Scottish Sheriff Court Decisions

[6] Mrs Bellfield was born on 22 March 1949. On 13 February 2009 she attended as an outpatient at the endoscopy unit at Victoria Hospital, Kirkcaldy in order to undergo an upper gastrointestinal endoscopy.

[7] The principal aim of the procedure was to investigate further an oesophageal stricture which had been identified by a previous endoscopy and to take biopsy tissue samples for testing. The consultant treating Mrs Bellfield, Dr Kishor Vaidya, suspected that a cancerous tumour might be present, but biopsy samples taken during a previous endoscopy had tested negative for malignancy. Depending on where exactly in the tissues the cancer is present, it can be difficult to diagnose cancer in the oesophagus. A secondary aim of the procedure was to dilate the narrowed area in order to provide symptomatic relief in respect of swallowing difficulties.

[8] Dilatation of the oesophagus is a process carried out by a medical instrument known as an endoscope, whereby a balloon is inserted into the oesophagus by means of a tube in the endoscope which is inserted into the patient's oesophagus through the mouth. The balloon is then inflated by the endoscopist, resulting in expansion of the oespophagus. This results in significant pressure being exerted on the oesophagus, and can result in significant air being introduced into the patient's abdomen. There is an increased risk of the oesophagus rupturing where cancer is present. Nonetheless, it is reasonable to carry out the procedure even when cancer is suspected, there being no other means of obtaining a diagnosis. The oesophagus must be expanded to enable it to be examined visually by a camera, also inserted into the patient through a channel in the endoscope. Further, unless symptomatic relief is provided to a patient who has difficulty swallowing (as did Mrs Bellfield), the patient will not receive necessary nutrition. Accordingly, it was reasonable for the procedure to be carried out on this occasion.

[9] The procedure was undertaken at 11.00am and lasted about 15 minutes. It was initially performed by Dr Sandeep Siddhi, specialist registrar. However, Mrs Bellfield appeared uncomfortable after the endoscope had been inserted and Dr Carlyle, a hospital practitioner who specialises in endoscopies, took over. After dilatation, the scope passed with difficulty into the oesophagus. Biopsies were taken and a visual examination carried out. No signs of cancer were noticeable (and in fact the biopsies again tested negative for cancer). During the procedure, Mrs Bellfield experienced moderate discomfort, which is not uncommon during dilatation. After the procedure, no difficulties were apparent and Mrs Bellfield was returned to the ward. Dr Carlyle saw her between about 12.00 and 12.30, when she was sitting up in bed and appeared well. Page 34 of Mrs Bellfield's medical records (lodged with the Fife Health Board second inventory of productions) is an accurate record of the procedure and of the observations carried out before and afterwards.

[10] The normal procedure after an endoscopy is to observe the patient for two hours, taking observations of pulse, temperature and blood pressure every half hour. If after two hours the patient is well and not complaining of pain, he or she will be discharged. That procedure was followed with Mrs Bellfield. Her observations throughout the first two hours were normal. However, although the precise chronology of events is unclear, at some stage Mrs Bellfield developed a pain in her lower abdomen which resulted in her discharge being delayed. Staff Nurse Anne Walker, who was her "named" nurse, saw her at about 14.30 and wondered why she was still there, as she would normally have expected her to have been discharged. Nurse Walker spoke to Mrs Bellfield, who complained of pain in her abdomen. At 2.45pm, Nurse Walker took observations and noted that Mrs Bellfield's blood pressure had dropped to 90/45. She was sufficiently concerned to ask Dr Birnie, a consultant endoscopist, who was carrying out his own list that afternoon, to examine Mrs Bellfield. Before that examination was carried out, another nurse, Charge Nurse Helena Davidson, who had previous experience of working in an endoscopy ward, and who happened to be in the ward, offered to look at Mrs Bellfield to provide a second opinion, which she duly did. She, too, noted pain in the lower abdomen. There were no signs of surgical emphysema (which results from air tracking into the subcutaneous tissues from the internal organs as a result of perforation - this process was variously described in evidence as having a similar sensation to bubble wrap when pressed, or rice-crispies). She attributed the pain to Mrs Bellfield having swallowed a lot of air during the procedure or pulling a muscle whilst moving around on the trolley.

[11] Dr Birnie saw Mrs Bellfield at about 4.05pm, having been asked by Mrs Walker to see her. She was complaining of pain in her abdomen and appeared distressed. He carried out a careful examination of her. He examined her abdomen. He was not certain that he could hear bowel sounds, causing him to be suspicious of a perforation of the oesophagus in the abdominal portion. He also checked her skin for signs of surgical emphysema, but could find none. He arranged for a chest x-ray to check for signs of leakage of air. Page 10 of the medical records is Dr Birnie's contemporaneous note of his examination of Mrs Bellfield.

[12] The chest x-ray was duly performed. Dr Birnie examined it for any sign of gas in the diaphragm, which would have indicated a leak from the oesophagus, but he could see none. He sought a second opinion from Dr Mark Jones, consultant radiologist. Dr Jones formed the view that the x-ray showed a hiatus hernia but normal structures, unremarkable lungs and no evidence of leakage. In short, there was no evidence of perforation.

[13] Dr Birnie then saw Mrs Bellfield again, at about 5.00pm. She appeared more settled. On this occasion, he could hear bowel sounds. He decided that it would be in order for her to be discharged. He asked the nurses to get Mrs Bellfield on to her feet and to discharge her, the final decision as to whether to discharge being theirs. Mr Bellfield was present, as he had been since about 2.00pm, but did not raise any concerns with Dr Birnie.

[14] Staff Nurse Walker did indeed discharge Mrs Bellfield, giving her the endoscopy report (medical records, page 11) along with an advice/instruction sheet (Fife Health Board First Inventory, No 6). That is a standard advice sheet, routinely issued to patients when they are discharge following an endoscopy. It includes advice to contact a doctor in the event of severe pain or vomiting, and that at weekends, either the patient's GP or NHS 24 should be contacted. Mrs Walker told Mrs Bellfield that if her discomfort returned she was to contact her GP or present herself at A&E.

