AD & AM (Fact-Finding: Re-Hearing)

This judgment concerns two children, a boy AD aged 4, and his older sister, AM who is just 7 years of age. They are the two children of MB (the mother) and FB (the father); they currently reside at home with their parents together with their paternal grandmother. In March 2014 I made a shared residence order in favour of all three adults in respect of the two children. There is currently a supervision order in favour of the London Borough of Newham (“the Local Authority”); that order has been extended throughout these proceedings.

This is the fourth substantive judgment which I have delivered in this case, and for a complete understanding of the evolution of this litigation all should be read. The first judgment was delivered in July 2013 at the conclusion of a five-day fact-finding hearing (Re AD & AM (Finding of Fact: Non-Accidental Injury) [2013] EWHC 4859 (Fam)); I there explained my finding that MB had inflicted near-fatal injuries on her son, who had suffered massive head wounds and spinal fractures. In March 2014 I conducted a welfare hearing at the conclusion of which I delivered a judgment explaining my reasons for endorsing the Local Authority’s plan, supported by the Children’s Guardian, to rehabilitate the subject children to the care of their parents and grandmother subject to strict conditions (see Re AD & AM (Non-Accidental Injury: Welfare) [2014] EWHC 4899 (Fam)). In March 2015, the Local Authority applied to the Family Court at East London for an extension of the supervision order, and in July 2015, within those proceedings, the mother applied for a re-hearing of the 2013 findings of fact. The applications were transferred to the High Court and listed before me in January 2016; I delivered judgment on 24 February 2016 (Re AD & AM (Fact-Finding hearing) (Application for re-hearing) [2016] EWHC 326 (Fam)) dealing with the mother’s application.

My February 2016 judgment addressed the first (“stage 1”) test on an application for a rehearing (see Re ZZ & Others [2014] EWFC 9 at [31] ("Re ZZ")); having read the newly-obtained medical reports (commissioned within the mother’s criminal proceedings) and heard argument, I was satisfied that there was “some real reason to believe that the earlier findings require revisiting” (per [12] and [80] of [2016] EWHC 326 (Fam)). By that time, it was known that a piece of AD’s skull, removed in the life-saving operation in October 2012 and then lost, had been discovered; this fragment was believed to reveal significant features of woven bones, possible signs of vitamin D deficiency, and unusual thinness at the fracture site. A number, eight in all, of additional medical experts (who were listed at [41] of the February 2016 judgment [2016] EWHC 326 (Fam)) had provided further opinion. In particular, I was persuaded that there was a basis for the mother to contend that:

I then gave directions for the conduct of ‘stage 2’ as follows (per [87] [2016] EWHC 326 (Fam)):

“Stage 2 must be conducted expeditiously, fairly, and proportionately in order to comply with the overriding objective of rule 1 FPR 2010. I believe that the "forensic context" of this particular case (…), requires an urgent experts' meeting (to include as many of the experts from both rounds of litigation: 2012/2013 and 2015/2016) as an important next step. I am not prepared to be drawn on the extent of the shape or extent of any enquiry following such a meeting; I will be guided by the outcome of the meeting and a consideration of the combined / informed views of the experts”.

Two experts’ meetings took place (by telephone, given the global distribution of the participants) in August 2016; the first meeting involved Prof Freemont, Dr Cohen, Prof Nussey, and Dr Cartlidge; the second meeting involved Mr Richards, Dr Fairhurst, Dr Watt, Dr Ramsay, Dr Teas, Dr. Van Ee, Dr Hyman, and Prof Hann. The meetings were both long, and intense, but were skilfully chaired by Dr. Margaret Bloom, barrister and former medical General Practitioner. The experts’ meetings exposed irreconcilable differences between some of the experts, and to some extent clarified and narrowed the issues between others.

By this judgment I address my reasoning for permitting a limited investigation and the calling (and re-calling) of evidence relevant to the facts (the ‘second stage’: Re ZZ at [34]), and my determination of the ‘stage 3’ process itself, that is to say the hearing of the review, where I have considered “the fresh evidence alongside the earlier material before coming to a conclusion in the light of the totality of the material before the court”.

I should state that the father supports the mother in her endeavour to set aside the findings, aware that on one outcome of the litigation, he could find himself in a worse situation than at the conclusion of the 2013 litigation, namely in the pool of perpetrators of injury. The Local Authority oppose the application, and contends that the evidence adduced in 2016 has proven the case more convincingly than the evidence adduced in 2013. The Guardian retains an essentially neutral stance, but has been invaluable in co-ordinating some of the expert witnesses. This re-hearing has required enormous effort on the part of the lawyers involved, whom I wish to credit for their endeavours.

Having conducted the relevant reading into the case (the new medical reports filled two of the ten lever arch files lodged with the court, the transcript of the experts’ meetings filled another – more than 200 pages of text), I heard submissions on the scope of ‘stage 2’. Mr. Storey QC and Mr. Bailey refined their objectives in this regard, importantly jettisoning any reliance on the evidence of Dr. Shaku Teas (Forensic Pathologist, Illinois, USA), whose views on a number of the medical issues attracted minimal if any support from any other expert. She had argued strongly that there had been a much older (possibly birth-related) fracture and a chronic bleed (which she dated to about the same time); that view, and other views, had attracted a certain degree of disparagement of other experts. Indeed Mr. Richards expressed the view to me in his oral evidence that he thought that some of her opinions were “bizarre” and “showed an ignorance to what happens in life”. No party indicated an intention to rely on the evidence of Dr. Chris Van Ee, Biomedical Engineer, Michigan, USA, recognising the limits of biomechanical evidence in an enquiry such as this. At the conclusion of submissions, I indicated to counsel that a proper basis had been made out for reconsidering the following issues:

The witness schedule was pared down further. Mr. Storey considered it unnecessary to call Dr. David Ramsey (Neuropathologist, Ontario, Canada). Professor Ian Hann (Consultant Paediatric Haematologist), Professor Tony Freemont (Professor of Osteoarticular pathology), Dr. Marta Cohen (Consultant Paediatric Histopathologist) were similarly de-warned. Those called to give oral evidence before me at this re-hearing were:

The parents indicated, through counsel, that they would not seek to give further evidence themselves. It follows that the ‘lay’ factual evidence, and the findings deriving from that evidence, remain untouched.

Before examining the evidence specific to the issues under investigation, I wish to make the following points about the applicable law, and the expert evidence.

