of hospital is not readily available. Similarly, it is difficult to comment on the casemix in private hospitals.

figure 7: type of hospital where the event

occurred, 2003–2005

12 OR 30% = operating (or procedure) room

1 PACU 2% = post anaesthesia care unit

7 POW 18% = postoperative general ward

20 HDU/ICU 49% = high dependency or intensive care unit

27 Metropolitan teaching

10 Private

2 Regional teaching

1 Rural

23

Type of anaesthesia

The majority of anaesthesia-related deaths (29) occurred in association with general anaesthesia, but it is worth noting that three deaths occurred in the setting of intravenous sedation.

figure 8: type of anaesthesia in

anaesthesia-related deaths, 2003–2005

29 General 72%

4 General & Regional 10%

3 Intravenous Sedation 8%

2 Regional 5%

2 Regional with Sedation 5%


24

Morbidity

Reports of morbidity 2003–2005

The Council considered 240 cases of morbidity of which 179 were classified as being related to anaesthesia. The vast majority of the morbidity cases listed

under categories 4−8 were surgical (cat 4). The Council also reviewed 23 critical incidents (cat 9).

table 8: primary causes of morbidity, 2003-2005

event category event type cat 1-3 cat 4-6 total

airway airway injury 2

failed intubation 3

loss of airway 4

obstruction 12 3

Airway total 21 3 24

cardiovascular arrhythmia 4 2

cardiac arrest 1

cardiac failure 1

embolism 6

haemorrhage 5

hypertension 1

hypotension 8 2

hypovolaemia 1

myocardial infarction 6 2

Cardiovascular total 21 18 39

drug-related anaphylaxis 22 2

other adverse reaction 3 2

drug error 23 1

drug sensitivity 1 2

Drug-related total 49 7 56

equipment-related 3 1 4

metabolic hypoglycaemia 1

mh syndromes 4

Metabolic total 5 1 6

25

event category event type cat 1-3 cat 4-6 total

neurological awareness 4

inadequate n/m reversal 1

neuropraxia 9

stroke 2

Neurological total 16 4 20

procedure-related complication from invasive

monitoring

1

neuraxial block minor

complication

10

neuraxial block minor sepsis 2

neuraxial block spinal cord injury 8

procedural error (wrong side

block)

2

seizure due to la regional block 3

toe-related oesophageal injury 8

Procedure-related total 34 1 35

respiratory aspiration 18

bronchospasm 1

hypoxia 1

pneumothorax 1 3

pulmonary oedema 4

respiratory depression 5

Respiratory total 30 3 33

Total 179 38 217


26

The Council identified a total of 219 cases in which anaesthesia was implicated in the outcome, comprised of 40 cases of

anaesthesia–related mortality and 179 cases of anaesthesia–related morbidity.

Analysis of causes of

anaesthesia-related mortality

and morbidity

table 9: causes of anaesthesia-related mortality and morbidity, 2003-2005

event category event type mortality morbidity total

airway airway injury 2 2

failed intubation 1 3 4

loss of airway 4 4

obstruction 1 12 13

Airway total 2 21 23

cardiovascular arrhythmia 4 4

cardiac arrest 4 1 5

cardiogenic shock 1 1

cardiomyopathy 1 1

embolism (amniotic fluid) 1 1

haemorrhage 2 2

hypotension 8 8

hypovolaemia 1 1

myocardial infarction 8 6 14

Cardiovascular total 16 21 37

drug-related anaphylaxis 5 22 27

other adverse reaction 3 3

drug error 23 23

drug sensitivity 1 1

Drug-related total 5 49 54

equipment capnograph 1 1

intrathecal pump 1 1

infusion pump 1 1

Equipment total 3 3

metabolic hypoglycaemia 1 1

mh syndromes 1 4 5

Metabolic total 1 5 6

27

event category event type mortality morbidity total

neurological awareness 4 4

inadequate n/m reversal 1 1

neuropraxia 9 9

stroke 4 2 6

Neurological total 4 16 20

procedure-related monitoring (arterial, pac,toe) 4 9 13

cnb minor complication 10 10

cnb mild sepsis 2 2

cnb spinal cord injury 8 8

procedural error 2 2

seizure from local anaesthetic 3 3

Procedure-related total 4 34 38

respiratory aspiration 3 18 21

bronchospasm 1 1

hypoxia 3 1 4

pneumothorax 1 1

pulmonary oedema 1 4 5

respiratory depression 5 5

respiratory failure 1 1

Respiratory total 8 30 38

Total 40 179 219

Where:

CNB = central neural blockade, MH = malignant hyperthermia, N/M = neuromuscular

PAC = pulmonary artery catheter, TOE = transoesophageal echocardiography

Table 9 demonstrates that the use of primary and secondary keywords allows identification of the types of adverse events associated with anaesthesia. When this is combined with detailed analysis of cases

using the Council’s classification system, the major clinical issues and contributory factors implicated in anaesthesia related mortality and morbidity can be reviewed.


28

Preoperative assessment

Careful preoperative assessment is a cornerstone of anaesthesia. Council identified eight mortality cases (20%) and 12 morbidity cases (7%) in which inadequate preoperative assessment and/or management contributed to the outcome. Cardiac disease is a major risk factor and failure to assess either underlying cardiac disease or acute cardiovascular status was noted in five mortality and six morbidity cases. These included not only ischaemic heart disease patients but also unrecognised cardiomyopathy and the potential effects

on pulmonary blood flow associated with anaesthesia for a patient with a mediastinal mass. It is also noteworthy that both underlying sepsis and apparently well-compensated hypovolaemia may occasionally be unmasked by anaesthesia. Inadequate airway assessment was contributory in at least three morbidity cases. Failure to account for the implications of obesity and chronic respiratory status also contributed to both mortality and morbidity. There were two procedural errors involving wrong side regional blocks and one case of aspiration in which non-fasting status was not identified. The rollout of the new World Health

Organisation Safe Surgery Checklist,10,17 including the ‘Sign in’ component prior to administration of anaesthesia, should help to further reduce such preventable events.

Airway-related events

Mortality

There were two airway-related deaths, both of which occurred in intensive care patients. These have been included with anaesthesia-related mortality despite the

fact that anaesthetists were not involved. One case involved unrecognised oesophageal intubation and hypoxic cardiac arrest. There was failure to interpret available information from capnography and organisational problems related to the level of medical supervision. The other case involved a patient who had aspirated during general anaesthesia with a laryngeal mask airway (LMA) and was recovering from respiratory failure when she developed acute obstruction of a tracheostomy tube with resultant hypoxia. Both cases demonstrate the mandatory requirement for strict protocols for airway management

Cardiac disease is a major risk factor and failure to assess either underlying cardiac disease or acute cardiovascular status was noted in five mortality and six morbidity cases.

29

in the high-dependency and intensive care environment, including improved deployment and application of both oximetry and capnography. The special care requirements of all patients with a tracheostomy means that such protocols also need to be used in general wards that look after these patients.

Morbidity

There are 21 airway-related events in the morbidity reports. Twelve of these cases involved airway obstruction. There were two cases in which patients had undergone prolonged laparoscopic colectomy and in the PACU developed severe obstruction due to glottic oedema. Another similar event occurred after an anterior cervical fusion. It is therefore important to take account of the effects of surgery and posture when planning postoperative airway management. Another case involved a patient in whom there was unexpected difficult intubation. The patient had not been informed by the previous anaesthetist about the likely need for fibre-optic intubation, and intubation was difficult, necessitating the use of a fastrach™ LMA.

