Inquest into the death George TRIANTAFILOPOULOS

STATE CORONER’S COURT OF NEW SOUTH WALES Inquest: Inquest into the death of George Triantafilopoulos Hearing dates: 8-9 December 2016 Date of findings: 20 December 2016 Place of findings: State Coroners Court, Glebe Findings of: Magistrate Harriet Grahame, Deputy State Coroner Catchwords: Coronial Law- fall, anticoagulation therapy and closed head injury File number: 2014/242135 Representation: Mr Stephen Kelly –Sergeant, Coronial Advocate Ms Jackie Sandford of counsel, instructed by HWL Ebsworth for Dr Karalasingham Mr Cameron Jackson of counsel, instructed by Avant for Dr Sivaseelan Findings: On the balance of probabilities, I find that George Triantafilopoulos died on 16 August 2014 at Royal Prince Alfred Hospital. He died of an acute right subdural haemorrhage, sustained in a fall. He was taking anticoagulant medication at the time of his death.

Recommendations I recommend that the Royal College of General Practitioners consider including an anonymised report of this death and the issues involved in its weekly newsletter to assist in the ongoing education of GPs on the subject of the assessment of closed head injuries in older people who are taking anticoagulation medication.

I request that a copy of these findings is sent to the Royal Australian College of General Practitioners for this purpose.

THIS DECISION HAS BEEN PREPARED WITHOUT THE BENEFIT OF A TRANSCRIPT.

IN THE STATE CORONER’S COURT GLEBE NSW SECTION 81 CORONERS ACT 2009 REASONS FOR DECISION

1. This inquest concerns the death of George Triantafilopoulos Introduction

2. George Triantafilopoulos was a 75 year old man who was living alone at Marrickville at the time of his death. He was born in Greece and had been living in Australia since

3. He married his wife Panayiota soon after coming to Australia and worked as a rail labourer. His wife was active in the Greek Orthodox Church in Newtown and introduced George to some of the friends she made there. Over the years George became friendly with George and Koula Koutsoubaris. He remained close to them after his wife’s death in 2005.

4. Over time George Koutsoubaris became George’s closest friend and carer. He visited George on a daily basis. He brought meals and helped him with shopping and attending appointments. Mr Koutsoubaris also assisted with translation when necessary. In 2011, George registered Mr Koutsoubaris as the beneficiary of his will and his legal next of kin. After George’s death Mr Koutsoubaris assisted the coronial investigation.

Unfortunately prior to the inquest commencing Mr Koutsoubaris died.1There appear to be no other family members or friends with an active interest in these proceedings.

4. George had a regular doctor, Dr Peter Calligeros, who had been treating him for many years. George had been treated for a number of conditions including anaemia and 1 Evidence of Constable Christopher Vea, 8 December 2016

diabetes mellitus. In 2001 or 2002 he had undergone cardiac surgery which involved the replacement of an aortic valve. He had been on warfarin since that time.2 The role of the coroner and scope of the inquest.

5. The role of the coroner is to make findings as to the identity of the nominated person, and in relation to the date and place of death. The coroner is also to address issues concerning the manner and cause of the person’s death.3 In addition, the coroner may make recommendations in relation to matters that have the capacity to improve public health and safety.4

6. In this case there is no dispute in relation to the identity of the deceased, or in relation to the time and place of death. The inquest focussed on the medical treatment George received. The inquest also touched on questions about whether his death could have been prevented.

The Inquest

7. A list of issues relevant to George’s death was circulated prior to the inquest commencing. The following questions were posed  Was the care and treatment provided by Dr Rupasenan Karalasingham at Marrickville Metro Medical Centre, Marrickville on 14 August 2014 appropriate in the circumstances?

 Was the care and treatment provided by Dr Thillai Sivaseelan at Primacare Family Medical Centre, Roselands on 15 August 2014 appropriate in the circumstances?

 Should Mr Triantafilopolous have been referred to an Emergency Department?

 Should either Dr Karalasingham or Dr Sivaseelan have referred Mr Triantafilopoulos for a CT scan?

