Polsen v Harrison

[2024] NSWCA 224

Court of Appeal

Supreme Court

New South Wales

Medium Neutral Citation:

Polsen v Harrison [2024] NSWCA 224

Hearing dates:

27 August 2024

Date of orders:

12 September 2024

Decision date:

12 September 2024

Before:

Gleeson JA at [1];
Harrison JA at [2];
Basten AJA at [3]

Decision:

(1)   Dismiss the appeal from the judgment and orders made and entered in the Common Law Division on 6 July 2023.

(2)   Order that the appellant pay the respondent’s costs in this Court.

Catchwords:

OCCUPATIONS – medical practitioners – professional negligence – bariatric surgery – patient discharged with intra-abdominal haematoma – peer professional opinion – whether discharge widely accepted as competent professional practice by peer professional opinion in Australia – Civil Liability Act 2002 (NSW), s 5O

Legislation Cited:

Civil Liability Act 2002 (NSW), ss 5B, 5C, 5D, 5I, 5O

Cases Cited:

Dobler v Halverson (2007) 70 NSWLR 151; [2007] NSWCA 335

South Western Sydney Local Health District v Gould (2018) 97 NSWLR 513; [2018] NSWCA 69

Sparks v Hobson; Gray v Hobson [2018] NSWCA 29; 361 ALR 115

Texts Cited:

N Chen, “Exploring the boundaries of professional practice: Sections 5O and 5P of the Civil Liability Act 2002 (NSW)”, (2022) 28 Torts LJ 1

Category:

Principal judgment

Parties:

Katrina Marie Polsen (Appellant)
Dr Richard Harrison (Respondent)

Representation:

Counsel:
D J Higgs SC / J A Hillier (Appellant)
M Windsor SC / M Hutchings (Respondent)

Solicitors:
Commins Hendriks (Appellant)
HWL Ebsworth (Respondent)

File Number(s):

2023/223536

 Decision under appeal 

Court or tribunal:

Supreme Court

Jurisdiction:

Common Law

Citation:

[2023] NSWSC 764

Date of Decision:

6 July 2023

Before:

Lonergan J

File Number(s):

2016/204451

HEADNOTE

[This headnote is not to be read as part of the judgment]

On 22 July 2013 Dr Richard Harrison (the respondent) performed a laproscopic sleeve gastrectomy on Katrina Polsen (the appellant) at Calvary Hospital, Wagga Wagga. She developed an intra-abdominal haematoma. On 25 July 2013, satisfied that the bleeding had ceased three days post-operation, the respondent discharged the appellant from hospital. On 31 July 2013, the appellant was admitted to Wagga Wagga Base Hospital with an infected intra-abdominal haematoma. Over the next three years, the appellant suffered a number of further complications arising from the initial procedure.

On 6 July 2016, the appellant commenced proceedings against the respondent seeking damages for negligence and breach of contract. The allegations of negligence covered many aspects of the respondent’s pre-operative and post-operative conduct, as well as his conduct of the operation. After a 28-day trial, on 6 July 2023 Lonergan J (the trial judge) dismissed the proceedings. On 5 October 2023 the appellant filed a notice of appeal confined to the trial judge’s finding that the respondent’s conduct in discharging the appellant from hospital on 25 July 2013 was widely accepted by peer professional opinion as competent professional practice, for the purposes of s 5O of the Civil Liability Act 2002 (NSW).

The issues raised on the appeal were whether the trial judge erred in:

rejecting evidence of the appellant’s experts as to the adequacy of the information and advice given on discharge;

placing on the appellant the onus to plead that the respondent had not informed her of the extent of her bleed and the symptoms of infection;

treating the lack of a hands-on examination of the appellant prior to discharge as immaterial;

failing to prefer one set of experts to the other; and

not finding that the acceptability of the respondent’s conduct was conditioned on satisfaction of (ii) and (iii).

The Court (Basten AJA, Gleeson and Harrison JJA agreeing), held, dismissing the appeal:

As to (i) – rejection of evidence

1. Inadequacy of information and instructions given to the appellant upon her discharge was not part of her pleaded case; nor was the respondent cross-examined on the issue: [56]. It was not open to the appellant to claim that expert opinion for the respondent relied upon the adequacy of the instructions, thus permitting her to lead evidence in the expert conclave as to their inadequacy: [57]. The trial judge did not err in rejecting the evidence: [58].

As to (ii) – pleading a failure to inform

1. The appellant’s assertion that the respondent bore the onus of proof on the s 5O defence misunderstood the nature of the statute. The section provides a standard against which a claim for breach of duty must be assessed, and qualifies the general principles outlined in s 5B for establishing a breach of duty. Although the appellant was not required to plead a claim which negatived that standard, she bore the onus of identifying the conduct that she alleged breached the duty of care: [54].

South Western Sydney Local Health District v Gould (2018) 97 NSWLR 513; [2018] NSWCA 69 applied.

As to (iii) – failure to undertake hands-on examination

1. The respondent’s evidence that he did not carry out a hands-on examination of the appellant did not contradict any assumption put to the conclave of experts. The evidence did not establish that the nursing staff who conducted examinations were not competent or able to advise the respondent as to the appellant’s condition. On one occasion an ICU doctor examined the appellant. No expert commented on the respondent’s practice as revealed in his evidence; the complaint on appeal lacked substance: [36].

