Mabior by her Next Friend Mary Kelei v Child and Adolescent Health Service
[2018] WADC 12
•2 FEBRUARY 2018
JURISDICTION : DISTRICT COURT OF WESTERN AUSTRALIA
IN CIVIL
LOCATION: PERTH
CITATION: MABIOR by her Next Friend MARY KELEI -v- CHILD AND ADOLESCENT HEALTH SERVICE [2018] WADC 12
CORAM: DERRICK DCJ
HEARD: 27-28 NOVEMBER, 1 & 4-8 DECEMBER 2017
SUPPLEMENTARY WRITTEN SUBMISSIONS FILED ON 18, 22 AND 23 JANUARY 2018
DELIVERED : 2 FEBRUARY 2018
FILE NO/S: CIV 3062 of 2008
BETWEEN: SUNDAY JOHN MABIOR by her Next Friend MARY KELEI
Plaintiff
AND
CHILD AND ADOLESCENT HEALTH SERVICE
Defendant
Catchwords:
Torts - Medical negligence - Standard of care - Breach of duty - Causation
Paediatric burns patient - Failure to diagnose sepsis - Delay in treatment for sepsis
Legislation:
Civil Liability Act 2002 (WA)
Evidence Act 1906 (WA)
Result:
Judgment for plaintiff
Representation:
Counsel:
Plaintiff: Mr T Lampropoulos SC
Defendant: Mr D R Clyne
Solicitors:
Plaintiff: Maurice Blackburn
Defendant: Panetta McGrath
Case(s) referred to in judgment(s):
Adeels Palace Pty Ltd v Moubarak [2009] HCA 48; (2009) 239 CLR 420
Amaca Pty Ltd (Formerly James Hardie & Co Pty Ltd) v Hannell [2007] WASCA 158; (2007) 34 WAR 109
Bennett v Minister of Community Welfare [1992] HCA 27; (1992) 176 CLR 408
Briginshaw v Briginshaw [1938] HCA 34; (1938) 60 CLR 336
Chappel v Hart [1998] HCA 55; (1998) 195 CLR 232
City of Stirling v Tremeer [2006] WASCA 73; (2006) 32 WAR 155
Commissioner of Main Roads v Jones [2005] HCA 27; (2005) 79 ALJR 1104
Department of Housing and Works v Smith [No 2] [2010] WASCA 25; (2010) 41 WAR 217
Dovuro Pty Ltd v Wilkins [2003] HCA 51; (2003) 215 CLR 317
Hunt & Hunt Lawyers v Mitchell Morgan Nominees Pty Ltd [2013] HCA 10; (2013) 247 CLR 613
La Macchia v Minister for Primary Industries & Energy (1992) 110 ALR 201
March v E & MH Stramare Pty Ltd [1991] HCA 12; (1991) 171 CLR 506
Medlin v The State Government Insurance Commission [1995] HCA 5; (1995) 182 CLR 1
Mustac v Medical Board of Western Australia [2007] WASCA 128
Naxakis v Western General Hospital [1999] HCA 22; (1999) 197 CLR 269
Panagoulias (by his next friend Fiona Averil Panagoulias) v The East Metropolitan Health Service [No 4] [2017] WADC 118
Purkess v Crittenden [1965] HCA 34; (1965) 114 CLR 164
Richard Warren Dorsett as Administrator of the Estate of Andrew Warren Dorsett (DEC) v Janeska [2005] WASCA 215
Roads and Traffic Authority (NSW) v Refrigerated Roadways Pty Ltd [2009] NSWCA 263; (2009) 77 NSWLR 360
Rogers v Whittaker [1992] HCA 58; (1992) 175 CLR 479
Rosenberg v Percival [2001] HCA 18; (2001) 205 CLR 434
Seltsam Pty Ltd v Ghaleb [2005] NSWCA 208
South Australia v Ellis [2008] WASCA 200; (2008) 37 WAR 1
South Metropolitan Health Service v Westcott [2016] WASCA 225
Strong v Woolworths Ltd t/as Big W [2012] HCA 5; (2012) 246 CLR 182
Undershaft (No 1) Ltd v Federal Commissioner of Taxation [2009] FCA 41; (2009) 175 FCR 150
Van Der Velde v Halloran [2011] WASCA 252
Wallace v Kam [2013] HCA 19; (2013) 250 CLR 375
Watts v Rake [1960] HCA 58; (1960) 108 CLR 158
Wright v Minister for Health [2016] WADC 93
Wyong Shire Council v Shirt [1980] HCA 12; (1980) 146 CLR 40
Table of Contents
Introduction
Factual background
The incident
The course of events following the plaintiff's arrival at PMH up until 9.45 pm on 11 December 2005
The course of events following the plaintiff's admission to the ICU
The cases of the parties – summary
The witnesses
Questions for determination
Medical terms and medical conditions
Infection
Oedema
Neutrophils
Neutropenia
CRP
Systemic Inflammatory Response Syndrome
ARDS
Sepsis
Septicaemia
Septic shock/toxic shock
Fluid overload
Did the plaintiff have sepsis on 10 December 2005 or at any time thereafter?
The relevant evidence
The plaintiff's expert witnesses
Dr Andrew Numa
Professor Alison Kesson
Professor Michael Ditchfield
The defendant's expert witnesses
Dr Jeffrey Prebble
Dr Hugh Allen
Associate Professor Mike Starr
Professor Roy Kimble
Associate Professor John Harvey
Dr Fiona Bettenay
Dr Conor Murray
The documentary evidence
The ICU notes
Four additional documents
Consultation request (exhibit 26)
Discharge report (exhibit 23)
Letter from Dr Jane Valentine (exhibit 24)
Letter from Dr Kay (exhibit 25)
Analysis and decision
What was the standard of care that the defendant had to meet?
Did the defendant breach its duty of care by failing to detect and treat the sepsis with antibiotics prior to the morning of 12 December 2005?
The relevant provisions of the Civil Liability Act 2002 (WA)
The relevant evidence
The plaintiff's expert witnesses
Dr Numa
Professor Kesson
The defendant's expert witnesses
Dr Prebble
Dr Allen
Associate Professor Starr
Professor Kimble
Analysis and decision
Was the breach of duty the cause of the injuries?
The relevant provisions of the Civil Liability Act 2002 (WA)
The relevant evidence
The plaintiff's expert witnesses
Dr Numa
Professor Kesson
The defendant's expert witnesses
Dr Allen
Associate Professor Starr
Analysis and decision
Conclusion
DERRICK DCJ:
Introduction
The plaintiff claims damages for injuries alleged to have been caused by the negligence of the team of doctors engaged by the defendant to care for and treat the plaintiff while she was a patient in the burns ward at Princess Margaret Hospital (PMH) during the period 9 to 11 December 2005. The court has, at this point, been asked to determine only the issue of liability.
Factual background
It is necessary at the outset to set out the non-contentious factual background to the plaintiff's claim.
The incident
The plaintiff was born on 8 August 2004 in Sudan. She is the third of six children born to her parents.
A short time after the birth of the plaintiff she and her family migrated to Australia.
In December 2005 the plaintiff was living with her mother, father and two siblings in Marangaroo.
At around 5.00 pm on Friday, 9 December 2005 the plaintiff suffered hot water scald burns to her chest, forearms, proximal regions of both hands and knees. The burns included partial thickness burns. In the course of suffering the burns the plaintiff did not inhale steam into her lungs. She therefore did not suffer any direct injury to her lungs.
The plaintiff's mother, Ms Mary Kelei, took the plaintiff to the Mirrabooka Medical Centre where she saw a doctor. The doctor provided the plaintiff with some initial treatment and referred her to the emergency department at PMH. He made arrangements for an ambulance to collect the plaintiff and her mother from the medical centre and to take them to PMH.
The plaintiff was taken to PMH. She arrived at the emergency department at about 6.20 pm.
The course of events following the plaintiff's arrival at PMH up until 9.45 pm on 11 December 2005
At 7.00 pm the plaintiff was seen by a surgery burns registrar in the emergency department. The plaintiff's temperature was 36.9 degrees Celsius. Her heart rate was 190 (beats per minute). Her respiratory rate was 50 (breaths per minute). Her oxygen saturation level while on oxygen therapy was 100%. Her lungs were clear. The registrar assessed the plaintiff as having superficial and partial thickness burns to approximately 18% of her total body surface area (TBSA).