The patient required emergency re-intubation in recovery due to a near cardiac arrest which was thought to be due to hyperinflation resulting from attempted bag and mask ventilation through an obstructed airway. A second intubation attempt proved even more difficult due to airway swelling. Nasal fibreoptic endotracheal intubation was

successful. An elective tracheostomy was performed in ICU 48 hours later.

An additional case involved postoperative airway obstruction due to oedema after a grade 4 laryngoscopy and difficult intubation. The patient required an emergency surgical airway in recovery. It is therefore important to consider the need for prolonged airway support in patients who have undergone a difficult intubation and may have had prolonged airway manipulation and instrumentation. It is recommended practice to notify both patients and their general practitioner regarding potential airway problems or difficult intubation. Some anaesthetists advise patients to obtain a medical alert bracelet.

There should be a mandatory requirement for strict protocols for airway management in the high-dependency and intensive care environment, including improved deployment and application of both oximetry and capnography.


30

There were three cases involving inappropriate anaesthetic planning and management. A patient with ankylosing spondylitis underwent a colonoscopy and became hypoxic during the procedure as a result of the loss of the airway. There was an inappropriate and unsuccessful attempt at cricothyroidotomy prior to LMA insertion which proved to be successful in ventilating the patient, who also developed aspiration and pulmonary collapse. In another case, the anaesthetist chose to perform a modified rapid sequence induction but avoided the use of relaxants due to concerns about histamine release. This resulted in a difficult intubation which was easily preventable. There was another report involving inappropriate airway management for a patient with features of predicted difficult bag and mask ventilation and intubation who was scheduled for general anaesthesia in the remote environment of a coronary care unit where there was no anaesthesia machine, monitoring or support facilities.

There were two cases of airway injury. One resulted from intubation by an intensive care registrar of a patient who required an invasive radiological procedure for gastro-intestinal bleeding. There was inadvertent endobronchial placement of the tube and tracheal rupture was diagnosed after demonstration of pneumothorax and pneumomediastinum. Case review revealed that as well as inadvertent endobronchial intubation the cuff of the endotracheal tube had been over-inflated. The patient required

a thoracotomy for repair of the posterior tracheal perforation. A second airway injury occurred in a female patient with multiple trauma, who developed ulceration of the tracheal wall after five days of prolonged intubation with an 8.5 mm endotracheal tube.

There were several cases of temporary loss of the airway during general anaesthesia. Inadvertent extubation occurred in one paediatric case during dental surgery. In another case, the patient underwent planned but difficult awake fibreoptic intubation prior to relaxant general anaesthesia which was complicated by inadvertent extubation during suction. The case was well-managed with suction and manual ventilation until reversal of neuromuscular blockade and cancellation of the surgery.

31

Mortality

There were three deaths attributed to pulmonary aspiration, which therefore continues to be a major risk in modern anaesthesia practice. All three deaths were in patients at high risk of aspiration, which occurred despite appropriate precautions including rapid sequence induction. One of these was in an elderly patient with a history of previous aspiration at the time of an emergency surgical repair of a volvulus of a hiatus hernia. On this occasion the bowel obstruction was due to a sigmoid volvulus, and during induction of anaesthesia there was regurgitation and immediate signs of significant aspiration pneumonitis.

The patient developed adult respiratory distress syndrome and died approximately one month after surgery.

Another patient presented with gastro-intestinal bleeding and an initial gastroscopy performed under sedation with no anaesthetist present, demonstrated gastric outlet obstruction. The procedure was abandoned due to the presence of residual stomach contents. A request was made for anaesthesia assistance in order to protect the airway, and upon induction, massive regurgitation occurred. The patient was

rapidly intubated and suction applied to the airway, but immediately progressed to cardiac arrest and was unable to be resuscitated.

In a third case, the patient deteriorated in recovery after a Hartman’s procedure for a large bowel obstruction due to carcinoma. Re-operation was undertaken and the patient regurgitated faeculent fluid during a rapid sequence induction, developed aspiration pneumonia and systemic sepsis, and died the next day. In such cases, it remains unclear whether insertion of a nasogastric tube to facilitate gastric drainage before induction of anaesthesia reduces this risk.

Morbidity

There were 18 morbidity reports of confirmed or suspected aspiration, most of which involved only minor morbidity. However, there were some patients who required either ICU admission or increased hospital length of stay. Ten cases were associated with the use of an LMA, and in more than half of them aspiration was attributed to either inadequate depth of anaesthesia or delayed gastric emptying due to trauma. There were three reports of aspiration during sedation or intravenous anaesthesia for colonoscopy, and in one of these the patient’s non-fasting status was not identified.

There were three deaths attributed to pulmonary aspiration, which therefore continues to be a major risk in modern anaesthesia practice.

Aspiration


32

It is important to note that there were a number of patients who had a cardiac arrest, but in several cases this was attributed to another primary cause such as anaphylaxis, hypoxia or acute myocardial infarction. These are reviewed elsewhere. There were, however, five cases in which intra-operative cardiac arrest was deemed to be the primary event, and four were fatal. Two of these involved underlying cardiomyopathy, one of which was unrecognised and resulted in delayed and inadequate crisis management and resuscitation. The other was in a patient with documented ischaemic cardiomyopathy and severe depression who was unable to be resuscitated after a cardiac arrest during anaesthesia for electroconvulsive therapy. There were two cases of cardiac arrest in postpartum obstetric patients, in which clear aetiology was not established, but underlying sepsis and compensated hypovolaemia were possibly implicated.

Drug–related events

There were 54 drug-related events, including 27 cases of anaphylaxis, of which five were fatal, and 23 morbidity reports of drug administration errors.

Anaphylaxis

Of the five mortalities due to anaphylaxis, three were associated with the use of suxamethonium. In one case, the patient, who was an obese smoker, required only minor surgery which would have been very suitable for local or regional anaesthesia. Despite the recommendation for a regional block by the anaesthetist, the surgeon

requested general anaesthesia. After failure to maintain a satisfactory airway with a laryngeal mask, suxamethonium was administered and endotracheal intubation performed. There was an immediate and profound collapse progressing rapidly to cardiac arrest. Resuscitation was prolonged and difficult and required large doses of adrenaline. Despite subsequent return of the circulation, the patient failed to recover neurologically. Serum tryptase was extremely high and there was a diffuse red rash.

In another case, after placement of an epidural catheter and induction of anaesthesia using suxamethonium followed by atracurium, the patient was initially reasonably stable but developed severe hypotension and cardiac arrest after the insertion of a chlorhexidine impregnated central venous catheter. Resuscitation was unsuccessful despite large doses of adrenaline. Autopsy revealed a high tryptase level and underlying coronary artery disease with ventricular hypertrophy. The third case involved a patient with gastro-oesophageal reflux, who developed a florid rash, bronchospasm and cardiac arrest immediately after receiving suxamethonium. Resuscitation was unsuccessful despite aggressive and prolonged treatment, including high dose adrenaline and noradrenaline. Autopsy revealed airway oedema, significant coronary artery disease and a high tryptase.

The other deaths were due to severe reactions to protamine and a re-exposure to aprotinin. In the latter case, the patient, who presented for urgent re-do thoracic

Cardiac arrest

33

aortic surgery, had received aprotinin only six weeks previously and suffered profound hypotension upon re-exposure to a test dose. Adrenaline was immediately administered and cardiopulmonary bypass was rapidly established. Surgery was prolonged and difficult and there was a significant requirement for vasopressor support attributed to both sepsis and anaphylaxis.

Postoperatively the haemodynamic status was satisfactory but the patient did not recover neurologically and had multiple haemorrhagic and non- haemorrhagic cerebral infarcts.