As to (iv) – assessing expert opinions

1. The assumption that the trial judge was required to conduct “an analysis of the relative merits of the competing approaches advocated by the expert witnesses” was misconceived: [59]. The judge’s function was not to prefer one set of opinions to the other but to decide whether the respondent’s expert evidence satisfied the s 5O criteria: [60], [62].

As to (v) – conditionality of expert opinions

1. None of the respondent’s experts expressed a conditional opinion in the conclave report or in their concurrent evidence. Explanations proffered in their concurrent evidence in explanation of their opinions given in the conclave, did not qualify their opinions as to the respondent’s conduct: [37], [41].

JUDGMENT

1. GLEESON JA: I agree with Basten AJA.

2. HARRISON JA: I agree with Basten AJA’s reasons and proposed orders.

3. BASTEN AJA: On 22 July 2013 the appellant, Katrina Marie Polsen, underwent a laparoscopic sleeve gastrectomy undertaken by the respondent, Dr Richard Harrison, at Calvary Hospital, Wagga Wagga. On post-operative day 3 (25 July 2013), the plaintiff was discharged from Calvary Hospital. Six days later, on 31 July 2013 she was admitted to Wagga Wagga Base Hospital with an infected intra-abdominal haematoma.

4. There were many subsequent complications, including multiple admissions to hospital. On 6 July 2016, the plaintiff commenced proceedings against the respondent, seeking damages in negligence and for breach of contract. The second further amended statement of claim, on which the trial was conducted, was filed on 16 September 2021. It identified 26 particulars of negligence. The trial, before Lonergan J, ran for 28 days between 15 February 2021 and 26 October 2022. By a judgment delivered on 6 July 2023, the trial judge dismissed the proceedings with costs. ^[1]

   1. Polsen v Harrison (No 8) [2023] NSWSC 764.

5. The primary issue on appeal was limited to one question, namely whether the respondent’s decision to discharge the plaintiff on 25 July 2013 was conduct which was widely accepted in Australia by peer professional opinion as competent professional practice, for the purposes of s 5O of the Civil Liability Act 2002 (NSW). The issue, thus confined, challenged a central part of the reasoning of the trial judge, but a limited aspect of the 475 paragraphs of the judgment, which extended over 129 pages. The two particulars relevant to this aspect of the case, as pleaded were as follows:

“b.   Failure to diagnose and/or recognise the complications suffered by the Plaintiff.

…

e.   Failure to treat the complications suffered by the Plaintiff in a competent manner.”

6. Further, those particulars were in substance limited to the action taken in discharging the plaintiff on 25 July 2013.

7. At the hearing of the appeal, counsel for the appellant explored in some detail subsequent pathology testing and medical opinion as to what had in fact caused the plaintiff to suffer from an infected intra-abdominal haematoma. That material was relevant to a question of causation of loss, if negligence were established, but not to the question of negligence: the conduct of the respondent on 25 July 2013 was to be assessed according to the then known condition of the plaintiff and the known risks attending the bariatric surgery which had been performed three days earlier.

8. Further, this Court was invited to undertake an assessment of the conflicting medical opinions of eight experts, four called by each party. The experts conducted a four-hour conclave on 10 February 2021 which addressed 32 questions agreed upon by the parties and produced a joint report on liability. The experts gave concurrent evidence on 18 May 2022, seven being in Court with one of the plaintiff’s experts, an upper gastrointestinal surgeon, Mr Andrew Jenkinson, giving evidence by AVL from the UK. ^[2]

   2. As a surgeon, he adopted the English practice of eschewing the title “Dr” for “Mr”.

9. As may be inferred from the history set out above, the evidence of the experts was not limited to the issue, the subject of the appeal. The trial judge noted “six main areas” which were the focus of the experts: the following two were relevant to the submissions on the appeal:

“339   The second area of focus was the discharge of Katrina initially on 25 July 2013 given her clinical state and relevant pathology results: (issue 10), the development of the intra-abdominal haematoma and whether it was diagnosed and treated in an appropriate fashion: (issue 11), and whether there was a causal connection between the infected haematoma and the development of the leak along the staple line: (issue 12).

340   The third area of focus was the timing of development of the leak and whether it was recognised, diagnosed and treated appropriately: (issue 13).”

10. Although evidence of a gastric “leak” as a source of the infection in the haematoma was addressed, no leak was found when the plaintiff was re-admitted and a further procedure undertaken on 1 August 2013, although the appropriate tests for a leak were then undertaken. The development of a leak was, at best relevant only to the causation question, identified as “issue 12”; neither it nor “issue 13” was relevant to the primary (and dispositive) issue on the appeal.

Standard for assessing professional negligence

11. The standard of care required of a professional person is identified in s 5O of the Civil Liability Act, in the following terms:

5O   Standard of care for professionals

(1)   A person practising a profession (a professional) does not incur a liability in negligence arising from the provision of a professional service if it is established that the professional acted in a manner that (at the time the service was provided) was widely accepted in Australia by peer professional opinion as competent professional practice.

(2)   However, peer professional opinion cannot be relied on for the purposes of this section if the court considers that the opinion is irrational.