At 8.40 pm the plaintiff was admitted to the PMH burns ward.
Throughout her time in the burns ward the plaintiff was given intravenous fluids to protect against fluid loss caused by her burns. She was also given appropriate pain medication.
At the time of her admission to the burns ward the plaintiff's temperature was 36.6 degrees Celsius (the normal range for a child of the plaintiff's age being 36.0 to 37.0 degrees Celsius). Her heart rate was 144 (the normal range for a child of the plaintiff's age being 90 to 140). Her respiratory rate was 28 (the normal range for a child of the plaintiff's age being 28 to 40). She was taken off oxygen therapy at this time.
At 8.50 pm various swabs were taken from burns on the plaintiff's chest, left knee and right knee. A screen of the 'nose and peri/t' for Multi Resistant Staphylococcus Aureus (MRSA) was also conducted. All of the swabs and the MRSA screen, save for the swab of the right knee burn, produced negative results.
The microscopic 'gram stain' examination of the right knee swab produced a positive result for bacteria (Gram-negative bacilli). This result was available to medical staff responsible for the plaintiff's care by the afternoon of 10 December 2005, if not earlier on that day.
The culturing of the right knee swab resulted in the identification of the 'Enterobacter' species of bacteria with '+++ growth'. The microbiology report relating to the right knee swab, which was available to medical staff responsible for the plaintiff's care approximately 48 hours after the swab had been taken, included a note in the following terms:
The significance of Gram-negative bacilli isolated from non-sterile specimens is difficult to ascertain. They may be normally resident or transiently colonising this type of specimen. Please contact the Microbiology laboratory if susceptibility results are needed for this isolate.
At 9.05 pm on the night of the plaintiff's admission blood samples were taken from the plaintiff so that the plaintiff's 'full blood picture' (FBP) could be ascertained. The analysis of the blood samples revealed, among other things, that the plaintiff's white blood cell count was 10.7 (the normal range being 6.0 to 17.5) and that her neutrophil count was 6.0 (the normal range being 1.0 to 8.5). The haematology report on the results of the FBP analysis, which was (as was the case for all FBP analysis results) available to medical staff responsible for the plaintiff's care within hours of the blood sample being taken, included a note that 'neutrophils show left shift with toxic changes'.
The plaintiff's burns were dressed with Silver Sulfadiazine cream which provides broad spectrum cover for topical skin bacteria.
Throughout the night of 9 and 10 December 2005 the plaintiff's vital signs were monitored on an hourly basis. The plaintiff remained asleep and settled.
At 8.00 am on 10 December 2005 the plaintiff's temperature was 38.6 degrees Celsius. Her heart rate was 182. Her respiratory rate was 40. Her oxygen saturation level was 95%. She was placed on oxygen therapy by nasal prongs.
At 8.20 am the plaintiff's temperature was 39.4 degrees Celsius.
At 9.00 am the plaintiff's temperature was 39.3 degrees Celsius. Her heart rate was 178. Her respiratory rate was 36. Her oxygen saturation level was 100%.
At 10.00 am the plaintiff's temperature was 39.4 degrees Celsius.
After 10.00 am the plaintiff's temperature dropped. Between 10.00 am and 9.00 pm the plaintiff's temperature ranged between 36.7 degrees Celsius and 38.8 degrees Celsius.
As to the remainder of the plaintiff's vital signs, between 9.00 am and 9.00 pm her heart rate ranged between 157 and 197, her respiratory rate ranged between 22 and 38, and her oxygen saturation levels while on oxygen therapy ranged from 89% to 99%.
At around 1.00 pm blood samples were taken from the plaintiff for analysis. However, the samples could not be analysed because the incorrect lids had been placed on the test tubes containing the samples with the result that the samples were considered by the pathology laboratory to be contaminated.
At 6.25 pm a blood sample was taken from the plaintiff. Analysis of this sample revealed, among other things, that the plaintiff's C-reactive protein level (CRP) was 108 mg/L (normal range being 0 - 10 mg/L).
At 9.00 pm a chest X-ray was performed on the plaintiff. The X-ray revealed mild patchy change in the right lung base medially but that the lungs were otherwise 'well expanded and essentially clear'.
At around 10.00 pm blood samples were taken from the plaintiff for FBP analysis. The analysis of the blood samples revealed, among other things, that the plaintiff's white cell count had lowered to 4.9 and that her neutrophil count had lowered to 0.7. The haematology report on the results of the FBP analysis, which were available to the medical staff responsible for the care of the plaintiff by some time after 1.45 am on 11 December 2005, included a note of 'mild neutropenia'.
Throughout the early morning hours of 11 December 2005 the plaintiff's vital signs were closely monitored. She remained on oxygen therapy throughout the night.
At 8.00 am on 11 December 2005 the plaintiff's temperature was 38.4 degrees Celsius, her heart rate was 178, her respiratory rate was 34 and her oxygen saturation level with oxygen therapy was 98%.
At 9.00 am the plaintiff had a burns bath and her wounds were dressed. She was very unsettled during this process.
From 9.00 am the plaintiff's condition began to deteriorate markedly.
At 11.00 am the plaintiff's heart rate was 182 ‑ 192, her respiratory rate was 40 and her oxygen saturation levels while receiving oxygen therapy were 84% ‑ 95%. By 12.00 pm the plaintiff's oxygen saturation levels had dropped further to 72% ‑ 92%.
Between 1.00 pm and 9.00 pm the plaintiff's heart rate ranged between 167 and 192, and her respiratory rate ranged between 32 and 74, the reading of 74 being at 9.00 pm. During the same period the plaintiff's oxygen saturation levels while being on various forms of oxygen therapy ranged between 89% and 100%, with levels of 89% and 93% at 8.00 pm and 9.00 pm respectively. At 4.00 pm the plaintiff's temperature was 37.8 degrees Celsius and at 8.00 pm it was 37.1 degrees Celsius.
At 5.55 pm, and then again at 8.55 pm, the plaintiff was given Frusemide (a diuretic). The Frusemide was given to the plaintiff because the medical staff treating her believed that her respiratory difficulties may be the result of her having been given excess fluid (commonly referred to as fluid overload).
On review by a medical registrar at around 6.00 pm the plaintiff was noted to have crepitation (crackling sounds) at the base of both lungs.
At about 8.30 pm a chest X-ray of the plaintiff was performed. The X‑ray revealed widespread patchy areas of alveolar consolidation in both of the plaintiff's lungs, particularly in the left lung.
At 9.45 pm the plaintiff, who by this time was in severe respiratory distress and very ill, was admitted to PMH's paediatric intensive care unit (the ICU).
At some point during 11 December 2005 nursing staff reassessed the burns suffered by the plaintiff to be to 12% of her TBSA. This reassessment was incorrect. The burns were to approximately 18% of the plaintiff's TBSA as originally assessed by the burns registrar on 9 December 2005.
The course of events following the plaintiff's admission to the ICU
At 10.00 pm on 11 December 2005 the plaintiff's temperature was 37.8 degrees Celsius, her heart rate was 187, her respiratory rate was 42 and her oxygen saturation level while on oxygen therapy was 95%.
At 11.00 pm the plaintiff's heart rate was 187 and her respiratory rate was 48. Her oxygen saturation level was 85%.
At 11.15 pm blood samples were taken from the plaintiff for FBP analysis and also for culturing. The FBP analysis of the blood samples revealed, among other things, that the plaintiff's white cell count had dropped to 1.3, that her neutrophil count had dropped to 0.1 and that her CRP level had increased to 242 mg/L. The results of the blood culture, which did not become available for 48 hours, were negative for bacteria.
At 12.00 am the plaintiff was intubated and oxygen was administered to her by bag and mask. She was by this time (if not earlier) gravely ill.
Over the ensuing early morning hours of 12 December 2005 the medical staff within the ICU worked to maintain the plaintiff's oxygen levels and to keep her alive. At 2.00 am she was placed on a ventilator in an attempt to combat her inadequate oxygen saturation levels. However, her levels continued to drop and to remain far too low. Indeed, by 9.00 am her oxygen saturation level had dropped to 33% despite her being on high frequency ventilation.
