England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEE'NS BENCH DIVISION
MR JUSTICE JACK
HQ02X02464

		Royal Courts of Justice Strand, London, WC2A 2LL
		07/12/2005

Lord Justice Latham :

Introduction

1. Alice is the appellants' second child, born on the 26^th April 1992. She suffers from a severe malformation of her brain caused in the early stages of foetal development by the failure of her fore-brain to divide into two. The condition is called holoprosencephaly. She is quadriplegic, and is unable to use her limbs or talk. She has a gastrostomy and requires twenty-four hour care. Despite her very severe difficulties, she can recognise and respond to those who look after her.

2. In this action the appellants claim damages for the loss, pain and suffering which they themselves have suffered as her parents consequent on Alice's birth. The trial with which we are concerned dealt solely with the issues of negligence and causation. The negligence alleged was that of Professor Rodeck, who has been Professor of Obstetrics and Gynaecology at University College London since 1990 and is head of the Department of Obstetrics and Gynaecology at University College Hospital. He is a man of considerable distinction; whilst he was in a previous post, at Kings College Hospital, he had set up what was possibly the first unit in the world devoted exclusively to foetal medicine. The negligence which was alleged in the claim was that he had failed when carrying out an ultra sound scan of Alice while she was in utero on the 3^rd December 1991 to appreciate that the scan showed abnormalities of Alice's brain indicative of holoprosencephaly and to have advised the appellants accordingly. Had they been so advised, the judge, Jack J, found that the pregnancy would have been terminated. There is no appeal against that finding. The judge, however concluded that the appellants had not established that Professor Rodeck was negligent. He therefore dismissed the claim. The appellants appeal on the basis that the judge, on a proper analysis of the facts, was wrong.  As a result, this court has had to consider in some detail the evidence at the trial and has done so acutely aware of the fact that it has not seen or had the opportunity to evaluate any of the witnesses as they gave their evidence.

   Holoprosencephaly

3. This is a comparatively rare condition which has been stated to have an incidence of between one in 10,000 and one in 30,000 births and to be responsible for one in 250 abortions. The judge, in his judgement, set out the description of the condition given in the report of Dr Peter Twining, Consultant Radiologist at the Queen's Medical Centre in Nottingham, who was called as an expert witness on behalf of the appellants. It has not been suggested that this description can be bettered, so I will use it for the purposes of this judgment. In paragraph 2 of his report he said:

   "2.1 The normal foetal brain consists of two halves, each half is called a cerebral hemisphere and each hemisphere contains a ventricle (a thin fluid filled space). These ventricles communicate towards the front of the brain and then there is a midline channel through which fluid passes (third ventricle) which then communicates with the posterior part of the brain (cerebellum). There is a further fluid space within the cerebellum (fourth ventricle) which communicates with the outside of the brain and this is where the fluid (cerebral spinal fluid) leaves the brain and then is absorbed over the surface of the brain.

   2.2. In holoprosencephaly the process by which the brain separates into two halves does not occur. The result is a spectrum of abnormalities in which many of the mid line structures are absent and there is a variable shaped single ventricular cavity. The brain usually separates into two halves during the fourth and fifth weeks of pregnancy. There are three main types of holoprosencephaly.

   2.3 The most severe type is alobar holoprosencephaly. In this condition there is a single ventricular cavity and the thalami are fused. There is also absence of the midline structures such as the cavum septum pellucidum, corpus callosum and falx. This diagnosis of holoprosencephaly is quite straightforward as the brain is severely disordered and there is a large central fluid filled space within the brain.

   2.4. Semi lobar holoprosencephaly is a more difficult diagnosis to make, however once again there is a single ventricular cavity around the thalami, which are partially fused. The anterior parts of the ventricles are fused in a sickle or horse shoe shape and there is absence of the corpus callosum, cavum septum pellucidum and anterior portion of the falx. The posterior parts of the lateral ventricles will appear relatively normal.

   2.5 Lobar holoprosencephaly is the least severe form and in this condition the ventricles are almost normally formed however there is fusion of the most anterior parts of the lateral ventricles. The corpus callosum, cavum septum pellucidum and part of the falx are absent.

   2.6. It should be noted therefore that in all forms of holoprosencephaly there is absence of the cavum septum pellucidum. In both lobar and semi lobar forms there is absence of the anterior horns of the lateral ventricles."

4. For the moment, it is only necessary to elucidate some of the terms used in these passages. The normal brain, formed as it is of two hemispheres, has a cleft between the two which is occupied by the structure known as the falx. This structure cannot develop if the brain has not divided. Hence its absence or partial absence depending on the extent of the holoprosencephaly where that condition is present. The corpus callosum is the structure which provides the communication between the two halves of the forebrain. This structure will accordingly be absent where division has not occurred. The cavum septum pellucidum (the CSP) develops in conjunction with the corpus callosum at the relevant period of gestation, which is 18 to 20 weeks. It consists of two parallel structures with a fluid filled space which ultimately join to form a fibrous sheet. While it is developing, it is identified on ultrasound scans by the presence of two parallel lines. Again, where the forebrain has not divided, there can be no CSP.

   Alice's Condition

5. MR scans of Alice's brain since her birth have established that she has semi lobar holoprosencephaly, towards the severe end of the spectrum. As a consequence she has no corpus callosum and has never had a CSP. Her falx extends no further forward in her brain than the line between her ears. The anterior horns of her ventricles are absent. The ventricles join in a horse shoe or sickle shape. She therefore has what is known as a monoventricle. It is the fact that these structures were absent at the time that Professor Rodeck performed his ultra-sound scan which is at the heart of this case.

   The History

6. This can be taken from paragraphs 7 to 32 of the judge's judgment, reported at [2004] EWHC 2452 (QB). As both parties are agreed that these paragraphs accurately set out the history, I gratefully adopt them:

   "The history in detail

   On 28 November 1991 Mrs Lillywhite attended at the West London Hospital for a routine abnormality scan. This was carried out by Mrs Janet Wright, the superintendent radiographer in charge of ultrasound services at the West London and Charing Cross Hospitals. Mrs Wright graduated as a radiographer in 1976 and in 1982 obtained a diploma of medical ultrasound. She was an able and experienced sonographer but did not have the training of a doctor. Mrs Wright remembers the scan for two reasons: first because Mrs Lillywhite was anxious that she might have an abnormal child because of her age (she was 36), and second it was the first and only time that Mrs Wright had not found a cavum septum: she remembered it as "the case of the absent septum pellucidum". Mrs Wright used a Hitachi EUB 340 machine, which she described as not one of the latest but a very good machine. Despite a careful search in the course of which, as was usual, she recorded pictures, she could not find a cavum septum. She thought that the front of the brain was filled with brain tissue. Her pictures are available and three relate to the skull and brain. They would not show the same definition and differences of shading as would have been apparent to Mrs Wright on the screen and they may have deteriorated with age. Each gives a single view as opposed to the multiple views Mrs. Wright would have obtained as she moved the probe.

   Mrs Wright scanned the skull and brain of the fetus in the axial or transverse plane, that is, so that she obtained a sonographic image of it at the level across the skull at which she was directing her probe, which represented a horizontal slice at that level. She found that she was able to obtain echoes which visualised the structures in the posterior brain. She found the midline echo representing the third ventricle but she could not find a cavum septum nor could she find the anterior horns of the lateral ventricles. She did not refer to the anterior part of the falx either in her evidence or expressly in the note she made on the examination form. That note read: 'unable to visualise septum pelucidum and normal anatomy in the anterior brain. ? absent corpus callosum ? normal'. Mrs Wright described how she found the cerebellum at the back as a marker and then moved up, in tiny movements, of perhaps 5 mm. She did not find echoes to represent the cavum. She took measurements for the biparietal diameter and the head circumference at the level at which she assessed that the cavum septum should have been. She measured the diameter as 39 mm, which I mention to give an idea of the size of the fetal head at this stage of development. She said that she tried moving the probe through a number of angles trying to find the cavum septum. She went on until she was satisfied that she could not find it. She did not try a coronal view, that is, to take vertical slices across the head, parallel to a line drawn ear-to-ear. She did not see that there was a single ventrical across the midline of the fore-brain. It would have shown up as a single fluid-filled cavity, had she been able to see it. She thought that the frontal part contained brain tissue only – as it did. She did not identify a small choroid plexus cyst, which was later found by Doctor Meire and by Professor Rodeck, but her recent training had been that these were of no importance at 18 weeks. Mrs Wright did not have any technical difficulties in obtaining a satisfactory picture. Mrs Lillywhite was not of a weight where her size presented any problem. Although that was raised in advance of the trial, it was accepted at the trial that with a body mass index of just under 25 Mrs Lillywhite's build should not have been the cause of any sonographic difficulty. Mrs Wright said that she extended the time she spent on the examination beyond the normal. Mrs Lillywhite said that the examination seemed to take a long time and was longer than the subsequent examinations by Dr Meire and Professor Rodeck.

   Following her examination Mrs Wright spoke to Mr Pawson, the obstetric consultant in charge. She believed that the absence of a cavum septum might mean the absence of the corpus callosum, but was uncertain as to the seriousness. It was her intention, with which Mr Pawson agreed, that he should refer Mrs Lillywhite to Professor Rodeck at University College Hospital. Such a referral was standard in such a situation. Mrs Wright had made numerous referrals to Professor Rodeck. She had a high opinion of him. The referral was made by Mr Pawson by telephone while Mrs Lillywhite was still at the West London Hospital and was for 3 December. The University College Hospital referral sheet was completed by Professor Rodeck's personal assistant, from what Mr Pawson told her. She noted 'can't visualise septum pelucidum. ?anterior horns. ? small head ? microcephaly'. The last two originated with Mr Pawson and were not matters which Mrs Wright had raised.

   Mrs Lillywhite had some discussion with Mrs Wright as to the outcome of the scan and knew that Mrs Wright had not been able to find the brain structures which she wished to find. Mrs Lillywhite made a note of the words 'corpus callosum'. She was very anxious and was concerned that, if she was to have the pregnancy terminated, it should be done as early as possible. She felt she could not wait to see Professor Rodeck on 3 December and so telephoned the Portland Hospital and arranged to see Dr Hylton Meire privately on 29 November. Dr Hylton Meire was consultant radiologist at Kings College Hospital and was also in charge of the ultrasound service at the Portland. He was a specialist in ultrasound scanning, both obstetric and non-obstetric. In 1991 his work was about 70% non-obstetric. He had written widely on ultrasound, and was the editor of a leading text book. He had lectured on ultrasound and Mrs Wright had attended his lectures. He had and has a high reputation.

   The Portland accepted Mrs Lillywhite without any formal referral but obtained the details of her G.P. Mr. Lillywhite accompanied her to see Dr. Meire. Mrs Lillywhite explained to Dr Meire why she had come, telling him what she could of the outcome of the scan by Mrs Wright. I find that he did not have any written report from the West London Hospital: for Mrs Lillywhite did not have that. It is nonetheless clear that Dr Meire understood that Mrs Wright had failed to find the appropriate structures in the fore-brain. That is what he proceeded in particular to examine though he scanned the whole fetus. Dr Meire has no recollection of carrying out the scan. He would have recorded a large number of pictures, between 20 and 30. These have been destroyed following the policy of the Portland to destroy such pictures after 6 years if they had not been examined in the interval. Dr Meire was unaware of that policy until he requested the pictures to assist with this case. Mrs Lillywhite's evidence was that she told Dr Meire of her appointment with Professor Rodeck on 3 December. Mr Lillywhite's oral evidence supported that. I will revert to this and I conclude that the probability is that Dr Meire did know.

   Mrs Lillywhite's recollection was that Dr Meire told her that the baby was high in the uterus and he had not got a good view, but overall it looked fine. He told her that he had seen a choroid cyst. Mr Lillywhite's recollection was that Dr Meire was fairly reassuring although he said that he had not got a good view. Mr Lillywhite remembered feeling unhappy that it was costing money but Dr Meire wasn't getting a clear view. He said that he did not feel particularly reassured but his wife was more cheerful following the appointment.

   Dr Meire wrote a report of his examination as follows:

   The fetal size is consistent with a gestation of just over 18 weeks. The fetal head was high in the fundus and imaging was not easy with reverberations partially obscuring the proximal hemisphere. The gross cerebral anatomy is normal and I was able to identify the thalami, septum pelucidum, falx, third ventricle and lateral ventricles which appear normal. There is a small (and probably insignificant) cyst within the left choroid plexus.

   If doubt about the cerebral anatomy persists a further scan in about 2 weeks would almost certainly be more informative.

