Demery v Cardiff and Vale NHS Trust

J U D G M E N T

1. LORD JUSTICE PILL: This is an appeal against a judgment of His Honour Judge Hickinbottom, sitting in the Cardiff County Court on 1st February 2006, in which he dismissed a claim for damages for personal injuries made by Ms Dawn Demery ("the appellant") against the Cardiff and Vale NHS Trust ("the respondents").

2. The facts and issues are set out succinctly and with clarity by the judge in the first two paragraphs of his judgment:

"1. On 16 May 2001, the Claimant Miss Dawn Demery fell whilst ice skating, injuring her left ankle. She attended the Accident & Emergency Department of the University Hospital, Cardiff, where an X-ray revealed a fracture of the fibula. That was treated (by the application of a below the knee plaster of Paris back slab), and she was sent home with instructions to go to the fracture clinic at the Royal Glamorgan Hospital the following week. Unfortunately, as well as the broken bone, the fall damaged important ligaments in her ankle, which were not identified or treated until she attended hospital a week later. By that time, her ankle had worsened. After a further delay of 4 days whilst the swelling in the ankle subsided, she had an operation to fix a plate in the broken bone and a diastasis screw approximating the fibula and tibia. Unfortunately, a number of later medical interventions were required, and now, 4½ years after her initial accident, Miss Demery still suffers from considerable disablement as the result of pain and weakness in her ankle.

2. The Claimant makes no complaint about her medical treatment after 16 May 2001. However, in this claim, she alleges that the ruptured ligaments of her ankle ought to have been diagnosed and treated on first admission at the University Hospital that day, and the Defendant was negligent in failing to do so. The Defendant admits the failure, and that that failure was in breach of duty. However, it avers that that breach gave rise to no injury or loss to Miss Demery, in that the medical difficulties suffered by her since her initial operation (it says) would have occurred in any event and the Defendant's admitted negligence has neither caused nor contributed to them. The issue of causation was tried before me on 16-18 January 2006, and this is the reserved judgment from that hearing."

3. On the stated issue the judge concluded, in the first sentence of paragraph 27:

"For these reasons, the Claimant has failed to satisfy me that her difficulties resulting from the failure of the syndesmosis in her left ankle properly to unite was caused by the delay in operation on that ankle, i.e. caused by the admitted negligence of the Defendant."

4. The appellant lives with her partner and young children, then aged five and nine.

5. In paragraph 7 of the judgment, the judge skilfully described the anatomy of the ankle:

"It may be helpful at this stage briefly to deal with the general anatomy of the ankle, and some of its parts which particularly feature in this case. The ankle is a major weight-bearing joint. The two bones of the lower leg (the tibia and fibula) form a deep socket at their lower end, into which fits the upper surface of the talus or ankle bone. These bones have strong supporting ligaments namely the deltoid or medial ligament (a particularly strong structure on the inside aspect of the ankle), the lateral ligament (on the outside of the ankle) and an interosseous membrane between the tibia and fibula (which, at its lower end, forms a complex ligament or syndesmosis, which is important in maintaining the stability of the ankle joint)."

6. The relevant facts are set out by the judge at paragraphs 8 and 10 of his judgment, which I repeat (without the references made in the paragraphs to documents produced). The doctors mentioned are experienced consultant orthopaedic surgeons, Mr Michael Pearse, who gave evidence for the appellant, Mr Thomas Smith, and Mr Peter Evans, who gave evidence for the respondents, Mr Evans mainly on an issue which is not relevant to the determination of this appeal:

"8. Miss Demery suffered a break about one-third up the fibula (i.e. a Weber C type fracture), rather than at the level of the syndesmosis (a Weber B type fracture). Weber C type fractures are relatively uncommon - Mr Smith said that, in his busy orthopaedic practice he saw only 4-5 per year - and they are associated with damage to the ankle ligaments and dislocations of the ankle. Mr Pearse and Mr Smith were agreed that Miss Demery's initial traumatic injuries were serious; and, when she presented to the hospital on 16 May 2001, she was suffering from a rupture of the deltoid ligament and syndesmosis as well as the fracture of the fibula ... [I add in parenthesis that the evidence of Mr Pearse was that he saw one or two C type cases a month.] She did not suffer any dislocation of the ankle but the x-rays taken that day revealed that there was lateral shift of the talus, i.e. traumatic displacement of the ankle joint. The experts agreed that, with such displacement, the medical staff ought to have appreciated that the ligaments had been compromised, causing instability of the ankle - which they ought to have treated by immobilising and elevating the leg, before operating for open reduction and internal fixation. Any swelling needs to be reduced by elevation before operation (primarily to prevent difficulties with stitching the skin at the end of the operation). Dependent upon the degree of swelling on 16 May - and the availability of facilities to operate - the experts agreed that the operation ought to have taken place within 4 days of 16th May and, during the waiting period, Miss Demery ought to have been admitted to hospital with her ankle rested and elevated.

7. I interpose that there is common ground as to what treatment should be given, first, that there should be reduction of the joint so far as possible; second, that the operation should have taken place within four days; and third, that during the waiting period the appellant should have been immobilised with her leg elevated:

"10. In fact, once the broken bone had been treated, she was allowed home, where she continued to look after her children. Her ankle was not elevated and to an extent she continued to weight-bear upon it. However, as the fibula was broken and important ligaments compromised, the ankle was instable. When she attended the hospital on 22 May, she was x-rayed again. This clearly showed that her ankle joint had suffered further displacement, which in turn led to further diastasis (widening of the gap between fibula and tibia, which the compromised syndesmosis was unable to prevent) and to increased swelling of the joint (which was compounded by the failure to elevate) ... These factors interacted, in the sense that increased swelling to the joint would have caused increased talar shift. As a result of these factors, the experts are agreed that the delay consequently made the subsequent surgery more difficult ... although they are not agreed as to the extent to which such difficulties would arise."

The surgery was delayed upon the return visit on 22nd May because of increased swelling which had developed as a result of the activity between 16th and 22nd May.

8. The subsequent course of events is set out at paragraphs 11 and 12:

"11. As can be seen from the above, the orthopaedic surgeons agreed much. As indicated above (Paragraph 8), they were agreed that, when she presented to the hospital on 22 May, Miss Demery was suffering from a ruptured syndesmosis. There is no evidence that the rupture or tear was made greater by the delay in treatment. In the operation of 26 May, a diastasis screw was put in place, which approximated the tibia and fibula without undue gap, with the hope and expectation that the syndesmosis would reunite. The x-ray of 30 May (with the screw in place) shows there to be no diastasis. However, the screw was taken out on 24 July 2001, and the x-ray of 15 February 2002 shows that, without a screw to prevent it, a gap had reopened. Mr Smith accepted in cross-examination that, as contended for by Mr Pearse, on the balance of probabilities this renewed diastasis was caused by a compromised syndesmosis.

12. Between 24 July 2001 and 15 February 2002, an x-ray was taken on 14 December 2001. Whilst not entirely clear, this appears to suggest some diastasis at that date. On 15 February 2002, the treating orthopaedic surgeon (Mr Anthony Jenkins) wrote to another of Miss Demery's physicians (Dr Pierrepoint) that the bone seemed 'to have drifted out over the past two months.' There is no compelling evidence that the syndesmosis failed before towards the end of 2001. At a consultation with Mr Jenkins' Registrar on 10 August 2001, she is noted as having been 'comfortable' and advised that she could go back to work. On all of this evidence, I find that the syndesmosis substantially held together for about 5 months after the diastasis screw was removed, but then failed again necessitating the insertion of a new screw in July 2002."

9. Mr Godsmark QC for the appellant challenges the use of the word "failed" in paragraph 12. He refers to agreed medical evidence, and submits that it was not part of it that there had been healing of the syndesmosis but then subsequent failure. Counsel has referred to the evidence of Mr Smith which confirms that view. This point is not central to the issues to be considered on the hearing of this appeal. That too is common ground. Mr Godsmark refers to it as some further of support (though he does not put it in the forefront of his case) for his submission that the learned judge has in certain respects misunderstood the evidence of the expert witnesses.