The ICU medical staff came to the view that bacterial infection (sepsis) was a possible cause of the plaintiff's difficulties even though the results of the blood culture (of the blood sample taken at 11.15 pm on 11 December 2005) were not available and would not be available for some time. Accordingly, between 12.20 am and 8.00 am a number of powerful broad spectrum antibiotics were administered to the plaintiff intravenously, namely Ceftriaxone (at 1.00 am), Gentamicin (at 3.00am), Vancomycin (at 6.00 am) and Meropenem (at 8.00 am). The plaintiff remained on a combination of at least some of these antibiotics until 20 December 2005.
At 5.00 am the plaintiff was given a blood infusion of packed red cells.
At 11.00 am the plaintiff, whose oxygen saturation level at the time was 68%, went into cardiac arrest. Cardiac compressions were commenced. The plaintiff was resuscitated.
At 1.30 pm the plaintiff went into cardiac arrest for a second time. Again, cardiac compressions were commenced and the plaintiff was resuscitated.
Unfortunately, given the plaintiff's prolonged low oxygen levels she developed multi-organ failure and brain damage secondary to hypoxia (lack of oxygen to the bodily tissues).
At around 2.30 pm blood samples were taken from the plaintiff for FBP analysis. The analysis of the blood revealed that the plaintiff's white cell count had risen to 6.3, and that the plaintiff's neutrophil count had risen to 0.5.
At around 6.30 pm blood samples were again taken from the plaintiff for FBP analysis. The analysis revealed that the plaintiff's white cell count had further risen to 12.6 and that her neutrophil count had risen to 1.6. The haematology report on the results of the FBP analysis included a note that the 'neutrophils show a left shift with marked toxic changes'.
On 13 December 2005, so at least around 24 hours after the plaintiff had first been commenced on antibiotics, various samples were taken from the plaintiff so that they could be tested for the presence of bacteria. The samples were of the plaintiff's peritoneal fluid, post-nasal aspirate, tracheal aspirate, pleural fluid and endotracheal aspirate. The testing of all of these samples produced negative results for bacteria.
Some time prior to 9.00 am on 13 December 2005 another blood sample was taken from the plaintiff for FBP analysis. The analysis revealed that the plaintiff's white cell count was 8.5 and that her neutrophil count was 2.0.
At 10.45 am on 13 December 2005 swabs were taken from the burn wounds to the plaintiff's right arm, right leg and knee, chest, left leg and knee and left arm. The swabs were taken so that they could be analysed for the presence of bacteria. Culturing of all of the swabs produced negative results for bacteria.
Additional blood cultures undertaken on blood samples taken from the plaintiff after the commencement of the administration of antibiotics produced negative results for bacteria.
The plaintiff remained in the ICU until 29 December 2005. On that date she was extubated and transferred to PMH's rehabilitation ward.
The cases of the parties – summary
Having outlined the non-contentious factual background to the plaintiff's claim, I turn to summarise the respective cases of the plaintiff and the defendant.
There is no dispute between the parties that the defendant owed to the plaintiff a duty to exercise reasonable care and skill in the provision of medical advice and treatment: Wallace v Kam [2013] HCA 19; (2013) 250 CLR 375 [8]. Nor is there any dispute that the defendant was liable for the acts and omissions of the team of doctors and nurses who were involved in the care and treatment of the plaintiff following her admission to PMH.
As to the standard of care required, the plaintiff alleges that the standard required was that to be expected of a tertiary hospital (which PMH was) employing and/or engaging medical practitioners and nurses expert in the area of paediatrics.
The plaintiff alleges that by the morning of 10 December 2005 she had, as a result of her burns, evolving and detectable sepsis, either by itself or in conjunction with an inflammatory response to the burns. The plaintiff alleges that the failure by the team of doctors engaged by the defendant who were responsible for her care and treatment in PMH's burns ward to recognise on the morning of 10 December 2005 that she was suffering from or might be suffering from sepsis, to test for signs of sepsis, and to commence antibiotic treatment amounted to a breach by the defendant of the duty of care that it owed to her. She alleges that this breach of duty then continued as an ongoing breach until sepsis was identified by doctors working in the ICU and treated by the administering to her of antibiotics shortly after 12.00 am on 12 December 2005.
In relation to the issue of causation, the plaintiff alleges that the defendant's breach of duty in failing to detect and treat her sepsis by the administration of antibiotics on the morning of 10 December 2005 and at any time prior to the morning of 12 December 2005 caused her to suffer Acute Respiratory Distress Syndrome (ARDS), which in turn caused her to suffer injuries comprised of cardiac arrest, multi-organ failure, brain damage and cerebral palsy (the injuries).
The defendant does not by its pleaded defence admit that the standard of care that it was required to meet was the standard alleged by the plaintiff. The defendant's case is that the standard of care that it was required to meet was that of a competent practitioner employed or engaged to work in the burns ward of a tertiary hospital.
In relation to the issue of breach of duty, although the defendant accepts that the plaintiff suffered an inflammatory response to her burns (indeed, on its case, a severe systemic inflammatory response) as well as ARDS, it denies that the plaintiff ever had sepsis. The defendant, therefore, necessarily also denies the plaintiff's allegation that the failure by the doctors engaged by the defendant to work in the burns ward to recognise on the morning of 10 December 2005 that the plaintiff was suffering from or might be suffering from sepsis, to test for sepsis and to commence antibiotic treatment amounted to breach by the defendant of the duty of care that it owed to the plaintiff. Further, the defendant's case is that even if contrary to its primary position the plaintiff did in fact have evolving sepsis from the morning of 10 December 2005, the failure by the doctors who were working in PMH's burns ward and who were responsible for the care and treatment of the plaintiff to recognise, test for and treat the sepsis by the administration of antibiotics did not amount to a breach by the defendant of the duty of care that it owed to the plaintiff.
As part of its denial of the plaintiff's allegation of breach of duty, the defendant asserts that the doctors and nursing staff working in PMH's burns ward who were responsible for the care and treatment of the plaintiff at all material times acted in accordance with a practice that was widely accepted as competent professional practice: Civil Liability Act 2002 (WA) (CLA), s 5PB.
As to the issue of causation, the defendant's case is that it was not sepsis, either alone or in combination with an inflammatory response, that caused the plaintiff to suffer ARDS, but rather the plaintiff's severe systemic inflammatory response to her burns. That is, the defendant's case is that it was the plaintiff's severe systemic inflammatory response in the absence of any infection (sometimes referred to as a sterile systemic inflammatory response) that caused the plaintiff to suffer ARDS and the consequential complications of that condition. Further, and alternatively, the defendant's case is that even if the plaintiff did have sepsis by the morning of 10 December 2005 or at any time thereafter, and even if the failure by the doctors and nursing staff to detect and treat the sepsis by the administration of antibiotics did amount to a breach by the defendant of its duty of care, this breach did not cause the ARDS and the injuries because the administration of antibiotics to the plaintiff on the morning of 10 December 2005 or at any time thereafter would not have made any difference to the plaintiff's adverse outcome.
I note that as part of its pleaded case against the defendant the plaintiff alleged that the defendant breached the duty of care that it owed to the plaintiff by reason of the team of doctors who were responsible for the care and treatment of the plaintiff in the burns ward administering too much fluid to her (fluid overload). The plaintiff alleged that the fluid overload caused the plaintiff to suffer ARDS and the injuries. However, prior to the commencement of the trial the plaintiff abandoned this aspect of her claim.
The witnesses
The plaintiff adduced evidence from the following witnesses: Ms Mary Kelei, the plaintiff's mother; Dr Andrew Numa, intensive care and paediatric respiratory physician; Associate Professor Alison Kesson, paediatric infectious diseases physician; and Associate Professor Michael Ditchfield, consultant paediatric radiologist.
Ms Kelei's evidence was adduced by tendering in her absence by consent an affidavit sworn by her on 2 November 2017.
Professor Kesson, Dr Numa and Professor Ditchfield all gave evidence as expert witnesses. Professor Ditchfield's evidence was adduced by tendering in his absence two reports prepared by him. Professor Ditchfield's reports were tendered under s 79C(1) of the Evidence Act 1906 (WA) in the absence of any objection by the defendant and on the basis that the defendant did not require him to be called and made available for cross-examination: s 79C(2)(e).
