Record of Investigation Into Death (Without Inquest)
Corners Act 1995
Coroners Rules 2006
Rule 11
I, Rod Chandler, Coroner, have investigated the death of Barbara Constance FOX
Find That:
(a) The identity of the deceased is Barbara Constance Fox (‘Mrs Fox’);
(b) Mrs Fox died on 26 April 2013 at the Royal Hobart Hospital (‘the RHH’) in Hobart;
(c) Mrs Fox was born on 7 February 1938 and was aged 75 years;
(d) Mrs Fox died as a result of a bleeding duodenal ulcer.
Circumstances Surrounding the Death:
Mrs Fox resided with her husband at 6/1 Orchard Avenue in Huonville. She ordinarily enjoyed good health. However, in March 2013 she attended the Emergency Department (‘ED’) of the RHH on three occasions. The first occurred on 13 March when she presented with shortness of breath and chest tightness. She was kept for observation overnight and then discharged. The second attendance was on 20 March when she presented with shortness of breath at night. A diagnosis of psychogenic hyperventilation was made related to anxiety and she was discharged to the care of her general practitioner. Three days later Mrs Fox re-presented at the ED. This time she was diagnosed as having had a panic attack and social help was put in place for her.
On 14 April 2013 Mrs Fox had a fall at her home. She was taken by ambulance to the RHH. Facial bruising was noted along with a laceration on the nose. A CT scan of the brain showed a minor traumatic subarachnoid haemorrhage. Blood and other tests showed Mrs Fox’s sodium level was low, she was dehydrated and she had a urinary tract infection. Her urea level was also elevated. She was kept in the ED overnight for observation. Testing of her blood pressure showed her to be hypotensive for many hours. On at least four occasions her blood pressure reached the “emergency call level.”
Mrs Fox was seen by the admitting medical registrar and a plan was written up for her care. However, his notes record; “Not for admission under general medicine team. Will review and consider take-over of care if appropriate.” Mrs Fox was then seen by a registrar from the neurosurgical team who made a note of Mrs Fox’s history and presentation. It was then recorded that a conservative approach should be taken in treating her subarachnoid haemorrhage. It was recommended that Mrs Fox be admitted. However, there was a dispute between the medical registrar and the neurosurgical registrar as to which ward should admit Mrs Fox. In the end she was admitted under the neurosurgical team.
In the early morning of 15 April a Medical Emergency Team (‘MET’) call was made when Mrs Fox’s blood pressure was recorded at 75/29. It was later noted that the blood pressure had been measured with the cuff on the patient’s thigh with the result that the reading was likely to have been 10 to 20% higher than if it had been measured on the arm. Mrs Fox was treated with IV fluids and the MET team was then stood down. After this Mrs Fox was transferred back to the care of the medical team as no ongoing neurosurgical issues appeared to be apparent. The medical team’s review noted that Mrs Fox was now complaining of abdominal pain and had begun to dry retch. Examination of her abdomen showed Mrs Fox to have a tender (not distended) epigastrium and right upper quadrant. Her abdomen was soft and her bowel sounds decreased. She remained hypotensive. No diagnosis was recorded.
A CT scan of the abdomen showed significant thickening of the bowel wall in the duodenum, probably related to duodenal inflammation. The associated air pocket may have represented a duodenal ulcer. The sigmoid colon showed significant diverticular disease. After the scan it was noted that her haemoglobin had dropped from 121 to 103 and her urea level was 20.8. The registrar queried whether Mrs Fox’s blood was diluted or whether she possibly had an internal bleed. A single dose of nexium was given. For the remainder of the admission she was administered ranitidine.
Later on 15 April Mrs Fox was reviewed by the gastroenterology team. They noted the falling haemoglobin level. Mrs Fox was transfused with packed red blood cells. A rectal examination revealed brown stool. A urinary tract infection was diagnosed and treated with antibiotics. The blood folate and vitamin B12 levels were measured but the iron level was not. The urine sodium was measured at <20 mmol/L but the urine osmolality was not measured.
Mrs Fox was seen early on 16 April by the gastroenterology team. It decided to delay a gastroscopy and institute a stool chart to monitor for melena (stomach acid altered blood in the stool). A mini mental-state examination revealed significant cognitive impairment.
