Leggett v Norfolk, Suffolk & Cambridgshire Strategic Health Authority

Alistair Leggett is a severely disabled young man who has just turned 16 years of age. In consequence of brain damage sustained by him soon after birth he suffers now from tetraplegic dystonic cerebral palsy, profound sensoneural hearing loss and epilepsy. In addition he has upward gaze paralysis, dental dysplasia and in all probability some degree of learning difficulty. Alistair needs permanent, full time care to help him to live with his many disabilities; and it is the belief of his mother, Mrs Deborah Green, that those disabilities, and the brain injury that caused them, have been sustained by Alistair because staff at the hospital where he was born failed to exercise reasonable care and skill in looking after him.

Alistair was a premature baby and was born after 34 weeks and 2 days gestation at 3.19 a.m. on the 15th March 1990 at the Norfolk and Norwich Hospital in Norwich, and there is no dispute that before he was discharged home in April 1990 he had sustained the cerebral insult that has so disabled him. Alistair’s mother, who with his father has cared for him devotedly ever since, issued a writ as his next friend on the 5^th June 1995 against the Norwich Health Authority, the body then responsible for the hospital where Alistair was born, and in the pleadings as then drafted she alleged that there had been negligence in his obstetric and neonatal care and that the hospital staff’s breach of duty had caused hypoglycaemia leading to cerebral palsy. As the litigation progressed and experts investigated what had happened in March 1990, it became apparent to those advising Mrs Green that the initial allegations could not be substantiated. By a curious turn of fortune, however, one of the Defendant’s witnesses, Dr Rosenbloom, suggested to Mrs Green that a possible unifying diagnosis for Alistair’s difficulties was that he had been caused to suffer a rare condition known as kernicterus, brought about by toxic levels of bilirubin penetrating his brain within a few days of birth.

So, on the basis of this new suggestion Mrs Green abandoned her original allegations of negligence and with the help of her legal and medical advisers set about investigating the question whether Alistair did indeed suffer from kernicterus and, if so, whether either the medical or nursing staff were in breach of their duty of care towards him in the way they addressed the bilirubin levels in his blood. It is now her case that the hospital staff were indeed in breach of that duty, and that had reasonable skill and care been exercised in March 1990 Alistair could and would have been saved from kernicterus. The pleadings have been amended accordingly; the Norfolk, Suffolk and Cambridgeshire Strategic Health Authority is, by amendment, the defendant now to the action; and by an order of Master Ungley dated 25 September 2005 I have been directed to determine at this stage all issues of liability and causation. It will come as no surprise to the parties and the witnesses to be told that judicial examination of events that occurred over 16 years ago, and judicial evaluation of the appropriate standards of care in 1990, is an exercise made especially difficult by that significant lapse of time.

Before I turn to the detailed history of the case I should note that although in the defendant’s written evidence there is some query as to whether all of Alistair’s disabilities are consistent with kernicterus, the defendant’s medical experts have deferred to Dr Rosenbloom’s opinion, and so it is now accepted on behalf of the defendant that there is no other cause. If, therefore, there was a breach of duty, the only question on causation is whether Alistair’s kernicterus would probably have been avoided had that duty been properly performed.

Bilirubin is a pigment carried in the blood which causes jaundice in babies when its level is elevated. If the serum level becomes excessively elevated to the point of toxicity, bilirubin can penetrate the brain/blood barrier and cause neurological damage. The condition that ensues is called kernicterus, (connoting a yellow staining of the brain) or, more correctly, bilirubin encephalopathy. Bilirubin, however, has two constituent elements, namely “conjugated” or “direct” bilirubin, and “unconjugated” or “indirect” bilirubin. The former is bound to albumin and is not toxic; the latter is capable of being toxic. As the evidence revealed there are circumstances in which it is now thought that “free” bilirubin, falling into neither category, may also be toxic, but for most cases in most circumstances it is unconjugated bilirubin that is potentially toxic to the brain.

Serum bilirubin is measured in micromols per litre, and references in this judgment are therefore to be taken as being in mmol/l units. Where the measurement includes both unconjugated and conjugated bilirubin the result is known as “total” bilirubin; where the separate unconjugated and conjugated levels are shown the result is known as “split” bilirubin. Split bilirubin can only be measured by tests carried out in the laboratory, but in the Norwich Hospital, as in the vast majority of hospitals, the SCBU had its own bilirubinometer for carrying out tests on site. The ward meter, however, could only measure total bilirubin in the blood and one of the questions debated in this case is whether or in what circumstances it is acceptable to have regard to the split as opposed to the total serum bilirubin results in deciding whether to intervene when the levels are elevated.

When a baby suffers from jaundice nature usually takes its course and there is no cause for concern. In some cases, and in neonatal units this is a common occurrence, the baby’s bilirubin level requires intervention by the provision of phototherapy. In other words the baby is put under “lights” in the incubator and the effect of light on the skin is to reduce the amount of bilirubin in the body. In other, very much rarer, cases, the serum bilirubin is so elevated that phototherapy may not prevent it rising to a toxic level and more substantial intervention is mandated. In such cases the baby is given a blood transfusion known as an “exchange transfusion” where over a period of time the whole of the baby’s blood is replaced with twice its volume. In this way sufficient bilirubin is eliminated from the blood circulation as to bring it down to a safe level for the baby.

In Alistair’s case there is no dispute that his serum bilirubin did rise to a toxic level and that kernicterus ensued, but there are six principal issues that have been debated in the evidence and that I have to decide:

When Alistair was born on the 15^th March 1990 he weighed 2.5 kg and so although he was premature, he was not significantly underweight. At birth he had a normal head circumference and there were no neurological concerns, but because of some respiratory distress he was admitted to the neonatal Special Care Baby Unit (“SCBU”) within the hour. During the mid-morning period he became collapsed and deeply cyanosed, so he was intubated and ventilated. In the afternoon he was thought to have Group B Streptococcal pneumonia and the clinical notes on the 16^th March 1990 identified four problems, namely sepsis, an extravasation injury, persistent hypoglycaemia and prematurity. The agreed medical evidence now is that Alistair probably did suffer from GBS pneumonia, but because his blood culture did not grow GBS it is unlikely that there was septicaemia-at any rate there is no contemporary evidence of septicaemia. His pneumonia was appropriately treated with antibiotics and his hypoglycaemia was corrected with dextrose.

On the 15^th March when he was 12 hours old Alistair’s total bilirubin level was 64 according to the ward meter. On the 16^th March the total bilirubin level was 101 on the ward meter when, according to the agreed evidence, Alistair was 30 hours old. According to the laboratory analysis the total was 97 at this time, of which 11 was conjugated (i.e. non-toxic). All of these readings, it is agreed, were within safe limits.

On the 17^th March Alistair’s total bilirubin was noted on the ward meter to be 163, and as I understand the agreed evidence this would have been at about 9.00 a.m. There is a nursing note that it was to be repeated, but there is no note of any other result that day. Again, it is accepted that 163 was not a reading that required any intervention by nurses or doctors apart from routine monitoring. However, it was noted that Alistair appeared to be increasingly jaundiced. Otherwise Alistair was noted to be improving and in the latter part of the day he was extubated.

On the 18^th March at about 9.00 a.m. Alistair’s total serum bilirubin according to the ward meter was 189, again within safe limits, but by about 8.15 p.m. it had risen to 213. At this point one of the doctors advised that phototherapy should be started and this was done. It is agreed that this was an appropriate response to the indicated bilirubin level. Also during the day Alistair had to be re-intubated. He was suffering from worsening acidosis and there must have been some concern about infection because an anitibiotic, Netelmicin, was added to his bolus. However, the clinical notes included the observation that “colour and general state improved after IPPU”. The events of the 18^th March into the 19^th March will require further analysis in this judgment, but for now it is sufficient to note that overnight Alistair was pyrexial, tachycardic, restless, intolerant of handling and still jaundiced. He remained in an incubator, ventilated and under phototherapy.

On the morning of the 19^th March 1990 Alistair was seen by the consultant paediatrician Dr Crowle on his morning ward round at about 9.00 a.m. The clinical notes report Alistair as being jaundiced and under lights, but there is no reference to any bilirubin level, although it is agreed that blood was taken through an arterial line at about 7.00 to 7.30 a.m. for blood gases to be tested in the laboratory and made available during the course of the morning ward round. Alistair was receiving full strength TPN, was sedated, and was being treated still with Penicillin and Netalmicin. Between 9.30 a.m. and 10.30 a.m. he suffered an episode of hypoglycaemia which was treated with extra dextrose and at 10.30 to 10.45 a.m. he was “noted to be making strange movements ? coughing/?fit”. The nursing notes record that during the morning he was “fighting the ventilator” despite the prescription of Triclofos, a sedative, and that “by 10.45 hours baby noted to be having slightly abnormal jerky movements”.

In the latter part of the morning Alistair was in pain with a distended stomach. The Registrar, Dr Roberts, at about mid-day suspected a malrotation with a volvulus, a condition which if not treated as an emergency could lead to the loss of the small intestine, and she called the consultant, Dr Brain. He confirmed the diagnosis and recommended an immediate laporotomy. A further blood sample was taken from Alistair for cross-matching before he went to the theatre for surgery and during the afternoon, between about 3.30 p.m.and 4.30 p.m., a laporotomy was performed. During the procedure Alistair was not receiving phototherapy, which otherwise had been continued, but on his return from surgery the evidence tends to suggest that he went back under lights and nobody has sought to argue otherwise. However, no malrotation or volvulus was discovered during the operation, although his liver was seen to be large and congested and a large gall bladder was emptied upon compression with the bowel, and so he was returned to the ward for his condition to be monitored. At this stage it is not necessary to recite all the findings made or the fluids and medication then prescribed, save perhaps to note that Alistair was still being treated for acidosis.

