OFFICE OF THE STATE CORONER FINDINGS OF INQUEST CITATION: Inquest into the death of Margaret Allan Johnston TITLE OF COURT: Coroner’s Court JURISDICTION: Brisbane FILE NO: COR 3297/06(3) DELIVERED ON: 20 November 2009 DELIVERED AT: Brisbane HEARING DATE(s): 2/9/08, 28/11/08, 1/12/08 – 5/12/08, 7/08/09, 17/8/09 – 21/08/09, 24/08/09, 25/08/09 and 28/08/09 FINDINGS OF: Coroner Lock CATCHWORDS: CORONERS: Inquest – Insulin, hypoglycaemic brain injury, reliability of immunoassays, medication error or maladministration REPRESENTATION: Counsel Assisting: Mr S Hamlyn-Harris For St Andrews War Memorial Mr G Diehm SC and Ms M Callaghan Hospital and its employees: instructed by Blake Dawson For Dr Hay, Dr Hill, Dr Love and Mr A Luchich instructed by Avant Dr Venkatesh: For RN Heelan and RN Moss: Mr G Rebetzke instructed by Roberts & Kane For Dr Johnston: Mr R Perry SC instructed by Spark Helmore For Mr Johnston and family: Mr P Clapin of Clapin Lawyers CORONERS FINDINGS AND DECISION 1. These are my findings in relation to the death of Margaret Allan Johnston who died at the St Andrews War Memorial Hospital on 2 December 2006. These findings seek to explain how the death occurred and consider whether any changes to policies or practices could reduce the likelihood of deaths occurring in similar circumstances in the future. Section 45 of the Coroners Act 2003 (“the Act”) provides that when an inquest is held into a death, the Coroner’s written findings must be given to the family of the person who died and to each of the persons or organisations granted leave to appear at the inquest. These findings will be distributed in accordance with the requirements of the Act and also placed on the website of the Office of the State Coroner. The scope of the Coroner’s inquiry and findings 2. A Coroner has jurisdiction to inquire into the cause and the circumstances of a reportable death. If possible he/she is required to find:- (a) whether a death in fact happened; (b) the identity of the deceased; (c) when, where and how the death occurred; and (d) what caused the person to die. 3. There has been considerable litigation concerning the extent of a Coroner’s jurisdiction to inquire into the circumstances of a death. The authorities clearly establish that the scope of an inquest goes beyond merely establishing the medical cause of death. 4. An inquest is not a trial between opposing parties but an inquiry into the death. In a leading English case it was described in this way:- “It is an inquisitorial process, a process of investigation quite unlike a criminal trial where the prosecutor accuses and the accused defends… The function of an inquest is to seek out and record as many of the facts concerning the death as the public interest requires.” 1 5. The focus is on discovering what happened, not on ascribing guilt, attributing blame or apportioning liability. The purpose is to inform the family and the public of how the death occurred, with a view to reducing the likelihood of similar deaths. As a result, the Act authorises a Coroner to make preventative recommendations concerning public health or safety, the administration of justice or ways to prevent deaths from 1 R v South London Coroner; ex parte Thompson (1982) 126 S.J. 625 Findings into the death of Margaret Allan Johnston 1 happening in similar circumstances in future.2 However, a Coroner must not include in the findings or any comments or recommendations, statements that a person is or may be guilty of an offence or is or maybe civilly liable for something.3 The Admissibility of Evidence and the Standard of Proof 6. Proceedings in a Coroner’s court are not bound by the rules of evidence because the Act provides that the court “may inform itself in any way it considers appropriate.”4 That does not mean that any and every piece of information, however unreliable, will be admitted into evidence and acted upon. However, it does give a coroner greater scope to receive information that may not be admissible in other proceedings and to have regard to its origin or source when determining what weight should be given to the information. 7. This flexibility has been explained as a consequence of an inquest being a fact- finding exercise rather than a means of apportioning guilt. 