New South Wales Court of Appeal CITATION : Johnson v Biggs [2000] NSWCA 338 FILE NUMBER(S) : CA 40917/99 HEARING DATE(S) : 27 September 2000 JUDGMENT DATE : 24 November 2000 PARTIES : Christine Heather Johnson - Appellant Michael Biggs - Respondent JUDGMENT OF : Giles JA at 1; Fitzgerald JA at 50; Santow AJA at 51 LOWER COURT JURISDICTION : District Court LOWER COURT DC 7644 of 1999 FILE NUMBER(S) : LOWER COURT P R Bell DCJ JUDICIAL OFFICER : COUNSEL : S Harben - Appellant P Brereton SC - Respondent SOLICITORS : Koops Martin, Coffs Harbour - Appellant Blake Dawson Waldron - Respondent CATCHWORDS : NEGLIGENCE - medical negligence - failure to warn - whether failure to warn of possible consequences of procedure - whether would have had procedure even if warned. D. Chappel v Hart (1999) 195 CLR 232; CASES CITED: Ellis v Wallsend District Hospital (1989) 17 NSWLR 553; Rogers v Whittaker (1992) 175 CLR 479. DECISION : (1) Appeal allowed. (2) Set aside the judgment in favour of the respondent and remit the proceedings to the District Court for a new trial limited to liability. (3) The respondent pay the appellant's costs of the appeal and have a certificate under the Suitors Fund Act if qualified. (4) Costs of the first trial follow the event of the new trial, subject to any application by reason of an offer of compromise. THE SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL CA 40917/99 DC 7644/98 GILES JA FITZGERALD JA SANTOW AJA Friday 24 November 2000 JOHNSON v BIGGS JUDGMENT 1 GILES JA: In 1993 the appellant Mrs Christine Johnson began to experience pain in her left side at the back just below the last rib. The pain got worse. The appellant consulted successive general practitioners, and through them a gastroenterologist, an orthopaedic surgeon, a neurosurgeon and a pain specialist. She received treatment from a chiropractor and from physiotherapists. A radiological examination, a CT scan, an MRI and myelogram were performed. 2 The early consultations and investigations were inconclusive and the treatment brought no relief. In the end the neurosurgeon, in consultation with the pain specialist, diagnosed "focal nerve root compression in the lower thoracic region on the left side". He recommended that if medication did not give relief neurectomy should be considered. Medication did not give relief, but nerve blocks then performed by the pain specialist at the T7/8 level did. The appellant was referred to another neurosurgeon, the respondent Dr Michael Biggs, for possible surgery. 3 The respondent saw the appellant on 6 June 1995, and later in June 1995 performed a thororascopic T7/8 and T9 intercostal neurectomy on the appellant. The surgery was successful, but the appellant encountered the complication of neurectomy known as "anaesthesia dolorosa", that is, the sensation of pain from an area not in fact sensitive to pain. Pain remained and was worse than before: the trial judge described the appellant as experiencing increased pain which she would experience for the rest of her life. 4 In her claim against the respondent the appellant alleged negligence by reason of - 1. failure to advise her (a) of all known possible alternative forms of treatment; (b) of all possible complications following surgery; and (c) specifically of the possible complication of developing anaesthesia dolorosa; 2. failure to warn her that she could be worse off after the operation; and 3. failure to properly diagnose and treat her condition. 5 The appellant's case did not involve failure to carry out the surgery with due care and skill, and the trial judge was satisfied that the diagnosis and treatment of her condition were without negligence. The trial judge was also satisfied that the focus of the referral to the respondent was surgery and that, given the investigation of the pain prior to that time, the only alternative form of treatment remaining of transcutaneous electrical nerve stimulation did not call for consideration; and he was satisfied that in any event if this form of treatment had been mentioned the appellant would not have pursued it. There was no appeal in relation to these matters. 6 The trial judge said that "the case on complications is essentially that of failure to warn that the severance of the nerve could result in a permanent condition of pain", and that he therefore treated allegations 1(b) and (c) as one. He dealt with failure to advise firstly in relation to possible complications, described as the complication of anaesthesia dolorosa, and secondly in relation to the possibility that the appellant could be worse off after the operation. 7 The appellant gave evidence that there was no advice as to the complication of anaesthesia dolorosa at all, and no warning that she could be worse off after the operation. The respondent gave evidence of the explanation he gave or would have given: I will come to the detail of the respondent's evidence. Resolving the accounts given by the appellant and the respondent called for referring to later communications between the appellant and the respondent and to other events. 8 As to the complication of anaesthesia dolorosa, his Honour preferred the evidence of the respondent to that of the appellant, and considered that the explanation was adequate. As to the possibility that the appellant could be worse off after the operation, which was treated as a separate matter from the complication of anaesthesia dolorosa, the trial judge found that the respondent did not warn the appellant that the pain could become worse. But his Honour found also that even if the appellant had been advised of the possibility that the pain could become worse, she would nonetheless have decided to undergo the surgery. 9 The appeal was concerned with the adequacy of the explanation and with causation. The appellant did not challenge the trial judge's preference for the evidence of the respondent in relation to advising about the complication of anaesthesia dolorosa, and the respondent did not challenge his Honour's finding that the respondent did not warn the appellant that the pain could become worse. The appellant submitted that, contrary to the trial judge's view, the explanation found to have been given was inadequate in a number of respects, including in its failure to make known that the appellant might be left with worse pain than was already being suffered. To the respondent's response that, even if the more complete explanation had been given, the finding that the appellant would nonetheless have decided to undergo the surgery would still apply, the appellant said that the finding could not be transposed to circumstances of more complete explanation. It will be apparent that the distinction made at the trial between advice as to the complication of anaesthesia dolorosa and warning of the possibility of worse pain was not a clear distinction. The advice and warning 10 There is no doubt that the appellant wished to be fully informed when she saw the respondent on 6 June 1995. She took a note to the consultation in which she listed thirteen questions about the possible surgery. They included "Any further complications after op? From op?"; "Very concerned that after op the pain will still be there?"; and "Can op cause any problems elsewhere?". The respondent had little actual recollection of the consultation, but did recall the appellant asking questions. 11 The respondent's notes of the consultation relevantly referred to "Opern explained" and "Risks", followed by reference to infection, haemorrhage, general anaesthetic, chest infection, thorocotomy, herpes zoster, phantom pain (denoted by the word "phantom"), and three days in hospital, but did not record more of what was said. In a report to the pain specialist dated 14 June 1995 the respondent said that he had explained "what that operation involves including the risk of infection, haemorrhage, general anaesthesia, postoperative chest problems, neuroma formation, deafferentation pain and the very small risk of haemorrhage requiring a thoracotomy". The respondent considered that the expressions "phantom pain" and "deafferentiation pain" meant basically the same thing as the expression "anaesthesia dolorosa", that is, the feeling of pain in an area of no sensation. 12 The respondent described his procedure in a consultation such as that with the appellant. According to the procedure, he went through the history, conducted an examination and reviewed x-rays, and explained "what the options are and if there is an operation involved, what the operation involves and the risks of such procedure, and that's a routine - its the same thing every time". He wrote notes as he went. 13 Over some pages in the transcript the respondent detailed this procedure as applied to the consultation with the appellant. Of immediate relevance is his evidence of the explanation which followed the explanation of the surgery itself involved - "Q. All right, then what in terms of the discussion that you had with Mrs Johnson? A. I then, as is always the case, explained what the risks of the operation are and whenever, once again, it's a routine thing, I do the same thing every time, where I always explain the three risks which apply with every operation and then I move on to the risks which are associated with this one operation. And the three risks I always explain, one, two, three, are the risks of the anaesthetic, the risk of a wound haemorrhage, and the risk of a wound infection, and that applies with any operation you do, any surgeon any operation. Q. What do you believe you said in relation to each of those three doctor, to Mrs Johnson? A. Well I can't exactly recall what I said, all I have written in my notes is that I explained the risk of the anaesthetic, risk of haemorrhage, a risk of infection. As a rule when I explain the risk of the general anaesthetic it's not always the same because it will depend on who you're operating on. If you're operating on an 18 year old healthy young man it's obviously the risk of the anaesthetic is not the same as an 84 year old that's had three strokes in the last half year and it's now [sic: not] always the same thing and I really, I can't recall exactly if I clarified what risks are associated with general anaesthesia, although normally I would, normally I would say something like, 'Well of course there's a risk with every general anaesthetic, risk