Of the 22 morbidity cases attributed to anaphylaxis, eight were due to muscle relaxants (four suxamethonium, four rocuronium). There were five reports of reactions to cephazolin, two to penicillin, four to gelofusine and others included haemaccel, methylene blue, intravenous contrast and another re-exposure to aprotinin. Despite the utility of obtaining blood samples for measurement of mast cell tryptase and the importance of follow-up skin testing, the most important diagnostic criterion for anaphylaxis is the clinical presentation.18 Hypotension, unresponsive to conventional vasopressors,

bronchoconstriction and rash are the most frequent manifestations, and successful outcomes rely on early recognition. Treatment must include immediate cardiopulmonary resuscitation and rapid escalation of adrenaline dosage to restore blood pressure and cardiac output. Aggressive fluid administration is also important.

Drug errors

Council received 23 reports relating to drug administration errors. There were seven reports involving inadvertent administration of muscle relaxants instead of either fentanyl or midazolam either before or during induction of anaesthesia. There were two reports of residual suxamethonium being flushed through the side-port of intravenous (IV) lines of patients who were in the PACU. Other cases included ephedrine instead of dexamethasone and morphine, metaraminol instead of heparin and ephedrine and phentolamine instead of metaraminol.

There were four reports involving inadvertent administration of drugs or fluids between intravenous lines and epidural catheters. Two involved nurses attaching epidural infusion solutions to IV lines. In another case, 5 ml of potassium chloride,

Treatment must include immediate cardiopulmonary resuscitation and rapid escalation of adrenaline dosage to restore blood pressure and cardiac output.


34

mistaken for a lignocaine top-up, was injected epidurally by the anaesthetist, resulting in severe back pain, transient paralysis and hypertension. In another case, metaraminol was injected into an epidural catheter by an anaesthetist. Despite previous efforts by this Council and others, it has not yet been possible to introduce a mandatory requirement that epidural infusion delivery devices be made to be incompatible with the connections for intravenous infusion lines. This is now an urgent issue.

The Council recognises that it is very likely that the actual number of drug administration errors is much higher than that identified though these 23 reports. It remains important for all anaesthetists, anaesthesia assistants and perioperative nurses to recognise this high risk and adopt multiple strategies to reduce it. Therefore it is appropriate to reinforce the suggestions outlined below and to also note the recent ANZCA document Guidelines for the Safe Administration of Injectable Drugs in Anaesthesia.8

The Council’s suggestions are to:

– be constantly vigilant in recognising the human factors involved, including fatigue, distraction, change of anaesthetist, and the added hazard of two anaesthetists (usually consultant and trainee)

– carefully check labels on ampoules including the name of the drug, the dose and concentration, and the expiry date. Be aware that ampoule size, shape and clarity of label can change. Pharmacy departments should be instructed to advise anaesthesia department directors, anaesthesia nurse unit managers and operating theatre administrative staff, whenever there is a change of drug brand or presentation so warnings can be promulgated

– never administer drugs that you have not made up or personally checked. If the level of assistance is adequate, consider double-checking with a nurse assistant or another anaesthetist. However, only one individual should be responsible for the administration

– label all syringes clearly with universal labels according to the Australian Standard. Such labels are distributed by a local supplier.19 It is important to use the recommended brand of labels to promote safer practice across different hospital sites. A marking pen should also be available for backup. Always read the label before injecting

– standardise syringe sizes for particular drugs. Always use dedicated red-barrelled 5 ml syringes for all muscle relaxants. For opioids, 10 ml syringes are suggested, and 20 ml syringes are suggested for intravenous anaesthetic agents. However, these recommendations may not always be applicable for paediatric patients

35

– keep ancillary drugs such as antibiotics, vasoactive drugs, anticoagulants and local anaesthetic agents labelled and on separate trays away from the agents used for general anaesthesia

– double-check any drugs employed in an anaesthetic holding area immediately before use, and use them sparingly. Never take relaxant drugs to a holding area.

Optimal strategies include bar coding, pre-loaded syringes, colour-coded infusions and dedicated connectors for all infusions. While advisable, these may not be universally possible in the short term. However, strict adherence to the above principles, as well as constant vigilance, should minimise the risk of administering the wrong drug.

Monitoring

The occurrence of any anaesthesia–related mortality or major morbidity attributable to monitoring techniques requires very careful analysis with respect to the principle of ‘prime non nocere’. There were three deaths resulting from pulmonary artery rupture from a PAC and one fatal oesophageal perforation from the intra-operative use of TOE.

There was a case in which a PAC was deployed for off-pump coronary artery bypass grafting and on arrival in the intensive care unit, the catheter was noted to be over-wedged and had to be withdrawn three times to obtain a satisfactory PA waveform. Immediately thereafter, the patient developed profuse bleeding from the endotracheal tube and urgent bronchoscopy showed massive bleeding which was greater on the right side. Heroic attempts to control the bleeding were unsuccessful.

Another case involved an elderly female patient who underwent coronary artery bypass grafts (CABG’s) and aortic valve replacement (AVR). Monitoring included a PAC and the pulmonary artery pressures were not elevated. After four hours of cardiopulmonary bypass (CPB), there was bleeding from the endotracheal tube. The PAC had already been withdrawn approximately 2 cm at the commencement of CPB, and was withdrawn a further 10 cm. Attempts to isolate the bleeding with a bronchial blocker were partially successful initially but pulmonary haemorrhage continued into the postoperative period. Subsequent haemostasis was achieved but gas exchange was poor and the patient required extracorporeal membrane oxygenation. She subsequently died

There were three deaths resulting from pulmonary artery rupture from a PAC and one fatal oesophageal perforation from the intra-operative use of TOE.


36

due to acute myocardial infarction and pulmonary haemorrhage.

The third mortality associated with the use of a PAC involved another elderly female patient who underwent CABG’s and tricuspid annuloplasty. The postoperative CXR showed the PAC to be coiled in either the right ventricle or the inferior vena cava. The PAC was then withdrawn, and repositioned until a wedge pressure tracing was obtained at 57 cm. Thereafter, profuse bleeding was noted in the endotracheal tube and the right chest drain. The patient had a cardiac arrest and was unable to be resuscitated.

The fourth mortality associated with monitoring in cardiac surgery involved an elderly female who underwent elective CABG’s during which a TOE probe was deployed. On the second postoperative day, she had abdominal pain and investigations revealed a perforation at the gastro-oesophageal junction. Despite laparotomy, she died from sepsis three days postoperatively.

There were a further eight morbidity reports of oesophageal injury attributed to the use of a TOE probe: four perforations and four mucosal tears, one of which also had a separate haematoma. One female patient presented with ingested fluid in the mediastinal drain on day one after elective CABG’s. Thoracic CT demonstrated oesophageal perforation, which was treated by surgical repair at thoracotomy. An elderly female patient with a history of peptic ulceration underwent AVR/CABG’s and

developed sepsis and respiratory failure on day four postoperatively. She had an empyaema due to oesophageal perforation and made a slow recovery after prolonged chest drainage and antibiotics.

A third female patient who required semi-urgent CABG’s after an acute myocardial infarction presented with nasogastric feed in the mediastinal drain on day two. She underwent surgical repair of the oesophagus through a right thoracotomy and fully recovered. An elderly male patient with a thoracic aortic dissection requiring prolonged CPB and deep hypothermic circulatory arrest had nasogastric bleeding in the early postoperative period. At gastroscopy, a pouch was observed in the lower oesophagus but no tear or perforation. A subsequent gastrograffin swallow reported a para-oesophageal hiatus hernia but no leakage was found. The patient remained moderately unwell requiring two further ICU admissions. On day 22, repeat gastrograffin swallow confirmed an oesophageal perforation which was repaired via a left thoracotomy. The patient had a very prolonged recovery and was eventually discharged to a rehabilitation facility after more than ten weeks. The four cases of mucosal tears of the oesophagus had much less morbidity, presenting with mostly minor gastro-intestinal bleeding and a fall in haemoglobin.