10. I summarise briefly Mr Godsmark's oral submissions, to some of which it will be necessary to return on reference to the judgment. He submits that the judge has, in reaching the conclusion in the first sentence of paragraph 27, left out of account material evidence. He submits that the issue did not turn solely on a resolution of disputes between the experts. The issue was whether, given proper treatment on 26th May, the joint would have held fully. Such conflicts as there were between the medical witnesses marred, he submits, in the judge's mind, the extent of the agreement between them. He relies on the fact that it is extremely rare, following an operation such as this, for the failure which has occurred to occur. He refers to the absence of explanations acceptable to the judge to account for the failure. He refers to the obvious factors present, the delay in treatment with the consequent worsening of the injury, as stated by the judge. The gap had increased in size and the degree of subluxation had also increased. After the re-entry to hospital a good job had been done by the doctors. On the balance of probability, he submits, earlier treatment would have prevented the damage which has in fact occurred.

11. This is not a case where a poor surgical intervention has played any part in the result. The surgery was as good as it could have been, following re-admission on 22nd May. Mr Godsmark accepts that there are, if the appellant succeeds in this appeal, further issues to be resolved. There is still an issue as to the extent to which the current disability is attributable to the negligence, the respondents contending that only the mechanical not the neuropathic pain should in any event be attributed. There would also be an issue as to loss of earnings and loss of earning capacity.

12. In the agreed evidence there are a number of factors which point strongly in the appellant's favour. As stated by the judge in the first sentence of paragraph 15 of his judgment:

"First, it is clear that, in the vast majority of operations for Weber C fractures, the syndesmosis does heal and a good recovery made. The experts agreed that, usually, when the relevant surgery is performed promptly, the ligament heals and recovery is good, with 3-4 months off work being typical."

It is necessary to return to the consequences of that finding.

13. Secondly, between 16th May and 22nd May, instead of the immobilisation and elevation of the leg which should have occurred prior to a prompt operation, the appellant was at home weight-bearing and looking after her partner and two young children. Moreover, further swelling resulted from that activity and that further delayed an operation following the return visit to hospital on 22nd May. There had been no significant swelling on 16th May on first presentation.

14. Third, between 16th and 22nd May the gap between the tibia and fibula had enlarged. It had become about 1 centimetre in extent, a gap which Mr Pearse described, without contradiction, as being "huge". Moreover, the degree of subluxation had also increased, as the judge found. The injury had become a more serious one.

15. Fourth, this was one of those very rare cases where surgery did not lead to a complete recovery.

16. In their evidence both parties sought to find an explanation as to what had happened consistent with the case they put forward. The judge rejected the evidence on behalf of the appellant that the formation of fibrous tissue during the period of delay contributed to the difficulty. No challenge is made to that finding. Secondly, the judge made a finding that the operation which was performed had not been converted into a different type of operation by reason of the delay. Thirdly, the judge rejected the respondents' reliance upon literature which they claimed had supporting their case. The judge found that an abstract submitted of a paper by G Konrath and others, from the Journal of Orthopaedic Trauma 1995, 9(5), pages 377 to 380, was not of assistance to him. He gave the reason for that:

"... the study involved the comparison of patients who had prompt operations and those whose surgery was delayed; but there is no indication whether the latter group were in the meantime confined to hospital and managed there by elevation of the ankle etc. Mr Smith supposed that they were, and, if so, they would not be comparable with Miss Demery who was allowed to weight-bear on her ankle during the period of delay."

17. The judge preferred the evidence of Mr Smith on the question whether prompt reduction, which was a component of the appropriate treatment, was solely to reduce the potential for swelling and pain for the patient or whether, as Mr Pearse contended, it was to maximise the chances of successful healing. That submission was not essential to the appellant's case. Neither were the others upon which the judge made specific findings against her. I would add that the very high success rate with this operation, whatever the reason for prompt operations, may still be a reflection of the value of prompt operative treatment.

