Antonucci Or McConnell v Ayrshire & Arran Health Board [2001] ScotCS 35 (14 February 2001)

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

[1] This action of damages for alleged medical negligence concerns the treatment of the pursuer by a consultant vascular surgeon, Mr Gordon Stewart, during the period between 1989 and 1991. The defenders are sued as Mr Stewart's employers, on the basis that they are responsible for his acts and omissions in the course of his employment.

[2] The action has had an unfortunate procedural history. It was initially raised in Ayr Sheriff Court during 1993. In February 1994 it was sisted, on the pursuer's motion. Three years passed. I was told by the pursuer's counsel that during that period the pursuer's solicitors were seeking the agreement of the Scottish Legal Aid Board to have the proceedings transferred to the Court of Session, legal aid having been granted initially for proceedings in the Sheriff Court. The Board eventually gave their agreement. In January 1997 the sheriff granted the pursuer's motion to have the action remitted to the Court of Session in view of its complexity. In August 1997 the pursuer's evidence was taken on commission. Later that month a proof before answer was allowed. The length of the proof was estimated by the parties at eight days, and a diet of proof was accordingly fixed for the eight days starting on 13 October 1998. The proof then proceeded, before a Temporary Lord Ordinary, but was not completed within the allocated time. The Temporary Lord Ordinary then became unwell. In December 1998 the Inner House pronounced an interlocutor in the following terms:

The length of the continued proof was estimated by the parties at twelve days, and a continued diet of proof was accordingly fixed for the twelve days starting on 22 February 2000. The first few days of that diet were unfortunately lost, due to the fact that a criminal trial over which I was presiding, and which began earlier in February 2000, lasted much longer than expected. Although a variety of suggestions were made by the court whereby lost time might be made up, the parties were unable to accept these, by reason of counsel's other commitments. The continued diet accordingly occupied the eight days starting on 29 February 2000. Parties then estimated the further time required at four days, and a continued diet was accordingly fixed for the four days starting on 19 June 2000. The first day of that diet was again lost because of a criminal trial taking longer than expected, but the period allocated in any event proved to be far too short. A further diet was then fixed for four days starting on 7 November 2000, when the proof was completed. The preparation of the present Opinion has been delayed since then by my responsibilities in respect of criminal trials.

[3] The exceptional delays involved in this case are unfortunate from the point of view of all concerned. The pursuer, who is not in good health, has had to wait many years to learn whether or not she has a right to damages. Mr Stewart has had a grave allegation of professional negligence hanging over his head for the same period. From the court's perspective, the delay makes it unrealistic to expect witnesses to have a complete and reliable recollection of events. The change in the identity of the judge after the first eight days of evidence also gives rise to particular issues, to which I shall return.

[4] On 14 June 1989 the pursuer, who was then aged 43, went to see her doctor, Dr Calum McCabe. She complained of cramp in her left calf when walking, which stopped when she rested. After examining the pursuer, Dr McCabe advised her to walk through the pain, to stop smoking and to come back in a month's time. The pursuer returned to the surgery on 12 July 1989. She was still complaining of cramp when walking, which Dr McCabe categorised as claudication. Dr McCabe advised her that she must stop smoking. The pursuer returned to the surgery on 11 October 1989, still complaining of left claudication. She said that she could only walk from Marks and Spencers to Boots in Ayr, a distance of several hundred yards, before suffering cramp. She also said that she was now smoking menthol cigarettes. Dr McCabe decided to refer her to Mr Stewart.

[5] On 6 November 1989 Mr Stewart saw the pursuer at his clinic at Ayr County Hospital, as an out-patient. The pursuer gave a history of claudication affecting her left calf which had been present for six months and appeared to be getting worse. She could walk about 300 yards before it came on. She was smoking 25 to 30 cigarettes per day. She had a cholesterol level of 5.9, which was abnormally high. Mr Stewart examined her. His findings suggested that there might be aorta-iliac disease, i.e. vascular disease of the main arteries above the groin. He decided that the pursuer should undergo a non-invasive assessment of her arterial tree, particularly above the groin; and accordingly arranged for the pursuer to undergo a vasoscan examination. The nature of a vasoscan is described in greater detail below, but it is essentially a non-invasive means of measuring the blood pressure and blood flow at a number of points in the arterial tree and comparing the measured pressures with those which would be expected in the absence of vascular disease. He also advised the pursuer emphatically about the importance of stopping smoking, smoking being a major risk factor for the progression of vascular disease, and something which jeopardises the success of vascular surgery. On 21 November 1989 the vasoscan was carried out. Mr Stewart reviewed the results and concluded that the femoral pressures, i.e. the pressure in the femoral artery of each leg, were normal on both sides, but that there was an abnormally high drop in pressure on the left side in the superficial femoral segment, indicating significant disease in the superficial femoral artery.

[6] At this point it is perhaps desirable to interrupt the chronology in order to explain some of the terminology, and in particular that relating to the arterial tree. The aorta descends from the heart and divides into two branches, one of which feeds the right leg and the other of which feeds the left leg. These branches are, respectively, the right and left common iliac artery. On each side of the body the common iliac artery forms, as it were, the beginning of the trunk of a tree. That trunk changes its name as it descends the leg. From time to time branches come off the trunk. Some of these branches are large and others are small. Some of these branches themselves have smaller branches coming off them. The first important branch, as the common iliac artery descends, is the internal iliac artery. Distal to that branch (i.e. moving further away from heart), the main trunk is known as the external iliac artery. Above the groin, the main trunk changes its name again and is known as the common femoral artery. The next important branch is the profunda femoris artery, which branches off the main trunk at a point below the hip. Distal to that point, the main trunk is known as the superficial femoral artery. At a point above the knee, the main trunk changes its name to the popliteal artery. Below the knee, the popliteal artery divides into three branches: the anterior tibial artery, the posterior tibial artery and the peroneal artery. This division was referred to by some of the witnesses as the trifurcation. Typically however the popliteal does not divide into the three branches at a single point: normally the anterior tibial artery branches off first, and the main trunk (known as the tibio-peroneal trunk) subsequently divides into the posterior tibial artery and the peroneal artery. At the foot, the anterior tibial changes its name to the dorsalis pedis.

[7] Returning to the vasoscan of 21 November 1989, Mr Stewart concluded that the pursuer's problem was more distal than he had originally thought, and that it would be advisable to proceed to arteriography , particularly if the pursuer stopped smoking. Arteriography (also known as angiography) is an invasive form of radiological investigation, involving the injection of a radioactive dye into the bloodstream. It produces a visual image of the artery, in the form of an x-ray, known as an angiogram or arteriogram, enabling a stenosis (narrowing) or occlusion (blockage) to be visualised. Mr Stewart arranged to see the pursuer as an out-patient in order to discuss the vasoscan findings with her.

[8] On 8 January 1990 Mr Stewart saw the pursuer again as an out-patient. She was still smoking. Her symptoms had worsened. Mr Stewart arranged for her to come in for arteriography, so as to have a clearer idea of the nature and position of the lesion.

[9] On 9 February 1990 the pursuer was admitted to Ballochmyle Hospital, where Mr Stewart had a ward for in-patients. A history was taken from her by Mr Stewart's house officer, Dr Maguire. He noted that the pursuer's father and four aunts all had arteriosclerosis and died of myocardial infarctions at about 45 years of age, and that all her sisters had high cholesterol. He also noted that she was currently smoking about ten cigarettes per day. On 12 February 1990 a left trans-femoral arteriogram was carried out by a radiologist, Dr Russell. His report was in the following terms:

The adjective "proximal" is the antithesis of "distal", meaning "closer to the heart". The ability of the arteriogram to disclose the position distal to the occlusion in the superficial femoral artery - in other words, beyond a blockage - is due to collateral circulation, i.e. circulation through vessels other than the main trunk, enabling the radioactive dye injected into the bloodstream to be carried beyond the blockage in the main trunk. In particular, substantial collateral circulation can be routed via the profunda femoris, which lay downstream of the point at which the dye had been injected. Collateral circulation is something to which I shall return.

[10] The finding that the pursuer had narrow arteries was significant, in that the narrow calibre of the vessels is a further factor (on top of a history of heavy smoking, a family history of vascular disease and a high cholesterol count) associated with a poor outcome: in particular, bypass surgery is less likely to be successful in such circumstances. Mr Stewart also had the results of a blood analysis carried out the same day, which disclosed a grossly raised cholesterol count, a high level of triglycerides, a high haemoglobin count and a grossly elevated red blood count. Mr Stewart inferred from the haemoglobin and red blood cell counts that the pursuer was still smoking heavily, despite what she had told the house officer. Mr Stewart concluded that the arteriogram showed a pattern of disease which would be suitable for treatment by a femoro-popliteal bypass graft (i.e. a graft connecting the femoral artery to the popliteal, bypassing the occlusion in the superficial femoral artery), but that the pursuer would have to stop smoking before she would be considered for such surgery because of the poor results of such surgery in patients who continued to smoke. She was accordingly told once more that she must stop smoking, and Mr Stewart arranged for her to be seen again as an out-patient in two months' time.

[11] On 19 April 1990 the pursuer was seen as an out-patient. Mr Stewart being on leave, the pursuer was seen by Mr A V Krishnan, an associate specialist in the Surgical Unit at Ballochmyle Hospital who assisted Mr Stewart, and himself an experienced surgeon. Mr Krishnan found that the pursuer's condition was unchanged, and he noted that she had stopped smoking. As Mr Stewart was not due to return from leave for three weeks, Mr Krishnan arranged for the pursuer to return in one month's time to be seen by Mr Stewart.

[12] On 24 May 1990 the pursuer was seen by Mr Stewart as an out-patient. He noted that she had stopped smoking and that she was suitable for an above-knee graft (i.e. a femoro-popliteal bypass graft taken to the popliteal artery above the knee, that being the preferred site because of the greater difficulty and poorer outcome, in general, of grafts taken to below the knee). Her pressure readings were found to be very similar to before. The pursuer was put on Mr Stewart's waiting list for surgery, in the "soon" category (there being four categories of patient on the list: "immediate", "urgent", "soon" and "routine"). As a patient in that category, she would be likely to be seen in about nine to twelve months, in the absence of further developments. It may be mentioned at this point that Mr Stewart had an exceptionally long waiting list at that time, as he was the only consultant vascular surgeon in Ayrshire and was awaiting the appointment of a second consultant. In the meantime, he was having to run, single-handed, vascular wards and clinics at several hospitals and had an extremely heavy operating schedule.

[13] On 1 August 1990 Dr McCabe wrote to Mr Stewart about the pursuer in the following terms:

The pursuer was at that time being treated with co-codamol, a paracetamol-based painkiller available without prescription. She was also being treated for depression, a complaint for which she had been receiving treatment since March 1988 (and for a time in 1981).

[14] On 14 August 1990 Mr Stewart replied to Mr McCabe's letter in the following terms:

"Doppler pressures" are measurements of blood pressure obtained using a hand-held instrument. Rest or night pain are symptoms which, as discussed below, would indicate that the foot was in danger. The pursuer was then placed in the "urgent" category on the waiting list. Patients in that category would typically be admitted within a period of weeks, but the pursuer was not typical of patients in that category, in that she was not assessed at that time as requiring limb salvage surgery (or aortic aneurysm surgery). In the event, the pursuer's admission to Ballochmyle Hospital was arranged for 12 November 1990. In the meantime, she was seen by Dr McCabe and other doctors at the surgery on a number of occasions. From 13 August 1990 she was prescribed dihydrocodeine, which is a stronger painkiller than co-codamol.

[15] On 12 November 1990 the pursuer was admitted to Ballochmyle Hospital. A history was taken from her by another house officer. He noted "zero rest pain", i.e. that according to the patient there was no rest pain. The claudication distance (i.e. the distance walked before the onset of pain), as reported by the pursuer, was noted to be 15 yards. Mr Stewart saw the pursuer the same day. Arrangements were made for her to undergo a further vasoscan examination, and that was carried out on 14 November 1990. Mr Stewart saw the pursuer again on 16 November 1990. He considered that the vasoscan findings strongly suggested a progression of the disease above the inguinal ligament (i.e. above the groin), and that it would be appropriate to arrange a further arteriogram to see what was happening there. The pursuer then went home for the weekend, and returned to the hospital on 19 November 2000, when she was again seen by Mr Stewart. A femoral arteriogram was carried out by another radiologist, Dr Rawlings, on 20 November 1990. His report stated:

The finding that the catheter (with a diameter of about one millimetre) had almost occluded the external iliac artery implied that there was a very tight stenosis (i.e. narrowing) in the external iliac artery, as well as the occlusion in the superficial femoral artery which had been shown previously. Mr Stewart considered it appropriate to deal with the most proximal disease first, so as to get the best possible inflow into distal parts of the limb. He therefore decided that the pursuer should undergo angioplasty of the iliac artery. This is a procedure whereby a balloon catheter is inserted into the artery and the balloon is then inflated, causing the diseased surface of the inner lining of the artery to be compressed and broken up, and expanding the lumen (i.e. the internal diameter) of the artery to its normal diameter.

[16] On 28 November 1990 the pursuer underwent left external iliac angioplasty. The procedure was carried out by another radiologist, Mr Murch. His report is in the following terms:

Mr Stewart saw the pursuer again on 27 and 29 November 1990. Following the angioplasty, and in light of Dr Murch's report, he decided that the pursuer should undergo a repeat vasoscan examination. That vasoscan was carried out on 30 November 1990. Mr Stewart saw the pursuer on 3 December 1990. He considered that the vasoscan findings showed that the angioplasty had significantly improved the inflow into the groin, but that superficial femoral artery disease remained a problem. Mr Stewart decided that the pursuer should undergo a left phenol sympathectomy, as a way of further improving the general blood flow into the leg. That procedure was carried out on 5 December 1990. On 7 December 1990 the pursuer was noted as saying that the leg felt the same. She was again seen by Mr Stewart that day, and was allowed home for the weekend. It was arranged that her "exercise Dopplers" (i.e. pressure readings measured after exercise) would be measured on her return. The pursuer returned to the hospital on 10 December 1990, when hand-held Doppler tests were done both before and after exercise. These showed a further improvement following the phenol sympathectomy. The claudication distance was about 60 yards. Mr Stewart concluded that the procedures undertaken had significantly improved the circulation to the lower leg on the left side. The pursuer was discharged on 11 December 1990 and returned home. She had been seen again by Mr Stewart that day. The district nurse was notified of the pursuer's discharge and was requested to have a health visitor observe the condition of the pursuer's legs. The discharge summary prepared by Mr Stewart, dated 27 December 1990, stated the following under the heading "Comments":

[17] A number of contemporaneous documents, discussed below, indicate that it was Mr Stewart's intention that the pursuer should be re-admitted, as envisaged in the last paragraph just quoted, in January 1991. In the event the pursuer was not re-admitted at that time, although there is an entry in the records, discussed below, recording that the pursuer failed to keep an appointment for admission to Mr Stewart's ward on 23 January 1991.

[18] Following her discharge from Ballochmyle, the pursuer was seen by Dr McCabe or other GPs on a number of occasions. She continued to be prescribed dihydrocodeine. No change in her symptoms was recorded.

[19] On 12 March 1991 the pursuer was re-admitted to Ballochmyle. It was an arranged admission. The history taken from her by the admitting house officer, as recorded, included the statement, "Past month has also been having night pain". The claudication distance was noted to be the length of the ward, i.e. about 25 to 30 yards. This history suggested a worsening of the pursuer's symptoms. Hand-held Dopplers were taken the following day, and showed a reduction in pressure when compared with the position when the pursuer had been discharged in December 1990. Mr Stewart concluded that the situation appeared to have deteriorated, and that the pursuer should in the first instance undergo a further vasoscan. This was carried out on 18 March 1991. The findings did not appear to Mr Stewart to reveal any new problem. In view of the complaint of night pain, Mr Stewart considered that surgery was appropriate. On 22 March 1991 Mr Stewart carried out an above knee left femoro-popliteal bypass graft. His operation note contains the following description of the procedure:

Post-operatively, the circulation to the left foot showed a marked improvement for several days, indicating to Mr Stewart that the graft was functioning well. On 5 April 1991 however the pursuer complained of a return of pain in the left calf on exercise. The Dopplers were found to be greatly reduced. It was concluded that the graft had become occluded.

[20] On 15 April 1991 the pursuer was discharged home, it being intended that she should be re-admitted some weeks later for exploration of the graft and possibly a further graft. The pursuer was re-admitted to Ballochmyle Hospital on 9 May 1991. On 17 May 1991 Mr Stewart carried out an exploration of the graft and a thrombectomy, i.e. a procedure to clean out the thrombus (or clot) within the femoral artery and to improve the flow into the profunda femoris, in order to improve the peripheral circulation and stabilise the situation.

[21] On 28 May 1991 the pursuer was transferred to the Western Infirmary in Glasgow, where another surgeon, Mr Quin, performed a below-knee graft on 31 May 1991. The pursuer was transferred back to Ballochmyle Hospital on 5 June 1991, where it was found that the second graft had also failed. Following the failure of the below-knee graft, there was no alternative to amputation of the pursuer's left leg. A below-knee amputation was carried out on 13 June 1991. That procedure did not resolve the pursuer's pain, and the amputation was converted into an above-knee amputation on 5 July 1991. On 29 August 1991 the pursuer underwent an unrelated procedure, cholecystectomy, to treat gall bladder problems. Persistent pain in the amputation stump was eventually traced to a trapped nerve in the stump, and the pursuer underwent a further revision of the stump, to resolve that problem, on 3 November 1992. The pursuer's vascular disease continued to progress. In June 1993 she was admitted to hospital on account of angina. While in hospital she sustained a severe stroke which affected her speech and caused paralysis of the right side of the body. She was also known to have suffered a heart attack at some point in the past. It was disclosed by an electrocardiograph in 1991. She continues to suffer from angina. Her life expectancy is curtailed.