5 8. A coroner should apply the civil standard of proof, namely the balance of probabilities but the approach referred to as the Briginshaw sliding scale is applicable.6 This means that the more significant the issue to be determined, the more serious an allegation or the more inherently unlikely an occurrence, the clearer and more persuasive is the evidence needed for the trier of fact to be sufficiently satisfied that it has been proven to the civil standard.7 This approach was approved by the Queensland Court of Appeal8 when specifically dealing with the coronial jurisdiction when it said that the “sliding scale of satisfaction test does not require a tribunal of fact to treat hypotheses that are reasonably available on the evidence as precluding it from reaching the conclusion that a particular fact is more probable than not.”9 9. It is also clear that a Coroner is obliged to comply with the rules of natural justice and to act judicially.10 This means that no findings adverse to the interest of any party may be made without that party first being given a right to be heard in opposition to that finding. As Annetts v McCann11 makes clear, that includes being given an 2 Section 46 of the Act 3 Sections 45(5) and 46(3) of the Act 4 Section 35 of the Act 5 R v South London Coroner; ex parte Thompson per Lord Lane CJ, (1982) 126 S.J. 625 6 Anderson v Blashki [1993] 2 VR 89 at 96 per Gobbo J 7 Briginshaw v Briginshaw (1938) 60 CLR 336 at 361 per Sir Owen Dixon J 8 Hurley v Clements & Ors [2009] QCA 167 9 Hurley v Clements & Ors [2009] QCA 167 at paragraph 27 10 Harmsworth v State Coroner [1989] VR 989 at 994 and see a useful discussion of the issue in Freckelton I., “Inquest Law” in The inquest handbook, Selby H., Federation Press, 1998 at 13 11 (1990) 65 ALJR 167 at 168 Findings into the death of Margaret Allan Johnston 2 opportunity to make submissions against findings that might be damaging to the reputation of any individual or organisation. 10. If, from information obtained at an inquest or during the investigation, a Coroner reasonably suspects a person has committed a criminal offence, the Coroner must give the information to the Director of Public Prosecutions in the case of an indictable offence, and to the Chief Executive of the department which administers legislation creating an offence which is not indictable.12 If, from information obtained at an inquest or during the investigation, a coroner reasonably believes that the information may cause a disciplinary body for a person’s profession or trade to inquire into or take steps in relation to the person’s conduct, then the coroner may give that information to that body.13 The Evidence Overview 11. Margaret Allan Johnston was admitted to St Andrew's War Memorial Hospital on 2 September 2006, after being referred by her general practitioner with a suspected urinary tract infection (“UTI”) and increased confusion. 12. At around 4:00 am on 4 September 2006 she was found by nursing staff to be unresponsive. Blood tests taken at her bedside showed that she had a very low blood glucose level and the clinical signs were consistent with her having suffered from a severe hypoglycaemic event. Subsequent tests revealed that Mrs Johnston had also suffered a series of cerebral strokes. She remained in a coma. Throughout the next three months she only had basic breathing and swallowing functions and after consultation with her family her feeding tube was withdrawn and she passed away on 2 December 2006. 13. An analysis of blood samples taken over the days following her admission revealed that on 3 and 4 September Mrs Johnston had unexplained high levels of insulin in her blood. She was not a diabetic and was not prescribed insulin. Dr De Voss, a pathologist at Queensland Medical Laboratory Pathology (“QML”), was responsible for conducting immunoassays on the blood samples and he discussed the results with Dr Hill, a consultant endocrinologist who was asked by the hospital to examine Mrs Johnston. As a result of the test results and the opinions of Dr De Voss and Dr Hill, combined with the clinical evidence, an initial investigation by the hospital commenced. This concluded that on the evidence available, Mrs Johnston had 12 Section 48(2) of the Act 13 Section 48(4) of the Act Findings into the death of Margaret Allan Johnston 3 probably been administered two separate doses of insulin exogenously (ie., insulin administered via an external source) with one on either the evening of 2 September 2006 or the morning of 3 September 2006 and the second dose on either the evening of 3 September 2006 or the early hours of 4 September 2006.. The hospital administration reported the matter to the police on the evening of 5 September 2006 as it could find no evidence of a medication administration error. 14. An extensive police investigation commenced and numerous statements were taken from security staff, nursing staff, medical staff, Mrs Johnston’s family and various experts. Professor Donald Chisholm, an eminent endocrinologist, gave an opinion in December 2006 supporting the view of Dr Hill that exogenous insulin had been administered. Detective Senior Constable Helen Wheatley was the lead police investigator and ultimately a very thorough and extensive report was forwarded to the Coroner. The accepted evidence at the conclusion of the police investigation was that a person or persons unknown had injected Mrs Johnston with either one or two doses of insulin, causing her to suffer a severe episode of hypoglycaemia resulting in irreversible and severe brain damage, which ultimately resulted in her death. No person could be identified as the person likely to be responsible for injecting Mrs Johnston. The only logical possibilities which remained were an accidental administration/medication error by hospital staff or malicious administration by hospital staff or an outsider. 