18. The judge preferred in certain respects the evidence of Mr Smith over that of Mr Pearse. He referred at paragraph 15, following the sentences I have already quoted, to the doctors' experience:

"However, albeit rarely, even without delay the ligament may not heal with there being no discernible reason for this failure. Mr Smith said that, in his 30 years experience of such fractures, he had seen 5-6 such cases. Mr Pearse said that he had not seen any cases in his practice, but he accepted that such failures were possible and Mr Smith was not challenged in respect of his personal experience. The failures of which Mr Smith had had experience were not related to delay in treatment. As he said, as with other bones and ligaments, even without delay, healing and good recovery cannot be guaranteed."

19. The judge also found, at paragraph 26, referring to Mr Smith's evidence:

"As I have indicated, his experience is that healing of ligaments is not significantly affected by a delay of the order of 6-10 days in reducing subluxation."

20. It is necessary to refer to the evidence given by Mr Pearse and Mr Smith, to decide whether those findings were justified. In his examination in chief Mr Pearse stated:

"A. Again we are dealing with a different fracture in that, unlike these, it is occurring and is becoming increasingly displaced with swelling whilst it's not properly immobilised or elevated.

Q. In this case?

A. In our case.

Q. Yes. Thank you. Now, you have demonstrated the separation on the x-rays. At p.129 of the bundle do you see Roman Numeral III?

A. Yes.

Q. 'Is it agreed that the diastasis could not have recurred if the ligaments had been soundly healed.' That was agreed between you and Mr Smith?

A. Yes.

Q. What is the significance of that, would you say?

A. If the ligaments had been soundly healed Dawn Demery would not have required as many subsequent operations as she had, would not have had as much problems as she is experiencing now."

21. Mr Pearse was properly cross-examined by Mr Davidson on behalf of the respondents:

"Q. One would normally hope or expect the syndesmosis to heal, but there will be occasions when it does not heal. You would accept that, would you not?

A. Yes.

Q. Regardless of any delay?

A. I do not have a case where the syndesmosis has not healed with prompt accurate reduction and an insertion of the diastasis screw. I do not have that experience. The only experience I have of problems with the syndesmosis are with a delayed presentation, or problems with surgical fixation.

Q. Are you saying that it does not occur?

A. In my experience. I have no experience of it. There are not many papers on it in the literature. We both looked for those papers.

Q. So you are unable to comment ----?

A. I am able to state that if it does not occur as a result of perfect treatment, it is an uncommon occurrence."

It is submitted that there should be a colon, semi-colon or gap, and that there are two propositions in that answer.

"Q. It can occur?

A. Potentially, yes, but I do not have experience of that which suggests it is quite uncommon."

And then following a question about the literature:

"Q. That is not the question, Mr Pearse?

A. You are focusing - excuse me - on the delay. My argument is that it is the delay combined with the increasing separation between the two bones either side of the ligament which will compromise the healing."

22. Mr Smith's evidence on this question, having referred to his 30 years' experience as a consultant:

"Q. In that time what experience have you had of injuries such as that suffered by Miss Demery - the original injury?

A. I must have seen, I suppose, three or four severe ankle injuries of this nature every year until fairly recently. In the last few years I have not been on call for fractures, so in the last few years I have seen fewer fractures like this.

...

Q. In your opinion, Mr Smith, had the operation [been] performed earlier, whether it be the day of the injury or four days thereafter, what is the likelihood of the syndesmosis failing to heal in any event?

A. I think there is always a risk of non-healing because nothing is guaranteed with surgery. It applies to any fixation of fractures or any operation to repair ligaments that you cannot guarantee healing.

Q. Do you have a view as to whether or not the syndesmosis would probably have failed in any event, or not?

A. I think it would, because I know of no evidence to show that a delay of surgery by six days or 10 days even would actually cause failure of healing.

...