This is no place for a rehearsal of all the legal principles in play in this case; the parties accept my summary of the key provisions of the law at [88] to [92] of the first fact-finding judgment ([2013] EWHC 4895 (Fam)). It is right that I nonetheless explicitly acknowledge the following core principles which are engaged:

Section 13 of the Children and Families Act 2014, rule 25 FPR 2010 and the associated Practice Direction (PD25) set clear parameters for the instruction of experts in family proceedings, including children proceedings. It is now clear that experts are only instructed where the court deems that such opinion evidence is “necessary” to resolve the proceedings justly (section 13(6): Re H-L [2013] EWCA Civ 655). Judicial scrutiny is brought to bear on all stages of the proposed instruction (per section 13(7)) – thus, the letter of instruction, the timeframe for reporting, the fee-scale of the expert is all made known to the judge at the point at which authorisation is sought. The same rigour is not brought to the instruction of the expert witness in the crown court (see Part 19 of the Criminal Procedure Rules 2015), and this was starkly apparent in the instruction of the experts on whom the mother relied in this case.

The point made at the end of the paragraph above is particularly pertinent to the instruction of medical experts from abroad, in this case two from the USA. As far as I am aware, such instructions are now relatively uncommon in family proceedings in England and Wales. Insofar as they occur at all, the instructions should, in my view, correspond with the guidance contained within the 2011 report of the Experts Committee of the Family Justice Council: ‘Guidelines for the Instruction of medical experts from Overseas’; with Dr. Teas out of forensic focus, I consider how those guidelines apply, or more accurately have not been applied, to the instruction of Dr. Hyman in this case:

In any case of this kind, the court will wish to consider opinions which are well-researched, supported by reference to the clinical notes, and referable to recognised current clinical practice and guard against the opinion of an "over-dogmatic expert, the expert whose reputation or amour proper is at stake, or the expert who has developed a scientific prejudice" (Re U (serious Injury: Standard of Proof); Re B [2004] EWCA Civ 567 [2004] 2 FLR 263). In this case, Dr. Hyman frankly spoke of positioning himself against what he described as the ‘Child Abuse Community’ – those experts who, he perceived, too readily identified child abuse; I sensed that his overall approach to this case reflected this position.

I would like to make two further short points about the expert evidence:

The factual matrix was investigated during the fact-finding hearing in July 2013, and rehearsed in my judgment delivered on 19 July 2013. There has been no re-examination of those facts, and those which are relevant to this re-hearing are as follows (though note my additional observation at (x) below, not included in the 2013 judgment):

The following severe injuries were noted (this much remains agreed in 2016):

I took the view that the vertebrae injuries and the skull fracture were likely to have been caused in a single incident (see [47] / [50] / [112] of the 2013 judgment). I found (in accordance with the schedule of expert agreement) that the force required to cause these injuries would be "substantial and considerable" ([51] ibid.).

I turn now to the three areas of the evidence (see [8] above) under review.

Earlier evidence: In my 2013 fact-finding judgment, I described the medical finding of osteopaenia on the skeletal X-ray of 31 October 2012, and reached the conclusion that its cause was most likely to have been the period of immobility and/or prescription of steroids in the immediately preceding 8 days. I said this:

“I note that the bone mineral density overall appeared somewhat reduced with "relatively poor bone mineralisation" on the scan of 31 October 2012, but I accept the view of Dr. Cartlidge and Dr. Fairhurst that this is likely to be attributable to his relative immobility in the immediate and critical post-event / pre-scan period.” ([63] [2013] EWHC 4859 (Fam)).

I discussed the vertebrae fractures in the July 2013 judgment (see [46] of the 2013 judgment); given my finding as to the cause of the osteopaenia above, I was of course proceeding on the basis that the vertebrae were of normal density at the time that AD sustained the spinal injuries. I recorded then that Drs Anslow, Cartlidge, Fairhurst and Richards had all agreed that the infliction of the injuries:

“… would have involved considerable force, and would have caused AD significant pain. The fractures probably occurred as a result of excessive flexion of the spine. When asked about this at the hearing, Dr. Fairhurst expressed the view that these injuries were "a different 'kettle of fish' from the skull fracture; they occur very rarely" (emphasis in the oral evidence) and that "a major event results in these fractures". She postulated the circumstances in which an injury of this type might be suffered would be where a child was in a road traffic accident having being restrained in the vehicle only by a lap belt, adding that "we know that they occur where children fall from a great height onto their feet". Whatever the event would be it would not be an event that "a parent would forget about"” ([46] [2013] EWHC 4859 (Fam)).

Dr. Fairhurst had earlier opined (2013) that the fracture to the vertebrae "can occur when a child is slammed down on a hard surface, or slammed against a wall", and further that "a fall would have been highly unlikely to cause the skull fractures but would not have caused the vertebral fractures" [53]. I indeed found that “It is likely that the vertebral fractures were caused with AD being slammed onto a surface, or thrown in such a manner that his body jack-knifed” [111].

I now turn to the evidence relevant to the vertebrae in 2016, and address the evidence by answering a series of relevant questions.

Is the presence of osteopaenia in the spine confirmed (2016)? The short answer is yes. No one takes issue that osteopaenia was indeed observed to affect the spine on the X-rays taken on 31 October 2012. Dr. Watt advised that the fact that osteopaenia was observable in the spine on the X-ray suggests that the bones were at least 20% demineralised (i.e. it has to be quite marked before it is noted radiologically), although he described the condition as “mild” in his report and in the experts meeting. Dr. Hyman by contrast described the osteopaenia as “significant”, and one of the “worst cases” that he had seen. The evidence of Dr. Fairhurst, which I accept (supported as it is by medical research), is that osteopaenia is more likely to affect, and therefore be identifiable in, the spine before other bones of the skeleton.

What has caused the osteopaenia? This aspect of the case has caused something of a conundrum. Dr. Cartlidge observed in the first experts’ meeting that this was “one aspect of this case which I find unusual”. He later expressed himself “uneased” about it, with a “lack of understanding” of why the vertebrae looked low in density. Prof Nussey indicated that he could not rationalise the presentation: it is “not easily explained”.

Through the written and oral evidence, a number of possibilities emerged as explanations for the osteopaenia in the spine:

It can be shortly stated that there were no obvious evidential indicators to point in favour of any one of the seven conditions adumbrated above; there have been no further fractures and no bone abnormality detected on subsequent X-ray. I deal briefly with each possible cause.

(i) Osteogenesis imperfecta (OI): There is no real evidence that AD suffered this condition; there has been no repeat of the fractures, which one would expect if there was an enduring condition of this kind; there was no evidence of the blue sclera of the eyes (a common finding in cases of OI); there is no family history of OI or other radiographic features to suggest an abnormal bone disease process or bone dysplasia (Dr. Watt).

(ii) Transient brittle bone disease: Dr. Hyman was alone among the experts in espousing the notion that AD had or may have suffered a form of transient brittle bone disorder or fragility. He explained it thus:

“transient bone fragility exists during the peripubertal period, it exists with prematurity, it exists with immobilisation, it exists with rickets, it exists with scurvy, it exists with children who are on steroids, it exists with children who are, you know, in sick bed. So transient bone fragility is well accepted in every place except the Society of Paediatric Radiology, at least in our country, the American Academy of Paediatrics where they made these statements”.