[22] As I have mentioned, the evidence of the pursuer was taken on commission in August 1997. Although the pursuer attended much of the proof, she was not led as a witness, and no objection was taken to the use of the evidence taken from her on commission. I was informed by the pursuer's counsel, in closing submissions, that the pursuer had considerable difficulties giving evidence, as she was severely affected by her stroke. She had difficulty in recollecting events, and had difficulty with dates. I was asked to have regard to a note prepared by the commissioner at the request of the pursuer's counsel. That note made some comments about the pursuer's apparent credibility:

In the circumstances, I accept that the pursuer's demeanour was such as to support her credibility. The reliability of her evidence, on the other hand, has to be assessed in the context of the evidence as a whole, and in particular by comparison with other evidence.

[23] A number of witnesses were led before the Temporary Lord Ordinary: Hugh McConnell (the pursuer's husband); Dr McCabe (the GP involved in the pursuer's care prior to her bypass surgery); Dr Murch (the consultant radiologist who carried out the angioplasty and arteriogram on 28 November 1990); Mr Quin (the consultant vascular surgeon at the Western Infirmary who carried out the below-knee bypass operation); Dr Derek Taylor (the GP involved in the pursuer's care since her bypass surgery); Mrs A Thomson (a payroll manager for Safeways, who were at one time the pursuer's employees); Mr Hugh Young (who gave evidence about the services required by the pursuer); and Miss Elizabeth Irvine (a consultant clinical psychologist who carried out an assessment of the pursuer). The evidence-in-chief of Mr John Drury, an expert witness for the pursuer, was also led in part before the Temporary Lord Ordinary, and completed before myself. Before myself, the pursuer also led Mr John McCormick, another expert witness, and Mrs Elizabeth Jones, a theatre sister at Ballochmyle Hospital. The defenders led Mr Stewart; Professor C V Ruckley, an expert witness; Miss Ann Bryden, a medical secretary at Ballochmyle Hospital; and Mr Robert Bryden, the Head of Health Record Services for Ayrshire and Arran Acute Hospitals NHS Trust (the successors to the defenders).

[24] I was not asked to recall witnesses led before the Temporary Lord Ordinary, and did not do so. I had available to me the notes which he had taken and I found that he had also made a few annotations on the documentary productions. None of this material revealed to me the impression which any witness had made upon him, and it has not influenced my own assessment of the evidence in any way. In these circumstances, I have proceeded on the basis that I must simply assess the evidence which I did not hear on the basis of the transcript, analysing its cogency and coherence and comparing it with the other evidence in the case. I have proceeded on the assumption that the witnesses I did not hear would have been likely to have made as good an impression on me, in terms of demeanour, as those whom I did hear. The assessment of the evidence in this case depends on what was said rather than the manner in which it was said.

[25] The next observation I should make is that the general approach adopted to the presentation of the pursuer's case had unfortunate consequences. Mr Stewart was not led as a witness for the pursuer; nor was any other direct evidence led, for the pursuer, of what Mr Stewart had actually done or why he had done it. Although Mr Stewart was cited as a witness for the pursuer, and was kept waiting outside the court one day, he was not called until the defenders' proof began. The strategy adopted, on behalf of the pursuer, was instead to have the expert witnesses, Mr Drury and Mr McCormick, give opinion evidence based on the records of the pursuer's treatment. Not surprisingly, as they had not been involved in the original events and were endeavouring to obtain a picture from the written record, they had to some extent misunderstood the position: either because they had misconstrued the records, or because the records themselves were incomplete or inaccurate. Such a misunderstanding can plainly taint, and indeed invalidate, the opinions expressed by the witness in his written reports and in his evidence-in-chief; and that happened in the present case. When a different state of facts was suggested to the witnesses in cross-examination, they were in some instances able to alter their previously-expressed opinion, on the hypothesis that what was being suggested to them was correct, and to express a revised opinion upon the correct basis. In other instances, however, the witnesses found it more difficult to adjust to the correct position in the witness box and to give a considered opinion upon it. The result was in many instances to weaken the persuasiveness of their evidence. This is not in reality the responsibility of the witnesses themselves, who are of course reputable experts in their field: it simply reflects the exceedingly difficult position in which they were placed.

[26] When Mr Stewart thereafter gave evidence, he explained that the records did not give a complete account of his actions or his thinking, and that some of the records should not be construed as the pursuer's experts had construed them. By then it was too late for the pursuer's experts to comment on the additional information provided by Mr Stewart. Instead, the pursuer's counsel attempted to prevent evidence being led of matters (including standard operating procedures) which had not been specifically recorded in the notes, and attacked the credibility of Mr Stewart's evidence insofar as it differed from the notes. He was challenged uncompromisingly and at length, on the basis that it was unbelievable that his notes should not be a complete and self-explanatory record.

[27] The danger, for a pursuer, of proceeding in this manner has been pointed out before. In Muir v Grampian Health Board, 17 March 2000, for example, Lord Abernethy observed:

Lord MacLean made the same point in Loughran v Lanarkshire Acute Hospitals NHS Trust, 6 April 2000:

[28] I respectfully agree with those observations, and would only add that a resumption of that practice might also lead to an earlier resolution of cases, since it is usually far easier for both parties to assess the prospects of success once the practitioner allegedly at fault has given his evidence. The public interest in the resolution of cases such as the present case, at the earliest possible time, is obvious: this proof not only occupied 23 days of court time, but it also occupied at least 25 working days which hospital consultants and general practitioners could otherwise have devoted to the care of patients. I would also add that, in my opinion, courts should treat with caution submissions which are made on the basis that medical records must be expected to be a complete record of events. Hospital records are not maintained by lawyers or for the use of lawyers: they are maintained for medical purposes. The courts should not in my opinion give any encouragement to the development of "defensive" record-keeping.

[29] The next observation I would make concerns opinion evidence. In this case, conflicting views on certain issues were expressed by Mr Drury and Mr McCormick on the one hand, as compared with Mr Stewart and Professor Ruckley on the other hand. In her closing submissions, counsel for the pursuer referred me almost entirely to the evidence of Mr Drury and Mr McCormick, and initially suggested no reason why I should reject the contrary opinion evidence. When I indicated that it was incumbent upon her to do so, some reasons were then suggested: principally, that I should disbelieve Mr Stewart's evidence on certain matters of fact, and therefore reject opinion evidence which was based on the assumption that Mr Stewart had given a true account. As counsel accepted, where there are conflicting bodies of evidence as to the professional acceptability of the medical treatment of a patient, with the two conflicting views being held by persons of appropriate professional standing, and each body of evidence being accepted as credible and reliable, the pursuer can succeed only if the opinion supportive of the treatment can be demonstrated to be untenable, for example because it is based on a mistaken or incomplete understanding of the relevant facts or has no logical basis (see e.g. Bolitho v City and Hackney Health Authority [1998] AC 232; and Gordon v Wilson, 1992 S.L.T. 849, at 852-853, where the earlier authorities are discussed).

[30] The final general observation which I should make about the evidence is that, although numerous objections were taken to the leading of evidence, and some of these objections resulted in evidence being allowed to be led under reservation of all questions concerning its competency and relevancy, none of the objections was renewed at the conclusion of the proof.

D THE WITNESSES
[31] Before considering the evidence in detail, it may be appropriate to make some general observations about the critical witnesses, namely those who dealt with practice in vascular surgery.

[32] The pursuer's first expert witness, Mr Drury, was a consultant general surgeon, with a specific interest in peripheral vascular surgery, at the Victoria Infirmary, Glasgow. He had been a consultant since 1987. He was well qualified to give opinion evidence, although he laboured under the disadvantage, mentioned above, that he could only comment on the hospital records and the inferences that he drew from them.

[33] It became apparent during the proof that Mr Drury's understanding of the facts of the case, on the basis of which he had prepared reports and had prepared to give evidence, was incomplete or, in some respects, incorrect. Specific examples are discussed later. That problem to some extent undermined certain of the conclusions which Mr Drury had reached.

[34] The pursuer's other expert witness, Mr McCormick, was Medical Director at Dumfries and Galloway Royal Infirmary. From 1974 to 1999 he was a consultant surgeon with a special interest in peripheral vascular disease, working initially at Dunfermline and West Fife Hospital and latterly in Dumfries and Galloway. Although it was established that Mr McCormick carried out vascular reconstructions, and femoro-popliteal bypasses in particular, much less frequently than Mr Stewart, he was nevertheless a very experienced vascular surgeon and well qualified to give opinion evidence about vascular surgery. Unfortunately, however, his ability to give as reliable evidence as the other expert witnesses in the circumstances of the present case appeared to me to be to some extent compromised by a number of factors.

[35] First, for whatever reason (and the responsibility is, of course, not necessarily Mr McCormick's), he had not prepared himself as fully as the other witnesses to give expert evidence about some of the relevant issues, and particularly by reference to the standards of 1990 rather than those of the present day. For example, an important issue in the case was the definition of critical limb ischaemia, since the presence of critical limb ischaemia would on any view require urgent surgical treatment. According to Mr McCormick's initial position, the definition in 1990 was gangrene of the toe, or a non-healing ulcer, plus an ankle pressure of less than 60mm Hg; or alternatively rest pain plus an ankle pressure of 40mm Hg. He was unable to give "chapter and verse" about that definition, "because it is not of relevance to today's practice". He had no clear memory of an International Vascular Symposium in 1981 on the definition of critical limb ischaemia, or of an editorial on its work published the following year in the British Journal of Surgery. Mr McCormick then said that there were a number of different definitions of critical limb ischaemia in 1990. He then said that a diagnosis of critical limb ischaemia was based on the patient's presentation and the information available to the surgeon, and could not be "written down in simple phrases". When asked next about the consensus document produced following a conference on the matter in 1989 in Berlin, Mr McCormick said that he had not felt that it was necessary to research earlier work, and that he had not appreciated that there was any dispute about the pursuer's critical ischaemia. The fact that there was such a dispute was plain from the Closed Record (cf. 29D-E and 32C-D), and had become very clear at the proof before Mr McCormick was led. The other expert witnesses, on the other hand (including Mr Stewart) were familiar with all this material in depth.

[36] Secondly, Mr McCormick was not entirely familiar with the history of the case. For example, in his evidence-in-chief Mr McCormick expressed the opinion that a vascular surgeon of ordinary competence would have responded to Dr McCabe's letter in August 1990 by seeing and examining the pursuer at the next available clinic. In cross-examination Mr McCormick was asked whether that opinion had been expressed on the understanding (which would be erroneous) that Mr Stewart had last seen the pursuer in February 1990. He replied that he believed it was. Asked if that understanding was wrong, he replied "I'm not sure, I haven't had time to peruse the notes to see whether there are any further entries." Shortly afterwards, he said "I did not have sufficient knowledge, in-depth knowledge of the notes at that particular time to make another comment." He then accepted that the view which he had expressed had been based, at least in part, on an erroneous understanding of the history of the case. He eventually conceded that Mr Stewart's response to Dr McCabe's letter had not been negligent. Several other misunderstandings of the history of the case could be mentioned. For example, Mr McCormick gave evidence that the pursuer had rest pain in February 1990, and referred counsel in that connection to a record that the pursuer had been diagnosed with "severe intermittent claudication left leg and early night pain". When counsel pointed out that this record was a discharge summary dated May 1991, he replied "I am grateful for your elucidation, I had just seen this." He then agreed that his earlier evidence about rest pain had been mistaken, and that (contrary to his earlier evidence) the pursuer had not had critical limb ischaemia in 1990. (As it happened, there had been an occasion in February 1990 when the pursuer had complained of pain at night, but that was pain in both legs, probably due to cramp, and was not ischaemic rest pain). Later in his evidence Mr McCormick remarked that it was "some years" since he had last looked at the pursuer's notes.

[37] Thirdly, Mr McCormick was not familiar with vasoscans, and it became apparent that he did not know how to interpret the vasoscan results. This did not prevent him from expressing views about those results during his examination-in-chief. After the defenders' counsel had explained to Mr McCormick how the results ought to be interpreted, Mr McCormick accepted that his previously expressed views were wrong, on the assumption that counsel's explanation was correct. Given that the vasoscan results had been an important basis of Mr Stewart's decisions (as I would have thought was apparent from the records), Mr McCormick's lack of a secure understanding of those results undermines some of his criticisms of the decisions which Mr Stewart took.

[38] Mr Stewart himself was principally a witness to fact, but also gave opinion evidence. He had initially worked in vascular surgery in the Royal Air Force, and had then been Registrar in Vascular Surgery at St Thomas's Hospital in London. He was Registrar there, and subsequently Senior Registrar, to a Mr Jamieson, who featured in much of the literature to which reference was made during the proof and was a vascular surgeon of international renown. Mr Stewart gained experience with other leading vascular surgeons at St Thomas's, and subsequently in Portsmouth. In 1987 he was appointed as a Consultant General Surgeon, with a special interest in vascular surgery, in Ayrshire. Part of his remit there was to develop a vascular service for Ayrshire. At that time very little vascular surgery was carried out in Ayrshire: such patients as were recognised by GPs as suffering from vascular disease were normally referred to hospitals in Glasgow, although some emergency cases were treated locally. Mr Stewart had to develop a vascular service virtually from scratch. This involved teaching the local GPs about vascular disease; teaching nursing staff about the management of vascular patients; and teaching the patients themselves. He gave tutorials to the nurses, ran a series of talks with GPs throughout the whole of Ayrshire, and spoke at hospital meetings in the three hospitals for which he had vascular responsibility. It also involved the introduction of new clinical techniques and equipment: there were no modern instruments in the theatre complex, and there were no grafts available for operating. Mr Stewart introduced to the hospitals where he worked (Ballochmyle, Ayr County and Crosshouse) new techniques, such as the concept of using Doppler ultrasound to measure ankle pressures, a pro-active treatment of heel care, and new methods of record-keeping specific to vascular records. He set up a High Dependency Unit in the vascular ward. He set up a graft bank. He introduced new equipment with which he was familiar from his previous experience, such as hand-held Dopplers, a Duplex scanner and a vaso scanner. The vasoscan was a non-invasive diagnostic tool which used Doppler ultrasound to measure blood pressures in segments of the arterial tree in the leg and simultaneously measured blood flow through each segment. It became apparent during the proof that this equipment, which Mr Stewart had used in England and which was also popular in the United States of America, was less well known in Scotland: Mr McCormick, as I have mentioned, had never encountered it, and even Professor Ruckley had not, he said, had the resources to instal it in his own unit in Edinburgh. Mr Stewart also introduced new methods of theatre management, such as all-day operating lists. Predictably, perhaps, Mr Steward rapidly became the victim of his own success, as a large waiting list of patients built up. It was apparent from the evidence that, by the time of the pursuer's treatment, the waiting list was longer than was desirable. Mr Stewart was at that time awaiting a decision by the defenders on his recommendation that a second consultant vascular surgeon be appointed.

[39] Until 1991 Mr Stewart remained the only consultant vascular surgeon in Ayrshire. A second surgeon was appointed in 1991, and a third in 1997. Both the National Medical Advisory Committee, in a report in about 1990, and the Scottish Office Acute Services Review in 1998, recognised the position achieved by the Vascular Unit in Ayrshire. The latter review recommended that, for reasons of quality and resources, the number of centres from which vascular services should be provided in Scotland should be reduced from 18 to 7, comprising the four main teaching centres and three district general hospitals. One of the three district units recommended for retention was the one in Ayr which Mr Stewart had developed from nothing during his time there, reflecting the view (as Professor Ruckley was able to explain) that Mr Stewart's unit provided a high quality service.

[40] In 1990 Mr Stewart carried out around 90 major vascular reconstructions, of which about 24 were femoro-popliteal grafts. The corresponding figures for 2000 were around 140 and 30, respectively. As is obvious from the foregoing, Mr Stewart is a very experienced (and dedicated) vascular surgeon.

[41] Mr Stewart made a strong impression on me while giving evidence. First, he was subjected to robust cross-examination over three days. He had earlier been kept waiting outside the court during the pursuer's proof. He had had allegations of negligence hanging over his head since at least 1992, when the pursuer's solicitors wrote to him. Despite these circumstances, Mr Stewart was patient and courteous throughout his evidence, and showed no sign of stress. Secondly, Mr Stewart's evidence, which lasted four days in all, consisted almost entirely of a detailed and rigorous examination of complex matters. Mr Stewart gave careful and considered answers throughout, and was plainly capable of sustained concentration. Thirdly, I was left with the clear impression of a caring and, above all, careful doctor who took his heavy responsibilities with all the seriousness that they require. These qualities are hardly surprising to find in someone in Mr Stewart's position, but they are nonetheless relevant, as will appear, when considering some of the suggestions made expressly or impliedly on behalf of the pursuer: in particular, that Mr Stewart had treated the pursuer carelessly as a consequence of the pressure and stress which he was under at that time, and that he had subsequently given an inaccurate account of his treatment of her.

[42] As I have indicated, I was impressed by the coherence of Mr Stewart's evidence. He was able to provide a cogent explanation of the reasoning behind each of his decisions. I was equally impressed by the care and precision with which he gave his evidence. Mr Stewart also had an advantage over all the other surgeons who gave evidence (as Mr Drury acknowledged), in that he was the only surgeon who saw and treated the pursuer at every material stage. As the other witnesses accepted, he was in the best position to describe the pursuer's condition at the time, insofar as he had a recollection of his treatment of her; whereas all the other witnesses could only draw inferences from the records. In short, I found Mr Stewart's evidence compelling. He was also, it seemed to me, less susceptible than some of the other expert witnesses to being led or coaxed by counsel (something which could be detected from time to time, for example during the evidence of Mr Drury, e.g. at Vol.6, pages 982 to 984, which can be compared with Vol.5, page 888 and Vol.9, pages 1431 to 1446).