On 17 April the medical unit team noted that Mrs Fox’s haemoglobin level had dropped further. However, it was felt that overall there had been an improvement in her condition and that she would probably be suitable for discharge the next day. Later that day she was again seen by the gastroenterology team who also noted the drop in the haemoglobin level. It was recorded that Mrs Fox told them that she was “sure” she had recently seen blood in her stools. That afternoon another unit of packed cells was administered. A nursing note made at 9.00pm describes a moderate volume of very dark coloured stool. A further rectal examination was not performed nor was faecal occult blood testing done.
Mrs Fox was again reviewed by the medical unit team on 18 April. It was noted that she was denying any pain and that she was eating and drinking. She was also reviewed by the gastroenterology team and arrangements made for her to have an outpatient gastroscopy in two weeks. That afternoon she was discharged. It seems no contact was made with her general practitioner concerning her hospitalisation and discharge.
In the early hours of 22 April 2013 Mrs Fox developed upper abdominal pain which she described as “sharp” and feeling “tight.” She was transported by ambulance to the ED at the RHH. The ambulance paramedic’s notes record the presence of melena (ie altered blood in the stool) consistent with a perforated and bleeding ulcer. Mrs Fox was examined and had a chest x-ray. Thereafter she was promptly taken to theatre where a laparotomy was performed by Dr Srini Yellapu and a perforated duodenal ulcer diagnosed. The perforation was closed by oversewing and an omental patch was applied over the ulcer. Dr Yellapu reports that at the time of surgery “there was no obvious active bleeding from the duodenal ulcer.”
Post-operatively Mrs Fox was transferred to the High Dependency Unit and she made a slow but steady improvement. On 26 April she was transferred to a ward. At 12:00 noon that day she was viewed by the ICU liaison nurse who reported her condition to be stable. However, at about 12:30pm a MET call was made when Mrs Fox’s blood pressure had fallen to 85/50 and her pulse rate was 135. The MET team recorded its impression of Mrs Fox in these terms:
“Increased right upper quadrant pain, falling haemoglobin and falling blood pressure and increasing pulse rate secondary to anaemia and hypervolemia.”
A surgical Registrar reviewed Mrs Fox and concluded that there was “no evidence of bleeding.” She was treated with IV fluids and transfused with packed red blood cells after her situation was discussed with the surgeon-on-call. A CT scan of the abdomen was then arranged.
At 5:30pm a further MET call was made when Mrs Fox’s blood pressure dropped to 60/30 and her heart rate was 135. Mrs Fox was transfused with further packed red blood cells and her hypertension was treated with metaraminol. She was then transferred to the CT scan suite. A CT scan showed a significant gastro-intestinal haemorrhage. The surgical staff suspected that Mrs Fox had bled from the omental patch site and decided upon conservative management with the aim of maintaining her blood pressure and general resuscitation. It was noted that there was no role for endoscopy at that stage and the plan was to ensure any coagulopathy was corrected.
Mrs Fox required inotropic medication to maintain her blood pressure. She received further intravenous fluids and had two units of blood transfused. At an unrecorded time she began to complain of back pain which was treated unsuccessfully with fentanyl. Further blood products were given. Mrs Fox’s haemoglobin fell from 95 to 67.
At about 8:00pm Mrs Fox was seen by surgeon, Mr Stephen Wilkinson. He recorded that she required further transfusion and resuscitation along with an anaesthetic assessment. He considered that when she was better resuscitated she would require a re-exploration of her abdomen. At 8:45pm anaesthetist, Dr Reid saw Mrs Fox in the Intensive Care Unit. He noted that she appeared pale, tachycardic, cold and in pain. He also noted that she was receiving cold blood via an iMed pump. She was urgently transferred to theatre.
On arrival in theatre Dr Reid noted Mrs Fox was still pale and hypotensive. When she was being transferred from her hospital bed to the operating table she vomited a large amount of “dark blood.” Suctioning could not clear her airway. At this point she lost consciousness. Cardiopulmonary resuscitation with medications proved ultimately futile. Mrs Fox died at 9:30pm.
Post Mortem Examination:
State Forensic Pathologist, Dr Christopher Lawrence undertook a post-mortem examination. In his opinion Mrs Fox died as a result of a bleeding duodenal ulcer. His report includes the following statement:
“Autopsy reveals a large duodenal ulcer which has perforated and been oversewn. The oversewing is intact. There is however a large branch of the gastro-duodenal ulcer which has been eroded (in the base of) the ulcer and is the source of bleeding. This is approximately 20mm away from the site of perforation and was presumably not seen when the perforation was repaired.