This potential cause was dismissed by all other experts who expressed a view about it.

(iii) Vitamin D deficiency / rickets: The thrust of the medical evidence is that AD was unlikely to be vitamin D deficient as at 23 October; there is certainly no evidence that he was. Prof Nussey was among those who most clearly expounded the view that it was unlikely that AD was vitamin D deficient as at 23 October 2012, and added, significantly, that there was no evidence of rickets in AD on any recognised basis – either clinical, radiological, biochemical, and/or histological. This is important, given the view he expressed to the first meeting of experts in August 2016 that “you don’t get fractures unless there is radiological rickets”; this was a re-statement of the evidence he had given in 2013: the "tendency to fracture is not measured in frank rickets, let alone subtle" (see [61] [2013] EWHC 4859 (Fam)). Dr Cartlidge agreed, on the balance of probability, as did Dr Cohen. That all said, Prof Nussey, both in the experts meeting and in his oral evidence to me, deferred to the two histopathologists in this respect. As to which, Dr Cohen confirmed that there was “no indication” that this child was vitamin D deficient at the date of his admission to hospital, although opined that the woven bone indicated previous vitamin D deficiency; Prof Freemont confirmed no evidence of vitamin D deficiency on histological examination. Interestingly Dr Fairhurst suggested that if AD were vitamin D deficient, to the extent that he had rickets, this would have the effect of softening the skull and potentially making it actually less likely to fracture than a skull unaffected by that condition; in this regard she cited the work of Prof Nick Bishop, the UK's only Professor of Paediatric Bone Disease, in Sheffield.

There is no dispute that AD was likely to have been vitamin D deficient in utero; Dr Hyman agreed, and opined that AD was suffering healing rickets of infancy.

(iv) Metabolic bone disease: Professor Nussey concluded that there was no evident endocrine disorder relevant to the causation of the injuries and no demonstrable metabolic bone disease. Dr. Fairhurst could not identify a likely cause other than immobility (see below) and in doing so, rejected as unlikely a generalised bone disease; she felt that had such existed it would have been apparent (as an evolving condition from 2012) in 2015 when in February, further X-rays and CT scans were taken of AD’s knee and distal left femoral metaphysis, and in June an X-ray of his left ankle; it was not.

(v) Clotting / bleeding disorders: These conditions had been excluded in the evidence filed for the previous (2013) hearing; per Professor Hann and Dr Cartlidge. I remind myself of Prof Hann’s evidence that the tests performed in this case represent the most comprehensive set that he had seen in reporting upon over 200 suspected NAI cases.

(vi) Immobility / steroid prescription: It was my view, on the evidence in 2013, that this was the likely cause of the osteopaenia. Dr. Watt (from whom, of course, I did not hear in 2013) was of the view that osteopaenia would only occur after a period of 2-3 weeks’ immobilisation; while it would be “possible” for it to occur after 7 to 8 days, it would not be “probable”; he acknowledged that it does occur in children admitted to intensive care for a short period of time, but this loss of bone density is generally minor and non-persistent. Dr Cartlidge similarly considered at this hearing that it was unlikely that immobility and/or steroid treatment could have caused the osteopaenia (he explained that if the demineralisation had been due to immobilisation, then this would normalise once AD was up and about again, which would in fact be consistent with the normal X-ray findings in 2015). Prof Nussey shared Dr Cartlidge’s view that immobility and/or steroid treatment were unlikely to have caused the osteopaenia; he felt that the studies of patients experiencing prolonged bed rest and/or loss of gravity described in the medical literature were not of any great value in understanding the cause of osteopaenia in this case, specifically in relation to a child of this age.

Of the experts giving evidence at this hearing (2016), it was Dr. Fairhurst who advocated this cause of the osteopaenia most strongly, as she had done previously. She considered that immobility and/or steroid prescription were a reasonable explanation for the osteopaenia, in the absence of any other identified cause; she was of the view that these factors had certainly contributed to it even if they did not provide the entire explanation. Dr. Fairhurst opined that given the “present medical knowledge, it is going to be extremely difficult if not impossible to determine what contribution the immobility played in the osteopaenia” (evidence in chief). Dr. Fairhurst commented that in a growing infant, a period of immobility can cause “rapid” demineralisation, because the body is not accreting minerals as it should in order to grow, and the immobility affects the absorption of minerals. She could offer no explanation for the osteopaenia other than the immobilisation; she felt that whatever else was found, the immobility would have made a “not insignificant” contribution over eight days to the finding of osteopaenia.

Natural variant: Dr Cartlidge considered that the presentation of osteopaenia in the spine as seen on the X-ray was at the “extreme end”, or variant, of a normal finding. Prof Nussey observed too that there is a wide variation in the presentation of a “normal” spine, of which this may be an example.

Unknown cause: Dr. Watt felt that there was an undiagnosed medical condition causing the osteopaenia; whatever it was, it was not an ongoing or persistent medical condition as it appears to have resolved (per X-rays of AD in February and June 2015 show normal long bones). Dr. Fairhurst expressed the view that there is much doubt about what could have caused the osteopaenia if it was not attributable to immobility. She said: “with our current medical knowledge, I have not identified anything in my opinion or in the opinions of the experts which suggests that there was any condition which could have caused the osteopaenia”.

The fractures: Dr Cartlidge told me that he comes across reduced bone density in clinical practice, but not vertebral fractures; these were a “rare finding”. If the bones in the spine were hypodense, it would not require the excessive flexion (to which I referred in the 2013 judgment: see [46] of [2013] EWHC 4859 (Fam)) to cause the fractures.

Is there evidence of osteopaenia or other bone weakness/unusual thinness in the skull? In my earlier judgment I recorded a degree of expert disagreement about this issue, noting:

“Dr. Anslow, Dr. Fairhurst and Mr Richards all agreed that the skull fracture would not be affected by any bone fragility … Professor Nussey and Dr Cartlidge opined that if there was generalised bone fragility, this would also affect the skull” [58]

Looking at this interesting question three years later, there remains medical controversy and division.

I take, first, the evidence of the experts as to the likelihood of osteopaenia in other parts of the skeleton. Dr. Watt, who provided a report for the criminal defence, and gave oral evidence before me, identified osteopaenia in the vertebrae and in some long bones. Dr. Fairhurst did not make this finding, in that she did not interpret the X-ray appearances in the same way so as to identify osteopaenia in the long bones; this may to some extent be referable to the fact that “radiographs are an insensitive modality for the assessment of bone density in children” (per Dr. Watt).