The witnesses from whom the defendant adduced evidence were as follows: Dr Deirdre Speldewinde, consultant paediatrician; Dr Simon Erickson, consultant paediatric intensivist; Dr Melissa Gillett, consultant chemical pathologist; Dr Jeffrey Prebble, consultant paediatrician; Dr Hugh Allen, respiratory and general paediatrician; Associate Professor Mike Starr, consultant paediatrician, infectious diseases physician and consultant in emergency medicine; Dr Fiona Bettenay, consultant radiologist; Dr Conor Murray, paediatric cardiothoracic and general radiologist; Professor Roy Kimble, paediatric surgeon; and Associate Professor John Harvey, general, thoracic and burns paediatric surgeon. All of the defendant's witnesses, save for Dr Speldewinde, Dr Erickson and Dr Gillett, gave evidence as expert witnesses. Dr Speldewinde, Dr Erickson and Dr Gillett were called only as witnesses of fact.
Dr Murray's evidence was adduced by tendering in his absence two reports prepared by him. Dr Murray's reports were tendered under s 79C(1) of the Evidence Act in the absence of any objection by the plaintiff and on the basis that the plaintiff did not require him to be called and made available for cross‑examination: s 79C(2)(e).
Questions for determination
Given the respective cases of the parties, the questions that arise for my determination are as follows:
1.Did the plaintiff have sepsis on 10 December 2005 or at any time thereafter?
2.What was the standard of care that the defendant had to meet?
3.Did the defendant breach its duty of care in the way alleged by the plaintiff?
4.If the defendant did breach its duty of care, was the breach of duty the cause of the injuries?
In posing the questions for determination in the way that I have, I recognise that the question whether or not the plaintiff had sepsis on 10 December 2005 or at any time thereafter forms part of the broader causation question in the sense that if the plaintiff does not prove that she developed sepsis she cannot prove her allegation that any breach of duty by the defendant caused her to suffer the injuries. I also recognise that a finding that the plaintiff did not have sepsis will not of itself be determinative of the question whether the failure by the doctors working in the burns ward to recognise that the plaintiff might be suffering from sepsis and to treat the possible sepsis with antibiotics amounted to a breach of duty by the defendant. Nonetheless, I propose to deal with the question whether the plaintiff had sepsis first and separately from the issue of causation. I propose to do so not only because of the understandable focus of attention that was placed on this question by the parties, but also because much of the expert evidence adduced at trial which bears upon this question is also relevant to, or forms part of the essential background to, the allegation of breach of duty. Therefore, to attempt to analyse and determine the allegation of breach of duty without having first dealt with the question whether or not the plaintiff had sepsis, and as part of doing so referring to the evidence which is relevant to the determination of this question, will not only be inconvenient but also somewhat artificial.
Medical terms and medical conditions
Before turning to deal with each of the above posed questions it is convenient to provide some further explanation of a number of the medical terms and medical conditions to which I have already referred or to which reference will be made hereafter. I note in this context that during the trial some of the terms and conditions were defined or described by various witnesses in slightly differing terms. However, the below explanations of the terms and conditions accurately reflect the substance of the meaning given to them by, if not all, the vast majority of the witnesses.
Infection
An infection is a condition which is caused by the entry of pathogens like bacteria, virus and fungi into the body. Infection can be limited to a body region or be widespread.
Oedema
Oedema is fluid in the tissues.
Neutrophils
White blood cells are made up of several subsets. The main subsets are neutrophils, lymphocytes and monocytes. Neutrophils are very important in the defence of a person against infection, particularly bacterial infection.
Neutropenia
Neutropenia is an abnormally low level of neutrophils.
CRP
CRP is a marker of inflammation or sepsis.
Systemic Inflammatory Response Syndrome
Systemic Inflammatory Response Syndrome (SIRS), in the context of the present case, is an inflammatory response of the body to trauma or insult including infection and/or sterile burns. Thus, a person can have SIRS without infection.
The reference to 'systemic' in this context is a reference to the whole of the body.
In the case of children SIRS has two or more of the following features: a temperature greater than 38.5 degrees Celsius or less than 36 degrees Celsius; tachycardia (fast heart rate); tachypnoea (fast breathing rate); leucocytosis (high white blood cell count); leukopenia (low white cell count).
ARDS
ARDS is an acute, diffuse and inflammatory lung injury that leads to increased vascular permeability with fluid in the alveoli (pulmonary oedema), a loss of aerated lung tissue and consequential hypoxia. ARDS is also sometimes referred to as Diffuse Alveoli Damage.
As ARDS develops breathing becomes laboured as the lungs work harder to get oxygen into the blood and carbon dioxide out of the body. As the oedema in the lungs increases it impedes access of oxygen into the blood and to the heart. The level of oxygen reaching the heart becomes inadequate (which causes the heart to pump harder to try and compensate) and consequently inadequate levels of oxygen reach bodily tissues including the brain.
ARDS has various causes including sepsis. Sepsis is a common trigger for ARDS. However, ARDS can also be caused by SIRS in the absence of infection or sepsis.
Sepsis
Sepsis is a serious complication of an infection and is a systemic response to the presence of infection.
Sepsis occurs when chemicals released into the bloodstream to fight the infection trigger inflammatory responses throughout the body. This inflammation can trigger a cascade of changes that can damage multiple organ systems causing them to fail.
The symptoms of sepsis include inflammation, an abnormally high or low body temperature, tachycardia, tachypnoea, a high or low neutrophil count, and pulmonary oedema (excess fluid in the lungs). Sepsis can be a complication of burns.
Sepsis can, although need not, involve bacteria from the site of the infection making their way into the bloodstream.
Septicaemia
Septicaemia is a form of sepsis. Septicaemia is a condition in which there are bacteria in the bloodstream which are actually growing in the bloodstream. Septicaemia is colloquially referred to as blood poisoning.
During the trial at least one of the witnesses (Professor Harvey) used the terms sepsis and septicaemia interchangeably. The two conditions are, however, distinct.
Septic shock/toxic shock
Septic or toxic shock occurs when the blood vessels, due to sepsis, become dilated with the result that the volume of blood being pumped around the body is inadequate. The blood pressure becomes low which in turn results in inadequate oxygenated blood flow to the body's tissues including vital organs which can in turn damage the tissues and organs.
Fluid overload
In lay terms, the effect of burns on a body is that fluid leaks both into and out of the body. Consequently, fluid is required to be replaced so as to maintain circulation.
Fluid overload occurs if too much fluid is administered to the patient with the result that the patient experiences oedema.
One of the purposes of estimating the TBSA affected by burns is that it determines the amount of fluid to be initially administered to the patient according to a well-recognised formula having regard to the age and weight of the patient. The estimate of the TBSA affected by burns is only a rough estimate for the purposes of determining the appropriate initial administration of fluid. The patient, the amount of fluid administered, and urine output are monitored on an ongoing basis and the level of fluid administered is then adjusted to maintain urine output within recognised parameters.
Did the plaintiff have sepsis on 10 December 2005 or at any time thereafter?
The relevant evidence
In order to deal with the question whether the plaintiff had sepsis on 10 December 2005 or at any time thereafter it is necessary, as a first step, to refer to the evidence adduced by the parties which bears upon the question. The evidence in this regard was detailed and extensive, and came from a number of witnesses. The evidence also included some documentary evidence (other than medical reports) adduced by the plaintiff under s 79C of the Evidence Act.
My below bracketed references to 'publication cited' indicate that the witness referred in his or her report to a paper, article or other publication, a copy of which was tendered, in support of the statement made, or opinion expressed, by him or her.
The plaintiff's expert witnesses
Dr Andrew Numa
Dr Numa is an intensive care and paediatric respiratory physician.
Dr Numa obtained his medical degree in 1983. He obtained his fellowship of the Royal Australian College of Paediatricians in 1990. He obtained his specialist qualifications in intensive care medicine in 1994. He obtained his specialist qualifications in paediatric respiratory medicine in 2003.