2 Exhibit 1, Tab 21 3 Section 81 Coroners Act 2009 (NSW) 4 Section 82 Coroners Act 2009 (NSW)

8. After a preliminary argument in relation to jurisdiction was resolved, the inquest proceeded over two hearing days. A number of statements and medical reports were tendered. Oral evidence was also received, including from independent experts and the doctors involved in George’s immediate care. Although initially it had been foreshadowed that Mr Koutsoubaris may give evidence in these proceedings, the Court was informed that he had died in September 2016. His statements were nevertheless tendered.

The first fall – Marrickville, 14 August 2014

9. On 14 August 2014 George Triantafilopoulos and his friend Mr Koutsoubaris were at the Aldi Supermarket at Marrickville Metro. According to Mr Koutsoubaris, as they were leaving George fell backwards and onto the ground, hitting his head on the floor.5 Mr Koutsoubaris reported that it made a loud banging noise. George told his friend that he felt a bit of pain. Aldi staff helped Mr Koutsoubaris take George to the Marrickville Medical Centre to get checked. This was partially captured on CCTV6.

10. Mr Koutsoubaris was present during the consultation. He said “I saw the Doctor check his blood pressure, look into his eyes and ears. I informed the doctor that he was on current medication for type 2 Diabetes, 4mg of Walfarin (sic) as well as insulin daily. I also informed the doctor that he has had 2 heart operations within the past 15 years.”7 According to Mr Koutsoubaris, the doctor told George that “everything would be okay”.

He prescribed Panadeine Forte and told him that if he felt dizzy or worse to call an ambulance.

11. The doctor that George saw on this occasion was Dr Rupasenan Karalasingham. He gave an account of the consultation firstly in a statement to police in November 2014 and later in oral evidence before this court 8. He stated that Mr Triantafilopoulos “spoke hardly any English” so most of the communication was through Mr Koutsoubaris. The doctor agreed he was told that George was taking various medicines. Dr Karalasingham stated that he knew George was on “walfarin(sic)” and inquired when the last INR test 5 See Exhibit 1, Tab 22, Statement of George Koutsoubaris, paragraph 7 onwards 6 Exhibit 1, Tab 28 7 Exhibit 1, Tab 22, Statement of George Koutsoubaris, paragraph 8 8 Oral Evidence and also see Exbibit 1, Tab17 Statement of Dr Rupasenan Karasingham

had been done. He was told that it had been done only days before and was in the therapeutic range.

12. Dr Karalasingham told police that he examined George and took his blood pressure, pulse and checked his respiratory rate. He did a “neurological examination”, noting that his pupils were equal and reactive and the power in his arms and legs were normal.

There was no unusual bleeding and he appeared alert and responsive. Dr Karalasingham stated that he gave him a Glasgow Coma scale (GCS) level of 15.

13. Dr Karalasingham stated that as a result of that examination he believed that George had suffered a minor concussion and he prescribed Panadeine Forte for headache.

However because George was taking warfarin, Dr Karalasingham was alert to the possibility of “the potential to bleed”. In his written statement, he said that he instructed Mr Koutsoubaris to keep a close eye on him for the rest of the day and take him to Royal Prince Alfred Hospital if there was any change in consciousness, worsening of the headache or severe vomiting.9 In oral evidence Dr Karalasingham confirmed that he was aware of the increased risk of bleeding on the brain due to the fact that George was taking anticoagulation therapy. He said that he explained there were two options, one was that he could write a letter and send George to casualty for an immediate cat scan and the other option involved close observation and attendance should symptoms develop. This was different to his earlier account which did not mention the cat scan option.

14. In oral evidence Dr Karalasingham also indicted that although he had not recorded it in his notes, he had been told that George had hit the back of his head on a plastic shopping trolley handle, not on the ground. This was in direct conflict with Mr Koutsoubaris’s account, which was that George had hit the back of his head on the ground. Given that Mr Koutsoubaris is no longer alive and available for cross examination, it is difficult to finally resolve this discrepancy, however given the complete lack of a written record in this regard I place little weight on Dr Karalasingham’s later recollections on this issue.