So far as bilirubin levels on the 19^th March are concerned there are four pieces of evidence relevant to an issue that has been very fully debated during the course of the trial, i.e. what was the actual level of serum bilirubin at material times during the day, and what were the levels known to the treating doctors and nurses :

So far as the clinical notes are concerned there is no reference to bilirubin until 11.00 p.m. when a doctor refers to “jaundice with ↑ with high conjugated SBR”. I shall return to this evidence in due course in order to decide what inferences can be drawn about the bilirubin levels at different times of the day.

What otherwise is significant about the 19^th March is that by 4.45 p.m. Alistair was found to be “very rigid and intolerant of handling: some odd movements observed-fitting-?pain?anaesthetic?rigors”. A nursing note made at 8.00 p.m. refers to “abnormal movements since return from theatre-rhythmic movements of arms and legs and lipsmacking…Definite fitting”.

On the 20^th March at 7.00 a.m. Alistair’s total bilirubin was then measured on the ward meter at 282; by 5.00 p.m. it had come down to 173; on the 21^st March it was further reduced to 166, although a laboratory analysis produced a total serum bilirubin of 137 with a split of 99 unconjugated and 38 conjugated.

The records for present purposes do not need further exposiiton. Although it unfortunately took many years for the diagnosis of bilirubin encephalopathy to be made, the summary of Alistair’s history does not need to go any further than this because it is agreed that by the end of the 19^th March 1990 he had sustained irreversible kernicterus.

With regard to the likely bilirubin levels at the relevant times I heard evidence from three nurses, Sister Lee, Staff Nurse Hope, and Nurse Practitioner Mynett. Evidence on the point was also given by Dr Crowle and Dr Roberts. The former was then a consultant paediatrician and had overall responsibility for Alistair’s care after he was born. The latter was a Paediatric Registrar who saw Alistair on the 19^th March 1990. Evidence was also given by the parties’ respective expert medical witnesses, Dr Ives on behalf of the claimant, and Dr Rennie on behalf of the defendant. Although called to assist the court on what was or was not acceptable practice in 1990, they both gave evidence as to their view of the inferences to be drawn about the progress of Alistair’s hyperbilirubinaemia over the relevant days in March 1990. Both expert witnesses are immensely qualified in the field of neonatology and indeed are colleagues, in the sense that Dr Ives has written the chapter on neonatal jaundice in the last two editions of Roberton’s Textbook of Neonatolgy, of which Dr Rennie is the editor. They are contemporaries in terms of their medical careers, with Dr Ives now a Consultant Neonatologist and Honorary Senior Lecturer in Paediatrics at the John Radcliffe Hospital in Oxford, while Dr Rennie is a Consultant and Senior Lecturer in Neonatal Medicine at University College London Hospitals. Such difference as there is in expertise comes down to this, it seems to me. Dr Ives has published extensively about bilirubin and penetration of the blood/brain barrier since 1992 and has a particular interest in kernicterus. Dr Rennie has a broader experience of paediatrics and neonatolgy in terms of the number and variety of publications and lectures, has a less specialised interest in kernicterus, but by virtue of her senior positions on advisory boards and committees over many years has extensive knowledge of practices in different neonatology units around the country. In particular she had responsibility for supervising standards of training for specialists in neonatology when she was Chairman of the College Special Advisory Committee for the Royal College of Paediatrics and Child Health between 2001 and 2005. Both, needless to say, were impressive witnesses.

In addition I have read the evidence of Alistair’s parents, Mr Leggett and Mrs Green, which can only provide a glimpse of the commitment they have made to Alistair.

The decision as to the precise bilirubin levels in Alistair’s blood over the 18^th and 19^th March 1990 is made difficult, it is agreed by Dr Ives and Dr Rennie, by the paucity of surviving evidence. The evidence of the medical and nursing staff is that in 1990, as now, a baby’s serum bilirubin level should have been entered on a chart to be kept at the foot of the incubator and noted in a bilirubin book. Although it appeared that at one time Dr Ives was accusing the staff at the Norwich Hospital of poor practice in not having such a chart, I think he now accepts that there was a chart on which Alistair’s bilirubin was recorded and that it has now been mislaid, as has the book. In any event, that is the finding I make because it is quite apparent from other charts for other babies born in early 1990 that such charts were routinely used, and there is no hint in the contemporaneous notes that Alistair was an exception. The extent to which such a chart would in fact have assisted becomes a matter of further speculation because Nurse Lee told me that sometimes the doctors who took the ward readings omitted to note them on the chart, and indeed there are some gaps in the anonymised notes of other babies consistent with such oversight. Furthermore, there is the evidence of Mrs Green and Staff Nurse Hope that the unit was particularly busy during that time in March 1990 when Alistair was born.

The records that do survive, therefore, are (a) the Data Flow Sheets, which do not by any means necessarily contain all the readings, (b) a chart of the blood gases taken during the day with timed entries, (c) such references to bilirubin as the nurses or doctors chose to make in their own notes, and (d) the analysis of three split results from the laboratory. The Data Flow Sheets contain some times for the taking of blood samples, but there are no times entered for the 18^th and 19^th March. The nursing notes provide a time frame within which the level was recorded, but the nursing evidence was that the notes were made according to a sequence of topics pre-determined by the care plan and so inferences cannot automatically be drawn from the place in the notes where a reference to bilirubin appears. And the medical notes do not refer to bilirubin at all during the precise hours under consideration.

What the evidence does disclose is this. On the 15^th March the Data Flow entry of 64 was made at about 3.20 p.m.. On the 16^th, 17^th and 18^th March the entries of 101, 163 and 189 were probably made, according to Dr Rennie, at about 9.00 a.m. On the 18^th March the further entry of 213 was probably made at about 8.15 p.m. when phototherapy was commenced. The entry of 267 for the 19^th March that appears on the Data Flow Sheet is (like those for the previous days) not timed. It is Dr Ives’ opinion that it was probably made in the morning at the time of the ward round, the sample having been taken earlier when the blood gases were sampled at about 7.00/7.30 a.m. (the timed entry on that sheet being a little difficult to decipher). It is Dr Rennie’s opinion that it was probably made some time at the end of the nurses’ shift at 12.30 to 1.00 p.m.

There is then the split laboratory test of the 19^th March with the result for total serum bilirubin of 321. There is no time on the printout to show when the sample was taken on the ward or analysed in the laboratory. However, both Dr Ives and Dr Rennie consider that the blood for the serum bilirubin test was probably taken with the other bloods recorded to have obtained through an arterial line at 7.00 to 7.30 a.m. Staff Nurse Hope had a different view and she recollected that the doctors on night shift would take the blood at about 6.00 a.m. so that it could be sent to the laboratory and be available for the morning ward round. But given that the chart in question shows that the previous blood sample on the 19^th March was at 3.00 a.m., her memory as to 6.00 a.m. is almost certainly wrong and it seems to me unlikely as well that 3.00 a.m. was the time when the laboratory sample was taken. Furthermore, it is Dr Crowle’s recollection, which he says he has checked with the laboratory staff, that in 1990 the laboratory did not open for routine assays until 9.00 a.m. and that tests were only done during the night in cases of emergency, when the on-call technician would be summoned. Although hearsay evidence given without notice is unsatisfactory, I have indicated that think that the nurse’s recollection is mistaken as to the timing of the sample and in my view she is likely to be mistaken as well about the hours kept by the laboratory. So, I have no hesitation in accepting that the 321 serum bilirubin sample was taken by arterial line from Alistair at about 7.30 a.m..

In terms of timing, therefore, the important questions to be answered are:

Both Dr Ives and Dr Rennie agree that the arterial blood sample taken at about 7.30 a.m. would probably have been used to assess the ward serum bilirubin level. So, the question is whether it was that sample that produced the 267 reading or another. Dr Rennie was of the opinion that because the nursing note at 12.30 p.m. referred to an elevated level remaining at 267, and the reference occurs right at the end of the notes, it is likely that this was the sample of blood recorded as having been taken at 11.30 a.m. rather than the 7.30 a.m. sample. The problem with that theory is that, if correct, it would mean that the doctor extracting blood at 7.30 a.m. did not do a ward serum bilirubin test, or that he failed to enter the morning result on the Data Flow Chart. Dr Rennie has noted that on the previous three days an early morning reading was recorded in the Data Flow Chart, and it would be very odd indeed if no reading was taken on the morning after phototherapy had been started, or, if taken, that the doctor forgot to put it on the Data Flow Chart. So, on the balance of probabilities I think that Dr Ives is right, and I find that 267 was the ward serum bilirubin reading from the sample taken at about 7.30 a.m.

As Mr Pittaway Q.C. has observed on behalf of the defendant, Dr Rennie thinks it likely anyway that the reading was between 260 and 270 at 7.30 a.m., and so in terms of the information available to the nursing and medical staff at the start of the day my finding makes no real difference. However, it does leave open the question, to be considered later, as to what, if any reading of ward serum bilirubin was available after bloods were taken at 11.30 a.m.