Dr Numa has held a number of positions in the area of paediatric intensive care. In 1991 he was a Fellow, paediatric intensive care, at the Children's Hospital Los Angeles, California. In 1994 he was promoted to the position of Senior Fellow at this hospital. The Children's Hospital Los Angeles is a large tertiary hospital. The hospital's intensive care unit, where Dr Numa worked, has around 2,500 admissions per year which is approximately twice the size of the largest intensive care units in Australia. The hospital's intensive care unit looked after all sub-speciality areas of paediatric medicine and surgery. The patients that he dealt with while working at the hospital included burns patients.
From 1996 to 2000 Dr Numa was the staff specialist in the intensive care unit at Sydney Children's Hospital. During this period he was also a lecturer in paediatrics at the University of New South Wales.
In 1998, and up until 2003, Dr Numa was the director of physician training at the Sydney Children's Hospital.
In 2001 Dr Numa was appointed the senior staff specialist, intensive care unit, at the Sydney Children's Hospital, an honorary consultant in respiratory medicine at the hospital and a senior lecturer in paediatrics at the University of New South Wales. He still holds these appointments.
Since 2008 Dr Numa has been the director of the Sydney Children's Hospital intensive care unit.
Dr Numa is currently an examiner for the College of Intensive Care Medicine.
Dr Numa has authored or co-authored a number of publications dealing with ARDS. One of the publications is the chapter on ARDS in the leading text Kendig and Chernick's Disorders of the Respiratory Tract in Children, 7th ed, WB Saunders, Philadelphia, 2006.
As part of his clinical experience Dr Numa has dealt extensively with sepsis, SIRS and ARDS.
Dr Numa's evidence was comprised of the content of three reports prepared by him for the plaintiff's solicitors supplemented by oral evidence. Dr Numa's three reports are dated 5 November 2016, 26 December 2016 and 27 July 2017.
Report dated 5 November 2016
Dr Numa prepared his report dated 5 November 2016 without having been provided with any of the expert reports that to that point in time had been exchanged between the plaintiff's and the defendant's solicitors. In his report Dr Numa makes the following statements and expresses the following opinions.
The plaintiff's deterioration over a period of approximately 50 hours from 7.00 pm on 9 December 2005 to 9.30 pm on 11 December 2005 was almost certainly related to either sepsis or toxic shock syndrome. The former diagnosis, sepsis, is more likely by a significant margin as an organism, 'Enterobacter species +++ growth' was identified on the swab taken from the plaintiff's right knee burn wound soon after her admission to hospital. Nevertheless, toxic shock, usually related to staphylococcal infection, presents similarly and is a well-recognised complication of burns in paediatric patients (publication cited).
The evidence for the plaintiff suffering sepsis consists of several significant spikes of temperature over the first 48 hours, persistent tachycardia, elevated inflammatory markers, evolving neutropenia, evolving respiratory distress and an oxygen requirement, and subsequent progression to multi-organ failure including circulatory collapse and ARDS in the ICU.
The spikes in temperature were 39.4 degrees Celsius at 10.00 am on 10 December 2005, 38.4 degrees Celsius at 5.00 pm on 10 December 2005, 38.1 degrees Celsius at 7.00 pm on 10 December 2005 and 38.4 degrees Celsius at 8.00 am on 11 December 2005.
As to the persistent tachycardia, observations recorded the plaintiff's heart rate largely in the range of 160 to 180 including during sleep. The resting heart rate for a 16-month-old child would usually be in the range of 80 to 120, and during sleep is much more likely to be at the lower end of this range.
Tachycardia is a non-specific sign, but is most usually associated with circulatory insufficiency or sepsis in this age group. The plaintiff's tachycardia was noted but was attributed to fluid overload. While fluid overload can cause tachycardia, there is little evidence for this being present. The plaintiff's fluid requirements were calculated appropriately on the basis of an 18% burn, but subsequently some doubt was cast on the area of the burn with the suggestion that it was closer to 12%. The 6% difference does not have a clinically significant impact on fluid administration.
On 11 December 2005 the diagnosis of fluid overload was put forward and diuretics were given with an impressive urine output of 410 ml. Despite this response the tachycardia did not settle. This suggests the cause was not fluid overload but was in fact evolving sepsis.
With respect to the elevated inflammatory markers, the blood sample taken from the plaintiff at 6.30 pm on 10 December 2005 revealed that the plaintiff's CRP level, a non-specific marker of infection/inflammation, was significantly elevated at 108 mg/L. By 11.15 pm on 11 December 2005 the level had further increased to 242 mg/L. The normal range for CRP is less than 10 mg/L.
The interpretation of CRP is somewhat difficult in burns patients as it may be elevated purely as a result of the burn itself without any infection being present. Nonetheless, an elevated CRP level is consistent with, although not diagnostic of, sepsis.
As to the evolving neutropenia, the total white blood cell counts fell over the period of the plaintiff's admission. This was especially so for the neutrophils. Low neutrophil counts (neutropenia) are a hallmark of sepsis. The plaintiff's neutrophil count of 0.1, revealed by the analysis of the blood sample taken from her at 11.15 am on 11 December 2005, was an extraordinarily low value that was highly suggestive of sepsis. Neutropenia is highly correlated with sepsis (but not burns alone) in animal models as well as adult and paediatric case series (publications cited).
In relation to the plaintiff's evolving respiratory distress and oxygen requirements, several possibilities can explain respiratory distress. However, the respiratory distress is consistent with, although not specific for, a diagnosis of sepsis. The medical staff caring for the plaintiff appeared to attribute the evolving respiratory distress to fluid overload. However, this is unlikely.
In relation to the plaintiff's subsequent progression to multi-organ failure including circulatory collapse and ARDS in the ICU, the plaintiff's course in the ICU from 12 December 2005 onwards was 'absolutely typical' for sepsis or toxic shock syndrome. No other diagnosis is likely to have caused the constellation of signs and symptoms observed. An alternative but 'implausible' hypothesis for the plaintiff's course is the well-recognised association between ARDS and burns. ARDS can be caused by smoke inhalation, but is also described as a response to burns not associated with any inhalation injury. However, the risk for ARDS is closely related to the area and severity of burns and in the case of the plaintiff, who had superficial and partial but no full thickness burns to 12% to 18% of her TBSA, the probability of ARDS developing purely as a response to the tissue injury alone is 'vanishingly small'. In a recent series, burns patients who developed ARDS had substantially more full thickness burn area than patients who did not develop ARDS (median 20.5% full thickness burn area in ARDS patients vs median 7% full thickness burn area in non-ARDS patients). Similarly, the total burn area was significantly larger in patients developing ARDS (publication cited). In this case where the plaintiff had no full thickness burns and burns to no more than 18% of her TBSA, ARDS is unlikely to have developed as a result of the burn itself.
Report dated 26 December 2016
Dr Numa prepared his report dated 26 December 2016 after being provided with the reports of the various expert witnesses of the defendant which had at the time been prepared and provided to the plaintiff's solicitors. In his report Dr Numa makes the following statements and expresses the following opinions in response to the reports of the defendant's expert witnesses that had been provided to him.
He supports the view expressed by the other report writers that the fluid administration to the plaintiff was neither excessive nor the cause of the plaintiff's ARDS.
The claim that a burn of the nature suffered by the plaintiff can result in ARDS is not supported by the literature. He has not been able to find any case reports of ARDS secondary to burns of the nature and area suffered by the plaintiff, unless accompanied by sepsis. Although it is not possible to entirely rule out a burn of less than 20% of the TBSA being a trigger for ARDS, this would be very rare sequelae whereas sepsis is a common complication of burns and a very common trigger for ARDS. Therefore, the balance of probabilities overwhelmingly favours sepsis as the cause of the plaintiff's ARDS rather than the burns per se.
None of the other opinions seriously address the possible issue of sepsis. The commonest cause of morbidity and mortality in paediatric burns patients is sepsis (publication cited). None of the opinions touch on the quite striking neutropenia which provides strong evidence for sepsis. The most likely cause for such a dramatic fall in white blood cell count is overwhelming sepsis.
None of the reports address the heavy growth of Enterobacter from the right knee swab which he finds 'an extraordinary omission'.
