England and Wales High Court (Queen's Bench Division) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION

		Royal Courts of Justice Strand, London, WC2A 2LL
		26/02/2008

The Honourable Mr. Justice Saunders:

1. This claim arises out of brain damage which it is said the Claimant (Marianna) suffered as a result of her treatment at Bristol Children's Hospital soon after her birth in 1985. She says that the Defendant was negligent in the way she was treated.
2. The Defendant denies negligence and also denies that the Claimant suffered damage as a consequence of her treatment.
3. The Claimant identifies three different times she says she was treated negligently:

1) The operation conducted by Mr. Dhasmana, a Senior Registrar employed by the Defendant, on 8^th May 1985.

2) Her care between the first operation and a second operation on 12^th May 1985.

3) The third operation carried out by Mr. Wisheart, a Consultant Surgeon employed by the Defendant, on 24^th March 1986.

4. The Claimant asserts that damage was caused by each negligent act, although it is accepted that only a small amount of damage can have been caused by 3.3) above. That Marianna has suffered brain damage is not in issue.
5. The Agreed Facts. Marianna was born on 5.5.85 at Royal Gwent Hospital at 1.24a.m. Her condition at birth was satisfactory. At 1a.m. on 6.5.85, doctors were alerted to a possible problem because Marianna was observed by nurses to be blue on feeding and crying. A test was carried out to see whether the cyanosis (blue colour) was caused by a failure in breathing or a heart defect. This test, a hyperoxia test, confirmed the cause as a heart defect. A blood analysis was done at Royal Gwent at 9.49a.m. (F6 p.643) which revealed low levels of oxygen saturation in the blood and a PH reading of 7.163. These readings indicated that Marianna had hypoxaemia (insufficient oxygen in the blood) and metabolic acidosis (a high level of acid in body fluids and tissues). Presence of metabolic acidosis may be an indication that hypoxia is severe. She was described as being 'somewhat jittery' which was a warning sign that there may be some neurological insult.
6. Marianna was transferred to Bristol Children's Hospital, as the regional centre for cardiac surgery, for treatment. On admission she was seen by Dr. Jordan, one of two Consultant Cardiologists who cared for both adult and child patients in Bristol; the other was Dr. Joffe.
7. Dr. Jordan examined Marianna; he carried out further tests and made a diagnosis of pulmonary valve atresia with intact intraventricular septum. The Claimant accepts that Dr. Jordan's diagnosis was correct. The Defendant, although Dr. Jordan was its witness, asserts that the correct diagnosis was pulmonary stenosis with intact intraventricular septum. That assertion is made on the basis of a finding by Mr. Dhasmana during the first operation that some blood was passing through the pulmonary valve. If Dr. Jordan was correct in his diagnosis, no blood was flowing through the pulmonary valve (atresia) if Mr. Dhasmana was correct in his record of the operation; some blood was flowing through the pulmonary valve although it was blocked to a significant degree (stenosis).
8. Whether it was a complete or partial blockage of the valve, the effect was that no or insufficient blood was passing through the valve into the lungs. As a result insufficient oxygen was getting into the blood and therefore to the brain. This is potentially a fatal condition.
9. Although surgery was required, it did not have to be carried out immediately, because Marianna was benefitting from a blood vessel called the ductus arteriosus (ductus) which carries blood from the pulmonary artery to the aorta by-passing the lungs. The foetus needs the ductus pre-birth as the lungs are not then working. After birth, if a baby's heart is functioning normally, the ductus is no longer required and it closes off. It is possible to keep the ductus open, and in some cases even re-open it, by introducing prostaglandin into the body. As a result of his diagnosis at 12.30p.m. on 6.5.85 Dr. Jordan began an infusion of prostaglandin into Marianna.
10. The effect of maintaining a flow of blood through the ductus is to keep the oxygen saturation of the blood at an acceptable level. It is not necessary for the purposes of this judgment to seek to explain the mechanism by which this occurs but treatment with prostaglandin can only produce a short term solution for a number of reasons:

1) It does not ensure a 'normal' oxygen saturation level in the blood.

2) The ductus can close and cease to function even though prostaglandin is being infused, and the consequence of that could be fatal.

3) There are some harmful effects of the use of prostaglandin including a raised pulse.

