WA Country Health Service v Wright [No 2] - [2010] WASCA 120

JURISDICTION : SUPREME COURT OF WESTERN AUSTRALIA

TITLE OF COURT : THE COURT OF APPEAL (WA)

CITATION: WA COUNTRY HEALTH SERVICE -v- WRIGHT [No 2] [2010] WASCA 120

CORAM: OWEN JA
NEWNES JA
JENKINS J

HEARD: 18 FEBRUARY 2010

DELIVERED : 30 JUNE 2010

FILE NO/S: CACV 43 of 2009

BETWEEN: WA COUNTRY HEALTH SERVICE
Appellant
AND
WILLIAM WRIGHT
Respondent

ON APPEAL FROM:
Jurisdiction              :  DISTRICT COURT OF WESTERN AUSTRALIA
Coram  :FENBURY DCJ
Citation  :WRIGHT -v- WA COUNTRY HEALTH SERVICE [2009] WADC 46
File No  :CIV 345 of 2006

Catchwords:

Negligence - Medical negligence - Appellant attended emergency department with stomach pain - Discharged after analgesics administered - Appellant alleged respondent negligent in failing to diagnose pneumonia - Appellant alleged pneumonia caused subsequent sepsis - Claim that appellant had pneumonia not made out - Respondent found negligent for failing to keep appellant under observation in emergency department until his illness diagnosed - Not the case appellant relied on at trial - Finding not open to trial judge - Turns on own facts

Legislation:

Nil

Result:

Appeal allowed

Category: B

Representation:

Counsel:

Appellant: Mr D R Clyne
Respondent: Mr J R Johnson

Solicitors:

Appellant: Minter Ellison
Respondent: Julian Johnson

Case(s) referred to in judgment(s):

Banque Commerciale SA (In Liq) v Akhil Holdings Ltd [1990] HCA 11; (1990) 169 CLR 279
Dare v Pulham [1982] HCA 70; (1982) 148 CLR 658
Leotta v Public Transport Commission (NSW) (1976) 9 ALR 437
Maloney v Commissioner for Railways (NSW) (1978) 18 ALR 147
Water Board v Moustakas [1988] HCA 12; (1988) 180 CLR 491
Wright v WA Country Health Service [2009] WADC 46

OWEN JA: I agree with Newnes JA.
NEWNES JA: On 3 July 2004 the respondent presented at Broome Hospital complaining of stomach pain. After a period of observation and treatment with analgesics, he was discharged later in the morning. In the early hours of 5 July, the respondent presented at Broome Hospital critically ill and was transferred by the Royal Flying Doctor Service to Royal Perth Hospital (RPH), where he was found to be suffering from septic shock complicated by cardiovascular and respiratory failure. The respondent spent 11 days in RPH before discharging himself. He subsequently brought a claim for damages for personal injury against the appellant in the District Court, alleging that his critical illness was caused by the negligence of staff at Broome Hospital. The Broome Hospital is owned and operated by the appellant.
In the District Court, Fenbury DCJ found the appellant liable in negligence to the respondent: Wright v WA Country Health Service [2009] WADC 46. The primary judge found that the appellant was negligent in discharging the respondent on 3 July in circumstances where he should have been kept in for observation. His Honour found that had the respondent been kept in for observation, whatever it was that caused his critical illness on 5 July would have been diagnosed and treated with the result that the critical illness would not have developed.
The appellant appeals against that decision, contending in substance, first, that the primary judge decided the case on a basis that was not pleaded or raised at the trial and, secondly, that in any event the evidence did not support a finding that if the respondent had been kept in for observation the critical illness he developed would have been avoided.
The respondent has filed a notice of contention, seeking to support the judgment on the additional ground that the primary judge should have found that (as pleaded by the respondent) at all material times the respondent was suffering from pneumonia and that had he been kept in for observation on 3 July the pneumonia would have been diagnosed and treated, thus avoiding the development of the critical illness.

The pleaded case

In his statement of claim, the respondent pleaded that he attended the emergency department of Broome Hospital on Saturday, 3 July 2004, following the sudden onset some two hours previously of severe left‑sided abdominal pain and shortness of breath. The respondent pleaded that a diagnosis of non‑specific abdominal pain was made by the doctor who

examined him at the hospital. He was given medication for pain relief and kept under observation for approximately 1.5 hours. The respondent said that during that period he coughed up red‑brown coloured blood and reported that to a nurse.