[15] It is important to note that throughout the afternoon, with one exception, Mrs Bellfield's observations were all satisfactory. The one exception is the drop in blood pressure recorded at 2.45pm. However the medical evidence given at the Inquiry was uniformly to the effect that a single drop in pressure did not of itself give rise for concern, provided it returned to normal, as Mrs Bellfield's did. Had she been in extreme pain, Mr MacKay confirmed that one would have expected that to have been objectively reflected in increased heart rate, and possibly a reduction in blood pressure and, perhaps, an increase in temperature. None of those symptoms were evident in Mrs Bellfield.

[16] It is clear that Mrs Bellfield was in pain from time to time after her procedure. However, she did not constantly complain of pain and when it was present, its intensity varied. I reach that conclusion having regard to Dr Carlyle's evidence that there were no apparent problems immediately after the procedure and the evidence of Dr Birnie, and Nurses Davidson and Walker, taken together.

[17] At this stage, I must deal with the conflict between the evidence of Dr Birnie, and Nurses Davidson and Walker on the one hand, and that of Mr Bellfield on the other. At the outset, I wish to make it clear that I accept Mr Bellfield as an honest witness, doing his best to give an accurate account of events as he remembered them. However, his recollection was clearly and objectively wrong on a number of counts (for wholly understandable reasons, as on any view, the events of that weekend must have been extremely traumatic for him). For example, he was definite in his evidence that his wife had complained of pain on the Saturday, resulting in the first NHS 24 doctor (Dr Stevens) attending on the Saturday evening, when in fact, as was plain from the contemporaneous record of Dr Stevens' visit, he attended on the Friday, a matter of hours after Mrs Bellfield had been discharged. That is not just a simple error as to date, since it involved a material mis-description of the length of time for which Mrs Bellfield suffered pain before the doctor was called. Likewise, Mr Bellfield could not remember Dr Stevens examining Mrs Bellfield, whereas as the contemporaneous notes also show, (and as Dr Stevens, whom I accepted as credible and reliable, said in evidence), a proper examination was carried out. Mr Bellfield also could not remember any doctors or nurses attending his wife during the course of the afternoon of 13 February, but I do accept the evidence of Nurses Walker and Davidson, and of Dr Birnie, that they did attend on Mrs Bellfield as set out more fully above. In addition, Mrs Bellfield must also have been attended upon regularly when his wife's observations were taken at half-hourly intervals.

[18] The element of Mr Bellfield's evidence on which much of the evidence at the inquiry was predicated was that his wife was in agony throughout the afternoon of 13 February and still in considerable pain when she was discharged. However, none of Dr Bellfield, Nurse Walker and Nurse Davidson observed Mrs Bellfield to be in agony or writhing in pain, as Mr Bellfield stated, and I do prefer their evidence to his on this issue, for the reasons just given. I accepted that all three were properly concerned for Mrs Bellfield's welfare and would not have discharged her had she been in the amount of pain described by Mr Bellfield. It may be that Mr Bellfield's recollection of his wife writhing in agony was of a different time during the course of the weekend, since I have no difficulty in accepting that she was in a great deal of pain at times during the weekend. As I have pointed out above, Mr Bellfield was confused in his evidence as to the day of Dr Stevens' visit. That confusion may also have extended to other events. That Mrs Bellfield was in pain after she returned home is borne out by the fact that not once, but twice, Mr Bellfield called NHS 24 which I do not consider he would have done for no good reason. Or it may be that the truth as to the amount of pain Mrs Bellfield was in on the afternoon of 13 February lies somewhere between the two accounts. It may be, for example, that she was complaining of more pain to her husband than to the medical professionals, and it is not inconceivable that, having been reassured by Dr Birnie that he could detect no problem she underplayed the extent of her pain in the face of her desire to get home. However, that is mere conjecture and I am satisfied that the picture presented objectively to Dr Birnie and to Nurse Walker shortly prior to, and at, the time of her discharge was that she was not at that time in pain, as borne out by the x-ray and the observations of her blood pressure, pulse and temperature. I should add, finally on this matter, that I did not accept that Dr Birnie was uninterested in investigating Mrs Bellfield's complaint of pain, as Mr Bellfield tended to suggest in his evidence. All the evidence was to the contrary effect, in that Dr Birnie broke off from his own list to see Mrs Bellfield following the reporting of concerns by nursing staff, carried out a thorough investigation of which he took a careful note, arranged for further investigations to be carried out and discharged Mrs Bellfield only because the result of that investigation was negative and he was satisfied in his own mind that she appeared to be better. I do not make any criticism whatsoever of Dr Birnie in his care of Mrs Bellfield or in his decision to discharge Mrs Bellfield.

[19] That said, it did not take long for the pain to resurface and Mr Bellfield summoned NHS 24. Dr Howard Stevens, a GP then of some 17 years' experience, attended at 7.15pm. He noted that Mrs Bellfield appeared to be in pain. As the history was of abdominal pain, he concentrated on examining her abdomen. He could not hear bowel sounds. She was able to speak, and to sit up and lie down apparently with little difficulty, which he found reassuring. He was also aware that she had been seen by Dr Birnie, whom he knew to be an experienced gastro-enterologist, that Dr Birnie had had an x-ray performed which had been clear, and that Dr Birnie had deemed her fit to be discharged. He had been reassured by that. He gave consideration to whether she had suffered a perforated stomach. He was then unaware of the potential complications of an oesophageal dilatation. He would have expected the abdomen to be hard and for there to be signs of vomiting. He attributed Mrs Bellfield's pain to the earlier procedure. He advised Mrs Bellfield to take the analgesics prescribed by the hospital and told Mr Bellfield to call back if there was any deterioration. Immediately after his visit, Dr Stevens prepared a note of his examination of Mrs Bellfield, a copy of which forms pages 3 and 4 of the medical records lodged by the Crown (Crown Production No 5).