including heart attack, stroke, et cetera.' Q. And in relation to infection? A. Wound haemorrhage or a wound infection, my usual line is that there's a three per cent risk of wound haemorrhage or wound infections. Q. Three? A. Yes, a three per cent risk which applies to all operations. Q. In relation to wound haemorrhage did you say? A. Yes, wound haemorrhage or infection, the risk of having a haemorrhage or an infection, the risk of having a haemorrhage or an infection in your wound is roughly three percent. Q. And what did you then move to, to talk about? A. Well they're the three risks that I apply with every procedure and I then went on to explain what risks I thought were associated with this intercostal neurectomy and I explained that there was a risk of chest infection, for example, pneumonia, and this was a direct result of the deflating of her lung and then having to re-inflate it, that there's a risk that she may have a pneumonia after the operation and I asked her whether she smoked, because obviously the risk of pneumonia is a lot higher in smokers than if you're a non smoker. I may have already explained throughout the course of explaining the operation, the risk of having to open the chest widely in case of a large amount of bleeding and I then went on to explain that there is a risk of having a flare-up of herpes zoster after this operation, I may have used the word shingles, it's the same thing, and I then went on and explained the risk that even after you divide a nerve, even though this area is completely numb, there is a risk that it may hurt in the area that you have rendered numb. As an example, if you remove a leg it can still hurt in the area that - the leg no longer exists but your mind is saying that 'it's hurting in my foot' even though the foot no longer exists. And I would have explained that and I then, I think, although I can't be absolutely sure of this, I then I think answered some questions that she had for me, and the reason that I say that is that I have written in my notes that she'd be in hospital three days, and its not something I would normally write, because I know how long she'll be in hospital and I don't usually write it in my notes, so she must have asked me how long will I be in hospital and I've written here three days." (emphasis added) 14 After he was referred to the word "phantom" in his notes, the respondent's evidence was - "Q. Are you able to say whether or not you used the word phantom when talking to Mrs Johnson? A. Well no I can't be certain, I may have. Q. And could you describe what phantom pain is and the way in which it's different, if at all, from anaesthesia dolorosa"? A. Anaesthesia dolorosa literally means, anaesthesia is no sensation, that's what anaesthesia means, no sensation. Dolorosa, I'm not sure whether it's Greek or something else, I'm not sure where it originates but it's a word for pain. So in other words, anaesthesia dolorosa is pain in an area where there isn't any sensation left. That's exactly what it means. … GREENWOOD: Q. What about deafferentation? A. Deafferentation is exactly the same thing. HIS HONOUR: Q. Well no let's put the two, the anaesthesia dolorosa together, we've got no sensation and pain sitting together, that sounds a bit like an oxymoron, opposites sitting together. So when you put the two words together what does in mean medically? A. Well exactly that your Honour, that's its pain in an area where there is no sensation. Q. Right? A. Touch sensation in other words. So in an area which is numb. And deafferentation pain is exactly the same think, it's just another word for it. Afferent fibres are the ones which run inwards from the skin to the spinal cord and they're the ones which carry the pain sensations. So if you deafferent someone you have divided those nerves carrying the afferent or the incoming sensation fibres. So once again it's pain in an area that you have deafferented or in an area which is numb. So it means exactly the same thing. Phantom pain is the same pain, the only difference is, rather than in an area which is numb, it is in an area which no longer exists. So an area that you've removed, like a leg you've removed. But all three are central mediated pain syndromes and all three are really to all intents and purposes I think, are interchangeable in this situation. GREENWOOD: Q. Just finally, in June of 1995 or thereabouts, did you use the expression anaesthesia dolorosa when talking to clients? A. No. Q. Why not? A. Well because it's not a term which they're familiar with, whereas when you explain it in a way that, you know, 'Have you heard of someone who's had their leg amputated and it still hurts in their foot?' The majority of patients will say, 'Oh yes I've heard of that, I understand that.' And they can understand it when you explain it in those words. But if you explain it as, 'Well, you know there's a risk you might get anaesthesia dolorosa' they don't know what that means." 15 This evidence was given in chief on the afternoon of 22 July 1999, at the end of which the proceedings were adjourned to 6 August 1999. Another witness was interposed on the morning of 6 August 1999, and the respondent gave further evidence in chief late in the morning. The respondent's counsel adverted to the earlier evidence and said that he wanted to take the respondent "particularly to the explanation that you recounted about risks of the particular operation that you described risks in relation to all operations and risks in relation to the particular operation". 16 The respondent then amplified the explanation he would have given in connection with herpes zoster, a viral infection causing a painful red rash in the area which the divided nerve supplied. He repeated that he may have used some other expression like shingles instead of herpes zoster. The explanation was ultimately in terms that there was "a risk that she could have a severe and worse pain in association with a rash in the area that these nerves supply and once again I showed her the area which would be numb and the area which these nerves supply" (emphasis added). The respondent had not in his earlier evidence described showing the appellant the area which would be numb and the area the nerves supplied. 17 The respondent's evidence continued - "Q. The matter of using the word deafferentiation pain, is that an expression that you believe you used with her or not? A. No its not, I don't use, as you've already heard this morning explaining that risk is an extremely hard concept to get through and as I explained when I was last up here, I do not use words like deafferentation pain or anaesthesia dolorosa because no-one understands what that means. I try and use ordinary English words with regards explaining that risk. I know that I explained it there is a risk you may be worse or have the same pain or in a wider area than what you had at present and this is a risk associated with the division of a nerve such as when you divide a nerve in an amputated limb you may have pain, a severe and worse pain in the limb which no longer exists. And that's how I try and explain it because I find, well I have found over the years that people have heard of that, they have heard that yes you can have your leg amputated and it still hurts or it hurts more in the foot which used to be there and I find the majority of people have at least heard of that happening and are able to comprehend that this occurs, when you explain it in those terms. Q. What about the distribution of the area that is going to be affected, did you say anything to Mrs Johnson about that? A. I'm unsure. HIS HONOUR: Distribution of what? GREENWOOD: Q. The pain. A. I know when I'm explaining that there's a risk of deafferentation I explain it's in the area where she feels the pain and that it may be in a wider area because it's - when you have a look at - when you look at any neuralgia it means there is pain in an area supplied by one nerve, so that's very localised, it's a very localised pain, very severe, very sharp, a localised pain. Relative to central pain such as anaesthesia dolorosa, or deafferentation pain where it isn't localised to one nerve. It's a pierce [sic] pain which often overlaps areas of supply, so often it's a wider area, and it's a pierce [sic] pain, it's not a sharp shooting pain that's associated with the neuralgia. Q. When I was talking about distribution I mean in terms of showing Mrs Johnson, do you believe you showed Mrs Johnson --- A. Yeah I showed her where the area would be numb and where if it was worse or at least the same where she would have pain , which is in that area. Q. I'm sorry you had your left hand there -- A. I'm sorry in the area your Honour that the nerves splice [sic: supply?] which is around the chest wall on the left side. HIS HONOUR: Q. Stand up please doctor and show me. A. Just in this area around the chest wall -- Q. Just a little slower please I'm trying to make notes of this, yes the lower chest wall? A. Coming from the spine and running around the chest wall into the mid line, that's the area that the nerves supply, the nerves we're talking about. I'm sorry were you not hearing that. That is the area that the nerve supplies running from the spinal cord running around the chest wall into the mid line." (emphasis added) 18 I have emphasised by underlining in the passages from the transcript I have set out the respondent's most direct evidence of the advice and warning he gave or would have given. The explanation recounted on 6 August 1999 was more full than that recounted on 22 July 1999. This was taken up by counsel for the appellant in cross-examination, and in due course was thought significant by the trial judge. 19 In cross-examination the respondent agreed that anaesthesia dolorosa or deafferentation pain was "a distinct possibility" with an operation of the relevant kind. He agreed that he would be at pains to ensure that his patient was given a full explanation of the procedure, and particularly of the risks, and that if a patient showed concern that after the operation the pain would still be there he would "address that fully". He affirmed that he would not use the word "deafferentation" to the patient, and said that deafferentation pain was what he referred to as phantom pain but that he was unable to say whether he used the words "phantom pain" to the appellant at the consultation. 