15. An inquest was set down to commence on 2 December 2008 to examine the circumstances surrounding Mrs Johnston's death. In the week before the commencement of the inquest a report from Professor Marks, a Professor of Clinical Biochemistry and world renowned expert in his field, offered a different opinion on the cause of Mrs Johnston's hypoglycaemia. He concluded that sepsis must be considered as a likely cause of hypoglycaemia in her case, and although the possibility that she had been given insulin on one or possibly two occasions could not be excluded with absolute certainty, he considered it to be unlikely. He challenged the reliability of the immunoassay results. 16. The report of Professor Marks was distributed to other expert witnesses. The opinion of Professor Marks was vehemently opposed by most of the medical experts who had provided statements to the investigating police and the Coroner. 17. The inquest had been listed, perhaps optimistically, for one week. As a result of the new evidence from Professor Marks, further evidence was sought. The issue of the Findings into the death of Margaret Allan Johnston 4 precise nature of the infective process that brought about Mrs Johnston’s admission needed to be investigated; along with what was “sepsis” and what was the causal relationship, if any, between sepsis and hypoglycaemia. An opinion from an infective diseases clinician (Associate Professor Allworth) was obtained. 18. Mrs Johnston had also suffered from a series of cerebral strokes (at about the same time as her hypoglycaemia) which would have further compromised her brain function. The significance of this information was not initially evident to me and during the first week I heard some evidence with respect to whether it was the hypoglycaemic event which caused her to suffer irreversible brain damage or whether it was the cerebral strokes or a combination of both. As a result the opinion of Professor Silburn, the neurologist who attended on Mrs Johnston, was sought to examine the relationship, if any, between hypoglycaemia and the onset of her strokes. 19. The respective expert opinions were then tested in a protracted court exercise. The medical issues were complex. 20. There was a significant delay in resuming this inquest after the first week, for all these reasons. This was a particularly unfortunate consequence for the family, causing them significant personal anguish and no doubt considerable further legal expense. As a result it has taken almost a further year to conclude the hearing of the evidence and for this decision to be handed down. It is almost 3 years since Mrs Johnston passed away. 21. I do not intend to set out in detail all the evidence that has been heard in this case. I have been greatly assisted in being provided with written submissions by counsel assisting Mr Hamlyn-Harris, Mr Perry SC representing Dr Johnston and Mr Diehm SC representing the hospital. Their respective oral submissions and that of Mr Rebetzke representing some of the nurses, also assisted. I have concluded, for the reasons that will be provided in my findings, that the opinions of Dr Hill and Professor Chisholm should be preferred to that of Professor Marks. I am satisfied that malicious administration of insulin by someone not connected to the hospital is most unlikely and has been excluded. Logically that only leaves open the other possibilities that there has been an administration/medication error or maladministration by hospital staff. I am unable to identify by whom and when the medication error or administration took place or which of those two possibilities is the most probable. I am cognisant that this is a serious finding to make but that this is the only conclusion Findings into the death of Margaret Allan Johnston 5 that can be made in accepting that the clinical findings of the morning of 4 September 2006, and subsequently, are consistent with hypoglycaemic brain damage and that the immunoassay results are clearly indicative of exogenous insulin administration. Social and Medical History 22. Margaret Allan Johnston was aged 82 years. She was married to Richard Johnston. Mrs Johnston was the mother of six children being two sons Craig and Garth and four daughters namely Amy, Sue, Jan and Zea. She has a number of grandchildren. It is evident that they were a close family with many of her children and Mr Johnston being present throughout most of the inquest. Importantly, none of the immediate family was prescribed insulin. Mrs Johnston’s daughter Jan was a non-insulin dependent diabetic and this condition was controlled by diet and exercise. No one within the family knows of any person that would want to intentionally harm Mrs Johnston. That possibility is in my view clearly excluded. 