The respondent pleaded that after 1.5 hours, although he was observed to be in significant pain and unsteady on his feet, he was provided with analgesic medication and discharged. He was advised to attend the Broome Region Aboriginal Medical Service on the following Monday, 5 July.
The respondent alleged that over the next 24 to 48 hours his medical condition deteriorated and he was taken back to the hospital at about 2.50 am on Monday, 5 July 2004. There he was found to be critically ill and, after some interim treatment, was flown to Perth by the Royal Flying Doctor Service and admitted to the intensive care unit (ICU) of RPH where it is alleged he was found to be suffering from pneumonia with septic shock complicated by cardiovascular and respiratory failure.
The respondent pleaded that at all material times he had community‑acquired pneumococcal pneumonia (pneumonia) which was responsible for symptoms of abdominal pain, shortness of breath and coughing up of blood in his presentation at the Broome Hospital on 3 July, and for the symptoms from which he was suffering on 5 July 2004 when he again presented at the Broome Hospital.
The respondent pleaded, in effect, that the appellant, by its servants or agents, was negligent in:
(a)discharging the respondent on 3 July 2004 without any reasonable diagnosis in circumstances where, on the basis of his symptoms, he should have been admitted for observation;
(b)failing to carry out a chest x‑ray after the respondent had coughed up blood, and/or in light of his elevated blood sugar level.
The respondent alleged that had he been admitted for observation, or had a chest x‑ray been carried out, pneumonia would have been 'promptly diagnosed' and appropriate intravenous antibiotics administered, avoiding the subsequent development of the critical illness.
In the particulars of his injuries arising from the appellant's negligence, the respondent pleaded, among other things, 'progression of his [pneumonia] with septic shock, complicated by cardiovascular and respiratory failure'.
The appellant denied that it was negligent. It admitted that the respondent attended the hospital on 3 July 2004 complaining of left‑sided abdominal pain of two hours duration but did not admit that he continued to suffer, or complained of suffering, on‑going pain. The appellant admitted that a diagnosis of non‑specific abdominal pain was made by the doctor who examined the respondent. The appellant pleaded that the respondent's presentation on 3 July was not indicative of pneumonia, or evolving pneumonia, and did not warrant admission to hospital. It denied that the respondent had pneumonia on 3 July or on 5 July.

The uncontested facts

A number of matters of fact were not in issue at trial and as they provide the background to the critical issues it is convenient to set them out. They are as follows.
The respondent attended the emergency department of Broome Hospital at approximately 10.38 am on 3 July 2004, complaining of left‑sided abdominal pain from which he had been suffering for two hours. The onset of the pain was sudden. The respondent had not vomited and did not have a fever. The respondent was seen by a triage nurse at 10.40 am and by a doctor at 10.55 am, and was observed by nursing staff to have been dry‑retching.
The respondent's blood sugar level was assessed by a preliminary test to be 18.8 mmol/L. It was common ground that that was a high level. When examined by an emergency department doctor, Dr Russell, the respondent indicated he had previously been well. Dr Russell found nothing remarkable on the respondent's skin, and his heart and lungs appeared normal. The respondent's blood pressure was elevated. The appellant complained of tenderness in the area of and above his left hip. He was able to cough and sit up, and had a normal gait.
Dr Russell entered in the respondent's medical notes that the respondent was suffering from:
1.non‑specific abdominal pain;
2.non‑insulin dependent diabetes mellitus (NIDDM);
3.hypertension.
Dr Russell also made a notation to the effect that he believed the level of pain complained of by the respondent was much greater than indicated by the clinical signs on examination.
The respondent was treated with a pain relief and anti‑vomiting injection at 11.10 am. At 11.15 am, 100 mg of Tramadol was administered by injection (rather than orally due to the respondent's nausea) and an anti‑nausea injection was given at 11.50 am. The respondent was discharged at 12.20 pm on 3 July.
On the basis of the medical evidence, the analgesic effect of the Tramadol is likely to have peaked about two hour after it was administered and would have had effect for up to six hours. It was likely to have been still climbing to its peak at the time of discharge.
At 2.53 am on 5 July 2004, the respondent re‑presented at the emergency department suffering from severe respiratory distress. He was subsequently transferred to RPH by the Royal Flying Doctor Service. At RPH he was found to be suffering from septic shock, complicated by cardiovascular and respiratory failure and was admitted to ICU. The septic shock resulted from sepsis. The respondent remained in RPH for 11 days before discharging himself.
The critical issues at trial were whether the respondent was suffering from pneumonia which should have been diagnosed and treated as a result of his attendance at Broome hospital on 3 July, and whether the sepsis which resulted in the septic shock and associated cardiovascular and respiratory failure from which he was found to be suffering on 5 July, had been caused by untreated pneumonia.