20 It was put to the respondent that his evidence in chief on 22 July 1999 had not included an explanation that the pain might be worse than before the operation, the respondent's evidence continuing - "Q. There is no reference there to the pain being much worse than it was before the operation is there? A. No. I was not asked though if it was worse or the same. Whenever I explain the risk of anaesthesia dolorosa I explain that it is sometimes a worse pain than the initial pain and then other times it's not. It may be the same or less, but it may be a worse pain than the presenting neuralgia. Q. But you didn't say that's what you told her on the last occasion you gave evidence. A. Correct. Q. You're now saying that that is what you would have told her? A. Yes. Q. Is your memory getting better doctor? A. No, it's just that I wasn't asked that on the last occasion. You're now asking me specifically what I would have said with regards to the severity of anaesthesia dolorosa. Q. On the last occasion your counsel asked you to recall as best you could the conversation you had with Mrs Johnson. That's correct isn't it? A. Yes. Q. You did your best from your system and notes to do that didn't you? A. Yes I did. Q. At no stage did you relate that I would have told her the pain could be much worse. A. If that's what you say. Q. Yet you're now saying that that's what you would have said. A. Yes I am." 21 The respondent also agreed that on the first occasion on which he gave evidence he had not given evidence of saying that the appellant could have a severe and worse pain in association with a rash in the area of the nerve supply. At the conclusion of the cross-examination it was put specifically that the respondent did not advise the appellant that her pain could be worse or of the possible development of anaesthesia dolorosa, which he denied. Findings as to the advice and warning 22 The trial judge referred to the respondent's notes of the consultation on 6 June 1995, and said - "[40] In his evidence in chief Dr Biggs said that he may have used the word 'shingles' instead of 'herpes zoster'. He said he would then proceed to explain that after a nerve is divided and the area was completely numb, there was a risk that it may hurt in the area rendered numb. He said he could not be certain that he used the word 'phantom' when talking to Mrs Johnson. The Defendant was asked to explain any difference between the expressions 'phantom pain', 'anaesthesia dolorosa' and 'deafferentation pain' and his answer was that they were basically the same thing, the feeling of pain in an area of no sensation. He said he did not use the expression 'anaesthesia dolorosa' to patients in June 1995 because it was not an expression they would be familiar with. [41] In summary the Defendant's evidence on this issue is that he explained there could be pain in an area where there was no sensation by referring to the example of the sensation of pain in a limb which no longer exists. He could not be definite he used any particular, specific word to describe this condition. … [43] There was also some challenge in the evidence as to the exact medical nature of deafferentation pain, anaesthesia dolorosa, and phantom pain. To Doctors Rushworth, Champion and Gronow there made [sic] be relevant medical differentiation. What has to be explained to the patient is what the patient can understand, not merely the medical terminology used - no law was cited to establish the contrary. In Dr Biggs [sic] view each condition is adequately described in the manner used. As it is not suggested by the Defendant that he used either expression (or even 'phantom pain') to Mrs Johnson I am of the view that this dispute does not assist me at all. What is relevant is an explanation that pain could be experienced from an area of no sensitivity, and if I accept what Dr Biggs described, then that description is adequate. [44] In any event, as I understand the Plaintiff's case, it is not that the words 'anaesthesia dolorosa' were not used, but that the condition of anaesthesia dolorosa was not explained, in any form of words." 23 After discussion of other matters bearing upon the differing accounts given by the appellant and the respondent, the trial judge indicated that he preferred the respondent's evidence. Given what followed, his Honour must have been referring only to the evidence summarised in para [41] set out above. 24 The trial judge then turned to the possibility of a worse condition, saying that his "findings as to the more reliable witness are reversed". For reasons which included that the respondent's first evidence of the consultation had not included that in explaining the risks of anaesthesia dolorosa he explained that it was sometimes a worse pain than the initial pain, but may be the same or less, his Honour said that he was satisfied on the balance of probabilities that the respondent advised the appellant "of the possibility that the pain might continue, but not that it could become worse". 25 I have earlier questioned the distinction made at the trial between advice as to the complication of anaesthesia dolorosa and warning of the possibility of worse pain. As can be seen from the evidence to which I have referred, warning of the possibility of worse pain was part of advice as to the complication of anaesthesia dolorosa. A composite finding would be that the respondent advised the appellant as to the complication of anaesthesia dolorosa, but did so inadequately because he omitted from the advice that the pain in the area with no sensation might be worse than the pain presently experienced; more simply, that the respondent failed to warn the appellant of the risk that the proposed surgery might leave her with worse pain than the pain presently experienced. 26 The trial judge must be taken to have declined to accept the respondent's evidence given on 6 August 1999, evidence of explanation not only of the possibility of worse pain but also of the possibility of pain in a wider area than presently experienced and of demonstration of the area which would be numb and where there could be the pain. On his Honour's findings, the explanation went no further than that there was a risk that it may hurt, meaning that there may be a continuation of the pain presently experienced, in the area rendered numb. This must be accepted for the purposes of the appeal. Adequacy of the advice and warning 27 The appellant submitted that the trial judge fell into error because he failed to explain a number of aspects of anaesthesia dolorosa, aspects which were not adequately conveyed by the expression "phantom pain" (if it was used), by reference to the sensation of pain in a limb which no longer exists, or by reference to a risk that it may hurt in the area rendered numb. The aspects were - (i) the pain can be worse; (ii) the pain can be in addition to the existing pain; (iii) the condition is permanent; (iv) the area affected can be sensitive to touch; (v) the drugs used for treatment are only sometimes effective; and (vi) the pain can extend to areas above and below the area serviced by the severed nerves. 28 I do not think it can be doubted, on the evidence, that anaesthesia dolorosa carried the possibility of worse pain than the pain which the severance of the nerve or nerves was intended to allay. Apart from the evidence to which I next refer, so much was inherent in the respondent's evidence that he would have told the appellant of possible worse pain and that the pain could be much worse. 29 Dr David Gronow, called for the respondent, said that he would attempt to explain the relevant risk by saying that the pain may come back even though the nerves have been cut "and it may not be as bad, it may be the same or it may be worse". He said that he would certainly warn the patient that the pain could be worse, although he considered the concept difficult to convey as most patients thought that their current pain was as bad as it could be. Dr Peter Blum, also called for the respondent, agreed that he would tell the patient that there was a risk that the pain could be worse after the operation, and his evidence included - "Q. If the patient said that she was very concerned that after the operation the pain will still be there, you would tell her quite clearly that yes it could be? A. Yes. Q. And that in fact it could be worse? A. The percent of people that develop deafferentation pain is actually quite low but you would mention it as a possibility. Q. So you would have to tell her it could be worse? A. Yes. Q. And you say the percentage is quite low. What percentage are we talking about? A. Somewhere between five and ten per cent. Q. So one in twenty or one in ten? A. Yes. Q. That's quite a high risk in medical terms isn't it doctor? A. Not for this sort of condition. HIS HONOUR: Q. So five to ten per cent experienced deafferentation pain but usually that pain is quite low. It that what you are saying? A. In this sort of circumstance where you don't know the cause of the pain. The per cent of deafferentation pain likelihood goes up to about five to ten per cent. Q. Yes but did I understand you to say that deafferentation pain is usually itself quite low? A. Yes. Q. But there's a possibility of it being worse and therefore you would say it could be worse? A. Yes." 30 I doubt that the evidence established a separate aspect of the pain being in addition to the existing pain. The meaning of the appellant's submission in this respect is not clear, but the evidence to which the appellant referred was that of Dr Blum, agreeing that he would explain to the patient that there was a possibility "there could be pain over and above or additional to that which [the patient] presents with". The context was set by the preceding unanswered question referring to the possibility of the pain being "more intense". This was, in my view, the same as the aspect of possibility of worse pain. 31 The permanency of the condition of anaesthesia dolorosa was found by the trial judge. The respondent accepted that it could endure for the rest of the patient's life, as did Dr Blum. Dr Blum agreed that he would explain that if the pain did develop it could be with the patient for the rest of the patient's life. The materiality of this, however, is debatable. It is unlikely that the patient would be considering a neurectomy unless the pain presently experienced was considered to be, at the least, of indefinite duration, and probably lifetime pain. Dr David Champion, called for the appellant, said that the likelihood of the appellant's pre-surgery pain spontaneously resolving without treatment was lower than 20 per cent, maybe much lower. On the other hand, if there is worse pain it would be said that permanency has particular significance and should be emphasised. 