11. While Professor Kirkham, the Paediatric Neurologist expressed the view that the prostaglandin infusion should have started at the Royal Gwent Hospital, no-one doubts the correctness of the management plan put in place by Dr. Jordan.
12. Having started prostaglandin to improve the oxygen saturation in the blood, an operation to insert a shunt was needed to connect the left subclavian artery to the left pulmonary artery, thus by-passing the pulmonary valve which was either completely or substantially closed. This would remove the necessity to keep the ductus open. The shunt also could only be a temporary remedy, and a valvotomy would be needed to repair the damaged valve and produce a normal flow of blood. A further operation may also be required at a later stage but that is not material to this case.
13. The infusion of prostaglandin had a beneficial effect in improving Marianna's colour and the first operation took place on 8th May between 2p.m. and 7.30p.m. The operation was carried out by Mr. Dhasmana who was the Senior Surgical Registrar attached to Mr. Wisheart. Between them they carried out all the cardio-thoracic operations for adults and children at the Royal Bristol Infirmary; both open heart and closed operations.
14. The way Mr. Dhasmana conducted that operation is said by the Claimant to be negligent and these issues will be discussed later. What is agreed is that Mr. Dhasmana did connect the left subclavian artery to the left pulmonary artery by means of a 5mm. Gore-Tex tube. Having done that, he did not carry out the proposed second part of the operation, the valvotomy, because he detected that some blood was passing through the pulmonary valve. In these circumstances the repair could not be carried out by a closed operation but would need open heart surgery. After discussion with Dr. Jordan, he decided not to proceed further.
15. At 8.00a.m. on 9^th May, Marianna was seen by Mr. Dhasmana. The note in the nursing records reads (F6 p.647) 'Bloods taken for U + ES, FBC + gases remains slightly cyanosed, S/B Mr. Dhasmarna (sic), to stop Prostin (prostaglandin) and observe, arterial line to be removed.' The prostaglandin infusion was not re-commenced until 11.00a.m. on 12^th May and the arterial line was not re-inserted.
16. The likely effect of stopping the prostaglandin would be to allow the ductus to close. The closure would reduce the oxygen saturation in the blood unless the introduction of the shunt resulted in a compensatory rise. The prostaglandin needed to be switched off so that proper assessment could be made as to whether the shunt was working.
17. The consequence of removing the arterial line was that samples of arterial blood could not be taken easily from Marianna. After the removal, the samples which were taken for blood/gas analysis were of venous or capillary blood. Previously, samples of arterial blood had been taken. So it was no longer possible to compare like with like and, further, arterial blood samples give the most accurate readings for blood/oxygen saturations because none of the oxygen has been used up by the body whereas in the other two samples it has to varying degrees.
18. There is no evidence that Mr. Dhasmana had any further contact with Marianna. When Mr. Dhasmana operated on 8^th May, Mr. Wisheart had been away in Edinburgh. Mr. Wisheart did a ward round at the Children's Hospital on the 10^th and from then on it is likely he became involved in Marianna's care.
19. Her general care was the responsibility of Dr. Jordan and, if he was not there, Dr. Joffe. There were regular ward rounds supervised by one or other Consultant accompanied by their team. Marianna's condition was monitored. The matter I will have to consider further is whether the right decisions were made, in particular whether the decision to operate a second time should have been made sooner than it was.
20. The clinical notes reveal that by the time of the evening ward round on 11^th May, consideration was being given to whether to perform a second shunt operation and a valvotomy because of the poor oxygen saturation levels in the blood. On 12^th May Dr. Jordan directed that the prostaglandin infusion should be re-started. It was re-started at 12 mid-day. At 3.30p.m., a cardiac catheter was carried out, the purpose of which was to ascertain whether the shunt was working. It revealed that the shunt was patent i.e. blood was passing through it but it also revealed that the shunt was kinked which inevitably restricted that flow (see discharge summary F 6 p.551).
21. On the evening of 12^th May as a result of these findings Mr. Wisheart carried out a right sided shunt operation. This operation was successful and immediately the oxygen saturation levels in the blood rose. No-one contends that the actions taken by the clinicians and surgeons on 12^th May were not correct. The Claimant's argument is that in failing to take these steps earlier, i.e. re-start prostaglandin, investigate the problem by means of a cardiac catheter and/or carry out a further shunt operation, they were negligent. In particular the Claimant says these steps should have been taken on the afternoon of 9^th May.
22. On 24^th March 1986, when Marianna was 10 months old, Mr. Wisheart carried out a further operation. The operation note is at F5 p.190. The description of the operation given there is 'Reconstruction of the right ventricular outflow tract by excision of the atretic pulmonary valve and insertion of a Dacron patch'. The purpose of this operation was to repair the pulmonary valve which was not allowing blood to pass into the lungs. It was that defect which had brought about the need for a shunt. This operation was open heart surgery and involved by-passing the heart. The operation was successful in achieving its objective and the repair was carried out. It is however accepted that Marianna sustained a neurological insult in the course of open heart surgery. That is not unusual, but the Claimant asserts that I should conclude that it was caused by an air embolism which could only have occurred if Mr. Wisheart was negligent in the manner in which he carried out this operation. The Defendant denies that any negligence has been demonstrated and even if it had, contends that it has not been demonstrated that the neurological insult caused lasting damage.
23. It is unnecessary for me to relate the history of Marianna's treatment further as I have covered the period during which the doctors are said to have been negligent. I am only dealing with liability so it is unnecessary to deal in detail with the damage suffered by Marianna save to say that it is clearly serious. Whether and to what extent it was caused by the actions of the Defendant will be dealt with later in the judgment.
24. I now turn to the specific allegations of negligence. The fact that these events happened in 1985 is relevant for two reasons:

1) Because the witnesses of fact, and in particular Dr. Jordan and Mr. Wisheart, cannot be expected to have a clear recollection of what went on. Their evidence has inevitably been a reconstruction from what they have seen in their notes and documentation. It would be unrealistic to expect them to have an independent recollection except in the most general terms. My approach to their evidence will reflect that.

2) I must judge the Doctors by the standards of normal practice and the state of medical knowledge in 1985. This is an area where significant research has been done since 1985 which informs the way Doctors treat neonates with congenital heart defects today. Although that later knowledge is relevant to questions of causation, it is not relevant when I decide whether the Doctors exercised reasonable care in 1985 which must be judged on what they knew at the time.

25. There was an Inquiry into children's heart surgery at the Bristol Royal Infirmary 1984-1995. The results of that Inquiry were published in July 2001 in a document entitled 'Learning from Bristol'.
26. I have been supplied with excerpts from that Inquiry (F. 3). The report makes it clear that it was not concerned with liability for any damage suffered by any child treated at Bristol. That was a matter for the Courts. In the event, the report has played a minimal part in this hearing. In closing submissions neither Counsel has referred me to any part of them and the report and its findings have played no part in the conclusions to which I have come.
27. The First Operation. There are 3 allegations of negligence in the Particulars of Claim relating to the first 'shunt' operation carried out by Mr. Dhasmana:

1) Caused or permitted a Senior Registrar (Mr. Dhasmana) to perform the first operation, which was technically demanding, alone and unsupervised;

2) Performed the operation negligently in allowing the shunt to be made too long, with the result that it kinked;

3) Failed to perform the planned valvotomy, due to technical incompetence and lack of assistance/supervision.