   It is clear that Dr Meire considered that he had seen echoes representing the cavum septum, the falx and the lateral ventricles. He would have used the echo of what he believed to be the cavum septum to give him the level at which to measure the biparietal diameter. But Alice had no cavum, no lateral ventricles nor did the falx extend into the front of the skull, which is where Dr Meire would have identified it. Dr Meire said that he would have tried to obtain a coronal view as well as the standard axial view. There was some discussion at the trial of the significance of the last sentence in his report, beginning 'If doubt persists …..'. Dr Meire said that this referred to the scan being two weeks earlier than he considered optimal, which was at 20 weeks. (I should say that there has been a shift of opinion as to the optimal time moving from 18 weeks to 20 weeks. Skull thickness does not begin to be a problem until 24 weeks. The downwards factor is the date by which any termination should be carried out.) He also said that it was a routine phrase which was not specific to Mrs Lillywhite. The question arises as to who was to have the doubts if Dr Meire himself was confident – for neither Mrs Lillywhite's GP nor obstetrician was in a position to question his conclusion. If Dr Meire knew that Professor Rodeck was seeing Mrs Lillywhite in four days, it is puzzling that he referred to the possibility of a further scan in two weeks. This has led me to question whether Mrs Lillywhite is right that she told Dr Meire of the further appointment. It seems to me that someone might be reticent in telling one consultant that they were seeing another consultant for the same purpose in a few days. I have also wondered what Mr and Mrs Lillywhite would have done if Dr Meire had told them that he was confident that Mrs Wright was correct: would they have still gone to Professor Rodeck for a third view? On the other hand some explanation was required why Mrs Lillywhite had come to see Dr Meire privately without a proper referral in a situation where it was standard procedure for there to be a referral to a tertiary referral centre. On balance I should accept that Mrs Lillywhite's recollection that Dr Meire knew of her appointment with Professor Rodeck is accurate.

   Dr Meire stated that his practice in cases where he could not be certain was to offer the patient a further scan without payment. He said that this happened in between 5 and 10% of cases. He said that he did not do that here because he was content in his mind that he had seen the structures he expected to see: if he had been uncertain, he would have offered a further scan. In the situation here, if he had been uncertain, there was also the forthcoming appointment with Professor Rodeck. He did not sound any note of uncertainty to be passed to Professor Rodeck. Nonetheless I conclude that the opening of his report on the scan he carried out and its closing do suggest that the scan was rather less satisfactory than it might have been. Mr. Lillywhite was left with that impression and was less reassured than his wife.

   Professor Rodeck has been Professor of Obstetrics and Gynaecology at University College London since 1990, and is Head of Department. When at King's College Hospital he had set up what was probably the first unit in the world devoted exclusively to fetal medicine. In 1986 he moved from King's to Queen Charlotte's Hospital where he established a second such unit. He then established the fetal medicine unit at University College Hospital. He has published very widely and has been involved with a number of developments in fetal medicine. He is a man of considerable distinction. In 1971-2 he was a research fellow at University College Hospital in fetal and perinatal pathology, so he is well-grounded in the anatomy of the normal and abnormal brain.

   Mr Lillywhite was not able to accompany his wife to Professor Rodeck on 3 December. Professor Rodeck has some recollection of the examination despite the passage of time, almost certainly because of his involvement after the birth of Alice. He said that when he heard how Alice was affected it came as a great shock and he went back to his records and had thought long and hard how it might have been possible that he had not diagnosed her condition.

   The paperwork which Professor Rodeck had, consisted of the referral sheet prepared by his personal assistant, and Dr Meire's report. I can be confident of the latter because it was included in the University College file for Mrs Lillywhite. She had been given a copy of the report at the Portland and must have provided it, or a further copy, to Professor Rodeck. Professor Rodeck had the assistance of Dr Martin Haeusler while he carried out the scan. Dr Haeusler is now consultant obstetrician and gynaecologist at the University of Graz in Austria. In 1991 he was senior clinical research fellow at University College Hospital and worked with Professor Rodeck in the fetal maternal unit from February 1991 to March 1992. Dr Haeusler completed most of the form recording the results of the scan.

   The record shows that Professor Rodeck measured the biparietal diameter, the head circumference, the distances of the lateral borders of the anterior and posterior horns of the lateral ventricles from the midline as against the distance from the midline to the skull at the same points (10/22 and 9/22 respectively), and the inter-orbital distances (30/14) and the nuchal fold. The record made by Dr Haeusler shows that Professor Rodeck revised his initial measurements for the biparietal diameter and the head circumference, which Professor Rodeck suggested showed the care he was taking. He clearly located what he thought were echoes representing the cavum septum and the anterior and posterior horns of the lateral ventricles. All the measurements were normal. Professor Rodeck said that he would have checked the cavum septum and the anterior horns in the axial and also the coronal planes. It was unusual to take the intra-orbital measurements. If the eyes had been closer than they should have been, which is named hypotelorism, that would have been an indication of semilobar holoprosencephaly. Professor Rodeck said that it occurred in between 2/3s and 3/4s of cases of semilobar holoprosencephaly. The distances were normal. The nuchal fold is increased in a number of abnormal conditions, which might include semilobar holoprosencephaly. In this case was not. Although there is no record, Professor Rodeck would also have found an echo which he took as representing the anterior falx. In addition to measuring the intra-orbital differences Professor Rodeck looked carefully at the echoes representing the face. Facial abnormalities are particularly associated with alobar holoprosencephaly, but may also be present with semilobar holoprosencephaly. Professor Rodeck found no facial abnormalities, and there were none for him to see.

   Professor Rodeck said that the image quality he obtained was good average: it was not poor or he would have made a written comment. The machine he was using was a Acuson XP128, which was as good as was then available and was better than the machine used by Mrs Wright. The machine would not produce images of the clarity produced today. In his evidence Professor Rodeck referred to the difficulty facing anyone conducting a fetal scan of this nature. If they conclude mistakenly that the brain is abnormal, it is likely that the pregnancy will be terminated with the loss of a healthy child. If they fail to diagnose an abnormality, a severely disabled child may be born to the mother. He said that here, after a long and detailed examination, he could not conclude that the fetus was other than normal. He had seen echoes consistent with the structures I have mentioned; he had found none of the abnormalities which are often associated with holoprosencephaly. He had felt confident in his conclusion. He described ultrasound as an elusive art and stated that there were few conditions that could be diagnosed with absolute certainty. He said that semilobar holoprosencephaly could sometimes be diagnosed. He said later that with ultra sound there was always a small question mark: it might be very small: he would not leave a patient with concerns if there was no real ground for them.

   Professor Rodeck did not record pictures during the scan. That accorded with the practice at that time at University College Hospital when the diagnosis had been one of normality. This practice was changed by Professor Rodeck following Alice's birth. There are thus no pictures recording any part of the images that were seen by Professor Rodeck and by Dr Meire during their scans.

   During the scan Professor Rodeck confirmed the presence of the choroid plexus cyst found by Dr Meire. He suggested a further scan in 5 weeks to check it. Because Mrs Lillywhite was visiting Australia it was arranged that the further scan should take place on 18 February 1992. It was carried out by Dr Haeusler. Dr Haeusler came from Austria to give evidence. He had no recollection of either scan with which he had been concerned. The record of his scan shows that he carried out a full scan involving the whole body of the fetus. He measured the biparietal diameter and the head circumference. He took measurements for the posterior horns of the lateral ventricles, but although he noted an 'A' for anterior he did not record a measurement for the anterior horns – he left a blank. He did not measure the intra-orbital distances. He did examine the facial profile. He found that the cyst was gone. His letter to Mr Marwood, Mrs Lillywhite's obstetrician at the West London Hospital, stated that 'Ultrasound examination revealed no obvious fetal malformation.' 'Obvious' must have been used in the sense of 'observable'.

   In a statement signed on 8 September 2003 Dr Haeusler said that he had scrutinised both the brain and fetal anatomy, that the measurements were reassuring, that he had specifically visualised the posterior horns of the lateral ventricles as this would have been where the cyst was if it still existed. He said it was not his practice to measure the anterior horns unless he suspected a cerebral abnormality, which he did not. On that basis I would not have expected him to have written an A with a space for the measurement. He suspected that he had not specifically sought the fetal facial profile but had obtained a good view and documented it for completeness. In a further statement signed on 28 September 2004 Dr Haeusler said that to measure the biparietal diameter and head circumference, he first identified the cavum septum to get the right level. He said that, if he had not identified it, he would have noted it and referred Mrs Lillywhite to Professor Rodeck. He also said that he was sure that he had seen the anterior horn of the lateral ventricle because, had he not done so, he would have noted it.

   Dr Haeusler said in his evidence that he had conducted a meticulous scan because he was aware of the referral. He would have been reminded of the referral by the record of Professor Rodeck's scan which was in the adjacent column on the sheet to his own and by the original referral sheet made out by Professor Lillywhite's personal assistant (which he had because he noted his letter to Dr Marwood at the bottom as having been sent with Mrs Lillywhite). Mrs Lillywhite herself recollected that Dr Haeusler took a long time with the scan. He struck me as a somewhat formal but very meticulous man. He said that at 29 weeks the picture would not be as clear as at 18 or 20 weeks – because of the increased skull thickness, but he obtained echoes that he was happy with. He accepted that his knowledge of the outcome of Professor Rodeck's scan could have had a reassuring effect.

   It was sought on behalf of the claimants to contrast Dr Haeusler's two statements suggesting that it was not credible that he had seen what he referred to in the second but omitted from the first. Given the medical technicalities which the case involves and the difficulties of dealing with a statement by telephone I do not think that the contents of the two statements are surprising. The second elaborates the first, explaining what was involved in the exercise which Dr Haeusler carried out. The only point on which I feel a degree of uncertainty is in relation to the anterior horn of the lateral ventricle.

   Alice was born on 26 April 1992. The birth itself was without difficulty. But it was later noticed that her head might be small and her forehead unusually sloping. Alice was X-rayed on 27 April. Because Mr and Mrs Lillywhite were unhappy at the treatment she was receiving, they arranged for Alice to be seen by Dr Meire at the Portland on 30 April. He examined her by ultrasound through the fontanelle. He reported as follows:

   The corpus callosum is absent and there is partial fusion of the bodies of the lateral ventricles. The anterior horns could not be confidently identified but I suspect are present but contain no csf [cerebrospinal fluid]. The posterior horns are present and normal and contain normal choroid plexus.

   The thalami and caudate nuclei are present and appear normal and the third ventricle appears filled with a large massa intermedia.

   The cerebellum and fourth ventricle are normal.

   There was no discussion at the trial of the ease or difficulty of making findings by this method at such a stage.

   It is plain that Mrs Lillywhite must have reminded Dr Meire that he had examined Alice as a fetus. He has, however, no recollection of this second examination. He said that he was amazed that he did not remember.

   It appears from a manuscript note made by Professor Rodeck's personal assistant and addressed to him that Dr Meire telephoned to report the result of the scan. I deduce that from the medical language used in the note and the statement 'Hilton Meire will be at home tomorrow Friday [telephone number] if you need to speak to him'. The note states that Dr Meire had confirmed agenesis of the corpus callosum. This must have come from Dr Meire himself because that diagnosis was not in his written note. The notes held by the University College Hospital were attached for Professor Rodeck. Professor Rodeck stated in his evidence that he did telephone Dr Meire at some point. He had no recollection of their conversation. He thought that they also had a meeting, but again he could not recollect it.

   Agenesis of the corpus callosum describes a different condition to holoprosencephaly. In this condition the division of the brain into two halves does occur. However the lateral ventricles are widely separated and are attenuated at the front giving them a drawn-out tear drop shape. There is no cavum septum or corpus callosum. The falx is present. The finding by Dr Meire that the lateral ventricles were partially fused appears inconsistent with this diagnosis, but the point was not raised during the trial.

   It was submitted that if the two men had thought that they had been misled by structures which had genuinely mimicked the echoes of the structures which Alice did not have they would have had much to discuss and that the case might have been written up. It was suggested that they would both have wanted to study Dr Meire's pictures. The point was made on the basis that the two men had each located a cavum septum that was absent, anterior horns that were absent from what should have been the lateral ventricles, and a falx that was absent. At this time, that is April 1992, the diagnosis was not of holoprosencephaly but of agenesis of the corpus callosum. So it was then thought that the anterior horns were present and that the falx was present. I refer to paragraph 3.2 of Dr Twining's report. On that basis the point may carry less weight. It is clear that both Professor Rodeck and Dr Meire were concerned that they had not diagnosed Alice's condition. Such a concern does not point to an appreciation that they had been negligent: it is equally consistent with a responsible approach to their work. Professor Rodeck later used Alice's case in his teaching as a warning of false positives appearing in ultrasound.

   When Professor Rodeck was informed of the problem with Alice he telephoned Mrs Lillywhite at her home. He made contact with her on 5 May and they had a conversation about which he remembers a little. He expressed his sympathy and said that Alice's condition could not always be detected in utero. Mrs Lillywhite was unhappy with the treatment that she had been receiving and Professor Rodeck wrote a letter of referral to Dr John Wilson at Great Ormond Street Hospital. In that letter he said 'I have spoken to her and had an initial discussion about the difficulty of antenatal diagnosis and also the variability in prognosis. I would be grateful if you would be able to fit her in as soon as possible and to have a word with you beforehand.' On 11 May 1992 Mr and Mrs Lillywhite went to see Professor Rodeck for a discussion of Alice's case. One of the questions which they had prepared to ask him was for his comments on the fact that Mrs Wright had raised the absence of a corpus callosum. In her statement she says that she asked if Alice's condition was detectable antenatally and that Professor Rodeck replied that radiology was an imprecise science: that he was disappointed at the failure to detect the abnormality. Professor Rodeck's concern for Mrs Lillywhite after Alice's birth caused her to write and thank him in December 1992 before she moved to Australia.