Q. How many of these cases have you dealt with where there has been a delay of 10 days in dealing with the ligaments?

A. I would say it was very unusual to see a delay like that.

Q. It's very unusual?

A. Indeed.

Q. How can you, therefore, conclude - how many cases have you dealt with where there has been a delay of 10 days in getting those ligaments back together?

A. I think in my 30 years as a surgeon maybe four or five cases - something like that. I can't quote dates or exact figures, but a small number.

Q. In those four or five cases ----?

A. Yes, I saw no problem with healing of the diastasis. I could say, conversely, that I have seen cases in which the diastasis was treated early on and healing still failed.

JUDGE HICKINBOTTOM: In these four to five cases where there was delay, there was delay in the operation, was there?

A. There was, your Honour, yes.

Q. During the period of the delay what was happening to the patients of these cases? Were they at hospital?

A. These were patients who were in hospital because of swelling and delay - as I've said before I have no doubt the delay in most cases was because the patients were suffering from a lot of swelling by the time they were admitted to hospital.

Q. So all of these four to five patients were in hospital and were being managed in hospital?

A. Yes, I can't recall a patient who was sent home for several days."

23. On the basis of that evidence the judge's findings with respect to both doctors, cited at paragraphs 17 and 18 of the judgment, were unjustified. His finding with respect to Mr Pearse (paragraph 15) that "he had not seen any cases in his practice" was not, with respect, accurate because Mr Pearse had said that there were such cases but that:

"The only experience I have of problems with the syndesmosis are with a delayed presentation, or problems with surgical fixation."

That important question of opinion was followed up by neither party.

24. The findings with respect to Mr Smith was also, with respect, in error, in that they failed to take into account the fact that the cases to which Mr Smith had referred were cases where appropriate treatment had been given and where, during the period of delay, the patient was kept in hospital. No fair comparison was therefore possible between that evidence and the present case.

25. Moreover, the judge's finding at paragraph 26 that Mr Smith's experience is "that healing of ligaments is not significantly affected by a delay of the order of 6-10 days in reducing subluxation" was not one which could be relied on as the judge did, because of Mr Smith's fair and accurate acceptance that cases with which he had dealt where there had been delay were cases where the weight-bearing and childminding intervention which occurred in this case had not occurred, but where the patient had been kept in hospital.

26. Notwithstanding both findings of the judge, the judge went on to say that he appreciated that:

"... theoretically, there may come a time when a rupture in a ligament has been present for so long that repair becomes less likely ..."

27. On the respondents' case (as the judge understood it) it is difficult to see how, if he accepted it, the judge can have come to that conclusion. It may have been an intuitive conclusion with which one would have great sympathy, because it is, with respect, counter-intuitive to believe, certainly in circumstances such as the present, that delay in treatment can have no effect upon the outcome which occurs. But that is not the point I am making. It is simply my difficulty in resolving the judge's finding, which I take to be a finding that he would in certain circumstances accept the adverse consequences of delay, with acceptance of what he understood, albeit wrongly, to be the evidence of Mr Smith. The point is a subsidiary one.

28. It is clear that there are very few cases where a delay such as occurred in this case have in the doctors' experience occurred. That is likely to be because, if a non-negligent diagnosis is made, the appropriate treatment, on which the medical experts agree, would be followed, and the present situation is unlikely to occur.

29. The judge was entitled to say (at the end of paragraph 26 of his judgment) that the case should not turn on assumptions:

"I have accepted that evidence [that is the evidence of Mr Smith just cited] and, upon the basis of it, I accept his evidence that no assumption can be made with regard to the chances of successful healing of ligaments after 10 days merely from the fact that it is common practice to reduce subluxation promptly."

30. However, the particular facts must be considered, in the light of the evidence about the medical condition, and a decision made upon a balance of probabilities.

31. I first turn to other criticisms of the judgment. The first I mention is in relation to the size of the gap which occurred by reason of the negligent diagnosis. It was not subject to a specific or reasoned finding by the judge. It is clear that Mr Pearse did rely on that increase in expressing the opinion he did referring to the delay:

"A. ... It is my view that the delay which led to the increased gap of the fibula and the tibia has compromised the potential for healing, which necessitated further surgery and led to problems."

