England and Wales High Court (Family Division) Decisions

IN THE HIGH COURT OF JUSTICE
FAMILY DIVISION

		Royal Courts of Justice Strand, London, WC2A 2LL
		5 December 2001

DAME ELIZABETH BUTLER-SLOSS P:

1. At 11.35 am on 8 February 2001 a 5-week-old baby boy, B, born on the 2 January 2001, was taken by his parents to the Accident and Emergency Department of their local hospital with a history of left-side convulsions. A cranial ultrasound scan was normal and a skeletal survey did not reveal any bony abnormality. A lumbar puncture examination at 1.30 pm of the cerebro-spinal fluid revealed that it was slightly bloodstained and yellow in colour. It is not entirely clear whether, on arrival at hospital, the anterior fontanelle was soft or tense. A note in the hospital records, made between 11.30 am and 12.00 noon, disclosed it to be soft but on examination in the paediatric unit it was recorded about 12.40 pm that the anterior fontanelle was tense but not bulging. This indicated some intra-cranial swelling. The swelling was transient and had resolved by 10 February 2001. Dr Gorham, the consultant paediatrician, in his oral evidence, suggested that the original examination in the Accident and Emergency Department may have been by someone inexperienced in examining young babies. There was a bruise to the right shin. A nurse recorded a bruise to the head, not otherwise observed or recorded.
2. He had a computerised tomography (CT) scan on 9 February 2001 which disclosed two relatively small and symmetrically situated acute subdural haematomas adjacent to the falx and superior sagittal sinus in the parietal region and a third small focus of fresh haemorrhage in or over the upper part of the frontal convexity. There was no intra-cranial swelling seen on the CT scan. He had further fits affecting the left side over several days. He received anti-convulsant treatment and had no further fits after 10 February 2001. On 13 February 2001 after an examination by an ophthalmologist B was found to have bilateral retinal haemorrhages which were more marked on the right side. The retinal bleeding had not previously been noted. B did not appear to the medical team at the hospital to have suffered an impact injury. They excluded the possibility of disease. B had clearly suffered an injury to the brain. The parents said that the elder sister C, nearly 3, had fallen on the baby's head 5 days before. This was not considered to be a probable cause of the injury. The parents had no other explanation and the treating paediatrician came to the conclusion that the injury appeared to be non-accidental and was probably caused by an adult shaking the baby.
3. The child protection procedure was put into effect and the police and the local authority became involved. By agreement with the parents and the extended family, the elder children, C and a brother J aged 7 went to different grandparents. B, on discharge from hospital, joined his elder brother at the home of the maternal grandparents where he has remained. Both the elder children have now returned home to both parents.

Background

4. The parents are not married although they hope to marry next year. The father of B is also the father of C but not of J. During the pregnancy of C the mother suffered from serious depression and was prescribed Prozac. Because of her problems the father gave up work and helped her care for the children. He had just gone back to work during the hearing of this case. The mother did not take Prozac during the pregnancy of B for the sake of the baby. Her state of health after the birth of B was such that she went back on Prozac 3 weeks after his birth, first one pill per day, but under the stress of his removal from her and the impending care proceedings she began, in April, to take two pills per day which she is taking at present. Both parents gave similar accounts of their care of the baby. The father looked after the three children including the baby during the day and the mother got up for the baby during the night which might be up to three times a night. The mother went back to bed while the father got the elder children up and J ready for school. She would also try to sleep during the day if she had an opportunity. This was not easy in a household with children and she clearly got very tired. B was an attention-seeking baby who cried a lot and more so than either of the other children. He would cry after 10 minutes if not given attention.

Account of events on 7/8 February

5. On the 7 February 2001 B had been generally well. He was fed by the father about 7.00 pm and brought the milk back. When the parents went to bed B was put to bed in the same upstairs room which was not heated. At 1.00 am the mother gave him a full bottle and part of a second bottle. On that occasion she fed him in the bedroom. At 4.00 am when the baby woke she took him downstairs where it was warmer and gave him a bottle sitting on the sofa. After he had finished his feed he began to twitch and the twitching affected his left face, left arm and left leg. She became alarmed and took him upstairs and woke the father. There were further fits. After J had gone to school, the parents took C and the baby to the maternal grandmother, left C with her and went to see the general practitioner. He advised waiting a day and if the fits continued they should return the next day. The parents were dissatisfied with the general practitioner's response to the condition of the baby. They sought the advice of both grandmothers who thought they should take B to hospital, which they did immediately.
6. The police interviewed both parents, each on two occasions. I should like to pay tribute to the careful and sensitive way in which both parents were questioned. The parents were asked in the interviews with the police and in evidence to the court whether there was any incident which might have caused the injury. They said that it might be due to one of the other children but were not sure. They both referred to C falling on the baby the previous Sunday. They referred to a bouncy chair in which the baby was placed. I refer below in more detail to the suggested explanations given by the parents. Both parents strongly denied shaking the baby.