The findings of the primary judge

On the question of whether, on 3 July, the respondent had pneumonia which should have been diagnosed, an important issue at the trial was whether the respondent had coughed up blood and, if so, whether staff at the Broome hospital were aware that he had. The respondent gave evidence that while in the emergency department on 3 July he had coughed up a quantity of fresh, bright‑red blood, without phlegm, into a tissue which he had then thrown into a bin. The respondent said that a nurse had gone to the bin, retrieved and looked at the tissue, and then put it back. The primary judge did not accept that evidence.
The primary judge said he was inclined to believe that the respondent did produce some blood from his mouth and wipe it on a tissue, but whether it was from coughing or his dentition problems was 'moot' [56]. He considered it was likely to have come from the respondent's teeth (33, pt 1). His Honour rejected the respondent's evidence that a nurse had looked at the tissue. He found that the appellant was not aware that the respondent had produced blood from his mouth [58].
The primary judge did, however, accept the respondent's evidence that when he was discharged from Broome Hospital on 3 July he was still suffering significant abdominal pain. His Honour also accepted the evidence of his condition after he left the hospital given by two witnesses, Ms Pindan and Ms Dolby, who saw him shortly afterwards. The primary judge found Ms Pindan and Ms Dolby to be truthful and accurate witnesses.
Ms Dolby gave evidence that she saw the respondent walking home from the hospital. She said he did not look well, although the respondent told her he was 'ok'. Ms Dolby said she did not believe him. She said he didn't look well and was not his normal self.
Ms Pindan said in evidence that the respondent arrived at her house at about lunchtime on 3 July, hunched over and looking weak. The respondent complained of being cold although it was a hot day. Ms Pindan gave him some blankets and he slept on a trampoline in her garden for a couple of hours before she took him home. At that stage he was weak and in pain. Ms Pindan said that when she saw him the following afternoon, Sunday 4 July, he was the same and was complaining of stomach pain (ts 32).
The primary judge found that the descriptions given by Ms Pindan and Ms Dolby were consistent with the respondent showing signs of abdominal pain, weakness and fever within about an hour of leaving the hospital. His Honour noted, however, that neither Ms Pindan nor Ms Dolby mentioned that the respondent was breathing irregularity or coughing [42].
His Honour found that while the respondent may not have complained about abdominal pain when he was discharged from Broome Hospital on 3 July he ought to have been, but was not, asked about it. His Honour did not accept Dr Russell's evidence that he would have asked the respondent whether he was in any pain. His Honour found that to discharge the respondent without asking about his pain fell short of reasonable care [72].
A number of expert witnesses gave evidence. Two specialists in emergency medicine, Dr Raftos and Dr Vinen, were called by the respondent. Both Dr Raftos and Dr Vinen have had a number of senior appointments in emergency medicine at Sydney hospitals.
The primary judge dealt first with the evidence of Dr Raftos. On the issue of whether the respondent had pneumonia, his Honour said:
It is clear that Dr Raftos placed some reliance on his belief that [the respondent] had a cough and was coughing significantly on 3 July and had coughed up blood. His view was that [the respondent's] symptoms were consistent with pneumonia. For the reasons I have set out above, I am not convinced [the respondent] coughed up blood. And there is no evidence that he had a cough on 3 July 2004. I put that matter aside [78].
However, his Honour accepted Dr Raftos' evidence that the respondent had an elevated blood sugar level which warranted urgent investigation on the basis that a common cause of an elevated blood sugar level is infection [79], (33, pt 4). The primary judge also appears to have accepted the evidence of Dr Raftos that the respondent should not have been discharged until the Tramadol had worn off and it could be ascertained whether his abdominal pain had returned. His Honour concluded that in those circumstances the respondent should have been kept in hospital and subjected to further investigation. He found that if the respondent had been kept in a little longer it is probable that the deterioration in his condition would have become evident and appropriate treatment administered, avoiding the serious illness he developed [81].
The primary judge turned then to the evidence of Dr Vinen, noting that much of Dr Vinen's evidence concerned whether the respondent had symptoms of pneumonia [98], [102]. His Honour observed that Dr Vinen believed that when the respondent presented on 5 July he was suffering from pneumonia and therefore that he must have had symptoms of pneumonia on 3 July [94]. His Honour found, however, that Dr Vinen's opinion that the respondent had symptoms of pneumonia on 3 July was based on the mistaken assumptions that one of the respondent's symptoms on presentation was shortness of breath [88], [95] and that he had coughed up blood [91]. His Honour again concluded that the issue of pneumonia did not need to be explored and put it aside [89], [98].
The primary judge observed that on occasions in cross-examination Dr Vinen 'appeared to lose confidence about some of his views', but his Honour concluded that Dr Vinen was unshaken in his opinion that the respondent's symptoms of pain were inadequately dealt with. The primary judge continued:
As I have mentioned, much of [Dr Vinen's] evidence involved discussion about pneumonia, whether the plaintiff was diagnosed with pneumonia on his re‑presentation, and whether he had symptoms of pneumonia on 3 July. In my view, as I have stated, if it is accepted that, for whatever reason, the plaintiff should have been kept at the Emergency Department, or perhaps admitted, for longer then he was, and if it is accepted that the sepsis from which he later suffered was treatable then the plaintiff's case when pared down is simplified. He was discharged prematurely. If he had not been discharged prematurely, given the history, his deteriorating symptoms would have been identified. If they had been identified they could have been treated, and he would not have suffered damage [102].
The primary judge then turned to the evidence of the two expert witnesses called by the appellant; Professor Jelinek, professor of emergency medicine at Sir Charles Gardiner Hospital, and Dr Herrington, the director of emergency medicine at Kalgoorlie Hospital.
The primary judge canvassed the evidence given by Dr Herrington and concluded:
The gist of Dr Herrington's evidence was that he did not see anything amiss in the contemporaneous records or in Dr Russell discharging [the respondent] home when he did. The latter half of his evidence related to the diagnosis when [the respondent] returned on 5 July, in respect of which, in my view, it does not matter much whether it was pneumonia or sepsis. What is important is that it was a curable or treatable infection [136].
His Honour referred to Dr Herrington's evidence that there was no clear evidence that the respondent was very sick on 3 July and that this was missed. His Honour said in relation to that evidence:
That may be so, it seems to me, however there was clear evidence that [the respondent's] condition commenced to deteriorate within a short time, maybe an hour following discharge. And it is clear that [the respondent's concerns that he was ill was well founded [138].
The primary judge described Professor Jelinek as 'an impressive witness' [150]. His Honour noted Professor Jelinek's evidence that he could find no evidence from the medical records of any indications that while the respondent was in the emergency department on 3 July he was suffering a serious underlying illness. His Honour also noted Professor Jelinek's opinion that on the respondent's presentation it was reasonable to make an assessment of non‑specific abdominal pain and to treat him with analgesics and antiemetics to relieve the symptoms.
His Honour then turned to some evidence of Professor Jelinek which his Honour considered to be of particular significance. His Honour noted that Professor Jelinek did not accept a statement by Dr Raftos that if the respondent had been kept in hospital for observation on 3 July it is likely that changes in his vital signs and clinical features would have been detected and that that would have led to a diagnosis of lower left lobe pneumonia some time later on 3 July or on 4 July. Professor Jelinek considered that it was not possible to know when the respondent's vital signs and clinical features changed after 3 July 2004, only that they were abnormal on presentation on 5 July. Professor Jelinek also gave evidence that sepsis can develop very rapidly and that on 5 July the respondent's illness appeared to be fulminate. Professor Jelinek went on to say that he considered it difficult to understand how the respondent's symptoms on 3 July could lead to a fulminate septic illness on 5 July (exhibit 5.6), even taking into account Ms Pindan's evidence that the respondent was feverish within about an hour of leaving the hospital (ts 85).
The primary judge considered, however, that it was possible to know when the respondent's vital signs and clinical features changed: Based on the evidence of Ms Pindan, within an hour or so of leaving Broome Hospital the respondent had developed signs of fever.
His Honour concluded that he could not find what caused the respondent's illness on 3 July or whether the respondent was suffering from pneumonia on 5 July, but he found that had the respondent been kept in for observation on 3 July, whatever it was that caused his critical illness on 5 July would have been diagnosed and treated, with the result that the critical illness would not have developed. He said:
It is not possible to form a view about what was causing the plaintiff's pain when he first presented. The diagnosis for his serious illness 36 hours later is also disputed. As I have mentioned, it might have been pneumonia on both occasions or it could have been some infection that resulted in the sepsis. Be all that as it may, it seems to me to be clear that if the plaintiff had been kept in the hospital for a brief time longer, then he would have avoided serious harm. There seems no dispute that whatever he suffered from on 5 July was curable.
In short, in my opinion, the plaintiff was still suffering pain when he was discharged and he had a blood sugar level that required an investigation for the possibility of infection. The failure to keep him in the Emergency Department longer to deal with his pain and/or test further for infection, both of which would have required his admission for in excess of the period of time by which his symptoms deteriorated, amounts to a failure to take proper medical care. The plaintiff may well not have been particularly communicative, but he did not receive appropriate attention on the day in question. I find for the plaintiff (33 ‑ 34).
The primary judge entered judgment for the respondent in the agreed sum of $37,500 and costs.