32. Later:

"A. You are focusing - excuse me - on the delay. My argument [said Mr Pearse] is that it is the delay combined with the increasing separation between the two bones either side of the ligament which will compromise the healing."

I have already referred to that answer in a different context.

33. The judge referred to the difference of opinion between the doctors on this point in the first part of paragraph 24. He then went on to consider (and I have cited the relevant paragraph) the question of delay, as to which there was a disagreement, and the judge gave a reasoned conclusion as to that. He did not specifically make a finding on the relevance of the increased size of the gap. It may be said that the general finding in the following paragraph was intended to cover that. However, what is clear is that there was no reasoned finding as to why the evidence of Mr Pearse was rejected on this point. There is also no finding by the judge as to the effect of the increase in subluxation which he found, in his statement of the facts, to have occurred.

34. Quite apart from that, it was in my judgment fundamental to the exercise which the judge was required to perform that he had regard to the generally successful treatment by the operation performed upon this type of injury. I cannot agree with the judge's finding at paragraph 16 that:

"The statistical rarity of such failures is of no assistance to me in itself."

35. It is in my judgment essential that the evidence of the doctors be considered against a background in which the great majority of cases have a successful outcome. If in a different situation one had a case where the experience of a particular form of surgery was 50/50 (that is only a 50% chance of a successful outcome), the onus upon a claimant to show that the case comes within the one 50% rather than the other is, or may be, a heavy one. But that is not the case here. The situation here is that in the great majority of cases, as the judge found on the agreed medical evidence, the outcome is a successful one. The judge needed to consider what evidence there was of factors present which allowed a very unusual result - that is failure - to occur in this case. Such explanations as were given he rejected.

36. Neither doctor had experience of this particular set of circumstances; that is delay which led to weight-bearing and the resulting increase in the size of the gap which was, in context, huge. The judge had to consider whether on the balance of probabilities the delay which occurred and the consequences of the delay turn what is a routine and normally successful operation into one of those very rare ones in which the outcome is unsuccessful. No other cause for the rare lack of success remains open to this court upon the judge's finding. The judge should have confronted this issue, in my judgment, and has not done so.

37. I summarise the other findings I have made. The judge has made no findings, and certainly no reasoned findings, on certain of the issues which the appellant relied on in support of her case. Further, the judge has in my judgment misunderstood, on an important point, the evidence which both Mr Pearse and Mr Smith gave. In those circumstances, action by this court is in my judgment required.

38. Lord Justice Neuberger has raised the question of whether, on findings such as those I have made, the appeal should simply be allowed. In the context which I have just summarised, the judge was, it is postulated, driven to the conclusion that the negligent diagnosis on 16th May 2001 has caused or contributed to the present disability. Mr Davidson opposes that course. He submits that there are too many open ends for the court to reach that conclusion. It is right that, even upon that conclusion, there would still be issues to be resolved.

39. The point having been raised in the course of the hearing, Mr Godsmark has accepted that, if he is successful on the resolution of the appeal, then remission is the appropriate course. He has in mind that there are other issues, including medical issues, which need to be resolved. He accepts that there are sufficient open ends in the matters on which he has relied to require that a judge considers generally the evidence in the case.

40. For the reasons I have given, I would allow the appeal and remit the case generally for consideration by a trial judge.

41. LADY JUSTICE SMITH: I agree that the appeal must be allowed, and also with regret that it must be remitted for a rehearing.

42. LORD JUSTICE NEUBERGER: I agree with Lord Justice Pill that the appeal should be allowed and the case should be remitted for rehearing. My reasoning is rather more narrowly based than his.

43. The primary reason arises from the evidence given by Mr Pearse, which has been quoted, and in particular two of his answers in cross-examination. The first is:

"The only experience I have of problems with the syndesmosis are with a delayed presentation, or problems with surgical fixation."