9 Statement of Dr Karalasingham, Exhibit 1, Tab13, paragraph 13

15. Dr Karalasingham gave evidence that as well as being a city GP, he had considerable experience working in the emergency departments of rural hospitals. He often relied on telephoning experts or senior colleagues to consult about patients who arrived for care. He had recently sought and received advice about an 84 year old patient on warfarin who had fallen and hit her head. Dr Karalasingham had been advised not to send her to the neighbouring town for a CT scan. This advice was at the forefront of his mind and seemed to weigh heavily on him when he was called upon to treat George.

Given that he believed George had a GCS of 15 and was in the constant care of his friend Mr Koutsoubaris, Dr Karalasingham explained that felt comfortable sending him home, under observation with instructions to seek help if anything changed.

16. In any event, Mr Koutsoubaris took George home after the appointment. George apparently decided that he did not want the pain killer and told his friend that he was fine. Given Mr Koutsoubaris’s solicitous care of his friend, one can safely assume that George was not suffering any extreme or obviously worrying symptoms or I am quite confident that Mr Koutsoubaris would not have left him alone.

The second fall – Roselands, 15 August 2014

17. Around 8.45 am on 15 August Mr Koutsoubaris met George at a coffee lounge on Illawarra road, Marrickville. George appeared to be moving and talking in a normal manner. The pair decided to go shopping and Mr Koutsoubaris drove them to Roselands Centro Shopping Centre. Mr Koutsoubaris parked in the disabled zone outside Coles and they got out of the car. Mr Koutsoubaris was unable to lock his car with the remote so he walked back to the car to manually close the door. When he turned back around he saw George was face down on the ground. He had a large scratch on his head and he was bleeding.10 He was on his hands and knees trying to get up. The men were quickly assisted by a staff member who offered to call an ambulance, which George refused. A wheel chair was obtained and first aid was commenced for the cut. Mr Koutsoubaris went to buy an iced tea as he thought the fall may have had something to do with George’s diabetes.

10 See Exhibit 1, Tab 22, Statement of George Koutsoubaris, paragraph 11 onwards

18. George was taken in a wheel chair to the Primacare Family Medical Centre where he was seen by Dr Thillai Sivaseelan. This is captured on CCTV. Again Mr Koutsoubaris was present and assisted with translation. He gave this doctor the same information about George’s medical history and current medications. According to Mr Koutsoubaris the doctor checked George’s blood pressure, his eyes and ears and a nurse checked his blood sugar. The blood sugar was a bit high and Mr Koutsoubaris told the doctor that he had just given George some iced tea. According to Mr Koutsoubaris the doctor told George he was okay and that it “was nothing serious”. The doctor told George that if he did not feel okay in four to five hours he should call the ambulance and go to hospital.

The doctor treated the scratch on George’s head and hand and they left soon after.

19. Dr Sivaseelan made an initial statement to police in November 2014. It appears to be taken from his contemporaneous clinical notes. He stated “in my opinion he had a closed head injury after a fall with no abnormal neurological signs. The head abration(sic) was cleaned and dressed. He was adviced(sic) to go to hopital(sic) if persistant (sic)head ache, persistant(sic) vomiting, unuasual drowsy(sic), fitting/seizure or if any concers(sic)11. In September 2016 he made a more comprehensive statement. In that statement, among other things, Dr Sivaseelan stated that the consultation lasted over half an hour and numerous investigations were carried out. He stated that in his assessment George had a GCS of 15, all neurological signs were normal, blood sugar levels were tested and the wound was cleaned and dressed. Dr Sivaseelan stated that he told George and Mr Koutsoubaris, “someone has to observe George closely for the next 4-5 hours, and keep an eye on him for the next 12 hours. If George suffers a persistent head-ache or vomiting, is unusually drowsy, or has a fit or seizure, or there are any concerns, you need to call an ambulance, or go to hospital”12