[7] There are no pending criminal proceedings. As a result of the sensible arrangements within the extended family it has not been necessary for the local authority to apply for interim care orders. B has made an unusually good recovery considering the severity of the injury and an encouraging aspect of this otherwise serious case, if the local authority proved its case, was that the hoped for outcome was the return of B to both his parents subject to assessments.

8. The application before me was by the local authority, and I was asked in this case to find that the Children Act 1989, s 31 threshold had been reached to enable the local authority to intervene. The guardian ad litem supported the local authority both as to s 31 criteria and the outcome of the case. In order to cross the threshold I had to decide whether either or both parents caused the injury by inappropriate handling as by shaking, or whether the cause of the injury was accidental in one of the ways suggested by the parents or was otherwise unexplained. I reminded myself that the standard of proof was the civil standard, the balance of probabilities, but that this was a serious allegation and the evidence must stand up to careful scrutiny and be cogent; see the speech of Lord Nicholls of Birkenhead in Re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, sub nom Re H and R (Child Sexual Abuse: Standard of Proof) [1996] 1 FLR 80, at 586 and 95 respectively. I informed counsel that I would try to give a decision after submissions in order to speed up the next stage of the case, if there was to be a next stage. After hearing the evidence of both parents and six expert witnesses and the written and oral submissions of counsel for the local authority, for the parents and for the guardian on behalf of B, I came to the conclusion that the local authority had proved the s 31 requirements to the requisite standard of proof but I was unable to come to a conclusion which of the two parents was responsible for the injury to B. In the light of the decision of the House of Lords in Lancashire County Council v B [2000] 2 AC 147, [2000] 1 FLR 583 it was sufficient to satisfy the threshold conditions for the local authority to show that the injured child was in the care of both parents as the primary carers. I reserved my reasons for coming to that conclusion which I now set out in this judgment.

Medical expert evidence

9. The question whether the s 31 criteria were proved by the local authority depended almost entirely upon the medical evidence. Seven consultants with different areas of expertise were asked to give written reports and six gave oral evidence. They were Dr Nelson, neuroradiologist, Dr Sprigg, radiologist, Dr Gorham, paediatrician with special interest in neurodevelopmental paediatrics, Mr Galloway, ophthalmologist, Professor Whitwell, forensic pathologist and Dr McCarthy, ophthalmic pathologist. The experts were unable, for various reasons, to meet as a group, although some of them did meet and I have extracted from the notes of agreement and from the oral evidence the points which seem to me to be agreed or not seriously contested. The uncontested evidence was that the baby had suffered an injury to the brain. The injury was likely to have been caused within 48 hours of the admission to hospital. The fits were the result and not the cause of the brain injury. The subdural bleeding was fresh. Both radiologists suggested it had occurred within 3 days. Dr Sprigg said it was possible it was up to 7 days but no more. Birth trauma was therefore excluded. The injury could not have been caused by C falling on top of her baby brother 5 days before. There was no evidence of disease and the experts unanimously excluded it. The presence of subdural haematomas might lead to the conclusion that the baby had suffered an impact or a 'shaking' or hyperextension flexion, otherwise whiplash, injury.

Impact injury

10. There was a suggestion by Miss Ball QC, for the parents, that there might have been an impact injury, based on the bruise recorded by the nurse and a right-side linear mark close to the bleeding at the front seen by the radiologists on the CT scan. Both radiologists were satisfied that the linear mark did not represent a fracture of the skull. I accept their evidence on this point. None of the forensic experts was able entirely to rule out the possibility of an impact injury. The parents gave evidence that after the fall of C on to her baby brother he had a raised purple vein on his head which looked like a bruise and when he cried it became more prominent. It is certainly possible that there was a bruise resulting from the earlier impact with the elder sister. It does not seem to me to have any relevance to the cause of the brain damage. The possibility of an impact injury was not pursued by Miss Ball QC in her final submissions. There was no evidence of impact and I do not consider it to be probable. I therefore exclude it.