The second is his next-but-one answer that he did not "have experience of that", which I think must be a reference to problems arising following "perfect" treatment. There was no further cross-examination on that topic and no re-examination on it, although it was touched on at the end of his examination in chief, and briefly, at the end of his evidence, by the judge.

44. Given the dearth of evidence as to the circumstances in which failure of this operation occurs, in particular a failure of the syndesmosis to unite, it seems to me that this was evidence which needed to be taken into account; indeed it cried out for further investigation. My Lord has quoted from paragraph 15 of the judgment, which refers to Mr Pearse's evidence. It seems to me that the reference is either inaccurate or significantly deficient in that it misses out important evidence. It is not quite clear to me whether the judge's reference to Mr Pearse saying he had not seen any cases in his practice is a reference to cases where the operation fails even though it is carried out promptly ("even without delay"), or whether it is a reference to failure generally. If it is the latter, then it is inaccurate. Reading it in context, I am inclined to agree with my Lord that it is inaccurate. But even if that is not right, there would also be an important omission, which also occurs in paragraph 25 of the judgment. The omission is of any reference to Mr Pearse's evidence I have just quoted in this connection. There is also the fact that the judge said that he appreciated that "theoretically, there may come a time when a rupture in a ligament has been present for so long that repair becomes less likely". That again raises more questions than it answers and makes his decision on the point susceptible to attack.

45. Further, rather than concentrating on the likelihood of the culpable delay in treatment having rendered the operation less likely to succeed, it seems to me that the judge concentrated on the reasons given by Mr Pearse for the delay having had that effect. The fact that none of those reasons was satisfactory in the judge's eyes does not mean that the delay did not have the alleged effect. In my judgment, on that basis the decision cannot stand and must be remitted.

46. Three final points. I would not myself allow the appeal for any other reason, other than (perhaps) the fact that the judge may have misunderstood Mr Smith's evidence. Again, to a substantial extent it turns on whether in paragraph 15, when discussing the two experts' experience, the judge was referring to their experience of operations that had gone wrong, even though carried out without delay, or operations which had gone wrong only with delay. The reference to "five to six" such cases in paragraph 15 is very difficult to link confidently to any particular part of the evidence, but it may be a reference to the four to five cases identified by Mr Smith in the evidence my Lord has quoted. If that is right, then it is a misunderstanding of that evidence, because it related not to operations that went wrong for no discernible reason, but operations which were delayed.

47. I do not believe that the judge was in any way wrong to proceed as he did on the statistical evidence. He held that:

"The statistical rarity [of the operation failing] is of no assistance to me in itself."

48. It seems to me that, if there was no other reason for believing that the six to 10 days' delay was responsible for the failure of the operation in this case, the fact that the operation was normally successful and was very rarely unsuccessful, is not of itself of any assistance in resolving the issue between the parties. The appellant's case in this connection could be said to be a classic example of post hoc propter hoc reasoning.

49. I do nor argue with any of the other criticisms which have been made of the judge's reasoning, which was otherwise full and careful. One can always improve and criticise a judgment, however carefully crafted and written. It seems to me that the judge effectively accepted the evidence of Mr Smith and rejected that of Mr Pearse where they disagreed. Were it not for the points which I have dealt with at the beginning of this judgment, I, for my part, would not have held the judgment to be open to attack.

50. Secondly, it is only fair to record that the judge is by no means on his own in overlooking the importance of the evidence of Mr Pearse that I have quoted. It was not, as I have said, followed up in cross-examination or picked up in re-examination, and we were told did not feature in the submissions of counsel to the judge.

51. Thirdly, as my Lord has said, I did raise the possibility of this appeal being allowed and the judge's decision simply being reversed. Certainly, in light of the view I have formed as to the reason for allowing this appeal, that would plainly be inappropriate. Mr Godsmark QC is in my view quite right, on this basis at any rate, in accepting that the matter should go back.

ORDER: Appeal allowed and case remitted generally for consideration by a trial judge; the appellant's costs of the appeal to be paid by the respondent; costs of the trial below to be costs in the case.

(Order not part of approved judgment)

______________________________