(3)   The fact that there are differing peer professional opinions widely accepted in Australia concerning a matter does not prevent any one or more (or all) of those opinions being relied on for the purposes of this section.

(4)   Peer professional opinion does not have to be universally accepted to be considered widely accepted.

12. The trial judge discussed the proper approach to the statutory standard by reference to judgments of this Court, at [415]-[419]. The judge found, consistently with authority, that the question of negligence was to be addressed by applying the test in s 5O and not by first asking whether the respondent was negligent in accordance with the terms of ss 5B and 5C of the Civil Liability Act, and then applying s 5O. The trial judge expressed her ultimate conclusion in the following terms:

“Did Dr Harrison act in a manner that (at the time his professional service was provided) was widely accepted in Australia by peer professional opinion as competent professional practice?

457   The short answer is yes. It cannot be suggested that the combined force of opinion of Professor Brown and Dr Garett Smith as bariatric surgeons practicing in 2013 in Australia, and the content and nature of their evidence is anything short of representing the standard applying to competent professional practice in 2013 that was widely accepted in Australia. Their opinions are entirely supportive of Dr Harrison’s management. That position was further supported by the analysis and evaluation of the clinical course undertaken by Dr Harrison, by the experience and expertise of Dr Byrnes and Dr Sethi from the point of view of a gastroenterologist.”

13. Consistently with that conclusion, the judge stated, succinctly, that the plaintiff having failed to prove negligence of the respondent, whether the harm suffered by the plaintiff was caused by the respondent, in accordance with s 5D, did not need to be considered: at [459].

14. The judge also concluded that “all of the complications that befell [the plaintiff] were well-known risks and sequelae of the gastric sleeve procedure” and were thus inherent risks for the materialisation of which the respondent was not liable, pursuant to s 5I of the Civil Liability Act: at [460]-[461].

Issues on appeal

15. It is convenient to set out in full the seven appeal grounds identified by the appellant in her further amended notice of appeal filed on 20 May 2024:

“1 The primary judge erred in finding that the respondent acted in a manner that, at the time the services he provided to the appellant from after the gastric sleeve gastrectomy of 22 July 2013 was undertaken by him until 31 August 2013 when the appellant was admitted to Wagga Wagga Base Hospital (WWBH), was widely accepted in Australia by peer professional opinion as competent professional practice as provided for in s. 5O of the Civil Liability Act 2002 (the Act).

2   In relation to those findings the primary judge failed to provide adequate reasons; including but not limited to an analysis of the relative merits of the competing approaches advocated by the expert witnesses.

3   The primary judge erred in ruling that the appellant's experts could not give evidence about the advice she received when discharged from the Calvary Riverina Hospital; and further failed to address this issue when raised in closing submissions by the appellant.

4   The primary judge erred in finding that any breach of duty by the respondent would likely not have caused any damage.

5   In relation to the findings in respect of the issue of causation, the primary judge failed to give adequate reasons.

6 In respect of the aforesaid period from 22 to 31 July 2013, the primary judge erred in finding that s 5I of the Act (inherent risk) provided a complete defence to this claim for damages.

7 In relation to the findings in respect of s 5I, the primary judge failed to give adequate reasons.”

16. Three initial points should be made in relation to these grounds. First, in relation to ground 1, counsel for the appellant accepted that the reference to “31 August 2013” should have been to “31 July 2013”. That is, the complaint did not extend to steps taken after 31 July 2013, the appellant’s infected intra-abdominal haematoma having been drained on 1 August 2013.

17. Secondly, in so far as ground 2 relates to a lack of reasons, the trial judge’s summary of the plaintiff’s submissions should be noted:

“431 The plaintiff’s written submissions filed on 30 September 2022 did not address the role or requirements of s 5O at all on the basis that s 5O was insufficiently pleaded in the Defence filed. That submission is misplaced, and I reject it.

432   The plaintiff’s written submissions fail to engage at all with the inquiry the Court has to make because they do not adequately acknowledge and deal with the expert evidence relied upon by the defendant that is supportive of the defendant’s management, other than to manufacture a conspiracy theory that the defendant’s experts were all ‘misled’ by the entry in Dr Harrison’s medical records regarding what he wrote to Dr Bartusek GP on 28 May 2013 (and what he told the Court), was his understanding as to what [the plaintiff] had told him regarding the reduction in her alcohol intake.”

18. Thirdly, with respect to grounds 4 and 5, an issue as to causation only arose if there was liability in negligence, that is assuming the appellant succeeded on grounds 1-3, or any one of them. Similarly, with respect to grounds 6 and 7, the absence of liability based on inherent risk also arose only if there were otherwise a finding of negligence on the part of the respondent.

19. It is convenient to deal with grounds 1-3 together.

Applying the s 5O standard of care

20. The case with respect to ground 1, as presented on the appeal, was twofold. First, it was suggested that the opinions expressed by the respondent’s experts did not address the facts as found at trial. This, it was submitted, included both the symptoms complained of by the appellant, and the conduct of the respondent. Secondly, it was submitted that the acceptance by the respondent’s experts of the conduct of the respondent as conforming to a standard of competent professional practice widely accepted in Australia was conditional and the conditions were not satisfied. For these reasons, it was submitted that the trial judge should not have accepted that the respondent complied with the standard prescribed by s 5O.