Parents' explanation of injury

11. In excluding the possibility of an impact injury, the question arose whether B was shaken inappropriately or suffered an injury accidentally of a 'shaking' kind in the home or from either his brother or his sister. The parents identified three possible explanations for the injury. The first was the fall by C on her brother's head 5 days before. That was excluded by all the experts. The second possibility was that the elder brother J had handled B inappropriately in picking him up and putting him on his lap. They said that they had the habit of placing B on a pillow on the sofa which he preferred to being in his Moses basket. On at least two occasions J had picked up the baby and sat him on his knee, although he had been told he was not to do so. When seen J was sitting peacefully with the baby but it was possible that there had been some rough handling of B by J. The father thought that was unlikely and there was no evidence to show in any event that J had done this, within the likely time frame of the injury within the previous 2 days before admission to hospital. The third possibility was the use of the baby bouncer. The parents described the way in which B was placed in a baby seat on the floor described as a baby bouncer. It was of an old-fashioned type with a strap across the baby's waist but not restraining the shoulders. C had the habit of putting her feet on the bar at the base of the baby bouncer and with her hands on the top bouncing the chair vigorously with the baby seated in the chair. She did this on many occasions although she was told not to do so whenever she was caught doing it. The last occasion she was seen to bounce on the back of the chair was probably about 2 days before the fits started. Miss Ball QC also suggested the possibility that the injury might have occurred in the ordinary rough and tumble of family life.

Medical evidence as to possible causes of injury

12. A number of different possible causes of the injury were carefully explored in the medical evidence. The presence of the subdural haematomas, the retinal haemorrhages and the transient intra-cranial swelling were the three most significant factors to be considered.

Retinal haemorrhages

13. At first sight the retinal bleeding appeared to be significant and has been, for some years, considered by paediatricians and ophthalmologists examining young children to be a significant pointer to the possibility of non-accidental injury. It has been found to be an important factor or the important factor in previous brain injury cases; see for instance Bracewell J in Re A (Non-accidental Injury: Medical Evidence) [2001] 2 FLR 657. Understandably, the presence of bilateral retinal haemorrhages in conjunction with subdural haematomas was seen by the ophthalmologist, Mr Galloway, the two radiologists and the paediatrician as highly significant and supportive of the conclusion that the brain injury was due to non-accidental injury. The significance of retinal haemorrhaging in the present case is, however, not altogether clear. Dr McCarthy, as a forensic pathologist, has had considerable experience in injuries to the eyes, seen, of course at the most serious end of the scale, in dead children. He raised a number of question marks over the presence of retinal bleeding in this case, and how it might have been caused. The possibilities raised included that the fits had continued over a period of 4 days, the duration of some of them being up to 45 minutes; that there may have been a sudden rise in intra-cranial and intra-ocular pressure, and the possible effect of the lumbar puncture in raising intra-cranial pressure. It was unfortunate that there was no record of observation of B's eyes at an early stage. The eyes were not examined by the hospital ophthalmologist until 13 February 2001 and the presence of retinal haemorrhaging was not recorded until 5 days after his admission. I do not therefore set out the evidence of Mr Galloway and Dr McCarthy, the medical experts on this subject, since the absence of any evidence about the eyes on admission to hospital or at least on the day that B was first examined raised, on the facts of this case, possible alternative causes for the retinal bleeding. It would be fruitless to explore further how the bleeding had been caused and whether the other possibilities had validity. I could not be satisfied to the requisite standard of proof what was the cause of the retinal bleeding and I did not therefore rely on it in coming to my conclusion that B suffered non-accidental injury. I do not, by my conclusion on the special facts of this case, wish to be casting doubt on the conclusions to be drawn in other cases as to the significance of retinal bleeding where allegations of baby-shaking are made.

Subdural haematomas

14. The conclusions to be drawn from the presence of subdural haematomas has been at the core of the dispute between the parties and between the experts. The presence of subdural haematomas in a young child or baby has become a serious indicator of non-accidental injury among clinicians dealing with young children, in the absence of an acceptable explanation for the injury. The view is widely held that subdural haematomas indicate a shaking injury and that the shaking must have been forceful to cause the tearing or shearing injury to the membranes in the space between the dura and the brain.