Dr. Cartlidge confirmed and expanded the view which he had given in 2013 (a view which appears to have been supported by Prof Nussey, Dr. Watt and by Dr. Cohen) that given the finding of osteopaenia in the spine, it is reasonable and logical to extrapolate the existence of some osteopaenia or other weakness in other parts of the skeleton, which would make them therefore more vulnerable to injury; Dr. Watt observed that “there is no evidence to suggest that the skull bones would be spared from any such process if found to be present”. None of the experts could quantify the extent of the bony vulnerability. Prof Freemont observed in the experts meeting that if there is metabolic bone disease, it should affect all of the bones but maybe not equally. Dr Fairhurst (supported in this regard by Dr. Cohen and Prof Freemont) advised the court that the skull bone forms in a different way from the long bones – the skull bones form from fibrous tissue (see Y449); put another way, “the ossification of the skull is intramembranous ossification (which) is different from the endochondral ossification that goes on in the other long bones like the wrist.” (per Dr. Cohen). This to some extent explains their reasoning for resisting any assumed extension of a condition.

Resolution of this issue has been informed by the closer study of the skull bones themselves for this 2016 hearing. Prof Freemont and Dr Cohen agreed, significantly in my judgment, that there is no histological evidence of thinning of the skull from the normal range; Dr Watt agreed that there is quite a variation in the thickness of AD’s skull but did not regard it as other than normal variation. He compared AD’s skull with the skulls of children of the same age, and he could not see “any convincing difference” between them; he added: “there’s nothing specifically that I could see that made me think the skull was abnormally thin”. He added that “some force” would be required to fracture the skull; it would have been a “significant impact event”.

As to its state of healthiness, it was expressly agreed in the first experts meeting between the histopathologists that there are no cellular abnormalities (‘no cellular changes’) in the skull cells which indicate current vitamin D deficiency (see Prof Freemont); they were of the view that the small areas of woven bone can be normal, or due to recovering vitamin D deficiency. In any event, the small pockets or islands of woven bone would not, in their opinion, have led to a reduction in bone strength. These are all important points of agreement. Dr. Watt confirmed that there was no pathology in the skull, and no evidence of vitamin D deficiency or rickets at the time of admission to hospital, a view shared by Dr. Cartlidge; Dr Cartlidge, Dr Cohen, Prof Freemont and Prof Nussey all further agreed that AD did not have a significantly weak skull. Dr. Hyman profoundly disagreed with this view; he was of the clear view (as he explained in his oral evidence) that woven bone is thinner than lamellar bone, and argued that the skull was therefore inherently weak.

Dr. Fairhurst discounted any reduced bone density, or hypodensity, in the skull for the following reasons:

Prof Freemont concluded that there was “no case to be made” that this child had craniotabes at the time of the injury (i.e. a condition that affects bone growth), and found no evidence of osteopaenia in the skull. In a subsequent written answer, he suggested that the presentation of AD’s skull fracture across a suture line indicated that the skull bone was not in fact weakened:

“The thinner the bone, the better able it is to fracture and the greater its ability to dissipate force over a wider area as the thinned bone cracks, breaks up and subsides. Thus the thinned bone, by having the ability to fracture “more” protects the sutures from the applied force to a greater extent than the harder, thicker bone. So were the bone to be thinner, the forces on the sutures would be less and they would not tear, either at all or until greater force still was applied to the bone… in an area of thinned bone, the sutures would be less likely to tear than in a bone of normal thickness”. (my emphasis).

Dr. Cartlidge confirmed (in cross-examination by Miss Cook QC) that this is a normally robust skull. Having listened to other experts discuss skull thickness and health during the first experts’ meeting, Dr Cartlidge summarised the position thus:

“… if both histology tests agree that the skull was about 2 mm thick and that is within normal range for all the tables that they use, which will include Caucasian babies in it, and that a thicker skull translates into a more robust skull, then regardless of racial origin, it is actually the skull thickness that is important and if the skull thickness is no different than normal, then it does not really matter what racial origin you are. That translates into normal skull robustness.”

Those present agreed with the accuracy and usefulness of this analysis.

Dr Hyman disagreed with these views, asserting in the second expert meeting that the skull was at the material time demineralised (not incidentally seen by Dr. Fairhurst) and that the identified woven bone is weaker, suggesting that it causes weakening to some undefined degree (he expressed it variously as between 1% and 20%). He said that “no one can guarantee that this child’s skull had normal bone strength”, his language exposing perhaps his more avowedly adversarial approach to cases of this kind, and his lack of familiarity with the civil standard of proof applicable in family law litigation here.

Skull and vertebrae fractures: caused at the same time or at different times? In my earlier judgment, I had concluded that it was more likely than not that the skull and vertebrae injuries were caused in a single incident. I said this:

“The medical opinion is to some extent divided on whether there was one or more than one incident. I discount any suggestion that AD sustained two separate assaults separated by any significant period of time. It seems possible to me that the injuries were caused in two separate assaults at more or less the same time; it is more probable in my judgment that these injuries – the head and vertebrae injuries – were all caused in the same incident” [112] [2013] EWHC 4859 (Fam)).

Mr. Richards advised me (per oral evidence at this hearing) that he recognised that the head injury and the spinal pathology may “not necessarily” be related; (see Y566):

“we have to look at the skull aspect in a separate way… we should divorce the vertebral fractures from the head injury because they have been seemingly linked together in other assessments and I don’t think that’s necessarily right.”

Dr Cartlidge by contrast who had indicated (per [47] of [2013] EWHC 4859 (Fam)) that "that the spinal fractures cannot be dated accurately" told me at this hearing that he always tries to seek a “unifying single diagnosis unless the evidence points against that”, and was tempted to default to a unifying diagnosis, “but that’s not because the alternative isn’t possible". Drs Fairhurst and Watt were unable to link the two sets of injuries on the basis of radiology alone, but Dr. Watt considered that the skull and vertebrae injuries probably occurred at the same time. Dr Hyman considered that the vertebral fractures “definitely did not, from a clinician’s standpoint, did not happen at the same time" as the skull fracture. It should be remembered that Dr Hyman’s speciality is not radiology and in expressing this view, particularly so definitively, he was straying somewhat outside his field of expertise.

Degree of force: vertebrae: From my earlier judgment ([2013] EWHC 4859 (Fam) at [58]) I collect the fact that “a very significant force” would be required to cause the vertebrae fractures to a healthy spine, but (per Dr. Anslow, Dr. Fairhurst and Mr Richards) there was agreement that less force would be needed to cause the vertebrae fractures, if there was any degree of bone fragility. This was broadly the position of the experts in this hearing. Dr Watt was of the view that the vertebral compression fractures are most commonly seen in this age group as the result of an underlying medical condition causing increased bone fragility resulting in fracturing with non-major trauma. Such conditions include osteogenesis imperfecta and rickets. Here arises the second major conundrum: there is no evidence of osteogenesis imperfecta and rickets (see above).

Degree of force: skull: There seems to be little doubt that this was on the combined views of the experts a “huge” fracture “that doesn’t just go through bone, that tears through fibrous tissue in a piece of bone…” (Prof Freemont), it was “severe” fracturing (Dr Cartlidge); its length and extent perhaps speaks for itself. Even Dr. Hyman – who championed the view that AD’s skull was inherently weak – nonetheless considered that “significant force” would have been required to cause the skull fracture.