21. These submissions will be addressed, but two further preliminary comments are appropriate. First, the case before the trial judge was not run on the basis now proposed. That may not be fatal, but it requires that care be taken in addressing the submissions based on facts found by the trial judge. Where the appellant sought to rely upon aspects of the evidence, and in particular symptoms complained of by the appellant which had not been the subject of findings by the trial judge, or were rejected, the submissions must fail: no ground challenged the findings of primary fact, nor sought that this Court make such findings.

22. Secondly, the submissions used the term “unreasonable” on a number of occasions, to refer to the expert evidence presented by the respondent. That terminology was apt to mislead: evidence of competent professional practice was not to be rejected on the basis that it was “unreasonable”, but only if it were “irrational” in the sense provided in s 5O(2). The concepts are different.

23. It is true that there may be circumstances in which the reasonableness or otherwise of a particular opinion may bear upon whether the Court is satisfied that the opinion is widely accepted in Australia. ^[3] However, that issue did not arise in the present case. The principal witness for the respondent was Professor Wendy Brown, a general surgeon with a subspecialty in upper gastrointestinal surgery and bariatric surgery who was actively practising bariatric surgery in Australia in 2013, was the President of the ANZ Oesophago-Gastric Association, a past president of the Obesity Surgery Society of Australia, the past senior examiner in the Court of General Surgery for the Royal Australian College of Surgeons, the Chair of the International Federation for the Surgery of Obesity Scientific Committee and the Federation’s Global Registry. She was eminently qualified to speak on questions of competent professional practice as widely accepted in Australia by peer professional opinion in relation to bariatric surgery. Her views were supported by Dr David Byrnes, (a gastroenterologist) who had interviewed and assessed the plaintiff in March 2017; Dr Siddarth Sethi (a gastroenterologist and hepatologist) and Dr Garett Smith (a gastrointestinal and bariatric surgeon of 17 years’ experience). As will be noted below, there was a significant degree of agreement between the plaintiff’s and respondent’s experts, though not uniformity of view as to whether the appellant should have been discharged on 25 July 2013.

    3. See, eg, Sparks v Hobson; Gray v Hobson [2018] NSWCA 29; 361 ALR 115 at [88].

20. The suggestion in ground 2 that it was the function of the trial judge to assess the “relative merits of the competing approaches advocated by the expert witnesses” does not reflect the express acknowledgement in s 5O(3) that there can be “differing peer professional opinions widely accepted in Australia”. ^[4] That some competent peers might have drained the haematoma before discharging the plaintiff was not inconsistent with satisfaction of the s 5O standard.

    4. The background to, and purpose of, s 5O is explained by N Chen, “Exploring the boundaries of professional practice: Sections 5O and 5P of the Civil Liability Act 2002 (NSW)” (2022) 28 Torts LJ 1, 2-3.

21. As to the basis of the experts’ reports, it must be emphasised that the subject of their opinions was the conduct of the respondent. That in turn was to be assessed by reference to the circumstances of the patient and relevant medical and hospital records available to him when he made the decision to discharge the patient, in the light of accepted risks. It was not to be judged by information obtained subsequently, nor by reference to evidence given by the patient nearly a decade later in the witness box. It was, of course, open to the experts to note that information was missing, and that further examinations or tests should have been undertaken before discharging the patient. Again, however, that was to be assessed as at the time the patient was discharged.

22. The critical issue in the present case was that the patient had suffered intra-abdominal bleeding after completion of the operation on 22 July. The loss of blood was obvious from the existence of the visible haematoma. The underlying concern was that, although blood was sterile, the haematoma could have become infected, either as a result of bacteria introduced at the time of the operation (such as staphylococcus aureus) or because of gastric leakage from the staple line created during the operation. Suggestions that the operation itself had not been competently performed, as had been pleaded, were not pursued on the appeal.

23. One issue raised on the evidence was whether the respondent should have ordered a CT scan before approving her discharge. As counsel for the appellant conceded in oral submissions: ^[5]

“It was debated between the experts as to whether or not the respondent should have taken a CT scan when the bleed was diagnosed. There was a body of evidence to the effect the CT scan wouldn’t have gone anywhere because the presumptive diagnosis was that it was a large bleed, it was near the leg and, therefore, it was not necessary. So I’m not taking that point.”

5. CA Tcpt, 27/08/24, p 27(30).

28. Counsel also conceded that there were “respectable points of view” as to whether to evacuate the haematoma; accepting that “if you went with a conservative approach, unless there was a suspicion of an infection, then you weren’t compelled to … evacuate … the haematoma”. ^[6]

    6. CA Tcpt, p 27(45).

29. In written submissions, the appellant identified the “conditions” imposed by the experts on the following bases: (i) that although the appellant had given evidence of severe abdominal pain, consistent with infection, the experts assumed there was no basis for inferring infection; (ii) that the absence of “rebound tenderness” as a sign of peritonism, was important; (iii) the appellant had been given good instructions and a planned follow-up; and (iv) proper advice on discharge would have included advice as to her haematoma and the need for her to report “ongoing significant abdominal pain”.