Evidence for infection is substantial. In addition to the evidence for infection that he referred to in his first report, the plaintiff's severe coagulopathy on her admission to the ICU adds weight to the diagnosis of sepsis. The pathology reports from 12 December 2005 show a severely disordered coagulation profile 'with an INR [International Normalised Ratio] = 8.5 and APTT [Activated Partial Thromboplastin Time] of 63.9 together with a borderline low fibrinogen (1.5g/L), suggesting disseminated intravascular coagulation'. This is a common finding in severe sepsis.
While none of these signs, apart from the positive wound culture, is specific for sepsis, the constellation of these signs in the plaintiff 'are essentially diagnostic of sepsis without question'.
The key issue in his mind is not whether the plaintiff's deterioration was related to sepsis, which is 'beyond any doubt', but whether the early symptoms and signs of sepsis should have been identified and acted on.
None of the reports provided to him alters his strong view that the plaintiff's deterioration was related to sepsis.
Report dated 27 July 2017
In his report dated 27 July 2017 Dr Numa makes the following statements and expresses the following opinions in response to some further expert reports provided to him, specifically the report of Professor Starr dated 25 June 2017 and the report of Professor Harvey dated 27 April 2017.
With respect to Professor Starr's assertion in his report that it appeared that the plaintiff's primary problem may have been neither sepsis nor fluid overload but rather ARDS, by definition ARDS is never a primary problem. Rather, it arises secondary to some triggering disease or insult. Recognised ARDS triggers are categorised into direct lung injury (for example, pneumonia, contusion, smoke inhalation, aspiration) and indirect (principally sepsis, severe trauma or transfusion reaction) (publications cited). In the paediatric population infection is by far the most common cause of ARDS. In a 2007 Australasian study, which he co-authored, local or systemic infection was the trigger for ARDS in 46.2% of cases compared to 5.1% of cases that were triggered by burns (publication cited).
He agrees that there can be difficulty differentiating between SIRS and sepsis (which is generally defined as SIRS plus evidence of bacterial infection). The syndrome of sepsis includes all the key markers of SIRS. That is, all 'septic' patients have SIRS but not all SIRS patients have sepsis. The differentiation is further hampered by the lag time between testing for bacterial infection and obtaining a result (often 24 - 48 hours), and the fact that even the best test for sepsis, blood cultures, is less than 100% sensitive.
The issue is one of clinical probabilities, that is, was there sufficient clinical evidence for an infection to be reasonably suspected? He believes that the plaintiff's constellation of clinical symptoms went beyond the level of those associated with burns alone and were more suggestive of burns with superimposed infection. Neutropenia, which was first identified at 10.55 pm on 10 December 2005, was a particularly worrying development.
In his experience the combination of catastrophic circulatory collapse, coagulopathy, ARDS and multi-organ failure experienced by the plaintiff could not have arisen purely from SIRS alone (and especially not from the degree of SIRS that is associated with burns of less than 20% of the TBSA).
With respect to Professor Harvey's statement in his report that the plaintiff did not deteriorate over the first 48 hours of admission, it was during the first 48 hours after her admission that the plaintiff experienced intermittent fevers, evolving tachycardia and tachypnoea, and a requirement for oxygen. This can only be described as deterioration. After the first 48 hours of relatively slow decline the plaintiff declined more precipitously. This pattern is typical of evolving sepsis.
He does not agree with Professor Harvey's assertion in his report that there is no evidence to support a diagnosis of toxic shock. The clinical signs and symptoms of toxic shock syndrome are essentially indistinguishable from sepsis. Toxic shock syndrome is usually caused by a staphylococcal infection. It is correct that staphylococci were not identified in the plaintiff. Although he thinks that toxic shock syndrome is less likely than sepsis as a diagnosis, it cannot be excluded. It is a well described complication of relatively small burns in paediatric patients (publication cited).
As to Professor Harvey's statement in his report that elevation of CRP and neutropenia may represent sepsis but that there is no evidence that either of these parameters are specifically related to sepsis, all laboratory markers of sepsis are to some degree non-specific. A perennial problem faced by physicians is to differentiate inflammation from infection. Rather than any single diagnostic test, physicians assess a constellation of symptoms, signs and laboratory test results. The combination of symptoms, signs and laboratory test results in the plaintiff were sufficiently suggestive of infection by the morning of 11 December 2005 to warrant institution of antibiotic therapy at that time.
Professor Harvey's assertion in his report that there does not appear to be 'good evidence to invoke sepsis as a cause of [the plaintiff's] deterioration in the face of repeated clinical evidence of a respiratory basis for her deterioration', fails to recognise that the respiratory deterioration is highly suggestive of sepsis. Patients do not develop respiratory failure and ARDS without an underlying trigger and sepsis is by far the most common cause of this disorder. Sepsis is also by far the most likely cause of the plaintiff's circulatory collapse.
The statement made by Professor Harvey in his report that ARDS may on occasion occur in more minor burns (20%), although the exact incidence is unknown, reflects the fact that the incidence is unknown because there are no reports of any such cases. He has been unable to identify in the literature any report of a paediatric patient with burns of less than 20% developing ARDS unrelated to sepsis. On his review of the last five published reports of the Burns Registry of Australia and New Zealand (covering 2009/10 through to 2013/14) there is not a single mention of ARDS.
The Burns Registry reports indicate that approximately 800 paediatric patients per year are admitted to hospitals in Australasia with burns. In a recent audit of all ARDS patients managed in paediatric intensive care units throughout Australasia conducted over a one-year period as part of the 2007 study which he co-authored, only six patients were identified with acute lung injury relating to burns. Acute lung injury is a disease definition that includes ARDS but the definition for acute lung injury is more broadly based. A rough estimate of the incidence of ARDS related to burns is therefore less than 1% of cases. Further, in his view such cases are largely associated with more severe burns and/or inhalation injury (that is, not scalds). In contrast, sepsis is a relatively common complication of burns. In a recent study of paediatric patients admitted to a single burns centre between 2008 and 2011, 30.6% had fever on at least one occasion and almost half (46.2%) of patients with fever were identified as having a serious bacterial infection (publication cited). Therefore, a burns patient with fever has a reasonably high probability of having a serious bacterial infection.
Oral evidence
In his evidence-in-chief Dr Numa said the following by way of elaboration of aspects of his reports.
In order to fulfil the diagnostic criteria for SIRS the patient needs to have at least two of four key elements, namely elevated temperature, elevated or reduced white blood cell count, tachycardia or tachypnoea. The most common cause of SIRS is sepsis. However, SIRS is also seen in other conditions including burns.
ARDS is, in essence, acute severe lung disease. ARDS can be looked at in two broad categorisations, pulmonary ARDS and non‑pulmonary ARDS. Pulmonary ARDS arises from a direct insult or injury to the lungs, for example, pneumonia or smoke inhalation. Non‑pulmonary ARDS arises from an insult elsewhere in the body. Examples of significant insults elsewhere in the body which can cause ARDS include burns and sepsis from whatever cause. A significant insult outside of the lungs can impact heavily on the lungs.
Toxic shock is a syndrome that presents very similarly to sepsis. However, it is specifically related to certain species of staphylococcal bacteria that produce a toxin which causes all the signs and symptoms of severe sepsis. He mentioned toxic shock in the first of his reports because it is certainly described in burns patients. It is a differential diagnosis. However, in the case of the plaintiff, with the Enterobacter recovered, toxic shock is not really 'on the table'. In the setting of the Enterobacter, toxic shock is ruled out and the diagnosis is sepsis.
Many, if not most, patients with burns will get some degree of fever. Nevertheless, a temperature of 39.4 degrees Celsius is a significant temperature. It is not a mildly increased temperature of, say, 37.5 degrees Celsius or 38 degrees Celsius that a patient might get with a moderate inflammatory insult such as a scald burn.
The plaintiff's elevated heart rate of 160 to 180 was very significant. Tachycardia should never be ignored. Tachycardia does not tell you what is wrong with the patient, but it tells you that something is wrong with the patient and a persistent tachycardia of 160 or 180 is a concern. Tachycardia may be the result of the contractibility of the heart being compromised so that the heart is unable to eject a sufficient amount of blood with each contraction with the result that it compensates for that by contracting more frequently. Alternatively, tachycardia may be the result of an increased demand for oxygen in the body, such as in the case of sepsis, so that even though the contractibility of the heart is satisfactory, it is having to pump a great deal more blood around the body to meet the demands of the tissues. Tachycardia is non-specific but it certainly represents a disordered physiology somewhere in the body.