It is important to note that several of these cases were identified through a targeted review of cases between 2001 and 2007 from

37

the database of the Australian Society of Cardiac and Thoracic Surgeons. The search involved identification of any patients who had undergone either upper G/I endoscopy, laparotomy or thoracotomy in the period after cardiac surgery. Any cases captured were reviewed and referred to Council if an oesophageal injury was found. An estimation of the number of cardiac surgical procedures was made from the same database in order to calculate the incidence of oesophageal injury associated with the intra-operative use of TOE. This survey identified an incidence

of oesophageal injury of 9 per 10,000, with a higher relative risk in elderly females.20 No such analysis of the frequency of pulmonary artery perforation from the use of PAC’s has been undertaken, but it is likely the incidence is extremely low. However, the three mortality cases demonstrate the catastrophic consequences of such an event. This detailed review of major complications associated with either PAC or TOE monitoring in cardiac surgery indicates that although such complications are rare, it is imperative that the decision to deploy these monitoring techniques must take account of the risk versus benefit in each patient. Clearly the information obtained from both TOE and

PAC has substantial benefits for most patients undergoing cardiac surgery, but neither of them should be considered routine.

Myocardial infarction

Given the prevalence of coronary artery disease in patients presenting for all types of surgery, strategies to minimise the risk of life-threatening cardiac complications, including myocardial infarction, have been widely deployed. Cardiac assessment and

risk evaluation has evolved over many years and there are regular updates to the guidelines that have been developed by the American College of Cardiology and the American Heart Association.21 There were 14 cases of perioperative myocardial infarction, including eight deaths. Preservation of adequate myocardial oxygen balance is essential in patients with coronary artery disease. There were four deaths due to acute myocardial infarction in which there was intra-operative hypotension resulting from induction of anaesthesia. It is important to recognise this risk, particularly in the elderly, in which the combination of preoperative vasodilator medication, the fasting state and

It is important to recognise this risk, particularly in the elderly, in which the combination of preoperative vasodilator medication, the fasting state and intravenous anaesthesia is very likely to cause hypotension.


38

intravenous anaesthesia is very likely to cause hypotension. There were also four cases in which there was documented ischaemic heart disease and fatal myocardial infarction occurred despite risk assessment, counselling and appropriate perioperative care planning. Transient hypotension however was also implicated in one of these deaths. There were six cases of non fatal AMI, and in at least three of these there was inadequate preoperative assessment. One case involved a failure to elicit a history of unstable angina prior to an elective colonoscopy, and two others involved failure of interdisciplinary planning during preoperative consultation in known high-risk patients.

Neurological complications

of central neural blockade

Council received 12 reports of minor morbidity associated with central neural blockade and these included inadvertent high block, post dural puncture headache, transient pain or sensory disturbance and mild sepsis.

However the major morbidity concern with either epidural or spinal anaesthesia and analgesia relates to the risk of paraplegia associated with spinal cord damage secondary to either epidural haematoma or abscess. This potentially catastrophic complication occurred in eight cases reported to Council during this triennium.

One case involved a patient who had an uncomplicated insertion of an L1/2 epidural catheter for a total knee replacement. In the

PACU she was pain free and able to move her feet. Two days later, and eight hours after the epidural infusion had been ceased, the patient was unable to move her legs. An urgent MRI revealed an epidural haematoma from L1 to L4, and at laminectomy a well-formed clot was removed. The patient only made a partial recovery, with considerable loss of motor function. There were organisational issues identified in relation to the timing of the first postoperative dose of low molecular weight heparin (LMWH) and postoperative neurological observations.

There were two patients in whom epidural/GA combination was used for thoracoscopic pleurodesis. In one of these, the patient had been on warfarin for atrial fibrillation. This had presumably been ceased before surgery but was recommenced on day one postoperatively after removal of the epidural catheter. Despite some perineal sensory loss the patient was discharged and presented three weeks later with back pain, progressive leg weakness and sensory loss. MRI showed an extensive epidural haematoma from T5 to L1 and after decompression there was progressive improvement in neurological function.

The second pleurodesis case involved a young patient who underwent the procedure for recurrent pneumothorax. An epidural catheter was in place for four days postoperatively and he was discharged home despite complaining of back pain. Upon return the next day, an MRI was negative. A further two days later, he re-presented with fever, leg weakness and back pain and a repeat MRI

39

demonstrated an epidural abscess from T1 to T10, with cord compression from T5 to T8. Thoracic laminectomy was performed and the patient’s motor function improved rapidly but back pain and urinary retention persisted and required prolonged treatment with antibiotics. Again there were organisational problems in the postoperative care of both these cases but it is also reasonable to question the appropriateness of the choice of epidural analgesia for thoracoscopic procedures.

Another patient underwent combined spinal/epidural analgesia for a total hip replacement. She had received the first dose of LMWH approximately 90 minutes after insertion of the epidural catheter and it was removed on day three, approximately 10½ hours after the last dose of LMWH. 90 minutes later, she developed lower limb paralysis. Urgent MRI revealed an epidural haematoma from T11 to L4 and a decompressive laminectomy was performed immediately. The patient made a slow but progressive recovery of motor function.

A patient who had an epidural inserted uneventfully at T9/10 for radical prostatectomy, had satisfactory postoperative analgesia. The epidural catheter was removed on day three and the patient was discharged the next day. Two days later, he re-presented to his local doctor with back pain, fever and local swelling at the epidural insertion site. Antibiotics were commenced after swabs were taken and on each of the next two days, the GP incised the abscess and then referred the patient back to the emergency department. The patient’s only

symptom was back pain, and an MRI showed an epidural abscess from T8 to T12, with evidence of anterior cord compression. After neurosurgical consultation, treatment was continued with antibiotics and there were no neurological sequelae.

An elderly female patient underwent combined spinal/epidural anaesthesia/analgesia for peripheral vascular surgery. After initial intra-thecal injection, there were some difficulties placing the epidural catheter, and a second attempt via a paramedian approach was successful. Postoperatively the epidural catheter was removed after 30 hours. After three days the patient was unable to void or walk and despite this, there was no contact with the anaesthetist until a further five days later. An MRI demonstrated a large epidural haematoma. The patient underwent an immediate laminectomy but there was no improvement and she remained paraplegic.

Another elderly female patient had spinal anaesthesia for internal fixation of a fractured neck of femur. Clopidogrel had been ceased five days previously but was recommenced postoperatively. She also received 70 mg b.d of enoxaparin. Approximately 24 hours postoperatively, she developed leg numbness and weakness. An MRI showed an extensive lumbar epidural haematoma with both spinal cord and cauda equina compression. The patient also had an acute on chronic brain syndrome and a brain CT revealed a subarachnoid haemorrhage. Palliative care was commenced and the patient died four days later.


40

An elderly male patient with multiple co-morbidities had an epidural/GA for a subtotal colectomy. The epidural insertion was difficult but there were no immediate complications. On day three postoperatively he developed leg weakness but a CT myelogram was not performed until day six. Some of this delay involved uncertainty about the appropriate investigation, as MRI was contra-indicated due to the patient’s thoracic aortic stent graft. The CT myelogram showed an epidural haematoma from T9 to T12. He was deemed unfit for surgical intervention, showed some progressive recovery of motor function, and died some weeks later from surgical complications.