   On 7 August 1992 Alice was examined by means of an MR scan at Great Ormond Street and the diagnosis of semilobar holoprosencephaly was made. No additional abnormalities were found."

   The Trial

7. At the commencement of the trial, the claimant's case was, and always remained, based upon the simple proposition that Dr Rodeck could not have exercised the standard of care to be expected from a specialist consultant in obstetric ultrasound in concluding that the brain showed no signs of abnormality save for the cyst and reassuring Mrs Lillywhite accordingly. Mrs Wright had not been able to visualise the relevant structures because they were not there. In support of those contentions, they called three expert witnesses, Professor Griffiths, a Neuroradiologist who was the professor in the Academic Unit of Radiology at the University of Sheffield, Dr Twining to whom I have already referred and Mr Walkinshaw, a consultant in Maternal and Foetal Medicine at the Liverpool Women's Hospital. Professor Griffiths' particular expertise was the development of the foetal brain. One of the most important parts of his evidence, which was the bedrock of the appellants' case, was that by the time Professor Rodeck carried out his scan, the normal foetal brain would have divided and the three critical structures, that is the CSP, and the corpus callosum, the anterior horns of the ventricles, and the anterior falx would have developed. The evidence of Dr Twining and Mr Walkinshaw was to the effect that, at least on tertiary referral where an experienced radiologist had failed to detect the presence of these structures, a consultant in the position of Professor Rodeck must have failed to exercise the appropriate degree of care if he concluded from his examination that there was no evidence of abnormality.

8. The respondent's case was, in one sense equally simple. Two extremely distinguished radiological experts, Dr Meire and Professor Rodeck had within days of each other concluded that there were no relevant abnormalities. Dr Haeusler, who had been present at the examination by Professor Rodeck and carried out the subsequent examination, albeit to confirm that the cyst which had been found was of no significance, had also found no relevant abnormality. In particular, both Dr Meire and Professor Rodeck asserted that they had identified structures which were consistent with normality. Dr Meire specifically reported identifying the CSP and the falx; and Professor Rodeck made measurements taken from what he thought were echoes of the CSP and the anterior horns of the ventricles. These findings, it was submitted, showed firstly, that taken together with the other evidence, both Dr Meire and in particular Professor Rodeck had carried out the examinations with due care, but secondly that those examinations had identified on the scans echoes which could reasonably have been interpreted as showing the presence of the relevant structures.

9. The judge's task in evaluating this expert evidence was intrinsically difficult. The events were fifteen years in the past. There were some still photographs from the scans carried out by Mrs Wright. Otherwise there was no contemporaneous material other than the notes of Dr Meire and Professor Rodeck. It was a striking feature of the case that none of the photographs which had been taken of the scans by Dr Meire remained in existence, even though it was clear that Professor Rodeck, at least, had been seriously concerned about the fact that the diagnosis had been missed and had spoken to Dr Meire at the time . It was also unfortunate, to say the least, that even if none of those photographs had been kept, there was no evidence from Dr Meire or Professor Rodeck to the effect that those photographs had been examined by them after the event in order to determine why it was that the diagnosis had been missed. It should be stressed, however, that there has been no suggestion at any stage that this had been the result of any desire to brush what had happened under the carpet. Indeed it is clear that Professor Rodeck was frank in what he said to the appellants after Alice's birth, and sympathetic and helpful in seeking to ensure that they had appropriate support and treatment for Alice once he knew what the true diagnosis was.

10. The judge's task was made even more difficult by the way in which the case developed. The respondent's experts were Mr Howe a Consultant in Foeto-Maternal medicine at the Wessex Foetal Medicine Unit in Southampton and Dr Sarah Russell a consultant radiologist at St Mary's Hospital, Manchester, specialising in pre-natal diagnosistic services. Their reports were essentially to the effect that in the early 1990s, holoprosencephaly was undetected on scans in between 40% and 60% of cases, and accordingly the failure to diagnose could not of itself indicate any professional failure on the part of Professor Rodeck. In particular Mr Howe ascribed the failure to identify it in the present case as being due to "the subtlety of its appearance". Their original reports were dated May 2004. On the 26^th August 2004, an experts' meeting was held between all five experts. The respondent's experts maintained the view that both Dr Meire and Professor Rodeck must have had sufficient material from the scans which they performed to be able to give the opinions that they did that the three relevant structures were present; the fact that they were absent did not of itself justify the conclusion that they had failed to exercise reasonable care. They declined to speculate on what it was that either Dr Meire or Professor Rodeck had seen on their scans which could have misled them.

11. Approaching trial, in September 2004, Dr Russell submitted material, in particular in relation to a patient identified as "D", born in April 1995 where she had failed to identify the absence of a CSP, and submitted sheets of ultra-sound images which she asserted showed images of a structure compatible with a CSP which had misled her.

12. At the trial itself, the judge expressed the position in paragraph 34 of his judgment as follows:

    "At the trial it was soon apparent to me and, I think to all involved that on 3 December 1992 Professor Rodeck had conducted an apparently painstaking scan and must have seen echoes which he took to represent the cavum septum and the anterior horns of the lateral ventricles. That followed from the measurements that he took. It was not suggested that such echoes could have been "artefact" that is, a product of the ultrasound alone not representing any actual structure in the brain. In the absence of any pictures of what the Professor saw, it was a matter of considering what structures could have produced echoes mimicking the structures Professor Rodeck was seeking, and whether he should have realised the echoes were mimics. This question had not been considered by any of the witnesses in the written material served prior to the trial"

13. In his evidence Professor Rodeck himself at first only raised one suggestion as to what echoes he might have seen which could have mimicked those of the missing structures, and that was an echo suggesting the presence of the anterior horn of the lateral ventricle, similar to an echo which was illustrated in "The Unborn Patient" by Harrison, Golbus and Filly, published in 1991. This became known during the trial as the 13-90 echo, which was the number of the illustration in the book. He then suggested that there was at the forefront of the anterior ventricle a structure which could have mimicked the CSP. This was later identified by the respondent's experts by reference to the MR Scan done on the 16^th August 1992, as "a kind of "beak" " in that position. No other suggestions were made either by Dr Meire or Professor Rodeck which were pursued. In particular, there was, as the judge put it, no clear candidate to mimic the falx. Despite the fact that none of these suggestions, nor indeed others which were raised but not pursued, were particularised in any document presented by the respondent until formulated in a document served on the court and the appellants on the 6^th October 2004, the judge permitted the respondent to rely upon them. The debate between the experts thereafter concentrated on the question of whether or not in the light of these suggestions, there was material which could have justified Professor Rodeck's opinion, in the sense that it was consistent with the exercise of proper skill and care.

14. Professor Griffiths was concerned mainly to establish the probable anatomy of Alice's brain at the time of the relevant scans. His consistent view was that there were no candidates as structures which could have produced echoes mimicking those whose absence were questioned by Mrs Wright. In particular, what became known as the "beak" was too far back and too high to justify the conclusion that any echoes from it could have been taken to be those from the CSP. A particularly important point of reference, in his view, was the position of the thalamus. The CSP would be positioned in front of the thalamus, whereas there was a portion of the thalamus which was anterior to the "beak".

15. Dr Twining agreed with Professor Griffiths as to the position of the "beak", in particular its position in relation to the thalamus. There were no structures which in his view could have mimicked the falx. And if Professor Rodeck had been misled by a 13-90 echo, he should not have been, as this was, by 1991 a well recognised pitfall.

16. Mr Walkinshaw gave similar evidence. It is clear that he was greatly exercised by the failure of Professor Rodeck to have identified the abnormalities, as he had great respect for him, describing him as being "one of the most experienced obstetric sonologists in this country and would have been regarded as one of the most skilled in the international community". He concluded, however, that the Professor must have been misled by echoes produced by the "beak", by a 13-90 echo and an echo for the falx near the top of the head. But in doing so, he could not, in his view, have exercised the skill and care to be expected at tertiary level because the position of the "beak" was wrong, the 13-90 echo was, as Dr Twining said, a well known pitfall and proper analysis of the echoes should have made it clear that the line of the falx did not extend to the anterior part of the brain.

17. Mr Howe, on behalf of the respondent, considered that the fact that Professor Rodeck had measured the anterior horns of the lateral ventricles made it clear that he had not only carried out a careful examination, but must have seen an echo or echoes which justified his taking those measurements. He made the point that the distances involved in the structures within a foetal head of this age, could be measured in millimetres. He also demonstrated on a video the significant difference in quality between scans taken in 1991 and contemporary scans, showing how the definition achieved had substantially improved over the years. He accepted that the position of the "beak" was further back than would be expected for the CSP, and that it would be necessary to use the thalamus as a marker. He accepted that the 13-90 echo was one which was known to be a problem; and he was unable to explain the absence of the anterior portion of the falx. Nonetheless he considered that the evidence given by Dr Meire and Professor Rodeck and their notes of their examinations indicated that these had been carried out with proper skill and care and that the fact that the diagnosis was missed could not justify the conclusion that Professor Rodeck had failed to exercise due care and skill in his interpretation of what he found. He pointed in particular to the fact that there were no positive signs of holoprosencephaly, such as dilation of the posterior horns of the lateral ventricles nor was there any malformation of the skull which often accompanies the failure of the brain to develop properly.

18. Dr Russell gave evidence as to the fact that holoprosencephaly was not diagnosed in every case even on tertiary referral and using contemporary equipment. She pointed in particular to case D to which I have already referred where a "beak" appeared to her to have mimicked the CSP. In Alice's case she accordingly concluded that the "beak" could have produced the same result and that the 13-90 echoes could have misled Professor Rodeck into considering that he had identified the anterior horn. She accepted that it was difficult to explain the failure to identify the absence of the falx in the anterior part of the brain.

19. The judge also referred to a study, published in 2004 of which Professor Griffiths was one of the authors, involving 100 women referred to Professor Griffiths' unit amongst which were a number where ultrasound had failed to identify the absence of a CSP on tertiary referral. However Professor Griffiths made it clear that these were all cases where there had been diagnostic difficulties at ultrasound which was the reason for referral to his unit where MR scanning was available. Professor Griffiths made the point that, although Dr Meire had, for the reasons indicated by the judge, expressed some reservations about the quality of his scan, no such reservations had been expressed by Professor Rodeck.

    The Judge's Conclusions as to the Expert Evidence

20. Neither Professor Rodeck nor Dr Meire expressly recorded that they had located any anterior falx; and both Dr Howe and Mr Russell found this aspect of the case difficult. The judge noted that there was a short line on one of Mrs Wright's pictures which was central and could have suggested the falx. But neither of the respondent's experts was prepared to say that that justified the conclusion that the falx was normal. As far as the 13-90 echoes were concerned, he accepted the evidence of Dr Twining and Mr Walkinshaw that these were known, certainly to sonologists accepting tertiary referrals, to be echoes which needed to be distinguished from echoes from the lateral ventricles. He concluded that it was possible that those echoes had been produced in Alice's case; but rather inconsequentially concluded that "it would not be so easy to distinguish them from echoes of the horns".

21. He accepted that the "beak" produced echoes which could have mimicked the CSP. He expressed his conclusion as follows:

    "68. The evidence of Professor Griffiths, Dr Twining and Mr Walkinshaw was that any echoes the "beak" might have provided would have been in the wrong position and of the wrong shape to have been taken by a careful sonologist for the cavum septum. I was initially impressed by this: it was clear and forcefully put. It was said that it was over the thalami which was the wrong place, and the thalami were visible. Dr Meire recorded that he could identify the thalami but that does not mean that he could see the outline of the whole: the posterior parts could have been visible but not the anterior. The whole does not show up clearly on Mrs Wright's pictures. So I am unconvinced that I should take the thalami as a rock upon which Dr Rodeck should have built.

    69. Taking the evidence as a whole, I do not think that the cavum septum always appears as the clear landmark that one might expect from some of what I heard. One line may be missing. The lines may not be sharp and neatly aligned. I have also to bear in mind the size of the brain under examination and the very small distances involved between the position of the cavum septum and the supposed position of the beak at 18 weeks. I fully accept that the structure of Alice's monoventrical was well established by the time of the scans, but do think that there is at least room for doubt as to where precisely where it and its beak was proportionately positioned at the time. Several witness emphasised the lack of knowledge as to the processes of development of abnormal brains. I think that, if I limit that by saying "proportional development" there are good grounds for it. I have to bear in mind also Dr Russell's cases A and D, particularly D."

    The Judges Conclusions

22. As to the law the judge, having referred to Penny Palmer and Cannon –v- The East Kent Health Authority [2000] PNLR 323, considered that the general principles set out in Hunter –v- Hanley [1955] SLT 213, Maynard –v- West Midlands Regional Labour Party [1984] 1 WLR 634 and Bolitho –v- City and Hackney Health Authority [1998] AC 232 were sufficient guidance in the circumstances of the case. He recorded that there was no dispute as to the standard of care to be exercised by Professor Rodeck, namely the ordinary skill of his speciality, that is a consultant sonologist at a tertiary referral centre. That had to be judged in the context of a referral by Mrs Wright which identified the structures which she considered to be missing. He further recorded that if it was accepted that the notes that Dr Rodeck made of his observations and measurements indicated that he had apparently carried out his examination in the manner to be expected

    "It is the task of the claimants to satisfy me on the balance of probabilities that, if he had exercised the skill and care to be expected of him, Professor Rodeck would have recognised the echoes as mimics."