30. The appellant’s contention that the respondent’s experts’ opinions were expressed conditionally, was explained in oral submissions: ^[7]

“Because one of the – the main issue in this case was whether or not it was appropriate to discharge the appellant on 25 July. We, in large part, rely upon the evidence of the respondent’s experts that were largely accepted by the trial judge, namely that on – in our submission, a precondition to the discharge was that she had been properly examined and she had been properly informed about what to look out for down the track once she was discharged.”

7. CA Tcpt, p 3(47).

31. Referring to the nursing notes having recorded that the respondent examined the appellant on 22 July, the appellant submitted that –

“… the respondent never examined the appellant by touching her, or palpating her abdomen, or doing anything of that sort. All he did was look at her, talk to her, and ask the nurses as to what they had done. And in our respectful submission, that was not part of what the respondent’s experts say needed to be done in order to justify an early discharge, because there was always the risk of an infection of the haematoma, the very large haematoma, being infected”. ^[8]

8. CA Tcpt, p 4(25).

The first challenge – basis of opinions

32. There was no question asked in the conclave directed to the respondent’s decision not to drain the haematoma prior to discharge of the patient on 25 July 2013. Question 16 asked the significance of the patient’s haemoglobin levels during her admission, all the experts agreeing that the post-operative drop in haemoglobin reflected a blood loss and a significant post-operative bleed. There was no issue that the respondent understood there to be a significant haematoma. A small haematoma had been noted at 3pm on 23 July, which had increased in size by 5.45pm. The nursing notes referred to the abdomen as remaining “lax and non-tender and clinical observations within normal limits”. The respondent reviewed the patient at that time. The following day the patient had developed a higher temperature (38⁰C) but with a normal pulse rate and blood pressure. The respondent referred to her as “haemodynamically stable”. At 6.30pm on 24 July the nursing notes recorded that the haematoma on the right side of the abdomen had increased in size, the patient remained febrile at 38⁰C but with a normal blood pressure and heart rate. She was seen by an ICU doctor who examined her abdomen which was soft, suggesting that there was no peritonitis. Blood tests were ordered. The tests, returned on 25 July, indicated that her haemoglobin level had dropped significantly, which suggested to Professor Brown that she had lost at least 500 ml of blood. ^[9]

    9. Report, 29 July 2019, p 3.4.

33. The respondent ordered further tests be undertaken on the morning of 25 July which indicated that the HG level had increased to 88G/L. Her temperature at 5.15am was normal (37⁰C) as were her pulse and blood pressure. She was described as haemodynamically stable, afebrile, ambulating and tolerating a liquid diet. The haematoma was described as “stable”.

34. In the course of cross-examination, senior counsel for the plaintiff put to the respondent: ^[10]

“Q.   … [I]f you don’t know the site of the bleed, and bleeds can reactivate in a small percentage of cases, isn’t it dangerous to discharge the patient?

A.   I wouldn’t discharge a patient in that scenario.

Q.   Isn’t it exactly that scenario that Mrs Polsen faced?

A.   No.

Q.   Why not?

A.   Because I was examining her and assessing her on a day-to-day basis and when it was clear that she had a haematoma with haemoglobin of 83 on the second post-operative day, I made the decision to hold her in hospital, to repeat those investigations and to watch her carefully for the next 24 hours. When a further review on the third post-operative day occurred and I checked that she had a stable, if not rising haemoglobin, normal hemodynamics, a return to bowel function, the abdominal wall that was safe and soft and she’d been tolerating a diet, she passed all criteria for discharge. To have kept her in hospital any longer, not requiring a drip, on a normal diet and progressing normally would have been unusual practice.”

10. Tcpt, 15/12/21, p 1088(45).

35. The cross-examination proceeded: ^[11]

    11. Tcpt, p 1090(32).

“Q.   It’s fair to say that you have not identified a site for this bleed? A site?

A.   Intra-abdominal.

Q.   You let her go, it seems, after you had been contacted by the nurse because she lived locally in Wagga, is that right, and she was readily accessible to you if need be?

A.   I let her go because it was safe to do so and she also lived, lived locally and had good instructions and planned follow-up.

Q.   You saw her on the last day, you say, at what time of day was it? About–

A.   I beg your pardon, the, the last post operative day?

Q.   Yes.

A.   It would have been my usual practice to see her between 7.30 and 8 o’clock in the morning.

Q.   Did you examine her on that occasion?

A.   Yes.

Q.   How did you examine her?

A.   I would have, as per my usual practice, I would have looked at the – her abdomen, abdomen, looked at her clinical parameters, her pulse, temperature, blood pressure, respiratory rate and ask questions of the nurse who was looking after her.”

36. The cross-examination took place on 15 December 2021; the concurrent evidence of the experts took place on 18 May 2022. No suggestion was made to any expert that the respondent’s evidence in the trial differed from the statement and records with which they had been briefed. There was no basis to think that they had misunderstood his statement, the hospital records or his evidence. The submission that he “never actually did any hands-on examination” did not indicate that there was a departure from the assumptions put to the experts. The Court was not taken to any expert evidence to suggest that the nurses were not competent and able to advise the doctor as to the patient’s condition. Further, the examination of the patient at 6.30pm on 24 July referring to the patient’s abdomen as “soft and dry” appeared to be a description provided by an intensive care unit doctor. There was a further record on the morning of her discharge, the nurse stating, “haematoma checked S&D [soft and dry] some area of extension on lower margin stated she was comfortable”. It is clear that some (but not all) of these records were of the nurses’ observations or examinations; the experts who had the hospital records must have known that. None raised any concern in that regard. The submission as to some change in the evidence resulting from the respondent’s cross-examination was without foundation. The point sought to be raised was entirely novel, had no immediate relevance or plausibility and must be rejected.