These cases all demonstrate that there are major risks associated with the provision of central neural blockade and in particular, epidural analgesia. Most importantly, such adverse outcomes are often attributable to organisational failure in this setting of postoperative pain management, including inadequate neurological observations, lack of coordination of perioperative anticoagulation or anti-platelet therapy and delays in the initiation of appropriate diagnostic imaging for epidural haematoma or abscess. It is essential that all staff responsible for the care of such patients be made aware of the importance of the

new onset of weakness, back pain, sensory disturbance and fever. Patients undergoing central neural blockade should also receive discharge advice regarding the importance of these clinical features.22

Organisational problems

The role played by organisational failure in clinical adverse events is one of the major issues in modern health care. The Council noted that organisational factors contributed to the outcome in 17 out of 40 (43%) mortality cases and 30 out of 179 (17%) morbidity cases. Lack of interdisciplinary planning and/or communication failure occurred in some cases during preoperative care. Failure to seek senior advice, lack of supervision and poor exchange of information during clinical handover were factors in three deaths that occurred in intensive care, and one intra-operative death due to massive haemorrhage. Many of the reports of drug administration errors also highlighted organisational problems. However the major area of concern relates to the provision of postoperative pain management. This has already been highlighted above relating to the risk of spinal cord injury after central neural blockade, but there are also major safety

It is essential that all staff responsible for the care of such patients be made aware of the importance of the new onset of weakness, back pain, sensory disturbance and fever.

41

concerns in the provision of postoperative care generally, but particularly with regard to pain management.

Postoperative care

and pain management

It is important to note that inadequate or inappropriate postoperative care was implicated in eight of the 40 mortality cases (20%) and in 15 of the 179 morbidity cases (8%).

Of major concern was one death that was due to hypoxia in a patient who had prolonged surgery lasting 14 hours and despite persistent drowsiness, tachycardia as well as needing medical assistance for airway control in the PACU, was discharged to the general ward after midnight. Some time later the patient required a MET call for hypoxia and suffered a cardiac arrest. During resuscitation there were initial problems with the airway and oxygenation remained poor even after transfer to intensive care. The patient failed to recover neurologically and died several weeks later. Another mortality case involved a patient who required re-

operation for bleeding after nasal surgery and although stable haemodynamically, was drowsy. Six hours after discharge from the PACU, it was documented that the patient had received anti-emetics and morphine 10 mg subcutaneously, followed 90 minutes later by sedation with nitrazepam 5 mg. Thirty minutes later, the patient was noted to be sleeping. Two hours later, at 1 am, the patient was found to be unresponsive and pulseless. A code blue was not called as the nursing staff deemed that the cardiac arrest had occurred some time earlier. Although an autopsy revealed co-existing coronary artery disease, death was attributed to hypoxia.

Two particular morbidity cases are noteworthy in relation to the current risk to patients receiving postoperative analgesia. One case involved a young male patient with lower limb fractures who received patient controlled analgesia (PCA) with morphine. He was in a medical ward, and after he was assessed as having poor pain control, the bolus dose was increased from 1 to 2 mg. The following morning he was found to be unresponsive with a Glasgow Coma Score (GCS) of three and respiratory obstruction. He had not been seen by nursing staff in the preceding four hours and a review of the

Failure to seek senior advice, lack of supervision and poor exchange of information during clinical handover were factors in three deaths that occurred in intensive care, and one intra-operative death due to massive haemorrhage.


42

PCA chart showed he had received 52 mg of morphine over the previous ten hours, as well as tramadol and ibuprofen. He did not respond neurologically to naloxone and was admitted to ICU. A subsequent CT brain was normal but he was thought to have suffered hypoxic brain injury and was finally discharged from hospital with a residual neurological deficit.

Another case involved a young female patient who underwent upper limb surgery and despite receiving 10 mg of morphine in the PACU, remained in severe pain. She was given a further dose of 10 mg morphine subcutaneously as well as another 10 mg intravenously as per protocol. Soon after, she had a pain score of 4/10 and was discharged to the ward with PCA. Initial observations were normal but 30 minutes later she was found to be pale, cyanosed and unresponsive with pin point pupils. A code blue was called and an anaesthetist performed bag and mask ventilation with immediate improvement in her conscious state. She was given two doses of naloxone (each of 40 mcg). Review of the PCA chart showed she had received 3 mg of morphine between arriving on the ward and the respiratory arrest. She had therefore

received a total of 33 mg of morphine in 100 minutes.

These cases graphically demonstrate that even in young, fit patients, the safe provision of postoperative analgesia is challenging. It is sobering to note that in this Council’s previous triennial report published in 2007,13 it was stated that

“an extremely common theme in morbidity case reports was respiratory depression due to the use of opioids in the setting of inadequate postoperative surveillance and/or inappropriate drug use. The Council remains extremely concerned about this ongoing risk to patients.” This extreme concern remains in 2011. The five morbidity reports of respiratory depression are most likely to be the tip of a vast iceberg.

These cases graphically demonstrate that even in young, fit patients, the safe provision of postoperative analgesia is challenging.

43

Maternal mortality

Data obtained from the Consultative Council on Obstetric and Paediatric Mortality and Morbidity show that during the period of 2003−2005, there were 190,375 confinements and 22 cases of maternal mortality in Victoria (seven direct and fifteen indirect). The Victorian maternal mortality ratio is 11.6 per 100,000 confinements.23 In comparison, the report Maternal deaths in Australia noted an Australian maternal mortality ratio of 8.4 per 100,000.24 There were no cases of maternal deaths directly related to anaesthesia in Victoria.

The Confidential Inquiry into Maternal Death in the United Kingdom over the same triennium (2003–2005) identified 295 cases (direct/indirect) of maternal mortality in over 2 million deliveries giving a maternal mortality ratio of 14 per 100,000. The report also identified six deaths directly related to anaesthesia in the total of 132 direct maternal deaths.25


44

45

References

1. Public Health and Wellbeing Act 2008 Victoria, ss.33-43

2. Australian Institute of Health and Welfare., Australian hospital statistics 2004–2005. Cat. No HSE 41. Canberra: AIHW; 2006 < http://www.aihw.gov.au/publication-detail/?id=6442467847>.

3. McNicol L, Story D, Leslie K, Myles P, Fink M, Shelton A, Clavisi O, Poustie S. Postoperative complications and mortality in older patients having non-cardiac surgery at three Melbourne teaching hospitals. Medical Journal of Australia May 2007: 186;(9):7.

4. Itani KM. Fifteen years of the National Surgical Quality Improvement Program in review. Am J Surg 2009;198 (5 Suppl):S9-S18.

5. K Wilkinson, I C Martin, M J Gough, J A D Stewart, S B Lucas, H Freeth, B Bull, M Mason. An age old problem. A review of the care received by elderly patients undergoing surgery. London: National Confidential Enquiry into Patient Outcome and Death; 2010.

6. Australian and New Zealand Audit of Surgical Mortality. National Report. Adelaide: Royal Australasian College of Surgeons; 2009.

7. Story DA, Leslie K, Myles PS, Fink M, Poustie SJ, Forbes A, et al. Complications and mortality in older surgical patients in Australia and New Zealand (the REASON study): a multicentre, prospective, observational study. Anaesthesia 2010, 65:1022-1030

8. Australian and New Zealand College of Anaesthetists. Guidelines for the Safe Administration of Injectable Drugs in Anaesthesia. Melbourne: ANZCA; 2009, <http://www.anzca.edu/resources/professional-documents/ps51.html>.