32 Not all the evidence to which the appellant referred in support of sensitivity to touch provided support. Dr Robin Rushworth, who provided a report for the appellant but did not give oral evidence, referred to the area supplied by the nerves in question being "still painful and dysaesthetic - this is called anaesthesia dolorosa". In a later report Dr Rushworth described anaesthesia dolorosa as pain experienced in an area which has been rendered insensitive to ordinary painful stimulae. In neither case does this necessarily mean sensitivity to touch. Dr Gronow said that he had expected the appellant to complain of sensitivity from clothes touching her, but that the appellant did not complain of it. But Dr Champion recorded complaint of sensitivity to touch after the surgery, and Dr Blum agreed that he would explain to the patient that the area affected can be sensitive to touch. There was therefore evidence of sensitivity to touch as an aspect of anaesthesia dolorosa. 33 Dr Blum said that in 1999 some relief could be obtained with a number of drugs "but in '95 there weren't many of these available"; he said sometimes the drugs do not work. He did not, nor did anyone else, give evidence that he would have given an explanation about relief from medication. The appellant had been unable to obtain relief from medication prior to the surgery, and the materiality of an explanation that the drugs used for treatment are only sometimes effective must be doubted. 34 As to the area of pain, both a report of Dr Rushworth and the oral evidence of Dr Gronow referred to pain which is "felt in an area whose nerve supply … has been interrupted, and therefore arises at some higher level in the nervous system". The appellant's submission linked this with the respondent's evidence concerning pain in a wider area and central pain not localised to one nerve . Dr Blum said that he would not expect pain to appear "except for the areas above or below your section", meaning the severing at T7/8 and T9. There was an evidentiary foundation for this aspect of anaesthesia dolorosa, although it was not particularly clear. 35 The respondent suggested in the appeal that the appellant's case at the trial had not been one of anaesthesia dolorosa carrying all the aspects upon which she now relied. The appellant's particulars of injuries were not particularly informative, referring only to "[p]ainful dyaesthesia in the distribution of the T7, 8 and 9 intercostal nerves". Failure to warn that the appellant could be worse off was a separate particular of negligence from the "possible complication of developing anaesthesia dolorosa". Much of the evidence for the aspects on which the appellant now relies came from the evidence of the respondent and Drs Gronow and Blum. The cross-examination of the respondent did not take up failure to advise as to aspects of anaesthesia dolorosa other than worse pain. 36 It may be that the trial judge's statement that, if he accepted what the respondent described, then the description was adequate, was because the emphasis of the appellant's case at the trial was different from that on appeal. Nonetheless the evidence was there, and from the cross-examination of Dr Blum failure to explain at least some of the aspects of anaesthesia dolorosa other than worse pain was part of the appellant's case. The appellant is able, in my view, to present the case I have described on appeal. 37 It was not in dispute that the respondent should have warned of a material risk inherent in the proposed surgery, and that a risk was material either if in the circumstances a reasonable person, if warned of it, would be likely to attach significance to it or if the respondent was or should reasonably have been aware that the appellant, if warned of it, would be likely to attach significance to it: Rogers v Whittaker (1992) 175 CLR 479 at 490. Imposition of too high a standard in the particular case, under the influence of hindsight, must be avoided, but in large part from his own evidence I consider that it must be found that the respondent's explanation was not adequate. 38 I do not necessarily embrace inadequacy as to all the aspects of anaesthesia dolorosa on which the appellant relied in the appeal. The respondent should have explained that there was a possibility that the pain would be worse than the pain which the neurectomy was intended to allay. Worse pain is not a precise concept, and can refer not only to intensity but also to area and nature: the explanation should have included that the area in which pain was experienced might be wider, and that the area might be sensitive to touch. It may be that the evidence warranted a finding that the respondent should have specifically explained that if the pain did develop it could be with the appellant for the rest of her life, as I think that would have been understood, or about relief from medication. The explanation should have included some measure of the risk, from the evidence of Dr Blum that five to ten per cent of patients encounter anaesthesia dolorosa but that usually the pain was quite low: so a much lesser percentage of patients would encounter pain worse as to intensity, area or nature. 39 On the trial judge's findings, the respondent's explanation was inadequate in relation to at least worse pain, the area in which pain might be experienced, and sensitivity to touch. That is enough for the purposes of this appeal, and because, as will be seen, I consider that there should be a new trial as to liability, it is not necessary to go further. At the new trial it will be for the trial judge first to determine precisely what warnings should have been given, and then to determine the question of causation in the light of the warnings which should have been given. Causation 40 The trial judge addressed what he described as the consequences of the finding of failure to advise that the operation would produce worse pain. He noted that the appellant submitted that the evidence established that she would have "declined to go ahead", and that various passages in the appellant's evidence and some other evidence, which he briefly identified, had been referred to in that connection. He said - "Whilst this evidence is primarily concerned with whether the Plaintiff would have undergone surgery having been told about the possible complication of anaesthesia dolorosa, it is more than reasonable to apply to the Plaintiff the same reaction in relation to the possibility of worse pain." 41 His Honour said, with reference to Chappel v Hart (1999) 195 CLR 232, that once there had been a breach followed by evidence of injury the respondent bore the burden of showing why the appellant should not recover damages. A little later he said that it was not the appellant's task "to prove her professed refusal of the operation even if it carried 10% risk of leaving her worse", and that the respondent had to establish that the appellant would have gone ahead with the operation. The respondent contended in the appeal that this was erroneous, and that it was for the appellant to establish that she would not have undergone the surgery had she been warned that her pain might be worse. It is unnecessary to deal with this matter. For what follows I assume the trial judge's approach, arguably too favourable to the appellant, without expressing a view. 42 On the approach so described, the trial judge concluded that on the probabilities the appellant "would have decided to undergo the operation", and so that she had not established that her condition of increased pain was caused by the failure of the respondent to advise her of the risk of that condition. 43 His Honour began by stating that it was "obvious common sense that no right thinking person would undergo an operation if told there was no benefit to it as they would emerge worse than before", and that "[a] question of degree is relevant". He said that he took into account, clearly enough meaning was persuaded to his conclusion on the probabilities by, a number of matters set out against dot points - ". she had already undergone various invasive surgical procedures in the course of investigation, . she had reported considerable difficulties with her life, as in walking and carrying and the experience of pain, . she had explored the alternative modalities of physiotherapy and chiropractory and had either received no benefit or suffered more, . she had received relief from the nerve block administered by Dr Segelov, who in turn had referred her to Dr Biggs, . it is reasonable to infer in the circumstances that the Plaintiff had confidence in Dr Biggs, . Dr Biggs advised her of the fact that the surgery involved the collapsing of her lung and an extra surgeon would be in attendance, . Dr Biggs advised her of the risks of haemorrhage, . Dr Biggs advised her of the risks associated with having a general anaesthetic - including death, . Dr Biggs advised her that there was a 20% chance she would not obtain relief from pain, . she had experienced pain for many months, then relief from that pain. The operation provided her with the prospect of ongoing relief, . in the Plaintiff's own evidence there are areas demonstrating unreliability and the need for caution in accepting what she maintains, even in the face of contradictory documentary material, . in relying upon her memory there was a distinct tendency in the Plaintiff's evidence for it to present a better picture than in fact actually existed. Her recollection provided a 'coloured' version." 