28. The trial on this aspect has centred on allegation (2). Although they were investigated, I think it is fair to say that by the end of the case, (1) and (3) were no longer vigorously pursued and in my judgment for good reasons.
29. The Claimant's case on (1) is based on the expert opinion of Mr. Pozzi. Mr. Pozzi is a distinguished Paediatric Cardiac Surgeon who presently holds the position of Senior Consultant Paediatric Cardiac Surgeon at Alder Hey Hospital in Liverpool and Clinical Lecturer at Liverpool University. His full CV is at F4 div.1. His opinion as set out in para. 40 of his report (F1 p.192) is that, while it was acceptable for a Senior Registrar to carry out a shunt operation, he should not carry out a valvotomy unsupervised because of the technically demanding nature of this operation. Mr. Dhasmana was unsupervised at the time of the operation because Mr. Wisheart was away in Edinburgh.
30. As events turned out, Mr. Dhasmana only carried out the shunt insertion part of the operation because, rightly or wrongly, he formed the view that the valve was not completely closed off and some blood was passing through. In these circumstances the valvotomy was not performed. If Mr. Pozzi is correct, that part of the operation should have awaited the return of Mr. Wisheart in any event. As it happens Mr. Dhasmana did not carry out an unsupervised valvotomy. It is therefore difficult to see how any damage could flow from this allegation of negligence even if it were to be made out.
31. Mr. Wisheart says that it was his decision to allow Mr. Dhasmana to carry out the operation. He did so because he believed him to be competent to do so. Mr. Dhasmana had had a considerable amount of experience and had spent time working and training at the University of Alabama and at Great Ormond Street with the leaders in the field of paediatric cardiac surgery. He became a Consultant the following January.
32. Mr. Stark is the expert Paediatric Cardio-thoracic Surgeon for the Defendant. His experience and expertise in this field is close to being unrivalled. He was a Consultant Cardio-thoracic Surgeon at Great Ormond Street Hospital from 1971 to 1999. He was a pioneer in the operations that I have been considering. He has published widely on the subject and is Co-Editor of one of the standard text books on surgery for congenital heart defects. His full CV is to be found at F4 div.2.
33. Clearly his opinions must command a great deal of respect although his distinguished career does not mean that his opinions must always be right. It is always necessary to examine the reasons behind any opinion.
34. Mr. Stark's view was that, with the training and experience he had had, Mr. Dhasmana was qualified to carry out both valvotomy and shunt operations.
35. While I do not criticise Mr. Pozzi for having an invariable rule in his own practice of not allowing Senior Registrars to carry out valvotomies, I do not consider it is negligent to allow a Senior Registrar to carry out such an operation provided he has the necessary experience, training and ability. No doubt some Senior Registrars are more skilled than some Consultants. In this case I am satisfied that Mr. Wisheart properly considered whether Mr. Dhasmana had the necessary experience, training and ability, and I am satisfied that he was entitled to reach the conclusion that he did.
36. It must follow from that, that particular (3) must also fail. I have not heard from Mr. Dhasmana. He has not been called by the Defendant. I do not know why, nor do I speculate what the reason may be. It does however follow that in relation to certain areas of the case I have no evidence from him as to what he did or did not do or why. We do however have the operation note from Mr. Dhasmana at F6 p.289/290. In relation to the decision not to continue with the valvotomy Mr. Dhasmana writes …… 'It was obvious that there was some blood coming through the valve into the pulmonary artery. …… I discussed the finding with Dr. Jordan and decided to abandon the procedure of trans-pulmonary valvotomy in favour of re-assessment at a later stage.' Dr. Jordan believes that he was probably in theatre during the operation and that, in the light of Mr. Dhasmana's finding, open heart surgery might be required so the decision was taken not to proceed.
37. Whether or not Mr. Dhasmana was right in his belief that there was blood coming through the valve, and there may be considerable doubt as to whether he was right, the decision not to proceed was not done on the basis of technical incompetence and a lack of assistance/supervision to carry out the planned operation.
38. Mr. Pozzi's opinion is that even if there was flow through the valve it must have been negligible and the operation as originally planned i.e. a closed operation could have gone ahead. Mr. Stark says that 'the decision whether or not to perform the pulmonary valvotomy or not at the time …… is a matter of surgical opinion. I do not think that can be classified as negligent.'
39. I think that in this regard, Mr. Pozzi has set the standard too high. Assessing the amount of flow during the operation, and therefore the likelihood of a successful closed valvotomy may be a matter of fine judgment. The safer course was to wait. I do not think it can be said that it was negligent to be cautious in these circumstances. If the shunt had been successful, that should have stabilised Marianna's condition enabling further consideration to be given to the appropriate way of carrying out the repair to the valve. In the event, that repair was eventually carried out by Mr. Wisheart with open heart surgery rather than the closed heart surgery proposed by Mr. Dhasmana.
40. In deciding that issue of negligence as with all others I have applied the test set out in the case of Bolam -v- Friern Barnet Hospital Management Committee [1957] 1 WLR 582 'The test is the standard of the ordinary skilled man exercising and professing to have that special skill. A man need not possess the highest expert skill; it is well established law that it is sufficient if he exercises the ordinary skill of an ordinary competent man exercising that particular art there may be one or more perfectly proper standards and if he conforms with one of these proper standards then he is not negligent.' That test has been followed and further explained by the House of Lords in Maynard -v- West Midlands Regional Health Authority [1984] 1 WLR 634 and Bolitho -v- City and Hackney Health Authority [1998] AC 232. The relevant passages are set out in Mr. Havers QC's closing submissions and I have treated these cases and these passages as encapsulating the relevant legal principles. Mr. Maskrey QC for the Claimant does not dissent from that.