There are a variety of mechanisms by which sepsis can cause a faster heart rate. One of the key pathologies in sepsis is a disruption of the integrity of the capillary endothelium such that the capillaries become leaky. Thus, fluid that is normally contained within the vascular space, within the circulation, tends to leak through the capillaries into the interstitial tissues. This results in a depletion of the intravascular volume in the same way that dehydration depletes the intravascular volume. In these circumstances, the heart has less volume to work with. Another mechanism by which sepsis causes a faster heart rate is that in sepsis the metabolic rate is increased and so the tissues require more oxygen. A third mechanism is that part of sepsis involves the release of inflammatory chemicals broadly called cytokines, some of which depress cardiac function, that is, impair the contractibility of the heart.
C-reactive protein is a good marker of serious infections. A normal range of CRP is less than 10. If a CRP of more than 30 is seen, this is certainly a concern. A CRP of 100 is significantly elevated.
CRP can be a non-specific marker of infection or inflammation or both. However, some sort of quantitative analysis is involved. A CRP of 100 starts to look more like infection than inflammation. A CRP of 200 or more is very high indeed and to his mind is far more suggestive of severe infection than anything else.
Burns patients will have an elevated CRP. The literature reveals that typical burns patients will have a CRP of up to around 100 to 120, this being a typical or a mean value.
He thinks that the finding of neutropenia in the plaintiff is very significant. Neutrophil counts usually either rise or fall in response to infection. They more commonly rise than fall. When a low neutrophil count is seen this tends to suggest that things are very severe indeed. Neutrophils kill bacteria but in the process of killing the bacteria they also die themselves. Neutrophils circulate in the bloodstream and are manufactured in the bone marrow. In the face of severe infection neutrophil counts will fall because they are attacking the bacteria and are then lost and cannot be replenished rapidly enough by the bone marrow. There is a limit on how many neutrophils the bone marrow can produce.
The first thing that is seen when the bone marrow produces more neutrophils is that more immature neutrophils are 'churned out'. This is the so‑called 'left shift'. So, a 'left shift' indicates that marrow production of neutrophils is 'revving up' and producing lots of immature neutrophils which are seen in circulation. But if the infection is severe, despite maximal production of new neutrophils by the bone marrow, the circulating population of neutrophils will be depleted and the neutrophil count will fall. This is a very worrying sign.
Most of the features of SIRS and sepsis are common to both conditions, namely fever, tachycardia, tachypnoea, elevated white blood cell count and an elevated neutrophil count. However, there is some good evidence from both animal and human studies to suggest that a low neutrophil count is more suggestive of sepsis than of SIRS.
In the plaintiff's case there was an alarming decrease in the plaintiff's neutrophil count which was highly suggestive of severe illness and, specifically, of severe sepsis. It was a 'real red flag' for action to be taken.
The plaintiff's neutrophil count of 0.1 as at 11.15 pm on 10 December 2005 was appallingly low. It is rarely seen. The plaintiff's ability to fight infection was severely compromised.
He has never seen neutropenia of that severity purely due to inflammation. The only time that he has seen neutropenia of this severity has been when it is related to sepsis or some other pre-existing condition such as a child who is having chemotherapy.
It is the severity and the breadth of the plaintiff's symptoms that cause him to conclude that the plaintiff's course in the ICU was absolutely typical for sepsis. The plaintiff had a catastrophic decline over a short period of time. To go from being admitted to hospital breathing room air with normal blood pressure to 48 hours later being in the ICU getting emergency intubation and being essentially unventilatable with the most severe lung disease imaginable does not add up to anything but sepsis. It is the severity of the plaintiff's symptoms that makes him think that she had sepsis rather than sterile inflammation.
The opinion expressed in his first report that an alternative but implausible hypothesis for the plaintiff's cause of deterioration might be the well‑recognised association between ARDS and burns is based on his reading of the literature. Although ARDS is certainly reported to occur in association with burns, his reading of the literature suggests that it is generally seen with burns of far greater severity than those suffered by the plaintiff. The median area of burn in most of the case reports of ARDS is around 35% to 40% and almost always involves full thickness burns. So, although ARDS can arise from burns, in response to the question whether ARDS could result from the level of burns suffered by the plaintiff, his answer is that although one never says never in medicine, it is tremendously unlikely. In addition, the most common cause of ARDS in intensive care units is sepsis by a significant margin.
The importance of the result of the gram negative bacteria on the right knee swab is that it allows it to be said that the plaintiff had sepsis because sepsis is SIRS plus a suspected or proven infection. So the swab demonstrated some proven infection. There were three pluses of gram negative bacteria on a knee swab.
The admission swabs are taken to make sure that there are no infective agents in the wound. Two tests are done. The first is a culture the results for which take at least 48 hours to produce. However, a more instantaneous test, a gram stain, is performed where the material from the wound is just plated onto a slide, some staining chemicals are applied, and it is looked at by a microbiologist through a microscope. This is done pretty much as soon as the specimen is received. The importance of this is twofold. Although it may not reveal exactly what organism there is in the wound, it does reveal whether there are any organisms at all and whether they are gram positive or gram negative. The fact that some organisms are seen on a gram stain is important because it means that there are a significant number of organisms present. If you only have a few organisms they may be missed on the gram stain but found two days later when the results of the culture come back. However, if organisms are seen on a gram stain it goes to the issue of quantity. In this case the quantity was three plus. Usually a grading system of one to four plus is used. So three plus is a decent number of pluses. Three plus bacteria on a gram stain is a highly significant result. It is a result that would have been available to the treating doctors if they had looked.
I note that the defendant's counsel in his closing submissions did not attempt to suggest that the principles to which I have referred in the preceding paragraphs do not accurately represent the 'established principles' for the purposes of s 5C(2). Nor did counsel attempt to contend that this is a case in which consideration of the matters specified in sub-paragraphs (a) and (b) of s 5C(2) should preclude a finding of factual causation in accordance with these established principles.
Section 5D embodies the principle that the plaintiff bears the onus of proving any fact relevant to the issue of causation. It is for the plaintiff to persuade the court on the evidence as a whole and on the balance of probabilities of the nature and extent of the injury and loss caused by the defendant's negligence: Watts v Rake [1960] HCA 58; (1960) 108 CLR 158; Purkess v Crittenden; Seltsam Pty Ltd v Ghaleb [2005] NSWCA 208 [94] - [100]; Van Der Velde v Halloran [2011] WASCA 252 [138] ‑ [139]; City of Stirling v Tremeer [70] - [80].
The relevant evidence
It is once again necessary, before turning to deal directly with the question posed, to refer to the evidence adduced by the parties, in addition to the evidence already referred to, that is relevant to the issue of causation.
The plaintiff's expert witnesses
Dr Numa
Report dated 5 November 2016
In his report dated 5 November 2016 Dr Numa makes the following statements and expresses the following opinions.
With the presumed Enterobacter sepsis it is likely that the plaintiff would still have become very unwell, and it is very likely that she may still have needed support in the ICU, even if action had been taken in a timely manner. Nonetheless, on the balance of probabilities earlier treatment, particularly if antibiotics had been commenced on the evening of 10 December 2005, would most likely have resulted in a less tumultuous course in the ICU with the possibility of avoiding periods of severe prolonged hypoxia and cardiac arrest. If this had occurred, it is likely that the plaintiff's subsequent course would have been improved to some degree.
Oral evidence
In his oral evidence Dr Numa said the following.
In his view the effect of not giving antibiotics to the plaintiff some 18 hours earlier than they were given was profound. On the morning of 11 December 2005 the plaintiff was sick but she was not shocked. On the evening of 11 December 2005 she was shocked. The issue is not whether the antibiotics should have been given within one hour or two hours of her admission to the ICU. A minor delay at that point is not the real issue. The real issue is the fact that the 'ship had sailed' 12 hours earlier.
If the antibiotics had been given at an earlier stage he would still have expected the plaintiff to have been unwell for many days and to have needed some time in the ICU. However, he does not think that she would have been anywhere near as unwell as she ultimately was.