The second challenge – conditional acceptance

37. The second submission was that the experts’ opinions were inadequate to support the respondent’s conduct because their evidence of the acceptability of his conduct was conditioned. The submission was puzzling because, although the experts each explained and justified (to an extent repetitively) the opinions each expressed, none expressly imposed a particular conditional assumption on his or her final opinion.

38. Question 29 answered in the conclave was as follows:

“Question 29

Was the management and treatment provided to Ms Polsen by Dr Harrison in accordance with competent professional practice as accepted by peers in 2013?”

39. The answers given clearly assumed that the question was directed to the whole of 2013, both pre-operatively and post-operatively and up to December 2013, which demonstrates the difficulties of focusing on a single event from a long-running course of conduct the subject of a lengthy trial, to identify error. The answers were as follows:

“Professor Brown opines that the treatment that was provided in 2013 was consistent with practice in Australia at that time.

Dr Byrnes, Dr Sethi and Dr Garett Smith agree.

Professor Morris disagrees and states that he regards both the lack of work-up in the first place and the eventual non-referral to a specialist unit as being less than competent professional practice.

Dr Miller opines that Dr Harrison’s management of the plaintiff up to 2013 [sic] was probably appropriate, because that was within less than six months from the surgery. He believes Dr Harrison he [sic] was entitled to continue treating her the way he was treating her for that first six months, but will have more to say on the following question.

Mr Jenkinson’s opinion is that the management and treatment provided was appropriate apart from the delay in diagnosis and treatment of the haematoma.

Dr Mar Fan agrees with Mr Jenkinson’s and Dr Miller’s comments.”

40. Whilst these conclusions were not focused upon the discharge on 25 July 2013, the only material criticism of the respondent’s conduct was Mr Jenkinson’s opinion concerning “treatment of the haematoma”. However, it was conceded that, in the absence of evidence of infection, a conservative approach (not immediately draining the haematoma) was acceptable. None of Mr Jenkinson’s evidence indicated a basis for considering that the haematoma should have been identified as infected prior to discharge on 25 July, but rather relied upon the risk of infection of which the respondent was aware, and which was accepted by all the experts.

41. So far as the opinions expressed in the concurrent evidence were concerned, none of the experts resiled from the opinions expressed in the conclave, although explanations were given in support of their various conclusions.

42. Some time was spent in the course of the appeal in establishing that the infection was caused by streptococcus milleri (an internal infection) rather than by staphylococcus aureus, which may be found on the skin and thus imported into the wound during the operation. ^[12] Counsel referred to a passage in Professor Brown’s evidence suggesting an internal source of infection, when the earliest pathology results identified the infection as staphylococcus aureus. If there were an error, which may be doubted as Professor Brown expressed hesitancy in expressing the opinion, ^[13] (and no one disagreed), it was an ex post facto analysis addressed to a question concerning the causal connection between the infected haematoma and the development of a leak along the staple line. ^[14] It had nothing to do with the question as to whether the respondent acted appropriately on 25 July 2013, before any pathology was available.

    12. See Tcpt, 18/05/22, pp 1395-1396.

    13. Tcpt, p 1394(45).

    14. Tcpt, p 1390(48).

43. Subject to the question as to the rejection of evidence, ground 1 must be rejected.

Ground 3 – rejection of evidence

44. It is convenient to address next ground 3, which challenged a ruling by the trial judge that the appellant’s experts could not give evidence about the advice she received when discharged from the hospital.

45. The impugned evidence arose in the following circumstances. In dealing with the appropriateness of the discharge of the appellant from the hospital, Professor Brown said: ^[15]

“So, I would’ve – and if the patient lived locally, they’re eating and drinking well, their vital signs have been stable for a period of time and the patient was reliable, then I would be happy for them to go home at that point.”

15. Tcpt, p 1366(30).

46. Dr Byrnes was “also comfortable with her discharge”. ^[16] Dr Sethi stated: ^[17]

“I agree with the comments made by Dr Byrnes, Dr Brown and Dr Garett Smith. Ms Polsen was adequately managed here, she was not actively bleeding at the time of discharge ....”

16. Tcpt, p 1366(36).

17. Tcpt, p 1366(50).

47. Professor Brown then added a further comment: ^[18]

“With a normal pulse rate and blood pressure, regardless of what the CT scan showed, I would be reluctant to take a patient back to theatre and I would adopt a, a what we call conservative approach, so monitoring the patient, and monitoring can occur in, in the hospital, as it did, and then discharge once stable. So, I don’t think a CT scan would change management regardless if it showed the site or the size of the bleed, in my hands.”