So far as the post operative reading of 272 is concerned, there is nothing in the Data Flow Chart or the nursing or clinical notes to put a time on it. The chart described as “Gasses and Vent settings” in the neonatal records notes that blood was taken after the operation at 5.00 p.m., 6.10 p.m., 7.45 p.m. and 10.15 p.m. The Data Flow Chart, which is not timed for the relevant entry, describes 272 as “post/op”. The chart with the timing when blood samples were taken has an annotation “Post Op” in an unknown hand for the 5.00 p.m. sample. And logic would suggest that in the afternoon, as in the morning, the Data Flow Chart would note the first result rather than a later one. So unless no serum bilirubin test was done immediately after the operation when blood was being taken for other tests, which seems to me improbable, 5.00 p.m. seems a more likely time for the 272 sample than 6.00 p.m. or 7.45 p.m. Unfortunately the blood gas printouts for the evening of the 19^th March have not been kept. Those that survive bear a date and time in somebody’s handwriting and one can cross check, for example Hb entries, to see whether or not that is the same sample as appears in Data Flow Chart. For the post operative sample, however, there is nothing to cross check against, and so on very exiguous evidence I find that 272 was serum the bilirubin reading at 5.00 p.m. and not any of the later times.

The next factual issue is when the split result was received on the ward. There is no reference to it in the medical notes until 11.00 p.m. when the entry mentions “Jaundice with ↑ with high conjugated SBR?” The note is written in the context of a query about hypoglycaemia and refers as well to the congested liver found during surgery. Although it is the conjugated element that is of interest to the note writer and not the unconjugated, the note plainly is written after the split result has been seen. But when was it seen?

Dr Crowle would not necessarily have expected to receive the laboratory results until late in the day, unless an urgent request had been put in. The results needed for ward rounds would be available in the morning, not at the start of the round but later on. The nurses’ recollection is that all laboratory results would be available for the morning ward round, but I am not convinced that their memory is necessarily so accurate as to distinguish between the ordinary blood tests and the bilirubin results, which are sent to the laboratory with different sample numbers. Neither Dr Crowle nor Dr Roberts has any recollection now of knowing the split result during the course of the day, but then neither was asked to recall any of the details concerning Alistair’s presentation with jaundice until very many years had passed, so their absence of memory is of little assistance.

Again on exiguous evidence I think on balance that the nursing and medical staff did not know of the split result until late on the 19^th March. My feeling is that if it had been available before the operation at 3.30 p.m. Dr Roberts would at the very least have referred in her notes to the conjugated element as being potentially relevant to Alistair’s stomach problem and would have wanted Dr Brain to know. Another slight straw in the wind is that the nurse who saw Alistair after the operation made copious notes, described him as “pink and jaundiced,” and might well have noted a laboratory finding if it related to that jaundice. So too the nurse who went off duty and made her notes at 12.30 p.m.: if she had the laboratory result, would she still have referred to the 267 result from the early morning ?

There is a separate issue as to whether the laboratory result should have been telephoned through in the morning and, if so, whether that would have made a difference to Alistair’s outcome, but I will deal with that point after considering whether the claimant has established a breach of duty on what was actually known about Alistair’s serum bilirubin levels at the time.

The principles are well established and set out in Bolam v. Friern Barnet Hospital Management [1957] 2 All ER 118, at 121-122, starting with a citation from Hunter v. Hanley:

    “If that statement of the true test is qualified by the words “in all the circumstances”, counsel for the plaintiff would not seek to say that that expression of opinion does not accord with English law. It is just a question of expression. I myself would prefer to put it this way: A doctor is not guilty of negligence if he has acted in accordance with a practice accepted as proper by a responsible body of medical men skilled in that particular art. I do not think there is much difference in sense. It is just a different way of expressing the same thought. Putting it the other way round, a doctor is not negligent, if he is acting in accordance with such a practice, merely because there is a body of opinion that takes a contrary view. At the same time, that does not mean that a medical man can obstinately and pig-headedly carry on with some old technique if it has been proved to be contrary to what is really substantially the whole of informed medical opinion. Otherwise you might get men today saying: “I don’t believe in anaesthetics. I don’t believe in antiseptics. I am going to continue to do my surgery in the way it was done in the eighteenth century”. That clearly would be wrong.”

Then in Maynard v. West Midland Regional Health Authority [1985] 1 All ER 63 Lord Scarman said:

In Bolitho v. City and Hackney Health Authority [1998] AC 222 Lord Browne-Wilkinson qualified the probative effect of expert evidence about the practice of respectable doctors at pp 241-242:

“Again, in the passage which I have cited from Maynard's case [1984] 1 W.L.R. 634, 639, Lord Scarman refers to a "respectable" body of professional opinion. The use of these adjectives - responsible, reasonable and respectable - all show that the court has to be satisfied that the exponents of the body of opinion relied upon can demonstrate that such opinion has a logical basis. In particular in cases involving, as they so often do, the weighing of risks against benefits, the judge before accepting a body of opinion as being responsible, reasonable or respectable, will need to be satisfied that, in forming their views, the experts have directed their minds to the question of comparative risks and benefits and have reached a defensible conclusion on the matter.”

On a separate issue, and one that arises on the evidence in this case, Lord Browne-Wilkinson dealt at p 240 with the problem of establishing causation where a doctor has failed to do something that he had a duty to do:

“I adopt the analysis of Hobhouse L.J. in Joyce v. Merton, Sutton and Wandsworth Health Authority [1996] 7 Med.L.R. 1. In commenting on the decision of the Court of Appeal in the present case, he said, at p. 20:

"Thus a plaintiff can discharge the burden of proof on causation by satisfying the court either that the relevant person would in fact have taken the requisite action (although she would not have been at fault if she had not) or that the proper discharge of the relevant person's duty towards the plaintiff required that she take that action. The former alternative calls for no explanation since it is simply the factual proof of the causative effect of the original fault. The latter is slightly more sophisticated: it involves the factual situation that the original fault did not itself cause the injury but that this was because there would have been some further fault on the part of the defendants; the plaintiff proves his case by proving that his injuries would have been avoided if proper care had continued to be taken. In the Bolitho case the plaintiff had to prove that the continuing exercise of proper care would have resulted in his being intubated."

   There were, therefore, two questions for the judge to decide on causation. (1) What would Dr. Horn have done, or authorised to be done, if she had attended Patrick? And (2) if she would not have intubated, would that have been negligent? The Bolam test has no relevance to the first of those questions but is central to the second.”

It is the claimant’s case that every competent nurse and doctor in a SCBU either knows or should know of the maximum serum bilirubin beyond which a baby cannot safely be left without an exchange transfusion. So much is uncontroversial. It is also part of the claimant’s case that before serum bilirubin ever reaches that level, whatever it is, the nursing and medical staff should intervene first with phototherapy to bring the bilirubin level down, and if that is not succeeding they should order blood in good time for the transfusion to be carried out. Again, there is no dispute about the general nature of the duty, and on the part of the claimant there is no complaint about the point at which Alistair was treated with phototherapy.

It is Dr Ives’ opinion, however, that even on the information that it is proved was available on the ward during the morning of the 19^th March the medical staff, and in particular Dr Crowle and Dr Roberts, fell below an acceptable standard of care in failing to order blood and in failing to arrange for Alistair to undergo an exchange transfusion. His evidence and reasoning may be summarised, without I hope doing any injustice to the conscientious way in which he demonstrated that his was not a conclusion arrived at lightly, as follows:

There are, it seems to me, some important general points to be made with regard to my task in judging what the acceptable standard of care was in 1990. First of all, as Dr Ives said in evidence, the paediatricians and neonatologists of the UK have, even in the year 2006, been unable to get together and agree some national standards. The result is that there is a variety of local standards and protocols and practices throughout the country’s hospitals. Dr Rennie told me that in discharging her responsibility to oversee training in neonatology departments she has herself seen a number of practices that she would not herself follow, but which responsible consultants consider to be reasonable. It has not been part of her brief, or that of anyone else, to eliminate such local variations in treating jaundice. The implication of Dr Ives’ evidence is that if such local practices fall short of the standard that he believes to be acceptable, then those departments and the consultants in charge of them are at risk of being held to be in breach of their duty of care to jaundiced babies, should anything like this happen again. It is in my view profoundly to be regretted that in a field of medicine where irreversible brain damage may occur, it has not proved possible to agree a straightforward national standard for intervention by exchange transfusion, and to stick to it.

Secondly, there has over the years been a debate internationally about the threshold at which clinicians should intervene with exchange transfusion. Dr Ives in his chapter in the 1999 edition of the Textbook of Neonatology edited by Roberton and Rennie refers to the “so-called ‘kinder, gentler approach to the jaundiced infant’”, notes that the American guidelines for healthy term infants have been criticized both for laxity and stringency, and warns against the potential dangers inherent in a move to a more relaxed approach.

On the other hand Dr Rennie warns against too hasty a rush to proceed to exchange transfusion because the procedure carries its own risks. Indeed Dr Ives refers at p. 729 of his chapter to the “perils of exchange transfusion”, and at p.728 he identifies them as follows: “In addition to the low risk of bloodbone infection, exchange transfusion carries a significant risk of morbidity and mortality from vascular accidents, cardiac complications and biochemical or haematological disturbance. This is especially the case in sick premature newborns.”