23. Applying that test to his findings, the judge's conclusion was that there was no fault in failing to observe that Alice had a monoventricle. The critical paragraph is paragraph 74, in which he said:

    "I have to put all these matters together in the whole context of the case, and decide whether on the balance of probabilities negligence is established. My answer is that it is not. I do not want to detract from the wider consideration of the issues which I have set out, but if I try to summarise my main reasons they are that the records suggest that Professor Rodeck approached his scan with particular care, the forceful circumstances that three experienced scanners reached a conclusion contrary to the correct conclusion of Mrs Wright, and that an investigation of what might have produced mimicking echoes has provided possible answers which do not point with clarity to negligence in failing to distinguishing them."

    The Appeal

24. Mr Badenoch, Q.C. on behalf of the appellants, submitted as I have already indicated in paragraph 2, that the judge was quite simply wrong on the evidence that he had heard to conclude that negligence had not been made out. Although he accepts that he cannot rely on the maxim res ipsa loquitur, he submits that the facts here are very close to establishing negligence in themselves. The structures which Professor Rodeck believed that he had identified and which formed the basis of his advice to the appellants were simply not there. Mrs Wright had correctly identified the anomalies in her scan; there was no reasonable explanation for Professor Rodeck failing to identify them himself. Where such an error has been made, there must be a heavy burden on the respondent to put forward evidence which would negate the otherwise inevitable conclusion that Professor Rodeck had not exercised due care and that burden had not been discharged. Mr Coghlan QC on behalf of the respondent, submits that the judge was entitled to conclude that the appellants had not discharged the burden of proof in the light of all the material before him, had made findings of fact with which this court should not interfere, and made no error of law.

    The Law

25. There is no significant dispute between the parties as to the law. Whilst Mr Badenoch, as I have said, referred to the maxim res ipsa loquitur, he did not suggest that, of itself, it provided a solution to this case. In medical negligence cases, where full evidence, including evidence from experts on both sides, has been heard, it is only in a rare case that it will assist the court. As Stuart Smith LJ said in Delaney –v- Southmead HA [1995] 6 Med LR 355 at 359:

    "For my part, I am doubtful whether it is of much assistance in medical negligence, at any rate when all the evidence in the case had been adduced. But even if Mr Stembridge is right in saying that at that stage the maxim applies, it is always open to a defendant to rebut a case of res ipsa loquitor either by giving an explanation of what happened which is inconsistent with negligence … or by showing that the defendant exercised all reasonable care."

26. Similar sentiments were expressed by Hobhouse LJ in Ratcliffe –v- Plymouth and Torbay Health Authority et al [1998] Lloyd's LR Med 162 at page 177, [1998] EWCA Civ 2000 :

    "Res ipsa loquitur is not a principle of law and it does not relate to or raise any presumption. It is merely a guide to help identify when a prima facie case is being made out. Where expert and factual evidence is being called on both sides at trial its usefulness will normally have been long since exhausted. "

27. In both of these cases plaintiffs had alleged that the injuries which they had sustained must have been the result of a negligently performed medical procedure. In Delaney the plaintiff had sustained a lesion of the brachial plexus after a successful operation to remove her gall bladder. She claimed that this must have been occasioned by hyper abduction of her left arm by the anaesthetist at some point during the administration, overseeing or conclusion of anaesthesia. The anaesthetist gave an explanation of his practice in carrying out anaesthesia which could not have involved hyper abduction of the plaintiff's left arm. The judge accepted his evidence. The injury was therefore one for which the defendant was unable to give an explanation. This court upheld the judge's finding that no negligence had been established on the basis that he had been entitled to find that the anaesthetist had exercised all due care in carrying out the procedure even though that left the injury unexplained. It was in that context that Stuart Smith LJ said what he did. In Ratcliffe, once again it was the procedure which was in issue. There the plaintiff was given a spinal anaesthetic, but subsequently suffered a serious neurological defect on the right side. The cause was a mystery. But an MRI Scan showed a lesion in the thoracic spine which the plaintiff claimed must have been the result of the spinal anaesthetic having been administered at the wrong level. The anaesthetist gave evidence that he had administered it at the lumbar spine. The judge accepted the anaesthetist's evidence and dismissed the plaintiff's claims.

28. Mr Coghlan, QC on behalf of the respondent submits that, although not referred to by the judge, the important passage in the judgments in this court in Ratcliffe, is at paragraph 49 where Brooke LJ dealing with the relevance of the maxim res ipsa loquitur said as follows:

    "(3) In practice, in contested medical negligence cases the evidence of the plaintiff, which establishes the res is likely to be buttressed by expert evidence to the effect that the matter complained of does not ordinarily occur in the absence of negligence.

    (4) The position may then be reached at the close of the plaintiff's case that the judge would be entitled to infer negligence on the defendant's part unless the defendant produces evidence which discharges this inference.

    (5) This evidence may be to the effect that there is a plausible explanation for what may have happened which does not connote any negligence on the defendant's part. This explanation must be a plausible one, and not a theoretically or remotely possible one, but the defendant certainly does not have to prove that his explanation is more likely to be correct than any other. If the plaintiff has no other evidence of negligence to rely upon, his claim will then fail.

    (6) Alternatively the defendant's evidence may satisfy the judge on the balance of probability that he did exercise proper care. If the untoward outcome is extremely rare, or was impossible to explain in the light of the current state of medical knowledge, the judge was bound to exercise great care in evaluating the evidence before making such a finding, but if he does so, the prima facie inference of negligence is rebutted and the plaintiff's claim will fail. The reason why the courts are willing to adopt this approach, particularly in very complex cases, is to be found in the judgments of Stuart-Smith and Dillon LJJ in Delaney…."

29. His submission accordingly is that this is a case in which the question which had to be asked by the judge was whether or not there was "a plausible explanation" for the outcome about which complaint is made. He submits that the judge, although not referred to Ratcliffe, had properly applied that principle. The dicta in both cases, have, however, to be read in the context of the factual situations confronting the courts in those cases. The important point is that in both, the judge had come to a clear conclusion on the facts as to the way in which the two procedures in question had been carried out which was, in itself, sufficient to absolve the anaesthetists in question of negligence.

30. But inherent in all these passages is, it seems to me, the fact that in some cases the evidence produced by a claimant may be such as to require the court to focus with some care on the explanation given by a defendant to displace that which would otherwise be the inevitable inference from the claimant's case that negligence has been established. A good example which has many echoes in the present case, is P–v- Leeds Teaching Hospitals NHS Trust [2004] EWHC 1392 (QB). In that case the claimant, then approximately eighteen to twenty weeks pregnant, underwent a routine abdominal ultra-sound scan at her local hospital. The sonographer was unable to visualise, in particular, the foetal bladder. As a result the claimant was referred to Leeds General Infirmary. The record of the scan taken there two days later shows that the bladder was noted under the heading "well visualised and appear normal". This scan had been taken by a specialist registrar. A second ultra-sound scan a month later conducted jointly by the same registrar and the consultant noted that "the bladder is seen within the pelvis". Two further ultra sound scans were taken in which no abnormality of the bladder was noted. In fact the child was born with a condition known as cloacal-exstrophy, a condition in which there is serious deformity resulting in the development of substantial parts of the abdominal content outside the abdominal wall. In that condition there was no bladder to be visualised. The judge, Holland J, was there confronted with a similar array of expert opinion as in the present case, including Mr Howe and Dr Meire on behalf of the defendant. He noted that the duty of care owed at the Leeds General Infirmary demanded a high standard of care and skill because the scan was, as he put it, "a scan with a focus". He concluded that "there is a heavy burden on Leeds when seeking to reconcile its incorrect visualisations with the exercise of all reasonable care and skill." He held that the claimant had established her case.

31. In my judgment, this case clearly falls into the same category. The duty of care owed by Professor Rodeck demanded a high standard of care and skill in the context of a focused referral based upon the concerns of Mrs Wright. I do not understand the respondents to deny that. The issue was of vital importance. If the structures were indeed absent, that would necessarily require consideration of a prompt intervention were Mrs Lillywhite to determine, as the judge found she would have done, to terminate the pregnancy. Equally, he would need to be confident of the absence of those structures in order to be able to advise Mrs Lillywhite as to termination, otherwise there would be the risk of aborting a healthy foetus.

    Conclusions

32. For the reasons that I have given above, this is not a case which could be solved by taking refuge in the maxim "res ipsa loquitur". However, when evaluating the way in which the judge approached the evidence, and the conclusions that he reached, I consider that we are entitled to look with care at the extent to which the judge properly reflected the strength of the claimant's argument based upon the simple proposition that Professor Rodeck purported to identify structures which were simply not there.

33. In my view, the judge correctly identified the problem that he had to resolve in paragraph 34 of his judgment, which is set out in paragraph 12 above. It seems to me that he was there acknowledging that up until such time as the respondent's witnesses grappled with the question of what structures could have been present to mimic those of the absent structures, there was really no defence to the claim. As the judge himself recognised, the mere fact that the examination had, in mechanical terms, been carried out with scrupulous care was only an answer to half the question unlike Delaney and Ratcliffe. As in the case of Penny (supra) which was a case involving the evaluation of smear tests, the real question was the extent to which a reasonable sonologist, given the information provided by the ultra-sound, could with reasonable care and skill have come to the conclusion that he did.

34. I see no reason for doubting the judge's finding that Professor Rodeck carried out the ultra sound procedure carefully. The question is whether or not he was entitled to decide that the claimants had failed to establish that Professor Rodeck cannot have exercised reasonable care and skill in concluding on the basis of that examination that the relevant structures were present. The basis of the judge's decision, expressed in paragraph 74 of the judgment, paragraph 23 above, was two fold. First, three experienced sonologists had reached effectively the same conclusion, and second that an investigation of what might have produced the mimicking echoes had "provided possible answers which do not point with clarity to negligence in failing to distinguish them." As to the first, that is undoubtedly a powerful consideration supporting the judge's decision. But, of itself, it could not be sufficient. And the judge did not suggest that it was. He had to answer the question that he had posed for himself in paragraph 34 of the judgment, namely what structures could have produced echoes mimicking the relevant structures and whether those structures could have produced echoes which Professor Rodeck could reasonably conclude were those of the relevant structures. In doing so, it seems to me that he had to approach the matter on the same basis that Holland J approach the problem in P (supra), namely bearing in mind that there was a heavy burden on Professor Rodeck to reconcile his incorrect conclusions with the exercise of all reasonable care and skill.

35. As to the falx, there was no structure which was capable of mimicking the anterior portion of a falx. The respondent's experts acknowledged that the conclusion that there was a falx was difficult to justify; and the judge, in my view correctly on the evidence, concluded in paragraph 67 of his judgment that "the idea that Professor Rodeck (or Dr Meire or Dr Haeusler) located the forward end of Alice's actual falx does not bear scrutiny." The only possible explanation which was suggested both at the trial and before us was that a line on the mid-line of one of the ultra-sound pictures produced by Mrs Wright could provide an explanation. But Mrs Wright did not suggest that the line had anything to do with the falx. Indeed she was asked no questions about it. The appellants' experts were of the view that it was in the wrong plane and Dr Howe clearly considered that of itself, it could not justify the conclusion that there was an anterior falx. Accordingly, the judge should have been driven to conclude that insofar as Professor Rodeck's case depended upon the assertion that he had indeed identified an anterior falx, there was no plausible explanation for how he could have done so in the exercise of reasonable care and skill.

36. As far as the lateral ventricles are concerned, the only echoes which were suggested as candidates were 13-90 echoes. It will be remembered that this was Professor Rodeck's own suggestion. As to those, the judge accepted the evidence from Dr Twining and Mr Walkinshaw that by 1991 it should have been known to tertiary sonologists practising in England that they were to be distinguished from the echoes from the lateral ventricle. The judge further concluded that on the evidence that he had heard, it was possible that such echoes had indeed been produced in Alice's case. In those circumstances, and in the absence of any further or other explanation, it is again difficult to see how the judge could have concluded other than that in purporting to identify the lateral ventricles, Professor Rodeck failed to exercise reasonable care.

37. I have set out the judge's conclusions in relation to the "beak" in paragraph 21 above. I have considerable difficulty in understanding the judge's reasoning. I can readily understand his concern that he had to evaluate that evidence in the light of the technology current at the time, which may not have produced such clearly defined images as might have been suggested. Equally, I can understand his concern to remind himself of the extremely small dimensions of the structures in question, and accordingly their relative positions. But the essence of his reasoning in rejecting the evidence of Professor Griffiths, Dr Twining and Mr Walkinshaw set out in paragraph 68 is in my judgment fundamentally flawed. The evidence of the appellants' experts was based upon the relative anatomical position of the CSP and the thalami which was not disputed by either of the respondent's experts. Neither Dr Meire nor Professor Rodeck suggested that they had only partially identified the thalami. This is of particular relevance in relation to the findings of Dr Meire, who expressly noted having visualised the thalami. And, in his evidence Professor Rodeck acknowledged having visualised them. There was no suggestion from either of those witnesses that they had had difficulty in so doing. There was, in my judgment, simply no evidence which could cast doubt on the conclusions of the appellants' experts.