[20] Mrs Bellfield did not improve over the weekend and Mr Bellfield again contacted NHS 24 on Sunday 15 February. On this occasion, Dr Daniel Smith, a GP then of some two years' experience, attended at about 14.31. His visit lasted half an hour. He considered perforation, but noted that the chest x-ray had shown no signs of it. She was not vomiting, which suggested that she was settling. She was neither writhing nor very distressed. Her temperature and pulse were both normal. He did not specifically check for surgical emphysema but probably would have noted it had it been present in the neck, when percussing the chest (in fact we know from Dr Leitch that there was no surgical emphysema present in the neck). The upper abdomen was mildly tender. Bowel sounds were present. He concluded that there was pain due to oesophageal stretch. He would have admitted Mrs Bellfield had he suspected a perforation. Dr Smith, too, prepared an immediate note of his examination of Mrs Bellfield, a copy of which forms pages 1 and 2 of Crown Production No 5.

[21] Later that evening, Mrs Bellfield collapsed in her house while attempting to go to the toilet. Paramedics were summoned but were unable to revive her. Mrs Bellfield died at 10.08pm that day.

[22] A post mortem was carried out by Dr Derek Leitch. A copy of his report is Crown production No. 2. He established that there was no surgical emphysema in the subcutaneous tissues of the chest. 2 litres of pus was found at the base of the lung, suggesting an acute inflammatory reaction with infection supervening. There was a 3mm perforation of the oesophagus where it entered the stomach. A stricture was present at the lower end of the oesophagus, 5cm from the gastric junction, affecting approximately 2 cm of its length where the oesophageal wall was thickened up to 0.9cm. The oesophagus was heavily diseased by a cancerous growth which would weaken it and make any strain more likely to perforate it. On average pus took 24 hours to form. The perforation could therefore have occurred after Dr Birnie's examination. After reviewing histological findings carried out after his post-mortem, Dr Leitch found the cause of death to be (a) suppurative mediastanitis and pleuritis (b) endoscopic oesophogeal perforation/rupture and (c) oeosphogeal squamous carcinoma.

ISSUES RAISED AT THE INQUIRY

[23] The expert evidence led at the Inquiry focussed on two issues. The first was whether Mrs Bellfield ought not to have been discharged by Dr Birnie on 13 February or whether he ought to have carried out further investigations. The second was the care bestowed upon Mrs Bellfield by the two out-of-hours general practitioners who examined her at home, Drs Stevens and Smith.

[24] There were at least two aspects to the first issue, namely, whether Mrs Bellfield ought to have been discharged when she was in the amount of pain described by Mr Bellfield, and, separately, whether it was reasonable for Dr Birnie to have had a chest x-ray performed as opposed to a gastrogafin swallow, as recommended by Mr MacKay, (who accepted that a CT scan enjoyed the same advantages as a gastrogafin swallow). I have dealt with the first of these topics above, where I have found that Mrs Bellfield was not in fact discharged at a time when she was presenting with the level of pain spoken to by Mr Bellfield. There is nothing more to be said on that issue, beyond the fact that any criticism made of the hospital for not giving due recognition to Mrs Bellfield's symptoms, or allegedly discharging her when she was in agony, is not made out.

[25] The second aspect is more complex. In addition to the evidence already described, the Inquiry also held evidence on this topic from Mr MacKay, Dr Rose and Dr Jones. Mr MacKay and Dr Rose spoke to the terms of their reports (respectively Crown Production No 4 and Fife Health Board Production No 3). Dr Jones gave evidence in relation to his interpretation of the chest x-ray carried out on Mrs Bellfield, but also gave some evidence about the merits of x-rays generally as opposed to swallow tests. I accepted that all were expert in their field, and qualified to give expert opinions to the Inquiry. To a large extent, their views coincided. However, Mr MacKay, who is a surgeon, has the most experience of dealing with oesophageal perforations and where his views differed from those of the others as to the likely time when the perforation occurred, I accepted his. It is clear from the expert evidence as a whole that there is no uniform practice across Scotland as to whether to investigate suspected perforation by means of chest x-ray, gastrogafin swallow or CT scan. At the time of Mrs Bellfield's death, the practice within Fife was to proceed by chest x-ray, at least as the first line of investigation. That was also the practice of Ayr Hospital, according to Dr Rose. The advantages of a chest x-ray as spoken to by Dr Rose and Dr Jones were, in short, that where the hole was small, gas was more likely to escape than was liquid and therefore that a chest x-ray in those case was more likely to detect a perforation than a swallow test. In addition, the patient is exposed to a smaller dose of radiation. As against that, Mr MacKay extolled the advantages of gastrogafin swallow as being that it should detect a hole in a case where air might not be detected as would be the case where there had been a posterior perforation or tear, which might result in air leaking but being contained within the mediastinum (Dr Rose who had apparently not previously considered the point, conceded that might be so); and, in addition, that the gastrogafin swallow test was observed by video camera in real time, whereas an x-ray gave a single snap shot at one point in time. Mr MacKay's view was that there was little point in carrying out a chest x-ray, since, while it can confirm that a perforation has occurred, there is a significant risk of a false negative, in other words, of the x-ray returning a negative result by failing to disclose evidence of air even though a perforation has occurred. The position is complicated by the fact that following a review of procedures carried out after Mrs Bellfield's death, Fife Health Board now routinely carries out a CT scan as the first line of investigation, with or without a contrast swallow. Dr Jones' evidence was that CT scans are the most sensitive of all the tests, that is, they are more likely to detect a perforation or tear than an x-ray. Various statistics were bandied about in evidence but none were based on reliable research. However, all the experts seemed to be in agreement that about one in ten perforations will go undetected by chest x-ray and that CT scans will reduce that number significantly. In addition, according to Dr Jones, a CT scan does not carry certain risks associated with gastrogafin swallow, such as the risk of liver impairment. Mr MacKay, although favouring gastrogafin swallow, agreed, as I have said, that it was reasonable to carry out a CT scan instead.