The grounds of appeal

The appellant relied upon the following grounds of appeal:
1.The learned trial judge erred in finding that the respondent had proven his case in circumstances where, as he also found, he was unable to conclude that the respondent had been suffering from Community Acquired Pneumonia (CAP) during the period complained of. That the respondent was suffering from CAP was the fundamental basis of the Respondent's case on liability, both as pleaded and as maintained at trial.
2.The learned trial judge erred in finding liability established on the basis that had the respondent been kept in hospital for a brief time longer on 3 July 2004 he would have avoided serious harm, when there was no finding of the cause of his illness. Without the cause of the illness being known there was no basis to support such a conclusion which in the circumstances was, accordingly, entirely speculative.
3.The finding that had the respondent been kept in hospital for a brief time longer on 3 July 2004 he would have avoided serious harm was in any event contrary to the evidence.

The disposition of the appeal

Ground 1

It is well‑established that, except in cases where the parties choose to disregard the pleadings and to fight the case on issues chosen at the trial, the relief which may be granted to a party must be founded on the pleadings: Dare v Pulham [1982] HCA 70 [6]; (1982) 148 CLR 658, 664.
In Banque Commerciale SA (In Liq) v Akhil Holdings Ltd [1990] HCA 11; (1990) 169 CLR 279, Mason CJ and Gaudron J explained the position as follows:
The function of pleadings is to state with sufficient clarity the case that must be met. In this way, pleadings serve to ensure the basic requirement of procedural fairness that a party should have the opportunity of meeting the case against him or her and, incidentally, to define the issues for decision. The rule that, in general, relief is confined to that available on the pleadings secures a party’s right to this basic requirement of procedural fairness. Accordingly, the circumstances in which a case may be decided on a basis different from that disclosed by the pleadings are limited to those in which the parties have deliberately chosen some different basis for the determination of their respective rights and liabilities [18], (286 ‑ 287). (citations omitted)
Where it is alleged that a point was not taken at trial it is necessary to look at the actual conduct of the proceedings. In deciding whether or not a point was raised at trial no narrow or technical view should be taken: Water Board v Moustakas [1988] HCA 12 [14]; (1988) 180 CLR 491, 497. It is necessary to determine what can fairly be regarded as having been identified by the pleadings as an issue to be fought at trial and in addition, whatever the content of the pleadings, what can fairly be regarded as an issue raised at trial and accepted by the parties as an issue to be determined.
In this instance, central to the respondent's pleaded case was the contention that the respondent had pneumonia on 3 July which, because it was not then diagnosed and treated, caused the critical illness from which he was found to be suffering on 5 July. It is also clear that at the trial there was no substantive departure from that case. On the appeal, counsel for the respondent conceded as much (ts 34). It is evident that that concession was properly made and indeed, in the circumstances, inevitable.
The written outline of the respondent's case at trial expressly identified the critical issue on causation as being whether the cause of the respondent's symptoms from 3 July 2004 onwards was pneumonia. It was said to be critical because it affected the likelihood that further investigation and treatment would have averted the serious illness which the respondent subsequently contracted.
In his written outline of closing submissions, counsel for the respondent pointed out that while it was not the respondent's case that the appellant was negligent in failing to diagnose pneumonia on the basis of the respondent's presenting symptoms on 3 July, it was the respondent's case that if the further observations and investigations which were called for by those symptoms had been carried out pneumonia would have been diagnosed (submissions [6] ‑ [9]). The respondent's case was put on the basis that the cause of his symptoms on 3 July was pneumonia; that this would have become apparent upon the further investigations that should have been carried out; that once it was diagnosed the respondent would have been started on a course of treatment with antibiotics; and that this treatment would have averted the critical illness which he developed (submissions [90]).
It is unnecessary to refer in any detail to the appellant's closing submissions at trial. Unsurprisingly, they reflected an understanding that the respondent's case turned on whether at all material times the respondent was suffering from pneumonia and to the extent there might have been any doubt about that, it was put to rest by the following exchange in the course of the oral closing submissions at the trial of Mr Clyne, counsel for the defendant (the current appellant) (ts 100 ‑ 101):
CLYNE, MR: … This plaintiff has pleaded that at all material times so on both 3 July and 5 July he had community acquired pneumococcal pneumonia. Now, if they haven't proven that the case is gone.
JOHNSTON, MR: I can just perhaps make a concession on that. I think the diagnosis of pneumonia is important because on a causation perspective it's important for us to show there was pneumonia so that antibiotics, had they been administered on Saturday, would have been effective in avoiding the progression in his illness. So if we're not able to identify what the illness is, no finding could be reached as to whether remedial steps would have been effective. So that's why pneumonia is important, and that's why it's part of our case.