41. It is accepted that the Gore-Tex tube inserted by Mr. Dhasmana, to form the shunt, kinked. It is also accepted that the reason it kinked was because it was too long. Mr. Pozzi is forthright in his view about this. At para. 43 of his report he says, 'As far as the execution of the shunt is concerned, there is no doubt that a significant mistake was made. The shunt is essentially a 5mm conduit which should be almost straight between the subclavian artery and the pulmonary artery. We know from the post-operative angiogramme (sic) that the shunt was kinked in the middle with almost complete occlusion (closure). This can only occur if the shunt is significantly longer than it ought to be. If it is only slightly longer, it will assume a slight curve, but not a kink. Such an excessive length would be obvious in theatre and, whilst it might happen in less experienced hands, it should be recognised and corrected immediately.'
42. During the case it has become apparent that the time at which the surgeon can see whether the shunt is too long, and therefore make any necessary correction, is after the chest has been partially closed. At that stage the shunt will be lying in its normal position within the body, and any excessive length will be apparent.
43. On viewing the angiogram, Mr. Pozzi concluded that not only was a kink visible but also there was a narrowing where the shunt was joined to the left pulmonary artery. Mr. Stark having viewed the same angiogram does not believe that there is any narrowing at this point.
44. Mr. Stark does however agree that there was a kink in the shunt caused because the Gore-Tex tube used was too long. However, he does not agree that a significant mistake has been made. He says that kinking is a recognised possible complication of this operation. He says that it is difficult to get the right length of tube. When he was pioneering this operation they tended to use too short a shunt which can have the effect of bending the pulmonary artery and so the tendency has been to make them longer rather than shorter. He agrees that it is both necessary and standard practice to check while closing up to see that the shunt is not too long but he says it is not always possible to get a good view of the shunt on closure. To support his contention that it is an accepted and unavoidable risk of the operation, he says that it has happened twice to him; that he has found it in about 3 out of 100 operations, although not after 1990, and he knows that it has happened to other experienced surgeons.
45. To help me decide this issue i.e. whether I should be satisfied on the balance of probabilities that the kinking was due to the negligence of the surgeon, I have been referred to a passage in one of the standard text books on Cardiac Surgery by Kirklin and Barratt-Boyes. Although published in 1986, there is no dispute but that this represented accepted good practice in 1985. At p. 762 the text relating to this operation reads as follows: 'A 5mm Gore-Tex tube is generally used …… Before any occluding devices are placed, the proper length of the Gore-Tex graft is determined. For this, the lung is partially inflated in order to bring the LPA (left pulmonary artery) into its usual position. When the anastomosis is completed, the Gore-Tex tube should lie without tension and without redundancy (and thus potential kinking) between the proximal one-half of the subclavian artery and the superior surfaces of the LPA.'
46. On the basis of what appears in that textbook, while kinking is an inherent risk in the operation, it is one which should be recognised and can be eliminated. Mr. Stark does not disagree with the contents of the text book but does say that it can sometimes be difficult to see clearly whether the Gore-Tex tube is lying without redundancy.
47. Mr. Stark is unable to say when it was that he performed the shunts which resulted in a kink, nor the reason for the kinking. He believes that they occurred between 1980 and 1990 and while he accepts the reasons would have been investigated and recorded, he is not able to re-call them. The figure of 3 kinks having occurred per 100 operations is Mr. Stark's estimate. It is not based on any records. We do not know when those operations occurred; who carried them out or why the shunts kinked.
48. In my judgment, bearing in mind the nature of the operation, there are only two possible explanations for this kinked Gore-Tex tube. Either Mr. Dhasmana closed the chest without properly carrying out the check described in Kirklin's book or, he carried out the check, but for some reason was unable to see the shunt clearly, and did not see the redundancy in the shunt which there must have been. On the basis of Mr. Stark's evidence that, even when exercising all reasonable care, it may not be possible to get a good enough view of the shunt, Mr. Havers QC submits that there being these two, what he would call equal alternatives, I cannot find negligence on the balance of probabilities.
49. What I have to do is to look at all the relevant facts to decide the issue. I do not consider that the maxim of 'res ipsa loquitor' assists. I have not heard from Mr. Dhasmana. He has not told me what steps he actually took. Did he look, before completely closing the chest, at the position of the shunt? If he had looked and got a clear view, he must on the evidence have seen redundancy. If he did not get a clear view of the shunt, why did he not? The only document I have to go on is his operation note. It is a detailed document. Having set out why a Gore-Tex tube had to be used, he goes on: 'The upper end (of the Gore-Tex tube) was cut obliquely and was anastomosed (joined) to the left subclavian artery in an end to side fashion using continuous 60 Prolene suture. Heparin at a dose of 1.5mg kg was given intravenously. Though the left subclavian artery at the side of the anastomosis appeared flattened there was a good flow through the graft. A proximal clamp was re-applied and the graft was cleaned with heparinised saline and filled. The left pulmonary artery was then clamped …… An opening was made in the pulmonary artery and the lower end of the Gore-Tex graft was anastomosed to this opening in an end to side fashion again using continuous 60 Prolene suture. Both clamps were then released; the distal first and there was a satisfactory flow through the graft into the pulmonary artery though I couldn't convince myself of a distant thrill'. The note then moves onto what was to have been the second part of the operation, the valvotomy. The finding of blood passing through the valve is recorded; the fact of a discussion with Dr. Jordan is recorded and the decision to abandon the valvotomy.