[43] The defenders' other expert witness, Professor Ruckley, had the unique advantage, among the expert witnesses, of giving evidence after having listened to Mr Stewart's evidence. Professor Ruckley is a vascular surgeon of great eminence. He was a consultant at the Western General Hospital, and subsequently the Royal Infirmary, in Edinburgh, from 1971 until his retirement from NHS practice in 1999. He has held a chair in vascular surgery at the University of Edinburgh since 1992. He was awarded the CBE for services to vascular surgery in 1998. One of Professor Ruckley's particular interests is vascular audit, i.e. the assessment and monitoring of standards of care in vascular surgery; and he was chairman of the Scottish Vascular Audit for a four year period in the 1990s. This experience made Professor Ruckley particularly well qualified to give opinion evidence about standards of care and management in respect of vascular surgery in Scotland. Professor Ruckley's publications include, as joint author, an article (the "Darke" study, as it was referred to in evidence) entitled "Femoro-popliteal versus femoro-distal bypass grafting for limb salvage in patients with an 'isolated' popliteal segment", to which repeated reference was made during the proof by all the expert witnesses. Another of his publications as joint author, the European Consensus Document on critical limb ischaemia, also featured extensively in the evidence of all the expert witnesses. His other publications include work on above-knee femoro-popliteal bypass grafts, on angioplasty for lower limb ischaemia, and on the quality of life of patients with claudication treated by angioplasty as compared with patients treated by arterial surgery. Professor Ruckley was also head of the vascular component of the Acute Services Review set up by the Scottish Office. In short, it is impossible to imagine anyone, not involved in the pursuer's treatment, better qualified to give expert opinion evidence in the present case.

[44] Professor Ruckley was criticised by the pursuer's counsel for having failed to append to his report a certificate of his understanding of the duties of expert witnesses. It was suggested that this was an appropriate practice. When I observed that the witness's taking the oath was good enough for me, counsel's response was that we were beyond that. I nevertheless view Professor Ruckley's failure to issue himself with such a certificate with indifference. His evidence made it clear that he was perfectly aware of his responsibilities as a witness.

[45] Professor Ruckley's evidence was based on a sound understanding of the facts of the pursuer's case and an exceptional understanding of vascular disease and surgical practice. He also had wide practical experience, and his evidence impressed me as being given with a clear-eyed and realistic appreciation of the practicalities of the situation facing Mr Stewart (and, for that matter, the pursuer). He gave his evidence with care and precision, to the extent that the more thoroughly he was cross-examined, the more compelling his evidence became.

[46] As I mentioned above, I am not entitled to "prefer" one body of credible and reliable opinion evidence to another, in a case of this kind, unless there is some demonstrable flaw in the rejected evidence. In the present case I accept the evidence of Mr Stewart and Professor Ruckley as credible and reliable. I have not found any reason to reject it. I therefore have no basis for "preferring" the contrary evidence of Mr Drury and Mr McCormick. I can however go further. As I have already indicated (and will explain further when considering the evidence in detail), I have reservations about the reliability of some of the evidence given by Mr Drury and, more especially, Mr McCormick. Insofar, however, as I accept the testimony of Mr Drury and Mr McCormick as credible and reliable, in the limited sense that it expressed opinions which were carefully considered, honestly held and accurately represented, I nevertheless have come to the conclusion that those opinions, so far as conflicting with those of Mr Stewart and Professor Ruckley, are flawed. My reasons for coming to that conclusion will appear more clearly when I come to discuss the evidence.

[47] By way of background, counsel for the pursuer emphasised that the events with which this case is concerned occurred at a time when Mr Stewart was extremely busy. He had to cover more than one hospital, and the work of his unit was undergoing a considerable increase at that time. He was the only consultant vascular surgeon in the unit. That is perfectly true. It does not however follow that Mr Stewart was careless or forgetful. As I have mentioned, he appeared to me (admittedly in the different context of a court room) to be someone who coped well under pressure and was careful and cautious. I do not however have to rely solely on my own impression. I refer later to the evidence of witnesses who worked with Mr Stewart at the time.

[48] There are a number of other matters which appear to me to be more significant in forming the context in which the pursuer's treatment has to be considered. First, it became abundantly clear during the evidence that vascular surgery was at the material time, and still is, a developing sphere of specialist skill. There was and remains considerable debate amongst vascular surgeons regarding both the diagnosis of the stage to which arterial disease has progressed and, thereafter, the appropriate treatment. The former point can be illustrated by the debate over the definition of critical limb ischaemia. The latter point can be illustrated by the fact that all the expert witnesses agreed that it would not have been negligent for Mr Stewart to have amputated the pursuer's left leg in November or December 1990 or March 1991: some vascular surgeons would have done exactly that, in accordance with accepted standards of treatment. As that fact dramatically illustrates, vascular surgery involves difficult and uncertain questions: it is a sphere where judgment is frequently called for and where it is often impossible to say that there is a single "correct" answer.

[49] Secondly, it is important to appreciate the remorseless nature of the pursuer's disease, and the manner in which it progresses. She suffered (and continues to suffer) from athero-sclerotic obliterative arterial disease (AOAD). The disease is incurable. The objective of treatment is the alleviation of symptoms. The pathological process, as explained by Professor Ruckley, involves the gradual deposition of a fatty material known as atheroma on the inner surface, and fibrotic change (sclerosis) within the wall, of arteries throughout the body. The artery affected by AOAD gradually narrows until it finally blocks by the clotting of blood (thrombosis) at the narrowest point. A clot is soft when it first forms, but solidifies (by a process known as organisation) over subsequent days. Patients with a serious prognosis include those with early onset disease and women with arteries of narrow calibre: the pursuer belongs to both those categories. Although AOAD is a progressive disease characterised by deterioration over a number of years, that progression is not a process of gradual decline. Rather, it is a stepwise progression, with points of acute deterioration, as new episodes of thrombosis cause sudden occlusion of new sections in the arterial tree. Each episode (assuming that it is not fatal) is followed by a period of stability, or even improvement, sometimes lasting for years, as the body compensates in various ways, for example by making better use of collateral vessels to carry blood past the blockage. As far as a limb is concerned, collateral channels may sustain viability, but they cannot convey normal pulsatile flow, and therefore do not provide normal function. When AOAD affects the legs, the early symptom is claudication. The leg symptoms of the majority of patients with claudication remain stable or improve over a long period of years; and such patients, although limited in their walking distance, can lead a reasonable life. In a small minority of patients the disease progresses rapidly. In contrast to claudication, critical ischaemia is serious and has a short course. The limb will normally be lost within days or weeks, in the absence of intervention to improve the circulation. The intolerable nature of the pain and the inability of even the strongest analgesics to control it, together with the danger of life-threatening gangrene, mean that action has to be taken quickly. Possible interventions include angioplasty and bypass surgery. Both have significant complication rates and limited success. They may fail for a wide variety of reasons. There is an early failure rate of 10 to 15 per cent, and less than 50 per cent of femoro-popliteal bypass grafts are patent after 5 years. When angioplasty or a graft fails, the outcome is usually amputation. Although some angioplasties and bypasses can be re-done when they fail, there is a progressively lower success rate with re-intervention.

[50] There was therefore never any possibility of the pursuer being "cured". Intervention would not prevent the further progression of the disease. The purpose of intervention was essentially to relieve pain (for a time at least) and to delay the period before she lost her leg.

[51] Thirdly, the progressive nature of the disease, together with the limited success and potential hazard of intervention, has implications for the management of patients. Experienced vascular surgeons are very reluctant to intervene in patients with claudication. An intervention which fails can result in the loss of a leg - a disastrous outcome, since claudication does not itself threaten either life or limb. Even greater caution applies in patients with known risk factors, which increase the odds against a favourable outcome. The known risk factors for AOAD include a family history of AOAD, smoking, hypertension and high cholesterol in the blood. On the information before Mr Stewart, the pursuer had all of these. In the great majority of cases, therefore, the vascular surgeon does not intervene unless the patient reaches the phase of the disease which threatens life and limb, namely critical ischaemia.

[52] Mr Stewart indeed explained that his approach towards the pursuer's treatment had been influenced throughout by the fact that she had so many factors placing her at high risk of progression of the disease (as manifested not only in peripheral vascular disease but also in heart attacks and strokes): onset of the disease at an early age; a woman with small arteries; a smoker; high cholesterol; and hypertension.

[53] A variety of grounds of fault are alleged against Mr Stewart in the pursuer's pleadings, but these came to be refined, and in part departed from, during the proof and in closing submissions, in the light of the evidence.

[54] The first ground of fault averred on Record concerns Mr Stewart's actions following his receiving Dr McCabe's letter of 1 August 1990:

This ground of fault was supported by both the pursuer's expert witnesses, Mr Drury and Mr McCormick, in their evidence-in-chief; but both witnesses abandoned that position under cross-examination. That ground of fault was then departed from by the pursuer's counsel.

[55] A further ground of fault averred on Record concerns the fact that the pursuer did not receive a below-knee bypass graft during her stay in hospital between 12 November and 11 December 1990 or by the end of December 1990:

[56] A further ground of fault averred on Record concerns the fact that the pursuer did not receive a bypass graft prior to March 1991:

A further ground of fault averred concerns the selection of an above-knee rather than below-knee graft:

[57] Certain other grounds of fault are averred on Record, but they were not supported by any evidence, and they were not made the subject of any submission.

[58] The three grounds of fault finally relied upon were aptly summarised by the pursuer's counsel as follows:

[59] In the following discussion, it seems to me that it may be useful (albeit rather unusual) to give a more detailed account of the position adopted on behalf of the pursuer in closing submissions, before narrating a summary of the evidence relevant to those submissions. I do so essentially because the quantity and complexity of the evidence is such that any narrative is perhaps best related to the specific submissions made.

[60] I turn now to consider in greater detail the first ground of fault alleged against Mr Stewart, namely that he was negligent in failing to carry out a below-knee bypass graft in December 1990.

[61] In support of this ground of fault, counsel for the pursuer made a number of points.

[62] The first important question of fact, so far as the present ground of fault is concerned, is whether the pursuer had critical limb ischaemia in November or December 1990. The pursuer's submissions are all either directed towards establishing that critical limb ischaemia existed, or take the existence of critical limb ischaemia as their starting point.

[63] "Critical" limb ischaemia means ischaemia which has reached the stage at which it endangers the leg or part of the leg. The definition of critical limb ischaemia in terms of measurable, objective, clinical criteria was much discussed in evidence. I have already mentioned a criterion which was given, at one stage in their evidence, by both Mr Drury and Professor Ruckley: an ankle pressure of 50mm Hg or an ABI or less than 0.4. I should say at once that, applying that criterion, there is no evidence that the pursuer was at the stage of critical limb ischaemia during the period in question, and a body of evidence demonstrating the contrary. According to the pursuer's case notes, whose accuracy in this respect was not disputed, the pursuer's ankle pressure was measured and recorded on three occasions during the period in question. On 14 November 1990 it was 58mm Hg; on 30 November (after the angioplasty had taken place) it was 62mm Hg; and on 10 December (after the sympathectomy, and the day before the pursuer's discharge home) it was 100mm Hg. The ABI was 0.44 on 14 November, 0.53 on 30 November and 0.71 on 10 December.

[64] There is, however, and was at the material time, more than one definition of critical limb ischaemia. Mr Stewart explained the definition which he was using, and the way in which it had evolved, as follows. A definition put forward in the early 1980s, following an international vascular symposium held in London in 1981, was of persistent recurring rest pain of greater than two weeks duration, requiring adequate analgesia, with an absolute ankle pressure of 40mm Hg or less; or the presence of ulcers or gangrene of the foot, with an absolute ankle pressure of 60mm Hg (I observe in parenthesis that the pursuer would not have met the requirements of that definition, since the only stage at which her ankle pressure was recorded as being below 60mm Hg was on 14 November, when she certainly had no ulcers or gangrene on her foot). Mr Stewart explained that the definition was further discussed during the 1980s and led to a meeting in Berlin in 1989, known as a "Consensus" meeting, at which leading vascular surgeons tried to arrive at an agreed definition. That meeting resulted in the publication of a document, known as the First Consensus Document, which put forward a definition of critical limb ischaemia in terms of persisting rest pain of greater than two weeks duration, requiring regular analgesia, with an ankle systolic pressure of 50mm Hg or less; or ulceration or gangrene with a pressure of 50mm Hg or less. I observe that the pursuer did not meet the requirements of that definition during the period in question, since her ankle pressure was never as low as 50mm Hg. Mr Stewart explained that discussion continued following the issue of the First Consensus Document, and resulted in some minor revisal of its proposed definition. The revised definition was published in 1992 in a Second Consensus Document. It was in substance the same as the 1989 definition: persistently recurring ischaemic rest pain needing regular adequate analgesia for more than two weeks, with an ankle systolic pressure of less than or equal to 50mm Hg and/or a toe systolic pressure less than or equal to 30mm Hg; or alternatively ulceration or gangrene of the foot or toes, with an ankle systolic pressure less than or equal to 50mm Hg or a toe systolic pressure less than or equal to 30mm Hg. I observe that the pursuer again did not meet the defined criteria, since her ankle pressure was never as low as 50mm Hg (there was no evidence as to her toe pressure). The essential requirements of the Consensus definition (as it was referred to in the evidence) are thus the same in both the initial version of 1989 and the slightly revised version of 1992: rest pain or ulcer/gangrene with an associated ankle pressure of 50mm Hg or less. The expert witnesses, including Mr Stewart, did not draw any distinction between the 1989 and 1992 versions; and Mr Stewart said that he used the Consensus definition, without making any such distinction. All the experts agreed that the Consensus definition (in either formulation) was a reasonable definition of critical limb ischaemia for a consultant vascular surgeon to be using in 1990 or 1991. As Professor Ruckley explained matters, the 1989 document was of the nature of a consultative draft, and the 1992 document was the finalised document, giving approval to what was already a recognised definition.

[65] For the sake of completeness, I should mention that Mr McCormick's definition of critical limb ischaemia was gangrene of the toe or an ulcer which was not healing, together with a systolic ankle pressure of less than 60mm Hg; or alternatively rest pain and a pressure of the order of 40mm Hg. That definition is the same as the one which Mr Stewart described as having been published in the early 1980s, following the London conference. As I have already mentioned, the pursuer did not meet the requirements of that definition during the period in question. Mr McCormick's use of this definition was one of several indications during the proof that his approach to diagnosis and treatment may have been that of a slightly older generation of surgeons than the generation to which Mr Stewart belongs.

[66] As I have mentioned, one essential element of the Consensus definition - which also constituted one of the "working" definitions mentioned by Mr Drury and Professor Ruckley - was an ankle pressure of 50mm Hg or less; and all the evidence indicates that the pursuer did not meet that criterion. In order to try to establish that the pursuer had critical limb ischaemia at the relevant time, and ought to have been so diagnosed by Mr Stewart, the pursuer's counsel therefore focused upon the other aspects mentioned in the Consensus definition - rest pain or ulceration or gangrene. These elements, even if proved to exist, would not, in the absence of an ankle pressure of 50mm Hg or less, establish critical limb ischaemia within the meaning of the Consensus definition, which was agreed to be a reasonable definition for Mr Stewart to use in treating the pursuer. Nevertheless, because it was accepted by all the experts that the surgeon had to have regard to the patient's clinical condition as a whole and not only to blood pressure measurements, the pursuer's case focused on these other aspects, which offered somewhat greater scope for debate: for example over the meaning and existence of "rest pain", and over the possible symptoms of actual or impending ulceration or gangrene. I shall therefore consider each of these matters in turn.

[67] The expert witnesses differed to some extent in their definition of rest pain. Both Mr Stewart and Professor Ruckley described it as an excruciating pain which required the most powerful painkillers - morphine or heroin - to control it. Mr Drury described it as one of the most excruciating pains known, like having one's foot on hot coals and being unable to take it away. Mr McCormick on the other hand also spoke of "mild" rest pain which could be controlled by much less powerful painkillers. What all the experts appeared to be agreed upon was that the pain was continuous and very unpleasant. The evidence of Mr Stewart and Professor Ruckley that rest pain results in sleep disturbance (unless adequately controlled by analgesia) would also appear to be common ground, since Mr McCormick described "night pain" as an early indication of rest pain. He explained that blood pressure generally dropped when the patient was falling asleep, causing the pain to start and the patient to awake with unpleasant discomfort.

[68] Mr Drury did not suggest that the pursuer had established rest pain during the period in question, but he did interpret the records as suggesting that the pursuer's pain was not being controlled, and regarded that as one of a number of symptoms that the pursuer's condition was deteriorating (a position which he however qualified under cross-examination, as I discuss later). This supposedly uncontrollable pain during December 1990 indicated critical limb ischaemia, in Mr Drury's opinion, and therefore an urgent necessity for surgery if the limb was to be saved. Under cross-examination, however, Mr Drury withdrew his opinion, as I explain later. At another point in his evidence Mr Drury said that he did not believe Mr Stewart's record in the discharge summary, that the pursuer "certainly seemed to be more comfortable and was not having pain at rest". Mr Drury asserted that the pursuer could not have been more comfortable if she had no apparent symptomatic relief, as Mr Stewart had also recorded in the previous sentence. I do not myself find it difficult to understand that Mr Stewart could have formed the view that the pursuer was in fact more comfortable although she reported no improvement: unlike Mr Drury, Mr Stewart actually saw the pursuer regularly during this period. Mr Drury also referred to a nursing note dated 29 November 1990, recording that the pursuer had required a painkiller in the middle of the night, as indicating that the pursuer was not having her pain controlled adequately. That entry, however, concerned the night following the angioplasty, a procedure which involves a wound to the groin and is liable to result in post-operative pain. Mr Drury had indeed explained earlier that the night dosage of analgesia that day would make sense, since the angioplasty would have been done under anaesthetic, which would then wear off. In short, there is nothing in Mr Drury's evidence which leads me to infer that the pursuer had rest pain, or critical limb ischaemia, during the period in question.