23. Mrs Johnston had previously suffered from atrial fibrillation and from angina attacks in 1998. She had suffered a minor cerebral stroke in 1999 from which she appeared to recover well. She had been complaining of some dizziness and light headedness during 2006 but was reported to be independently mobile. She was in relatively good health, albeit frail. Events of 1 to 6 September 2006 24. On 1 September 2006 Mr Richard Johnston telephoned his son Dr Craig Johnston. He was concerned about the condition of his wife who was confused and irrational. Dr Johnston visited his mother that evening and diagnosed a possible transient ischaemic attack (TIA) or a urinary tract infection, and suggested that she have an early consultation with her GP the next morning. Dr Johnston was telephoned shortly after midday the next day by Mrs Johnston’s GP who advised him that he was sending Mrs Johnston to St Andrews Hospital for admission with a provisional diagnosis of UTI. By late that afternoon many members of the family were present at the hospital. Dr Sonya Rose saw Mrs Johnston in the Emergency Department and arranged for her to be admitted to ward 4B. Dr Rose wrote out a medication order for gentamicin, an anti-biotic which is commonly prescribed for treatment of a UTI. 25. Dr James Love saw Mrs Johnston on the ward and diagnosed a possible kidney infection (pyelonephritis) and ordered another dose of gentamicin. Mrs Johnston seemed to be improving but as she continued to be confused, the nursing supervisor organised for her to be ‘specialled' overnight. This involves having an assistant in Findings into the death of Margaret Allan Johnston 6 nursing (“AIN”) being present in the patient’s room, or in the immediate vicinity, to ensure that the patient did not get out of bed or wander off. At this stage the treatment of Mrs Johnston followed standard good quality medical practice. 26. Dr Craig Johnston is a specialist anaesthetist who works at various hospitals including St Andrews War Memorial Hospital. During the investigation it was revealed that Dr Johnston assisted in the resiting of a cannula in the arm of his mother in the early evening of 2 September 2006. Nothing turns on that incident. 27. On 3 September 2006 Mrs Johnston remained confused. At around 11:00 am she had a short period of unconsciousness when she was being assisted to the toilet. Her pulse was rapid and irregular. Dr Johnston was present and thought that her unconsciousness was probably due to postural hypertension associated with rapid atrial fibrillation and suggested to medical staff that an ECG be performed. An ECG was organised and confirmed that she was suffering from atrial fibrillation. Following standard practice, Dr Love started Mrs Johnston on digoxin. 28. One issue which became quite significant during the course of the inquest was whether or not Mrs Johnston had any breakfast or fruit juice on the morning of 3 September 2006. Dr Johnston stated that he had been told on his arrival at the hospital on 3 September that his mother had eaten breakfast. The progress notes record that Mrs Johnston was given a drink at 5:25 am. The observation charts show that she was sitting in a chair between 7:15 am and 7:45 am, which may be consistent with her having got up to have some breakfast at that time. Endorsed Enrolled Nurse (“EEN”) Pierce (who specialled Mrs Johnston between 6:45 am and 3:15 pm on 3 September 2006) noted in her statement that she showered Mrs Johnston “after breakfast”. At 9:00 am a blood collector from QML took a sample of blood and noted that it was a “random test”. This is consistent with the collector having been told that the patient had eaten something and therefore the sample would not be regarded as a fasting test. 29. EEN Pierce, gave evidence that Mrs Johnston had not eaten any breakfast and was very confused. Dr Johnston recalls that a nurse had told him that his mother had eaten a good breakfast. The importance of this issue will become evident during an analysis of the significance of the results from various immunoassay tests that were later conducted. I am satisfied on the balance of the evidence that Mrs Johnston had taken some fruit juice, and perhaps some breakfast (even if only a small amount), on Findings into the death of Margaret Allan Johnston 7 the morning of 3 September, and therefore the blood sample taken at 9:00 am was not a fasting analysis. 