Finally he deals with closure and says,

'Haemostasis was satisfactory. The pericardium was approximated by interrupted sutures. One basal pleural drain was inserted and the chest was closed in layers in our usual method. Skin by subcuticular Dexon. Swabs and instruments etc. were correct.'

50. What is not in that note is any mention of any check being made for redundancy of the shunt during closure. If he had checked he should have seen redundancy and should have rectified it. If he had been unable to check for some reason, it would have been important to note it. Kinking is a recognised risk and it should have been recorded on the note that he had been unable to exclude the possibility of it happening. It is something the clinicians needed to know. If there was a risk of kinking, which Mr. Dhasmana had been unable to eliminate, it might affect their subsequent care of Marianna.
51. It may be that having moved on to consider the next part of the operation and, having discussed that with Dr. Jordan, he simply forgot to do the check.
52. Mr. Dhasmana's behaviour afterward does not suggest that he was aware that there was still a risk of kinking which he had been unable to exclude. At 8.00a.m., according to the Nursing Records (F6 p.647) Mr. Dhasmana directed that the prostaglandin infusion should stop and the arterial line be removed. I think it is inconceivable that Mr. Dhasmana would have done that if his state of mind was that there was a risk of a kink which he had been unable to eliminate. The only blood gas readings taken since the operation had shown no improvement in oxygen saturation in the blood, which can hardly have re-assured him that there wasn't a kink. Stopping prostaglandin at that stage would allow the 'ductus' to begin to die. This would result in a drop in oxygen saturation and, if no blood was passing through the shunt, the possible death of the child. If he had believed there was a risk of a kink, he would have kept the prostaglandin going until such time as there was an improvement in oxygen saturation.
53. In relation to the kink, I am satisfied, on the balance of probabilities, that Mr. Dhasmana was negligent in failing to realise before completely closing the chest that the shunt was too long and therefore liable to kink.
54. In so far as criticism was founded on the suggestion that there was a narrowing at the site of the join of the shunt and the left pulmonary artery, I am not satisfied on the balance of probabilities that there was any such narrowing. Mr. Stark could not see any such narrowing and during his evidence Mr. Pozzi seemed to become a great deal less convinced that there was one.
55. Particulars 5 and 6 of the Particulars of Negligence allege that the Defendant was negligent in its treatment of Marianna after the first shunt operation in that:

5) Failed, once it had become clear that the shunt was not working properly (which was revealed by the low blood oxygenation readings on 9.5.85) to re-assess the Claimant and re-operate to correct the fault in the shunt, and,

6) Failed to maintain the Claimant's prostaglandin medication in the period 9 – 12 May 1985.

56. These criticisms have become somewhat refined by Mr. Pozzi on whose evidence the Claimant relies for this part of the case.
57. The prostaglandin was stopped completely on the morning of the 9^th May at 9.00a.m., after a gradual weaning off after the operation on the 8^th. Mr. Pozzi agrees (see answer to Q.13 of the experts' meeting) that the timing of the withdrawal of the prostaglandin infusions was appropriate. What he does say is that, because the saturation levels remained low, immediate action should have been taken and the prostaglandin should have been re-started on the evening of the 9^th May.
58. The Claimant's case is that the Doctors should have taken the action they took on the morning of 12^th May on the evening of the 9^th, i.e. re-start prostaglandin; investigate whether the shunt was working and operate.
59. The Defendant's case is that the doctors were entitled to wait until the morning of the 12^th May before taking the action they did. Although the Defendant accepts that the oxygen saturation figures were low, it is argued that there were other indications that the hypoxia was moderate rather than severe and therefore the risk of brain damage was not great. The indications relied on are:

1) The PH and BE readings on the blood gas analysis taken on 8^th - 12^th May (F6 p.517) had improved since the reading taken on 6^th May at Gwent Hospital (F6 p.643). So, whereas there had been a marked metabolic acidosis on the 6^th; there was not between 8^th and 12^th May.

2) The general condition of Marianna immediately after the operation seemed to improve as reflected in her ability to feed: her colour and general apparent well being.

3) Hypoxia can also be caused by respiratory difficulties. X-rays taken on 9^th May did indicate some congestion in the lungs. Physio was given to Marianna which did produce matter which could have been causing congestion.