It seems to me therefore that registrars and consultants who have to make the decision whether to proceed to an exchange transfusion in the case of a particular baby are, albeit within certain parameters, pulled in different directions, and there is room for the reasonable clinician to believe in either a more or a less interventionist approach depending on his or her experience, training and judgment. It is in that context that one has to consider whether a particular set of circumstances mandated an exchange transfusion.

I deal first with the issue of the rate of rise in Alistair’s serum bilirubin overnight on the 18^th/19^th March 1990 from 213 to 267 according to the ward bilirubinometer. The former was from blood taken at 8.15 p.m., the latter from blood taken at about 7.30 a.m., a little under 12 hours. The rise in total serum bilirubin was therefore apparently 54, or just over 4.5 mmols/l/h. What is the evidence that this rate called either for an immediate transfusion or for the ordering of blood?

Within the bundles of documents there are exhibited a number of lecture notes, protocols and draft protocols for treating jaundice that were compiled either by medical practitioners or nursing staff or both at various times. There is also the text of Dr Ives’ chapter in the 1999 edition of Dr Rennie’s textbook, and there is material from “Neonatal Jaundice” by Dr Maisels, who Dr Ives acknowledges to be an expert in the field.

The first thing to note is that in the lecture notes prepared by Dr Upton and Nurse Mynett there is specific advice that in the case of a haemolytic baby (which Alistair was not), “If rate of rise of SBR ›20/hr will almost certainly require exchange”. Dr Ives in his chapter writes at p.728 “If, despite double phototherapy, the serum bilirubin continues to rise by more than 10 mmols/l/h exchange transfusion is to be anticipated”. I accept his evidence that his comment here refers to haemolytic jaundice and is not to be construed more widely. But the fact remains that although there is clear guidance in the case of the haemolytic baby, (albeit the rate of rise requiring exchange in such babies seems to have been very much greater in 1990 than 1999), there is no guidance in anything like those terms in relation to babies who appear to have physiological jaundice. Dr Maisels it is said gives specific advice in his book and warns that a rate of rise of more than 17 mmols/l/h may mandate earlier intervention by exchange transfusion, although there is an issue again as to whether that refers only to haemolysis; but to my mind it is significant that there are no specific references in Dr Ives’ chapter or in the materials available to the staff in 1990 at the defendant’s hospital stating that any particular rate of rise in serum bilirubin results requires intervention.

It is true that there are references both in Dr Ives’ published work and in the materials available to the doctors and nurses in 1990 in which a rate of rise of more than 100 mmols/l over 24 hours is identified as a significant clinical feature (see p.723 of the 1999 edition), but in each case that rate of rise is described at the head of the relevant table as, “Clinical features that suggest a pathological cause of jaundice in the newborn”. In other words, the reader is not being told to prepare for transfusion, but to recognise that treatment for pathological rather than physiological jaundice may be required. Such treatment may of course require transfusion, but on its own the rate of rise does not according to the text mandate that degree of intervention. Indeed Dr Ives said in his evidence in chief that a rate of rise of 54 mmols/l/12h, leading to a potential rate of 108mmols/l/24 hours, suggests pathological jaundice and so should be investigated further.

So, although Dr Ives is of the view that Dr Maisels is wrong about 17 mmols/l/h being the threshold that difference of opinion, he says, relates to haemolytic babies and is irrelevant to this issue. But, as I have indicated his textbook does not advise in terms that there is any specific rate of rise of serum bilirubin in babies like Alistair that mandates exchange transfusion, and there is nothing in any material available in 1990 that I have seen to put the nurse or medical practitioner on notice that any particular rate of rise requires intervention of that nature. In my judgment the highest it can be put is that 54 mmols/l/h is a rate of rise which, if persisted in when serum bilirubin was next tested, would have been a significant feature warranting closer monitoring of the serum bilirubin level than otherwise would be the case. There was no breach of duty therefore in the failure of the doctors, when the 267 reading was obtained, to move to exchange transfusion because of the apparent rate of rise. The same applies, a fortiori, to any results obtained between 8.15 p.m. on the 18^th and 7.30 a.m. on the 19^th March, there being no evidence to establish that there was a steeper rate of increase during the night.

The next question is whether the reading of 267 was so high that, with some evidence from the rate of rise being suggestive of pathological jaundice, blood should have been ordered with a view to being ready to exchange if the levels continued to go up. There is a here a conflict of opinion between Dr Ives and Dr Rennie, not so much as to the threshold for exchange transfusion in healthy premature babies, but as to whether acceptable practice in 1990 required that threshold to be lowered if the baby was ill. Within the materials available to staff at the defendant’s hospital at different times over the years, and within the textbooks and journals, different opinions have been given at different times about the appropriate threshold in the case of healthy premature babies, with some practitioners adjusting for gestational age and/or weight and others not. Some authors provide large bands within which clinical judgment is to be exercised; others provide narrower bands. In some of the material of the latter sort bilirubin levels are suggested for babies between 32 and 34 weeks gestational age, and then between 34 to 36 weeks. In other such material (e.g. a draft protocol discussed by Dr Crowle and others in 1999) the serum bilirubin levels were set out for babies of 24, 28, 32 and 36 weeks. So a baby born just after 34 weeks gestational age would potentially fall between two bands on one table or on the dividing line in another. Fortunately, both Dr Ives and Dr Rennie agree that the chart apparently used by the nursing staff indicating that exchange transfusion should take place at 330 mmols/l may properly be taken as the benchmark for exchange transfusion in a healthy baby of Alistair’s age and weight.

Dr Ives’ evidence is that 50 mmols/l should have been deducted from 330 because Alistair was ill, and therefore the threshold for transfusion would have been 280. So, even if the rate of rise was of the order of 4.5 to 5 per hour, Alistair would have been in the danger zone in about 2.5 hours after the morning sample had been taken, i.e. by the time more or less of the morning ward round. So, given that on any view of the evidence it would take at least three hours to order blood, transport it from Cambridge and set up a transfusion, there arose an immediate duty to order that blood.

Dr Rennie said that there was a difference of opinion between responsible practitioners as to whether 50 should be deducted, and if so in what circumstances, and this was based on her experience of visiting a significant number of neonatal units. She herself would deduct 50 if the baby fell into a category of illness that warranted the deduction, but explained that the reason given by other responsible practitioners for not requiring such a deduction is that it leaves open a large area of subjective clinical judgment as to whether the baby is ill and if so, how ill, and those doctors who are less interventionist, having regard to the risks of transfusion, prefer to apply thresholds which are clearly and objectively understood.

The charts produced by Nurse Hope support a deduction if the baby is “ill (septic, hypoxic, or acidotic)”. The charts used when Alistair was born, as produced by Nurse Mynett contain no such warning. There were no protocols as such at the hospital in 1990, but a later protocol recommended a deduction if the baby was “very ill”, although the draft 1999 protocol referred to “sick” babies. Nurse Mynett told me that the defendant hospital does not now operate a 50 mmol/l deduction at all for illness. Given the wide variety of opinion and practice and accepting Dr Rennie’s expertise in assessing the different practices, I accept the submission by Mr Pittaway Q.C. on behalf of the defendant that the evidence falls short of proving that those doctors who decline to deduct 50 represent a body of irresponsible practitioners. Taking into account the absence of national standards, the continuing debate about the risk of intervening too early and the risk of leaving it too late, and the legitimate desire to find, if possible, straightforward criteria to assist doctors to know when they should intervene, I conclude that those who deduct 50 have not been proved to be acting illogically and without due regard for the relevant risks and benefits.

What, however, gives rise to a possibly more difficult question is the evidence of Dr Crowle and Dr Roberts that they would in fact have deducted 50 mmols/l themselves, whatever advice was or was not written on the hospital charts in 1990, if in their clinical judgment the baby was sufficiently ill to justify it. The interesting legal question then is whether there is a duty upon a doctor, actionable at the suit of a patient, to use reasonable care and skill in following a local practice which there was no duty otherwise laid down. Whether this question needs to be answered depends on whether Dr Crowle on his ward round on the morning of the 19^th March would or should have made the deduction. Both he and Dr Roberts say that they would not in fact have deducted 50 mmols/l, because Alistair did not fall into that category of ill babies calling for the deduction. Was that, or would that have been, an unacceptable clinical judgment to have made? Dr Rennie’s evidence is that it would in her professional opinion have been reasonable for the doctors to have decided on the morning in question that Alistair was not so ill as to warrant the deduction. They could properly therefore have applied the normal threshold when considering what steps, if any, should be taken to intervene in the light of the elevated bilirubin level of 267 and the rise overnight from 213. What evidence is there to the contrary?