38. The only suggestion that was made to the contrary was that the "beak", found on the MRI scan to be in a position in relation to the thalami which was not consistent with the position of a CSP, may not have been in that position at the time of Professor Rodeck's scan. That is wholly insufficient, in my judgment, to justify the conclusion expressed by the judge that this might explain the mistake. This explanation was not accepted by Professor Griffiths, who was the expert specifically called to deal with the anatomy of the foetal brain and whose expertise in this respect was accepted by both the respondent's experts. In the context of a trial where the suggestion in question arose out of the blue, in the sense that it formed no part of the case before trial, and was not an explanation put forward by Professor Rodeck, it could not, and should not, have cast doubt on the evidence of the appellants' experts. That is particularly so since Case D, used by Dr Russell in support of her evidence, as a case where she said she had been mislead by a "beak" at the relevant period of gestation was one in which she conceded that she had made "an error", although I accept that she was not by that agreeing that she had failed to exercise due care. The appellant's experts pointed out that it too was in an inappropriate position and should not have been identified as a CSP. Further, the "beak" in that case was substantially different in shape, position and size from that in the question in this case. Accordingly, in my view the judge could not have concluded that the "beak" exonerated Professor Rodeck.

39. The fundamental difficulty for the respondent was that in order to be able to establish, using the words of Brooke LJ in Ratcliffe, a plausible explanation, a plausible explanation had to be found for all three of the absent structures. It may be that a judge could have concluded that the mistake as to the extent of the falx, or the presence of a CSP, on the anterior ventricle was not, of itself, sufficient to establish negligence. In the light of the evidence to which I have referred, the respondent would however have faced considerable difficulty in relation to each. But what seems to me to be insurmountable is the hurdle of establishing that there were plausible explanations for all three.

40. In these circumstances, the only answer which the judge could properly have given to the question that he posed for himself in paragraph 34 of his judgment was that the only structures which had been identified which could have produced the echoes which Professor Rodeck interpreted as being the echoes of the relevant structures should not, had he exercised reasonable care and skill, have been so interpreted. I do so with some diffidence, for two reasons. First, we have not had the opportunity of hearing the evidence ourselves. On the other hand we have had the significant benefit of a full transcript of all the relevant evidence which has enabled us to consider with care the judge's reasoning. And secondly, the result is a conclusion that an extremely distinguished doctor has been found to have fallen below the appropriate standard of care and skill in what was undoubtedly a difficult case where he showed great compassion and gave considerable assistance and support to the appellants once he knew what the true position was. But those considerations do not deflect me from concluding that this appeal should be allowed.

    Lady Justice Arden:

41. A healthy, normal child was no doubt what Mrs Lillywhite hoped for at the end of her pregnancy. She went to her local maternity unit for an ultrasound scan when she was then some 18 weeks pregnant. She knew, at that stage, that there was a two per cent chance that the baby would have Down's syndrome. The ultrasound was performed by Mrs Wright. She could not find echoes of two of the three structures which she needed to find in order to be satisfied that the baby had normal brain formation. Those two structures were cavum septum pellicudum (which I will call the cavum septum) and the anterior horns of the lateral ventricles. She did not record the absence of the third structure, called the falx. Indeed the pictures she took contain an echo which is in the same position as an echo of part of the falx would be. So, because she could not make a diagnosis Mrs Wright caused Mrs Lillywhite to be referred Professor Rodeck, a leading consultant sonologist. His diagnosis was that he could not conclude that the baby was other than normal. As a result, Mrs Lillywhite did not proceed to a termination, as the judge accepted she would have done if she had been given the contrary advice. In due course, the baby was born and named Alice.

42. Contrary to Professor Rodeck's conclusions from his scan, Alice has a severe congenital malformation of the brain called semi-lobar holoprosencephaly, which is described in paragraphs 3, 4 and 5 of Latham LJ's judgment. In this condition, the cavum septum and the anterior horns could not have been present at any stage in Alice's pre-natal development. As to the falx, in Alice's case, this was severely reduced and, if visualised on the scan performed by Professor Rodeck, would have been seen to be abnormal. The anterior part of the falx was absent.

43. Needless to say, the arrival of Alice has been a life-changing event for Mr and Mrs Lillywhite. While, no doubt, they have courageously sought to come to terms with the disability that Alice has, I recognise the severe effect that it has had on them. However, this appeal is concerned only with the question whether this court should set aside the judge's conclusion that Professor Rodeck had not been negligent.

44. As to the background of the events leading to this claim, and the expert evidence at trial, I gratefully adopt what Latham LJ has already said.

45. The referral to Professor Rodeck was a tertiary referral. As a result, Professor Rodeck owed a heightened duty of care, i.e. a duty to take the reasonable care that would be expected of a consultant sonologist to whom a case had been referred because indications of abnormality had already been found. As Holland J put it in the P case in the passage cited below, it is a scan with a focus, although it is not only a scan with a focus. A consultant on a tertiary referral will also carry out a full scan and reach an independent opinion of his or her own. In fact, as I have said, Professor Rodeck found no abnormality.

46. As Latham LJ has explained, before Mrs Lillywhite saw Professor Rodeck, she had another scan with another expert, Dr Meire. While he was unable to obtain a perfect image of Alice's brain, he too did not diagnose the abnormality. He saw echoes of all three of the structures in question. Moreover, some weeks after the scan by Professor Rodeck, a third expert, Dr Haeusler, performed a scan. He likewise found that there was no obvious malformation of Alice's brain. In particular, he, like Professor Rodeck and Dr Meire, but unlike Mrs Wright, found an echo of what he took to be the cavum septum. The judge described the fact that none of these three experts made the correct diagnosis as a result of the scan as a "forceful circumstance" (judgment, paragraph 74). This was so even though Mrs Wright had not been able to satisfy herself that Alice's brain formation was normal.

47. There are two crucial findings by the judge about Professor Rodeck's ultrasound examination:-

    i) He performed the ultrasound with great care. The records of his examination were made by Dr Haeusler, who was also in attendance, and they show that at one point Professor Rodeck revised his initial measurements for the biparietal diameter and head circumference, which in the judge's judgment showed the care he was taking (judgment, paragraph 19).

    ii) The judge found that Professor Rodeck located what he thought were echoes representing the cavum septum and the anterior horns, and indeed taken normal measurements of the distance between them. The judge also accepted that, although there was no record of his doing so, Professor Rodeck would also have found an echo which represented the anterior falx. This was Professor Rodeck's evidence, which the judge accepted (judgment, paragraphs 19, 30 and 63). On the appellant's case, there were no structures that he could have measured.

48. So the issue, therefore, was not whether Professor Rodeck had seen any echoes representing the three vital structures (because he could not have done so) but whether he should have realised that the echoes which he had seen were not ones on which he could rely. The echoes he saw were obviously only mimics, that is, true and constant echoes mimicking some other structure, and not echoes of the actual structures he needed to visualise. As the judge held:-

    "As I have described, the disagreement between the expert witnesses was whether Professor Rodeck should have discovered that the echoes he must have taken as representing the cavum septum, the anterior horns of the lateral ventricles and the anterior falx were in fact mimics of those structures and not echoes of those structures. I saw "must have taken" because there is no record of what Professor Rodeck saw, and it is a matter of deduction as to what occurred. It is the task of the claimants to satisfy me on the balance of probability that, if he had exercised the skill and care to be expected of him, Professor Rodeck would have recognised the echoes as mimics." (Judgment, paragraph 62)

49. Anatomically it was not impossible, but it was extremely unlikely, that there were structures in Alice's brain which would satisfactorily mimic absent structures (see, for example, the evidence of Professor Griffiths, Day 5, pages 75-76, and Day 6, page 8).

50. Buxton and Latham LJJ have stressed a number of important points. For example, there are no photographic images of what Professor Rodeck saw. Professor Rodeck did not keep pictures of the scans he took of Alice. Dr Meire's pictures were destroyed in the course of a normal record destruction policy. It is unfortunate that nothing was done to stop this. However, the appellant did not suggest that these images were improperly destroyed. There is no evidence suggesting anyone is to blame for that loss in this case. Buxton and Latham LJJ also lay considerable emphasis on the fact that in the course of the trial the respondent changed its explanation of what Professor Rodeck saw and in the end had no explanation. However, the force of that point depends on who had the onus of proof, whether on the evidence any clear medical explanation was possible and whether the judge was entitled to find that the respondent's evidence in effect provided any necessary explanation. Buxton LJ has also emphasised the lateness of the evidence as to the possible reasons for the mimicking echoes. But the fact is that the expert evidence was lacking on both sides (see the passage from paragraphs 34 of the judgment cited by Latham LJ at paragraph 12 above). If there is no established explanation for mimicking echoes, it is less surprising than at first sight it may seem that explanations were not attempted prior to the trial. I agree with Buxton LJ that one would have expected Dr Meire and Professor Rodeck to have investigated why they had not found holoprosencephaly but it is right to recall that they were both of the view that 100% accuracy of diagnosis from ultrasound examination was simply not possible.

51. Was the judge correct to say in the passage which I have already quoted from paragraph 62 that the onus of proof lay on the claimant? The appellant's case is that the circumstances called for an explanation by the respondent. Latham and Buxton LJJ agree that in the circumstances it fell to Professor Rodeck to provide an explanation. They illustrate this proposition by reference to the decision of Holland J in P v Leeds Hospital NHS Trust [2004] EWHC 1392 (QB). In this case, a pregnant woman had been referred for an ultrasound to Leeds Hospital, following the discovery by her local maternity unit that the child she was carrying appeared to have no bladder. The expert at Leeds Hospital performed an ultrasound and concluded that the bladder was visible. Later ultrasounds were taken at Leeds Hospital which did not mention the bladder. In fact the child had a serious abnormality known as cloacal exstrophy, involving the extrusion of the bladder and bowel from the body wall. The question was whether the hospital was liable in negligence.

52. In my judgment, the analogy between this case and the P case should not be taken too far. The crucial part of the judgment of Holland J in the P case is as follows:-

    "Turning to execution to that high standard, I am satisfied that in the prevailing circumstances it was not enough to visualise a bladder simply by reference to shape and position. I accept that no more could necessarily be expected in the course of a standard anomaly scan. I write 'necessarily': the evidence showed that some practitioners would routinely go further, but I cannot find that that which was seemingly done was outwith the range of conduct appropriate for the standard anomaly scan. This , however, was not such a scan, it was a scan with a focus and I am satisfied that with care and skill commensurate to that focus and its importance that which was mistakenly identified as a bladder could have been exposed as something other than that organ. I accept Dr. Loughna's contention (as to which there was really no dispute) that with persistence that which was visualised as a bladder could have been checked for filling and emptying a test seemingly definitive. Again, with like persistence as demanded by the importance of the issue, the shape could and should have been checked by scanning in different planes and the Doppler could have been activated so as to obtain the guidance that can be furnished by visualising the relationship between the organ in question and the umbilical arteries. Granted that no one involved in the Leeds scans has now any real recollection of such, and granted that all now say that they must have been alive to the potential for cloacal exstrophy, it is worrying (whether or not significantly) that there is no reference to this condition in the contemporaneous documentation that none of the images depicts that which was visualised as a bladder; and that the terms of the letter of the 6^th May are as cited. Finally, the Claimant is entitled to point to the fact of cloacal exstrophy and to visualisations at Whitby correctly consistent with such: there is a heavy burden on Leeds when seeking to reconcile its incorrect visualisations with the exercise of all reasonable care and skill."

53. In my judgment, that passage has to be read as a whole. In the P case, it is clear that there were procedures which could have been used to test whether or not the echoes which were visualised in fact were echoes of the bladder. In addition, the results of the scan taken at Whitby were known. When the judge held that there was a heavy burden on Leeds to reconcile its incorrect visualisations with the exercise of all reasonable care and skill, in my judgment he was holding that on those facts Leeds would be liable unless there was some good reason why it did not perform these procedures. Although there is a parallel in that the results of Mrs Wright's scan were known, the situation in the P case was a different situation from the present case. Here Mrs Lillywhite has to show that there was some step that Professor Rodeck failed to take or that he ought to have realised that the echoes he saw were only mimics. In addition, this is not a case where there was 100% success, even at the tertiary level, in diagnosing holoprosencephaly at the relevant time. Nor is it the case that Professor Rodeck cannot recall the ultrasound examination of Mrs Lillywhite. On the contrary, the judge found that he had carried out the ultrasound with great care. His evidence was that he had not ignored Mrs Wright's findings. The P case is accordingly not on all fours with the present case.

54. In my judgment, the judge was correct to direct himself, as he did, as follows:-

    "60. I do not think that the law provides any difficulty. It was considered by the Court of Appeal in a case which has some analogies with the present in that it was concerned with examination and interpretation of cervical smears, though there the smears were still available for examination. The case is Penney, Palmer and Cannon v East Kent Health Authority [2000] PNLR 323. In the course of his judgment Lord Woolf MR quoted the following passages from earlier authorities as setting out the relevant law:

    (a) Hunter v Hanley [1955] SLT 213 per Lord President Clyde:

    In the realm of diagnosis and treatment there is ample scope for genuine difference of opinion and one man clearly is not negligent merely because his conclusion differs from that of other professional men … The true test for establishing negligence in diagnosis or treatment on the part of the doctor is whether he has been proved to be guilty of such failure as no doctor of ordinary skill would be guilty of, if acting with ordinary care.