60. The Defendant had to balance the risk of a wait and see policy against the risk of another operation or the re-starting of prostaglandin. The risks of another operation for such a young baby so soon after the first one are obvious. The risks of re-starting prostaglandin are less obvious, but, there are some, and re-starting prostaglandin may make it more difficult to reach a firm conclusion as to whether the shunt had failed.
61. Dr. Jordan, in so far as he could, gave evidence of the management of Marianna's case. I say, as far as he could, because he could not be expected to remember the content of contemporaneous discussions or why decisions had been taken. Instead he had to try and re-construct what happened. He also gave evidence of the day to day running of the hospital and the way in which decisions would have been taken.
62. He said the procedure was that there were 2 ward rounds per day at 8.30a.m. and 5.30p.m. They were conducted by one or both of the Consultant Cardiologists. Also present would be a Paediatrician (Senior Registrar or Consultant); Anaesthetist (Senior Registrar or Consultant); as well as all the SHOs who might be concerned with the management of Marianna over the next few hours. Also present would be the Senior Sister in ITU (where Marianna was) and any Nurses concerned with her care. Reports would be given by the SHO and Nursing staff of Marianna's condition since the last ward round. There would be discussion, and then decisions made as to Marianna's management until the next ward round.
63. Dr. Jordan, unsurprisingly, cannot say whether he actually took part in any of the ward rounds on 9^th, 10^th or 11^th. All he can do is try and construct from the clinical and nursing notes what must have been the rationale of the treating clinicians.
64. He accepted that after any shunt operation there would always be concern as to whether it had worked or whether a further operation would be required. He said that in his experience the insertion of a shunt did not always have an immediate beneficial effect. He accepted that there was no evidence that the operation had brought about an improvement in oxygen saturation in the blood and that pulmonary flow was inadequate. He accepted that the removal of the arterial line prevented the clinicians having an accurate measurement of oxygen saturation in the blood. He said that he was aware of a loose connection between hypoxia and brain damage. He did not claim that the clinicians believed that respiratory factors were making a major contribution to the hypoxia but, until the lungs were clear, it was impossible to see exactly how great the contribution was.
65. Mr. Stark supported the Defendant's position. Having studied the notes, he believed the decision to wait and not restore prostaglandin and/or operate before the 12^th May was justified because of the general condition of Marianna, her colour, the fact she was taking oral feeds; and because tests suggested a lung component to the hypoxia and cyanosis which needed to be excluded. His opinion was that there could be a gradual increase in oxygen saturation in the blood after the shunt operation because of a decrease in resistance in the lungs. His opinion was that re-starting prostaglandin could lead to respiratory arrest and therefore a decision which should not be taken lightly. He also believed the degree of hypoxia to be moderate and he would be looking for a progressive acidosis as an indication that prostaglandin should be re-started. Mr. Stark therefore concluded that the care of Marianna between the operations did not fall below a proper standard. Indeed he says that, although he would have kept the arterial line in for another 24 hours, his treatment of Marianna would have been the same.

That opinion coming from one of the leading experts in the field is highly persuasive, but clearly it is necessary for me to look at the reasons for that opinion and decide whether I consider it to be justified.

66. Mr. Stark's objectivity is attacked by the Claimant. There is no doubt that he made a significant factual mistake in his report which the Claimant says has coloured his opinion. On the 4^th page of his report (b.1 p.201), he sets out chronologically the relevant facts. Under an entry for 8.5.85; in a reference to the first shunt operation, he says, 'Her initial postoperative progress was uncomplicated. Her arterial oxygen saturation improved to 71.3%.'

That entry is factually incorrect. There was no improvement in arterial oxygen saturation after the first operation. That reading comes from the second operation on 12^th May. That operation, as everybody agrees was a success and 71.3% is indicative of a successful shunt operation. So, the argument goes, having arrived at his opinion using an inaccurate figure, he feels he has to stick by it, even though his true opinion would have been different had he been working on the correct figure.

67. I have considered that argument with care, but I do not accept it because:

i) The most significant readings on which Mr. Stark relied were not taken immediately after the operation but when prostaglandin had been stopped, and Mr. Stark acknowledged that those readings were low. Those readings came not only from venous and capillary blood but also the transcutaneous PO2 readings which were continuously produced by a monitor attached to Marianna. It does not seem to me, on the evidence, that the incorrect initial reading materially affected Mr. Stark's conclusions

ii) Mr. Stark was adamant that if his report had included the correct information his opinion would have been the same. Mr. Wisheart made the same mistake in his witness statement. Although that may suggest that they had both been supplied with the same incorrect information, as it is unlikely they would have made an identical mistake, that does not help me in my conclusions.

68. I have not found this part of the case easy. For reasons which I will come to later, I am satisfied that during the period when the prostaglandin was stopped i.e. between 9.00a.m. on 9^th May and 12 midday on 12^th May, Marianna did suffer damage from hypoxia. So, with the benefit of hindsight, the decision not to re-start prostaglandin on the evening of the 9^th was wrong. But it does not necessarily follow that the decision to wait until the 12^th was made negligently.
69. In the end I am not satisfied that the decision was made negligently for the following reasons:

a) The state of knowledge of the connection between hypoxia and brain damage is greater now than it was in 1985. In 1985 the doctors were entitled to take the view that, because there was no acidosis, the degree of hypoxia was not such as was likely to cause brain damage. There is a causal connection between severe hypoxia and acidosis. That was therefore a relevant and proper consideration. It is likely that now, the absence of acidosis would be regarded as less significant.

b) Common sense dictates that doctors were entitled to give some weight to the general condition of the child, particularly the improvement in colour which would demonstrate a decrease in cyanosis, together with the improvement in feeding.

c) The doctors were entitled to take the view, which is supported by Mr. Stark and Dr. Jordan, that the flow through the shunt might improve.

d) Although no-one could have believed that respiratory problems were the principle cause of the hypoxia, the doctors were entitled to conclude that they may be contributing to it and needed to be eliminated.

70. When to re-start prostaglandin and/or to operate were decisions of fine clinical judgment. The clinical and nursing records demonstrate that considerable care was being taken over the treatment of Marianna. The clinicians, I am satisfied, were concerned about the low oxygen saturation in the blood. They were constantly and conscientiously monitoring her condition. In these circumstances Courts need to look very carefully, particularly when we have the benefit of 23 years hindsight, at all the circumstances and the state of knowledge at the time, before deciding that a decision was made negligently even when it can be shown, as it can in this case, to have been the wrong decision.
71. The remaining allegations of negligence relate to the 3^rd operation which was carried out by Mr. Wisheart. They are:

7) Left open the right shunt during the procedure including the time when the heart was arrested: this caused a serious drop in perfusion pressure and produced a distension of the left ventricle:

8) Removed the aortic cross clamp 'once the suturing of the patch was begun' and before it was completed, when the ventricular cavity was open in the presence of interatrial communication: Mr. Wisheart should have completely closed and de-aired the heart before removing the cross-clamp and, in particular, before the heart restarted contracting, to minimise the risk of air embolism.