There is some residual bruising from an alleged fall on 17/04/2013 and a resolving area of subarachnoid haemorrhage in the left parietal area but this is unlikely to have contributed significantly to death. There was an unexpected drop in haemoglobin during the earlier admission, a MET call for low blood pressure and she was tender in the epigastrium and I suspect that the duodenal ulcer may have been bleeding during the first admission.”
Investigation and Outcome:
The circumstances of Mrs Fox’s death raise the question why Mrs Fox’s bleeding duodenal ulcer was not diagnosed and treated at an earlier time? A further question is whether Mrs Fox received optimal post-operative care following her surgery on 22 April 2013. The answers to these questions have been the focus of my investigation. That investigation has included the following:
• The interview by telephone of Mrs Fox’s husband.
• The provision of a report by Colorectal, Laparoscopic and General Surgeon, Dr Yellapu.
• The provision of a report by Mr Wilkinson.
• A review of the RHH records undertaken by research nurse Ms Libby Newman.
• Compilation of a report by Dr A J Bell, as medical adviser to the Coroner.
• A review meeting attended by myself, Ms Newman, Dr Bell, Dr Lawrence, and Forensic Pathologist, Dr Donald Ritchey.
When Mrs Fox presented at the RHH on 14 April 2013 her head injury was the medical staff’s initial focus. However, during her admission Mrs Fox had signs and symptoms suggestive of probable gastrointestinal bleeding. Most obvious was her persistent hypotension or low blood pressure which most concernedly was the catalyst for a MET call made in the early hours of 15 April. (It’s pertinent to note that during her three previous visits to the ED Mrs Fox was recorded to have a high or normal blood pressure.) Mrs Fox remained hypotensive over the following days. Her falling haemoglobin level persisted despite transfusion, she had a tender epigastrium, she reported seeing blood in her stool and nursing staff recorded observations of her stool consistent with the presence of blood. Notwithstanding these matters no faecal testing for blood was done nor was any assessment undertaken upon the risk of re-bleeding. Finally, the CT scan taken on 15 April indicated duodenal pathology. Despite these matters Mrs Fox was discharged home without any diagnosis being made to explain her signs and symptomatology.
In his post-mortem report Dr Lawrence suggests that Mrs Fox was suffering from a bleeding duodenal ulcer during this admission. Dr Bell advises me that in his view this was certainly so. I accept this opinion. Dr Bell further advises that the attempts to make and prove a diagnosis for Mrs Fox were “pitiful.” It is a term which, in my opinion, accurately describes the situation.
In my view the medical staff caring for Mrs Fox after her admission on 14 April most certainly should have recognised that she was suffering from upper gastrointestinal bleeding and that she required treatment. That should have involved the immediate re-administration of a proton pump inhibitor (she had previously been given one dose of nexium) to prevent stomach acid secretion and thereby reduce the risk of ongoing bleeding. It also required Mrs Fox to have an endoscopy to confirm the diagnosis and to identify any pathology which required endoscopic attention. Dr Bell advises that had these steps been taken it is likely that Mrs Fox would have made a full recovery from her bleeding duodenal ulcer. I accept this opinion.
When Mrs Fox was returned to the RHH on 22 April 2013 her perforated duodenal ulcer was promptly diagnosed and surgically treated. Dr Lawrence has since reported that the source of Mrs Fox’s duodenal bleeding was identified at post-mortem to be an eroded branch of the gastro-duodenal ulcer sited about 20mm from the site of the perforation. Regrettably, despite its proximity, this bleeding site was not noticed during the course of the surgery and hence was not repaired.
Mrs Fox remained in the High Dependency Unit until 26 April when she was transferred to a ward. At this time her condition was stable as recorded by the liaison nurse. However, shortly afterwards she rapidly deteriorated necessitating the involvement of the MET team. Dr Bell has reported upon this involvement in these terms:
“The first MET call was an appropriate response of the nursing staff to an emergency situation; the sudden deterioration of a patient. The clinical signs clearly show the patient had a substantial bleed with approximately 40% of the circulating blood volume lost. The previous CT scans of the abdomen did not show aortic or vascular disease, thus the diagnosis was almost certainly a gastrointestinal bleed.
The MET doctor(s) and the surgical registrar did not understand how severe the bleeding was. Again supine hypotension indicates that the patient has lost 40% of the blood volume and in this case rapidly. The check by the ICU liaison nurse showed the patient had normal vital signs 50 minutes prior to the MET call.