That all said, I accept that if the evidence demonstrates that AD’s skull was weaker to any or any significant extent as at 23 October as a result of osteopaenia or other compromise of bone density, it may well be that a lesser degree of force would be required to cause the extensive injuries to AD’s skull and brain. As Mr Richards put it in the second experts meeting: “if the court determines the bone was weaker for metabolic reasons, then, yes, it’s common sense to say that less force may have a greater effect… if it’s (the skull) weaker than normal, it may affect the outcome.” He referred later to this ‘common sense’ approach as ‘theoretical’.

In this context, Dr Cartlidge propounded a hypothesis in his oral evidence which conjured up an accidental cause for the injury. He suggested that at 4.30pm, with the father briefly out of the room, AM could have picked her baby brother (AD) up, and dropped him (more or less 2’) head-first onto the metal leg of the TV base. This proposition had been ventilated first at the first experts meeting, having been flagged in Dr. Hyman’s report: he there had expressed the view that a normal skull could fracture at a free-fall distance of 2 feet, and that conceivably a fracture may occur with a fall of less distance if the skull was more fragile than usual. However, he later conceded that he had only “occasionally” (on about two or three occasions during his career), come across babies with a fractured skull from falls of this distance adding: “I can’t remember the extent of that fracture, whether it’s as severe as it is in this case, and of course we wouldn’t know whether it propagated through the fibrous suture because we won’t see that on an X-ray…”; he further drew on 19^th century research on cadavers, and on Dr. Van Ee’s experiments. Although he postulated in oral evidence that AD may have fallen against the long metal leg of the television stand, he had in fact earlier opined that the fracture was consistent with impact with a flat or near flat object (i.e. a floor), a view shared by Prof Freemont.

As a matter of fact, a point I return to below, I caused research of the documents (particularly the parties’ statements and police documents) filed for the 2013 proceedings to be done to search for any account of AM’s interest in AD, and specifically of her holding or picking up AD or attempting to do so. The fruits of this research were limited to short extracts from the police interviews, where the mother confirmed that she would not allow AM to pick up AD. She said that she had seen her try to do so, but that “she wouldn’t properly… be able to pick him up properly … she would struggle I think”. She later says that she would not be able to assess how AM would manage picking him up, and that “she’s too young for… you know, she’s too young to be picking up baby, you know”.

Prof Freemont challenged Dr. Cartlidge’s hypothesis, and the degree of force required to inflict the injuries observed in AD. While acknowledging in the experts’ meeting that he does not know enough about the precise forces that are needed to tear a suture, he nonetheless contributed this view to the meeting:

“… there is also a long tear through the fibrous tissue and that’s really very strong tissue indeed and it doesn’t have some of the weaknesses that bone has, of course, which relates to the fact that it’s mineralised and therefore, you know, it is more easily cracked, if you like, than it is to tear.”

He agreed with the formulation by Dr. Bloom at the meeting that the fact that the fracture propagated right through the fibrous tissue as well as the bone indicated a greater rather than a lesser fall, adding:

“I’m not sure of the significance of this, but the tear through the fibrous tissue is at right angles to the bone fracture… Which implies, I think, you know, for something to turn that far, I think in a certain type of impact and quite a significant impact as well”

During this hearing, Professor Freemont was specifically asked to consider Dr. Cartlidge’s hypothesis by way of a written question. He responded:

“… in order to get diastatic skull fractures (ones that involve the sutures as well as the bone) in a relatively mature infant skull, such as that in a 9-10 month old child, the forces required are greater than those in the scenario described in the question”.

He added, by way of explanation:

“I have not seen a large, full thickness separated fracture with propagation completely across a bone (i.e. several inches long) and into a suture in a fall of less than 5 feet on to a hard surface of any type, whether a tube with an oblong cross section such as the leg of this table or a hard surface such as was illustrated in the images provided to me, in a child of this age… Following a fall onto the head, the bone is more likely to be fractured than the sutures are to be torn, presumably because the bone is better able to dissipate force than the suture, or perhaps, put better, it requires more force to initiate tearing of a suture than to initiate fracturing of an adjacent bone”.

And later in the same written response:

“…much more force than that generated by a fall from 2 feet, even on to the table leg, would have been needed to cause this fracture and also to tear the dense, gristly, fibrous tissue of the suture for a length of 2.5-3cms (1 inch plus)” (emphasis by underlining in the original).

Dr Cohen agreed (so far as it was in her field of expertise) that a fracture passing through the fibrous tissue indicated a greater degree of force than if it had only been a fracture through the bone. Dr Hyman himself expressed the view that there was “a significant impact to cause all that damage".

The histopathological experts agreed (this is recorded as a specific point of agreement No.5) that the propagation of the fracture across the suture line at right angles, in addition to the bone fractures, is important and suggests a high level of force. Mr. Richards speculated that the tear through the suture may be attributable to the increased intra-cerebral pressure and/or surgery, but this would not explain the continuation of the fracture beyond the suture.

The histopathologists agree that it is not possible to identify the precise timing of the injury by reference to the changes seen in the skull; Prof Freemont and Dr Cohen agree that the injury almost certainly occurred between 8 and 24 hours prior to the craniectomy. So it is to the evidence of clinical presentation, and to some extent of external skull swelling (see below), to which I turn to assist with this important question.

Mr. Richards remained firmly of the view, given in 2013, that given the whole picture of injury it was unlikely that there could or would have been a lucid interval. His earlier evidence is summarised in [65] of [2013] EWHC 4859 (Fam), note specifically [65(i)], wherein he said that it would have happened “after the child was last seen to be behaving normally” adding that after the incident causing the injuries there would not have been any “normal behaviour”. He expressed himself at the second experts’ meeting thus:

“the whole picture of a child who was not causing the parents concern who suddenly became acutely encephalopathic with features associated clinically with skull fractures… Radiologically a skull fracture, on direct surgical inspection the dural laceration and an acute brain that was swollen, and an intracerebral haematoma, acute subdural blood, it all points to an acute traumatic event close to the time of the clinical change…shortly before the emergency services were summoned” (emphasis by underling added).