72. During the course of the hearing allegation 7 was withdrawn, after examination of a perfusion chart which recorded pressures during the operation. That demonstrated that pressure fell during the operation for a period which was insufficient to cause any neurological insult to Marianna. So whether or not Mr. Wisheart was negligent, it cannot have been causative of any damage.
73. As to 8, Mr. Wisheart accepts that he removed the cross-clamp before completing the de-airing procedure, but asserts that he did not create a risk of air embolism. He says that that risk could only arise if he allowed blood to gather in the heart and eject into the aorta. He accepts that if an ejection of blood takes place before the de-airing has been completed, there is a risk, if not a likelihood, that air will also be ejected, which could cause an air embolism. Mr. Wisheart says he avoided that by not allowing blood into the heart until the de-airing had been completed.
74. Mr. Pozzi says that this technique was wrong. He says that the cross-clamp should be kept in place until the suturing and a first de-airing procedure were completed. To remove the cross-clamp and allow the heart to beat before de-airing carries an unnecessary risk of air embolism because you can never completely exclude the risk of blood getting into the heart and an ejection taking place. Mr. Pozzi relies on passages in Mr. Stark's book as setting out the correct procedure.
75. Mr. Stark says the method used by Mr. Wisheart was a perfectly acceptable method and he said he had used it himself. He said there was no risk of air escaping using Mr. Wisheart's method, if all precautions were taken. He agreed that if air did escape it demonstrated that either all precautions had not been taken or the precautions did not work.
76. This part of the case, it seems to me, can be reduced to a question of fact as to whether, on the balance of probabilities, I am satisfied that Marianna suffered a neurological insult as a result of an air-embolism.
77. I am satisfied on the balance of probabilities that Marianna suffered a neurological insult. There is agreement between the Paediatric Neurologists about that, but was it caused by an air embolism?
78. There is no direct evidence of an air-embolism. The Claimant's case is that the neurological insult was caused by something. It cannot have been caused by low pressure; therefore it must have been caused by an air embolism.
79. Mr. Stark's evidence was that there are numerous possible reasons for a neurological insult happening during the course of or after open heart surgery. In his report Mr. Stark quotes research which revealed that neurological damage was reported after open-heart surgery in between 2% and 25% (mean 8%) of patients and there have been similar results with patients at Great Ormond St.
80. That there are a number of different possible reasons for a neurological insult during and immediately after open heart surgery was confirmed by Professor Kirkham, the Claimant's Paediatric Neurological expert.
81. Having considered all the relevant evidence, I am not satisfied on the balance of probabilities that an air embolism was the cause of the neurological insult.
82. Causation. Although I have found that the Defendant was negligent in the way the first operation was conducted, the Claimant can only succeed if I am satisfied that on the balance of probabilities, Marianna suffered some damage after and as a result of the first operation.
83. To help me decide this issue, I have heard evidence from two distinguished Paediatric Neurologists. Professor Kirkham, who is Professor of Paediatric Neurology at the Institute of Child Health in London and a Consultant Paediatric Neurologist at Southampton General Hospital, was called by the Claimant. Dr. Neil Thomas a Consultant Paediatric Neurologist to Southampton University Hospitals NHS Trust was called by the Defendant.
84. While these two experts disagreed as to the central issue, namely whether any damage was suffered after the first shunt operation, there was substantial agreement between them about other matters. In particular they agreed:

1) Marianna has suffered two different forms of brain damage:

a) Periventricular leukomalacia (PVL) which is damage to the white matter of the brain and caused in this case by hypoxia or hypoxic-ischaemia. That is the damage which the Claimant attributes to the Defendant's negligence, and,

b) Global damage, the result of which is an 8 - 9 point reduction in IQ. This damage is the result of Marianna having a congenital heart defect and for which the Defendant cannot be responsible.

2) The PVL was sustained by Marianna sometime between her birth on 5^th May 1985 and the second shunt operation on 12^th May.

3) They agree that throughout the period 5^th – 12^th May, Marianna was hypoxic 'and perhaps more hypoxic in the period between the operations'.

4) In the agreed answers after their meeting they said that 'the only recognised factor for the development of PVL present in Marianna was hypoxia.' In his evidence, as I understand it, Dr. Thomas qualified that answer to say that ischaemia would also have to be present which means for the purposes of this case a reduction in the supply of blood to the brain.

85. Where they disagreed was that Professor Kirkham said that it was more likely that the PVL developed in the period between the operations rather than before the operation whereas Dr. Thomas said that it was more likely that the PVL occurred before the operation.
86. Professor Kirkham's reasons for this were based on empirical research as well as theoretical deduction. The most recent research into children with congenital heart disease who were operated on soon after birth, shows that16% to 28% of these babies were found to have pre-operative PVL whereas the number who developed PVL or whose PVL got worse post-operatively was more than 50%. Therefore, argued Professor Kirkham, it was statistically more likely that the PVL was caused post-operatively. Further, as the hypoxia in Marianna's case went on for a longer period between the operations than before, it is more likely that it was caused after than before. In her view, as agreed by Dr. Thomas, the hypoxia was at least as severe after the operation as before. The relevant periods were

1) (Before the operation) From 1.00a.m. on 6^th May when Marianna was first noted to be cyanosed to the commencement of the prostaglandin infusion sometime between 12.00p.m. and 1.00p.m., and,

2) (After the operation) From 9.00a.m. on 9^th May when the prostaglandin infusion was stopped to 12.00p.m. on 12^th May when it was re-started.