Dr Nelson

15. Dr Nelson said his expertise was in the area of subdural haematomas. He described the injury as seen on the CT scan. The baby presented with two haematomas symmetrically situated in the parietal region. The third area of bleeding was close to the anterior fontanelle. In his view they occurred not more than 3 days before the scan, (on 9 February 2001). The lumbar puncture supported bleeding as seen on the CT scan. He would expect the bleeding to be seen on examination by a lumbar puncture about 6 hours after the injury but he would accept it might take 12 hours for the blood in the cerebro-spinal fluid to begin to deteriorate. Fitting can occur at the time of the shearing of the membranes. Bilateral haematomas at the back of the head were, in his experience, characteristic of a shaking injury. Shaking was likely to be backwards and forwards by holding child on both sides of the chest. There were two effects of shaking. First a lower brainstem hypoxic/ischaemic event where the shaking created tension to the brainstem/cervical cord at the cranio-cervical junction and a disturbance and transient interference to the flow of blood and oxygen to the brain. In the present case, that was shown by the intra-cranial swelling over the anterior fontanelle. The second effect was a shearing force on the cerebral veins bridging the subdural space. The shearing of the membranes would be concurrent with but separate from the hypoxic/ischaemic event. Fitting might start at the moment of the injury or it might follow up to 24 hours after the injury. It was more likely that the subdural haematomas were due to shaking prior to the start of the fitting. He considered that the site of the subdural haematomas in the parietal region and their symmetry were indicators of a shaking injury. The good neurological outcome was less characteristic of such an injury.
16. He accepted that a relatively small amount of force was needed to create a hypoxic/ischaemic insult and did not disagree with the paper by Geddes et al (to which I refer below). He pointed out that the paper did not address subdural haematomas or retinal bleeding which required a greater level of force. From his clinical experience this injury did not occur through normal handling of a baby, even rough and vigorous play with other children. In his report and addenda Dr Nelson suggested a greater degree of force was required than he said was required when he came to give his oral evidence, but he was firm that the degree of force would be inappropriate and more than normal play.

Dr Sprigg

17. Dr Sprigg confirmed in general the evidence of Dr Nelson. He agreed that the most likely cause of the injury was shaking prior to the start of fitting. It was not normal to shake a baby. The handling of the baby would be inappropriate and unacceptable and the person causing the injury should have realised that he/she was holding and shaking the child in an inappropriate way. He disagreed with Dr Nelson on the significance of the site and symmetry of the subdural haematomas. In his experience bleeding might be found in any part of the subdural space. Although he did not rely on symmetry or the site of subdural bleeding it was very common to find bilateral haematomas at the back of the head. Subdural haematomas on a CT scan were markers of significant head injury. They were not to be found on routine CT scans and their presence was a very significant finding. He did not consider that the bouncy chair used as described could cause subdural haematomas. The baby's head would be supported and there was not likely to be a hyper-extension flexion movement sufficient to cause them. There was no satisfactory explanation of the cause of the injury. It was most likely to be shaking for a constellation of reasons: no external swelling, no skull fracture, the presence of subdural haematomas and no explanation.

Dr Gorham

18. In the conclusions to his written report to the court, Dr Gorham stated:

'Unacceptable and excessive force would be needed by an adult to cause such a severe injury in a small infant of this age. This injury could not have occurred as a result of normal handling. The carer who was responsible for the shaking would have been fully aware that he/she had used excessive force on the infant. This injury is thought to occur most commonly during a sudden outburst of temper with a small baby.

In the light of current medical evidence the most likely explanation of this child's injury would be that it had occurred as a result of a severe shaking injury by an adult several hours prior to admission to [hospital].'

19. He agreed, in his oral evidence, that there was no experimental work done with very young babies and no precise estimate of the force needed to cause such an injury. There was very little research generally in all infants. In his clinical experience it was not a common finding in the absence of a history of injury. Where there was no explanation and retinal haemorrhages and subdural haematomas were present there was a very high index of suspicion. They were known to be associated with injuries which had been clearly identified as non-accidental. He accepted that where there was no explanation and no admission it might not be non-accidental. As a paediatrician it was his role to co-ordinate the overall picture. In clinical practice it was extremely unusual to see a neonate with subdural haematomas, retinal bleeding and fitting. A subdural haematoma was a very rare occurrence in babies and extremely rare in a 5-week-old baby or a neonate. It would not appear to be an injury from normal everyday handling. It would be an event over and above normal handling.
20. The mechanism of the injury was sudden force, not force over a period of time. There was a debate over the mechanism of subdural haematomas. It had to be an event and one would look for an explanation. A common explanation was a baby crying in the middle of the night and the parent lost control and violently shook the baby before realising what he/she had done. He/she might not realise the seriousness of the shaking action. Fitting was the common sign of such an injury. Another sign was apnoea. In his view the most important element of the injury was the presence of subdural haematomas. It was common to have a fit very quickly after an injury to the brain. In normal circumstances fitting reduced with time. There could be a 'lucid interval' between the injury and the fitting. The experts agreed that 24 hours was the cut-off time but it was an arbitrary cut off. From his clinical experience small babies with head injuries commonly have fits immediately after the injury. In normal cases of injury a child would be affected in feeding. The child would be ill but it would not always be recognised by carers or by the general practitioner. It would be recognised by paediatricians. If a minor trauma could cause subdural haematomas then they would be seen more often. In cases where there has been a significant degree of rough handling it did not lead to an injury such as this. It would be beyond the normal bracket of household injury. Children suffering from quite severe trauma from a household incident such as falling downstairs or off settees would not have suffered subdural haematomas. The level of forceful trauma was such that, in the light of the current medical knowledge, a person causing it would know it was unacceptable. It was highly likely that less force was used in a case where the child had a good outcome. Most cases have far more severe results.
21. He was asked about the bouncy chair. He said that millions of babies use not dissimilar chairs without injury. It was common practice for paediatricians to check on baby equipment, such as baby walkers, which had been criticised. Even with C bouncing on the back of the bouncy chair he did not consider it likely that the injury occurred in that way. He had seen children who had been bounced and had fallen downstairs without suffering subdural haematomas. In his opinion it was extremely unlikely that the injury was caused to B in the bouncy chair in the way described.