[26] All the experts were in agreement that no test was perfect and that the crux of the matter was what was described by them as the index of suspicion: ultimately it is a matter of clinical judgement for the doctor treating the patient as to whether further investigation is required or not. That of necessity is based upon a variety of factors including the patient's temperature, blood pressure and pulse; the result of any x-ray, swallow or CT scan; the extent to which pain is present; and critically, the doctor's own examination of the patient. Mr MacKay, whose report suggested that further investigation ought to have been undertaken but who qualified that in his evidence, freely accepted that he had not seen the patient whereas Dr Birnie had, and also that he was not in a position to judge the extent to which Mrs Bellfield was in fact complaining of pain.

[27] Mr MacKay also suggested that the index of suspicion ought to have been raised by the fact that cancer was suspected but that is tempered by the fact that no such diagnosis had in fact been made. It must also be pointed out that, as was acknowledged by all the experts, Mrs Bellfield's symptom of lower abdominal pain was atypical of a thoracic perforation. She did not display any of the other typical signs of a perforation, such as a drop in blood pressure, or an increase in heart rate.

[28] A still further complication is that it is not possible to tell with certainty when the perforation occurred. As will be seen, this is a key factual issue in deciding whether the carrying out of a CT scan might have prevented Mrs Bellfield's death. Dr Leitch put it as no more than a possibility that the endoscopy had caused the perforation. He said that it might have been triggered by a cough or burp. However, Mr MacKay, who has more experience of oesophageal perforations, viewed that opinion as surprising. I accepted his evidence that it was most likely that the perforation was caused due to the "instrumentation going down", as he put it, and that spontaneous perforations were rare. Dr Rose agreed that the endoscopy at least started the process of the rupture, pointing out that the size of the hole approximated to the size of the tip of the endoscope. However, he also postulated that it was possible that the hole had been created but been temporarily sealed by the pressure of the surrounding tissues. Mr MacKay for his part agreed that that was a possibility. The main positive source of evidence that a complete perforation had not occurred by the time of the x-ray was the statement to that effect in Dr Rose's report (Fife Health Board Production No 3), which was based on inference from the fact that no air was detected on the chest x-ray. However, he departed to some extent from that position in evidence when Mr MacKay's alternative explanation - that air had escaped posterially and was not uncontained - was put to him, and which he had not previously considered, which somewhat diluted the value of his opinion on this particular point. When coupled with the evidence of pain, which Mr MacKay said was a significant symptom of a perforation, and having regard to the evidence of the prevalence of false negatives returned by x-rays (in cases of known perforation) I do prefer Mr MacKay's evidence on this whole issue. As I understood his position overall, while he accepted that other explanations were possible, it was likely that the perforation had occurred by the time of the chest x-ray and probably would have been detected by a gastrogafin swallow test (or a CT scan).

[29] If a perforation had been diagnosed, or strongly suspected, there was again general agreement that the appropriate treatment in the first instance would have been to admit Mrs Bellfield to hospital, prescribe "nil by mouth", and administer antibiotics pending a further examination by CT scan or contrast swallow. It is not possible to state that the death would not have occurred had that course been followed but on any view of the evidence, it might have been prevented. Mr MacKay stated that treatment would have significantly increased Mrs Bellfield's chances. The immediate cause of death was sepsis in the mediastinum, and as Mr MacKay said, that ought to have been controlled by antibiotics.

[30] Insofar as the second issue is concerned, the level of care administered by Drs Stevens and Smith, the Inquiry heard from only one independent witness, namely Dr Norman Wallace, who was led as an expert witness by the Crown. He spoke to the terms of his report, Crown Production No 3. Insofar as Mr MacKay expressed an opinion in his written report as to what Drs Stevens and Smith should have done, he conceded in his evidence that he is not himself a general practitioner and therefore not in a position to give an expert opinion on the level of skill to be shown by a GP. Accordingly, when assessing the conduct of the two general practitioners, and considering what level of knowledge they ought to have possessed, I do prefer the evidence of Dr Wallace to that of Mr MacKay, where the two were at odds.

[31] Dealing with Dr Stevens first, Dr Wallace opined that he had taken a careful history and was aware of the endoscopy procedure and of Dr Birnie's examination. He had carried out a competent examination and had given a good safety net, as Dr Wallace put it, by instructing the Bellfields to call again if Mrs Bellfield did not improve. It was not easy for a general practitioner, who might never come across an oesophageal perforation. Dr Wallace himself had only ever seen one, although patients often complained of discomfort or pain after an endoscopy had been performed. Unfortunately, Mrs Bellfield did not have some of the more classical symptoms of a perforation, such as surgical emphysema. It was not unreasonable for a general practitioner to ascribe the symptoms to the procedure itself. For completeness, Mr MacKay said that the procedure should not have causes the level of pain being complained of by Mrs Bellfield, but on the basis of Dr Wallace's evidence, I have to accept, as did Mr MacKay, that that is not something which a general practitioner should be expected to know.

[32] As for Dr Smith, Dr Wallace said that the symptoms should have been settling after 48 hours and it would be right to be concerned, as Mrs Bellfield was obviously still distressed. However the examination was well recorded and competent. All the observations had been normal. With hindsight, the wrong conclusion had been reached but it was a very good assessment.