9. McNicol L, Mackay P. Special Article. Anaesthesia-related morbidity in Victoria:1990 to 2005, Anaesthesia and Intensive Care September 2010: 38(5): 837-848.

10. The World Health Organisation (WHO) Surgical Safety Check List, Geneva: WHO; 2009. <http://www.who.int/patientsafety/safesurgery/tools_resources/SSSL_Checklist_finalJun08.pdf >.

11. Victorian Surgical Consultative Council. Victorian Postoperative Orders Form. Melbourne: VSCC, <http://www.health.vic.gov.au/vscc/victorian-hospitals-post-operative-orders-form>.


46

12. Victorian Quality Council. Acute pain management measurement toolkit. Melbourne: Department of Human Services, February 2007, <www.health.vic.gov.au/qualitycouncil>.

13. Victorian Consultative Council on Anaesthetic Mortality and Morbidity. Ninth report of the Victorian Consultative Council on Anaesthetic Mortality and Morbidity.McNicol L, ed. Melbourne: VCCAMM; 2007 <www.health.vic.gov.au/vccamm>

14. Coroners Act 2008 Version No. 001 as at 1 November 2009

15. Gibbs N, ed. Australian and New Zealand College of Anaesthetists, Safety of Anaesthesia in Australia – a review of anaesthesia-related mortality 2003–2005. Melbourne: Australian and New Zealand College of Anaesthetists; 2009 <http://www.anzca.edu.au/resources/books-and-publications/ANZCA%20Mortality%20Report.pdf>

16. American Society of Anesthesiologists, Manual for anesthesia, Department Organisation and Management, 2003–2004, (excerpt). Park Ridge, IL: American Society of Anesthesiologists; 2007 <www.asahq.org/clinical/physicalstatus.htm>.

17. Haynes AB, Wieser TG, Berry WR. A Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population, for the Safer Surgery Saves Lives Study Group. N Engl J Med 29 January 2009; 360(5):491-499.

18. Sampson IH, Munoz-Furlong A, Campbell R. Second symposium on the definition and management of anaphylaxis: Summary report-Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol Feb 2006; 117(2):391-7.

19. Promedica P/L, part of Thermo Fisher Scientific, 5 Caribbean Drive. Scoresby, Vic 3179 (1300 735 292)

20. Piercy M, McNicol L, Dinh DT, Story DA, Smith JA. Major complications related to the use of transoesophageal echocardiography in cardiac surgery J Cardiothorac Vasc Anesth 2009; 23: 62-65.

21. Fleisher LA et al. ACC/AHA 2007 Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery. Journal of the American College of Cardiology October 23 2007; 50(17): 1707-1732.

47

22. Discharge advice for patients receiving regional blockade <http://www.health.vic.gov.au/vccamm/downloads/discharge.pdf>

23. Oats JN, King J, The Consultative Council on Obstetric and Paediatric Mortality and Morbidity, Annual Report for the year 2006, incorporating the 45th Survey of Perinatal Deaths in Victoria. Melbourne: Consultative Council on Obstetric and Paediatric Mortality and Morbidity; 2008.

24. Sullivan E, Hall B, King J. Maternal Deaths in Australia 2003-2005. Canberra: Australian Institute of Health and Welfare; 2008.

25. Lewis G, ed. The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving Mothers’ Lives: reviewing maternal deaths to make motherhood safer – 2003-3005. The Seventh Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. London: CEMACH; 2007.


48

Appendix 1: Council membership

The membership comprises:– The chairman: a specialist anaesthetist

nominated by the Australian and New Zealand College of Anaesthetists, recommended by the Minister and appointed by Cabinet.

– Sixteen specialist anaesthetists, appointed by the Minister, comprising three nominated by the Australian and New Zealand College of Anaesthetists, three by the Australian Society of Anaesthetists, and ten (including a rural practitioner) by the Victorian Teaching and Regional Hospitals.

– Six additional members, appointed by the Minister, comprising a nominated representative from each of the Royal Australasian College of Surgeons, the Australian and New Zealand Intensive Care Society and/or the College of Intensive Care Medicine, the Royal College of Pathologists of Australasia and/or the Victorian Institute of Forensic Medicine, the Australasian College of Emergency Medicine, the Royal Australian College of General Practitioners and/or the Rural Visiting Medical Officer, and the Department of Health.

Chairman:

A/Prof Larry McNicol* (2005)Dr Patricia Mackay* (1991–2005)

Deputy Chairman: Dr Anthony Weaver (1995-2005)

Emeritus Consultant: Dr Patricia Mackay* (2005)

Confidential Project Officers: Ms Pauline Berryman (2001–2006)Ms Zenaida Magtoto (2007-2010)Ms Elena Alexandrova* (2010)

Nominated representatives

The Australian and New Zealand College of Anaesthetists Dr Patricia Mackay* (1991) Dr Mark Langley (1994–2007)Dr Phillip Ragg* (2000)Dr Christopher Bain* (2008)A/Prof Larry McNicol* (2004)

The Australian Society of AnaesthetistsDr Patrick Hughes (1994–2007)Dr Gaylene Heard* (2004)Dr Anthony Weaver (1994–2005)Dr Alex Babarczy*(1998–2004)(2007) Dr Rob Beavis* (2007)

* Denotes currently serving members in 2011

49

Victorian Teaching and Regional Hospitals and the Rural Visiting Medical Officer Dr Jennifer Carden* (2001)Dr John Monagle* (2004)Dr Margaret Griggs (2001-2011)Dr Andrew Ross* (2004)Dr Barbara Robertson (2001-2008)Dr Paul Francis* (2004)A/Prof Terence Loughnan* (2004)Dr Simon Tomlinson* (2008)Dr David Beilby* (2008)Dr Maggie Wong* (2008)Dr Robert Dawson* (2008)

Representatives nominated by the Australian and New Zealand Intensive Care Society and/or the College of Intensive Care MedicineDr John Santamaria (1994–2006)Dr Helen Opdam (2007-2009)Dr David Charlesworth* (2010)

Representatives nominated by the Royal Australasian College of SurgeonsA/Prof David Francis (2001-2008)Mr Michael Dobson* (2008)

Representatives nominated by the Royal Australian College of General Practitioners Dr Anthony McCarthy (1994–2004)Dr Quentin Tibballs* (2010)

Representatives nominated by the Royal College of Pathologists of AustralasiaA/Prof David Ranson (1998-2004)Dr Alpha Tsui (2003-2004)

Representative nominated by the Victorian Institute of Forensic MedicineDr Shelley Robertson* (2004)

Representatives nominated by the Australasian College for Emergency MedicineDr Marcus Kennedy (2001–2003) Dr Stephen Priestley (2003–2006)Dr Gino Toncich* (2007)

Representative nominated by the Department of Health Dr Michael Ackland (1994-2010)

Members are appointed for a three-year term, and are eligible for re-appointment.

Secondments

Seconded Provisional Fellows Dr Mathew Piercy (2004-2007)Dr Leona Leong (2010-2011)Dr Dana Pakrou (2010-2011)Dr Louise Simpson (2010-2011)

Seconded Confidential Project OfficerMr Andrew Shelton (2006-2008)

* Denotes currently serving members in 2011


50

Appendix 2: Operation of the council

Under the auspices of the Department of Health, the Council meets monthly and discusses the cases of mortality and morbidity prepared by the Chairman. The functions of the Council are to identify avoidable causes of mortality or morbidity related to anaesthesia, and to disseminate as widely as possible the results of deliberations and possible strategies for prevention. The Council’s role is not to apportion blame but to establish the contribution of anaesthesia-related factors to the death or morbidity, and to improve the safety of clinical practice through dissemination of relevant information. It is therefore a ministerial advisory committee with a specific role of quality assurance for the clinical practice of anaesthesia.