Professor Whitwell

22. She is a co-author of research in two papers published, under the heading 'Neuropathology of inflicted head injury in children' by Geddes et al, in the medical journal, (2001) 124 Brain 1290 et seq. She provided a report to the court together with the two published papers.
23. In her written report to the court she explained her research:

'In infants there is a spectrum of injury ranging from clearly at the very serious end, fatalities, to those with relatively minor injuries. With respect to my own research the major importance of this is that we have demonstrated that the brain damage in fatal infants is not due to severe primary traumatic brain injury but due to lack of oxygen secondary, in some cases, to neck injury. I do not think it is possible to quantify the degree of shaking necessary to produce either brain damage due to lack of oxygen and neck injury and subdural haematomas in young infants. I do not agree with the term violent. It may well be that a fairly simple, but excessive hyperextension/flexion movement takes place. It may be that in this case that is what we are looking at, although there is no clear evidence as such. I would point out that this field is extremely difficult and fraught, with in many cases, a lack of scientific objectivity. The series, of which the vast majority of cases were my own has attempted to elucidate some of the fundamental issues raised, in particular the causation of the brain damage. The mechanism of the subdural haematomas in this age group is still open for considerable debate.

It is well recognised that subdural haematomas together with retinal haemorrhages may be caused by shaking or shaking impact or impact to the head … I note the history of the bouncy chair and the child possibly being bounced. The conventional teaching is that this does not produce sufficient forces to cause subdural haematomas. However, it has to be said that there is no experimental work available to indicate the degree of force necessary to produce subdural haematomas, particularly in the young up to a few months old. In general these subdurals are very thin films and whilst in adults it was recognised that significant force, usually with impact, is required to cause a subdural as a result of tearing to the underlying bridging veins, the evidence for this in infants is less clear and indeed it is currently impossible to give any even reasonably, precise estimate as to the degree of force. None the less, it is fair to say that many children are subject to being thrown up and down in the air and/or being in bouncy chairs etc and do not appear to suffer subdural haematomas …

As regards hypoxic-ischaemic injury as this is likely related to neck injury, there is no experimental data to quantitate force. It is, however, the currently accepted view that this would be in excess of normal handling. How many "shakes" is impossible to say. It is possible one or two uncontrolled neck movements in a young infant may be sufficient to produce damage to the brain stem/upper cord with breathing irregularities.'

24. In her oral evidence she described how the research on 53 children who had died had been a breakthrough on the previously accepted view that injuries caused to the brain had been as a result of direct trauma. The basis of the two papers related to the issue of severe, very violent shaking type brain damage which the research showed was due to lack of oxygen rather than primarily traumatic. The damage was not caused directly to the brain but to the neck with the effect of starving of oxygen, causing a hypoxic event in the brain by swelling and damage to the axons. The brain damage was due to lack of oxygen, not to the application of direct force to the brain. Since the injury was not directly to the brain but to the neck, the force required to cause serious injury to the brain need not be severe.
25. She agreed that they had not concentrated in their research on conclusions about subdural haematomas. Subdural haematomas were not caused by the hypoxic event but, as a result of the same movements that cause injury to the neck, they might occur at the same time as the injury to the brain arises from the hypoxic event. Subdural haematomas were not caused by swelling in the brain. She raised the question whether the degree of force required to cause subdural haematomas or retinal bleeding needed to be as severe as was previously considered necessary in much of the earlier research. She said that, in the absence of disease, subdural haematomas were most likely due to trauma or injury. In adults the physics of acceleration and deceleration and the tearing of the bridging veins between the brain and the dura had been researched. Such research had not been carried out in respect of young children. It was not known in the very young what was the mechanism for this tearing or shearing to occur to the small blood vessels between the brain and the dura. One could not compare a baby with an adult.