Dr Ives at p.729 of the 1999 edition of his chapter in the textbook advises a reduction for sick babies, defined in Table 31.8 as those with “Rhesus disease, perinatal asphyxia, hypoxia, acidosis, hypercapnia”. Nurse Hope’s chart referred to sepsis as well as to hypoxia and acidosis. So, the criteria by which a baby is judged to be ill have varied between professionals, and indeed Dr Rennie referred to a paper by Hansen in 1996 in which 19 different “sickness” criteria were identified in an international survey. Be that is it may, if sepsis is one of the criteria, in Alistair’s case he was not in fact septic in that there was no evidence of septicaemia. Dr Ives relies, however, on the presence of acidosis which was being corrected but had yet to correct completely. Both Dr Ives and Dr Rennie agreed, as I understand it, at the expert’s meeting that acidosis is not uncommonly something that has to be treated in premature babies, and it follows from that, it seems to me, that the degree and persistence of acidosis is a matter of clinical assessment. Dr Rennie considers that it was a mild respiratory acidosis and not such as to predispose Alistair to toxicity. Dr Crowle plainly has no specific memory of his thought processes on the 19^th March 1990, but looking at the records now he says that he too would not have considered Alistair to be so ill as to fall within the group of babies for whom a deduction of 50 should be made. He says that it would be a rare baby that would come within that group, and because Alistair was not persistently acidotic, he fell outside the cohort requiring a lower threshold of intervention. Dr Rennie, as I have indicated, agrees that that would have been a reasonable clinical judgment for Dr Crowle to have made. Dr Ives disagrees about the significance of the acidosis as it is recorded in the notes but crucially, when he came to answer the direct question about acceptable practice, he said words to the effect, “I agree that Dr Rennie’s view that Alistair’s sickness does not fulfil the criteria for lowering the threshold is within the spectrum of views that may be held, but I strongly disagree with it”.

This was a very fair and thoughtful and difficult concession for Dr Ives to have made, but in my judgment it leaves the evidence in a state where (a) Dr Crowle’s evidence that he would not in fact have lowered the threshold cannot be gainsaid; and (b) there is insufficient expert evidence to find that he would have been in breach of duty had he formed that view, and the balance of the evidence is to the opposite effect. I find therefore that there was no duty to deduct 50 mmols/l from the threshold of 330, and I find that neither Dr Crowle nor Dr Roberts would have done so.

I am left therefore with the evidence that Alistair would have been assessed as a baby whose bilirubin levels had risen overnight from 213 to 267 while he was under lights, but at the time of the morning ward round he was over 60mmols/l below the threshold for intervention by exchange transfusion. At an apparent rate of rise of 4.5 to 5 per hour a straight line projection would not have put him in the transfusion zone for 12 hours, with plenty of time to order blood if the rise persisted. In my judgment therefore the claimant has failed to prove a breach of duty either to proceed immediately to exchange transfusion or to order blood.

I now have to consider the case as further developed by the claimant, which is that the treating doctors should have been told by the laboratory about the 321 reading and should have been told about it during the early morning of the 19^th March. If Dr Crowle or any other doctor had been told, then he or she, it is submitted, would have been under a duty to order blood and to proceed to exchange transfusion immediately.

The defendant argues that there is no evidence upon which a finding can properly be made that the laboratory staff were under a duty to telephone the results through once the assays had been carried out, having regard to the evidence of Dr Rennie that 321mmols/l could properly be interpreted as within the normal range of readings for a term baby, and the technicians would have no reason to think that there was any sort of emergency indicated by the result.

I am not so sure. Dr Ives was astonished at the idea that the laboratory would have waited until the late afternoon to communicate the results. Plainly there is an evidential difficulty in that it was not fully appreciated until the trial that it was going to be suggested that the 321 result was not known during the morning of the 19^th March, as I have found in all probability was what happened. For myself, the inference that I draw is that I would expect a laboratory to telephone serum bilirubin test results through to the SCBU (and here one can see from the printout that the laboratory knew that Alistair was receiving total parenteral nutrition), so that the split result can be of use while the baby is being monitored. It would seem odd to wait for 6 to 8 hours if it is possible that the baby’s bilirubin is being monitored 4 to 6 hourly I would therefore expect the laboratory to phone the result through while it was likely to have been of assistance to the treating nurses and doctors, and in the absence of any cogent reason as to why that was not done I would hold that the laboratory was under a duty to do so. The evidence that the request was for a routine test does not persuade me to the contrary. So although the evidence tends to suggest that the results were probably not telephoned through, as I have found above, in my judgment they should have been.

At what time then should the doctors have been made aware of the 321 result? The first question, though not one explored much if at all in the evidence, is what level of urgency was attached to the request to the laboratory? Dr Crowle thought that it was probably a routine request, in the absence of anything on the printout to show that it was an emergency. Dr Rennie considered too that it would have been a routine request, and properly transmitted as such. There is no direct evidence as to who sent the serum bilirubin sample off to the laboratory, or whether that person knew the ward result when the laboratory sample was dispatched. In my judgment, therefore, to the extent that Dr Ives suggested that the person who sent it to the laboratory should have marked it “emergency” or “urgent”, there is not in my view the evidence to make that finding.

I am then left with Dr Crowle’s evidence that the laboratory equipment in 1990 was “antiquated” and that it could take three hours for a routine assay and even up to two hours for an urgent request to be turned around. I am not asked to find that the defendant was negligent in its provision of laboratory equipment, but I am asked to be sceptical of what Dr Crowle now says, having regard to Dr Ives’ view that, in effect, it is difficult to believe that the defendant’s laboratory was quite as bad as that. Dr Crowle was a consultant charged with training nurses before 1990, and then later with helping to draw up protocols to do with the management of neonatal jaundice. I am not prepared to reject his evidence about the delays associated with the laboratory in 1990 in the absence of some cogent evidence that his memory must be seriously at fault.

My finding, therefore, is that the 321 result should have been telephoned through between two to three hours after the laboratory opened, (which I have found to be at 9.00 a.m.) i.e. between 11.00 a.m. and mid-day on the 19^th March, say 11.30 a.m.

Any nurse or doctor, on being told of the 321 reading, could have realised it came from a blood sample taken at about 7.30 a.m. if the chart with timed blood samples had been referred to. Alistair’s total serum bilirubin had risen from 213 at 8.15 p.m. on the 18^th March to 321 by 7.30 a.m. on the 19^th March, if the both of those results were accurate. That would appear to be a rise of over 9mmols/l/h. As I have already found, there was no formal learning or training in 1990 that associated a high rate of rise in serum bilirubin levels with the need for immediate exchange, as opposed to the need to determine whether the jaundice was pathological and to be treated accordingly. On the other hand, there is force in Dr Ives’ opinion that a very high rate of rise in a case of apparently physiological jaundice would be very concerning. So what would or should the treating doctor have done?

The unusual feature of this case is that both Dr Ives and Dr Rennie agree that it is quite possible for the ward bilirubinometer and the laboratory assay to differ for various reasons by the difference between 267 and 321. As noted earlier in this judgment, the earlier and later split results also demonstrate different readings, with the ward reading having apparently been higher on other occasions than the laboratory result. So one is entering into a world of some conjecture as how reliable the readings were at any one time or in what direction, a feature that adds an unfortunate element of speculation, not least because of the absence of any pattern. Nonetheless, in the light of the fact that Alistair did suffer from kernicterus, it seems to me that the laboratory result at 321 for total serum bilirubin was more likely to be right than the ward reading at 267. But what would or should the nurses and doctors have made of it between 11.00 a.m. and 12.00 p.m. on the 19^th March?

If no other ward results were available by that time, Dr Crowle, or whichever SHO or registrar was informed, would have had two very divergent total serum bilirubin results from the 7.30 a.m. sample. If the ward readings were correct, the gradient from 163 on the 17^th to 189 on the morning of the 18^th to 213 on the evening of the 18^th, to 267 on the 19^th represented a concerning but not a steep gradient. However, if 321 was right, the gradient overnight had got very much steeper and, if continued, would already have taken Alistair into the danger zone of 330 by 8.30 a.m. to 9.00 a.m. if the total conjugated and unconjugated levels had to be added together. But the doctor would only realise that if he or she appreciated that the blood for the 321 result had been taken at 7.30 a.m., but for present purposes I assume that he would have checked that or have presumed that both readings came from the same sample..

If Dr Ives is right, the risk of Alistair’s serum bilirubin level going above 330 was so high at that point that it would appear on the face of it obvious that blood should have been ordered and an exchange transfusion set up. The defendant, however, contends that if only the unconjugated component of 266 had to be taken into account by the treating doctor, Alistair was still well outside the danger zone when that sample was tested, and so an immediate transfusion was not required.

Dr Rennie considers, and as I understand it Dr Ives does not disagree, that faced with the 321 reading and the split result, the doctor on the ward would at some point run a check with the ward bilirubinometer. In his chapter in the textbook Dr Ives in fact says that one should go further and obtain an up-to-date laboratory result before proceeding to exchange transfusion, but in fairness to him qualifies that by advising that if such a facility is not immediately available the transfusion should go ahead nonetheless.

Dr Crowle’s evidence was that that if he had been presented with the 321 reading he would not have done an immediate test but would have taken comfort from the fact that the unconjugated element was 266, and so still well short of the exchange threshold of 330 total serum bilirubin. In general it is agreed that the conjugated element is usually such a small proportion of the total serum bilirubin that the charts used by nurses and doctors and the working advice given to them for treating patients can be given in total serum bilirubin, rather than in a split result, because for all practical purposes the two are usually more or less the same. However, there is a conflict of evidence between Dr Ives on the one hand, and Dr Rennie and Dr Crowle on the other, as to whether it would be unacceptable to make a clinical judgment based purely on the unconjugated element.