    (b) Maynard v West Midlands Regional Health Authority [1984] 1 WLR 634 at 648 per Lord Scarman:

    … that a doctor who professes to exercise a special skill must exercise the ordinary skill must exercise the ordinary skill of his specialty. Differences of opinion and practice exist, and will always exist, in the medical as in other professions. There is seldom any one answer exclusive of all others to problems of professional judgment. A court may prefer one body of opinion to another: but that is no basis for conclusion of negligence"

    (c) Bolitho v City and Hackney Health Authority [1998] AC 232 at 11588H per Lord Browne-Wilkinson:

    "In my view the court is not bound to hold that a defendant doctor escapes liability for negligent treatment or diagnosis just because he leads evidence from a number of medical experts who are genuinely of the opinion that the defendants' treatment or diagnosis accorded with sound medical practice … the court has to be satisfied that the exponents of the body of opinion relied on can demonstrate that such opinion has a logical basis. In particular in cases involving, as they 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.""

55. These quotations from well-known authorities show, among other matters, that it is not enough that other experts would have reached a different diagnosis from Professor Rodeck. It has to be shown on a balance of probabilities that no reasonably competent tertiary sonologist would have made the same error, i.e. treated what he saw as echoes of the three brain structures that he was seeking to find, namely the cavum septum, the anterior horns and the falx. For this purpose, however, it would be enough to show that no reasonably competent sonologist would have misinterpreted any one of the three relevant echoes since, any one of them would have put him on notice that the other two were potentially flawed. The Bolam principle mattered because the appellant's case was that Professor Rodeck, Dr Meire and Dr Haeusler all fell below the standard of care to be expect of a reasonably competent tertiary sonologist, and had all made elementary mistakes. They visualised and measured echoes which they ought to have realised were not echoes of the actual structures they needed to find.

56. The judge made no reference to the doctrine of res ipsa loquitur. For that doctrine to apply, the mere occurrence of an accident must of itself be sufficient to establish a prima facie case of negligence. To avoid liability, the defendant must produce satisfactory evidence in answer to rebut the claimant's case. Nor does the appellant rely on this doctrine. In my judgment, it does not usually assist the process of fact-finding to speak in terms of this presumption. Nor, in general, is it necessary to do so. The standard of proof required in civil cases is generally the balance of probabilities and inherent in that standard is flexibility. The stronger the claimant's case is, the greater the need for the defendant to lead evidence or arguments to show that negligence did not occur.

57. I recognise that in some circumstances a prima facie case of negligence will be rebutted by proof that the defendant acted carefully: see in particular the Ratcliffe v Plymouth and Torbay Health Authority [1998] Lloyd's LR Med 162 at page 173 per Brooke LJ and page 175 per Hobhouse LJ. However, there are special factors in this case which take it out of that category of case. The fact that Mrs Wright was not able to detect all the requisite brain structures some five days before Professor Rodeck's scan and that her conclusions were accurate indicates that as a matter of common sense some explanation is required. (That said, the fact that Mrs Wright had not been able to identify all the necessary structures should be seen in context. Professor Rodeck's evidence was that he was able to visualise structures in over half the referrals made to him on the basis that vital structures were absent. In addition, Dr Meire had by then carried out his ultrasound examination.) The judge approached the case on the basis that some explanation for the misdiagnosis was needed. But the requirement for an explanation does not mean that the finding that Professor Rodeck performed the scan carefully has no role to play. On the contrary, it adds weight to the possibility of the occurrence of any explanation the respondent puts forward.

58. In this case, once it was accepted that Professor Rodeck performed a careful scan and that he must have seen mimicking echoes, the scan which he performed, even with its incorrect diagnosis, could not without more establish a prima facie case of negligence. It all depended on what those mimicking echoes were, and how they should have been interpreted by a reasonably competent tertiary sonologist. The appellant had to adduce expert evidence about this and the respondent had to adduce expert evidence in response. The judge then had to decide, on a balance of probabilities, whether Professor Rodeck was negligent. In that assessment, the respondent did not in my judgment have to show that, on a balance of probabilities, Professor Rodeck was not negligent. The onus of proof remains on the claimant.

59. This court considered the doctrine of res ipsa loquitur Ratcliffe v Plymouth & Torbay Health Authority, which also illustrates the point which I have just made about the onus of proof. In the course of his judgment, Brooke LJ, with whom Sir John Vinelott agreed, held:

    "(4) The position may then be reached at the close of the plaintiff's case that the judge would be entitled to infer negligence on the defendant's part unless the defendant adduces evidence which discharges this inference.

    (5) This evidence may be to the effect that there is a plausible explanation of what may have happened which does not connote any negligence on the defendant's part. The explanation must be a plausible one and not a theoretically or remotely possible one, but the defendant certainly does not have to prove that his explanation is more likely to be correct than any other. If the plaintiff has no other evidence of negligence to rely on, his claim will then fail."

60. Importantly, Brooke LJ contemplated that the explanation need not be a probable one, and that it was enough if it was plausible. In my judgment, this follows from the burden and standard of proof. Where, as here, the claimant cannot show what the doctor did, and the court is satisfied that the doctor acted carefully, and there is evidence that one possible explanation for the misdiagnosis involves no negligence on his part, the court may, depending on the circumstances, properly conclude that it is not more likely than not that the misdiagnosis was caused by the doctor's negligence.

61. In the context of clinical negligence, care needs in my judgment to be taken about what constitutes a possible explanation for this purpose. As is well known, medicine is not an exact science. Events can occur for which doctors have no scientific explanation. So, in the context of clinical negligence, a defendant may be able to do no more than point to the possible occurrence of a medical phenomenon, the aetiology for which is not known, as an explanation for what happened. That may be sufficient to lead the trial judge to the conclusion that, on a balance of probabilities, the claimant's case has not been established. If, in an area of medicine where there are no clear answers, the court were to require the doctor, who is said to have been negligent, to provide a full clinical explanation of what might have happened in order to rebut a prima facie case of negligence, the court would be requiring the doctor to go beyond the bounds of known medical science. In my judgment, that would be wrong in principle since it would throw on to medical practitioners and those who are vicariously liable in law for them the risk of not being able to provide explanations for phenomena which in the present state of medical knowledge cannot be satisfactorily explained. A distinction has to be drawn in such a case between a finding that a phenomenon occurs, which can be misinterpreted by a reasonably competent practitioner, and a finding as to what causes such a phenomenon to occur. Clearly, in any such case, the judge must be satisfied that the state of medical knowledge is as it has been represented to him. But there was evidence of that in this case. There was, for example, the lack of any medical explanation for the mimicking mid-line echo of the anterior falx on Mrs Wright's picture and the lack of any medical explanation for 13-90 echoes (which were also mimics). In addition, the judge in any such case must be satisfied that the doctor's error was due to a phenomenon that he could not reasonably have appreciated was a false signal, and that he was not in fact through negligence missing some condition which he ought to have recognised. However, in this case, other eminent sonologists had come to the same conclusion as Professor Rodeck, and Professor Rodeck had made measurements in the course of his scan consistent with mimicking echoes of the relevant structures in their normal position. So there was here an evidential basis for the respondent's case that Professor Rodeck had seen appropriate mimics which he could not reasonably have realised were such.

62. The principal reasons given by the judge for his assessment of the balance of probabilities are to be found in paragraph 74 of his judgment, which Latham LJ has set out in paragraph 23 of his judgment. In this paragraph, the judge held that the claimant failed to establish her case on a balance of probabilities because (1) there were possible ways in which the mimicking echoes could have occurred and could reasonably have been interpreted as true echoes of the relevant structures, (2) Professor Rodeck had performed the scan with care and (3) not only Professor Rodeck but two other sonologists had reached the same conclusion. Accordingly, the judge made his assessment as to where the balance of probabilities lay by reference to a combination of factors. As I see it, these factors must be seen together: the judge's evaluation of the evidence about possible explanations cannot be viewed in isolation from the judge's findings as to the care with which Professor Rodeck performed his scan or the view which the judge on the evidence formed of the care taken by Dr Meire and Dr Haeusler.

63. The appellant, however, concentrates her attack on the judge's findings about possible explanations for Professor Rodeck's misdiagnosis. Accordingly, I turn next to examine the judge's detailed findings about the mimicking echoes for each of the three structures in turn.

64. As to the cavum septum, the respondent called an expert, Dr Russell, who gave evidence that in two cases, case A and case D, she thought she had seen echoes representing the cavum septum although it was absent. Of the two cases produced by Dr Russell, case D was closer than case A because of the particular abnormality present. (By way of explanation, according to Professor Griffiths, Alice's condition is a very extreme version of semi-lobar holoprosencephaly while case D was a case of alobar holoprosencephaly at the subtle end of the spectrum.) The appellant's experts sought to undermine Dr Russell's evidence by suggesting that in her original scan of case D Dr Russell could not have exercised appropriate care. However, Dr Russell maintained her position in cross examination (Day 9 transcript page 104) and the judge described the appellant's case on this point merely a suggestion. The judge accepted the evidence of all the experts save where he expressly stated to the contrary (see his judgment, paragraphs 39 and 70). He clearly accepted, through Dr Russell's evidence, that there was a reputable body of opinion among tertiary sonologists that there could be echoes which were mimics of the cavum septum and not echoes of an actual structure, and that such mimics could without want of care on the sonologist's part lead to a wrong diagnosis.

65. The experts were in general agreement that any mimic of the cavum septum would have to have come from the monoventrical beak. The appellant's experts suggested that an echo from the beak protruding from Alice's monoventricle would have signalled to Professor Rodeck that he could not be visualising echoes of the true structures because of the beak's relationship to the thalami. The judge dealt with this in paragraphs 68 and 69 of his judgment, which are set out in paragraph 21 of the judgment of Latham LJ. The judge's conclusion was that for the reasons there given it would not necessarily have been possible to test the echoes by reference to the thalami, and accordingly that this could not be grounds for finding that Professor Rodeck was necessarily negligent. As to paragraph 68 of the judgment, while Dr Meire and Professor Rodeck did not refer to partial visualisation of the thalami, the evidence of Dr Russell was that the images taken by Mrs Wright did not enable the viewer clearly to define part of the anterior margin of the thalami. Moreover, it was not only Professor Rodeck but also Dr Meire who saw the thalamus in the course of the scan and yet they did not appreciate that the mimicking echo, which they took to be of the cavum septum, was in the wrong location. Professor Rodeck's measurement of the biparietal diameter depended on the location of the cavum septum and it was normal. In paragraph 69, the judge made a number of points. Several witnesses, including Dr Russell, Dr Meire and Professor Rodeck, expressed the view that one could not assume linear growth of the relevant brain structures. The judge was entitled to accept this evidence as sonologists must have expert knowledge of anatomy, and the judge's point about location of structures relates only to the monoventricle and its beak. In addition, Dr Russell had herself been misled into thinking that scans she had performed of Cases A and D were normal and that she had seen the cavum septum. The judge clearly did not accept that her error was the result of lack of reasonable care as a tertiary sonologist.

66. Latham LJ has made the point that paragraphs 68 and 69 of the judge's judgment are very difficult to follow. I do not find them easy either, but I bear in mind that the judge had the benefit of evidence from leading experts in this field who gave a great deal of evidence by reference to pictures taken in the court of prenatal and postnatal ultrasound examinations of the brains of babies. We have not had that benefit. But I also bear in mind that the last point in paragraph 69, at least, is clear, namely that Dr Russell had herself been led to believe that she had seen echoes consistent with the cavum septum in the brains of babies with holoprosencephaly who could not possibly have had actual structures to produce those echoes.

67. As to the falx, the judge also accepted that, in one of Mrs Wright's pictures, there was an additional mid-line echo which could have mimicked part of the anterior falx. Furthermore, her picture shows an echo continuing in the direction of the front of the brain, beyond the point at which Alice's falx stopped. Dr Russell gave evidence that there was a mid-line echo mimicking the falx in the scan of case A, which was also a case of holoprosencephaly. In the course of his scan, Professor Rodeck took a measurement which was consistent with the presence of a normal falx (see the evidence of Mr David Howe, Day 9, page 16), even though in Alice's case the falx must have stopped somewhat short of that point. Mrs Wright may also have accepted that the mid-line echo in her picture represented the anterior falx (judgment, paragraph 72, Day 2, page 43, and see also Dr Meire's evidence at Day 3, page 30).

68. It is said that the judge made a finding that Professor Rodeck could not have visualised a mimic of Alice's anterior falx in paragraph 67 of his judgment, where the judge said:

    "Next, what does the evidence that is available suggest that Professor Rodeck may have seen? At the end there was largely agreement between Mr Walkinshaw, Mr Howe and Dr Russell. The only candidates that survived were the echo of the monoventrical beak, to mimic the cavum septum, the 13-90 echoes to which mimic the anterior horns of the lateral ventricles, and the small piece of midline on Mrs Wright's picture (vi) for the anterior falx. I have said largely agreed because there was not agreement as to the last. But I do not think that the idea that Professor Rodeck (or Dr Meire or Dr Haeusler) located the forward end of Alice's actual falx bears scrutiny. It ran out at the highest point of the cranium and has disappeared in the last coronal view on the MR scan (9 of 16) where the beak of the monoventrical appears, or as Professor Griffiths said, at the line of the ears."