Mr. Richards was challenged by Mr. Storey about this evidence; he was asked to consider the incidence of skull fractures in babies at birth, which are ‘subtle’ or ‘silent’ in the sense that they can remain undetected and not notably symptomatic. Mr. Richards agreed that such fractures (and subdural bleeds) do occur but that those fractures at birth to which he referred are generally “very short linear” fractures, not of the scale/size of the fracture on AD’s parietal bone. He accepted that there were some indicators of general unwellness in AD the period after 4.30pm (crying for 5-6 minutes, vomit after the feed, possibly a dose of ibuprofen given, see [17(x)] above) but felt that these were not particular indicators of anything sinister; he advised that it was “inconceivable” that AD would have suffered his near-catastrophic injury at 4.30 yet gone on to be assessed by the grandfather at 5.30pm as essentially normal, and then go on to eat some supper and have a bath. Specifically he opined that the severity of the injuries particularly when viewed in combination contra-indicated a lucid interval; he said that he found it “difficult to accept” that the baby would have been conscious after sustaining these injuries, and equally inconceivable that, if conscious, he would have been able to feed; later in his evidence, in answer to questions from Miss Cook, he said that “it would be very, very, very unusual if [the baby] was not unconscious” immediately after suffering injuries of the kind suffered by AD, “and if conscious, [he] would not be eating and bathing normally”. He commented that the “whole neurological picture” was “incompatible” with a lucid interval; he had expected any conscious infant to have been “inconsolable until his level of consciousness took the pain away”. He attached some importance to the presence of subarachnoid bleeding in the gyri and the sylvian fissure; subarachnoid blood is “very irritative and painful… it is one of the most painful things that a patient can have”; it is, he said, “exquisitely painful”, and would have been likely therefore to have manifested in the child’s behaviours, and general state of extreme distress.

Dr. Cartlidge expressed himself in similar terms. He indicated that he regarded as “very likely” that the collapse would have happened “straight away”, and told me that he held this opinion “with a high degree of confidence”. He specifically rejected the notion that the descriptions of AD’s presentation during the late afternoon of 23 October could be a description of a child who is going through a lucid interval:

“… a lucid period is not in terms of normal behaviour… I find it inconceivable, to be frank, that there could be a delayed intracranial injury and I therefore think that the intracranial injuries are very likely to have been sustained at the same time as the skull fracture and the totality of the symptoms would have been both caused by the intracranial injuries, i.e. the child collapsing…”

And later:

“I can’t see that this could be clinically silent or more or less silent between 4:30 and then a sudden dramatic collapse shortly before the ambulance was called… I…cannot understand how he could have sustained these skull fractures and the intracranial injuries at 4:30 and for the clinical presentation to be as it was.”

He later excluded a lucid period: “I don’t think it’s possible”. He summarised the position at the first experts meeting by identifying that the histopathology, the X-ray, and the swelling can “only give you quite a wide range of possible timings”, but that the timing can be narrowed by:

“an assessment of the history and if the court eventually decides that [MB] and [FB] are describing a child who is behaving more or less normally between 4:30 and 9 o’clock in the evening, I find it difficult for any paediatrician … or a neurosurgeon to believe that these injuries could have been sustained at that time”.

Although Dr. Cartlidge had hypothesised a possible scenario for the trauma occurring accidentally at 4.30pm (see above [17(v)]), he was quick to point out that this does not overcome the difficulty of explaining the significant gap in time between such an incident and the onset of the overwhelming and overt symptoms. He rejected Mr. Storey’s suggestion of a brief loss of consciousness followed by crying; he expressed the view that it would be unusual for a child to come round from a period of unconsciousness. He said that he did not recognise Dr. Hyman’s theory (see below) as “plausible”, and found it difficult to “follow the flow of the argument”.

Dr. Hyman expressed a contrary view. He was of the opinion that AD had suffered an older fracture (a view propounded initially by Dr. Teas see [3(iii)] above), which had then become depressed (albeit “clinically silent”) in the fall at the 4.30pm incident; this fall, and the possible incursion of a piece of bone into the dura, had led to a growing cerebral oedema which had only become symptomatic at 9.30pm. Mr. Storey directed my attention to the neatest summary of Dr. Hyman’s view in the experts meeting:

“… In situations of lucid intervals, and perhaps in this child’s case from what the family history is, the three members of the family’s history, is that it took some time for the devastating events to occur, that the initial fall in this room could have caused the fracture, extended the fracture et cetera. It could have taken some time for the volume, the intracranial volume, to have an increase of the smaller amounts of subdural and subarachnoid blood for whenever the time the parenchymal haematoma was caused, perhaps due to the perforation of the bone fragment, that there was this interval and certainly the time when he went into extremis could have been during his sleep time so you have a bit of a shorter interval and then you have a child that comes in with a GCS [Glasgow Coma Scale] of 3.” Y531

The reference to “extended the fracture” (underlined above added) refers back to an earlier answer which reveals Dr Hyman’s view that there had been an existing skull fracture which was ‘extended’ in a 4.30pm fall.

Dr. Hyman explained that he believed that the fracture had grown in length as a result of the cerebral swelling (a view not shared by any other expert), and that it had not in fact crossed the coronal suture (again, a view not shared by any other expert). He explained that the clinically silent fracture had accounted for this lucid interval between the major injury and the onset of the overt symptoms in this way; Dr. Hyman was of the view that there were sufficient indicators of upset in AD to suggest that all was not well with him. His thesis had initially been put rather tentatively in the experts meeting: “I don’t think that one can absolutely say that the event has to happen right before the paramedics came”, but it gained in momentum as the meeting progressed, developing to a confident assertion by the end that there was indeed a lucid interval between 4:30pm and 8.00pm. In his oral evidence he described it as a “neurological cascade” befalling the child in the period after the 4.30pm incident. He placed reliance on the paper by Denton and Mileusnic (“Delayed sudden death in an infant following an accidental fall” (2003)) but this paper (as is acknowledged by counsel) is clear in its acknowledgment that what was described was a “rare phenomenon”.

Dr. Hyman commented in his written report that after a significant intracranial injury it “is probably true in the majority of cases” that profound and abnormal symptoms will present immediately; he nonetheless relies on “enough documented exceptions to invalidate the concept that symptoms will always occur immediately". I could not help but feel that, in some respects, Dr Hyman enjoyed playing devil’s advocate.

Evidence had been given about skull swelling in July 2013. I recorded the observation of the ambulance technician that the swelling had “started” to grow “while in resus”, and Mr. Richards’ assessment that this was a significant observation. I continued [67]:

“Although scalp swelling: "can occur within minutes but may take several hours…" Mr. Richards described this as a "rapidly evolving situation", and (in answer to questions from Judge) indicated that "this [reference to 'started to grow'] would put the timing [of the incident] nearer to the time of the 999 call than further". The observation at (set out at paragraph 66 above) is consistent also with the comments of the anaesthetist who noted at 10.48p.m. that the "scalp swelling was increasing in size".

There is no expressed doubt that the soft tissue swelling was a result of trauma. In the discussion of experts in the first meeting Dr. Cartlidge opined that the onset of swelling is variable in rapidity and extent, and was reluctant to distinguish between an incident at 4:30pm and an incident shortly before 9:30pm simply by reference to the fact that the swelling was seen to be growing (as discussed above). He said that “it’s likely to be an onset within a few hours, possibly within a few minutes and reaches a maximum after about 24 to 48 hours”. However, he went on to tell me that if I considered the ambulance entry (viz: “slight swelling noticed on ® temporal area which started to grow while in resus”) to be ‘robust’ evidence then that would be more consistent with a very recent incident immediately prior to the observation.