44 The appellant's submissions included that the trial judge's references to unreliability and the need for caution in accepting what she maintained, and to a tendency to present a better picture than existed, were unwarranted. At an earlier part in his reasons the trial judge had said that the appellant's cross-examination revealed a picture of her pre-surgery condition which was significantly different from that stated by her in her evidence in chief, so that her evidence had to be approached with caution, and had also detailed a number of other aspects of her evidence which he found unsatisfactory. His Honour had the advantage of seeing and hearing the appellant, and despite what was put to us on her behalf in the appeal I consider that his assessment of the appellant's evidence can not be departed from. 45 The appellant gave evidence in chief that she would not have had the surgery if she had been informed of the possibility of anaesthesia dolorosa, if she had been informed that if the operation was unsuccessful she would be worse off, or if she had been aware of a risk that she would end up post-operatively as she in fact did. In cross-examination she generally repeated this. The difficulties inherent in retrospective evidence of this kind are, of course, well known, see Ellis v Wallsend District Hospital (1989) 17 NSWLR 553 at 560, 588 and Chappel v Hart at 246. It may be added that the firmness of the appellant's evidence was slightly undermined when she was asked in cross-examination whether she would have undergone the surgery if told of a 90 per cent chance of being better but a 10 per cent risk she would be worse off. She at first said that she was not sure, but then said that she would not have taken the risk. The appeal must proceed on the basis that the trial judge declined to accept the appellant's evidence at least in relation to being worse off. 46 The appellant stated her "reaction in relation to the possibility of worse pain", and the trial judge's reasons for finding that, on the probabilities, the appellant would have decided to undergo the operation encompassed the hypothesis that she had been told about the possible complication of anaesthesia dolorosa and warned of the possibility of worse pain. The finding would hold good if the hypothesis were that the respondent had explained to the appellant not just that there could be pain in an area where there was no sensation but also that the pain could be worse than the pain which the neurectomy was intended to allay. The question is, then, whether it can be said on appeal that the finding would also hold good if the hypothesis included explanation that the area in which pain was experienced might be wider and the area might be sensitive to touch, and explanation that while 5 to 10 per cent of patients encountered anaesthesia dolorosa usually the pain was quite low, so that a much lesser percentage of patients would encounter pain which was more intense, was experienced in a wider area, and involved sensitivity of the area to touch. 47 On one view, where the trial judge declined to accept the appellant's evidence in the respects earlier noted this Court is in a position to determine, for example, by weighing up the other matters which the trial judge took into account, what the appellant would have done if the more complete explanation had been given. (It should be said that it was not suggested that the appellant claimed for loss of a chance, cf Chappel v Hart at 237-9, 252, 260, 278-9, 288-9.) However, there are difficulties in that course. First, the appellant had given evidence of a wider area of pain and of sensitivity to touch post-operatively, but the trial judge's attention was not focussed on this evidence so far as he considered what she would have done if she had been aware of a risk that she would end up post-operatively as she in fact did. These aspects of anaesthesia dolorosa could be significant to a person in the position of the appellant, and what the trial judge said about unreliability etcetera does not necessarily apply to the evidence. Secondly, it is not clear that sensitivity to touch was only post-operative - there was evidence both ways - and a finding of fact is appropriate, if not necessary, as to that (and possibly other matters) in order properly to assess what the appellant would have done if more fully advised and warned. Thirdly, there must have been an element of the trial judge's assessment of the appellant's demeanour in making his finding as to causation, something of which we are deprived. 48 It may be that, on a re-trial, the same result will be arrived at. But I think that there must be a new trial, and that causation can not properly disposed of on appeal. Damages were assessed in an amount which neither party questioned, and the new trial should be limited to liability. But I do not think that the new trial can realistically be further limited to causation alone. I have held back from complete appellate views as to warnings, in part because of the rather incomplete dealing with the matter at the trial, and in any event because the matter will be readdressed at the new trial. The evidence and findings at the new trial as to warnings and causation will be inter-woven, and I repeat that it will be for the trial judge first to determine precisely what warnings should have been given and then to determine the question of causation in the light of the warnings which should have been given. The appeal has been dealt with on the evidence at the first trial, and the views I have expressed as to warnings are founded on that evidence. The new trial will have its own evidence. 49 I propose orders - 1. Appeal allowed. 2. Set aside the judgment in favour of the respondent and remit the proceedings to the District Court for a new trial limited to liability. 3. The respondent pay the appellant's costs of the appeal and have a certificate under the Suitors Fund Act if qualified. 4. Costs of the first trial follow the event of the new trial, subject to any application by reason of an offer of compromise. 50 FITZGERALD JA: I agree with Giles JA. SANTOW AJA: OVERVIEW 51 The law now recognises in Australia that "a doctor has a duty to warn a patient of a material risk inherent in the proposed treatment; a risk is material if, in the circumstances of the particular case, a reasonable person in the patient's position, if warned of the risk, would be likely to attach significance to it or if the medical practitioner is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it. This duty is subject to the therapeutic privilege." See Rogers v Whitaker [1992] 175 CLR 479 at 490 per Mason CJ and Brennan, Dawson, Toohey and McHugh JJ. 52 In so doing, subject to therapeutic privilege, the patient is entitled to a truthful answer to specific queries (Whitaker (supra) at 487). Unlike the United Kingdom, the standard of care in fulfilling that duty to warn is not determined solely or even primarily by reference to the practice followed or supported by a responsible body of opinion in the medical profession, though it is a useful guide. Rather this question is ultimately for the court to determine (Whitaker at 487). 53 It is important to put the duty to warn in its therapeutic context. Risks are not always calculable but the patient needs the doctor's assistance to make an informed choice. When seeking advice about the risks of an investment, the concern is predominantly with the financial risk of making the investment. On the other hand, a duty to warn in relation to a prospective surgical operation encompasses both a duty to warn of the risks of proceeding as well as the risks of not proceeding, where the patient's physical well-being may be at stake in either case. Those risks need to be put in perspective. They require the clinician's overall assessment of their likelihood and gravity of consequence expressed so as intelligible to the lay patient and responsive to that patient's expressed concerns and questions. The duty to warn, like the duty that attends actually carrying out a surgical procedure, is assessed according to that standard of reasonable care and skill required of an ordinary skilled person exercising and professing to have the relevant special skill (Whitaker at 483). The Respondent in the present case is a neurosurgeon with specialist skills and is judged against that standard. In assessing after the event the adequacy of a surgeon's warning, it must be remembered that medicine is a science where knowledge is imperfect. It is also an art in which judgment is brought to bear by the doctor, taking proper account of the patient's questions and concerns, in giving the patient the wherewithal to reach an informed decision. Not every detail, or every contingency if of minor consequence, is necessarily relevant or useful if a warning is to be balanced and informative. 54 The present appeal is concerned with the scope of the warning that should have been given by the surgeon prior to embarking on a medical procedure, in order for the surgeon to discharge that duty to warn. It is a duty that arises in the fulfilment of the doctor's duty of care for the patient. 55 Essentially the grounds of appeal involve a twofold attack upon the trial judge's findings: (a) that the trial judge did not go far enough in identifying the deficiencies in the warning given by the Respondent surgeon, so (b) erring in deciding that the Appellant, properly warned, would still have undergone the surgery. 56 In broad terms, these questions arise in the following circumstances. The medical procedure which the Appellant underwent was a neurectomy whereby certain nerves were severed in order to relieve the Appellant's pain. That pain had been of long-standing. Though it was, as an operation, carried out without negligence, the Appellant in consequence of the operation suffered the condition known as anaesthesia dolorosa, sometimes also called deafferentation pain or in lay terms, phantom pain. In the Appellant's case, this manifested itself in worse and different pain than she was suffering before the operation and involved a number of specific complications. The risk of what was termed by the expert (Dr Blum) deafferentation pain was estimated as somewhere between 5% to 10%. Usually that pain is quite low, though there remains the possibility, which here occurred, that it could be worse than before; see Black Book, 195 expert evidence of Dr Blum. No percentage estimate was given of that latter risk. The trial judge noted that while there "is no evidence of quantification of the degree of risk of increased pain, there is evidence of the percentage of successful outcome of the neurectomy (80%)"; Red Book, 35J. The trial judge held that the risk of worse pain was not part of the warning given and should have been. Nonetheless the trial judge concluded that the Appellant would have had the operation anyway having regard to a number of factors pertaining to the patient, her circumstances and concerns. 