[69] Mr McCormick stated that dihydrocodeine, at the dosage prescribed for the pursuer on her admission to Ballochmyle in November 1990, would be prescribed for mild to moderate rest pain, and not for claudication. The higher dosage prescribed at the time of the pursuer's discharge in December 1990 would indicate that the rest pain was becoming severe. The analgesic effect of the dihydrocodeine could in his opinion also be increased (or "potentiated") by amitriptyline, which the pursuer was also taking for her depression. Although he had earlier described ischaemic rest pain as "one of the worst types of pain....relentless, very unpleasant pain", he did not agree that ischaemic rest pain could only be controlled by morphine derivatives: "mild" ischaemic rest pain did not in his opinion require morphine. This evidence is inconsistent with that of Professor Ruckley (and also that of Mr Stewart), which I have no reason to reject. In cross-examination, Mr McCormick accepted that the pursuer's prescription of dihydrocodeine was a moderate dose of a moderate painkiller. Eventually, in cross-examination, he agreed that the pursuer did not have rest pain by the time she was discharged in December 1990. In view of that evidence, I do not feel able to place reliance upon his earlier evidence to the opposite effect, or his subsequent evidence in re-examination, when he reverted to his original position.

[70] Neither Mr Drury nor Mr McCormick carried out a detailed analysis of the pursuer's analgesia requirements, as recorded in the hospital notes. Such an analysis was however provided by Mr Stewart. It was revealing. First, the strongest analgesia prescribed during the admission in November and December 1990 was dihydrocodeine. The dosage increased significantly for a time following the angioplasty, and again for a time following the sympathectomy. According to Mr Stewart, both those procedures can result in post-operative pain. The pursuer's analgesic requirements did not otherwise change greatly during that admission: the dosage of dihydrocodeine increased, but it remained, in Mr Stewart's words, "a moderate dose of a moderate painkiller" (a description which Mr Drury also accepted). The only occasions on which analgesia was provided during the night were on 29 November (following the angioplasty) and on 6 December (following the sympathectomy). Far from supporting the opinion that the pursuer had critical limb ischaemia, these records in fact tend to suggest the opposite, especially as the pursuer did not require analgesia at night except in the aftermath of specific procedures.

[71] The pursuer's counsel, and also Mr McCormick in his evidence, founded strongly on the fact that in a letter dated 8 September 1992 Mr Stewart had, in the course of summarising the pursuer's medical history, stated:

I note that the letter in question was not a clinical record and was written long after 1990. It was concerned principally with more recent events. As Mr Stewart explained, it was sent to an insurance company from which the pursuer was claiming benefit under a permanent health insurance policy arranged by her former employers, Safeway. This letter forms part of the Ayr Hospital records, Mr Stewart having a clinic at Ayr Hospital where he treated the pursuer from time to time. The records of the pursuer's admission to Ballochmyle Hospital in November and December 1990 form part of the Ballochmyle Hospital records, and are not duplicated in the Ayr Hospital file. Mr Stewart's evidence was that the pursuer did not have rest pain in November 1990. When this letter was put to him in cross-examination, however, he said:

The question of what Mr Stewart meant by "early" rest pain, and how it differed from the rest pain forming one of the elements of critical limb ischaemia, was not explored. From the answer just quoted, however, and also the reference in the letter itself to "severe intermittent claudication", it is apparent that, whatever Mr Stewart meant, he did not mean that the pursuer had reached the stage of critical limb ischaemia. More fundamentally, however, and despite Mr Stewart's reluctance to depart from what he had written in the letter when strongly pressed in cross-examination, it seems to me that the emphasis placed on this letter by the pursuer's counsel was disproportionate. The letter was written long after November 1990. There is no evidence that Mr Stewart wrote it on the basis of a consideration of the 1990 medical records. Those records expressly record (in the admission note and in the discharge summary) that the pursuer did not have rest pain in November or December 1990; and that is consistent with all the other recorded information about, for example, ankle pressure, ABIs, walking distances and analgesia.

[72] Professor Ruckley's evidence on this topic was that rest pain, as a diagnostic feature of critical limb ischaemia, required morphine or heroin to control it: it was an excruciating pain. Dihydrocodeine was the next step up from aspirin. Having checked the pharmaceutical industry's Data Compendium, there was no suggestion that the effect of dihydrocodeine was potentiated by amitriptyline. I have no reason to reject this evidence.

[73] In the whole circumstances, I am entirely satisfied that the pursuer did not at the material time have rest pain as that expression is used in the context of the definition of critical limb ischaemia.

[74] Considerable reliance was placed by the pursuer's counsel, and indeed by Mr Drury and Mr McCormick, on the supposed development and deterioration of a pressure sore on the pursuer's heel. This was suggested to indicate actual, or at least imminent, ulceration and gangrene. It was therefore, the argument ran, evidence of critical limb ischaemia, and consequently of an urgent need for surgery.

"Safe environment" is a nursing term indicating that the entry is based on observation and assessment of the patient. Granuflex is a protective dressing applied to protect the heel and to spread pressure on the heel. An entry for 7 December 1990 recorded:

"Communication" is another nursing term, indicating that the entry is based on what the patient has told the nurse. A pressure area information chart contained the following entry, also dated 7 December 1990:

[75] Mr Drury interpreted the entry for 2 December as suggesting that the skin on the heel was on the point of breaking down. He presumed, incorrectly as it turned out, that the phenol sympathectomy performed on 5 December was intended to counter this problem by improving the flow of blood to the skin. He interpreted the entry for 7 December as meaning that the skin and the underlying tissues were dying: the phenol sympathectomy had therefore failed to alleviate the problem, which had become significantly worse. According to Mr Drury, by 10 December there was already localised gangrene forming in the heel. In cross-examination, Mr Drury confirmed that if the problem with the left heel had been necrosis or ulceration or critical ischaemia, he would have expected it to have got worse after the pursuer's discharge. He would have expected her to have symptoms requiring her to contact her GP. He had not seen the GP records for several years. When these were put to him, he confirmed that they contained no entry, subsequent to the pursuer's discharge, concerning her heel. When the March 1991 hospital records were put to Mr Drury, he accepted that they also were consistent with the heel not being a source of concern. In these circumstances, Mr Drury's earlier opinion that there was necrosis or gangrene in the heel cannot be correct.

[76] Mr McCormick similarly regarded the entry for 7 December as suggestive of necrosis. He also had not seen the GP records for some years. He also accepted in cross-examination that there was no record of any problem with the heel subsequent to the pursuer's discharge in December 1990, up until her re-admission to hospital in March 1991. If the pursuer had been discharged with the necrosis or ulceration associated with critical limb ischaemia, he would have expected the condition to her heel to have deteriorated. He considered that the explanation might be that the angioplasty or phenol sympathectomy had assisted in the healing of any problem with the heel. He also accepted that he did not know what the nature of the lesion on the heel had been, whereas Mr Stewart had had the opportunity to examine the pursuer at the time. He eventually accepted that it would be reasonable to conclude that the pursuer did not have necrosis or ulceration or gangrene. Although, in re-examination, Mr McCormick to some extent reverted to his position in examination-in-chief, and maintained that as at 11 December there would still have been a pressure sore which was unlikely to heal and should have been viewed as a matter of grave concern, that opinion appears to me to lack a reasonable foundation. I am in any event not prepared to place reliance on it, given his own evidence in cross-examination and the evidence of Mr Stewart and Professor Ruckley, which I see no reason to reject.

[77] Mr Stewart, in his evidence, explained that one of the steps he had taken after his appointment to Ballochmyle was to introduce a heel care management system: in other words, the nursing staff were trained to be alert to heel problems and to respond to them promptly. As he explained, it was well recognised in vascular circles that a problem with the skin on the heel could often lead to amputation. Hospital beds being hard, it was necessary to maintain a clear understanding, on the part of nursing staff, of the importance of heel care. The nursing staff on his ward, where the pursuer was being treated, would have understood the importance of heel care. It was one of the issues which was discussed during his ward rounds. The patient was also made aware of the importance of heel care, and was told to bring any concern about the heels to the attention of the nursing staff. Various steps were taken to avoid pressure on the heel, including the use of sheepskin anklets on patients who were at risk or had early signs of problems. The pursuer, in her evidence, spoke of having worn such an anklet. A problem with the heel having been noted on 2 December 1990, the nursing care record contained the following entry, with a start date of 3 December 1990:

Mr Stewart explained that the purpose of that entry was to inform nurses looking after the pursuer that they had to check her heel or consider it when they were dealing with her. The nursing record of care given contained the letter B on 3, 6 and 7 December, meaning that the nursing staff had on those days observed or done something to the heel (other entries recorded "self care", implying that no particular issues had arisen with regard to the heel). The entry in the nursing notes for 7 December was headed "Communication", meaning that it was a record of what the patient had said to the nurse, rather than what the nurse had observed. In the entry, the word "said" meant "the patient said". The nurse who had made that entry had also made the entry in the pressure area information chart on the same date, recording that the heel was dark and inflamed and that Granuflex had been applied. On the same date, the nursing care plan was altered to identify that there was a problem with the left heel and to record the action to be taken:

The remaining entries in the nursing notes, prior to the pursuer's discharge, made no mention of the condition or appearance of her left heel. There was no record, at any stage, of any report of any further deterioration in the left heel. When the pursuer was discharged home on 11 December 1990, the district nurse referral form was completed by Staff Nurse Crearie, a capable staff nurse experienced in the care of vascular patients. She had written nothing in the form about the heel. The purpose of the form was to inform the community nurse about areas requiring to be monitored or treated. Staff Nurse Crearie had also made an entry, dated 11 December 1990, on the nursing care plan, immediately below the entry dated 7 December concerning the pursuer's heel: she must therefore have been aware of the earlier treatment of the heel. Mr Stewart had also personally seen the pursuer on 3, 7 and 11 December (7 December in particular being the date when the pursuer complained about her heel). In these circumstances, Mr Stewart inferred that there could not have been a continuing problem with the heel on 11 December: there appeared to have been some early pressure changes in the heel which were picked up and managed and which then disappeared. This appears to me to be a reasonable inference. It is fortified by the subsequent records concerning the pursuer's visits to her GP. As Dr McCabe said, under reference to the records, the pursuer contacted the practice on several occasions between her discharge from hospital on 11 December 1990 and her re-admission on 12 March 1991: on 12 December, 21 December, 1 February, 12 February and 11 March. Various complaints were recorded, but none concerning her heel. Moreover, when the pursuer was re-admitted to Ballochmyle on 12 March 1991, the history given by the pursuer to the house officer contained no reference to her heel. This is significant in view of the pursuer's having in December 1990 brought the heel problem to the attention of the nursing staff. The notes made by the house officer also indicate that he examined the pursuer's feet: he commented that the feet were cool and that the tips of the first and second toes of the left foot were pale. He also assessed the pulses at ankle level. The nursing admission document of the same date states, in terms, "skin appears intact". That document would have been completed by nursing staff who were aware of the importance of looking at the pursuer's feet, and who would have been aware (from the same document) that the pursuer had a vascular problem affecting her left leg. That document also contains a history taken from the pursuer by Staff Nurse Campbell, in which a number of concerns and problems are described. The history does not contain any mention of the heel. A pressure area information chart was completed the same day by the same staff nurse, and did not identify any problem with any pressure area. That chart would not have been completed without an assessment of the pursuer for any problem with a pressure area. Finally, the nursing notes for the period from 12 March until 22 March (when the bypass was performed) contain no reference to any problem with the pursuer's left heel.

[78] In Professor Ruckley's opinion the argument that the "dark and inflamed" heel was an indication for immediate surgery was shown by subsequent events to be incorrect. There was no evidence that this lesion ever amounted to a pressure sore or ulcer, i.e. an actual break in the skin. The GP records suggested that it was not a problem following discharge, and no such lesion was recorded when the pursuer was re-admitted in March 1991. All the available evidence therefore pointed to the discolouration in the heel having resolved in the interim. Had it been an area of actual necrosis (i.e. gangrene) it would not have resolved. Neither would it have resolved if the limb had been critically ischaemic. Indeed the favourable progress of this lesion suggested that the angioplasty and sympathectomy had a beneficial effect on the circulation of the foot, and confirmed that the pursuer did not have critical limb ischaemia at the time of discharge on 11 December 1990.

[79] I accept the evidence of Mr Stewart and Professor Ruckley. There appears to have been a short period in December 1990 when there were some minor early pressure changes in the heel, which were treated by the nursing staff. They appear to have been transient. There is nothing to suggest that they progressed to anything significant: on the contrary, the only reasonable inference that can be drawn from the records, considered as a whole, is that the problem resolved itself completely. There is no reasonable basis for the suggestion that the pursuer suffered skin necrosis or indeed any problem with her heel which would indicate a threat to the limb.

[80] I have thought it best to deal in turn with each of the major topics discussed in evidence in relation to the first ground of fault. Having done so, I now turn to consider the criticisms made of the pursuer's treatment at the material time, by Mr Drury and Mr McCormick, and the responses of Mr Stewart and Professor Ruckley. This involves some repetition, insofar as opinions were expressed about the treatment on the basis that there was, or was not, critical limb ischaemia, rest pain or ulceration/gangrene. Having already covered those issues, however, it will be more readily apparent why certain opinions appear to me to have firmer foundations, in the facts of the case, than others.

[81] In examination-in-chief, Mr Drury initially defined critical limb ischaemia in terms of tissue loss or skin necrosis together with an ankle pressure of 50mm Hg, or an ABI of less than 0.4. He noted that pressure measurements had been taken by vasoscan on 14 November 1990. The ankle pressure was 58mm Hg and the ABI was 0.44. By his initial definition, critical limb ischaemia therefore did not exist at that time, although there was what he described as "severe" ischaemia. He found no fault with the decisions to carry out arteriography, angioplasty and phenol sympathectomy, but said that the "rider" to that opinion was "the delay involved in the lady who was, in vascular terms, becoming much more severely incapacitated". The delay in question was that from discharge in December 1990 to the bypass operation in March 1991. Having seen that there was no improvement following the angioplasty and phenol sympathectomy, a competent vascular surgeon would in his opinion have had two treatment options in December 1990: to amputate the leg; or to try further surgical measures to save the leg, since it was critically ischaemic. One such measure would have been a bypass graft. Such a surgeon would have sent the pursuer home in December 1990, for surgery at a later date, only if there was evidence that the leg was surviving and a degree of stability about her symptoms. The records however suggested that her pain was not being controlled, she was not mobile and she had a deteriorating pressure sore, all of which suggested that her symptoms were deteriorating rather than stable. The problem with the pursuer's heel, in particular, was "the beginnings of tissue destruction" and "localised gangrene". It would be acceptable to send her home to see whether there was a benefit from the treatment already carried out, but only for a matter of days: within the limits of critical limb ischaemia the time window is very short.

[82] Mr Drury's criticisms of Mr Stewart's treatment of the pursuer, in relation to this chapter of the case, are therefore based on the following propositions: there was no improvement following the angioplasty and phenol sympathectomy; the heel problem showed that the leg was dying; and the overall clinical picture - the lack of improvement, the heel problem and the degree of pain - showed that the leg was critically ischaemic. I will deal in a moment with the issue of whether there was an improvement in the pursuer's condition following the angioplasty and phenol sympathectomy, although I have already noted the improvement in the recorded ankle pressure and ABI. So far as the other matters are concerned, for the reasons I have already explained I am satisfied that Mr Drury had misunderstood, and exaggerated, the seriousness of the heel problem: he had not, in particular, considered the significance in this context of the GP records and the hospital notes relating to the period subsequent to the pursuer's discharge on 11 December 1990. When they were put to him, he accepted that they were consistent with the heel not being a source of concern, and were not what he would have expected if the problem had been as serious as he had earlier thought. I am also satisfied, for the reasons already explained, that the pursuer's pain was not of the level of severity associated with critical ischaemia, but was consistent with a less severe degree of ischaemia causing claudication. Mr Drury had in particular attached undue significance to an entry recording pain at night, without apparently appreciating that it followed immediately upon a procedure (the angioplasty) which left a surgical wound and bruising in the groin, and that there was an absence of any evidence in the records of pain at night in November or December 1990 except in the immediate aftermath of invasive treatment. I am also satisfied, for reasons already explained, that the pursuer did not meet the criteria of critical limb ischaemia as defined in the Consensus definition at any time in 1990 (as Mr Drury indeed acknowledged), and that that was a reasonable definition for Mr Stewart to use, and act upon, in his treatment of the pursuer. Mr Drury's diagnosis of critical limb ischaemia, on the basis of a more general assessment of the pursuer's clinical condition, is in any event flawed by his misunderstanding of the position as regards the heel and as regards pain (and also, as I will explain in a moment, his misunderstanding of the pursuer's supposed deterioration following the angioplasty and sympathectomy).

[83] So far as the issue of improvement or deterioration is concerned, in cross-examination Mr Drury accepted that post-angioplasty Dopplers were noted on 7 December 1990 to be "better"; that Dopplers recorded on 10 December 1990 showed an AB1 of 0.71 (as compared with 0.44 in the pre-angioplasty vasoscan), and that the claudication distance was recorded then as 60 yards (compared with a reported 15 yards pre-angioplasty). He agreed that there had been an improvement in the ABI, and also in the absolute pressure: for example, the dorsalis pedis and posterior tibial pressures were recorded in the pre-angioplasty vasoscan as each being 58mm Hg but as being 90 and 100mm Hg respectively in the post-angioplasty Dopplers. He agreed that the readings showed a significant improvement in the patient's condition following the procedures undertaken. Given that concession, the foundation of his earlier opinion disappears. After being taken through all the records concerning the pursuer's heel, Mr Drury also accepted that when the pursuer was discharged from hospital on 11 December 1990 she did not have critical limb ischaemia as defined in the Consensus Document. Crucially, Mr Drury further accepted in terms that it was reasonable, and not negligent, for Mr Stewart to consider at the time of the pursuer's discharge that a bypass was not an urgent necessity.