In the 1970s bilirubin charts were drawn by reference to the unconjugated element only, and the defendant and Dr Rennie relied on a paper by Cockington to illustrate this. Those charts continued to be in use for many years, although at the defendants’ hospital they had been replaced with charts based on total serum bilirubin by 1990. Nonetheless, the tenor of Dr Rennie’s evidence and that of Dr Crowle was that there were exceptional cases for which routine charts did not cater, where the conjugated, i.e. non-toxic bilirubin level was so high that a clinical decision could properly be made after deducting that element, provided that the conjugated component exceeded 50 mmols/l. So if that were done in Alistair’s case, his conjugated serum bilirubin at about 7.30 a.m. on the 19^th was 266 (i.e. 321 less 55) and so well outside the danger zone.

Dr Ives’ very firm opinion was that this was a dangerous and unacceptable practice, particularly if no deduction had been made for illness. His evidence was that we do not know enough about what happens in cases of high levels of conjugated bilirubin to take the risk of excluding the direct bilirubin in making a decision as to treatment. Conjugated bilirubin competes for binding to albumin and it is not known how much may fail to bind and become free and capable of passing the brain/blood barrier. And so for those reasons, more fully developed in his chapter in the textbook, clinicians should avoid the risk.

In his writings, however, Dr Ives refers to the debate on this topic and indicates that various views have been expressed in favour of deducting the conjugated element, whether at 50 mmols/l or above, or by reference to percentages of the total serum bilirubin. Mr Pittaway Q.C. relies on the following passage in the 1999 edition: “Most treatment guidelines rely on total bilirubin level, but faced with the decision as to whether or not to perform an exchange transfusion, subtraction of a conjugated component of › 50 mmol/l would appear logical. Others recommend the more cautious approach of not subtracting the direct bilirubin concentration from the total until it reaches 50%.” Now I accept Dr Ives’ evidence that he was not meaning in that passage to convey the message that because it seems logical to subtract the direct component, therefore he recommends or even agrees with the practice, and in the 2005 edition he has made his position much less ambiguous. But the fact remains that his 1999 edition conveys a clear impression that the question of subtraction or no is still part of the debate among responsible neonatologists, and that there are the two schools of thought. Later on in that section he recommends that guidelines should be agreed and adhered to and that excuses for not doing exchange transfusions should be discouraged; but he does not spell out what the guideline should be in this regard, even for ill babies, and his views as expressed in that chapter fall well short in my judgment of putting the responsible practitioner on notice that subtraction of the conjugated component is a practice that should be abandoned unless the component exceeds 50% of the total serum bilirubin level.

By contrast, Dr Rennie and Dr Crowle both say that they would have deducted the 55 conjugated component. Dr Roberts says that if she had seen the split result her treatment would have been the same when she came to see Alistair in the early afternoon, and she too would have taken the conjugated element into account. Dr Rennie also relied on the 1986 edition of Roberton’sTextbook on Neonatology and the chapter written by Dr Mowatt (whom Dr Ives succeeded) in which the unconjugated component only is identified as indicating the need for exchange.

Mr Utley submits that I should treat the evidence of Dr Crowle and Dr Roberts as to what they would actually have done in 1990, if they had known about the 321 reading, with a degree of scepticism because it is difficult for even the most conscientious doctor to admit to something against his or her interest when he or she may be open to criticism. But even if their replies were disingenuous, which I do not believe, even Dr Ives seemed to me to be somewhat equivocal about whether the subtraction of direct bilirubin would fall outside the bounds of acceptable practice. His answer in re-examination was firm enough, but in cross-examination he spoke of heightened awareness of risk and changing views rather than breach of duty. And so, even if I were to accept that there are now strong grounds for taking the cautious approach and abandoning the “subtraction” methodology, the weight of the evidence to my mind comes down firmly on the side of the proposition that in 1990 there were different schools of thought and that it was perfectly legitimate to hold the view that if the non-toxic component exceeded 50, it could properly be deducted before deciding to embark on an exchange transfusion.

To some extent, however, this conflict of expert opinion, though hotly debated, may not be all that relevant, because it seems to me improbable that Dr Crowle or Dr Roberts or any other doctor on the ward when the 321 result was phoned through would have made a final decision one way or the other about an exchange transfusion without any other information. There are two possibilities:

The starting point is what was or should have been the monitoring regime in the light of the 267 total serum bilirubin result earlier in the morning. Dr Crowle’s evidence was that he would have expected Alistair’s serum bilirubin level to be checked every 6 hours on the basis of a reading of 267, but would have increased that to every 4 hours after receiving the 321 reading. Dr Ives said that at 267 he would do the next ward test at 4 hours and then decide whether or not to relax the gaps between tests in the light of the result. The nursing evidence was not really precise enough to know whether in fact the ward tests were being done 4 or 6 hourly on the morning of the 19^th March, but on the basis of Dr Crowle’s evidence I think that it is more likely that the ward tests were being done every 6 hours.

However, the question remains as to what the doctor on the ward would or should have done on receipt of the 321 reading, given that a moment’s reflection would have highlighted the fact that there were two very discrepant readings for morning of the 19^th March. If the doctor knew that the samples related to blood taken at 7.30 a.m., he or she could easily have worked out that, if the 321 result was right, the overall rate of increase from the night before was over 9 mmols/l/h.

The unconjugated component, however, was 266, and the doctor could properly take that as the starting point for the worst prediction for Alistair, as I have found. He would not know how much the indirect component had risen overnight or at what rate in the absence of a split result from the night before, and he may not have checked the precise times the bloods were taken to establish the exact rate of rise as between the two results. But if the true rate of increase was 9 mmols/l/h and, say, 8 mmols/l/h/was attributable to the unconjugated component, then there was a risk that in about 8 hours Alistair would be in the danger zone. So, if a test had been done on blood taken at 7.30 a.m., it should have been within his contemplation that Alistair might be in danger by about 3.30.p.m

At one point in his evidence Dr Ives indicated that he would have done a check on the ward bilirubinometer on seeing the 321 result, but in cross-examination he appeared to accept without demur the choice of seeking another laboratory test, though that would have involved a two hour delay. So I am a little unclear as to what he says the duty would have been in terms of the urgency of carrying out a fresh test upon receipt of the 321 result.

There is the possibility, referred to above, that on 6 hourly testing Alistair could be due for another test anyway at 1.00 p.m. to 1.30 p.m., a time at which bloods were again taken according to the chart, giving the doctor the option of waiting until then to check whether it had been the 267 or the 321 result that was more likely to be right.

The claimant’s real case is that the doctor on receiving the split bilirubin results should have proceeded to an exchange transfusion without delay, but having regard to the difference between the two readings from the 7.30 a.m. sample, and to Dr Ives’ own opinion as I understood it that one checks the result anyway before moving to transfusion, I am sure that any doctor would want to clarify whether 321 was or was not an “outlier,” as it was called in the evidence. The only question then is whether the doctor would or should have done it on the ward or in the laboratory, and whether he would or should have done it immediately.

If the 321 result had been telephoned through before 11.30 a.m. and ward test done immediately, then the outcome would have been known at around 11.30 a.m. (when blood was actually taken through the arterial line). If the doctor went back to the laboratory the re-test would not have come back until 1.30 p.m. or thereabouts. If the doctor waited for the next 6 hourly check he or she would know between 1.00 and 1.30 p.m. So unless the doctor had been able to use the 11.30 a.m. sample to check, or was under a duty to check the result on the ward immediately if the 11.30 a.m. blood had already been sent off by the time the 321 level was known, the re-check would not be available until 1.00 to 1.30 p.m. The difference in timing of course is only material if, on the balance of probabilities, the serum bilirubin level would still have been going up, because there no suggestion by Dr Ives that the medical staff should have proceeded to an exchange transfusion if the level had fallen. It is also only material if it would have made a difference to Alistair’s outcome.

In my judgment, having heard Dr Crowle, I think it likely that he would have waited for the next ward test before deciding whether to proceed to an exchange transfusion or to continue to monitor the serum bilirubin levels, and it seems to me that that would have been within the band of decisions that it was professionally acceptable to take, given the high conjugated component of the 321 reading, even if it meant waiting until 1.00 p.m. Whether in fact he would have used the 11.30 a.m. blood sample or waited until 1.00 p.m. is, however, purely speculative, because on the evidence there is no way of deciding if the 321 result would probably have come to him before or after 11.30 a.m.

Would he have ordered blood? In his evidence he said that he would have considered it, but probably would not have done so that stage. I noted that in cross-examination he paused for a very long time before giving his final answer, and I have come to the view that, whether it took 3, 4 or 5 hours for whole fresh blood to be received from Cambridge, it is likely that with the potential for a further rapid rise in the bilirubin level he would in fact have taken the precautionary approach and ordered blood.

So, if a ward test had been done at 11.30 a.m. or 1.00 p.m. what would it have shown? Thiat involves consideration of following factual questions:

In order to form a view as to what happened or should have happened, and in particular whether the serum bilirubin level was still rising, it seems to me that I have to look here at all the evidence including some of that relating to causation.