It is also unnecessary at this stage to canvass the expert medical evidence. Suffice it to say that, while it also turned on occasions to other issues, the primary focus of the expert evidence was on the issue of pneumonia. Indeed, that is apparent from observations made by the primary judge in his reasons for judgment.
As I have mentioned, the primary judge concluded that on the evidence he was unable to find that the respondent was suffering from pneumonia at any stage and he put the issue of pneumonia aside entirely. It appears that his Honour also formed the view that the issue of pneumonia was unnecessary and had made the respondent's task at trial more difficult than it needed to be. His Honour's view that the respondent's case could be put more simply is encapsulated in a passage set out earlier but which it is convenient to repeat. His Honour said:
In my view, as I have stated, if it is accepted that, for whatever reason, the plaintiff should have been kept at the Emergency Department, or perhaps admitted, for longer then he was, and if it is accepted that the sepsis from which he later suffered was treatable then the plaintiff's case when pared down is simplified. He was discharged prematurely. If he had not been discharged prematurely, given the history, his deteriorating symptoms would have been identified. If they had been identified they could have been treated, and he would not have suffered damage [102].
Consistent with that approach, his Honour went on to find, in effect, first, that the respondent should have been kept longer in the emergency department to ascertain whether his stomach pain subsided and whether his elevated blood sugar level was due to infection; secondly, if the respondent had been kept in a little longer for observation on 3 July whatever it was that caused the sepsis is likely to have been diagnosed by means of investigations that the respondent's symptoms should have prompted; and thirdly, it would have responded to treatment such that the respondent would not have developed the sepsis which led to the critical illness.
That, however, in my respectful view, was not simply to 'pare down' the respondent's case; it was to depart very substantially from it. The effect was to discard the foundation of the respondent's case that the respondent's critical illness was caused by pneumonia which could and should have been diagnosed and treated before it resulted in sepsis, and to proceed instead on the basis that whatever the respondent was suffering from on 3 July, it could and should have been diagnosed and treated before it resulted in sepsis.
But (as counsel for the respondent had made clear at the trial) the respondent's case had been deliberately framed on the former rather than the latter basis, because it was considered that unless the illness on 3 July could be identified, it would not be possible to prove that it could have been treated so as to prevent sepsis from developing. It was for that reason that the focus of the medical evidence was on whether the respondent had pneumonia and not on the case as found by the primary judge.
On behalf of the respondent, however, it was submitted that once the primary judge found that the appellant's breach of duty caused the respondent's injury, it was open to his Honour to find for the respondent and it was not to the point that he did so on a basis that fell outside the respondent's pleaded case. Counsel argued that the decision in Dare v Pulham supported that proposition.
I do not accept that submission. As appears from the judgment of the majority in Moustakas, the decision in Dare v Pulham (and the decision in Leotta v Public Transport Commission (NSW) (1976) 9 ALR 437, 446) is authority for the proposition that where, in cases alleging the breach of a duty of care, the evidence has been allowed to travel beyond the issues raised in the particulars, failure to amend the particulars will not necessarily preclude a verdict upon the facts as they have emerged. In Moustakas, their Honours observed, however, that that presupposed that the new issue or the new way of particularising the existing issue had emerged at the trial and had been litigated (497). They referred with evident approval to the judgment of Jacobs J in Maloney v Commissioner for Railways (NSW) (1978) 18 ALR 147, where Jacobs J (with whom the other members of the court agreed) said:
All it means is that a defendant cannot lie by as the evidence supporting the case emerged and then at the end of the trial submit that the issues of fact raised by that case should not be submitted to the jury or even wait for an appeal and then claim that the case which emerged should not have been left to the jury. It is entirely different from saying either that a judge of his own motion is bound to look out a case not only different from the facts pleaded but also different from that which the plaintiff seeks to have submitted to the jury or that a court on appeal may as of course be prepared to seek out from the evidence a case different from that which the plaintiff relied on at the trial (151 ‑ 152).
In the present case, it is plain that the case found by the primary judge was not a case litigated at the trial. It was not the case the appellant had come to meet and it was not a case which the appellant had had an opportunity to meet at trial. It cannot now be said that the appellant would have been unable to answer such a case. In my view, it was not open to the primary judge to determine the case on the basis he did.
I would uphold the first ground of appeal. In view of that finding it is unnecessary to deal with the other grounds of appeal. It is appropriate, however, to say something about them.