The MET was ceased at 13:40 hours for reasons that are not documented. The patient was hypotensive and tachycardic, there was no diagnosis and a minimal plan. There is no documentation that the ICU consultant was involved in the patient’s care. There is minimal record of the events after the initial MET call until the second MET call. The care provided was substandard. The sudden large bleed that appears to suddenly stop is typical of arterial bleeding. The treatment provided was substandard in all aspects.
The MET was called again at 16:12 hours. The MET call team appears to have performed minimal resuscitation and the patient remained hypotensive and tachycardic when transferred to the CT scanner. The patient was transfused and treated with the drug metaraminol. This drug should not be used until the circulating blood volume has been restored. The management was determined to be conservative but the patient either continued to bleed or had a further substantial bleed. Sudden large bleeds that cease rapidly are typical of arterial bleeding. The care provided was substandard. The second major (sudden) haemorrhage indicates that the patient has an arterial bleed presumably related to the known ulcer. This situation is an indication for surgery.”
In his report Mr Wilkinson includes these observations specific to the involvement of the MET team:
• “In retrospect, it would have been more appropriate after the first MET call to transfer the patient to ICU for more vigorous fluid management, then to CT scanning from there. Because there were no indications of the source of any possible bleeding, it appears as though the volume loss was under-appreciated. With such an elderly patient with many co-morbidities, it would have been a better choice to transfer her to ICU early.”
And;
• “This case indicates that there need (sic) to be changes to the MET call protocol, whereby consultant level notification of the MET call needs to be mandated.”
It is clear at the time of the first attendance of the MET team at 12.30pm that Mrs Fox had suffered a sudden and very significant bleed. As Dr Bell has noted, it represented a 40% loss of her circulating blood volume. The MET team responded by commencing a transfusion of blood and ordering a CT scan of the abdomen. Mrs Fox was left on the ward pending the scan. This response was grossly inadequate. Instead, it should have been appreciated that Mrs Fox most likely required urgent open surgery to repair the bleed and preparatory steps needed to be immediately taken. This required Mrs Fox’s transfer to ICU and her resuscitation in a closely monitored setting. It also required an urgent CT scan of the abdomen to identify the source of the bleeding. Rather than initiate these steps the MET team stood itself down and Mrs Fox was left on her ward pending the non-urgent CT scan. This scan had not been done by 5.30pm when the second MET call was made after Mrs Fox had a further major bleed. By this time Mrs Fox’s condition had clearly worsened and the opportunity had been lost to salvage the situation.
I am advised by Dr Bell that a significant mortality rate (in the region of 10 to 25%) attaches to the arterial bleed suffered by Mrs Fox. However, he further advises that if surgical repair had been promptly initiated following the first MET call there was a real probability that Mrs Fox would have survived. I accept this advice.
Conclusion:
When Mrs Fox was first admitted to the RHH she was suffering from a bleeding duodenal ulcer. This is not ordinarily a fatal condition and can usually be readily and successfully treated. Unfortunately it was not diagnosed when it should have been. This set in motion a train of events which culminated in Mrs Fox’s death. There was a final opportunity to salvage the situation when Mrs Fox had a substantial bleed on 26 April 2013 but this opportunity was not taken because of the inadequate response of the MET team. In the result there has been a death which should have been prevented. This is a most regrettable outcome.
Comments and Recommendations:
In 2010 I handed down findings into the death of Mrs Beryl Walters. That case also involved the RHH’s MET team. I was advised that following its own investigation of that death the RHH introduced a protocol whereby the MET team was to be controlled by an ICU consultant and the team could only be stood down by that consultant. It is apparent from my investigation of Mrs Fox’s death that this protocol has been abandoned. I venture the view that had it been in place for Mrs Fox the shortcomings surrounding her first MET call may have been avoided. This leads me to recommend that the RHH conduct a review of its MET team with a view to adopting changes which address those deficiencies that have presented in this case.
I have decided not to hold a public inquest into this death because my investigations have sufficiently disclosed the identity of the deceased, the date, the place, cause of death, relevant circumstances concerning how her death occurred and the particulars needed to register her death under the Births, Deaths & Marriages Registration Act 1999. I do not consider that the holding of a public inquest would elicit significant information further to that disclosed by the investigations conducted by me. The circumstances of the death do not require me to make any further comment of to make other recommendations.
I conclude this matter by conveying my sincere condolences to Mrs Fox’s family.
DATED: 17 November 2014 at Hobart in the state of Tasmania.