It is the opinion of Dr Ives that the serum bilirubin level must have continued to rise after 7.30 a.m. because the level reached at that time was too low to have caused kernicterus. If 321 was the correct reading for total serum bilirubin at 7.30 a.m., then according to his calculations it could have reached 386 by the time of the operation at 3.30 p.m. If 267 was the correct reading it could have risen to 293 at the time of the operation and 302 by the end at 4.30 p.m.. Dr Rennie, on the other hand, says that the serim bilirubin had reached a plateau and that “low bilirubin kernicterus”, though very rare, is a recognised phenomenon and one which she has personally come across twice before, once in practice and once in advising in a medico-legal capacity. In her view it was not rising and whenever a re-test would have occurred, it would not have suggested danger.

Both of these conclusions involve degrees of improbability. Neither Dr Ives nor the defendant’s expert paediatric neurologist, Dr Rosenbloom, had ever come across low bilirubin kernicterus . In the case of the baby under Dr Rennie’s care who developed a low bilirubin kernicterus, and in respect of whom Dr Rennie presented a reconstructed chart, we do not know, as Mr Utley submitted, whether the degree of acidosis or growth retardation in that case significantly increased the risk that lower levels of bilirubin would be toxic. There is, it is agreed, no medical literature in the United States or Europe in which any baby has been identified with kernicterus after a peak bilirubin reading of 267. There are, however, some Japanese case studies which Dr Rennie considers to be reliable because of the extensive use of MRI in Japan, but which Dr Ives considers, for the reasons he gave in answer to Q.61 at the experts’ meeting, not to be clinically robust. And he makes the point that if clinicians really did come across such cases there would more pressure to lower the threshold for intervention rather to relax them. Dr Rennie, however, is very experienced and in my judgment did not commit herself lightly to her opinion.

The difficult with Dr Ives’ hypothesis, as he foreshadowed in paragraph 2.24 of his report, is that the total serum bilirubin level after the operation (which I have found to have been tested at 5.00 p.m.) was 272. His answer to the apparently dramatic fall from 386 to 272 (assuming that the serum bilirubin was still rising at 3.30 p.m.) is that a significant proportion of Alistair’s bloodstream was diluted with fluids before, during and after surgery, consisting of plasma, saline solution and sodium bicarbonate, with the result that a diuretic, frusemide, was infused and he was diet restricted. He also hypothesised an efflux of bilirubin from the vascular space to the brain or a severe acidotic episode, although he found no evidence in support.

Dr Rennie, to my mind convincingly, explained that bilirubin is not very solvent in water and that as it is removed from the circulating blood, further bilirubin is absorbed from the tissues. Hence the need to replace the whole of the baby’s blood twice in an exchange transfusion. In addition there is no evidence that sodium, which is soluble, went down during the operation, and the blood tests taken after surgery showed a rise in sodium. Dr Ives’ theory that the rise could be accounted for by the amount of sodium bicarbonate administered after Alistair’s return from theatre was weakened by his recognition that he had mistakenly calculated the solution at 8 mls when in fact it was 4 mls. There was also the point that for about two hours Alistair would have been taken off phototherapy and therefore the additional effect of the lights would have been absent. Dr Rennie also relied on the fact that the serum bilirubin reading was 282 on the 20^th March and that this tended to suggest that there was no dramatic decline attributable to dilution. Looking at the matter in the round, quite apart from the problems identified by Dr Rennie, it seems to me that Dr Ives’ opinion involves a hypothesis that has not been considered in any literature and neither he nor Dr Rennie have actually come across the suggested effect in practice.

I prefer Dr Rennie’s evidence, therefore, on this aspect of the case but on any view Alistair’s presentation, so far as serum bilirubin levels are concerned, was atypical.

Of course, it does not follow even on Dr Ives’ view that the serum bilirubin level was necessarily rising right up to the moment of surgery, because kernicterus develops, according to the unchallenged evidence, not immediately upon a peak being reached but as a result of a combination of a peak level, the persistence of a high level and the predisposition to toxicity that may exist in cases of severe illness. So too it does not follow that the total serum bilirubin must have been at a plateau throughout the day, as Dr Rennie considered to be the case, simply because there was a reading of 267 at 7.30 a.m. and 272 at 5.00 p.m. For example, if the rise was 9 mmols/l/h from the true reading of 321 at 7.30.a.m., the total serum bilirubin by 9.00 a.m. would be of the order of 335, a reading that would be within the range culled by Dr Ives from some of the literature, and referred to by him in paragraph 4.15 of his report, even if one allows for the high conjugated component.

I also noted Dr Ives’ evidence that at 302mmols/l it is unexpected for bilirubin to pass the blood/brain barrier, but it can happen. Although Dr Ives did not say in terms that his figure of 302, as a possible level at which kernicterus can occur, was or was not a total serum bilirubin readingI, I have assumed from his starting point of 267 that that is what he meant. In assessing therefore statistically whether low bilirubin kernicterus exists and whether a particular case comes within an unusual but possible category, or within one that is so rare as to be labelled “low bilirubin kernicterus”, it seemed to me that on Dr Ives’ evidence there might be cases that would appear to be in the former category even though the conjugated element might take them below 300 for the unconjugated component.

If, therefore, there had been some further but not persistent increase, the level could then have started to decline from 335 or some figure in that region and the reading of 272 at 5.00 p.m. becomes easier to explain. Of course, it still remains the case that the conventional teaching according to Nurse Mynett and Dr Upson’s notes is that kernicterus is very unlikely “if SBR ‹ 340 mmols/l” (p 115i of the trial bundle), a proposition with which Dr Ives would agree, but whatever the explanation for Alistair’s kernicterus it has to involve some degree of improbability, because even on Dr Ives’ evidence, as I understood it, he accepts that kernicterus could have occurred at a peak level of 302.

In the course of the trial I canvassed the possibility that the peak might have been reached before 7.30 a.m. Dr Ives considered that that would have involved a rate of rise more consistent with “aggressive haemolysis”. Dr Rennie referred to the possibility without adverse comment as something that could have occurred, and Dr Rosenbloom also thought it possible: but I have to accept on the evidence and on the basis of Dr Ives’ expertise that if there had been an earlier peak, it cannot have been much above 321.

What significance, if any, should then be attached to the absence of any mention in the nursing or medical notes to any serum bilirubin level between 7.30 a.m. and 5.00 p.m.? Dr Ives in his report had wrongly assumed in the absence of a bilirubin chart that the hospital were simply taking no adequate steps to monitor serum bilirubin at regular intervals at all, and therefore, as I understand his evidence, he took it that the medical staff negligently failed to carry out the test that would have shown that Alistair was in the danger zone on a rising gradient. Mr Utley’s submission now is that other charts disclosed by the defendant, including one for the 19^th March, do establish that there was a system of regular monitoring, but they show significant gaps in time between the levels being marked on them. So, even though the hospital had a system, the inference, he submits, is that nobody actually carried out a check between 7.30 a.m. and 5.00 p.m., and it may be that the eye was taken off that particular ball because of Dr Roberts’ concern about Alistair’s possible volvulus, and the preparations for the laparotomy.

The evidence of the nursing staff, however, is that a baby under lights would be monitored every four to six hours, but sometimes the doctors forgot to make an entry on the chart. Staff Nurse Hope, Nurse Lee and Nurse Practitioner Mynett all seemed to me to be dedicated and competent, and I was told that either a doctor or a nurse might do the ward bilirubin test with blood from an arterial line, depending on who was available. Dr Roberts wrote full notes and was an impressive witness; and Dr Crowle was, as I have noted before, a consultant relied upon to assist in the drafting of protocols and in the training of nurses. Looking at the events of the 18^th/19^th March apart from the allegations with which I am concerned, the nursing notes are good, the various charts properly completed (apart from an absence of times on some of the columns in the Data Flow Chart), and nothing stands out as indicating that systems were not adhered to, save for doctors who may have failed to enter bilirubin results on the chart at the foot of the incubator.

On this evidence I find it very unlikely that no ward serum bilirubin test would have been done at all between 7.30 a.m. and 5.00 p.m., particularly as bloods were taken at 11.30 a.m. and 1.00 p.m. Dr Crowle said that in his judgment, on a ward reading of 267 six hourly checks would have been sufficient. Nurse Lee also said that when the baby was under lights the bilirubin tests were done every six hours, and I have already found that it is probable that the tests were being done every six hours.

Dr Ives said in his report that having regard to the rate of rise and the level of 267 serum bilirubin should have been tested three hourly, although having heard the nursing evidence he was content with testing every four hours. I am not persuaded however that testing every six hours, if that is what happened, would have involved a breach of the duty of care towards Alistair.

So, whether the routine test was at 11.30 or at 1.30 p.m., was the reading, which is now lost in the absence of the chart, such that on the balance of probabilities it mandated an immediate move to exchange transfusion. In other words did Dr Brain and Dr Roberts (if the test was at 1.30 p.m.) or the SHO previously on duty at 11.30 a.m. (if the test was done then) have available a test result that required an exchange transfusion? If they did, then plainly they overlooked its significance and they failed to note it in the medical records, and the nursing staff also missed it.

On behalf of the claimant it can properly be argued that where an error has occurred it is not surprising that nobody has made a note of it at the time. On the other hand there is nothing in the notes either during the 19^th March or subsequently to suggest that anybody might have overlooked the significance of a bilirubin result that would have been entered in the bilirubin book; nor is there any hint by either the nurses or doctors that someone had failed to check Alistair’s bilirubin after 7.30 a.m. If either the ward serum bilirubin test had been missed for some reason, or it plainly showed a reading in excess of 330, the failure on anyone’s part to advert to it would, it seems to me, have involved a marked failure of the systems in place.