30. Mrs Johnston continued to be “specialled” throughout her stay on ward 4B. AIN Manisha Smith specialled her from 10:30 pm to 7:00 am on the evenings of 2 and 3 September 2006. She was informed that Mrs Johnston was a falls risk. Other than when she was on a break AIN Smith remained just outside the doorway and completed an observation chart throughout the night. She does not recall anything unusual happening during either of those shifts. At one stage late on the evening of 3 September she recalled that Mrs Johnston was snoring loudly, and thinks this was probably around 11:30 and 11:45 pm, as the observation chart has the words “restless” crossed out and replaced with “sleeping”. She recalled asking the Registered Nurse what was the appropriate comment to make in the records and changed it according to that advice. She did not administer any medications and no- one, other than nursing or medical staff, entered the room. She recalls a registered nurse came in to change her drip on the morning of 4 September and that was when the emergency call was made. 31. Registered Nurse (“RN”) Kim was working on ward 4B. She worked from 3:00 pm to 10.30 pm on Saturday 2 September and from 8.30 pm on 3 September until 7:00 am on 4 September. In her first statement she noted that she checked Mrs Johnston at about 2:00 am on 4 September but she later stated that it would have been about midnight when she says Mrs Johnston was snoring heavily and loudly. Almost certainly, RN Kim was the nurse who AIN Smith spoke to. RN Kim says she took Mrs Johnston’s observations (temperature, blood pressure etc) at midnight which were normal, but which were not noted in the records, despite Mrs Johnston’s chart being located with the AIN outside Mrs Johnston’s door. At around 4:00 am when RN Kim was doing her next set of observations she thought Mrs Johnston’s breathing had changed. Mrs Johnston was incontinent to urine and was unresponsive to voice or pain stimuli. A medical emergency (MERT) was called. 32. Dr Rosemary Hay attended the MERT call at about 4:50 am. She was the Assistant Director of Intensive Care at St Andrews. When she arrived it was obvious to her that Mrs Johnston was in a serious condition. Mrs Johnston was unresponsive and febrile and she had laboured breathing. Mrs Johnston also had decerebrate posturing (hyper extension of all limbs) which is a classic sign of severe neurological impairment. Dr Hay immediately transferred Mrs Johnson to the ICU and asked the QML blood collectors to take a series of tests. She also asked the nurses to do a Findings into the death of Margaret Allan Johnston 8 bedside glucose test on the last drop of blood in the syringe which the blood collectors had used. Dr Hay was thinking at this stage that Mrs Johnston had had either a stroke or a cerebral infection. She recalls a nurse then informing her that Mrs Johnston’s blood glucose was 1.7 mmol/L (millimoles per Litre) which was very low, so Dr Hay immediately ordered 50% dextrose intravenously. This did not wake her. Dr Hay ordered various tests to be performed and was still on duty when the blood results from QML revealed Mrs Johnston’s blood glucose level was 0.8 mmol/L. A later c-peptide test result introduced the possibility of exogenous insulin having been administered. 33. Dr Hay was aware that hypoglycaemia can be a consequence of severe sepsis. The only sign which she considered could be consistent with sepsis was Mrs Johnston’s fever. Blood cultures had negative growth. A CT scan done that day showed no evidence of a thrombotic stroke but a subsequent CT scan 4 days later showed evidence of a stroke. Dr Hay had no further involvement in her care. 34. Dr Love saw Mrs Johnston in the ICU later the same day and noted Mrs Johnston’s extremely low blood sugar level. He had never seen anybody present with a blood sugar level that low. The normal range for blood sugar is between 3.2 - 6 mmol/L. He noted that some medical conditions can cause an extremely low blood sugar level such as insulinoma and those with severe liver or adrenal failure, both of which were absent here. Dr Love considered the first report of Professor Marks and was of the view that it had introduced for him a number of interesting possibilities. 35. Professor Peter Silburn is a consultant neurologist. He was requested to provide a neurological opinion on Mrs Johnston by the ICU. He noted her history and progress notes and reviewed her on 6 September 2006. She had a normal CT scan of the brain. He noted no spontaneous movements and there was symmetrical decerebrate posturing to painful stimuli. Her brain stem reflexes were intact. He felt that she had severe diffuse brain dysfunction and damage. He felt her prognosis for useful recovery was poor but felt a repeat brain scan was necessary. The repeat CT scan demonstrated a right occipito-parietal ischaemic infarct which he felt was in the branch of the right middle cerebral artery. There was documented evidence on CT scanning of the brain of a recent cerebral infarction on a background of prior global CNS disturbance. He felt that Mrs Johnston had a very poor prognosis for useful recovery. I will refer to Professor Silburn’s evidence later in this decision. Findings into the death of Margaret Allan Johnston 9