The Council is fully committed to working collaboratively with all Victorian anaesthetists in order to (i) maximise the reporting of potentially avoidable anaesthesia related adverse events, (ii) continue to maintain strict confidentiality, and (iii) provide direct feedback to the reporting practitioner as well as wide dissemination of information obtained from analysis and review of collated cases.

Data collection and analysis

The Council obtains case reports from three sources: direct reports from anaesthetists, QA cases referred from hospital departments of anaesthesia, and coronial cases. In Victoria, direct reporting to the Council of mortality and morbidity related to anaesthesia has

always been voluntary. After evaluation of the case by the chairman and, if required to obtain additional information, follow-up with the reporting practitioner, the direct report case is de-identified and presented to the Council for deliberation and classification. QA reports are referred as already de-identified cases and are similarly presented to the Council. The chairman also has access to the coroner’s files, which are public documents. The chairman uses these files to gather further information, as necessary, by consulting directly with the anaesthetist, surgeon or other practitioner identified from the medical deposition to the coroner, and then reviewing appropriate sections of the hospital record and obtaining autopsy reports. Coroner’s cases are then de-identified and presented to the Council. The Council chairman is able to obtain such information through the cooperation of the Coroners Court of Victoria, but it is important to note that the legislative provisions of the Council under sections 42 and 43 of the Health Act1 prohibit the coroner from access to the Council’s case reports and deliberations.

After complete de-identification, the cases are presented to the Council for discussion and classification. In the case of direct referrals, the Council’s deliberations are conveyed to the medical practitioner concerned. After each meeting, the chairman documents the Council’s deliberation for each case and these minutes of meetings are retained, and all relevant data entered into a secure and dedicated electronic database. No information in the computer database

51

identifies patient, doctor or hospital. In addition to classification, keywords are allocated for each case report, and this information helps in the identification of clinical issues that emerge from case analysis.

Validity of data

The level of voluntary and QA reporting has been maintained at the same level as in the ninth report. It is however noteworthy that there has been reduced access to coronial cases. This raises some concerns that the actual number of anaesthesia-related deaths may be higher than 40. Measures have been undertaken, with the co-operation and support of the State Coroner, Judge Jennifer Coate to improve the access to coroner’s cases.

Despite this reduced availability of coronial information, the Council considers that the level of reporting overall is sufficiently valid to draw some reasonable conclusions and justify certain recommendations.

It is also important to note that the Council remains committed to collecting case reports on procedures outside the operating room, including sedation and anaesthesia for invasive procedures in endoscopy, radiology and cardiology. Similarly, there are also complications that occur in other areas such as pain management, intensive care, emergency medicine and resuscitation, and it is likely that that these are under reported. Although it could be argued that inclusion of all of these as part of anaesthesia-related morbidity and mortality may lead to an over

estimation of adverse event rates, it is the view of this Council that it is important to include them under the current terms of reference.

It is recognised that a major priority for the profession of anaesthesia is to identify all cases involving significant morbid adverse events and mortality, as well as obtaining the most accurate data possible for the number of anaesthesia-related procedures. Only then can such improved estimates of anaesthesia-related mortality and frequency of specified complications be relied upon to inform anaesthetists and the community about risk. This priority is currently being pursued through the Quality and Safety Committee and the ANZCA National Mortality Working Group, and the Council is collaborating with these committees in this process.

The president of ANZCA is the chair of the National Mortality Working Group and the membership is comprised of the chairman of the ANZCA Safety and Quality Committee as well as the chairs of each state’s Mortality Committee or Consultative Council. It is interesting to note that the Victorian Consultative Council on Anaesthetic Mortality and Morbidity remains the only such body that reviews morbidity data as well as mortality. The National Mortality Working Group publishes a triennial report based on mortality data forwarded from each state, and the most recent publication, Safety of anaesthesia in Australia: a review of anaesthesia-related mortality 2003–2005, was distributed in 2010.15


52

New legislation under the Public Health and

Wellbeing Act 2008 and Coroners Act 2008

The Council was established in 1976 under section 13 of the Public Health Act 1958 and the legislative provisions have recently been updated in sections 33-43 of the Public Health and Wellbeing Act 2008.1 These revised provisions have taken into account the essential requirement for the preservation of confidentiality, as well as recognising the need to improve systematic reporting of potentially anaesthesia-related morbidity and mortality. Historically, in the Victorian coronial legislation, there has been considerable confusion regarding the definition of reportable deaths, particularly for deaths that may have been deemed to be associated with anaesthesia.

(1) Preservation of confidentiality

The Council is aware that reporting of mortality and morbidity has always been on a voluntary basis and that the speciality of anaesthesia has a long history of participation in audit and quality assurance activities. The level of reporting has remained constant over many years, due mainly to the high level of trust between practising anaesthetists and the Council. Sections 42 and 43 describe the confidentiality obligations which preclude the identification of a person from whom, or in relation to whom, the information was obtained. In addition to these legislative confidentiality provisions, the Council has imposed an additional layer of security in that only the Council chairman and the Council’s confidential project officer are privy to the identity of the reporting practitioner, the patient, and the hospital. All identifiable information is deleted from the case reports prior to presentation to

Council for deliberation. However, it remains important for the Council chairman to have direct contact with the reporting anaesthetist in order to obtain the most accurate information regarding the case.

(2) Enhanced systematic reporting

Although there is a strong track record of spontaneous direct reporting by anaesthetists, it is important to maximise the level of case acquisition as required by the Council’s terms of reference. There are new sections in the legislative provisions which are designed to improve the systematic reporting of anaesthesia-related mortality and morbidity by Victorian hospitals. Under section 39 of the Public Health and Wellbeing Act 2008,1 the Council chairman may request, by written notice, a health service provider, to provide general or specific information (anaesthesia-related morbidity or mortality). Under section 40, the health service provider must provide such requested information.

In August 2010, letters were sent from the chairman of Council to all health services in Victoria, outlining these requirements. It is anticipated that compliance with this legislation will be achieved through hospital department of anaesthesia QA co-ordinators and it is hoped this will increase the overall level of reporting.

53

(3) Improved coronial legislation

The new Coroners Act 200814 includes improved definitions of reportable deaths, and in Part 1, section 4, 2 (b), a reportable death includes, a death that occurs -– during a medical procedure; or– following a medical procedure where

the death is or may be causally related to the medical procedure – and a registered medical practitioner would not, immediately before the procedure was undertaken, have reasonably expected the death.

Included within the term medical procedure is any diagnostic or therapeutic procedure as well as the administration of any anaesthetic, including general, local, conscious sedation, regional anaesthetic, intensive care sedation, spinal or epidural anaesthetic or other.

This improved definition of reportable death is welcomed by Council and it is more specifically aligned with our own definitions of anaesthesia-related mortality. Under the new arrangements with the Coroners Court of Victoria, the chairman of Council has access to the medical depositions submitted to the coroner in all cases in which any of the above anaesthetic administrations have occurred. All such depositions can then be screened by the chairman and when deemed appropriate, further information can be obtained for cases that require deliberation by Council.

Relationships with

specialist organisations

The Council receives strong support from both the Australian and New Zealand College of Anaesthetists (ANZCA) and the Australian Society of Anaesthetists (ASA). The chairman is a co-opted member of the Victorian Regional Committee of ANZCA, which provides assistance in publicising the Council’s activities.