57 However, the Appellant contends that not only should the Respondent surgeon have warned that the pain could be worse. The warning should also have described in what way the pain would be worse, or different, from the pain she had already experienced by reference to the specific complications of anaesthesia dolorosa. In particular, the warning should have encompassed the following specific complications of anaesthesia dolorosa; additional or different pain (though here replacing the existing pain), permanent in nature, sensitive to touch, with drug treatment only sometimes effective and extending to areas above and below (but not much beyond) the area serviced by the severed nerves; see 42 below. It is not disputed in the present appeal that these are the complications of anaesthesia dolorosa which the Appellant suffered. 58 The Appellant's case is that, had these complications been specifically identified, the Appellant would not have proceeded with surgery. 59 It is common ground that if this Court concludes that there was a failure in duty to warn beyond that identified by the trial judge, in or to the effect of the respects earlier identified, the matter should be remitted to the lower court to consider whether, had the proper warning been given, the Appellant would, or would not, have proceeded with the surgery. 60 The Respondent did not contend that the so-called therapeutic privilege applied, namely that applicable where disclosure would prove damaging to the patient, who is "unusually nervous, disturbed or volatile"; Rogers v Whitaker (supra) at 490. 61 I shall deal later (52 below) with the final ground of appeal. This was essentially that the trial judge erred in failing to make any finding at all on the contractual pleading in the Amended Statement of Claim and thus failed to deal with the consequence that the Respondent was in breach of that contract in failing to warn of the chance of increased or worse pain following surgery or of the possible complications of anaesthesia dolorosa and that damages should therefore have been awarded to the Appellant for one or both of these breaches of contract. 62 I turn now to the specific grounds of appeal as stated in the Amended Notice of Appeal. GROUNDS OF APPEAL 63 The first two grounds of appeal were expressed in these terms: 1. His Honour erred in failing to find that:- (a) the defendant owed a duty to the plaintiff to warn her of the possible complication of anaesthesia dolorosa. (b) the defendant failed to warn the plaintiff of the possible complications of anaesthesia dolorosa. (c) If warned the plaintiff would not have proceeded with surgery. 2. His Honour erred in holding that even if warned of the possibility of increased or worse pain following surgery the plaintiff would have still proceeded with surgery. 64 In argument, it became clear that what was meant was not that the judge was in error in eschewing the technical term anaesthesia dolorosa, likely by itself to be unintelligible to the lay person. As the High Court said in Rogers v Whitaker (supra) at 490: "… the skill is in communicating the relevant information to the patient in terms which are reasonably adequate for that purpose having regard to the patients' apprehended capacity to understand that information." Rather the crux of the appeal was that the doctor's warning of the possible complications from the operation should have described the specific features or complications of that pain condition anaesthesia dolorosa. As appears (42 below) such description could be in lay terms. Thus the Appellant contended that the judge fell into error by failing to take into account, in identifying the warning required, of those adverse features of that particular condition, anaesthesia dolorosa which the patient suffered. Their omission from the warning was therefore in breach of the doctor's duty of care. 65 The Appellant's contention is that the trial judge misdirected himself in posing the question, necessarily hypothetical, would the Appellant have had the operation had she been warned that "the pain may become worse". The question should rather have been posed, would the Appellant have had the operation if warned not only that "the pain may become worse" but also of those adverse features of anaesthesia dolorosa. 66 The Appellant then contends that since the wrong question was posed the consequence is that the trial judge's finding, that she would have proceeded with the operation, must be set aside and a new trial ordered. Posing the correct question is essential to determining whether, in causal terms, the Appellant's injury can be brought home to the Respondent's failure to give the appropriate warning of the material risks of the operation. 67 The Respondent takes issue with whether the warning need have gone further than to describe, in lay terms, as the surgeon did, the possibility of anaesthesia dolorosa and, as the surgeon did not, that it might involve worse pain. Thus, though the warning was held by the Trial Judge not to have included the possibility of worse pain, the Trial Judge took that omission fully into account. He determined that disclosure of that possibility of worse pain would have made no difference to the Appellant's decision to have the operation given her history, attitude and particular circumstances, all of which he carefully weighed up. It is as I have said common ground that if this Court is satisfied on that first ground of appeal, it should not attempt to determine the causality issue for itself but should remit the case for a new trial. 68 The final ground of appeal concerned the failure to make any finding at all on the contractual pleading in the Amended Statement of Claim, as the trial judge dealt with the matter purely as a claim in tort; see 52 below. 69 I turn now to such elaboration of the factual circumstances as is necessary to deal with the grounds of appeal. FACTUAL CIRCUMSTANCES 70 The Appellant was born in April 1955. She was forty years old when she underwent an intercostal neurectomy operation carried out by the Respondent Dr Biggs on 20 June 1995. That was an operation in which nerves — in this case the seventh, eighth and ninth intercostal nerves — were divided or sectioned by surgical incision. This was with the intended object of curing the Appellant's longstanding condition of debilitating left side rib pain. 71 Earlier in 1983 the Appellant had hit her head on a car door and had time off from work. Ten years later in December 1993 the Appellant fell down two steps at home. 72 As appears in the notes and reports of various doctors, the most contemporaneous being those of Dr Biggs on 6 June 1995 made at the time of the consultation before her operation (Blue Book, 67), the Appellant developed, by January 1994, worsening rib pain. It was felt in the left sub-costal region intermittently. She experienced difficulty with bending, twisting and sitting. Her medical history is subsequently elaborated with Dr Gronow on 1 July 1998 (Blue Book, 86) and Dr Champion 22 March 1999 (Blue Book, 102). These were both specialists, briefed respectively by Respondent and Appellant. 73 She returned to work in February 1994. It appears from wage records she was working three days per week (Blue Book, 8). 74 Still working, in April 1994 she went on holidays to Fiji. There she suffered food poisoning which became worse (notes of Dr Biggs of 6 June 1995, Blue Book, 67). That led to her consulting a general practitioner who diagnosed bowel pain following which she consulted various specialists and various scans and tests were carried out. 75 Her left side pain did then get worse. She cut down her work progressively from four days per week through three, two and finally one day per week actually ceasing work on 22 September 1994; see Appellant's cross-examination (Black Book, 41) and her evidence in chief, (Black Book, 8—9) and wage records (Blue Book, 8). I will return to her account of her work history later, in dealing with matters of credibility relied upon by the trial judge. 76 During 1994 she had extensive chiropractic and physiotherapeutic treatments. None of this gave her any lasting relief. 77 She then had further medical investigations in January and February 1995. Her attempts to deal with the pain problem by analgesics were not successful. 78 However, on 11 May 1995 the Appellant was admitted to hospital for thoracic nerve root blocks carried out at the T8 and T9 levels by Dr Salmon. His report to the referring doctor (Dr Segelov) is dated 30 May 1995 (Blue Book, 66). It describes the pain she had previously experienced. The report concluded that she had considerable relief of pain following that operation. The description Dr Salmon gives of her pain before that procedure was: "of a constant pain in her left low back and anterior chest wall. The pain was constant, and varied from a mild dull ache, to a severe, sharp, piercing sensation. The severe pain could last for days. It was aggravated by bending, lifting, movement, coughing, sneezing and rotation especially to the right …. . It was associated with burning of the back when the pain was severe". 79 However, when the Appellant saw Dr Biggs on 6 June 1995 she was, according to his report of 14 June 1995 to her referring doctor, Dr Salmon (Blue Book, 70), then experiencing "left intercostal nerve pain of a constant nature as well as a burning sensation at the middle of her back to the medial scapular region". Dr Biggs adds, "I understand from your letter that she has had complete relief from a T8 nerve block". The trial judge attached significance to her positive experience from that operation. 80 Thus the consultation occurred, as found by the trial judge, against a background where the Appellant "had been through an extensive amount of medical examination and investigation and she had been specifically referred to a neurosurgeon for consideration of surgery and her evidence was that she was looking forward to exploring that option" (Red Book, 23T). 