20. In oral evidence Dr Sivaseelan confirmed that he had no recollection of having been informed about the fall the previous day. He stated it would have been his practice to have noted that kind of important information and there was no record of it. It is very surprising to me that during a consultation that is said to have lasted half an hour, Mr 11 Statement of Dr Sivaseelan, Exhibit 1, Tab18 12 Statement of Dr Sivaseelan (dated 21/9/16) paragraph 29

Koutsoubaris did not mention the events of the previous day. It seems inherently implausible. It was an important part of George’s medical profile at that point and if raised, should have caused further concern. Mr Koutsoubaris made a second statement in March 2016 13 where he confirmed that he told Dr Sivaseelan about the first fall.

However, he was not available to shed further light on this issue at the inquest and the exact circumstances in relation to the making of that statement are unclear.

21. After seeing Dr Sivaseelan, George and Mr Koutsoubaris continued shopping and then had lunch at Mr Koutsoubaris’s house. George appeared to be talking and walking normally and Mr Koutsoubaris had no concerns when he dropped George home. He told George to call him if anything happened and Mr Koutsoubaris would get an ambulance.

22. Around 3pm it appears that George called and spoke to Mrs Koutsoubaris. He told her that he was vomiting and had called an ambulance.

Admission to Hospital

23. An ambulance was called to George’s address. It appears that George made the call just after 3pm. He was described as conscious and breathing but he was “not completely alert”, “not responding appropriately”.14

24. By the time the ambulance arrived, George was unconscious and there was vomit on the floor. Ambulance officer commenced treatment and George was taken to Royal Prince Alfred Hospital.

25. George was admitted to Hospital and examined by Dr Lee Juan Chiang15 in the Emergency Department. At that time he was unresponsive with a Glasgow Coma scale of 4/15 and severe hypertension. He was already intubated and was placed on a ventilator.

26. An urgent CT scan of his brain was ordered. It showed that George had a large right sided subdural haematoma overlying the right cerebral hemisphere with significant 13 Statement of George Koutsoubaris, Exhibit 1, Tab 23 (Statement of 31/3/2016) 14 See Ambulance Incident Report, Exhibit 1, Tab 26 15 See Statement of Dr Lee Juan Chiang, Exhibit 1, Tab 19

mass effect. Subdural blood was also noted in other areas of the brain. There was no skull fracture.16 The neurosurgical team consulted and were of the view that surgical intervention would be futile. Palliative care was commenced.

27. George was pronounced life extinct at 2.15 am on Saturday 16 August 2014, by Dr Fiona Duncan.17 The Autopsy Report

28. An autopsy was subsequently conducted.18 The direct cause of death was given as an acute right subdural haemorrhage, arising from a fall. It was noted that George had been receiving anticoagulation therapy for his metallic valve replacement and that there was a presence of severe three vessel coronary artery atherosclerosis.

Dr Vincent Roche’s evidence

29. Dr Roche provided a short report and gave more expansive oral evidence. He was an experienced general practitioner and had worked extensively as a visiting medical officer in the Accident and Emergency Department at the Bowral and District Hospital for many years.

30. Dr Roche made the point that not only were doctors now dealing with an increasingly aging population, but that there had been a massive increase in the number of people taking anticoagulant medications of one sort or another over the last 10 years. In general terms he doubted the utility of GPs requesting a CT scan of every patient who presented, who had suffered a fall and was over 65 and on anti-coagulant medication.

He believed that GPs often had a filtering role in the health care system and their task was slightly different to that of an emergency doctor. In his view, taking a full history was more useful than following a guideline which mandated scanning on all occasions.