87. Dr. Thomas' final position was that, in his opinion, hypoxaemia would not cause PVL unless associated with ischaemia, and the combined effect would have to be of sufficient severity to cause acidosis. So, in his view, there needed to be a combination of a reduced level of oxygen in the blood (hypoxaemia) and a reduced rate of supply of blood (ischaemia) sufficient to cause PVL and acidosis. Because the only demonstrated incidence of acidosis was before the operation on 6^th May, and the PH readings had improved by the time of the first operation and maintained that level until the second operation, he concluded that the PVL must all have been caused before the first operation.
88. Professor Kirkham did not agree either that PVL could not occur without the presence of ischaemia as well as hypoxaemia or that PVL would only occur when acidosis was also present.
89. Her reasons for this were:

1) That it was common ground between her and Dr. Thomas that Global damage (i.e. brain damage which was not PVL) could be caused by hypoxia without ischaemia and acidosis. If that was the case for Global damage why not for PVL.

2) The research, although limited, suggested that excessive acids could develop in the brain when acidosis was not present elsewhere in the body.

3) That the absence of acidosis between the operations was likely to be the result of the body (and in particular the kidneys) compensating for the chronic hypoxia and reducing the acidosis. The acidosis on 6^th May resulted from an acute hypoxia to which the body had no opportunity to compensate.

90. In supporting Professor Kirkham's view against Dr. Thomas', Mr. Maskrey QC correctly points out that Professor Kirkham has always expressed the view that it was more likely that the PVL was caused after the first operation whereas Dr. Thomas' view until he wrote the letter dated 27^th September 2007 was that he could only say the PVL was sustained between birth and the second operation and he was unable to say it was more likely to be before or after the first operation. Indeed in his evidence he said that was really the position he was most comfortable with. There is nothing intrinsically wrong with an expert changing his or her view on more reflection or on the basis of fresh information; indeed, it might be regarded as a refreshing change, but in this case the two views he has expressed are inconsistent. If it is correct to say that without acidosis PVL cannot be caused then it could never have been possible that it was caused after the first operation.
91. On balance I prefer the evidence of Professor Kirkham that PVL is caused by hypoxia, which does not have to be associated with ischaemia and that it can occur in the absence of acidosis. I have concluded that her view is more consistent with the small amount of empirical evidence which we have and her explanation for the absence of acidosis between the operations makes sense.
92. I am however unconvinced by Professor Kirkham's argument that PVL is only likely to have been caused after the first operation. It is based on the 16% to 28% figure to which I have already referred. These findings were based on relatively small numbers of patients although according to Professor Kirkham a sufficient number to be statistically significant. In my judgment the most likely explanation why some babies got PVL pre-operation and why some did not is that those who did were more hypoxic than those who did not or were more sensitive to hypoxia. There does not seem to be any logical basis for any other explanation.
93. On the evidence I have heard the likelihood is that Marianna suffered severe hypoxia. On the balance of probabilities, I am satisfied that Marianna suffered damage throughout the period from 1.00a.m. on 6^th May to the second operation save for the times when she was on a prostaglandin infusion.
94. If, on the evidence, I am able to decide on the balance of probabilities what proportion of the damage was caused before the operation, then the Defendant will not be liable for that damage. If, however, I am unable to make any apportionment then the Claimant is entitled to recover in full. That is, as I understand it, agreed by the parties to represent the law on the authority of Dingle -v- Associated Newspapers [1961] 2 QB 169
95. One possible way of apportioning the damage would be by reference to the time that Marianna suffered from hypoxia before the first operation as compared to the whole period. That works out at 15%. That has the merit of simplicity and has the attraction that the Defendant would not have to compensate the Claimant for damage she has suffered for which they are not responsible. In cross-examination Professor Kirkham seemed to accept that that was a possible approach but in re-examination said that from a scientific point of view there is no way the damage suffered before the first operation can be accurately calculated.
96. Dr. Thomas was also asked the same question. By now his position was that all the damage was caused before the first operation. He was therefore asked: if he was wrong about that and damage was caused over the whole period of time, what percentage would be caused before the first operation. As I have not accepted his evidence as to when the damage was caused, it follows that I must reject this evidence as well. Dr. Thomas himself said it was guesswork.
97. The matter is further complicated by the fact that research has shown that, in many cases where the cause of the hypoxia is treated successfully, PVL which has been suffered, is reversed.
98. These questions arise:

1) May the initial acute hypoxic episode cause more damage than later hypoxia for which the body may compensate?

2) Alternatively, is the body's resistance worn down by prolonged chronic hypoxia so that the amount of damage suffered increases exponentially?

3) If the first operation had been successful, would any PVL suffered have been reversed or a proportion of it and if so, how much?

99. Medical science is unable at present to answer these questions. On the balance of probabilities I do not think that the damage would have been suffered equally over the whole period. Indeed I think it is extremely unlikely that it was. I do not believe it is possible to apportion the damage and, on the basis of what I understand to be the agreed position in law, the Claimant is entitled to recover in full for the PVL.
100. As I have already indicated the Claimant is not entitled to recover for the Global damage which accounts for 8 to 9 points of IQ. I have already found that the Claimant has no claim in relation to the third operation as I am not satisfied there was an air-embolism. Even if I had been so satisfied, I would not have been satisfied that any damage was suffered by the Claimant arising from it. It is agreed it cannot have caused PVL. If it caused any damage it would be Global damage, and there is no evidence that Marianna has suffered Global damage beyond what you would expect to find with any child with congenital heart disease.
101. Both Professor Kirkham and Dr. Thomas agreed that there was some neurological insult during or just after the third operation. They also agreed that a neurological insult can cause permanent damage but may not. Professor Kirkham said that permanent damage was more likely to be caused in Marianna's case because of her pre-existing brain damage. However, the evidence does not satisfy me on the balance of probabilities that damage was suffered by Marianna arising from any neurological insult arising from the third operation.