81 The Appellant took a note with her to this consultation containing 13 questions (Blue Book, 117). Her asking those questions in that way led the trial judge to conclude that the Appellant "demonstrated her desire to be fully informed over the operation" (Red Book, 23L). That conclusion was not challenged. 82 Dr Biggs then suggested that she undergo a left T7, 8 and 9 thoracoscopic intercostal neurectomy, being the procedure that she subsequently underwent. In that same report he says: "I have explained what that operation involves including the risk of infection, haemorrhage, general anaesthesia, post operative chest problems, neuroma formation, deafferentation pain and the very small risk of haemorrhage requiring a thoracotomy." [emphasis added] 83 The expression "deafferentation pain" was thus the technical expression used by Dr Biggs in his notes though his evidence was that they were not the terms of his lay explanation. The trial judge observed that there may have been some medical differentiation as between deafferentation pain, anaesthesia dolorosa and phantom pain, but appears to have attached no significance to that; Red Book, 27G referring to the evidence of Doctors Rushworth, Champion and Gronow who suggested that medical differentiation. The trial judge accepted that the explanation to be given was what "the patient can understand, not merely the medical terminology used" and that a lay explanation was given (Red Book, 27G to P and see Black Book, 136Q to 137G for the Respondent's evidence of what he said). He refers to Dr Biggs' view, without critical comment, that "each condition is adequately described in the manner he used". That explanation was essentially that "even after you divide a nerve, even though this area is completely numb, there is a risk that it may hurt in the area you have rendered numb"; see 37.4 below (Black Book, 136Q to 137G). 84 After the unsuccessful operation and experiencing further, worse pain, the Appellant saw another specialist, Dr Terenty in July 1995 whose report appears at Blue Book, 30. 85 I turn now to the grounds of appeal. GROUNDS OF APPEAL 86 To deal with these grounds of appeal in their context, I set out below a summation of the trial judge's principal findings as they bear upon them. 1. On the Appellant's credibility, her evidence had to be approached with caution. There were areas demonstrating unreliability, though expressly no finding was made that she was deliberately untruthful. These matters principally pertained to the picture she gave of her pre-surgery condition in the face of contradictory documentary material, and omissions from a chronology submitted on her behalf at the commencement of trial. See Red Book, 18 — 20 paras (11) to (16) and Red Book, 36M. 2. There was no failure to advise on alternative forms of treatment. The Appellant had been through an extensive amount of medical examination and investigation and had been specifically referred to the Appellant neurosurgeon for consideration of surgery and was looking forward to exploring that option; Red Book, 23T. 3. A complication of neurectomy, being the complication that in effect ensued, is "anaesthesia dolorosa", described generally as "the sensation of pain in an area which is not sensitive"; Red Book, 25S. 4. There was no failure to advise that the complication of anaesthesia dolorosa could ensue. The advice was given in lay terms. This was because "what has to be explained to the patient is what the patient can understand, not merely the medical terminology used …". The explanation was essentially that "pain could be experienced from an area of no sensitivity". That finding was based upon contemporary records and evidence as to the routine followed by the Respondent, as against the Appellant's memory; see Red Book, 25O — 31E and below. It can be taken that the trial judge accepted the following evidence given by the Respondent as to the warning he had given to the Appellant in lay terms (Black Book, 136Q — 137G): "I then went on and explained the risk that even after you divide a nerve, even though this area is completely numb, there is a risk that it may hurt in the area you have rendered numb. As an example, if you remove a leg it can still hurt in the area that — the leg no longer exists but your mind is saying that 'it's hurting in my foot even though the foot' no longer exists." The contemporaneous notes of the Respondent are in Blue Book, 68 being notes of 6 June 1995 of the Appellant and in Blue Book, 70 being letter from the Respondent to Dr Salmon of 14 June 1995 reporting on the explanation given and finally Blue Book, 72 being record of phone call by Respondent to the Appellant dated 28 July 1995 following the Appellant's consultation with Dr Terenty. 5. On the balance of probabilities, the Respondent did advise the Appellant of the possibility that the pain might continue, but not that it could become worse (Red Book, 33J). The evidential onus thus passed to the Respondent to show why the Appellant as plaintiff should not recover damages; (Red Book, 34S). 6. On the balance of probabilities, had instead the Appellant been warned of the possibility that the pain might not only continue but could become worse, the Appellant would have still decided to undergo the operation; Red Book, 36W. In that regard, the trial judge took into account not just the Appellant's credibility but also his findings on a number of the factors of a more objective nature. These included that "Dr Biggs advised that there was a twenty percent chance that she would not obtain relief from pain and other factors pertaining to her past medical history and her difficulties with her life from her experience of pain"; see Red Book, 35N — 36R. 87 The Appellant took issue with the trial judge's findings as to credibility. The contention was that he over-relied on omissions from the chronology submitted by her legal advisers and failed to take sufficiently into account what the Appellant volunteered by way of doctor's reports and in examination in chief. However I do not consider that the trial judge's conclusions on credibility should be interfered with, based as they were on a number of factors, and given the trial judge's advantage on that score. The trial judge was suitably cautious, and a degree of caution attends the appraisal of any exercise in retrospective reasoning by a patient grappling with the misfortunes that have followed treatment. As Kirby P put it in Ellis v Wallsend District Council (1989) 17 NSWLR 553 at 560: "It is true that answering that question involves an exercise in retrospective reasoning. The patient cannot, when the mishap leading to damage and litigation has occurred, determine the answer authoritatively by the response in court to the question of what he or she would have done had only full and proper advice been given. However honest the patient may try to be, self-interest and the knowledge of the misfortunes that have followed the treatment will necessarily colour the patient's response to that question. Nonetheless, the answer remains an important ingredient in the decision by the fact finding tribunal as to what it thinks the patient, subjectively and at the time before operation, would have done if properly and fully advised." Likewise an investor who loses money on a misleading prospectus , as Buckley J said in Broome v Speak [1903] 1 Ch 586 at 604: "Be the man the most honest man possible, it is so easy to be wise after the event, that it is difficult for any man to say what he would have done under circumstances which did not arise." I am not however suggesting that caution should be exaggerated or one-sided. A professional person, questioned about the amplitude of his or her prior warning of risk, may also have an understandable tendency for retrospective reasoning. More recently Kirby J in Chappel v Hart (1998) 72 ALJR 1344 at 1367 whilst acknowledging in the case of a patient "the malleability of recollection of such a witness", added the caveat that "these dangers should not be overstated". 88 Moreover the trial judge's reliance upon credibility evidence against the Appellant did not prevent the trial judge preferring the evidence of the Appellant when he considered that justified. Thus he found against the Respondent in concluding that the Respondent had failed to warn that the pain could become worse. 89 I do not consider that the trial judge fell into error in his understanding of what was meant by anaesthesia dolorosa insofar as that contention was based on his broad statement of that condition at paragraph 37 of the judgment: "the sensation of pain in an area which is not sensitive" (Red Book, 25S). That as a general description was in accordance with the description of anaesthesia dolorosa in the Appellant's Counsel's opening (see Black Book, 1) and with the evidence of the Appellant's own medical expert, Dr Rushforth (Blue Book, 79 report of 26 May 1998). However, that statement needed amplification when it comes to identifying the specific complications of that condition (see 42 below). 90 That leads to the main thrust of the Appellant's case. It was that the trial judge fell into error because, in allowing as sufficient a lay explanation of anaesthesia dolorosa around the notion of "phantom pain" (though not using those terms), the trial judge overlooked the significance of omitting the specific attributes of anaesthesia dolorosa with its adverse features from the warning given. It was not enough that the trial judge identified the omission from the warning that the pain could become worse. Those adverse features should also have been stated. They could readily have been expressed in the lay terms, as their following description demonstrates. The Respondent did not take issue with these being actual or potential complications of anaesthesia dolorosa, provided the last of these were qualified by adding such words as, "but not much beyond" after the word "below". 91 I set out these complications below: (i) The pain can be in addition to the existing pain. (Black Book, 195T) (ii) The condition is permanent (Red Book, 29I; Black Book, 182K; 195W) though some relief may be available with drugs. (Black Book, 196D and E but not (iv) below) (iii) The area affected can be sensitive to touch. (Exhibit J1 Blue Book, 72 (Dr Rushforth); 74 (Dr Salmon); 76 at 77O (Dr Rushforth); 91 (Dr Davis) and Black Book, 145E; 196B; Exhibit J2 Blue Book, 79T. (iv) The drugs used for treatment are only sometimes effective (Black Book, 196D (Dr Blum). (v) The pain can extend to areas above and below (but not much beyond) the area serviced by the severed nerves. (Blue Book, Exhibit J1 76 at 77O; Black Book, 151G (Dr Biggs); 159O (Dr Biggs); 195O (Dr Blum). 