31. Dr Roche referred to the potential resourcing difficulties that could flow from a policy where there was a blanket rule mandating a cat scan for each and every older person who was taking an anticoagulant medication and fell. He referred the court to the 16 Statement of Dr Lee Juan Chiang, Exhibit 1, Tab 19 17 See statement of Dr Fiona Duncan, Exhibit 1, Tab 20 18 Autopsy Report, Exhibit 1, Tab 35

possible repercussions of a recent coronial decision in South Australia.19 That decision dealt with a nursing home patient who fell from a chair. She was receiving warfarin at the time. While she was observed following the fall, she was not sent for a scan. She later died of an acute subdural haemorrhage. It is noted that she suffered a number of other health issues, in particular dementia, which can make assessing whether or not there has been cognitive impairment after a fall more difficult. Dr Roche expressed the view that introducing a blanket policy of cat scanning patients in these circumstances would result in an enormous number of scans being undertaken for no benefit, in other words he said, “the yield is low”.

32. Dr Roche reviewed Dr Karalasingham’s treatment of George. He could not find fault with Dr Karalasingham’s history taking, examination or management plan. He found it “exemplary and more comprehensive than that of an average GP in this situation.”20 He did not believe that Dr Karalasingham ought to have referred George for a CT scan, given his presentation on 14 August 2014. He maintained this position even knowing that George had not just fallen, he was reported to have hit his head.

33. Dr Roche only specifically commented on Dr Karalasingham’s consultation, but it followed from what he said that he believed that both of the GPs involved had acted appropriately in the circumstances they struck.

Professor Raftos’s Opinion.

34. The Court was also assisted by the evidence of Associate Professor Dr John Raftos, an experienced emergency medicine specialist.21 He reviewed the medical records and circumstances existing before George’s death. He identified George’s age and the fact that he was taking anticoagulation medication (warfarin) as significant risk factors. In his view, in the circumstances of this case, a CT brain scan should have been performed urgently on each occasion. In his written report, Dr Raftos expressed the view that failure to order a scan, in these circumstances, was “a departure from the accepted standard of care and would expose the patient to an unnecessarily increased risk of 19 Findings of an Inquest into the death of Marie Janet Ford (8 February 2016) Decision of Mark Frederick Johns, State Coroner of South Australia. Exhibit 3 20 Report of Dr Vincent Roche, Exhibit 2 21 Expert report of Associate Professor John Raftos, Exhibit 1, Tab16

death or neurological disability because of an undiagnosed and expanding subdural haematoma”

35. During his oral evidence Dr Raftos’s position softened to some degree. He remained of the view that patients over 65, who strike their heads and are on anticoagulants should be scanned, however he agreed that there were appreciable differences between the resources and options available to GPs on the one hand and doctors working within emergency departments on the other. He appeared to agree that in practice GPs helped filter the case load presenting to hospital. He expressed the view that sometimes taking a detailed history was difficult in a hospital emergency department environment and that it was often faster to send patients for a quick CT scan, which might take 5 minutes.

36. While Dr Raftos was not generally supportive of prescriptive guidelines, and preferred an approach where doctors relied on their clinical judgement, he was aware of the NSW Health Policy “Initial Management of Closed Head Injury in Adults”22 and thought the document was useful. He was taken specifically to a guideline or algorithm, entitled “Initial Assessment of Adult Closed Head Injury”23, which outlined the risk factors and offered pathways to initial treatment and care. In his view it was a “good document” and could usefully be circulated to GPs in their weekly newsletter or by some other means.

37. Dr Raftos stated that early reversal of the anticoagulation and the possibility of neurosurgical evacuation of the subdural haematoma would have increased George’s chances of survival, 24 I accept that position. However, in my view, it is impossible to say, with any certainty, whether or not George would have survived and if so, in what condition if a scan had been done at any given time before his death. In oral evidence the experts appeared to agree that even if a bleed had been identified at an earlier stage, the prognosis was not good. It is certainly possible that George may have survived, but it is also possible that while the anticoagulation problem could have been 22 Initial Management of Closed Head Injury in Adults” NSW Ministry of Health, Exhibit 1, Tab 23 Initial Management of Closed Head Injury in Adults” NSW Ministry of Health, Exhibit 1, Tab 31, page 8 24 Report of Dr John Raftos, Exhibit 1, Tab 15

addressed, there may nevertheless have been no appropriate surgical treatment available, given George’s age and co-morbidities. A scan may have been useful in diagnosis or in clearing the way for appropriate palliative care, but may not have saved his life.