If the possibility of sepsis had been considered on the morning of 11 December 2005, antibiotic therapy could have been commenced some 17 to 18 hours earlier than the antibiotic therapy did commence. This would have given the plaintiff a much better chance of a good outcome. Sepsis evolves rapidly and it is a widely accepted principle that early initiation of treatment provides the best chance of survival. A delay of 18 hours is totally unacceptable. In his view this delay is likely to have significantly worsened the plaintiff's clinical course and denied her the best chance of a good recovery.
He believes that on the balance of probabilities earlier treatment with antibiotics commenced on the morning of 11 December 2005 is likely to have led to a better outcome in the plaintiff's case.
Professor Kesson
Report dated 25 January 2017
In her report dated 25 January 2017 Professor Kesson states that on the balance of probabilities the administration of antibiotics on the morning of 10 December 2005 would have prevented the plaintiff's presumed infection escalating to septic shock (with the consequential brain injury).
Report dated 21 July 2017
In her report dated 21 July 2017 Professor Kesson makes the following statements and expresses the following opinions.
On the balance of probabilities the earlier administration of antibiotics to the plaintiff would have decreased the risk of morbidity including the development of ARDS.
The pathophysiology of sepsis is complex and not always fully understood. Essentially, there is a decompensation of several interdependent body systems. This progresses to vasodilation and septic shock. These changes will result in ischaemia (that is, low blood flow with resulting low tissue oxygen concentration) and the process becomes a progressive and lethal cycle unless interrupted by specific interventions. Therefore, early treatment with antibiotics is essential in halting the initial stimulus to the pathophysiological cascade leading to septic shock and ultimately death. Delays in antimicrobial treatment 'will result in a worse outcome because of increasing ischaemic tissue and organ injury with time or ultimately in patient death because the pathophysiological processes have progressed to a point where they cannot be reversed'.
Oral evidence
In her oral evidence Professor Kesson said the following.
Infection, shock and death can develop very quickly.
It is possible that if antibiotics had been given to the plaintiff earlier this would not have made any difference to her outcome. However, she thinks that this is improbable. The degree to which the giving of antibiotics at an earlier point may or may not have prevented the outcome is something that can never be known. It is not possible to do an experiment. You would need two of the plaintiffs with one getting antibiotics and one not getting antibiotics to see what would actually happen. So, the answer to the question will never be known. However, she thinks that if the plaintiff had received antibiotics at an earlier point in time the plaintiff would have had a significantly better outcome. Certainly she considers this to be well beyond the balance of probabilities.
The longer the delay in administering antibiotics, the greater the risk of serious damage or death.
If the plaintiff had not ultimately been given antibiotics the process that was causing the SIRS and ARDS would not have stopped.
The defendant's expert witnesses
Dr Allen
Report dated 26 April 2017
In his report dated 26 April 2017 Dr Allen states that the earlier administration of antibiotics may have resulted in an improved outcome, but that equally there may have been no change in the outcome.
Oral evidence
In his oral evidence Dr Allen said the following.
If one assumes that sepsis was present, the earlier administration of antibiotics might have made a difference to the plaintiff. It is known that in children who do have documented sepsis the administration of antibiotics does not stop the profound inflammatory process occurring. The cascading effect is a response that could not be turned off in 2005 and usually cannot be turned off in 2017. If the patient has sepsis and the antibiotics are given before the cascading effect, this will produce a better outcome for the patient.
Associate Professor Starr
Report dated 25 June 2017
In his report dated 25 June 2017 Professor Starr states that he does not agree with Professor Kesson's opinion that earlier administration of antibiotics to the plaintiff would have been likely to alter her outcome. He further states that on the balance of probabilities the plaintiff's complications would have occurred whether or not antibiotics had been started earlier.
Oral evidence
In his oral evidence Professor Starr said the following.
He accepts that as a general rule if somebody has sepsis, the longer the delay in giving antibiotics the greater the risk of morbidity and mortality. He accepts that as a general rule the longer you leave the administration of antibiotics to a patient who has sepsis, the greater the risks of some permanent damage to the patient.
If it is assumed that the plaintiff did have sepsis, he does not know if the outcome would have been the same if she had been given antibiotics earlier. Nobody can know this because there is no evidence either way.
When he states in his report dated 25 June 2017 that the plaintiff's outcome would not have been altered by starting antibiotics earlier, he is intending to convey that because the plaintiff's condition did not, in his view, turn out to be sepsis, it would not have made any difference if the antibiotics were administered earlier. He accepts that if the plaintiff did have sepsis, and if the antibiotics had been started earlier, namely on the evening of 10 December 2005, the probability is that this would have affected the plaintiff's outcome. If the plaintiff did in fact have sepsis, then the administration of antibiotics some 24 hours earlier is likely to have resulted in a better outcome for the plaintiff.
Analysis and decision
At trial the defendant did not seek to contend that if the plaintiff did have sepsis, the sepsis did not cause, or at least materially contribute to, the development of the plaintiff's ARDS and consequentially the injuries. Rather, and as I have already indicated in summarising the respective cases of the plaintiff and the defendant, the defendant's position was that even if the plaintiff did have sepsis, and even if it did breach its duty as alleged, the breach did not cause the injuries because the administration of antibiotics to the plaintiff on the evening of 10 December 2005 or at any time thereafter would not have made any difference to the plaintiff's adverse outcome.
The defendant's decision not to contend that if the plaintiff did have sepsis, the sepsis did not cause, or at least materially contribute to, the development of her ARDS is, in light of all of the evidence to which I have referred, unsurprising. In any event, I state to avoid uncertainty on the issue that I am satisfied on the evidence to which I have referred that the plaintiff's sepsis was, if not the sole cause, then a materially contributing cause of her ARDS and consequently the injuries.
It is clear from the evidence to which I have just referred that none of the expert witnesses were able to say with certainty that if the plaintiff had been commenced on antibiotics earlier than she was she would not, as a result of her sepsis, have proceeded to develop ARDS and the injuries. However, it is, in my view, equally clear from the evidence to which I have just referred that given that the plaintiff's condition began to markedly deteriorate on the morning of 11 December 2005, it is more probable than not that if the plaintiff had been commenced on antibiotics by around 3.00 am on 11 December 2005 she would not have proceeded, as a result of her sepsis, to develop ARDS to the extent that she did and consequently would also not have suffered the injuries. Accordingly, I am satisfied on the balance of probabilities that but for the negligent conduct of the team of doctors who were responsible for the care and treatment of the plaintiff in the burns ward, the plaintiff would not have developed ARDS to the extent that she did and consequently would not have suffered the injuries: CLA, s 5C(1). More specifically, I am satisfied that if the doctors responsible for the care and treatment of the plaintiff in the burns ward had commenced to administer antibiotics to the plaintiff by around 3.00 am on 11 December 2005, the cascading effects of her sepsis would have been avoided or reduced with the result that she would not have developed ARDS to the extent that she did and consequently would not have suffered the injuries.
Alternatively, if contrary to my above expressed finding the plaintiff has failed to prove that but for the negligent conduct of the team of doctors who were responsible for her care and treatment in the burns ward she would not have developed ARDS to the extent that she did and would not have suffered the injuries, I am satisfied on the balance of probabilities on the basis of the evidence to which I have referred, of the following:
1.The failure by the doctors to administer antibiotics to the plaintiff by around 3.00 am on 11 December 2005 materially increased the risk of the plaintiff suffering ARDS to the extent that she did and consequently the injuries;
2.The ARDS and the injuries were within the area of foreseeable risk;
3.The defendant has not discharged the evidential burden of showing that the plaintiff would have suffered ARDS and consequently the injuries even if antibiotics had been administered to the plaintiff by around 3.00 am on 11 December 2005;
4.There is no reason why responsibility for the injuries should not be imposed on the defendant; and
5.Accordingly, the negligent conduct of the team of doctors caused the ARDS and the injuries: CLA, s 5C(2).
In summary, I am satisfied that the defendant's breach of the duty of care that it owed to the plaintiff caused the injuries.
Conclusion
For the reasons I have stated, I would uphold the plaintiff's claim and enter judgment accordingly.
4
2