92 I consider that not only should the warning given have included that there was a risk, admittedly below 5% to 10%, there could be worse pain; that much the trial judge held. That warning should have gone further. It should have stated in broad terms the known adverse attributes of that worse pain insofar as material. Thus it should have stated that such pain could be in addition to the existing pain (for the patient may treat the pain she knew differently from an unknown potentially worse and different pain), permanent, sensitive to touch and for which drugs were only sometimes effective. If each of these adverse features had been mentioned I doubt whether omitting the last complication (the possible not much extended location of the pain) would have mattered. It fits into the category of minor detail by itself not of such consequence as to require specific mention; see 4 above. Certainly the first four adverse features of the pain, particularly its permanence would have been material to this patient as well as to a reasonable patient in this patient's position. It is one thing to live with a risk that the old pain would recur , even perhaps that it might be worse; it is another to have to grapple with the possibility of permanent, worse pain, not necessarily alleviated by drugs. This low risk, as against the likelihood of continued pain without the operation, might or might not have induced the patient, so informed, to go ahead. 93 It must be remembered that material risk in this context has both a subjective and objective aspect. Consider the first aspect with this particular patient. She, as the trial judge held, demonstrated her desire to be fully informed about the operation. So too in Rogers v Whitaker (at 491), the High Court placed emphasis on the fact that the patient was keenly interested in the outcome of the suggested eye procedure. It is significant in terms of the complications she ended up suffering was that amongst the Appellant's questions in the written note which she took to her pre-operation consultation with Dr Biggs were these: "Any further complications after op?" and "Very concerned that after op the pain still be there?" and "Can op cause any problems elsewhere?" (Blue Book, 197) Even if as a layman she could not be expected to ask the precise question as to the risk that eventuated, that of new, different and worse pain, her questions were clearly directed at that kind of risk. So too said the High Court with the patient in Rogers v Whitaker who wanted to know if there was a danger of unintended or accidental interference with her good eye. 94 The same result is reached if one applies an objective test to determine what is a material risk. I am satisfied that a reasonable person in the patient's position, if warned of those concomitant aspects of the pain that ensued, enumerated in 43 above, would have attached significance to them. That is not to say, one way or the other, what decision would then have been taken about proceeding or not with the operation. 95 Once it be accepted, as I do, that a warning failed to include these specific complications in a way that would have been understood by a patient in the position of the Appellant, then it must follow the first ground of appeal is made out. Merely to have told her that she might suffer worse pain, would not have been enough. Hence the trial judge's finding that, if so warned, she would still have had the operation, could not carry with it any finding of what she would have done if warned not only that she might suffer worse pain, but also of its adverse features. Moreover, the omission that was found by the trial judge, that the pain might be worse has two consequences. First it denied this concerned patient the prompting to enquire further and be informed about the nature of that worse pain, such as its permanence and amenability to alleviation. Second, once a warning is given of the possibility of worse pain, its permanence and other adverse features are likely to be of heightened significance, as compared to those features being associated just with the original pain. Thus permanent worse pain following the operation, though of low risk, may be viewed more seriously than permanence associated with the original lesser pain. Similarly with non-susceptibility to alleviation by medication. 96 To put such warning in perspective for the patient, where some reasonably reliable indication of degree of risk is feasible, one would expect that risk indication also to be included in the warning. Otherwise there is a danger that an undifferentiated catalogue of potentially adverse complications will not contribute to an informed patient decision. It may even deter the patient from undertaking a needed operation with beneficial effects, where the patient would have gone ahead if the risks were put in some overall perspective. 97 As King CJ said in F v R (1983) 33 SASR 189 at 193-4 (expressly approved in Whitaker at 489): "What is under consideration here is the patient's right to know what risks are involved in undergoing or foregoing certain surgery or other treatment." [emphasis added] The therapeutic context in which the doctor's duty to warn arises is very relevant and informs both its content and the manner of communication. The doctor is not issuing a prospectus. Rather, the doctor has a fiduciary relationship with the patient, in which the patient places trust in the doctor to listen to the patient's concerns. The doctor is then trusted to give a fair and balanced picture, concentrating on what is material for that patient, as to the risks and benefits of proceeding or not proceeding with the operation. Then the patient can make an informed choice. 98 That leads to the second ground of appeal, namely that the trial judge erred in holding that, even if warned of the possibility of increased or worse pain following surgery, the Appellant would have still proceeded with surgery. The Respondent fairly acknowledged that to maintain the finding of the lower court, he would have to establish that, had the fuller warning been given, the trial judge must necessarily have reached the same conclusion, namely that the Appellant would have still proceeded with surgery. Here the test is a subjective one in contrast to determining what risks are material where materiality may be satisfied according to either an objective or subjective basis. Thus causality depends on what this particular patient's decision would have been, had proper information been given; Whitaker at 490 followed for example by Kirby J in Chappel v Hart (supra) at 1366-7. 99 I am satisfied that the trial judge's conclusion that the Appellant would have had the operation, if properly warned, must now be re-determined in a fresh trial; one at which the deficiencies in that warning as determined in the light of the evidence are fully taken into account. I agree with Giles JA that the new trial cannot realistically be limited to causation but will have to be on liability. Given that outcome, the question of the burden of proof and any shifting onus does not call for resolution in the present case, though raised by the Respondent's Notice of Contention. It suffices simply to note the lack of consensus in the various judgments in Chappel v Hart as to whether the coincidence of the failure to warn against a risk and that risk eventuating suffices to establish causation. Thus one view is that this coincidence is sufficient to shift the evidentiary onus to the defendant; (Gummow J at 1357-8, Kirby J at 1367 and McHugh J 1352). Another shading came from Gaudron J (at 1347); that once the plaintiff gives evidence to the effect that he or she would have acted to avoid or minimise the risk of injury, then liability follows. Gaudron J appears to leave open whether that evidence would be capable of refutation by contrary evidence from the defendant. Against these views favouring the plaintiff in varying degrees, is that expressed by Hayne J (1374-5 and 1377): that mere coincidence of patient query (about what was omitted from the warning) coupled with failure to warn and that unwarned risk eventuating are, without more, insufficient to establish causation and liability. Hayne J concludes (at 1377) that the law still requires a comparison between "the facts of what happened and the hypothetical facts of what would have happened if there had been no negligent act or omission". 100 It has not been established one way or the other, what the patient would have done if properly warned, for the trial judge did not identify all of the deficiencies in the warning. It was common ground that this court should not determine that matter. Regrettably that means that a fresh trial must be ordered. 101 The third ground of appeal needs only brief consideration. The Appellant's complaint is that the trial judge erred in failing to make any finding at all on the contractual pleading in the Amended Statement of Claim, notwithstanding that it was on all fours with the claim in tort. Given that the Appellant succeeds on the first and second grounds, the matter is in any event moot. But I should record my conclusion that requiring the trial judge to determine the matter on the basis of a breach of contract would not reflect the way in which the matter was conducted in the lower court. All that would arise were the matter determined on the basis of a breach of contract is that nominal damages would be awarded. There is nothing before us to indicate that this would have any cost consequences. In any event, such a course would be of no practical purpose on the basis that grounds one and two succeed and the matter is remitted for a new trial. CONCLUSION AND ORDERS 102 I would conclude that grounds one and two of the Appellant's Amended Notice of Appeal succeed. I would therefore order as follows: 1. The Appellant is successful in grounds one and two of her Amended Notice of Appeal. 2. The trial judge's decision should be set aside and a new trial ordered. 3. The Respondent should pay the Appellant's costs of appeal. ___________ DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.