[84] Mr McCormick's evidence added nothing of significance so far as this chapter of the case is concerned. He had provided his report, and had prepared to give evidence, on the assumption that the pursuer had critical limb ischaemia at all material times: his criticisms of Mr Stewart's treatment of the pursuer were based on that assumption. He had not addressed his mind to the question whether the pursuer in fact had critical limb ischaemia, or as to how critical limb ischaemia was defined at that time. Insofar as he attempted to support this assumption by reference to his recollection of the records, his recollection proved to be less than wholly reliable. He accepted, eventually, that the pursuer did not have rest pain in November 1990, and that that was a strong pointer to her not having critical limb ischaemia as defined in the Consensus definition. Indeed, the definition of critical limb ischaemia which he had himself put forward in his examination-in-chief - gangrene of the toe or an ulcer which was not healing together with a systolic ankle pressure of less than 60mm Hg; or alternatively rest pain and a pressure of the order of 40mm Hg - was itself not satisfied in November or December 1990, since he accepted that it would be reasonable to conclude that there was no gangrene or ulceration, and he also accepted that there was no rest pain. He further accepted that the Consensus definition of critical limb ischaemia was an acceptable definition. He subsequently conceded that, following the angioplasty and the phenol sympathectomy, it would be reasonable for a consultant surgeon to have taken the view that the pursuer's condition had improved, and that she did not have critical limb ischaemia as at 11 December 1990.

[85] Mr Stewart explained in his evidence that he saw the pursuer on the day of her admission and discussed with her the deterioration of her situation. In view of the progressive nature of the pursuer's disease, the first step he took was to re-assess the state of her disease, initially by a vasoscan, since arteriography carries a risk of damage to the artery and a risk of death. The vasoscan (of 14 November 1990) suggested progression of the disease in the iliac artery. He then decided to arrange an arteriogram to see what was happening there. The arteriogram (of 20 November 1990) confirmed that there was a very tight narrowing in the external iliac artery. He decided that the next step was to deal with that stenosis by angioplasty, in accordance with the standard practice of dealing with the most proximal disease first. The angioplasty was carried out by Dr Murch on 28 November 1990, and Dr Murch reported that it effected some improvement. Mr Stewart then decided to repeat the vasoscan, in order to check that the angioplasty was working and to see how it was affecting the peripheral circulation. That vasoscan (of 30 November 1990) confirmed that the angioplasty had significantly improved the inflow into the groin (where the femoral artery pressure was back to normal) and had also improved the ankle pressure (where the absolute pressure had risen since the 4 November vasoscan from 58 to 72mm Hg, and the ABI from 0.44 to 0.53). There remained disease in the superficial femoral artery, and possibly an occlusion in the popliteal artery. He considered it possible that the pursuer might require a bypass graft, but given the risks involved (particularly in view of the narrow nature of her vessels) he did not want to undertake that procedure without trying other options first. He therefore decided to try phenol sympathectomy in the first instance, as it was a recognised means of increasing the blood flow into a limb following a proximal intervention. The phenol sympathectomy was done on 5 December 1990. Hand-held Dopplers were done on 10 December, and showed an ABI of 0.71: a significant improvement not only from admission (0.44) but also from the position following the angioplasty (0.53). The absolute pressure was also up: it had been 58mm Hg on 14 November and 72mm Hg on 30 November (after the angioplasty) and was now 100mm Hg. The measures undertaken had therefore significantly improved the circulation in the left leg. The pursuer (who had, as I have mentioned, a history of depression) nevertheless reported that the leg felt much the same. Mr Stewart therefore had her walking distance objectively checked, and found a significant improvement over the distance reported on admission (60 yards as compared with 15 yards). The pursuer was discharged the following day. In his discharge summary, Mr Stewart noted that the angioplasty and sympathectomy had been carried out "without any apparent symptomatic relief", recording the patient's perception; but, he added, "Having said this she certainly seemed to be more comfortable and was not having pain at rest." As I have already mentioned, Mr Stewart had seen the pursuer frequently during her stay in hospital, both before and after the angioplasty and sympathectomy. He had seen her on the day she was discharged. He was in no doubt that she did not have critical limb ischaemia on discharge. Considering each element of the Consensus definition, she did not have persistent recurring ischaemic rest pain which required regular analgesic treatment for more than two weeks: on the contrary, she did not have rest pain at all. She did not have ulceration or gangrene of the foot or toes. The ankle systolic pressure was not 50mm Hg or less: at 100mm Hg, it was double that level.

[86] Asked in cross-examination why he had not carried out a bypass in December 1990, Mr Stewart explained that he was uncertain about the accuracy of Dr Murch's report suggesting that there was a popliteal occlusion, partly because the vasoscan suggested otherwise. That is an issue to which I return below. He knew however that if Dr Murch was correct, then any graft would probably have to be below-knee rather than above-knee, with a consequent reduction in the prospects of success. He was aware that the measures already undertaken had resulted in objective signs of improvement in the patient. He therefore decided to wait and see whether a bypass would prove to be necessary. The measures taken during that admission had resulted in a degree of improvement; his judgment was that there was no requirement for further intervention at that point, and that it was appropriate to allow a period of time to see whether there was further improvement.

[87] Mr Stewart was questioned about the fact that the house officer's admission note was headed "Arranged admission for fem-pop bypass graft". Mr Stewart explained that the house officer would have obtained that information from the admission list. It did not imply that the intention was that the pursuer should necessarily undergo such an operation. I accept that explanation.

[88] The pursuer's counsel, and Mr McCormick, also placed some emphasis on the fact that in the discharge summary dated 27 December 1990 Mr Stewart had recorded, under the heading "Primary Diagnosis", the words "Critical ischaemic left foot". This was interpreted, mistakenly as it turned out, as Mr Stewart's diagnosis of the pursuer's clinical status when discharged in December 1990. In evidence, however, Mr Stewart described the "Primary Diagnosis" as the admitting diagnosis; but it was a diagnosis assigned to the case retrospectively, following the patient's discharge - "the diagnosis, having reviewed the notes, that I felt was appropriate for her when she came into hospital". The pursuer did not quite fit the definition of critical limb ischaemia on admission but was heading in that direction. Mr Stewart considered that he had an option of putting down either severe claudication or critical limb ischaemia, and he put down the latter because the pursuer had a progressive disease which was heading in that direction. He was quite clear that the pursuer did not have critical limb ischaemia when she was discharged. Mr Stewart also referred to "coding", but that matter was not explored with him.

[89] Professor Ruckley, when asked about this matter, described the "Primary Diagnosis" as a coding label and said that he had seen no evidence which would justify such a clinical diagnosis. Explaining the expression "coding label", Professor Ruckley said that all discharge documents generated in Scottish hospitals were coded according to national coding numbers and were then entered into national databases for the purposes of a national audit exercise. The coding requirements necessitated that each case be given a diagnostic label - such as "claudication" or "critical ischaemia" - describing what the patient had been brought in for. That label was then entered into the database along with other codes representing the procedures carried out to treat the condition for which the patient had been admitted. Since the pursuer was, on admission, at the very serious end of the claudication scale, into a rather indeterminate zone between advanced claudication and the development of critical ischaemia, it was reasonable to use "critical ischaemia" as a diagnostic label for audit purposes. The treatment of the patient, on the other hand, was not determined by diagnostic labels, but by the patient's clinical status. A diagnostic label, in other words, was not a clinical diagnosis.

[90] I accept Professor Ruckley's evidence that there was no basis for diagnosing critical limb ischaemia, as defined in the Consensus document, in November 1990. I am disinclined to attach much significance to an ex post facto label given to the pursuer's case for statistical purposes. Given that the pursuer's condition was not very far short of critical limb ischaemia on the date of admission, and that (in the absence of appropriate treatment) it was progressing fairly rapidly in that direction, Mr Stewart's use of that "label" for audit purposes was understandable. In any event, even if one accepted that the pursuer had been at the stage of critical limb ischaemia when admitted in November 1990, it is plain that her condition thereafter improved, so that she was not at the stage of critical limb ischaemia when discharged in December 1990.

[91] Professor Ruckley also of course gave more general evidence about the acceptability of Mr Stewart's treatment of the pursuer. Mr Stewart's cautious approach was strongly supported by Professor Ruckley in his evidence. The admission note indicated that a femoro-popliteal bypass had been in contemplation when the pursuer was admitted on 12 November. Some surgeons might have proceeded directly to carry out such a bypass. That would have been a mistake. It was judicious for Mr Stewart to re-assess the patient. The vasoscan findings suggested significant disease at iliac level, which was subsequently confirmed by the angiogram of 20 November. Thus the pursuer's disease had progressed, and now required intervention at a different site. The vasoscan findings on 14 November (with an ankle pressure of 58mm Hg), as well as the absence of rest pain, showed that the pursuer had not reached the stage of critical limb ischaemia. It would not have been appropriate to have performed a femoro-popliteal bypass at this stage, since the narrowing at iliac level compromised the inflow of blood and therefore the likely success of a more distal bypass. It was important to correct the iliac narrowing first, and this was done by angioplasty. Phenol sympathectomy was added as a possible means of enhancing blood flow through the widened iliac artery and into the leg. Very few vascular surgeons in 1990, in Professor Ruckley's opinion, would have disagreed with these interventions. A judgment then had to be made as to what to do next. Although the pursuer did not report any improvement in symptoms, there was objective evidence of improvement in the circulation in the limb in that her walking distance had increased to 60 yards and her resting Doppler pressure had increased to 100mm Hg - a level 50mm above the critical limb ischaemia level, and the sort of value normally associated with not very severe claudication. She was still taking only medium range analgesia and did not apparently require morphine or heroin - an indication that she did not have rest pain. She did not therefore need immediate surgery. It was unclear at this time whether the improvement would be continued or whether it would be reflected in symptomatic relief. A short interlude before contemplating any further arterial reconstruction was a wise precaution at this stage for several reasons. The symptomatic benefits of angioplasty and sympathectomy, as experienced by the patient, may take many days or weeks to accrue. There was a possibility that the pursuer might derive enough benefit from the angioplasty and sympathectomy that bypass could be avoided or postponed. Since bypass surgery carries significant complications, and its failure can precipitate the loss of the leg, experienced surgeons in these circumstances are anxious to do the minimum to achieve relief of symptoms. It was proper for Mr Stewart to be particularly cautious about operating on the pursuer when she was at the stage of claudication rather than critical limb ischaemia, since she was at high risk of a poor outcome. Professor Ruckley agreed entirely with Mr Stewart's approach. I accept Professor Ruckley's evidence.

[92] For the reasons which I have explained in the course of discussing the evidence, I reject the evidence critical of Mr Stewart's treatment of the pursuer in November and December 1990, and reject the first ground of fault. Returning to the particular points relied on by the pursuer and summarised in para.61 above, I conclude that the pursuer did not have critical limb ischaemia in November or December 1990, despite what Mr Stewart wrote in the Discharge Summary; she did not have rest pain in November 1990, despite what Mr Stewart wrote in the letter of 8 September 1992 for the insurance company; she did not, in November and December 1990, have any clinical features indicative of critical limb ischaemia; the lesion on her heel, in particular, did not imply critical limb ischaemia; and the implications of critical limb ischaemia therefore do not arise. In reality, the pursuer's first ground of fault is an edifice erected on exceedingly slender foundations. It has been built up from misunderstood entries in the medical records, without a proper or complete understanding of the context. Indeed, even if one confines one's attention to the records, I am quite satisfied that these records, read as a whole, do not reveal the kind of picture, of acute and deteriorating ischaemia, reaching the stage even of incipient gangrene, contended for by the pursuer's witnesses.

[93] I consider next the second ground of fault alleged against Mr Stewart, namely that he was negligent in failing to carry out a below-knee bypass graft prior to March 1991.

[94] In support of this ground of fault, counsel for the pursuer relied upon a number of points.

[95] I shall summarise the evidence relating to the treatment the pursuer received before considering the evidence commenting on the standard of that treatment.

[96] Dr McCabe, the general practitioner, gave evidence that the pursuer had not required treatment with morphine until after the bypass had blocked in March 1991. He had seen the pursuer on 12 December 1990 and had written in the notes, "For surgery in New Year". He would use that expression to mean more or less after the holiday period had finished in early January 1991. He suspected that he had been told that by the pursuer. Either the pursuer had told him, or he had been aware from some other source, that it was planned to re-admit the pursuer at that time. If the pursuer had been given an appointment for 23 January, that would fit with his note. The pursuer had been at the practice on a number of occasions during December 1990 and January, February and March 1991. The records contained no suggestion that her leg was becoming any more painful.

[97] The hospital's internal copy of the discharge summary prepared by Mr Stewart and dated 27 December 1990 bore the postscript:

That postscript was absent from copies of the discharge summary sent outside the hospital, such as the copy sent to Dr McCabe. Mr Stewart explained that he had dictated the discharge summary. What was in his mind was that the pursuer should be re-admitted for further assessment and probable surgery. Depending on the findings, she would require either a right iliac to left femoral procedure or a left femoro-popliteal bypass graft. The former procedure would be appropriate if the iliac angioplasty carried out in November 1990 was not working well: in that event, it would be necessary to improve the inflow into the leg, and that could be achieved by means of a graft from the right iliac artery into the left femoral artery. If, on the other hand, the angioplasty was working, then the problem in the femoro-popliteal segment could then be addressed, by means of a femoro-popliteal bypass graft. The postscript was a waiting list note. It was a note given by Mr Stewart to his secretary, probably by dictation, informing her that he wanted to re-admit the pursuer in January 1991, and requesting her to remind him. It had either been dictated by Mr Stewart along with the summary, or it had been added when he came to sign the summary. Mr Stewart had earlier explained how his waiting list, with its various categories, was maintained by his secretary. He would discuss admissions and the waiting list with her on a regular basis, and take decisions as to when patients could be fitted into his operating schedule and should be admitted. The patient would then be sent an admission letter by Mr Stewart's secretary, specifying the date, time and place of admission. Early in January 1991 he and his secretary would have sat down and looked at the programme for the month of January, and given the pursuer an appointment. The note on the discharge summary would have been a reminder that the pursuer was to be admitted then. Mr Stewart said that he would be very surprised if the pursuer had not been sent an admission letter. Her failure to attend might be attributable to a subjective improvement in her condition, as well as the objective improvement already seen, having regard to the GP records. This possibility, incidentally, was also accepted by the pursuer's husband, although he could not recollect the pursuer missing an appointment in January 1991.

[98] The Ballochmyle records contain, near the beginning of the file, a document headed "Record of Attendances". It records, in the first column, a series of dates - 9 February 1990, 19 April 1990, 24 May 1990, 12 November 1990, 23 January 1991, 12 March 1991, and subsequent dates. These dates - with the exception of 23 January 1991 - are the dates on which the pursuer was seen, by appointment, at Ballochmyle Hospital. The second column records the ward and in most cases the consultant: Ward 3 (Mr Stewart's ward for female patients) and Mr Stewart. In the line for 23 January 1991, the words "Wd 3", and dittos indicating that the consultant was again Mr Stewart, have been entered. Above them, in a different ink, have been entered the letters "DNA".

[99] Mr Stewart explained that, when an admission had been arranged, the medical records office would send the patient's case notes up to the ward a few days before the admission. If the patient failed to attend, the senior nurse or sister on the ward would ask the medical records office to retrieve the notes, and the letters "DNA" would be written in the notes to record the fact that the patient had not arrived. Those letters, standing for "Did Not Attend", were a standard entry.

[100] The Ballochmyle records do not contain any copy of an admission letter for 23 January 1991, but neither do they contain a copy of any other admission letter. Mr Stewart explained that it was not normal practice to put a copy of any admission letter into the case notes. The entry for 23 January 1991 on the Record of Attendances would not have been made unless there was correspondence indicating that the pursuer was due to be admitted that day. The pursuer having failed to attend that day, she would have been given another appointment, bearing in mind the other admissions already arranged, the resources available and the relative degrees of urgency of the various patients as they appeared at the time. In the event, she was given an appointment for 12 March 1991.

[101] Evidence was also given by Mr Stewart's secretary in 1990-1991, Anne Brydon. She had typed the discharge summary. It had been dictated by Mr Stewart. She described the procedure followed at that time. If Mr Stewart wanted to put a patient on the waiting list, he would dictate at the end of the document what she had to do. After she had typed the document, she would then type his instructions at the foot of the copy of the document which was to go into the case notes, and she would put the patient on the waiting list. The Ballochmyle copy of the discharge summary in question had at the foot the typed instructions about the waiting list. The waiting list was computerised. In order to carry out Mr Stewart's instruction, she would access the computer, put the patient's details on the waiting list which Mr Stewart had instructed (i.e. "urgent", "soon" etc.), and add details of the procedure which the patient was to have and any matters of which Mr Stewart had asked to be reminded. A waiting list printout would then be run off the computer, giving a hard copy of the waiting list and the other details entered into the computer. She would provide Mr Stewart with a copy of the printout. He would then mark the admission dates on the printout, and return it to her. She then had all the information needed before contacting the patient about her admission. She then used the computer to generate two further documents: an admission list, containing details of all patients to be admitted and the date of admission; and an admission letter, to be sent to the patient. She then sent the admission letter to the patient. No copy of that letter was kept with the patient's notes. The admission list was photocopied, and copies were sent to the ward and the medical records office. A copy was also sellotaped into Mr Stewart's diary. The medical records office would then send the case notes to the ward in time for the admission date. If the patient failed to attend, a "DNA" entry would be made, and Mr Stewart would be informed. He would then consider what action to take. A tick on the discharge summary was Miss Brydon's sign that she had put the patient on the waiting list as instructed. Such a tick was present on the discharge summary in the pursuer's case.

[102] Evidence was also led from Robert Brydon, who had been responsible for medical records in Ayrshire hospitals since 1989 and was familiar with the record-keeping system at Ballochmyle Hospital in 1990 and 1991. He confirmed the system of having a copy of the admission list sent by the secretary to the records office, and the records office then sending the case notes to the ward prior to the admission date. The "DNA" entry in the notes meant that the patient did not arrive for admission on that date.