[33] Dr Wallace further pointed out that both Dr Stevens and Dr Smith had been reassured by the fact that Dr Birnie had carried out an assessment of whether there had been a perforation and had ruled that out. It was not unreasonable for them to have attached considerable weight to that. Although this applied to both doctors, Dr Stevens had more cause to be reassured than Dr Smith, as he had seen Mrs Bellfield only one and a half hours after her discharge. Having said that, Dr Smith had carried out an extensive examination. Dr Wallace candidly admitted that he himself may well have acted as did Drs Stevens and Smith.

SUBMISSIONS

[34] Against the background of the foregoing evidence, parties made submissions as to what findings if any should be made, having regard to the function of the inquiry, as set out above.

[35] Parties were all agreed as to the formal determinations to be made in respect of (a) and (b). Insofar as the other potential determinations were concerned, Ms Yousef for the Crown submitted first that there were reasonable precautions which might have prevented Mrs Bellfield's death. She might have been kept in overnight, or a CT scan or contrast swallow might have been carried out. Had either step been taken, life-saving treatment could then have been given. As for Dr Stevens, alarm bells should have started ringing to the effect that she was suffering from something more than general discomfort and he should not have been as reassured as he was by the fact of Dr Birnie's examination. A reasonable precaution which he could have taken was to re-admit Mrs Bellfield. Similarly, Dr Smith should have readmitted Mrs Bellfield to hospital, given that she was complaining of pain some 48 hours after the procedure. As for whether there had been defects in any system, Mrs Youself reiterated that a CT scan or contrast swallow should have been carried out rather than a chest X-ray. Turning to section 6(1)(e), Ms Yousef submitted that there were facts relevant to the circumstances of the death. These were:

The lack of training or refresher training offered to gastro-enterologists;

The lack of training of General Practitioners in the recognition of post-operative complications following oesophageal procedures;

A lack of discharge advice;

The lack of cognisance of Mr Bellfield's concerns at the hospital;

CT scans and contrast swallows are now more readily done.

[36] Mr Stewart, for Dr Birnie, submitted that I should make no findings in terms of section 6 (c), (d), or (e). He submitted that there were no reasonable precautions which Dr Birnie could have taken which might have prevented Mrs Bellfield's death. Dr Birnie had acted reasonably in deciding not to admit Mrs Bellfield overnight. That was demonstrated by Dr Rose's evidence and, ultimately, also that of Mr MacKay. Further, there was no certainty that Mrs Bellfield's life could have been saved even had the perforation been recognised. There had to be a "real and lively possibility" that the death could have been avoided, for any finding to be made under section 6(c). As for (d), this was not truly relevant, but in any event Dr Birnie had followed established hospital procedure. Finally, there were no facts relevant to the circumstances of the death.

[37] Ms Ritchie, for Drs Stevens and Smith, also submitted that I should make no findings in terms of section 6(c) or (d). She submitted that those doctors, too, had acted reasonably and could not be criticised. It was not appropriate to make any finding regarding a precautionary measure which might have prevented the death where doctors had been exercising clinical judgment. Where a doctor was faced with more than one reasonable option, and had to choose between or among them, in choosing one it was inevitable that alternatives would be disregarded. Provided their actions were in all the circumstances reasonable, the omission to take an alternative course could not properly be regarded as a failure to take a reasonable precaution. Moreover, there had been a stark absence of any evidence to show that the death might have been avoided had Mrs Bellfield been admitted to hospital by either GP. There had been no evidence as to what treatment she would have received in A & E or what level of expertise would have been possessed by the doctors who might then have treated her. As far as Dr Smith was concerned, there was evidence from Dr Wallace that sadly by the time he saw her, it was probably too late for any action to have been taken which would have saved her life. Finally, any finding in relation to section 6(e) should be confined to findings in relation to the carcinoma of the oesophagus.

[38] Mr Fitzpatrick, for Fife Health Board, aligned himself to the submissions of Mr Stewart and Ms Ritchie. He pointed out that the function of the Inquiry was not to apportion blame or to consider whether any person had been negligent: Black v Scott 1990 SLT 612, per Lord President Hope at p 615. The court should be cautious of drawing sweeping conclusions from evidence which may be incomplete. He submitted that a precaution was a measure designed to address a foreseeable risk; and a reasonable precaution was one which was fairly to be expected or called for in the circumstances. The court should specify as a reasonable precaution only any step which it was unreasonable to omit. He acknowledged that there was some judicial support for the contrary view that hindsight could be taken into account: Sharman Weir FAI¸ per Sheriff Reith. However, it would make little sense to do so in cases involving the exercise of clinical judgment, where hindsight was never available and where the clinician must always approach the matter prospectively. While the court might well take the view that a contrast swallow (or CT scan or overnight admission) might have saved Mrs Bellfield's life, it would not be helpful to anyone for a finding to be made that any of these actions was a reasonable precaution which might have avoided the death. None of those actions would have been unreasonable but the difficulty for the clinicians was in assessing the evidence to decide whether any of them was the appropriate thing to do. A section 6(1)(c) determination would appear to say that one or more of those actions should have been taken. A more helpful approach would be to make a recommendation in terms of section 6(1)(e). As far as section 6(1)(d) was concerned, that provision seemed most appropriately applicable to deaths in the workplace, although the provision was not expressly so confined. Before any finding could be made, the court would require to be satisfied that any defect in any system of working probably did contribute to the death. Finally, as for section 6(1)(e), any recommendation for the future could be made under that provision.