Very young babies had brains with a much higher water content and immature fibres. Blood vessels within the brain were less developed. Injuries to the neck had been seen in association with subdural haematomas. It was not known how much force was needed to cause subdural haematomas.

26. It was inappropriate behaviour towards a baby to shake it but a rapid flop back or forward without supporting the neck might cause injury. A single movement might be sufficient to cause subdural haematomas. She just did not know what degree of force was required. In earlier research a high proportion of children with subdural haematomas showed indications of non-accidental injury. In 20% of cases there was no clear evidence of non-accidental injury or even trauma. No one could say in that group how it happened. She expressed concern that if subdural haematomas were found by scan and there was no explanation of their presence then the injury was labelled as non-accidental injury, because rough play in the household was not accepted as a possible explanation. She recognised that if subdural haematomas occurred as a result of rough play there would be large numbers in the Accident and Emergency Departments of hospitals.
27. In the present case the child's fits over the period of 4 days could be due to minor swelling, but there was no major hypoxic event. The small amount of intra-cranial pressure was likely to have resulted from the subdural haematomas. The presence of subdural haematomas, retinal haemorrhages and swelling showed that the most likely cause was some trauma to the child. It required a degree of trauma for subdural haematomas to occur and that might be as a result of 'shaking', impact or impact with shaking. There was no evidence of impact. She was asked whether the actions of C on the bouncy chair, causing a movement of the baby's neck, could have caused the injury to B. She said that she could not say it would not be an explanation. She could not rule out the bouncy chair. It was possible that the injury could have been caused by the eldest child J picking up the baby and holding him in an inappropriate way. There was no explanation. A baby in the age range of under 4 months was where 'shaking' in the broadest sense with no impact was seen. With older children an impact was more likely. She agreed that the cause of the injury would be in excess of normal handling and there would have been some trauma to the child. She agreed that it would point the finger at someone causing the trauma but there were a significant number of cases where there was no explanation. This was of concern. She questioned what was the link between force and the mechanism of retinal haemorrhages and subdural haematomas. She agreed that subdural haematomas were a serious matter and, if present, it was likely that there were symptoms to be seen and it was not likely that children were walking around with subdural haematomas. But if there was no explanation one should be cautious about labelling the injury as non-accidental.

Dr McCarthy

28. Dr McCarthy was called to give evidence about the significance of the retinal haemorrhaging. In his written report to the court he stated:

'It is well recognised and generally accepted that subdural haemorrhages together with retinal haemorrhages may be caused by shaking an infant or shaking with impact or with an impact to the head of an infant. However, there are many areas of uncertainty and speculation with regard to the degree of force, the nature of the event causing the damage and changes, and the interplay with other events that are present in each individual case …

The first factor that requires consideration is the age of the infant, which will generally embrace issues such as size and development, as there are differences at different stages in the causation of non-accidental injury and its effects.'

At the end of his report he commented:

'Accepted current medical practice associates subdural haemorrhages and retinal haemorrhages in infants with the shaken or shaken/impact baby syndrome …

There is, however, a deficiency or lack of scientific medical evidence regarding the force that is necessary to cause these lesions …

In the case of B, while shaking cannot be excluded as the cause of his illness, other factors are present which must be given due consideration.'

29. As a forensic pathologist he had considerable experience in the very young baby and was asked to answer more general questions. He said that 5 years ago he would have agreed that the presence of retinal haemorrhaging and subdural haematomas showed that the injury was likely to be by shaking or shaking impact. Now he was not so certain what was the causation of injury to young babies just after birth. He pointed out that B, 5 weeks old and born at 39 weeks, was only just outside the 4-week definition of neonate and should really be considered as such. There was a lack of knowledge about the cause of damage and the mechanisms affecting very young babies.
30. The problem was that the evidence was available at the severe end in children who had died and where the pathologist was able to inspect the brain. There was a lack of evidence at the other end where the trauma was mild or minimal. The fits were caused by the intra-cranial trauma. The transient intra-cranial pressure may have been due to the gradual increase in the size of the subdural haematomas. Since they were small when seen on the CT scan the pressure on the brain may have been gradual. The neonatal brain was not fully developed. The question was what degree of force was required to cause an injury such as occurred to B. Some force was necessary to cause damage. Something happened to the child but he asked the question: how abnormal was the behaviour that caused the injury? There must have been some event to cause it. There were not a lot of children with subdural haematomas and such a child would be an ill child. It was too easy to say that the baby was shaken. He accepted that clinicians faced with these problems were adhering to the present practice on the basis of the medical knowledge available and were acting appropriately. Subdural haematomas and retinal haemorrhaging did not, however, equal non-accidental injury.
31. He considered the possible explanations proffered by the parents. He dismissed the possibility of J picking up the baby as a likely cause. Like the other forensic experts, he looked carefully at the bouncy chair which was on view in court. His conclusion in his report was that it was an attractive theory but highly unlikely. In his oral evidence he said it was more likely that the injury did not happen in the bouncy chair. There was no explanation for the injury. The most likely cause was some event which involved trauma within the past 48 hours, something out of the ordinary. It needed force in excess of normal handling but not the degree of force hitherto regarded as necessary to cause such an injury. The issue in the case was the presence of acute subdural haematomas which were a matter of the greatest concern.