Certainly the doctor who wrote in the medical notes at 11.00 p.m. to query whether Alistair’s stomach problem might have something to do with the high conjugated component revealed in the laboratory test, the results of which were by then available, did not apparently suspect that something might have gone wrong during the day with the serum bilirubin testing.

On this evidence, is Dr Ives’ reasoning about the elevation in serum bilirubin that must have occurred to cause kernicterus so compelling that it should lead to the conclusion that either the nurses and doctors forgot to carry out the test that would have revealed a level well above 321, or that such a test was carried out but its significance negligently ignored?

For Dr Ives’ conclusion to hold good, his speculative theory about the dilutional effect of infused fluids has to hold good as well, and I find Dr Rennie’s response to that hypothesis convincing. I do, however, think that a ward serum bilirubin test was probably carried out at about 1.00 p.m., and so if as Dr Ives postulates the level was still rising at 9 mmols/l/h, then after five and a half hours from 7.30 a.m. the reading would have been over 370.

If Dr Ives is right, therefore, the doctor and/or nurse responsible for carrying out the test, noting it in the bilirubin book and entering it on the chart, missed the significance of a reading well in excess of the compulsory exchange threshold or for some reason failed to respond to it, and no other doctor or nurse ever commented on the error. For my part, there is so much speculation required in reconstructing what probably happened to Alistair’s serum bilirubin during the day of the 19^th March, and what the times were at which tests were or were not carried out, that the evidence does not establish on the balance of probabilities that his serum bilirubin was still rising at the time when the routine ward test either was or should have been carried out. It seems to me that if the serum bilirubin had risen above 321 for a time after 7.30 a.m. and then fallen back to a range of to 290 or thereabouts by 1.00 p.m. when the routine test was probably done, or if it had peaked before 7.30 a.m. and had fallen back, Dr Crowle would not have been alarmed at the routine result and would not have been required to proceed to exchange transfusion. What he would have been expected to do would be to monitor the results carefully, and the fact is that on my findings the next ward test was done at 5.00 p.m., which is four hours after the 1.00 p.m. test which I believe was carried out. So whoever was on duty at the time, the timings would consistent with a decision to test the ward serum bilirubin four hourly. In my judgment, such evidence as there is consistent with the medical staff doing their duty with a ward serum bilirubin test that never produced a reading above 300.

So to return to the hypothetical scenario of what would or should have haappened if a doctor had seen the 321 result at about 11.30 a.m., there is every chance that whether he re-tested then and there or waited until 1.00 p.m. the total serum bilirubin according to the re-test would have been lower than the 321 reading from 7.30 a.m., unless of course the ward bilirubinometer went haywire again. If, say, the level had risen at a rate of 9 mmols/l/h to 335 by 9.00 a.m., by 11.30.a.m it could have fallen back at the same rate just under 300. On a falling gradient, with a high unconjugated component Dr Crowle or Dr Roberts, or whoever was on duty, could then properly have given an instruction to check the bilirubin level after four hours, and so when it was checked at 5.00 p.m. and found to be 272 it would appear that Alistair was not in danger, and that indeed is what everybody believed to be the case until the diagnosis of kernicterus came to be made.

I emphasise that in postulating results of 335 at 9.00a.m. and so on I am not making findings. My purpose is simply to demonstrate that in my judgment there are so many variables where the direct evidence is now lacking that the court is being invited to speculate too much. I am, for the reasons given, not persuaded by Dr Ives that an inference should be drawn on the balance of probabilities that there was a rising level of bilirubin after 7.30 a.m. on the 19^th March 1990 at the times when the doctors were under a duty to check the serum bilirubin level. The evidence is consistent with serum bilirubin appearing on the ward to be within safe levels, and it seems to me more likely that Alistair’s kernicterus was caused with a peak total serum bilirubin at a level higher than 321, albeit with a significant conjugated element, than with an obvious rise well into the danger zone that was missed by the nursing and medical staff.

In fact, it is my view on causation that Alistair’s brain damage had probably occurred before the afternoon of the 19^th March 1990. Dr Rosenbloom, on whose evidence the defendant relied, is a vastly experienced Consultant Paediatric Neurologist. In his report dated the 9^th March 2006 he concluded that “the timing of this brain damage is likely to have been in association with the neurological abnormalities that were apparent on 18 and 19 March 1990”. In a letter dated the 4^th April 2006 Dr Rosenbloom indicated that :

Dr Rennie expressed a similar opinion in her evidence, and in her report considered that Alistair developed seizures on the 19^th March before the operation in the afternoon, referring in particular to “abnormal movements” that were thought to be “fits”. This was a reference to a medical note timed at 10.30 to 10.45 a.m.: “Noted to be making strange movements ?coughing ? fit”.

Dr Ives accepted in his evidence that the signs on the 19^th March were neurological signs, although he did not accept that proposition in relation to the nurse’s note the previous day, which he considered to be a commonplace observation in a SCBU. In his opinion the neurological signs in the morning were caused by an episode of hypoglycaemia, and he relied on the fact that until very much more obvious signs were noted after the operation in the afternoon there were no further reports during the day suggesting neurological impairment. He also relied on the entries in the notes made after the operation in which it is apparent that very much more obvious and dramatic signs of neurological damage were being recorded.

Dr Rosenbloom responded that Alistair had had previous episodes of hypoglycaemia without signs suggestive of neurological damage being noted, and that during the morning of the 19^th his blood glucose went down to 1.4 but not to zero, and that it recovered with extra dextrose. In his opinion, and he said he had experience of neonatology as well as paediatric neurology, the entry “?coughing ? fit” was a very unusual entry and suggested abnormal neurological behaviour, and he did not think a hypoglycaemic convulsion was the explanation. He did not consider that the absence of any further reference in the notes to “subtle” signs of neurological damage was significant. Dr Rennie supported his reasoning and conclusion.

There is no doubt that Dr Rennie’s opinion is weakened by the fact that in her report she said that Alistair had had seizures on the 19^th March, without referring to the nursing notes for 18^th , whereas in her oral evidence she said that the notes for the 18^th suggested neurological impairment. And it does seem to me that the notes for the 18^th are much more equivocal in terms of what they denote than the nursing and medical notes for the 19^th March. The real difficulty for the claimant is whether Dr Ives’ reasoning as a neonatoligist is more convincing than Dr Rosenbloom’s as a paediatric neurologist, with some neonatal experience. Dr Rosenbloom, it will be recalled, was the defendant’s expert who alerted Mrs Green to the diagnosis of kernicterus and I am satisfied that there is no element of conscious or unconscious partiality in his evidence. In the circumstances I do place considerable reliance on Dr Rosenbloom’s evidence, and I find that Alistair was suffering neurological damage by 10.45 a.m. on the 19^th March 1990. Having rejected the evidence that such damage was caused by hypoglycaemia, and there being no evidence of any other cause apart from kernicterus, I conclude that by 10.45 a.m. toxic levels of bilirubin had penetrated Alistair’s blood/brain barrier.

To my mind such a finding is consistent with Alistair’s total serum bilirubin having peaked somewhere in excess of 321, whether before or after 7.30 a.m., and then declining during the rest of the day to 272. For reasons which it is impossible now to know, that level was enough to be toxic, and nothing that Dr Crowle or any other doctor could have done after that time could have saved Alistair from his grievous disabilities.

In the result, the question of the time it would take to receive the blood from Cambridge to carry out an exchange transfusion falls away. Were it necessary to make findings, I would not disagree with Mr Uttley’s submission that on the evidence as a whole it would have taken three to three and a half hours from ordering blood to starting the transfusion. If, contrary to my findings, a ward serum bilirubin test at 1.00 p.m. had been so high as to mandate transfusion, the exchange would have therefore have been started at 4.00 to 4.30 p.m., which even on the claimant’s case would have been too late. If the laboratory test result had been available at 11.30 a.m. and Dr Crowle had ordered blood, the transfusion could then have been started by 2.30 to 3.00 p.m., which is so close even on the claimant’s case to when the damage occurred that I could not find that it would probably have averted the kernicterus. And in any event, for all the reasons I have given, the 267 reading was insufficient to warrant the ordering of blood, and nothing else occurred or should have occurred between 7.30 a.m. and 11.30 a.m. to create an expectation or a duty to order blood.

Kernicterus is, or should be, a preventable condition and Dr Ives, I am sure, is wise to promote the precautionary principle in setting treatment guidelines and to argue against the relaxation of the intervention thresholds. I am conscious that Alistair’s parents will feel exactly the same. They have looked after their son with devotion for many years, and I am sure hoped very much that the court would not only recognise their devotion but ease the future for them and for Alistair by awarding compensation for all his suffering and loss. It is, however, my sad conclusion that the evidence of what happened on the 19^th March 1990 to cause kernicterus contains so many gaps and is now so difficult to reconstruct with confidence, and the degree of speculation has become so great, that I am quite unable find that there was, on the balance of probabilities, a lack of reasonable care on the part of Dr Crowle, Dr Roberts or any other doctor on duty over the relevant period of time.. There was a lack of care on the part of the laboratory, but even if the split bilirubin result had been telephoned through during the morning of the 19^th March it would, I am afraid, have made no difference to the outcome.

The claim is therefore dismissed.