Mr. Richards expressed the view that swelling “can come on over a few hours or perhaps even longer in a few cases".

Professor Nussey observed that if the swelling at the site of the fracture is not only due to the breaking of the bone but also due to the impact on the skin (as Prof Freemont and Dr Cohen opined) "then it would seem to me that the impact on the skin must have been quite substantial".

I have considered the evidence adduced at this hearing with considerable care. I have thought it right to set out the medical debate extensively above to demonstrate the scope and range of the views expressed. The evidence has been technical, and has been meticulously and thoroughly picked over by counsel. The quality of the expert evidence has generally been of a high quality. This is a case in which, in the end, much turned on the extent to which I would accept the broadly contrary views of Dr. Hyman. I studied his evidence thoroughly, but considered that his contributions – both as to his views on the precise science as much as his overall approach – were less helpful than the contributions of his medical colleagues. I say so for the following reasons, in combination:

I turn therefore to the three areas under review.

That osteopaenia was identified in X-rays of AD’s spine is not in doubt; the extent of it is more contentious. Overall, I was impressed by the evidence of Dr. Watt who spoke with authority and balance on the issues; he described the osteopaenia as “mild”. Dr. Hyman considered it to be more significant than that, but he is not of course a radiologist, and had indeed deferred to a colleague on issues of radiology when preparing his evidence. I accept that Dr. Watt had observed evidence of osteopaenia in the long bones of the skeleton, through lucencies of the metaphyses on X-ray; these lucencies had been observed by Dr. Fairhurst but not interpreted by her in the same way, which she ascribes to an essentially subjective assessment of the bony structures using a blunt instrument. Insofar as there is evidence of osteopaenia elsewhere than the spine, I am prepared to accept Dr. Watt’s opinion, though consider that such evidence as there was of osteopaenia in the long bones, it was of a very modest nature.

The evidence which has been filed and called for this hearing has caused me to review my conclusion as to the cause of the osteopaenia. Challenge is now raised to my earlier finding that the bony changes observed in the X-ray on 31 October would have been caused by a combination of immobility and steroid prescription in the immediately preceding 8-day period; indeed, this would not now accord with the clinical experience of any of the experts called. Of the limited medical research papers available to me, none (with the possible exception of Giangregorio and Blimkie (2002)) support such a short degree of immobility as a likely cause of this condition. The preponderance of the evidence (certainly from Dr. Watt, Dr. Cartlidge and Professor Nussey) is that a period of no less than 2-3 weeks of immobility would need to elapse before osteopaenic changes would be observed; Dr. Watt (as I indicated above) regards it as only ‘possible’ but certainly not ‘probable’ that an 8-day period may cause a degree of mild osteopaenia. Dr. Fairhurst makes a strong case that immobility remains a distinctly possible, if not probable, cause, having regard (in combination) to the fast growing bones at AD’s young age, causing the loss of accretion and absorption; her position appears to be supported to some extent by the paper of Giangregorio and Blimkie (2002) (“Skeletal Adaptations to Alterations in Weight-Bearing Activity: a comparison of models of disuse osteoporosis”) which addresses the demineralisation of bones in rats, indicating that (a) a short period of immobilisation can result in significant mineral loss in the growing skeleton and, (b) vertebral mineral loss, at least in their experiment, occurred at nearly twice the rate it did in the tibia.

As a child who was not known to be particularly mobile, it is possible of course that the osteopaenic changes (cause by a degree of immobility) had commenced in the spine before 24 October. I draw from my earlier factual findings (see [14] on [2013] EWHC 4859 (Fam)) that the maternal grandmother had referred to AD as not making “any effort to do things; he was lethargic”; she said that if anyone tried to encourage him to move he would cry. Later she said that he "didn't enjoy making the effort" to mobilise, and that it would be unusual for him to mobilise without encouragement (see [71(i)] of the 2013 judgment). All of this makes it possible that the mineralisation deficits were the result of – or as Dr. Fairhurst made clear materially contributed to – a degree of immobility over a period of time which preceded 23 October. This was suggested in the submissions, not having been actively explored in the evidence, and I cannot (and do not) attribute weight to it as a theory.

Taking the evidence as a whole, and on the basis of the evidence summarised at [76] above, I am no longer satisfied AD’s immobility in the hospital ward in the period from 23 October to 31 October is the probable cause of the osteopaenia.

I am equally not satisfied that there is any, or any adequate or material, evidence to support the existence of osteogenesis imperfecta, transient brittle bone disease, current vitamin D deficiency or rickets; nor is there sufficient evidence to support a metabolic bone disease or clotting disorder. None of these can be said to be probable causes. We do know that in 2015 the long bones were seen, on two occasions, to be normal which contra-indicates any pervasive disease or disorder. I am attracted by the view that the low level of mineralisation seen on 31 October 2012 was an extreme variant of normal, as advocated by Dr. Cartlidge and Professor Nussey, but cannot regard this, without more, as the probable explanation.

In the end, and for the reasons outlined, I am left to conclude that there is no identifiable probable cause found for the osteopaenia in the spine, and I replace my earlier finding with the finding that the osteopaenia was of ‘unknown cause’.

Without an identified cause for the osteopaenia, I must look with care at the evidence which may suggest low mineralisation in other bones of the skeleton, including – most significantly – the skull. As earlier indicated, there remains controversy over the superficially attractive logic, or medical reasoning, that bone weakness in the spine necessarily involves bone weakness in all other bones including the skull; however not all of the experts agree with this proposition (notably Prof Freemont and Dr. Fairhurst). Against a background of a finding that there is some very modest bone weakness in the long bones (see [75] above), I must consider the evidence in relation to the skull with particular care. For a number of reasons, six in all, I have reached the conclusion that the bone weakness identified in the spine has not affected, or materially affected, the skull. I say so for the following reasons in combination:

It follows that there is no proper basis for adjusting my 2013 conclusion about the degree of force required to generate such a significant fracture. Indeed, on any view, including that of Dr. Hyman, it was caused by an extremely significant impact.

I turn then to consider Dr. Cartlidge’s hypothesis for the 4.30pm accident. First, in view of my finding above, I am satisfied that AD’s skull was of normal or essentially normal strength, and it follows that (even on Dr. Cartlidge’s view) the fall would really need to have been of a height of 2’ or more in order to inflict the observed damage; even then, it would be an unusually but not unprecedentedly low, height from which a child would need to fall to be likely to suffer injuries of this kind. Given the diastatic nature of the skull fracture, I prefer Prof Freemont’s view that the forces required are greater than those in the hypothesis described above (see [58] above), and that a fall of 5’ or more would be more likely. Overall, I am further unpersuaded by Dr. Cartlidge’s hypothesis for the following reasons:

The evidence was clear in 2013 that there would have been little if any effluxion of time between the acute traumatic event causing the injuries and the acute clinical change, i.e. the collapse. Frankly, with the exception of the evidence of Dr. Hyman to which I turn in a moment, the evidence was as strong if not stronger on this point at this hearing.