DISCUSSION

[39] In considering the submissions, and the approach to be taken in considering what findings to make, it is necessary to bear in mind the function of this Inquiry. As has often been said, it is not the function of an inquiry under the Fatal Accidents and Sudden Deaths Inquiry (Scotland) Act 1976 to consider fault or to attribute blame. Rather, the purpose of such an inquiry is for the Sheriff to make one or more determination on those circumstances of the death which are set out in section 6(1) of the Act, referred to above. Moreover, as was stated by Sheriff Reith in her determination following a Fatal Accident Inquiry into the death of Sharman Weir, with whose opinion I respectfully agree:

"...a Fatal Accident Inquiry is very much an exercise in applying the wisdom of hindsight. It is for the sheriff to identify the reasonable precautions, if any, whereby the death might have been avoided. The sheriff is required to proceed on the basis of the evidence adduced without any regard to the state of knowledge at the time of the death. The statutory provisions are concerned with the existence of reasonable precautions at the time of the death and are not concerned with whether they could or should have been recognised. They do not relate to the question of foreseeability of risk which would be a concept relevant in the context of a fault-finding exercise, which this is not. The statutory provisions are widely drawn and are intended to permit retrospective consideration of matters with the benefit of hindsight and on the basis of information and evidence available at the time of the Inquiry. There is no question of the reasonableness of any precaution depending on the foreseeability of risk. In my opinion the question of reasonableness relates to the question of availability and suitability of the precautions involved....The purpose of any conclusions drawn is to assist those legitimately interested in the circumstances of the death to look to the future. They, armed with the benefit of hindsight, the evidence led at the Inquiry and the Determination of the Inquiry, may be persuaded to take steps to prevent any recurrence of such a death in the future."

[40] In considering whether to find, in terms of section 6(1)(c) of the Act, that there were reasonable precautions whereby Mrs Bellfield's death might have been avoided, it is necessary to decide what is meant by reasonable precautions, and whether a precaution may be considered to be reasonable only if the need for it could have been anticipated at the time of the events leading to the death. The latter task is not made any easier by the divergence of judicial opinion on the issue. Since a fatal accident inquiry is not concerned with questions of negligence or fault, and there is no need to consider whether an accident or death was foreseeable, then it seems to me that there is no reason in principle why, in certain circumstances, a section 6(1)(c) finding should not be made in relation to a precaution which should not have been foreseen as necessary, particularly when the finding does not carry any connotation that the failure to take the precaution was negligent. Accordingly, I consider that Mr Fitzpatrick's submission, that a precaution is a measure designed to address a known or foreseeable risk, was overstating the position. I can conceive that there will be some situations where a risk which was not foreseeable could have been prevented by the taking of a reasonable precaution, which should be taken in future. In such situations, it can easily be seen that there is some purpose to the making of a finding under section 6(1)(c), in order that lessons might be learnt so that future deaths might be avoided in similar circumstances.

[41] However, that is not to say that every single thing which might have been done and which might have avoided the death should, if it was a reasonable step to have taken, make its way into a finding under section 6(1)(c). Not only would that not be helpful in avoiding future deaths, but it would involve placing an unjustifiably wide construction on the word "precaution". Whatever that word means, it must place some limit on the sort of acts or events which should be included in a 6(1)(c) finding. The natural meaning of "precaution" is an action or measure taken beforehand against a possible danger or risk. Further, since one purpose of a fatal accident inquiry is to inform those with an interest of what precautions should be taken in future, a finding under section 6(1)(c) must carry with it the implication that the precaution ought, with the benefit of hindsight, to have been taken in the case which resulted in the death, albeit without any necessary implication that the failure to take it was negligent. That being so, I agree that when one has a situation which solely involves the exercise of clinical judgment, where a range of reasonable actions might be taken, and the choice as to which to take rests on the skill and experience of a doctor based upon such information as is available to him at the time, and the doctor happens to choose a course which results in death, it would be wrong to hold that the selection of another option within the range, which might have prevented the death, was a reasonable precaution which ought to have been taken. Not only does that involve straining the meaning of precaution, but such a finding would be of no real practical benefit to others in the future. A Fatal Accident Inquiry cannot prescribe how doctors or nurses should exercise their judgment. Put another way, the true precaution which ought to be taken in any given case may simply be a requirement that a patient is seen by a suitably skilled doctor, rather than how the doctor exercises his skill and judgment thereafter.

[42] For this reason, I am not prepared to hold that it would have been a reasonable precaution for either Dr Stevens or Dr Smith to have admitted Mrs Bellfield to hospital, whether or not such a step might have avoided the death (and at least in the case of Dr Smith it is likely that it would not; in the case of Dr Stevens, I find it unnecessary to reach a concluded view on Ms Ritchie's submission that there was a lack of evidence to make that finding). Having regard to the evidence of Dr Wallace, the only independent GP to give evidence, those doctors' respective assessments of Mrs Bellfield were reasonable, and it would not be reasonable for all patients who complained of pain to be readmitted to hospital. Dr Wallace also stated that general practitioners could not be expected to recognise an oesophageal perforation as they happen so rarely, and while to some extent that may cause one to question the value of advising a patient who has undergone an oesophageal dilatation to contact his or her general practitioner in the event of pain, I did not hear sufficient evidence to entitle me to criticise that advice or to recommend that it be changed in some way. However, I do consider that it was a circumstance relevant to the death that, during a period of some 48 hours throughout which she was experiencing severe pain, Mrs Bellfield was attended upon by two general practitioners who did not have, nor could be expected to have, knowledge or experience of oesophageal perforations such as to enable them to make a correct diagnosis, and I have made a finding to that effect in terms of section 6(1)(e).

[43] For the same reason, I am not prepared to hold that it would have been a reasonable precaution for Dr Birnie to have instructed further investigation of Mrs Bellfield after examining the chest x-ray which he had instructed, whether by having her re-admitted or otherwise. His assessment of her at that time was reasonable and cannot be criticised. Further, he took the precaution of consulting with Dr Jones, a consultant radiologist, in case he had missed something on the x-ray, and Dr Jones, who himself has considerable experience and who has looked at the x-ray many times since, confirmed that there was, quite simply, no sign of any leakage of gas. By that time, Mrs Bellfield appeared to be better than she had been, and Dr Birnie had no reason for his "index of suspicion" to be at such a level that he ought to have admitted her to the hospital overnight. Although Mr MacKay suggested otherwise in his report his view was considerably watered down in evidence, when he accepted that his report was predicated on his understanding of the level of pain manifested by Mrs Bellfield, both during and after the procedure, and as a matter of fact, that level of pain was not established at the Inquiry, for the reasons given above. Accordingly, I am satisfied that what Dr Birnie did following his instruction of the chest x-ray was reasonable, and there was nothing else which he ought to have done thereafter. Moreover, in instructing a chest x-ray in the first place, as the first line of investigation, he was not only acting in accordance with established practice at the Victoria Hospital at that time but the practice itself was a reasonable one.