The Council works collaboratively with the other consultative councils, the Victorian Surgical Consultative Council and the Consultative Council on Obstetric and Paediatric Mortality and Morbidity. When appropriate, there is provision under the Public Health and Wellbeing Act 1, for sharing of confidential information between the three consultative councils.

The ANZCA also provides considerable support through the formation of the National Mortality Working Group. It is also important to recognise that with the support of both the ANZCA and the Joint Consultative Committee on Anaesthesia of the Royal Australian College of General Practitioners (JCCA) a practitioner who provides a direct report to Council is eligible for two points toward their respective continuous professional development program. To avoid compromising confidentiality, the chairman provides the anaesthetist concerned with confirmation of points’ eligibility in a letter separate from the one which contains the Council’s deliberations on the case.


54

Output of the Council

The Council publishes both annual and triennial reports. Annual reports provide details of the Council’s workload and activities for the relevant year. They involve some preliminary analysis of the cases that were reviewed during that year but some of the cases may not have actually occurred during that particular year. By contrast, triennial reports specifically review all cases that actually occurred during that three-year period. They involve more in-depth analysis of the clinical issues and trends that emerge from the collated case reports after formal classification and allocation of keywords. Triennial reports have historically been delayed, mostly due to the time constraints involved in coronial cases, but also due to the time required to undertake detailed collation of accumulated case reports. It remains challenging to reduce the time lag between the actual event and the publication of reports, but it is now a major priority. In addition, a useful tool for more rapid communication of issues to emerge from case reviews is through alerts and articles posted on the Council website.

55

Appendix 3: Classification of cases

reported to the council

All cases were classified according to the criteria agreed by the National Working Party on Anaesthetic Mortality convened by ANZCA and by using the same glossary of terms. The same classification was used for morbidity, which is studied only in Victoria.

Categories of mortality and morbidity

deaths or morbidity attributable to anaesthesia

category 1 Where it is reasonably certain that death or morbidity was caused by

the anaesthesia or other factors under the control of the anaesthetist

category 2 Where there is some doubt whether death or morbidity was entirely

attributable to the anaesthesia or other factors under the control

of the anaesthetist

category 3 Where death or morbidity was caused by both medical/surgical

and anaesthesia factors

Notes

1. The intention of the classification is not to apportion blame in individual cases but

to establish the contribution of the anaesthesia factors to the death or morbidity.

2. The above classification is applied regardless of the patient’s condition before the

procedure. However, if the medical condition is considered to have substantially

contributed to the anaesthesia-related death or morbidity, then subcategory

H should also be applied.

3. If no factor under the control of the anaesthetists is identified that could

or should have been done better, then subcategory G should also be applied.


56

death or morbidity in which anaesthesia played no part

category 4 surgical death or morbidity where the administration of the anaesthesia

was not contributory and where surgical or other procedural factors

were implicated

category 5 inevitable death or morbidity that would have occurred irrespective

of anaesthesia or surgical procedures

category 6 fortuitous death or morbidity that 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 anaesthetist or surgeon

un-assessable death/morbidity

category 7 those cases that cannot be assessed despite considerable data

because information is conflicting or key data are missing

category 8 cases that cannot be assessed because there is inadequate data

category 9 a critical incident where a problem was identified but

no morbidity occurs

57

Causal or contributory factors in anaesthesia-related mortality

and morbidity − subcategories

Note: It is usual for more than one factor to be identified in the case

of anaesthesia-attributable death or morbidity

a. preoperative

(i) assessment this may involve failure to take an adequate history, perform an

adequate examination, undertake appropriate investigation or

consultation, or make an adequate assessment of the volume

status of the patient in an emergency. Where this is also a

surgical responsibility, the case may be classified in category 3

(ii) management this may involve failure to administer appropriate therapy or

resuscitation. urgency and the responsibility of the surgeon may

also modify this classification

b. anaesthesia technique

(i) choice or application there was inappropriate choice of technique where it was

contraindicated or the incorrect application of a technique that

was correctly chosen

(ii) airway maintenance,

including pulmonary

aspiration

there was inappropriate choice of an artificial airway or failure

to maintain or provide adequate protection of the airway, or to

recognise the misplacement or occlusion of an artificial airway

(iii) ventilation death or morbidity was caused by failure of ventilation of the

lungs for any reason. this would include inadequate ventilator

settings and failure to reinstitute proper respiratory support

after deliberate hypoventilation (for example, bypass)

(iv) circulatory support death or morbidity was caused by failure to provide adequate

support where there was haemodynamic instability, particularly

in relation to techniques involving sympathetic blockade


58

c. anaesthesia drugs

(i) selection this involves the administration of a wrong drug or one that was

contraindicated or inappropriate. this category includes ‘syringe

swap’ errors

(ii) dose this may be due to incorrect dose, absolute or relative to the

patient’s size, age and condition. in practice, it is usually an

overdose

(iii) adverse drug reaction this includes all fatal drug reactions, both acute (such as

anaphylaxis) and the delayed effects of anaesthesia agents such

as the volatile agents

(iv) inadequate reversal this includes relaxant, narcotic and tranquillising agents where

reversal was indicated

(v) incomplete recovery a prolonged coma is an example

d. anaesthesia management

(i) crisis management this involves inadequate management of unexpected

occurrences during anaesthesia or in other situations which, if

uncorrected, could have led to death or severe injury

(ii) inadequate monitoring this involves failure to observe minimum standards (as

enunciated in the anZca policy document) or to undertake

additional monitoring when indicated (for example, the use of a

pac in the case of left ventricular failure)

(iii) equipment failure death or morbidity was a result of failure to check equipment or

due to failure of an item of anaesthesia equipment

(iv) inadequate resuscitation death or morbidity was a result of failure to provide adequate

resuscitation in an emergency situation

(v) hypothermia death or morbidity was a result of failure to maintain adequate

body temperature within recognised limits

59

e. postoperative

(i) management death or morbidity was a result of inappropriate intervention

or omission of active intervention by the anaesthetist or a

person under their direction (for example, a recovery or pain

management nurse) in some matter related to the patient’s

anaesthesia, pain management or resuscitation

(ii) supervision death or morbidity was due to inadequate supervision or

monitoring. the anaesthetist has ongoing responsibility but the

surgical role must also be assessed

(iii) inadequate resuscitation death or morbidity was due to inadequate management of

hypovolaemia or hypoxaemia, or failure to perform proper

cardiopulmonary resuscitation

f. organisational

(i) inadequate supervision,

inexperience or

assistance

death or morbidity was a result of the anaesthetist, whether a

trainee, a non-specialist or a specialist, undertaking an unfamiliar

procedure. the criterion of adequacy of supervision of a trainee

is based on the anZca policy document

(ii) poor organisation

of the service

death or morbidity was due to inappropriate delegation, poor

rostering and/or fatigue

(iii) failure of interdisciplinary

planning

death or morbidity was a result of poor communication in

perioperative management and failure to anticipate need for

high-dependency care

G. no correctable factor identified

death or morbidity was due to anaesthesia factors but no better technique could be suggested

h. medical condition of the patient

the patient’s medical condition was a significant factor in the anaesthesia-related

death or morbidity

60

Consultative Council on Anaesthetic Mortality and Morbiditygpo Box 4923, Melbourne, Victoria, Australia 3001Website: http://www.health.vic.gov.au/vccamm/ Email: [email protected]

Documents

Tenth report of the Victorian Consultative Council on ...docs2.health.vic.gov.au/docs/doc... · tenth report of the victorian consultative council on anaesthetic mortality and morbidity