69. In my judgment, it is clear from the context that in the last two sentences of this paragraph the judge was referring here to an echo of the falx as a real structure, not to a mimicking echo. He dealt with the evidence specifically directed to the possibility of a mimicking echo for the falx later in his judgment, at paragraphs 71 and 72. But, whichever it was, it is relevant to note that Dr Russell's evidence had been that it was not necessary to find an echo of the whole of the anterior falx:

    "Yes, my reason for defining that is there is this perception that we should see the falx and know it is the falx because it hits the skull and actually it is a midline structure, and it does not, as we have seen on the images, always – although it clearly does anatomically extend to the skull, on the ultrasound image it is not a requirement to see it extending anteriorly to the skull to know it is the falx." (Day 7, page 88)

70. Dr Russell's evidence on this point was not challenged, and the judge was entitled to accept it. Mr Walkinshaw appears to have accepted that the front of the anterior falx could be obliterated in an ultrasound picture (Day 7, page 43). Likewise, Mr Howe accepted it (Day 9, page 34).

71. That leaves the anterior horns. There was no explanation for the echo which Professor Rodeck saw of the anterior horns. There was evidence that there was a phenomenon, called 13-90 echoes, which could be mistaken for the lateral ventricles and that the 13-90 echoes could be the explanation for an echo mimicking the anterior horns in Alice's case. However, the judge held that 13-90 echoes were a known phenomenon by the time of the ultrasound in question and therefore should not have misled Professor Rodeck if lateral ventricles had been present. However, in Alice's case, there were no lateral ventricles. The judge held as follows:

    "On the basis of the evidence which I heard it is possible that these echoes [the 13-90 echoes] were produced in Alice's case and no other candidate to mimic the anterior lateral ventricular horns was established. It is not known how they would appear in a case where there were no anterior horns. Further, in the absence of horns it would not be so easy to distinguish them from echoes of the horns: for the contrasting structure is not there." (Judgment, paragraph 70)

72. The reason put forward in the last sentence was one which the judge was entitled to give on the basis of Dr Russell's evidence:

    "The linear echoes that it is suggested could mimic the anterior horns we know do exist and therefore I think it is reasonable to put forward the proposition that they would be included in a view that also contained a pseudo cavum septum pellucidum, the thalami and the third ventricle.

    Jack J: I am sorry, I need to ask you to repeat that. The linear echoes could suggest mimicking the anterior horns - ?

    A. These linear echoes that we know are – the anterior margin of them in the normal stance from the roof of the lateral ventricle, including the anterior horn – as we know they exist and that they are there. I do think that in Alice's case it is reasonable to propose that they would be in a position that could mislead in the absence of the anterior horns. " (Day 9, page 62)

73. This evidence was not challenged. The judge had established with Mr Walkinshaw that the images of the 13-90 mimics in evidence were of a brain with anterior horns (Day 7, page 94). Mr Howe gave evidence similar to Dr Russell's (Day 9, page 41). in addition, it has to be borne in mind that Dr Meire, who is also a leading sonologist, had also found echoes mimicking the anterior horns. That factor, plus the factor that the judge was satisfied that Professor Rodeck, whom he had seen in the witness box for about two days, had conducted his scan with all requisite care and the factor that mimicking echoes may have been present for the falx and cavum septum, must be thrown in the balance which evaluating this possibility. In my judgment, the judge was entitled to come to the conclusion which he did on this point. He did not reach his conclusion as to a possible explanation for mimicking echoes of anterior horns in Alice's case as the basis of the 13-90 echoes, but on the basis that the evidence did not establish how those echoes would appear in Alice's particular case.

74. The judge regarded the bases, on which he found that mimicking echoes could have occurred which could reasonably have deceived Professor Rodeck, as no more than possible explanations for what happened (judgment, paragraph 74). That was an overall assessment of the evidence. It depended in part on the judge's assessment of the expert witnesses. However, in my judgment, the judge must also inevitably have been influenced in this assessment by the fact that he had found that Professor Rodeck had performed his scan with care and that Dr Meire and Dr Haeusler had made the same diagnosis. For the reasons given above, I do not consider that, once the judge was satisfied that a reasonably competent tertiary sonologist could have interpreted the mimics as Professor Rodeck did, the respondent had to go further and produce a sound medical explanation for the mimicking echoes.

75. The judge considered whether Professor Rodeck ought to have observed that Alice had a monoventrical brain. The judge's conclusion, however, on the evidence was that this was something which on occasion might be observed but equally might not (judgment, paragraph 73). There is no appeal on this point.

76. The judge also found that ultrasound does not produce as clear an image as magnetic resonance imagining scan and that the clarity of images has substantially improved since 1991. "Abnormalities which might be regarded as routinely detectable in 2004 would have been rarely diagnosed even 10 years ago." (judgment, paragraph 59). As Dr Russell said:

    "We know that holoprosencephaly is a very difficult diagnosis to achieve, because we have evidence from the literature which we have reviewed in this court which illustrates that in secondary and tertiary centres that this is not a diagnosis that, despite the skill and experience of the operator, that is always achieved. So whilst the abnormality may well be barn door [viz.obvious] once it has been diagnosed and we have got a magnetic resonance image, that is not the same as saying that it is barn door in ultrasound diagnostic terms at the time at which the attempted diagnosis is made." (Day 9, page 101)

77. The position on the judge's findings was this. The appellants could not say what Professor Rodeck had seen. They had to say that it was more likely than not that he negligently misinterpreted the echoes that he saw. But two other experts, Dr Meire and Dr Haeusler, had reached the same conclusion as he had. The judge was satisfied that Professor Rodeck had taken all appropriate care in performing the ultrasound. In addition the evidence provided possible answers to the question of what might have produced mimicking echoes. Diagnosis of holoprosencephaly by ultrasound was not 100% successful, even at the tertiary level. Another eminent sonologist, Dr Russell, had missed a case of holoprosencephaly by ultrasound. On his assessment of all the relevant evidence in the case, the judge concluded that it was not more probable than not that Professor Rodeck had been negligent (judgment, paragraph 74).

78. The judgment of the judge is concise and condensed. He heard some nine days of evidence. The questions he asked show that he took a close interest in the detailed medical evidence. His judgment is highly compressed but closely reasoned. The fact that he gave greater weight to some of the evidence than this court might have done is not a ground for setting his order aside. It is necessary for any appellate court to bear in mind that the evaluation of the balance of probabilities at the end of a trial represents an overall judgment on all the arguments or evidence heard in the course of the trial even though not all those arguments or all that evidence is articulated in the judgment. As Lord Hoffmann said in Biogen Inc v Medeva plc (1996) 38 BMLR 149 at 165:

    "The need for appellate caution in reversing the judge's evaluation of the facts is based upon much more solid grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance … of which time and language do not permit exact expression, but which may play an important part in the judge's overall evaluation."

79. It follows that this court should in my judgment substitute its evaluation of the balance of probabilities for that of the judge only if it is satisfied that, notwithstanding the judge's superior knowledge of the case, his evaluation was against the weight of all the evidence in this case. That is a very strong conclusion to draw in a case and involves holding that there was no evidential basis on which the judge's conclusions can be justified. For the reasons given above, I am not so satisfied in this case, and accordingly I would dismiss this appeal.

    Lord Justice Buxton :

80. I agree with the judgment of Latham LJ. Since the court is divided, and since the majority would reverse the judge, I add some observations of my own.

    The duty of care

81. Latham LJ has set out the law relevant to this case, but it is worth returning to the great weight placed by the Trust on the decision of this court in Ratcliffe v Plymouth, etc HA [1998] Lloyd's LR Med 162, , [1998] EWCA Civ 2000 and in particular on the observations of Brooke LJ at §§ 22 and 52 that:

    "The Greek tragedian Aeschylus addressed the unforeseen predicaments of human frailty in terms of the sport of the gods. In a modern scientific age, the wisest of experts will sometimes have to say: 'I simply do not know what happened'. The courts would be doing the practice of medicine a considerable disservice if in such a case, because the patient has suffered a grievous and unexpected outturn from a visit to a hospital, a careful doctor is ordered to pay him compensations as if he had been negligent in the care he afforded to his patient…..In this case however the judge made the positive finding that the anaesthetist had performed the spinal injection in the appropriate place with all proper care. In those circumstances any possible inference of negligence falls away. "

82. Ratcliffe was not cited to the judge, but Mr Coghlan said that he had adventitiously, but correctly, applied its jurisprudence. The present case was also a case of a careful doctor faced with an unexpected outturn. On the basis of Ratcliffe, the judge's finding, in his paragraph 74, that Professor Rodeck approached his scan with "particular care" concluded the case in favour of the Trust.

83. In fact, the judge did not apply the jurisprudence inherent in Ratcliffe in the way contended for by the Trust. I would respectfully adopt the analysis of his judgment in that regard to be found in Latham LJ's paragraphs 29-30 above. The mere fact that the examination had, in mechanical terms, been carried out with scrupulous care was a necessary condition of the examination not having been negligent, but it was not sufficient for that purpose. I venture for two reasons to set out in some further detail why "care" in the conduct of the examination plays only that limited role. First, Arden LJ places some weight on the judge's finding of a careful examination: see her paragraphs [18], [20], [22], [34] and [37] above. Second, elucidation of this issue is important in understanding not only the incidence and nature of the burden of proof, but also the judge's route to his conclusion.

84. The fact that Ratcliffe was not cited to the judge in a trial attended by two specialist leaders suggests very strongly to me that the case was not seen by the profession as carrying the general significance that is now sought to be attributed to it. If that was the profession's perception it was, with respect, correct. A professional man is required, in the law of negligence as well as in the law of contract, to display not just care, but care and skill. That is so familiar a proposition that citation is hardly needed. But generations of professional negligence lawyers have grown up on McNair J's classic statement in Bolam, [1957] 1 WLR at p 586:

    How do you test whether this act or failure is negligent? In an ordinary case it is generally said you judge it by the action of the man in the street. He is the ordinary man…..But where you get a situation which involves some special skill or competence, then the test of whether there has been negligence or not is not the test of the man on the top of the Clapham omnibus, because he has not got this special skill. The test is the standard of the ordinary skilled man exercising and professing to have that special skill.

    And indeed the parties in the present case were agreed that skill as well as care was in issue. As the judge recorded in his paragraph 61:

    There was no dispute here as to the standard of care to be exercised by Professor Rodeck. It was the ordinary skill of his speciality, that is, as a consultant sonologist at a tertiary referral centre.

85. A case of professional negligence can therefore only be concluded by a finding of care alone when the facts are of the unusual nature of those in Ratcliffe, and in Delaney v Southmead HA (1995) 6 Med LR 355 which preceded it: where a routine, orthodox and in effect mechanical procedure, in Ratcliffe the administration of a spinal anaesthetic, produces entirely unexpected damage. In such a case, there are only two possible explanations: either the doctor was physically careless in performing the operation; or there is some underlying condition, unknown to medicine, on which a properly performed operation reacts adversely and for which the doctor plainly is not responsible. A finding that the operation was performed with due care leaves the latter as the only explanation. That is what this court held in Ratcliffe, and that is the category of case addressed by Brooke LJ in his observations cited above.

86. This is plainly not such a case. The reading of ultrasound images requires not just care, but also skill and judgement. The complaint in this case is that the requisite level of skill and judgement simply could not have been exercised, given that the results produced were so disastrously wrong; and given that an earlier reader, in the person of Mrs Wright, had not made the same errors. As my Lord says, this is not a case of res ipsa loquitur. But it is a case in which the outcome that Dr Rodeck attributed to his reading called for an explanation.

87. To say that an explanation is required is not to reverse the burden of proof, any more than the doctrine of res ipsa where it applies (being a rule of evidence and not a rule of law) reverses the burden: see on that point the judgment of Hobhouse LJ in Ratcliffe, cited by Latham LJ in his §26, and the treatment in Clerk & Lindsell (18^th edition), § 7-196. The claimant retains the burden of persuasion; but, given the need for an explanation the defendant must, in the first instance, produce an explanation that at least goes far enough to raise questions about the initial assumption of negligence. That is the approach to be found for instance in P v Leeds NHS Trust, cited by Latham LJ in his paragraphs 30, though it is by no means confined to facts of the type that occurred in that case.

88. How the explanation is to be weighed against the initial assumption, within the claimant's obligation to prove his case, may be an issue of some difficulty, more likely to be solved by a pragmatic approach to the evidence rather than by the application of any rule. This court in Ratcliffe in a general exposition, not limited to the particular facts of that case as discussed above, went no further than to say, at p 173 of the report, that:

    "(4) The position may then be reached at the close of the plaintiff's case that the judge would be entitled to infer negligence on the defendant's part unless the defendant adduces evidence which discharges this inference

    (5) This evidence may be to the effect that there is a plausible explanation of what may have happened which does not connote any negligence on the defendant's part. The explanation must be a plausible one and not a theoretically or remotely possible one, but the defendant certainly does not have to prove that his explanation is more likely to be correct than any other. If the plaintiff has no other evidence of negligence to rely on, his claim will then fail."