[103] I turn next to the opinion evidence commenting on the acceptability of the pursuer's treatment. As I have already mentioned, Mr Drury was of the opinion (at one point in his evidence) that the pursuer should have been seen again within a matter of days of being discharged on 11 December 1990. That opinion rested on the assumption that there was critical limb ischaemia, as evidenced by uncontrolled pain and a deteriorating pressure sore. I reject those assumptions, and therefore cannot accept Mr Drury's opinion on this point. As I have mentioned, he himself withdrew that opinion under cross-examination.

[104] Mr McCormick expressed the opinion that the pursuer should have been brought back within two or three weeks of the (supposed) pressure sore being identified in December 1990: a delay until 23 January would have been too long. Since I reject the assumption that there was a pressure sore, I am unable to accept Mr McCormick's opinion on this point.

[105] Professor Ruckley accepted that an ordinarily competent vascular surgeon, having discharged the pursuer in December 1990, would have wished to review her within a few weeks. A period of three to four weeks would have been appropriate. If, when re-assessed, the pursuer was found to have established rest pain, then it would be appropriate to intervene either by amputation or by performing a bypass. Professor Ruckley did not however accept that the pursuer had in fact had established rest pain within a few weeks of her discharge in December 1990: rest pain had not in his opinion emerged until late February 1991.

[106] Professor Ruckley considered that the pursuer had reached the state of critical limb ischaemia some time in late February 1991, on the basis that when admitted on 12 March 1991 she gave a history that for the past month she had been having night pain. At the same time, Professor Ruckley acknowledged that even as late as March 1991 some of the clinical findings were inconsistent with critical limb ischaemia as defined in the Consensus document: the ankle pressure was measured on 18 March 1991 as 81mm Hg, well above the level of 50mm Hg mentioned in the Consensus definition; and her walking distance was measured the same day as 120 yards, which would be surprising for someone with critical limb ischaemia. I note that Professor Ruckley explained that the latest Consensus definition of critical limb ischaemia, published in 2000, uses rest pain (or ulceration or gangrene) as the only definition for clinical purposes, without reference to ankle pressures. That may have influenced his opinion that critical limb ischaemia existed in February or March 1991, despite the fact that the earlier Consensus definition was not satisfied: Professor Ruckley may well have been expressing his opinion as to the time at which the pursuer reached the stage of critical limb ischaemia, not the time at which any competent surgeon would in 1990 or 1991 have diagnosed the pursuer as having reached the stage of critical limb ischaemia. Be that as it may, one implication of Professor Ruckley's evidence is that the pursuer had not reached the stage of critical limb ischaemia on 23 January 1991. That conclusion is also supported by Dr McCabe's evidence that the GP records of the pursuer's various complaints around that time did not include any indication that her leg pain was getting worse. In these circumstances, even if the pursuer had been admitted to hospital on 23 January 1991, it does not follow that she would have undergone a bypass graft any sooner than in fact she did.

[107] My conclusions on this chapter of the case, and on the points relied on by the pursuer's counsel, can be summarised as follows.

[108] More generally, in relation to both the first and second grounds of fault, my conclusion is that the pursuer did not have critical limb ischaemia as defined by the Consensus document (which represented a responsible body of opinion of consultant surgeons) before February, or more likely March, 1991 at the earliest. That is because there was no rest pain until mid-February (and therefore no established rest pain until about early March), the ankle pressure readings were well above 50mm Hg, and there was no necrosis or ulceration. Since the pursuer did not have critical limb ischaemia, it was at least reasonable for Mr Stewart to take the view, at all stages prior to March 1991, that surgical intervention was not justified. Since it is only the development of critical limb ischaemia, as defined by the Consensus document, that would have mandated surgical intervention - either by amputation or by grafting - according to such a responsible body of opinion, it follows that Mr Stewart cannot be held to have been negligent for not having performed a bypass graft before March 1991.

[109] I consider next the third and final ground of fault alleged against Mr Stewart, namely that he was negligent in failing to carry out a below-knee graft, rather than an above-knee graft, on 22 March 1991.

[111] Dr Murch gave evidence about the angiogram he carried out on 28 November 1990. He had no recollection of what the film showed: he was reliant upon the terms of his report. His report has been quoted in para.16 above. In evidence, he said that as the popliteal artery crossed the knee joint, he spotted a short occluded segment. He had no recollection of its dimensions (the angiogram - i.e. the x-ray photograph - had gone missing during the course of the action, but prior to the proof, at a point when the only witnesses to have seen it were Dr Murch, Mr Stewart and Mr Drury), He had no doubt in his mind that that occlusion was evident on the angiogram. It was not possible that there was no occlusion. He accepted that sometimes an angiogram showed the presence of an occlusion which turned out not to be there - an artefactual occlusion - but in his experience it would be very unusual to have a very short occluded segment that was not in fact present. He accepted that the region of the knee was partly obscured by bones, but at the knee joint the bones stopped, and the layer of cartilage between them was transparent to x-ray. He had total confidence that the arteriogram demonstrated an occluded vessel. In his opinion, a graft would therefore best be placed below the knee: that was, as he put it, "my view, such as it is, as a radiologist", or "for what it's worth, I suppose, my opinion". Surgery was however not his field. It was a surgical decision, and it would not be appropriate for him, as a radiologist, to indicate what had to happen in a surgical procedure. He was perfectly prepared to be overruled by surgeons if they had other information and other reasons for not doing as he had suggested. His view that the graft would have to go below the knee, if there was an occlusion at the knee joint, might be mistaken: that was a matter lying within the field of expertise of a surgeon. If no occlusion was found by the surgeon when he went into the artery, one possibility was that the occlusion had cleared spontaneously: that would however be unusual. Another possibility, if the occlusion was caused by clot adhering to the vessel, was that the passage of a catheter down the artery by the surgeon might have dislodged the clot, so that the vessel was re-opened. That was one possible explanation of the fact that, in an arteriogram performed on 23 May 1991, only two calf vessels had been seen, whereas three had been seen in November 1990: the catheter might have poked the clot down one of the calf arteries. Dr Murch acknowledged however that it was difficult to express a confident view, as the disease was a dynamic process, and the November 1990 angiogram and the March 1991 bypass surgery had been several months apart. The dislodging of clot was one of several possible explanations for the fact that the graft worked very well (for a time), something which suggested that there was no occlusion beyond the distal anastomosis (i.e. the lower point at which the graft was plugged in), despite the fact that the November arteriogram had shown an occlusion. This theory, that the catheter had dislodged a clot, was described by Mr Murch as "speculative" and "hypothetical". If clot was removed, the atheroma on the wall of the artery would remain, and might have attached more clot and blocked up again. In relation to the calf vessels, Dr Murch said that his arteriogram on 28 November 1990 had shown these to be of small calibre but apparently healthy.

[112] Mr Drury had seen the angiogram produced by Dr Murch in November 1990. He confirmed that Dr Murch's report was an accurate representation of what he had seen: a blockage of 1 to 2cm at the level of the knee joint. Given that information, there would be no point in putting a bypass above the obstruction: the graft would have to go below the knee. Given the information from the angiogram, no ordinarily competent vascular surgeon would have been of the view that a bypass graft could appropriately be positioned above the knee. The decision to perform an above-knee graft was wrong, because there was "knowledge" from the November 1990 angiogram that the pursuer had an occlusion in the popliteal artery at the knee joint. No ordinarily competent vascular surgeon would have performed an above-knee graft, given the information that he had from the angiogram: he would have either performed a below-knee graft, or amputated the leg.

[113] Mr Drury was referred in this context to a study (the "Darke" study) concerned with patients who had isolated popliteal segments, defined for the purposes of the study as patients in whom the popliteal artery and the calf vessel were separated by an occluded or significantly stenosed trifurcation. The study compared the results of grafting above the knee, into an isolated popliteal segment, with those of grafting below the popliteal artery (i.e. into the trifurcation, or into one of the calf vessels). Mr Drury understood the study to show that, in such patients, there was no benefit in grafting below the occlusion as compared with grafting above it (and relying on collateral circulation to provide the outflow from the graft). This reflected the fact that grafts were technically more difficult, and had a higher rate of complications, the further down the leg the distal anastomosis was located. The pursuer did not have an isolated popliteal segment (IPS) as so defined, since her distal popliteal artery was patent i.e. there was no occlusion separating the patent section of the popliteal artery from the trifurcation.

[114] Asked about the defenders' averments to the effect that Mr Stewart had checked the patency of the artery by passing a catheter down it and injecting heparinised saline, Mr Drury said that the passage of a catheter down the artery would not be normal practice: Mr Drury would expect the operation note to record that that had been done, and the reason for its being done. He accepted that a catheter could be used to rest the patency of the artery distal to the proposed site of the graph. Mr Drury would however be extremely surprised if a catheter had passed down the popliteal artery, since the lesion shown on the angiogram looked hard and dense. Mr Drury found it difficult to believe that a catheter could pass through any occlusion without resistance. It might be possible to push it through a soft area of atheroma or occlusion, if the surgeon pushed hard enough, but not without resistance. If the clot were like soft jelly, then it might be possible for the catheter to go through the clot, or to push the clot down the artery. To push the catheter through an occlusion of the size shown in the November 1990 angiogram would take a degree of effort which could be traumatic to the other arteries. If the blockage were soft jelly-like clot, however, there would be much less difficulty. In that event, on the other hand, the passage of the catheter through the clot would not justify grafting above the knee, because the surgeon would have to investigate why clot had lodged at that particular point. If the situation found by the surgeon on the operating table was different from what was disclosed by the angiogram, then the ordinarily competent surgeon would record that in his operation note.

[115] Thus far, Mr Drury's opinion evidence was predicated on the view that, as a matter of fact, the angiogram had established the existence of an occlusion, which must still have been present in March 1991. He plainly doubted whether the suggestion that patency had been checked with a catheter could be true, since a catheter would not have passed through the occlusion at all, or at least not without extreme force being used. The possibility of the catheter passing through soft clot was put to him, but without any discussion of whether the occlusion in question might have been soft clot, beyond his evidence that the occlusion had looked hard and dense on the angiogram. His opinion was also influenced by his belief that the use of a catheter to test patency was abnormal and would therefore be expected to be recorded.

[116] In cross-examination, Mr Drury was asked to consider the vasoscans of 14 and 30 November 1990 and 18 March 1991. He agreed that they showed no drop in pressure below the knee compared with the value which would be predicted in the absence of disease (for example, on 30 November the pressure reading above the knee was 117mm Hg, from which a pressure of around 87mm Hg would be predicted below the knee: the actual reading at that point was 90mm Hg). He agreed that that information did not support the existence of an occlusion in the popliteal artery across the knee. In his opinion it did not however preclude the existence of such an occlusion, since it was possible that collateral circulation might explain the normal pressure reading.

[117] Mr Drury was also asked to consider the findings following the bypass operation. He accepted that the apparent success of the operation, for a time at least, might also indicate that there had been no occlusion as described by Dr Murch, but not necessarily: Mr Drury considered it possible that the proximal anastomosis might have widened the common femoral artery and so increased the flow of blood into the profunda femoris, and that in turn might (via collateral circulation) have increased the flow into the vessels below the knee. Importantly, Mr Drury accepted that Mr Stewart could not be criticised for putting the graft above the knee if he established that the popliteal artery was open. If Mr Stewart had followed the procedure described - i.e. the passage of a catheter - then Mr Drury would not criticise him for being negligent.

[118] Mr McCormick, in his evidence, was similarly critical of the decision to perform an above-knee graft, given that it appeared from the angiogram (which he had not seen) that there was an occlusion in the popliteal artery at the level of the knee joint. Given such an occlusion, one would expect occlusion of the graft by thrombosis within a few days. He would therefore anticipate that any vascular surgeon seeing the x-ray "and agreeing the appearances" would perform a below-knee graft. Mr McCormick accepted that the surgeon had to make his own assessment of the validity of the radiologist's opinion and was entitled to question Dr Murch's report. Mr McCormick also accepted that Dr Murch's report was inconsistent with the vasoscan information.

[119] Mr McCormick considered that Mr Stewart's operation note was not as full as he would have expected, if Mr Stewart had carried out all the steps which he claimed to have done. At the same time, Mr McCormick acknowledged that practice varied greatly between surgeons as to the content of their operation notes, and also acknowledged that in practice it was not unusual to have notes which did not record matters as fully as he would like.

[120] Mr McCormick accepted that the insertion of a catheter into the artery, and the introduction through it of heparinised saline in order to detect any obstruction, was a well-established procedure, and a reasonable test for an experienced surgeon to use. Using that method, the surgeon would be aware of an occlusion, unless it was very soft thrombus.

[121] Mr Stewart, in his evidence, said that he had asked Dr Murch to get him a better view of the popliteal artery than had been obtained in the previous angiograms. The x-ray which Dr Murch then obtained suggested that there was a block in the popliteal artery around the knee joint. This conflicted with the vasoscan of 14 November 1990, which had shown that there was no disease there: the measured pressure above the knee was 82mm Hg, from which the predicted pressure below the knee, in the absence of disease, would be 52mm - exactly the same as the measured reading below the knee. In other words, the measured drop in pressure from the segment of the main arterial tree immediately above the knee to the segment immediately below the knee was exactly what one would find in people without arterial disease - a normal gradient of about 30mm Hg. There would appear, from those findings, to be no significant disease in the popliteal artery. Having reviewed the x-ray, he was not entirely comfortable with the appearance of the block in the popliteal artery: it was rather short, it was sitting over the knee joint, and he did not feel that it had the right pattern for vascular disease. An additional complication was the presence of the bones of the knee joint on the x-ray. The location of the reported occlusion - at the junction of the bones of the joint - was the most difficult place to interpret appearances. A repeat vasoscan was done on 30 November. It again showed a pressure gradient across the knee which was normal (a drop from 117mm Hg to 90mm Hg) from which he would again conclude that there was no disease in that segment. Both vasoscans therefore conflicted with Dr Murch's report. No surgery was of course performed in the popliteal region during the November-December 1990 admission.

[122] When the pursuer was re-admitted on 12 March 1991, she was complaining for the first time of night pain, which suggested a deterioration. Hand-held Dopplers were carried out on 13 March, and showed an ankle pressure of 60mm Hg - a reduction from 100mm Hg in December. Mr Stewart decided to carry out, in the first instance, a further non-invasive assessment by vasoscan. That was done on 18 March. The results again showed a normal gradient across the knee - a drop from 100mm Hg in the segment above the knee to 78mm Hg in the segment below the knee. There again appeared to be no blockage in that part of the popliteal artery. The readings also indicated a satisfactory inflow into the femoro-popliteal segment: in other words, the angioplasty and sympathectomy carried out during the previous admission had resulted in a lasting improvement at that level. The pursuer's walking distance was also measured, and turned out to be 120 yards: much more than she had said on admission, and a distance which was surprising if, as she said, she had rest pain. In Mr Stewart's opinion, the pursuer certainly did not have established rest pain, but the night pain which she reported might be an early sign of rest pain. Her complaint of night pain was however at variance with the results of the investigations. Nevertheless, Mr Stewart felt that surgery was indicated because of the pursuer's expression of night pain and the other signs of a deterioration in her condition. He carried out the bypass operation four days later.

[123] Mr Stewart described his operating procedures in detail. The patient's notes and x-rays, including angiograms, would be in theatre, and the x-rays would be displayed on a viewing screen. In the pursuer's case, Dr Murch's angiogram would have been displayed, since there was a question in Mr Stewart's mind about the popliteal artery and the possible occlusion reported by Dr Murch. After the pursuer was anaesthetised, the first step was to explore the femoral artery, in order to confirm that there was adequate inflow into the femoral artery. That having been confirmed, it followed that the appropriate graft was a femoro-popliteal graft (i.e. a graft connecting the common femoral artery directly to the popliteal artery, bypassing the disease in the superficial femoral artery in the left leg), rather than a cross-over graft (i.e. one bringing blood from the right iliac artery to the left femoral artery). The next step would be to consider where the distal part of the anastomosis was going to take place, i.e. whether the lower end of the graft was to be connected to the popliteal artery above the knee or below the knee. The first step in that process would be to make an incision above the knee. An illustration to which Mr Stewart referred showed this to be a long incision ending at about the level of the knee joint. The segment of the popliteal artery lying below the incision would then be exposed. Having controlled the artery with elastic slings, the next step would be to assess the suitability of the exposed section as the point of anastomosis. As the artery was going to be occluded - i.e. the circulation would have to be closed off with clamps in order to operate on the artery - the anaesthetist would then be asked to give heparin (an anticoagulant) intravenously, to prevent the blood from clotting once the circulation was closed off. The artery would then be clamped at each end of the exposed section. An incision, known as an arteriotomy, would then be made in the artery. The distal clamp would then be taken off, so as to assess the backflow. Mr Stewart explained that that procedure was a simple way of ensuring that the artery was patent beyond the incision: if it was not, blood would not flow back to the incision. Mr Stewart said that what he next did, in every case where he operated on the popliteal and distal segment, was to assess the run-off. To do so, a catheter, with an external diameter of 2.66mm, was gently inserted into the artery. A syringe with heparinised saline was attached to the catheter. Mr Stewart would gently press on the syringe, to see how easily the fluid would go in, and pull back on it to see if blood came back up. He would gently push the catheter down the artery, stopping every few millimetres to check the position. This was a tactile assessment. If he came against an obstruction he would stop, bring the catheter out and measure it (from markings on the catheter) and so determine where the obstruction was located. The purpose of doing this was to check that there was adequate patency below the point where he intended to put the anastomosis. That was important in every case, because the run-off was important to the outcome of any bypass graft; and it was not unusual to begin an operation at a certain level and then have to move downwards, following this procedure, in order to get an adequate vessel which would provide reasonable run-off. In this particular case he also had particular cause to check whether there was an occlusion in the popliteal artery. The catheter would be inserted until it came against the vessels in the trifurcation, at which point it would be brought out and measured. The trifurcation could not be confused with an occlusion at the knee joint, because of the much greater distance involved. If there had been an occlusion, he would have detected it using this technique. The catheter could not have been pushed through organised arterial disease

[124] Mr Stewart accepted that his operation note (quoted in para.19 above) did not record everything he had done, but it was in his standard form. The only non-standard thing he had done was to check the femoral artery pressure by invasive methods, and so he had specifically recorded that. The only other important aspects requiring to be recorded were the locations of the proximal and distal anastomoses, and the type of prosthesis used. All of these matters were recorded.