Conclusion

38. Having reviewed the evidence I am of the view that both general practitioners took adequate histories from a patient they had never encountered before. While I am somewhat troubled by Dr Sivaseelan’s assertion that he did not know George had also fallen the day before, it is a question of fact that cannot now be resolved. I accept both doctors conducted appropriate physical examinations and were of the view that George was alert and responding normally when he left their respective surgeries. They both considered the risk involved in his anticoagulation. George clearly did not want to go to Hospital and had already refused an ambulance. After some consideration both doctors left him in the care of a dedicated and seemingly capable friend. Mr Koutsoubaris, was himself content to leave George at home on the first day, and continue shopping on the second day. It strikes me as quite implausible that Mr Koutsoubaris would have done this if he had identified any significant change in his friend or any cognitive depreciation. This certainly supports the view that George was not exhibiting any concerning cognitive deficiency or any lowered level of consciousness when he was with either doctor.

39. I accept that both doctors exercised adequate care, under the circumstances. While their decisions were based on their clinical findings and the history they took, in my view it was certainly open to them to have taken a more conservative approach. It is a matter of delicate clinical judgement. Abundant caution would have seen them refer George to a local hospital, but given his presentation I understand the decisions they made at the time. In retrospect more emphasis could have been placed on the mechanism of injury. It was not just that George fell, in the first instance Mr Koutsoubaris reported that he fell on concrete and it made a loud bang.25On the second occasion he had an obvious head injury and an abrasion was treated. After each fall, George appeared relatively normal, however at some stage over the two days he began 25 I accept that this is not Dr Karalasingham’s memory of the report.

to bleed in his brain. The effects were not immediate and therein lies the need for caution. It is interesting to note that both doctors expressed, in their oral evidence, that with the benefit of hindsight if faced with the situation again, they would both send George for an immediate CT scan. It strikes me that other general practitioners might also find the opportunity to review this case study useful.

40. The College of General Practitioners did not appear at the inquest, but provided information to the court, indicating that it had no current written policy guiding members on the initial assessment of closed head injuries.

41. I have carefully considered whether the evidence before me suggests that recommendations should be made in relation the assessment of closed head injuries by general practitioners. One must always be wary of making broad recommendations on the facts of a single case. There is no evidence to suggest that this is a widespread problem, however given what Dr Raftos described as an “explosion” in the number of older people taking anticoagulation therapy, it is certainly an issue worth considering very carefully. What emerged was the importance of clinical judgement based on a comprehensive history and physical examination. It is also essential that doctors are aware of the known risk factors and approach the decision of whether to order a scan from an evidence-based understanding of those risks. As more people in the community continue to take anticoagulation therapies, the real number of old people at serious risk of life threatening bleeds will continue to rise. It is a situation worth ongoing review.

42. I am of the view that some general practitioners may benefit from re-thinking their approach to this issue, in the light of the issues raised in this inquest. While I do not suggest that the College should necessarily adopt the Department of Health policy, it is as Dr Raftos said “a good document” and one well worth considering. On the limited evidence before me, arising from a single case, I do no more than alert the College to the issues raised.

Findings

43. On the balance of probabilities, I find that George Triantafilopoulos died on 16 August 2014 at Royal Prince Alfred Hospital. He died of an acute right subdural haemorrhage, sustained in a fall. He was taking anticoagulant medication at the time of his death.

Recommendations

44. I recommend that the Royal College of General Practitioners consider including an anonymised report of this death and the issues involved in its weekly newsletter to assist in the ongoing education of GPs on the subject of the assessment of closed head injuries in older people who are taking anticoagulation medication. I request that a copy of these findings is sent to the Royal Australian College of General Practitioners for this purpose.

45. I close this inquest.

Magistrate Harriet Grahame Deputy State Coroner 20 December 2016