Conclusions on medical evidence

32. The distinguished research carried out by Professor Whitwell and others and set out in the papers by Geddes et al, is clearly of the greatest significance in demonstrating that diffuse axonal injury (DAI) was not likely to be the cause of brain injury in many child deaths. The primary cause was identified as injury to the neck causing hypoxic/ischaemic damage to the brain, by starving of oxygen or blood causing raised intra-cranial pressure and/or a brain shift. It followed that it was not necessary to have the very considerable degree of force from direct impact to the head, assessed in some earlier research as the equivalent of a road traffic accident, in order to cause serious intra-cranial pressure and, in the cases seen by Professor Whitwell, death of the child. Put simply, once it is appreciated that the primary injury is to the neck and not to the head, one can see that far less force to the neck is needed, than would be to the head, in order to cause serious damage and death to children.
33. This research was however not directed to the degree of force required to cause subdural haematomas. The mechanism for causing these collections of blood between the dura and the brain in babies is not yet explored, but they are undoubtedly separate from and unconnected with the intra-cranial swelling and any other injuries within the brain, other than that they may be caused at the same time as the brain injury. None the less the recognition that the force is primarily directed to the neck, which is vulnerable in young children and particularly in babies must raise questions as to the degree of force required to cause subdural haematomas. There is so far no scientific research available to provide the answers.
34. It is clear from the medical evidence that it would be highly desirable for some research to be undertaken specifically in the field of injuries to young babies of 3 months or under, since these are not unusual cases to come before the family courts and the issues raised in this case deserve further consideration.
35. As Dr McCarthy said, well-recognised and accepted current medical practice associates the cause of subdural haemorrhages with shaking an infant. Professor Whitwell and Dr McCarthy raised the question whether the accepted view was necessarily correct in some of the cases where there was no explanation. It is clearly an important question to be answered especially in a court and judges in family cases must take the greatest care not to jump to conclusions in this difficult area.
36. Professor Whitwell's evidence was interesting and challenging. She posed questions and gave few answers. She challenged the existing medical view on the significance of the subdural haematoma and, in the absence of scientific research, refused to exclude any possibility, and I would add, however improbable it might be. I did not find her evidence helpful and where she differed from Dr McCarthy, I accept his evidence which I found to be impressive.
37. Dr Gorham, Dr Nelson and Dr Sprigg support the widely recognised medical practice in the present case. They all accept that there is a lack of research. They all modified their initial view that considerable force was needed to cause subdural haematomas and recognised in oral evidence that less force than previously believed necessary might cause an injury such as B suffered. The fact that B was in effect a neonate with an undeveloped brain was an added factor in the degree of force required. It would seem particularly so in B's case since he made an unusually good recovery and his injury was obviously at the mild end of the spectrum of injury. None the less, all three experts were agreed that some force was necessary, that it was inappropriate handling of a child and ought to be known to be so by the adult causing the injury. Rough household play would not be sufficient to cause the injury. The most likely cause was shaking. Dr McCarthy agreed that some force, in excess of normal handling, was necessary to cause damage and there had to be an event to cause it. He saw the presence of acute subdural haematomas as a matter of the greatest concern. Professor Whitwell raised the question whether household play would be sufficient, but was not supported in this theory by the other medical experts. None the less she also agreed that the injury would be caused by handling that was in excess of normal handling. All the experts accepted that the event would make B an ill baby and that children are not walking about with subdural haematomas. I accept the evidence of Dr Gorham, supported by Dr Nelson and Dr Sprigg, that it is important to look at the whole of the clinical picture in order to come to a conclusion.
38. The bouncy chair, as the cause of the injury, was discounted by all the medical experts other than Professor Whitwell. She left it 'in the frame' only to the extent of not excluding it as a possibility. I accept the majority evidence about the bouncy chair and do not consider that it was probable, or even possible that the bouncy chair caused the injury. One matter which appears to me to be of some significance is that C bounced on the back of the chair on several occasions seen by each parent, sometimes more and sometimes less violently. It does not appear to me to be probable that, with a baby of only 5 weeks, it was not until the last occasion that this scenario caused this serious injury. I would have expected the serious damage to have occurred much earlier.