[125] Post-operatively, the pursuer was noted on 23 March 1991 to be pain free. On 25 March, the foot was noted to be warm, which was a good sign. On 27 March, Dopplers were recorded. They showed that the pressures in the left foot had returned to the normal range: 110mm Hg, as compared with 60mm Hg on 13 March; and an ABI of 0.92 on 27 March, compared with 0.43 on 13 March. The nursing notes depicted a similar situation. On 22 March and the following days the foot was noted to be warm, with a palpable pulse. That implied that there was pulsatile flow at the ankle, and therefore that there must be a functioning graft, with a pulsatile flow from the graft through a patent arterial system to the ankle. In other words, the post-operative findings implied, in Mr Stewart's opinion, that the graft itself was functioning well, and that between the popliteal anastomosis and the foot there was a patent arterial tree. The findings were not consistent, in his opinion, with there having been an occlusion in the position described by Dr Murch. They could not be explained by collateral circulation. It was not unusual for the operative findings to differ from the radiologist's report. Surgeons were trained not to make decisions simply on the basis of radiographs: they had to make their decisions at operation, based on all the information that they had.

[126] Professor Ruckley considered that, when the pursuer was re-admitted on 12 March 1991, it was evident that her foot had deteriorated since she reported that for the previous month she had had night pain. An attempt to further improve the circulation by femoro-popliteal bypass had to be undertaken. This was the first evidence that the pursuer had reached the stage of critical ischaemia, and therefore the first time she had reached the stage at which many vascular surgeons would consider it appropriate to operate on the ischaemic leg.

[127] In relation to the location of the graft, Professor Ruckley confirmed that it was well known among vascular surgeons that better results were achieved by above knee rather than below knee femoro-popliteal bypasses. In principle therefore an above knee site was to be preferred. In practice, surgeons had to make their decision using the diagnostic tools that they had trained with, knowing their value and limitations. No diagnostic test, including angiography, was infallible; and there was no true "gold standard" (Mr Drury's and Mr McCormick's description of angiography), since different tests measured different parameters and registered different types of image. The cautious surgeon therefore did not base his decisions on a single observation but on all the evidence available.

[128] In particular, an angiogram could indicate occlusions which were not present. It had been quite a common experience for Professor Ruckley to explore arteries and find them to be patent when the radiologist had described them as occluded. There could be a number of reasons why a short segment of artery might appear to have an occlusion which did not exist. The commonest cause of artefact was overlying shadows. At the knee joint, the gap between the joint surfaces was small, and the bone density at the articular ends of the two bones was very high. The contour of the joint was a series of convexities and concavities, and in the middle of the joint there were bony spurs coming out, casting an irregular radiographic shadow. That was a possible cause of artefact in x-rays of the knee joint, which he had seen in practice.

[129] Professor Ruckley had not seen Dr Murch's angiogram before it disappeared. He however shared Mr Stewart's opinion that a blockage of 1 to 2 centimetres at the level of the knee joint was not typical of atheromatous disease. He could not recollect ever having seen a patient with arterial disease with that problem. He would expect a blockage to be longer and more proximal. He also confirmed that angiographic appearances below other blockages (in this case, at the level of the superficial femoral artery) were difficult to interpret and fallible, particularly where the contrast material crossed the articular surfaces of the two bones which met at the joint. The vasoscan, on the other hand, showed that there was no block or tight narrowing at that level. Given that none of the vasoscans provided any supporting evidence of a popliteal occlusion, it was Professor Ruckley's view that Mr Stewart's decision to assess the situation at operation was the correct one. The final decision precisely where to locate the top and bottom ends of bypass grafts was invariably influenced by the findings at operation. No competent surgeon went into an operation with a fixed ideas as to how to position the graft: it was a matter of judgment at the time of surgery. There were a number of ways of assessing the status of the outflow vessels. The choice was determined by the availability of facilities and the techniques with which the surgeon had been trained or had personally developed. Professor Ruckley's own normal practice, like Mr Stewart's, was to pass a catheter down the vessel and inject some heparinised saline and to assess the patency and freedom of flow. It was his opinion therefore that Mr Stewart's approach to this operation was sound.

[130] Critically, Professor Ruckley, like Mr Stewart, considered that the post-operative findings were inconsistent with the presence of an occlusion: they could not, in particular, be explained by collateral circulation. Although there was a network of collateral channels which crossed the knee joint (known as the geniculate arteries), they were very small vessels and sustained a small volume blood flow. A popliteal artery blockage resulted in very severe impairment of circulation to the lower leg. Reference was made in this context to the Darke study, in which the ankle pressures had been recorded of patients who had a blocked popliteal artery and a graft taken to above the block, and were therefore dependent on collateral circulation for blood flow to the distal vessels. The pressures were far lower than the pursuer's post-operative pressures. Professor Ruckley had never seen a patient with a blocked popliteal artery who had palpable pulses at the ankle or ankle Doppler pressures remotely approaching normality. In the pursuer's case, however, the post-operative observations were of palpable pulses and Doppler pressures in the normal range. They were values which could only be obtained if there were patent main stem vessels, and in particular a patent popliteal artery. In Professor Ruckley's words, "this is conclusive evidence that there was no blockage in the popliteal artery below the graft".

[131] Responding to the suggestion that there might have been an occlusion constituted by a blood clot forming on a diseased artery wall, and that the clot had moved, Professor Ruckley said that he would not expect a clot of that kind (i.e. thrombosis on pre-existing disease) to move spontaneously, although it might move if it were pushed by a catheter. That however could only happen when the clot was fresh. In its early stage, a clot was a jelly of coagulated blood; but it then underwent a process of organisation in which it became steadily more solid and more adherent to the wall of the artery. After a day or two of its formation, it would become progressively more difficult to dislodge. Vascular surgeons knew that if a clot formed in an artery they had got a few days to remove it, after which it became impossible to dislodge it, unless it were cut out of the artery or burned away. If there was a clot present on 28 November 1990, it was impossible that it could have been dislodged by the passage of a catheter on 22 March 1991. That theory could be ruled out.

[132] On a related point, Professor Ruckley did not accept that a surgeon ought to go below the knee merely because of significant disease in the popliteal artery at the knee which could give rise to clot. There was always going to be active arterial disease throughout the arterial system. The preferred site for a graft would be above the knee, unless there was an actual occlusion at the knee.

[133] So far as the operation note was concerned, Professor Ruckley said that the style and content of surgical operation notes was extremely variable. Mr Stewart took the view that the essential ingredients were the location of the upper and lower anastomoses and the nature of the graft, these being the key items of information needed by any surgeon who had to re-operate later. Many surgeons would agree, although others would write in more detail. Professor Ruckley did not consider that Mr Stewart's note keeping was at variance from that of a great many other vascular surgeons.

[134] Professor Ruckley accepted however that if the surgeon found that the artery was patent and suitable for grafting into, whereas the radiologist had said that it looked as if it was unsuitable, then that would normally be recorded; but the reality of life was that that did not always happen. Mr Stewart's note contained the essential information which a subsequent treating doctor might require.

[135] Professor Ruckley was one of the authors of the Darke study of patients with isolated popliteal segments. He regarded this study as relevant to the pursuer's case, on the hypothesis - contrary to his evidence - that there was indeed a popliteal blockage present at the level of the knee joint as reported by Dr Murch. This would have created an isolated segment of patent upper popliteal artery. To have inserted a graft into the patent segment above this blockage would not in his opinion have been negligent, given the known capacity of isolated segments to sustain functioning grafts, a capacity illustrated by the Darke study. If Dr Murch were correct, there was an isolated popliteal segment, comprising the upper two-thirds of the popliteal artery; and such a segment was capable of sustaining a graft. In other words, if the graft was going to fail, it was not going to be because of where the distal anastomosis was located.

[136] The Darke study was a trial designed to study the relative merits of a short graft placed from the groin into an isolated popliteal segment around the knee level (relying on collateral circulation to link up with vessels distal to the isolated segment), compared with a longer graft running from the groin to a small artery in the lower calf or foot. It was found that there was no significant difference between the two groups of patients in terms of outcome: the results were however somewhat better for the patients who received a graft into the isolated popliteal segment, and the study therefore concluded that that was the preferred site. Because the purpose of the study was to compare a short bypass into the isolated popliteal segment with a long bypass further down the calf, the study looked only at patients in whom the choice of treatment lay between those alternatives, i.e. patients in whom the occluded segment of the popliteal artery extended to the trifurcation. The pursuer was not such a patient, since, even if Dr Murch were correct, the distal popliteal artery was patent. That difference did not however affect Professor Ruckley's opinion: the relevance of the study was that it showed how well an isolated popliteal segment was capable of sustaining a graft. In other words, although the pursuer did not have the particular type of isolated popliteal segment studied by Darke and his colleagues, she did (if Dr Murch were correct) have an isolated popliteal segment; and such a segment was capable of supporting a graft. In cross-examination, Professor Ruckley accepted that, assuming that there was an occlusion at the knee and adequate run-off below the knee, most vascular surgeons would perform a below-knee graft. Asked if the ordinarily competent vascular surgeon would not fail to do that, he replied, "I think that's probably what they would do, yes." Subsequently he explained that the surgeon would know that an isolated popliteal segment was feasible, but that (assuming free outflow in the distal vessels) it would be more likely to fail than an outflow which was totally unimpeded. In the light of those answers, I have difficulty understanding Professor Ruckley's earlier opinion that to graft above a mid-popliteal occlusion would not be negligent. I recognise that the explanation for this may well lie in the deficiencies of my own medical understanding, but the consequence is that I find myself unable to accept his evidence on this point.

[137] Finally, Professor Ruckley said that many vascular surgeons, although not the majority, would not have offered bypass surgery to the pursuer because of the risk factors and the poor prognosis of arterial disease in this category of patient. They would not have offered surgery for claudication. Once the condition had progressed to critical ischaemia, amputation would have been recommended. This opinion was consistent with evidence given by Mr Drury and Mr McCormick.

[138] I should also mention the evidence of Mrs Elizabeth Jones, who was led as a witness for the pursuer. She was a theatre sister of 21 years experience. She had worked at Ballochmyle Hospital from 1963 until 1991. Between 1987 and 1991 she had assisted Mr Stewart, and in particular had assisted in femoro-popliteal bypass operations two or three times a week. According to her evidence, Mr Stewart always used a catheter, in every femoro-popliteal bypass operation. It was the practice to display angiograms during surgery. She had no specific recollection of the pursuer's operation. Mrs Jones confirmed that Mr Stewart took time to teach nursing staff about vascular procedures. She described Mr Stewart as "very committed indeed". He left his staff in no doubt of what was expected of them. Her impression of Mr Stewart as a vascular surgeon was that he was "meticulous". He was devoted to vascular surgery. He had a good working relationship with other staff. Asked if he was a caring doctor, she replied, "very, very much so, yes, indeed, yes." Mrs Jones's observation that the adjective "meticulous" summed up Mr Stewart quite well accords with my own impression. This evidence is of some significance in view of the insinuation by the pursuer's counsel that, because Mr Stewart was extremely busy during the period in question, he might have been careless.

[139] The records are equally eloquent of Mr Stewart's dedication and care in his treatment of the pursuer. When she was initially referred to his clinic, and on subsequent outpatient visits, she was seen not by a junior doctor but by Mr Stewart himself (other than on one occasion, when Mr Stewart was on leave, and she was seen by another experienced surgeon). Despite his formidable workload and responsibilities, on each occasion that she was admitted to hospital she was seen by him soon after admission, prior to discharge and at regular intervals throughout her hospital stay. He personally carried out the arterial surgery. At each phase of her care he was careful to re-assess her condition and to modify his plans in the light of clinical and investigation findings. Professor Ruckley observed that this adds up to a picture of a standard of clinical care and a level of personal attention by the consultant which is greatly superior to the common practice of most hospitals.

[140] In general terms, it is common ground that the decision as to the siting of the distal anastomosis of the pursuer's bypass graft was a question of judgment for the operating surgeon. It is also common ground that Mr Stewart had conflicting information: Dr Murch's angiogram showed what appeared to be an occlusion in the popliteal artery, but the vasoscans showed no significant arterial disease at that level. I also accept that the appearance shown on the angiogram was uncharacteristic of the pursuer's disease. In these circumstances I accept that it was not merely reasonable, but undoubtedly right, for Mr Stewart to satisfy himself during surgery as to the adequacy of the inflow and outflow to support a bypass graft. I accept Mr Stewart's evidence that he did so, by performing the catheter test. I have no reason to disbelieve his evidence on this matter, and I regarded him as a credible and reliable witness. In particular, I do not share Mr Drury's doubts on this point, which were due partly to his belief that such a test would not be normal (contrary to the evidence of Mr McCormick and Professor Ruckley, as well as Mr Stewart) and partly to his firm, and even dogmatic, belief that the angiogram had definitely shown a block. In accepting Mr Stewart's evidence on this point I am also influenced by my general impression of him as a careful and cautious doctor; Mrs Jones's evidence that he was, in her word, "meticulous"; and her evidence that it was his invariable practice to use the catheter test when performing femoro-popliteal bypass operations. If Mr Stewart performed that test, then the artery cannot have been occluded as Dr Murch reported: it is plain from the evidence of all the expert witnesses that the presence of an occlusion close to the arteriotomy would have been detected using that test.

[141] My confidence in the conclusion that there was no occlusion in the popliteal artery is strengthened by the fact that I am also driven to the same conclusion by an entirely separate body of evidence, concerning the post-operative findings. Mr Stewart and Professor Ruckley both gave evidence that the blood flow and pressure in the distal vessels after the operation were consistent only with a patent arterial tree from the distal anastomosis downwards: they could not be explained by collateral circulation. Mr Drury gave contrary evidence. I am not however persuaded that Mr Drury's evidence on this matter is to be accepted: I have no reason to reject the evidence of Mr Stewart and Professor Ruckley, and it is supported also by the figures in the Darke study. I also reject the suggestion that there might have been an occlusion as shown by Dr Murch, but that it was moved by the catheter: that possibility can be excluded, since any occlusion present in November would have been immoveable by March, even if it had been fresh (contrary to Mr Drury's evidence about its radiological appearance) in November. I accordingly conclude, from the post-operative findings, that there cannot have been an occlusion in the popliteal artery.

[142] Mr Drury criticised Mr Stewart's decision to carry out an above-knee graft solely on the basis that there was an occlusion: "if there was no occlusion I wouldn't criticise him". Mr McCormick's criticism was made on the same assumption, and he accepted that the catheter test was a reasonable test to perform. Since I am satisfied that there was no occlusion, I am unable to accept their criticism of Mr Stewart's decision. Their criticism effectively placed the angiogram on a pedestal; and counsel for the pursuer erected an even higher pedestal for it in her submissions. Mr Stewart's decision as to the siting of the distal anastomosis, on the other hand, was correctly based on his interpretation of the totality of the pre- and intra- operative findings. Having regard to these findings, it was reasonable - indeed, correct - for Mr Stewart to conclude that the distal anastomosis should be sited above the knee.

[143] More particularly, my conclusions in respect of the pursuer's submissions (as summarised in para.110 above) can be summarised as follows.

[144] In case my conclusions on the issue of fault should be held to be flawed, I have to deal also with the remaining issues in the case, namely causation and the quantification of damages, on the hypothesis that fault is established.

The averments of loss then detail a number of elements: severe ischaemic limb pain from December 1990 to July 1991, when the amputation was converted to an above-knee amputation; further pain in the stump until it was revised again in November 1992; the amputation operation itself, and the two subsequent revisions; severe disability and lack of mobility; a depressive illness due to the leg pain and the loss of the leg; loss of earnings; a poor quality of life; various expenses; and a claim in respect of services provided by the pursuer's husband. In essence, therefore, it is alleged that Mr Stewart's negligence caused (1) a prolongation of pain and (2) the loss of the leg. The other elements of the claim are consequential upon those two primary elements, and especially upon the loss of the leg.

[146] In closing submissions, counsel for the pursuer summarised her position as follows:

Counsel acknowledged that it would not have been negligent of Mr Stewart to have amputated the pursuer's leg in November or December 1990 or March 1991, but submitted that that was irrelevant. This was a situation in which two treatments could competently be offered - a below-knee bypass, or amputation - and the practitioner had selected the bypass. He having made that selection, the patient was then entitled to have the treatment performed competently; and if it was not, the patient was then entitled to contend that that negligence had deprived her of the outcome which would have resulted from that treatment if it had been competently performed. The fact that the outcome of the negligent treatment might be the same as the outcome of the alternative treatment, which the practitioner might properly have selected in the first place, was beside the point.

[147] Counsel also acknowledged that the leg could probably not have been saved indefinitely (as indeed counsel had acknowledged during the evidence of Mr Drury). A below-knee graft would, it was submitted, have delayed amputation.

[148] Counsel for the defender began by emphasising, under reference to Wilsher v Essex Area Health Authority [1988] AC 1074, that the pursuer must establish that the breach of duty was a material cause of the adverse result of which she complains. In particular, if she is to recover for the loss of her leg, she must establish that she would not have lost it but for Mr Stewart's negligence; and if she is to recover for her poor mobility and quality of life, she must establish that she would not have been in that condition but for such negligence.