Grounds 2 and 3

These grounds advance, in substance, the same contention. The appellant submits, in effect, that the finding of the primary judge that the respondent would not have suffered the critical illness if he had been kept longer in the emergency department on 3 July was not one that was open to him on the evidence. In my view, the appellant's submission must be accepted.
It is unnecessary to canvass the question of whether the evidence was capable of establishing that whatever the cause of the sepsis might have been, it was present and diagnosable on 3 July. It is sufficient for present purposes that there was no evidence capable of supporting a finding that whatever caused the sepsis, it could have been treated so that sepsis would not have developed. It is clear from the expert evidence that there are a number of possible causes of sepsis. Whether, as from 3 July, any of those possible causes could have been treated so as to prevent sepsis from developing is a matter that could only be determined upon the basis of expert medical evidence on that question. There was no such evidence. That is not surprising in circumstances where the respondent's case had been put quite specifically on the basis that the sepsis had been caused by pneumonia because it was considered that unless the illness could be identified causation could not be established.
I would uphold these grounds of appeal.

The notice of contention

The respondent seeks to support the judgment on the basis that the primary judge should have found, as pleaded by the respondent, that at all material times the respondent had pneumonia and that had proper investigations been carried out on 3 July, it would have been diagnosed and treated, with the result that the sepsis he developed would have been avoided.
It was submitted on behalf of the respondent that the expert evidence established on the balance of probabilities that when he presented at Broome Hospital on 3 July the respondent had pneumonia and that by 5 July the pneumonia had led to sepsis, resulting in septic shock. Counsel submitted that both Dr Raftos and Dr Vinen had remained firm in their evidence at trial that at his presentation at the emergency department on 3 July the respondent had pneumonia. Although when initially reaching that conclusion they believed that the respondent had coughed up blood, both considered that even if that did not occur, pneumonia was the most likely cause of the respondent's sepsis and it was most likely that he had pneumonia at the time of his presentation on 3 July. Dr Phillips, the doctor who treated the respondent at the emergency department on 5 July, was '95%' certain at that time that the respondent had pneumonia (ts 72).
Counsel for the respondent argued that the opinions of Dr Herrington and Professor Jelinek were based on the emergency department medical notes and overlooked the respondent's evidence that he had experienced coughing and shortness of breath, those symptoms not being recorded in the medical notes. It was submitted that the primary judge accepted that the respondent was an honest witness and it followed that his Honour accepted the respondent's evidence that he had such symptoms. Counsel also argued that while Professor Jelinek had expressed the view that pneumonia was unlikely he had not offered any alternative explanation for the sepsis. It was further suggested that Professor Jelinek's evidence was coloured by the fact that he had approached the issue as a question of scientific certainty rather than probabilities.
It is convenient to deal first with the evidence of Dr Phillips. Dr Phillips' diagnosis of pneumonia was contained in the Broome Hospital patient discharge summary of 5 July which she had prepared. At trial, Dr Phillips said that that diagnosis was based on the respondent's high temperature, a very high CRP (C‑reactive protein) result and a chest x ray consistent with opacification. (The x‑ray could not be found and was not therefore available to any of the expert witnesses at trial.)
As the case was argued, the critical element in that diagnosis was the result of the chest x‑ray. That is consistent with the medical evidence that while an elevated CRP result indicates infection or inflammation, it is a 'non‑specific' or 'multi‑factorial' marker of illness (ts 23, 29, 81). The chest x‑ray was taken on a mobile x‑ray machine. In relation to the chest x‑ray result, Dr Phillips had noted on the patient discharge summary the caveat 'although mobile film with poor chest expansion'.
In her evidence, Dr Phillips acknowledged that none of the radiological reports on the chest x‑rays taken at RPH (exhibit 1) reported the consolidation in the lung which had appeared to be present on the x‑ray taken in Broome. Dr Phillips explained that mobile chest x‑rays 'are not often the best' (ts 71) and that on the x‑ray taken at Broome Hospital there was also poor chest expansion. Dr Phillips said the x rays taken at RPH would have been better. In light of that, I do not think a great deal of weight can be given to Dr Phillip's diagnosis of 5 July.
I turn then to the evidence of the expert witnesses. In his report of 28 May 2005 (exhibit 2.2), Dr Raftos listed 12 'risk factors' for pneumonia and six 'typical clinical features'. When he prepared that report Dr Raftos understood that at his presentation to the emergency department on 3 July the respondent fitted two of the risk factors (he was Aboriginal and he was over 50 years of age) and had two of the typical clinical features; namely, he had a cough and he had produced 'reddy‑brown' or blood‑stained sputum. In his report, Dr Raftos said that 'rust coloured or bloody sputum' or 'reddy‑brown' sputum is typical of pneumococcal pneumonia.
The evidence, however, did not establish that the respondent had any of the typical clinical features. The primary judge found that, contrary to Dr Raftos's understanding, the respondent did not have a cough on 3 July [87]. That finding is not challenged. In addition, the evidence of the respondent was that he had produced fresh blood rather than coloured or blood‑stained sputum. In cross‑examination, Dr Raftos acknowledged that his instructions had been that the respondent had coughed up blood. He said he may have assumed it was 'reddy‑brown' sputum (ts 65). Dr Raftos said that rust‑coloured or 'bloody' sputum is typical of pneumonia, but a patient with pneumonia can cough up fresh blood (ts 63). He agreed that bad dentition can also lead to blood being expectorated and he also agreed that bad dentition can cause sepsis (ts 64). Dr Raftos accepted in cross‑examination that if the respondent had not coughed up blood then there was nothing on his presentation on 3 July to indicate that he had pneumonia (ts 62).
Dr Raftos agreed in cross‑examination that his conclusion that the respondent had pneumonia on 3 July was based on what had happened subsequently (ts 62). Dr Raftos maintained that the chest x‑ray taken on 5 July on the respondent's admission to RPH disclosed opacification which indicated pneumonia. He agreed that the radiologist's report on the x‑ray did not report that opacification but Dr Raftos maintained that it was nevertheless present. While the discharge letter from RPH noted that the various chest x‑rays had shown no lung consolidation and the focus of the respondent's sepsis was never able to be determined, Dr Raftos maintained his view that the sepsis had resulted from pneumonia.
In his report of 30 April 2008 (exhibit 2.8), Dr Vinen had proceeded on the basis of his instructions that the respondent had coughed up blood and that he was suffering from shortness of breath on 3 July. He considered that in the light of symptoms of unexplained abdominal pain, shortness of breath and coughing up blood or sputum, the respondent's presenting signs on 3 July were consistent with pneumonia. In Dr Vinen's opinion, if a chest x‑ray had been carried out on 3 July it is probable that pneumonia would have been diagnosed. He also considered that if it had then been treated appropriately the likelihood of it progressing to sepsis would have been significantly reduced and the respondent would have made a full recovery.
In cross‑examination, Dr Vinen agreed that someone presenting with pneumonia would usually (although not necessarily) have a cough, chest or upper abdominal pain, and shortness of breath (ts 32 ‑ 34). He said he would not expect abdominal pain caused by pneumonia to be in the left iliac fossa (that is, in the lower abdominal region (ts 6)) where the respondent had complained of pain. On the issue of coughing up blood, Dr Vinen acknowledged that the poor dentition from which the respondent was suffering could cause bleeding in the mouth and he also acknowledged that it could be a source of sepsis (ts 36).