Who caused the injury?

39. It is impossible to say when the injury occurred save that it was within 48 hours of admission to hospital. From the timing of the lumbar puncture at 1.30 pm on 8 February 2001, and accepting, as I do from the opinion of Dr Sprigg and Professor Whitwell, that it required at least 12 hours for the cerebro-spinal fluid to become bloodstained, the event that caused the injury was likely to have happened before 1.30 am. It follows that the fits did not immediately follow the event and that the injury was unlikely to have occurred at or after the 4.00 am feed.
40. I do not consider that either of the elder children had anything to do with the injury. Both parents in their evidence dwelt on the bouncy chair and the possibility of J or C being the unwitting culprit. I am satisfied that it was improbable that this injury was caused by rough household play. If it was sufficient, subdural haematomas would be seen in many cases and I accept the evidence of Dr Gorham on this matter. Although Professor Whitwell did not exclude the rough and tumble of family life, she was a lone voice and I do not consider it likely in the face of the other evidence that something utterly unremarkable could cause severe neurological damage with three collections of blood in the subdural region of the brain. No event other than those I have found to be improbable has been suggested by either parent. Neither parent gave me any possible other event that might have caused the injury. This was a 5-week-old baby who was not old enough to do anything to himself. Some event happened in that period of 48 hours about which I have not been told. He suffered some traumatic incident which could not, in my judgment have been unremarkable or overlooked, although the extent of its severity might have been under-appreciated. I do not therefore accept the submission of Miss Ball QC that an event might have occurred about which neither parent had any knowledge. The failure of either parent to identify any situation in which he or she had taken part prior to their young baby suffering this trauma makes it difficult to accept the truth of the evidence of either. It is impossible to say whether it is one or both who is or are not telling me what really happened. I recognise, as Miss Ball QC rightly pointed out, that these are respectable parents but I found myself unable to accept their evidence as credible. I therefore came to the conclusion, for the reasons set out above in this judgment, that one or both of the parents was responsible for the injury to their younger son and he/she or both knew of the incident that occurred and have concealed it from the court. The most probable cause of the injury was some form of shaking incident. Although the shaking may not have been severe and the degree of force was at the lower end of the scale, it was clearly unacceptable and inappropriate handling of a 5-week-old baby. It is however entirely possible that the serious consequences of the inappropriate handling might not have been understood at the time. It remains a seriously worrying aspect of these 'shaking' cases that adult carers still do not appear to understand just how serious are the consequences of losing control and shaking a baby, particularly a very young baby like B. In the present case, the long-term outcome of his injury is unclear but there appears to be a good prospect that he will not suffer long-term consequences. B has been unusually fortunate. I was therefore satisfied, for the reasons set out above, that B suffered significant harm sufficient to cross the threshold requirements of s 31 of the Children Act 1989.

Comments

41. I add a few general comments. The Geddes et al research has posed questions in the related field of subdural haematomas but it has not provided answers. As I said earlier, further research on the mechanism of subdural haematomas and the degree of force required to cause them in young children and babies would be very helpful for the medical profession faced with the results of injury in hospital, for the child protection teams and for the judges and magistrates who try these cases. We must be careful not to jump to conclusions nor to accept too readily the diagnosis of non- accidental injury in these 'brain injury' cases. Equally there is no research that entitles us as judges to dismiss out of hand the clinical experience of paediatricians and other medical experts derived from examining in hospital these children and their injuries and investigating the circumstances in which the injuries were said to be caused. All the experts in this case came round to the opinion that the degree of force required to cause subdural haematomas need not be as great as previously believed. It remains however equally clear that the force used must be out of the normal rough and tumble of family life and must be unacceptable and inappropriate and obviously so. Each case of course has to be decided on its own facts. This is likely to be an evolving area of research. These are difficult cases: babies must be protected and parents or other carers must not be unfairly treated. The courts must however continue to deal with medical evidence on the basis of generally recognised medical opinion, giving due weight in the individual case to any advances in medical knowledge. This is a difficult area and in any case where contested issues on brain injury, subdural haematomas or retinal haemorrhages are raised in the immediate future, it would be wise for the court to consider where it would best be tried, depending upon the facts of the case and a realistic approach to the likely medical evidence to be called.

The threshold condition under s 31 of the Children Act 1989 found to be satisfied.

PHILIPPA JOHNSON

Barrister

The permission for BAILII to publish the text of this judgment
was granted by Jordan Publishing Limited
Their assistance is gratefully acknowledged.