[149] Counsel submitted, in the first place, that no better result could have been expected in the particular circumstances of this case. This point was made in a number of ways, which might be summarised thus:

[150] Secondly, counsel submitted that, on a balance of probabilities, a below-knee bypass would not have made any difference. The pursuer's small vessels, her poor outflow circulation, the aggressive nature of her disease and her risk factor profile all pointed to a strong probability that the outcome would have been no different. It was not enough for the pursuer to establish that a below-knee bypass would have had better prospects of success, and that she had lost a chance of saving her leg: she had to show that there would have been a different outcome.

[151] Thirdly, counsel submitted that the major part of the pursuer's disability had been caused by her stroke, and was not attributable to anything Mr Stewart had done or failed to do.

[152] Mr Drury stated that a successful graft operation removed rest pain virtually immediately (as indeed is confirmed by the fact that the pursuer was herself free of pain in her leg while her graft remained patent). The pursuer would not, after a successful below-knee bypass in November or December 1990, have needed so many painkillers and would have been more comfortable. She would still have had pain. If sufficiently motivated, she would have been sufficiently mobile to return to work within three months. If operated on successfully in March 1991, the recovery would have been delayed by several months, because the heel ulcer had progressed by then. If a below-knee graft had been performed either in December 1990 or in March 1991, it was probable that she would not have lost her leg "shortly after surgery". A below-knee vein bypass graft would have had a 70 per cent chance of patency at one year. The leg could probably have been saved for several years. The 70 per cent figure came from the world literature and would be reproduced in most textbooks. It appeared from Mr Drury's evidence that that figure was a general statistic for the results of patients who undergo such a bypass, rather than an assessment of the pursuer's prospects in particular. Mr Drury said that even if the graft ceased to be patent, it did not follow that the limb would necessarily be lost, if the graft had opened up the collateral circulation before it became blocked. Considering the pursuer's case specifically, Mr Drury based his prognosis on the view that "there is no information available to suggest that she had severe distal disease in the left leg beyond her occlusion". In those circumstances, he considered that, on a balance of probabilities, the pursuer would have kept her leg for somewhere between two and four years.

[153] I observe that the statistical evidence has to be treated with caution, and not only because it is not based on an assessment of the pursuer's case in particular. Mr Quin, for example, who performed a below-knee bypass on the pursuer following the occlusion of the above-knee graft performed by Mr Stewart, said that the likely success of his operation, given the previous failed operation, was no better than 50 per cent. In reality, however, it would appear that the operation was doomed to failure, because the state of the pursuer's arterial system was such that it was unable to maintain the graft in a patent condition. In other words, although 50 per cent of below-knee grafts, in patients with a recent failed above-knee graft, might succeed, the pursuer was always in the other 50 per cent. The actual likelihood of her operation succeeding, if the aetiology of graft failure was perfectly understood and all the material facts were known, was nil. Similarly, when Mr Drury said that there was a 70 per cent patency rate after one year, that does not necessarily imply that the pursuer's graft was more likely than not to be patent after one year: the question remains whether she would have been one of the 70 per cent, or one of the 30 per cent whose grafts became occluded. I refer to the observations of Lord Mackay of Clashfern on this issue in Hotson v East Berkshire Area Health Authority [1987] AC 750 at pages 784-785. If one is unable to form any view about that question, beyond the overall statistical chances, then the balance of probabilities favours patency: but other evidence in the case may enable one to decide that the pursuer was more likely to be amongst the 30 per cent than the 70 per cent. Mr Drury's prognosis for the pursuer, as I have mentioned, was based on an absence of information suggesting severe distal disease.

[154] In relation to the latter point, Mr Drury spoke to a report which he had prepared, in which he recorded his own interpretation of the pursuer's angiograms. It will be recollected that the first angiogram report was by Dr Russell on 12 February 1990. It stated, in relation to the vessels below the popliteal artery:

Dr Murch's angiogram report of 28 November 1990, following the angioplasty, stated:

I note that none of these x-rays appears to have shown patent vessels extending to the lower calf, ankle or foot. This was a point, as will appear later, which was emphasised by Professor Ruckley.

[155] Mr McCormick similarly gave evidence that a successful below-knee graft would have removed rest pain. If the pursuer had such a graft in December, she would have been walking a month later. In his own patients, the success rate was of about 75 per cent patency at one year. It dropped below 50 per cent after about five years. The risk of amputation was only of the order of 30 per cent after the first year. Had the pursuer had a below-knee graft in December or March the patency rate would have been of the order of 75 per cent at one year.

[156] Mr Quin had carried out a below-knee bypass graft, following the occlusion of the above-knee graft performed by Mr Stewart. Mr Quin's graft also had become occluded. He said that one explanation might have been that the distal popliteal artery was small in calibre. In a patient with slender arterial vessels, the prospects of success of arterial bypass surgery might not be good.

[157] In relation to causation, Professor Ruckley observed at the outset that there is no evidence whatever that surgical intervention alters the progression of the disease in the arteries: a bypass is put in to relieve symptoms, and does not prevent (indeed, it may accelerate) the deterioration of the pathological process in the remainder of the arterial tree. The pursuer's prognosis in general was determined by the progression of her arterial disease. Her risk factors included the effects of previous smoking, hypertension, hypercholesterolaemia, a family history of vascular disease, early onset disease and small calibre arteries. The significance of this was apparent in the subsequent development of coronary artery disease (i.e. angina and ECG evidence of heart attack), stroke and deterioration of the circulation in her right leg. She was severely disabled by stroke, to such an extent that, having prior to the stroke been graduating to walking sticks instead of crutches and contemplating going back to work, subsequent to the stroke she required assistance with all her personal hygiene needs and household and domestic needs and was not able to walk properly at all. The major part of her disability was due to her stroke.

[158] Considering the question whether it would have made any difference if bypass surgery had been performed in December 1990 or before March 1991, Professor Ruckley answered that question on the basis that between August 1990 and February 1991 the pursuer was in the grey zone between end stage claudication and impending critical ischaemia, and that rest pain had started in late February 1991. In those circumstances, Professor Ruckley said that there was no evidence whatever in the world literature or within his own clinical experience that indicated that the outcome for the pursuer would have been different if she had had her interventions at different times within this period.

[159] Considering the question whether the outcome would have been different if Dr Murch had been correct about the presence of an occlusion, and the bypass had been positioned at the below knee level, Professor Ruckley said that the necessary evidence to answer this question was not available. All vascular surgeons agreed that the success of a graft in terms of patency and limb salvage depended on the patency of both the inflow and the run-off vessels. Mr Stewart had corrected the narrowing in the inflow by angioplasty. A successful below-knee bypass also needed good quality outflow vessels, not only in the calf but also at the level of the ankle and foot, in order to provide continuity of outflow. In the pursuer's case, Professor Ruckley would have no confidence that a below-knee graft could function any better than a graft positioned above the knee, because all the radiological evidence he had seen suggested that the run-off below the knee in this patient was extremely poor. She had poor distal vessels. None of the angiograms gave any indication that any of the three arteries running down the calf - the anterior tibial, the posterior tibial and the peroneal - were patent beyond the lower calf. In these circumstances he would be very concerned indeed that a bypass graft might not remain open because of the poor quality of run-off in the lower calf and in the foot. The angiogram reports were very unpropitious for the success of a bypass graft, wherever the lower end was sited.

[160] In Professor Ruckley's opinion, there was no evidence that a below knee bypass performed any time between August 1990 and March 1991 would have functioned more successfully than did the above knee graft. The risk factor profile and the clear evidence of the aggressive nature of the pursuer's arterial disease pointed to the strong probability that the outcome would not have been any different.

[161] Conflicting views have been expressed, as between Mr Drury and Mr McCormick on the one hand, and Professor Ruckley on the other hand, as to whether it would have made any difference if the pursuer had undergone a below-knee bypass graft, on the hypothesis that such a procedure ought to have been carried out. As I have mentioned before, since the burden of proof rests on the pursuer, she cannot succeed unless I am in a position to reject the opinion expressed by Professor Ruckley. I was invited by the pursuer's counsel to do so on the basis that Professor Ruckley, unlike Mr Drury, had not seen the angiogram of 28 November 1990, which had given a clearer picture of the distal run-off than the earlier angiograms. Given that Mr Drury, who had seen all the angiograms, had formed the opinion that there was adequate run-off to support a graft, Professor Ruckley's opinion to the contrary had no adequate basis and should be rejected. I am unable to accept that submission. Professor Ruckley's opinion was based, primarily at least, on three aspects of the pursuer's case: her risk factor profile; the aggressive nature of her disease, as demonstrated by its history both prior and subsequent to the events in question; and the failure of the angiograms to demonstrate patent vessels below the lower calf. I had the impression that he also regarded as relevant the fact that the below-knee graft performed by Mr Quin also became occluded within a short time. Professor Ruckley was not in error in relation to any of these matters of fact. The conclusion which he drew, and expressed in his opinion as regards causation, was not incoherent or illogical. In those circumstances, I have no proper basis for rejecting it. I am also bound to say that it also fits with my own strong impression, looking at the evidence as a whole, that this unfortunate lady had a relentless and aggressive disease.

[162] I would be more hesitant about accepting Mr Drury's opinion. In so far as it was based upon statistics, I have already commented on the limited usefulness of general statistics in determining the likely outcome in a particular case. Professor Ruckley's opinion, on the other hand, was based on the specific features of the pursuer's case, and his experience of such cases. Mr Drury's opinion was also based upon the view that there was nothing to suggest that the pursuer had severe distal disease beyond the supposed occlusion. It is unfortunate that Mr Drury was not questioned, in that regard, about the appearance (or, rather, non-appearance) of the more distal vessels in the angiogram, which might on one view raise a doubt, at least, about Mr Drury's assertion. For the same reasons, but with rather greater force, I would also be hesitant about accepting Mr McCormick's opinion, which was based largely on the general statistics. Accordingly, I am not prepared to accept, even on the hypothesis that there was an occlusion in the popliteal artery which ought to have been bypassed, that such a bypass would have been more successful than the graft actually performed by Mr Stewart. I therefore do not accept that the pursuer would have avoided any pain which she in fact suffered, or would have avoided an amputation and its sequelae, if a below-knee graft had been performed. It follows that, as well as failing to establish negligence, the pursuer has also failed to establish that, even assuming negligence, any injury was caused.

[163] I should finally deal with the argument that the pursuer cannot recover damages for the loss of her leg because amputation would have been an acceptable alternative to bypass surgery. My observations on this point are of course obiter dicta.

[164] According to the pursuer's submission, there is simply no legally recognised way in which this argument can gain a foothold. If Mr Stewart failed to carry out the bypass operation with reasonable care, then he committed a legal wrong. He is then liable for any injury caused by his failure to take reasonable care, including (ex hypothesi) the loss of a leg which a carefully performed operation would have saved. The consideration that Mr Stewart might properly have amputated the leg, rather than performing a bypass at all, has no bearing on the question of causation. Nor does it enter into the consideration of whether Mr Stewart was negligent. By the time he behaved negligently, the option of amputation was simply part of the history of the case: it was, as it were, the road not taken.

[165] The defenders' approach, on the other hand, oscillated between an argument directed towards causation ("but for the alleged negligence the pursuer would still have lost her leg) and an argument directed towards duty of care ("failure to save the leg was not culpable").

[166] There is in my opinion considerable force in the contention that Mr Stewart should not be held to have incurred a liability for the loss of the leg by trying unsuccessfully - even negligently - to save the leg, when it would have been reasonable and proper for him not to have attempted to save it. It would, for obvious reasons, be undesirable if the law were to discourage those who are prepared to undertake such interventions. In the circumstances of the present case, in particular, it would be undesirable if the law were to have the effect that surgeons (and health service trusts) who carried out bypasses ran a risk of legal liability which could be avoided by performing amputations instead. I am unaware of any exactly analogous situation in any decided case, and none was cited to me. The principle upon which the case of East Suffolk Rivers Catchment Board v Kent [1941] AC 74 was decided, however, appears to me to be relevant. In that case, the plaintiff's land was flooded by sea-water and the defendant Board, although under no duty to repair the breach in the sea wall, did undertake such repairs but so negligently that the land remained flooded for much longer than it would have done had the repair been efficient. The House of Lords held that the Board was not liable, because its neglect did not inflict any more damage than would have resulted from total inaction on their part. Their Lordships acknowledged the importance of not creating a situation in which prudent Boards would do nothing so as to avoid exposing themselves to the risk of liability. On the other hand, their Lordships accepted that if additional damage had been caused by the negligent conduct of the repair, there would have been liability for that additional element. Like the defenders' submission in the present case, their Lordships' speeches in the East Suffolk case analysed the issue both in terms of causation and in terms of duty. Viscount Simon L.C.'s speech focused (at pages 87-88) on causation:

Lord Thankerton adopted a similar analysis: the causa causans of the prolonged flooding was the tide, rather than the Board's negligent repair. Lord Romer, on the other hand, focused (at page 102) on the scope of the Board's duty:

Lord Porter adopted a similar approach. Lord Atkin delivered a dissenting speech.

[167] The decision, and the wider issue which it illustrates, are discussed in Hart and Honoré, Causation in the Law (2^nd ed., 1985, page 140). The authors suggest that the "duty" analysis should be preferred to the "causation" analysis, on the basis that issues of causation are to be solved by the use of common-sense notions and are eminently suitable for submission to a properly instructed jury, whereas policy limitations upon responsibility are reflected in other concepts, notably duty of care:

I am inclined to agree that it is preferable to deal with causation as a question of fact, "dealt with broadly, and upon common-sense principles as a jury would probably deal with it" (The Volute v Admiralty Commissioners [1922] A.C. 129, 136 per Lord Birkenhead L.C.), and therefore to deal with the type of issue raised by the East Suffolk case in terms of the scope of any duty of care. The East Suffolk case arose in a setting of statutory powers. It was followed, in a context of common-law duties, in the Canadian case of Horsley v MacLaren [1970] 1 Lloyd's Rep. 257, a decision of the Court of Appeal of Ontario which concerned a negligently conducted rescue attempt.

[168] The principle which can be extracted from the East Suffolk case might be expressed thus: where a person is under no duty to intervene to assist the pursuer but chooses to do so, he will incur no liability provided his intervention does not leave the pursuer worse off (I adopt that formulation from Clerk & Lindsell on Torts, 17^th ed., 1995, para.7-32; other formulations are of course possible). So expressed, the principle does not cover the present case: Mr Stewart did not intervene gratuitously and without any duty to do so; he undoubtedly owed a duty to the pursuer, as his patient, to treat her and to provide her with a standard of care which met the minimum reasonably to be expected of an ordinarily competent vascular surgeon. Nevertheless, the principle of the East Suffolk case seems to be capable of extension to cover the extremely unusual facts of the present case. An extended version of the principle might be expressed thus: where a person is under no duty to undertake a particular intervention to assist the pursuer but chooses to do so, he will incur no liability provided his intervention does not leave the pursuer worse off than she would have been if that person had acted in some other manner compatible with any duties incumbent upon him.

[169] On the evidence in the present case, it is not in dispute that Mr Stewart was under no duty to undertake bypass surgery. The alternative treatment, which would have been compatible with the duties incumbent upon him, would have been amputation. In these circumstances, even if the bypass surgery had been negligently performed. I do not consider that Mr Stewart would be liable for all the consequences of the surgery: he would be liable only to the extent that those consequences were worse than the consequences of an amputation would have been. In expressing this view, I am conscious that amputation could not have been carried out without the pursuer's consent; and it is conceivable that the issue of informed consent might involve the possibility of referral to another consultant prepared to carry out bypass surgery. Issues of this kind were not however explored in evidence or submissions, and those submissions proceeded on the basis that bypass surgery and amputation were, in a straightforward sense, alternative treatment options. I have therefore dealt with the issue on that basis.

[170] I was addressed in detail on the quantification of damages. Not surprisingly, parties' submissions were based on widely differing assumptions as to causation: for example, whether the pursuer's leg could have been saved, and if so, for how long; whether, if an amputation had proved necessary at some stage, it would have required the various revisions which the pursuer in fact underwent; the extent to which the pursuer's loss of mobility and poor quality of life were attributable to the amputation or to the stroke; the extent to which her depression was attributable to these aspects of her condition, or to a pre-existing condition; and her likely employment history if a graft had worked, at least temporarily. Parties' respective assessments of damages were correspondingly disparate, and a wide range of figures were suggested for a variety of scenarios. Given my conclusions in respect of fault and causation, I see no point in going through all the possible situations which might arise if those conclusions were held to be flawed, and assessing the damages appropriate to every considerable situation. The possibilities are, if not endless, at least manifold. In these unusual circumstances, I shall not offer a view as to the appropriate quantum of damages.

[171] Mr Stewart was called upon to make a series of judgments during his treatment of the pursuer. At each stage his approach was cautious and considered. When examined years later, it stands up to rigorous analysis. In my opinion it cannot be said that in reaching any decision, or in making any judgment, Mr Stewart was negligent. Indeed I would go further: on the evidence, I am satisfied that Mr Stewart made the right decision at every stage.

[172] The pursuer is deserving of much sympathy for all that she has had to endure. But what she has endured has been the consequence of the relentless disease from which she has the misfortune to suffer. It has not been the consequence of her treatment at Ballochmyle Hospital. On the contrary, her treatment there does credit to all concerned, and especially to Mr Stewart. It is always regrettable that someone who has devoted his professional life to the care of his patients should be unjustifiably blamed for problems which he has done his best to alleviate. The heavy responsibility of making crucial decisions in relation to the care of the seriously ill is not made any easier when those who bear that responsibility are subject to ill-founded criticism of their professional judgment and conduct.

[173] In the circumstances, I shall sustain the defenders' second and third pleas-in-law, repel the pursuer's pleas-in-law, and pronounce decree of absolvitor.

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	Scottish Court of Session Decisions
You are here: BAILII >> Databases >> Scottish Court of Session Decisions >> Antonucci Or McConnell v Ayrshire & Arran Health Board [2001] ScotCS 35 (14 February 2001) URL: http://www.bailii.org/scot/cases/ScotCS/2001/35.html Cite as: [2001] ScotCS 35

Scottish Court of Session Decisions