[44] However, I have more difficulty in resolving the question as to whether the taking of a CT scan would have been a reasonable precaution to have taken, and if so whether it might have prevented Mrs Bellfield's death. The decision to take a chest x-ray was not so much an exercise of clinical judgment, but rather a case of Dr Birnie simply following the usual practice in force in Fife at that time, which practice, as I have already pointed out, has since been changed. There is also no getting away from the fact that everything that happened after the chest x-ray had been performed was, at least to some extent, attributable to the fact that the x-ray was clear. Dr Birnie's own concerns as to whether there had been a perforation were allayed. That was later compounded by Dr Steven's reliance on Dr Birnie's having been reassured that there was no perforation when he might otherwise have had Mrs Bellfield re-admitted; and further compounded by Dr Smith's placing similar reliance on Dr Birnie's examination. With the benefit of hindsight, this seems to me to involve the placing of a considerable, and unjustifiable, amount of reliance, by a number of different doctors, at different times, on a test which is not only known to result in false negatives in a not insignificant number of cases, but is also known to be less sensitive than a CT scan.

[45] I consider that the use of a CT scan as the first line of investigation, in cases where a perforation is suspected, so as better to inform the clinician's judgment in assessing the patient thereafter, can properly be regarded as a precaution. (I am focussing on CT scans rather than contrast swallow tests, simply because Fife Health Board now routinely uses the former, and the question is therefore whether this is a precaution which could reasonably have been taken sooner.) I have already expressed the view that the risk against which the precaution is taken does not necessarily require to be foreseeable. However, in the present case, I would point out that the risk of a perforation having occurred was foreseeable, and it was also known that the risks were higher when dilatation had been performed, as here, and where cancer was (or might) be present, as here. It was also known that if a perforation had occurred and went undetected, there was a considerable risk to Mrs Bellfield's life. The taking of a CT scan to detect whether there had been a perforation therefore can properly be regarded as a measure to guard against a known risk.

[46] The next question which arises is whether it was in this case a reasonable precaution. In deciding that question, I must deal with the submissions presented to me to the effect that it would be open to me to find that a CT scan was reasonable only if I reached the view that what was done was unreasonable. With respect, I do not consider that to be correct. I have already pointed out that negligence is not in issue and that it is not the function of this inquiry to attribute blame. It is therefore nothing to the point to inquire as to whether what was done was reasonable, and it seems to me to involve a non sequitur to hold that a precaution which was not taken can be held to have been reasonable only if what was done was not reasonable. To take that approach respectfully seems to me to apply the principles and language of negligence, which are irrelevant for the purposes of this inquiry. I do not see why it is not open to me to hold that, even though what was done was reasonable, other reasonable precautions might also have been taken which might have prevented the death.

[47] In considering the question of reasonableness, it is relevant to note from the evidence that although the occurrence of oesophageal perforation is rare, the consequences of its going undetected carry a high mortality rate. Dr Rose gave the example of the rabid bat: the bat probably does not have rabies, but if it does, any bite will be fatal, and so one should guard against the possibility. Mr MacKay and Dr Rose agreed that one in ten chest x-rays will fail to detect a perforation which has occurred. That seems to me to give rise to a sufficiently large number of patients with a potentially fatal (if untreated) condition that it is better to err on the side of caution by carrying out as sensitive a test as possible as the first, rather than the second, line of investigation. This is the approach taken in Glasgow. Of all the experts who gave evidence, Mr MacKay was the one who had the most experience of oesophageal perforations and I accept his evidence that there is little point in carrying out a chest x-ray in the first instance. That the carrying out of a CT scan is a reasonable precaution is in my view reinforced by the fact that those involved in primary care, general practitioners, are unlikely to have the necessary knowledge and experience to recognise a perforation, if the patient is discharged - in other words, there is little prospect of a perforation which has gone undetected by a chest x-ray subsequently being diagnosed by a GP with little or no experience of oesophageal perforations.

[48] It is also relevant in considering reasonableness that, since Mrs Bellfield's death, all endoscopy units in NHS Fife have in fact changed their procedures. The guidance issued by the Endoscopy Users Group as at August 2010 (Fife Health Board Production No 4) points out the risks of perforation where dilatation or stent insertion has occurred, and that perforation although rare is associated with significant morbidity and mortality. Various measures are recommended following such procedures, including the following:

"If on medical assessment there is any likelihood of perforation then the patient should be discussed with a radiologist and urgent CT scan of the chest and abdomen should be arranged. The radiologist may or may not decide that contrast is required".

It can be seen that not every patient should receive a CT scan, simply those in respect of whom a medical opinion has been reached that there is "any likelihood" of perforation. Mrs Bellfield of course fell into that category since Dr Birnie had expressly considered that possibility when he examined her at 16.05 hours on 13 February 2009.

[49] For all these reasons, I find that the taking of a CT scan of Mrs Bellfield's chest and abdomen was a reasonable precaution which could, and with the benefit and wisdom of hindsight, should have been taken once Dr Birnie had formed the opinion that an oesophageal perforation had possibly occurred.