In the course of cross‑examination, Dr Vinen appeared to accept that there was an element of speculation in his conclusion that the respondent had pneumonia (ts 37). He seemed to identify the respondent's elevated respiratory rate and elevated blood sugar level as the possible symptoms of pneumonia (ts 41, 45, 47 ‑ 48). Dr Vinen acknowledged, however, that those symptoms could be caused by a number of illnesses (ts 41, 47) and that an elevated respiratory rate can be due to pain (ts 45). Dr Vinen appeared ultimately to fall back on the report of the x‑ray taken at Broome Hospital as confirming pneumonia (ts 48). In respect of the RPH x‑ray reports, which did not find signs of pneumonia, Dr Vinen suggested that the treatment the respondent had received at Broome Hospital on 5 July may have resulted in some resolution of the infection and made consolidation in the lung on the x‑rays at RPH less visible or not apparent at all (ts 48, 50).
I have earlier referred to the submission by the respondent that the opinions of Dr Herrington and Professor Jelinek were unreliable because they failed to take into account evidence accepted by the primary judge that on 3 July the respondent was short of breath and coughing. In my view, there is no substance in that submission. There was no finding by the primary judge that the respondent had those symptoms and the evidence did not support such a finding. While, as the respondent submitted, his Honour accepted that the respondent was an honest witness, it does not follow that his Honour accepted all of the respondent's evidence. Indeed it is clear that he did not. On this issue, his Honour expressly noted that neither Ms Pindan nor Ms Dolby referred to the respondent coughing or having any breathing irregularity when they saw him. His Honour accepted that their evidence was truthful and accurate. Both Ms Pindan and Ms Dolby saw the respondent on 3 July on his way home from the Broome Hospital and Ms Pindan saw him again that evening and on 4 July. Dr Russell also gave evidence that at the time of his presentation on 3 July the respondent did not have a cough or shortness of breath (ts 42). His Honour expressly found that there was no evidence the respondent had a cough on 3 July [78]. It is also apparent that his Honour did not accept that the respondent was suffering shortness of breath on presentation on 3 July [88]. There was no finding that he suffered shortness of breath at any other time on 3 July.
Professor Jelinek considered there was no indication in the medical records of any serious underlying illness on the respondent's presentation to Broome Hospital on 3 July. He accepted that coughing up blood may occur with pneumonia (although he considered that coughing up blood alone would be very unusual) but said that it may also occur in a variety of other conditions, many of them non‑infective. Typically with pneumonia the discharge was not blood but rusty brown or blood‑flecked purulent sputum (exhibit 5.3, (9)). It was also his evidence that while abdominal pain sometimes occurs in children with pneumonia, it is very uncommon in adults. Where it does occur it would be very unlikely to be the lower abdominal pain reported by the respondent (exhibit 5.3, (10, 11)).
Professor Jelinek considered that apart from the assertions of shortness of breath and coughing up blood, there was nothing in the respondent's presentation on 3 July to indicate pneumonia (exhibits 5.5, 5.6). The respondent appeared well, had no fever, hypotension or tachycardia, his chest examination was normal and he had a normal oxygen saturation result. Professor Jelinek said that a lowered oxygen saturation rate was usually a very good indicator of infection in the lung (ts 77). The respondent's normal oxygen saturation rate excluded significant respiratory infection (exhibit 5.5). Professor Jelinek considered that the combination of the clinical findings on the respondent's presentation on 3 July made pneumonia very unlikely (exhibit 5.5).
In relation to the respondent's condition on 5 July, Professor Jelinek expressed reservations about the reliability of a diagnosis based on the x‑ray taken at Broome Hospital (ts 87). Professor Jelinek noted that while the report on the x‑ray raised the possibility of pneumonia being present, further x‑rays and investigations over 11 days at RPH (including multiple septic screens) had failed to confirm pneumonia or any focus of infection (exhibit 5.3, (10)). Professor Jelinek gave evidence that it is common for mobile films in an emergency department to be difficult to interpret, particularly where there is heart failure, as appears to have been the case with the respondent (exhibit 5.3, (7)). Professor Jelinek said that if pneumonia was present he would have expected it to show up on at least the initial x‑rays at RPH (ts 75). Professor Jelinek said he had viewed the RPH x‑rays and in his opinion they did not indicate the presence of pneumonia (exhibit 8).
I do not consider there is any substance in the respondent's submission that Professor Jelinek approached the issue as a question of scientific certainty rather than probabilities. I understand that to be a complaint that Professor Jelinek refused to accept the possibility of pneumonia unless it was established beyond any question. Having read Professor Jelinek's reports and the transcript of his oral evidence, I can find nothing which supports such a contention. Nor do I consider there is any substance in the criticism that Professor Jelinek did not provide an alternative explanation for the sepsis. Professor Jelinek considered it was very unlikely the respondent had pneumonia. On the respondent's pleaded case it was unnecessary to determine what else might have caused the sepsis. There was, however, ample evidence that sepsis can arise from a number of sources other than pneumonia, including bad dentition which it appears the respondent had.
Dr Herrington considered there were no symptoms consistent with pneumonia recorded in the medical notes of 3 July (exhibit 5.9, ts 5). Dr Herrington also considered that the respondent's clinical history was not consistent with pneumonia on 5 July (ts 18). In his opinion, the respondent's symptoms on 3 July and his apparent feverishness shortly after leaving hospital were not respiratory symptoms consistent with pneumonia, and the shortness of breath on 5 July could well have been due to sepsis. Dr Herrington said that infection from multiple causes could lead to symptoms very similar to those the respondent had on 5 July (ts 18 ‑ 19). Dr Herrington considered that it was impossible to say what the cause of the sepsis was but he did not consider that pneumonia was the most likely cause (ts 21). He suggested a number of possibilities including diverticulitis, an abscess, or poor dentition or skin lesions both which he said were common in Indigenous patients (ts 24).
In my view, on the evidence it could not be concluded on the balance of probabilities that the respondent had pneumonia. The evidence of Professor Jelinek was to the contrary. So too was the evidence of Dr Herrington. The investigations carried out at RPH, including chest x‑rays and multiple septic screens over the 11 days that the respondent was an in‑patient, failed to find pneumonia or any particular source of the sepsis.
Dr Vinen, as the primary judge observed [102] and is apparent from the transcript of his evidence, appeared in the course of cross‑examination to lose confidence in the firm view he had expressed in his report that the respondent had pneumonia. In the absence of a finding that the respondent had coughed up blood or that he had shortness of breath on 3 July, the basis for Dr Vinen's view that the respondent had pneumonia largely evaporated.
The opinion of Dr Raftos that the respondent had pneumonia depended upon his understanding that the respondent had coughed up blood and that he had a cough. The evidence did not support an affirmative finding in respect of either. As Dr Raftos acknowledged, if the respondent had not coughed up blood he had exhibited none of the typical clinical features of pneumonia on 3 July (ts 61). While Dr Raftos maintained that the initial x‑ray at RPH disclosed signs of pneumonia on 5 July, that was not supported by the report on the x‑ray by the RPH radiologist, and Professor Jelinek was of a contrary view. The septic screens undertaken at RPH did not assist Dr Raftos' contention. The doubts attached to the report of the x‑ray taken at Broome Hospital were expressly noted by Dr Phillips at the time it was taken and were confirmed in the evidence of Professor Jelinek and Dr Herrington.
In my respectful view, the primary judge correctly found that the respondent had failed to make out a case that he had pneumonia on 3 July. I would dismiss the notice of contention.

Should there be a new trial?

It was submitted on behalf of the respondent that if the appeal was upheld the action should be remitted to the District Court for retrial. In my opinion, that submission should be rejected.
Where a party raises on appeal a case that it did not put at trial, a new trial to allow that case to be litigated will be ordered only in exceptional circumstances: Maloney (152); Moustakas (497 ‑ 498). In the present case there are no grounds which would justify an order for a new trial. The respondent's case at trial was advanced, for the reasons explained by the respondent's counsel, quite deliberately on the basis that the respondent's injury was caused by the failure of the appellant to diagnose and treat the respondent for pneumonia on 3 July. The primary judge, correctly in my respectful opinion, concluded that on the evidence he was unable to find that the respondent had pneumonia. The respondent's case was not put on the basis found by the primary judge because counsel for the respondent considered that on such a case causation could not be established. It would be quite unfair to the appellant for the respondent now to be allowed a new trial to put the case on a basis that he had deliberately eschewed at the original trial.
In my view, as the respondent failed to make out his claim the action should be dismissed.

Conclusion

I would:
(a)allow the appeal;
(b)dismiss the notice of contention; and
(c)dismiss the respondent's claim against the appellant.
JENKINS J: I agree with Newnes JA.