89. When this court spoke of the need for a plausible explanation, it was hardly using that expression in its Dictionary sense of having a [mere] show of truth. The sense is perhaps better captured by the judge's formulation in this case, in § 74 of his judgment, in terms of a "possible" answer or explanation. As this court said in its principle (5) set out above, mere assertion will not do; but neither need the explanation be shown to be the probable or likeliest answer. An explanation to that modest standard has to be reasonably available on the evidence taken in the round. As indicated above, how the court proceeds once such an explanation is established may in some cases be a matter of difficulty. That problem does not, however, arise in our case, because as Latham LJ has demonstrated, and I respectfully agree, the defendant did not pass the test of plausibility or possibility, as it did not succeed in adducing explanatory material that put the initial assumption of negligence under question. I return to that point at paragraph 101 below.

90. Ratcliffe therefore recognises, as did P, that cases will arise where the outcome of medical treatment calls for an explanation from the defendant. And it is clear that the judge indeed thought, and the Trust, albeit somewhat late in the day, agreed, that this was such a case, and that an explanation had to be given. That explanation was advanced by the Trust at trial in terms of "mimics" of the vital structures, that did or might have misled an expert. As the judge put it in his paragraph 34, set out by Latham LJ in his paragraph 12 above:

    "In the absence of any pictures of what the Professor saw, it was a matter of considering what structures could have produced the echoes mimicking the structures Professor Rodeck was seeking, and whether he should have realised the echoes were mimics"

    In following that agenda, it is necessary first to look at the way in which that evidence emerged, and then to consider whether it did indeed perform the task that the defence attributed to it.

    The emergence of the evidence

91. My Lord has described in paragraph 8 above the limited nature of the case originally advanced by the Trust. It followed from the Trust's resistance to this claim that, in its view, a highly skilled expert employed by it, following a standard procedure completely without negligence, had nonetheless given a patient advice that was seriously wrong, and which had led to grievous consequences for her and for her family. We were told by counsel, on specific instructions, that despite that no enquiry had been instituted by the Trust when, comparatively shortly after the scan was taken, the error became manifest. If elementary steps had been taken of peer review and the re-reading of such of the images as were available before they were destroyed, the Trust would have been able to say then, and would thereafter have been able to tell the judge, whether it could explain what had occurred in terms that did not inculpate it as negligent.

92. That was therefore an unpromising start to any attempt to explain the case now. And there are two further difficulties that stand in the Trust's way. First, the judge was told that Professor Rodeck and Dr Meire had discussed the problem when they first realised what had happened, but had not been able to find any explanation for the incorrect readings. If there indeed is, as is now claimed, an explanation for those readings, it is very surprising that it was not readily apparent to two specialists anxiously considering the case ten years ago.

93. Second, as is now customary good practice, a meeting was held between the two sides' experts shortly before the trial. The judge in my view correctly reported the position of the defence experts at that meeting in his §33 as having been that although Professor Rodeck must have seen echoes that seemed to him to be echoes of the structures that he wrongly thought that he had identified, in the absence of the original ultrasound pictures any view as to what other structures might have produced the echoes that the Professor had misunderstood would be merely speculation. That position was consistent with the lack of any reference to an explanation in terms of mimics in the defence experts' reports: see the judge's comment at the end of his paragraph 34.

94. It was that lack of explanation that caused the claimants' solicitors to invite the Trust to withdraw its defence. The solicitors were wholly justified in taking that step, and in my view the Trust was wrong not to accept that, on its evidence as it then stood, it could not succeed. We have seen that the judge held, and the conduct of the trial itself accepted, that the defence must not only identify a mimic for each of the three, separate, missing structures, but must also show that it was possible to follow that mimic without being negligent in so doing. The posture of the defence experts at the experts' meeting was that the Trust could not even adduce evidence answering the first of those question, let alone that addressed the second of those questions.

95. Nonetheless, the trial proceeded. Up to then, there had been no identification of what existing structures in the brain might have misled Professor Rodeck, and had caused the mimics that were later relied on. Nor was anything said about that in the witness statements of Professor Rodeck or Dr Meire. The first move in that direction did not come until Dr Meire gave his oral evidence, when he suggested that the echo that had been taken for the anterior horns could have come from the insula that forms at the side of the brain. Dr Rodeck adopted that suggestion in his own evidence, and added the possibility that the foremost part of the single ventricle might have produced an echo mimicking the cavum septum. It is a very singular fact that both of these propositions were later abandoned. We might be forgiven for thinking that that says something about the whole of this process.

96. But, once the possibility of some sorts of mimics had been ventilated, the defence experts returned to their task, and now produced explanations in terms of mimic that did or might address all the three mistakes: a task, it will be recalled, that had previously been characterised as speculative. These developments, that took place in the actual course of the trial, were extremely unsatisfactory. None of this evidence, which opened up a new, detailed and very difficult case, had been properly considered in advance. None of it had been contained in the original expert reports. However, the judge was not asked by the claimants to exclude the evidence or the line of defence that it supported, and in those circumstances he is not to be criticised for not taking what would have been the heroic step of excluding it of his own motion. What the judge very properly did do, in an attempt to reduce the trial to some sort of order, was to insist that the defence experts set out in writing what the case now was. That document was produced overnight on the fourth day of the trial, and the claimants' experts were invited to respond, in broadly the same time-scale.

97. Even forty years ago, the production of highly technical evidence crucial to the case half-way through the trial would have caused comment. In the days of Woolf, with advance exchange of expert reports; experts' meetings to limit areas of dispute; and detailed pre-reading and technical study by the court; litigation in that form is completely unacceptable. The court's procedures give parties ample time to prepare their cases and to assemble evidence in their support, and ample time to subject that evidence to scrutiny. Failure to respect those procedures threatens unfairness in removing late evidence from that scrutiny. That is not only very unfair to the opposite party and to the court, but is also unfair to the (in this case, very distinguished) experts who attend to assist the court. I will cite only one example. On the eighth day of the trial it was put for the first time, in the cross-examination of Mr Walkinshaw, that the "13-90" echoes (see paragraph 13 above) might have been caused by groups of blood vessels. (It may be noted that that theory was later abandoned, it being agreed that no clear explanation could be given for the 13-90 echoes). Mr Walkinshaw [Day 8, page 12] pointed out that he was being asked to think on his feet, as the claimants' experts had been required to do throughout the trial, but on that basis said that the only word to describe the theory was "absurd". The following exchange then took place

    The Witness: I feel, my Lord, quite strongly that this is a proposition that should have been put to Professor Griffiths who has significant knowledge of the anatomy of the brain over and above the knowledge that the other four experts have and I am not sure why this possibility was not broached when he was here.

    Counsel for the NHS Trust: I can tell you it was not put because it was not something which, to be absolutely frank with you, that my experts had thought about at that time.

    Counsel's observation says all that needs to be said about the nature of the process into which the court and the witnesses were drawn.

98. For the reasons that I have already indicated, the judge is not to be blamed for this; it was only because of his intervention that the trial stayed in any sort of proper form. But in future, parties must not shrink from inviting the court to exclude expert evidence that is produced actually in the course of the trial, and more particularly where its assumptions appear to contradict a position previously taken; and courts must take such objections seriously. All that said, however, in the event in this case the evidence was given, without objections that went to the length of seeking to exclude it, and the judge relied on that evidence. I must therefore go on to consider whether he was justified in the conclusions that he drew from it.

    The case in negligence

99. I put out of my mind, as has Latham LJ, the otherwise very strong scepticism about the value of the defence evidence that would follow from its lateness, and from the fact that the theories that it espoused had occurred to no-one who had considered the case, including the very people who were giving the evidence, until the eleventh, one might say the thirteenth, hour. I exercise that restraint because the judge, who heard the witnesses, accepted what they said at face value, despite strong submissions by the claimants that, for the reasons suggested above, the evidence was unreliable.

100. As indicated in paragraph 5 above, Alice's brain was defective in three respects: the falx; the CSP; and the anterior horns of the lateral ventricles. The scan had identified none of those defects. In order to take the first step towards establishing that the advice given to Mrs Lillywhite was not negligent, an explanation had to be provided for all three of those failings; because the identification of any one of the defects would have put the doctor conducting the scan on notice. I have already indicated, in paragraph [10] above, that in line with the guidance given by this court in Ratcliffe the explanation has to be a possible one, in the sense of being reasonably available on the evidence taken as a whole, without needing to be unavoidable or even probable. The case before the judge was conducted on the basis that an explanation had to be identified in those terms in respect of each of the mistakes. He summarised the agenda as it emerged at the end of the trial in these terms, in paragraph 67 of his judgment:

     "what does the evidence that is available suggest that Professor Rodeck may have seen? At the end there was largely agreement between Mr Walkinshaw, Mr Howe and Dr Russell. The only candidates that survived were the echo of the monoventrical beak, to mimic the cavum septum, the 13-90 echoes to which mimic the anterior horns of the lateral ventricles, and the small pieces of mid-line on Mrs Wright's picture (vi) for the anterior falx. I say largely agreed because there was not agreement as to the last"

101. It is therefore necessary to consider whether, in each separate case, it was open to the judge to conclude, as he did, that the particular explanation that the had been identified as meeting the test set out above was properly available. That process has to be undertaken applying the approach, and the limits on the power of this court, that Arden LJ has set out in her paragraph [39].

102. In the case of the falx, what was missing was its anterior part. The only candidate for a mimic for that part of the falx was Mrs Wright's "mid-line", to which Arden LJ refers in her paragraphs [27] and following. It is, however, clear from the judge's paragraphs 71 and 72, read together, and indeed from the defendant's experts' mid-trial statement (see paragraph [14] above), that no claim was made that Mrs Wright's line could be taken to represent the anterior falx; and as Latham LJ points out in paragraph 35 above, not even Mrs Wright herself suggested that "her" line had anything to do with the falx. And further, I am unable to read the evidence referred to by Arden LJ as displacing that failure of the witnesses to rely on the mid-line, whatever else it may have said about the identification of the falx. There was therefore no evidence on which the judge could return a positive answer to the question that he had posed to himself, as set out in paragraph 21 above.

103. As to the CSP, I respectfully agree with Latham LJ in his paragraphs 37-38 that the evidence did not justify even the tentative conclusion drawn by the judge.

104. As the judge said, the only candidate in respect of the lateral ventricles was the 13-90 echoes. But the judge in his paragraph 70 specifically accepted the evidence of Dr Twining and Mr Walkinshaw that by 1991, as the relevant article itself made clear, the possibility of confusion with 13-90 echoes should have been known to tertiary-level sonologists. A claim that Professor Rodeck might have been misled by 13-90 echoes could not therefore assist the defence, because to adopt it would in itself inculpate the sonologist of negligence. Arden LJ suggests in her paragraphs [31-33] that the judge nonetheless accepted, in the last sentence of his paragraph 70, that it might not have been negligent in a case where anterior horns were not present to mistake a 13-90 echo for a representation of those horns. The judge's observation seems to have been based on one sentence in Dr Russell's evidence, that Arden LJ has set out. If the judge did take that view I have no hesitation in saying that on the evidence, and as explained by him, that course was not open to him. The evidence that he accepted as to the state of knowledge about 13-90 echoes at the time of Professor Rodeck's scan was authoritative, set out in general terms without qualification, and expressed very strongly. The whole of the evidence of Dr Twining and Mr Walkinshaw on this point bears inspection, but I would merely cite one observation of Dr Twining under cross-examination [Day 7, page 10]:

     "We have it here in black and white in the literature that is being presented to us that this is a well recognised pitfall. This is not something that should be a surprise to any expert who is carrying out a scan. This is in black and white, written in 1990"

105. In the circumstances of this case it is perhaps otiose to observe that Dr Russell's theory, if such it was, was not advanced until the ninth day of the trial, well after Dr Twining and Mr Walkinshaw had given their evidence and departed. The theory was never put to them, and one suspects that it would have received short shrift if it had been. But leaving all that aside, it simply was not open to the judge to prefer Dr Russell (if that was his source: her evidence is never identified in this part of his judgment) to Dr Twining and Mr Walkinshaw without making clear the basis on which he reached that conclusion and the reasoning by which he justified departing from expert evidence that he had earlier said in terms that he accepted. The judge never even entered upon that process.

106. In each case, therefore, the explanations advanced by the defence were neither possible nor plausible, and so did not discharge the inference referred to in paragraph (4) of the Ratcliffe formulation. In the case of the falx and the CSP the explanations were not possible when looked at in the context of the evidence as a whole; and in the case of the lateral ventricles the only suggested explanation indicated negligent unawareness of the then state of the art. The defence therefore did not reach even the starting-point of positing an explanation that the claimant, bearing the burden of proof, had then to displace.

107. I accept that this was a very burdensome case, in which the judge had the advantage, which this court has not had, of seeing and hearing the witnesses. Nonetheless I find myself driven, like Latham LJ, to be unable to uphold the judge's judgment. On the material before him he should have found for the claimants, and I would substitute such an order for the order that he made.