[50] The final question to consider in this part of my determination is whether a CT scan might have prevented the death. As all parties accepted, it cannot be said that it would have prevented Mrs Bellfield's death. Mr MacKay stated that the outcome might have been no different, even if the perforation had been detected. However, the test in considering whether to make a section 6(1)(c) determination is not one of probability but of possibility: the question is whether the death might have been prevented. Parties were in agreement that the test was whether there was a "real and lively possibility" that the death might have been avoided, making reference to a test which has been used before. That so-called test has no statutory basis. It has also been said that there must have been a realistic possibility that the death might have been prevented. Whatever precise meaning is given to the words, it is generally accepted that there must be something more than mere speculation that the death might have been avoided, before a finding under section 6(1)(c) can be made.

[51] The issue of whether Mrs Bellfield's death might have been prevented by the carrying out of a CT scan must be analysed in two stages: would the scan have disclosed the perforation, and what would have happened thereafter if it had? As regards the first stage, I have already referred to the evidence that there was a possibility that the perforation might either not have occurred by the time of the x-ray, or that it had occurred but been temporarily sealed, and in either event, even a CT scan would have shown nothing that the x-ray did not, and the death would still have occurred. However, for the reasons which I give in paragraph 28 above, I have found that it is probable that the perforation had occurred by the time the chest x-ray had been carried out. Even if I am wrong in that, there is at the very least a real and lively possibility that the perforation had occurred, and was discoverable. Having regard to the evidence as to the merits of a CT scan over a chest x-ray, and to Mr MacKay's evidence in particular, I further consider that if the perforation had been present, then it would probably have been detected by a CT scan. Turning then to the question of what would have happened thereafter, I have no difficulty in holding that if the perforation had been detected on the afternoon of Friday 13 February, prior to Mrs Bellfield's discharge from hospital, there was a real and lively possibility, or a realistic possibility, of her death having been prevented. As Mr MacKay said, we still don't know what would have happened, and the outcome might still have been no different. But we do know that the appropriate treatment would have been administered: nil by mouth and a course of antibiotics. As I have found above at paragraph 29, that would have significantly increased Mrs Bellfield's chances of survival. In short, the whole of the foregoing evidence is sufficient for me to find that had a CT scan been carried out, Mrs Bellfield's death might have been prevented.

[52] For all these reasons, I have found that the taking of a CT scan was a reasonable precaution which might have prevented Mrs Bellfield's death. In making that finding, I again wish to stress that I have been able to do so only because at this Inquiry I am entitled, and indeed bound, to inquire into what happened with the benefit of hindsight, so that in the future others may learn from what happened here; and I think it is not insignificant that Fife Health Board had itself, by the time of the Inquiry, already changed its procedures in the light of Mrs Bellfield's tragic death. I am not imputing any blame whatsoever to Dr Birnie, the exercise of whose judgment in light of the chest x-ray cannot be criticised. Nonetheless, it was established on the evidence that a clinician's judgment is more likely to be better informed by a CT scan (or gastric swallow) than by a chest x-ray, and it is appropriate that my findings reflect that.

[53] On the question of whether there were any defects in any system of working which were relevant to Mrs Bellfield's death, I do wonder about the value of a post-discharge system which effectively transfers out-of-hours post-operative care to general practitioners who do not have and cannot be expected to have sufficient knowledge and experience to recognise a perforation. However, on the basis of the evidence led, I do not feel able to make any finding that there was a defect in any system of working.

[54] I consider that there were other circumstances which were relevant to the death, in addition to that already mentioned at paragraph 42. First and foremost, the fact that cancer was present in the oesophagus, causing it to weaken and to be more susceptible to perforating, was clearly a relevant circumstance. However, I accept the evidence of Dr Vaidya and others that cancer in the oesophagus can be - and in this instance, was - difficult to diagnose, and it is telling that even the biopsies taken on 13 February 2009 tested negative for malignancy. Dr Vaidya had a suspicion of cancer but the endoscopy on 13 February was a necessary and reasonable step to try to confirm his suspicion. Nothing else could have been done, either to diagnose the cancer earlier, or indeed to avoid the perforation occurring. For the avoidance of doubt, there was no suggestion at any time, nor any evidence whatsoever, that the endoscopy itself was carried out with anything other than due skill and care.

[55] Next, the fact that Mrs Bellfield had symptoms atypical of a thoracic perforation was also a relevant circumstance, inasmuch as it made the perforation more difficult to diagnose. Had she displayed more of the typical symptoms (other than pain) then it is more likely that she would have been admitted to a ward and received the recommended treatment. There was no real medical explanation as to why her symptoms were atypical, beyond the fact that pain on occasion appears somewhere other than the site of the perforation.

[55] I am not prepared to make any of the other findings proposed by the Crown. In particular, there was insufficient evidence to enable me to make considered recommendations, which could have far-reaching implications, for the training of the medical profession, and I could not sensibly make any proposals for training of general practitioners or gastro-enterologists. I have given consideration as to whether it would be appropriate to make any recommendations regarding the use of CT scans or contrast swallows, but, since that depends at least to some extent on the resources available to health boards, I have come to the view that I should not make any formal recommendations. That said, I hope that the tragic circumstances of Mrs Bellfield's death, which, as I have found, might have been avoided by the use of a CT scan, will be taken into account by others involved in endoscopies, in deciding how suspected perforations should best be investigated in future. As for the discharge advice, I consider that adequate advice was given to Mrs Bellfield. I did accept the evidence of Nurse Walker that Mrs Bellfield was handed the report of her procedure along with the standard discharge advice, and the note taken by Dr Smith in particular, although not conclusive, would tend to suggest that he had sight of the former. In any event, the point is in a sense academic given that Mr Bellfield did in fact call NHS 24 to attend to his wife.

[56] Finally, I wish to thank parties' legal representatives for the manner in which the Inquiry was conducted and for their helpful submissions. I also again extend my condolences to Mr Bellfield and the other family members. Although they did not play a formal part in the inquiry, they attended all the proceedings and conducted themselves with quiet dignity throughout.