This was not a case in which the range and seriousness of the injuries would be likely to remain clinically silent; these were not fractures and subdural bleeds of a type or extent which are sometimes known to have occurred at birth – in so concluding, I have in mind Mr. Richards’ answers to Mr. Storey in cross-examination (see [63] above), and the Whitby, Griffiths, Rutter et al paper of 2003, the Looney, Smith et al paper of 2007, and the Rooks et al paper of 2008 all which was referred to during the evidence and in submissions. Overall I conclude that AD was relatively normal after the 4.30pm incident. He may have been niggly; the mother may have thought that he was teething, and given him ibuprofen, but I accept Mr. Richards’ evidence, supported by Dr. Cartlidge, that these were not signs of any sinister and profound evolving brain and subdural damage brewing under a seriously injured skull. I further accept the evidence of Mr. Richards and Dr. Cartlidge that AD would have been likely to have become almost immediately unconscious following the trauma – and insofar as he was conscious at all (which would be “very, very, very unusual”: see [63]) it would have been for a short period of acute and exquisite pain rendering him “inconsolable until his level of consciousness took the pain away” (ibid.).

Dr. Hyman’s opinion of the neurological cascade was predicated upon the existence of an older fracture – a finding which no party sought to support. He was isolated in his views as to the length of the fracture and its direction (through the suture) which perhaps reflected his lack of expertise in the interpretation of the radiology, and/or a lack of objectivity demonstrated by the circumstances in which he accepted his instruction. In fairness to Dr. Hyman, he had given a rather more realistic appraisal of the likelihood of a lucid interval in his written report (see [68]) and at an earlier stage of the second experts’ meeting (“I don’t think that one can absolutely say that the event has to happen right before the paramedics came”) than his resting point revealed at its conclusion, and in his oral evidence to me. He placed reliance on the Denton paper (see above), but as earlier pointed out, this described a rare phenomenon – an outlier.

Mr. Storey realistically acknowledged that this aspect of the case was “the most difficult one that we face on this re-opening application” (closing submissions). I am conscious that it is not for the parents to prove that there was a lucid interval, Mr Storey nonetheless suggested that I may prefer the case advanced by the parents by following one of two essential routes: either (a) I prefer the evidence of ‘neurological cascade’ of Dr Hyman over the evidence of Mr Richards and Dr Cartlidge, or (b) I conclude that there are sufficiently unusual features of this particular case to treat it as “an outlier”. I have already explained (see [73] above) why I regarded Dr Hyman as less satisfactory and convincing as an expert witness than the English counterparts who have given evidence before me in this hearing; that aside, I am satisfied that both Mr. Richards and Dr. Cartlidge have objectively re-evaluated the conclusions which they formed in 2013, from positions of great authority and current clinical experience, and have arrived at the same location. I venture to suggest that they did so with greater confidence. I am unable to say that there are any, or any material, unusual features of AD’s presentation to justify the conclusion that this case should be treated as an exception to the general, accepted, and expected medical norm.

I therefore conclude that there is no basis on which to alter my finding that the significant traumatic event in which AD suffered his life-threatening injuries occurred very shortly before the paramedics were called.

The opinions expressed in the experts' meetings about the skull swelling gave cause to believe that there may be a wider timeframe within which the material swelling should be assessed in this case. The evidence is that the swelling could develop over a number of hours. The key evidence is that of Mr Richards and Dr Cartlidge who expressed the view that if the recording of the paramedic was “robust”, this would be more consistent with a very recent incident. I do regard that recording as reliable, or, to use Dr Cartlidge’s word, “robust". I say so because it entirely accords with the absence of swelling at 5:30pm, and the continued evolution of the boggy swelling in hospital after AD’s admission.

Having reviewed the evidence with care, I see no basis for diverging from the essential finding I made in 2013 that the mother inflicted serious head injuries upon her son.

Specifically, my finding about the 4.30pm incident at [94]-[97] ([2013] EWHC 4859 (Fam)) must stand, and there is also no basis for altering the factual conclusions reached at [98] (ibid.) in which I describe how “AD was observed by family members to behave more or less normally” after the fall in the bedroom. Crucially, I uphold my earlier finding that the medical evidence remains “clear” that the description of AD’s behaviour between 4.30pm and 9.30pm:-

“… is wholly inconsistent with a child who has, in the hour or two immediately beforehand (and in the absence of evidence of a deterioration thereafter), just sustained serious and life-threatening injuries”.

Having found earlier in this judgment that there is no proper basis for concluding that AD’s skull was thin, or materially weakened by disease or deficiency, I maintain my finding that the forces required to cause the extensive fracture must be significant and in the absence of any account of an event of the severity to cause such catastrophic injury, further conclude that they must be non-accidental. I reject the theory that there has been a lucid interval, for the reasons set out above, and accordingly, my finding at [109] ([2013] EWHC 4859 (Fam)) that it was the mother who inflicted these serious injuries to her infant son, AD must also stand. The summary of factors which inform this conclusion remain intact, save that in place of “unambiguous” in [108](iii), I would now say “significantly preponderant” to reflect the fact that Dr. Hyman did not agree with the views of his medical colleagues. I remain of the view that I should not be drawn into speculating on precisely how AD’s head was injured.

Paragraph [111] of the 2013 judgment continued:

“It is likely that the vertebral fractures were caused with AD being slammed onto a surface, or thrown in such a manner that his body jack-knifed.”

In view of my inability now to identify the cause of the osteopaenia in the spine, I cannot say whether AD was affected by this condition, and if so to what extent, in the days, weeks and/or even months prior to 23 October. I cannot rule out that he may have had the condition for some time. Lesser forces would have been required to cause him to suffer thoracic vertebral fractures if his spine was hypodense in the way described by the radiologists. I am no longer satisfied that the vertebral and skull fractures were caused in the same incident (see [112] of the 2013 judgment), and in view of my current findings, I am not satisfied on the balance of probabilities that these thoracic fractures were non-accidental. I therefore withdraw that finding. Having done so, I have re-considered the evidence relevant to the skull fracture, and reflected on whether it affects my view of causation. I am satisfied that it does not.

Save, therefore, for the limited extent set out in [93] above, the essential outcome of this fact-finding hearing is much as it was before. AD can be reassured in the future that his situation, and specifically the causation of his injuries which have left him to some extent impaired, has been given the closest and most anxious scrutiny through an intense and repeated medical and legal process. He must, as the family must, live with the sad and highly regrettable truth of what happened to him.

That is my judgment.