18. Tcpt, p 1367(42).

48. Mr Jenkinson then raised an issue that “many of the experts have suggested that it was okay to discharge the patient because she lived locally”. ^[19] He added: ^[20]

“Actually, no, just a, a slight separate point. I’m not sure if Mrs Polsen was, Ms Polsen was aware that she had lost a large volume of blood when she was discharged and whether the failure of disclosure of that was appropriate as well.

…

To, to, the failure to disclose to the patient that she had lost a large volume of blood and was being discharged.”

19. Tcpt, p 1368(2).

20. Tcpt, p 1369(5).

49. Counsel for the respondent objected to the evidence and the judge stated:

“Yes, that volunteered material is struck out. Only because it’s not part of the pleaded case, Mr Jenkinson, I have to stick to the pleaded case.”

50. Dr Garett Smith then gave a further opinion in strongly expressed terms that there was no dispute that there had been an intra-abdominal bleed and no doubt as to the diagnosis. Dr Morris then intervened and stated: ^[21]

“In a patient with a 4 or 5 gram haemoglobin drop, would it not be appropriate to discuss that with the patient, warn them of the potential, especially if you’re going to send them home?”

21. Tcpt, p 1370(45).

51. Counsel for the respondent objected and, without expressly rejecting the evidence, the judge noted that she understood the objection and was required to focus on the allegations of negligence.

52. In the course of the appeal, counsel’s attention was drawn to the absence of any particular of negligence relating to a failure to warn. Senior counsel for the appellant described that as “an incorrect characterisation”, stating that it was “a failure to provide information about what she should be on the lookout for, and to discuss how serious the situation was”. ^[22]

    22. CA Tcpt, p 28(45).

53. When pressed as to the pleaded particulars, counsel submitted that it was not for the plaintiff to plead a case because the respondent bore the onus of proof on the s 5O defence. The evidence became relevant because Professor Brown had stated that the discharge was acceptable in circumstances where the patient was fully informed of her condition and the symptoms for which she should be alert. ^[23] In summary, counsel submitted that the preconditions to an acceptable discharge were (i) a proper examination for localised tenderness and (ii) an explanation to the patient of the seriousness of the bleed. ^[24]

    23. CA Tcpt, p 29(25)-(35).

    24. CA Tcpt, p 31(40)-(45).

54. As explained in South Western Sydney Local Health District v Gould,^[25] characterising s 5O as providing a “defence”, with the implication that a legal burden of proof is placed on the defendant, misunderstands the operation of the statute. ^[26] Rather, the section provides the standard against which a claim of breach of duty of care must be assessed, and in that respect qualifies the general principles stated in s 5B for establishing a breach of a duty of care. While the plaintiff does not have to plead a claim which negatives that standard, the plaintiff must establish the conduct relied upon as a breach of a duty of care. Those factual circumstances must be pleaded. To allow a new case to be put forward in the course of concurrent evidence of experts, will usually be unacceptable. One consequence of allowing such a course would likely be to require the respondent to be recalled for further cross-examination. Indeed, there might be more serious disruptive consequences in relation to other evidence.

    25. (2018) 97 NSWLR 513; [2018] NSWCA 69 at [119]-[126] (Leeming JA, Meagher JA and I agreeing) (Gould).

    26. Gould at [124], citing Dobler v Halverson (2007) 70 NSWLR 151; [2007] NSWCA 335 at [59]-[61] (Giles JA, Ipp JA and I agreeing); see also Sparks v Hobson at [17].

55. The respondent gave evidence by way of a statement dated 20 November 2020 which included the following:

“69   Ms Polsen was provided … with written post-operative discharge instructions. Attached and marked ‘M’ is a copy of that sheet. Ms Polsen was given mobile contact details for my bariatric nurses, who could be contacted 24 hours per day. When I saw Ms Polsen on the morning of her discharge, I encouraged her to call if she had any concerns. My practice is to say to patients ‘call any time. If I don’t hear from you, I assume everything is fine’. I also reminded her that she could contact Calvary Hospital 24 hours a day, and I could be contacted through the hospital if necessary. I also asked that arrangements be made for Ms Polsen to see me for a follow up consultation in four weeks.”

56. This Court was not taken to any cross-examination of the respondent in relation to that statement, nor as to the content of the post-operative instruction sheet. Significantly, when the expert evidence was objected to and rejected, senior counsel for the plaintiff did not intervene to say that there was a “failure to warn” case which he was pursuing.

57. It follows that, there being no issue at trial as to whether the instructions given to the appellant on her discharge on 25 July 2013 were adequate, it is not open to the appellant now to submit that expert opinion by the respondent relied upon the adequacy of the instructions, thus permitting the appellant to belatedly assert that they were inadequate, so as to demonstrate that a condition of the experts’ acceptance of the respondent’s conduct had not been satisfied. That submission did not merely seek to raise an issue not the subject of the trial, but sought to impose on the respondent an onus of proof which is not found within s 5O of the Civil Liability Act.

58. Ground 3 must be rejected.

Ground 2 – assessing conflicting expert evidence

59. Ground 2 involved an implicit assumption that the trial judge was required to conduct “an analysis of the relative merits of the competing approaches advocated by the expert witnesses”. That assumption is misconceived. ^[27] As the cases discussing the operation of s 5O reveal, the exercise required of the trial judge is not simply to adopt a view expressed by an expert as to competent professional practice so long as it is not “irrational”. First, the judge must be satisfied that the evidence of the experts was correctly directed to the “manner” in which the respondent provided a professional service. Usually, as in this case, the experts will be provided with the statement of claim in the proceedings, and the evidential statements or affidavits prepared by the plaintiff and by the professional service provider. That material may well give rise to evidential conflicts, the resolution of which cannot be known until judgment is delivered unless, which is usually awkward and inefficient, the trial is conducted by reference to separate questions on a staged basis. However, this is not a difficulty which is unique to cases in which s 5O is invoked: it is a commonplace issue where expert evidence is relied upon in relation to an alleged breach of duty of care (and other causes of action). There is no doubt that the judge must be satisfied that the expert opinions address conduct consistent with the court’s findings.

    27. See [24] above.

60. Further, the judge will need to decide whether the evidence supports a finding based on peer professional opinion, widely accepted in Australia, at the relevant time. What is “widely accepted” need not be universally accepted, as recognised by s 5O(3). However, the test is imprecise and may require an evaluative judgment by the trial judge. Various elements may affect such a judgment, including, self-evidently, the seniority, the practices, knowledge and experience of the experts offering an assessment of peer professional opinion. It is not merely a matter of considering whether the opinions are not irrational; the fact that the experts address possible conflicting views may demonstrate a basis for acceptance that the opinions are widely accepted, though not necessarily universally accepted, in Australia.

61. The issue in the present case was not whether the patient should have been discharged following the bariatric surgery, but rather whether some additional step, such as a CT scan, should have been undertaken before the patient was discharged. That the experts were divided as to the necessity for that course was self-evident and each party demonstrated the reasons for or against the need for such a step to be taken. Similarly, it was common ground that there was a risk of infection of the haematoma and two views were expressed as to whether the haematoma should have been drained before the appellant was discharged or whether the “conversative” approach was available, consistently with competent professional practice. On the appeal the appellant did not seek to challenge the evidence of the experts for the respondent, either as not expressing peer professional opinion as to competent professional practice, or as not expressing an opinion that was widely accepted in Australia.

62. The assumption that the judge was required to prefer one set of opinions to another was inconsistent with the language and purpose of s 5O. That she did not do so cannot be a ground of appeal and ground 2 must be rejected.

Other issues – causation and inherent risk

63. On the other hand, the trial judge did make findings as to the conclusions drawn from the lay and expert evidence which she had set out in considerable detail. Relevantly for present purposes, she stated:

“394   There was post-surgical bleeding which stopped. On 25 July 2013 when discharged, [the plaintiff] was stable and it was appropriate to discharge her.

395   A haematoma formed which was appropriately treated. Gastric leak was a known complication of the procedure, and a leak was found and appropriately treated.”

64. Those conclusions were not, however, dispositive. The judge set out the statutory framework provided by ss 5O and 5I at [414]-[428]. The trial judge then noted that “neither the defendant’s nor the plaintiff’s written or oral submissions approached the case on the basis prescribed by the Court of Appeal …, that the standard of care in professional negligence matters has been assimilated into the test stated in s 5O of the Act”: at [429]. Nevertheless, the judge did set out aspects of the plaintiff’s submissions at [431]-[447], addressing them by way of appropriate findings.

65. The judge adopted the same approach with the defendant’s submissions at [448]-[456], again making appropriate findings in respect of the particular submissions. As neither party had dealt appropriately with s 5O, the judge addressed it separately at [457], which has been set out above.

66. While it is true that the judge did not address in terms the matters now relied upon in relation to the appellant’s discharge on 25 July 2013, that was undoubtedly because, in a wide-ranging case, they received little attention from the parties. In addressing, briefly, the harm which was said to have flowed from the supposedly premature discharge, counsel could only say that the opportunity for a better outcome was lost. However, in circumstances where it was accepted by the appellant that a “conservative” approach of not seeking to drain the haematoma immediately was widely accepted as competent professional practice, the significance of the matters raised on the appeal are by no means obvious. There was no evidence that the haematoma was infected at the time of the appellant’s discharge: it was common ground that there was a risk of infection. On the appeal, counsel sought to demonstrate the likelihood that sepsis was caused by the introduction of bacteria from the skin during the operation, rather than an internal leak from the gut. However, that was simply not an issue at trial.

67. There was no submission that the intra-abdominal bleeding which led to the creation of a haematoma was other than an inherent risk of the operation and was not attributable to negligence on the part of the respondent. Even Mr Jenkinson was satisfied that once sepsis set in it was properly diagnosed and treated.

68. Given that the challenge to the application of s 5O and the finding that the respondent was not negligent must be rejected, there is no purpose in addressing the further grounds of appeal, which do not arise. It is true that the trial judge dealt with the application of s 5I (dealing with inherent risk) briefly and without full reasons: that was because the issue was not dispositive. This Court should be wary of adopting a different approach, lest it be treated in some future case as authoritative.

Orders

69. The Court should make the following orders:

1. Dismiss the appeal from the judgment and orders made and entered in the Common Law Division on 6 July 2023.

2. Order that the appellant pay the respondent’s costs in this Court.

**********

Endnotes

Decision last updated: 12 September 2024