| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Commentary |
Consultant Neuroradiologist, Department of Neuroradiology, Frenchay Hospital, Bristol BS16 1LE, UK
 |
Abstract |
|---|
|
Introduction |
|---|
The main role of the Expert Witness is to interpret and explain to the Court material which is within the Expert's area of expertise, but which is outwith the experience of the Judge or the jury. The Expert Witness is in the privileged position of being allowed to give opinion evidence to the Court, but this brings a responsibility to ensure that the opinion given is both reasonable and capable of withstanding logical analysis. An Expert Witness has an overriding duty to the Court that takes precedence over any obligation to those Instructing the Expert. The Expert should be independent, impartial and confine their opinions to their particular area of expertise. All sides in Court, will rightly question this evidence and the Expert must be able to justify their opinion. However, just as lawyers' opinions on the interpretation of points of law sometimes differ, doctors reading the same scientific papers may come to different conclusions; hence controversy!
|
What are the "typical" neuroimaging features of NAHI? |
|---|
A fairly common finding in cases of NAHI is of shallow subdural haematomas (SDH) at various, separate sites over the cerebral hemispheres and often in the posterior fossa. The subdural blood may be of different ages, although this is by no means always the case. The most common site for SDH following both accidental trauma and NAHI is over the cerebral convexities. SDH following all but severe accidental trauma, however, tends to be unifocal whereas SDH in NAHI tends to be multifocal. SDH at sites other than over the convexities, especially if in the posterior interhemispheric fissure or posterior fossa, is more likely to occur following NAHI than following accidental trauma [1]. Subdural blood can be seen following an impact injury from any cause, but it is usually related to the site of the impact and is often associated with a fracture. Subdural blood seen away from the point of impact is unusual in accidental trauma, unless severe.
Focal parenchymal lesions such as contusions, haematomas and shearing injuries may be seen following both accidental and non-accidental head trauma; when seen following accidental trauma there is usually a very clear history of a major traumatic event. Shearing injuries (diffuse axonal injury or the larger gliding contusions) were thought to be very common in NAHI. However, recent neuropathological and neuroimaging evidence suggests that axonal injury is uncommon even in infants who die following NAHI and that hypoxic–ischaemic changes are more common in these infants [2, 3].
Often there is evidence of reduced grey-white differentiation focally or more generalized on head scans following NAHI and, given the neuropathology, it would seem likely that these scan changes are related to hypoxic–ischaemic changes in the brain.
|
There is no absolute scientific evidence that shaking causes the injuries in "shaken baby syndrome" is there, doctor? |
|---|
There is no absolute scientific proof that shaking causes these injuries because it is not possible to perform the relevant scientific experiment, of shaking infants of different ages and sizes with different degrees of force and then performing sequential scans; and then studying them clinically and correlating this with scan appearances. In the absence of such data it is necessary to rely on various other sources of evidence including animal experiments, from the neuroimaging appearances of infants following witnessed accidental trauma and imaging evidence from cases of NAHI in which there has been a reliable confession.
Primate experiments have been performed in which the animals were subjected to pure translational and/or rotational forces without impact. The head injuries seen in the animals were very similar to those seen in NAHI [5]. Scans of infants and children who have sustained head injuries following accidental head trauma may show the same pattern of SDHs at different sites (as in NAHI), but usually only where the accidents involved severe forces such as following road traffic accidents or major falls. Most domestic accidents do not involve this degree of force and, given the number of domestic incidents that occur daily involving children bumping their heads, the fact that our departments are not inundated with scan requests for infants suggests that the majority of these falls are neurologically benign, as has been shown in the literature [6].
There is debate as to whether shaking alone is sufficient to cause all of the features of NAHI or whether some impact (even against a soft surface) is always necessary. Some of the animal work cited above suggests that impact is not required. There are biomechanical models that show that the amount of force generated during a pure shake is much less than following an impact, but there are also reports in the literature of fatal cases of NAHI where there has been no evidence of impact injury even on post-mortem. The focal brain injuries sometimes seen in NAHI may reflect associated impact or the result of very severe shaking.
Some perpetrators admit to shaking and many a loving parent must have been pretty close to it, having been up for hours at night with an inconsolable infant that they have done everything to try to settle. It is possible that only the minority of cases are due to wilful cruelty and some authors have advocated changing the way in which we classify abuse partly because of factors such as these [7].
|
The subdurals are very thin and not causing significant mass effect, so what is the cause of the presenting symptoms? |
|---|
This neuroimaging evidence is supported by the neuropathological literature. Geddes et al published two papers after studying a cohort of fatal cases of NAHI. One paper [2] was a review of 53 cases, 37 infants ranging in age from 20 days to 9 months and 16 children ages ranging from 13 months to 8 years. The authors demonstrated that most of the brain damage seen in these fatal cases was due to hypoxic vascular damage rather than traumatic axonal damage. Traumatic diffuse axonal injury was only seen in three cases.
A second paper by this group [3] was based upon 39 infants aged 9 months or under where the authors compared the neuropathology with that seen in 14 control patients. Traumatic diffuse axonal injury was only found in two of the NAHI patients, both of who had suffered severe head injury with multiple fractures.
If hypoxic–ischaemic injury is the major cause of death in fatal cases, it is logical to assume that it is an important factor in non-fatal cases, and that the degree of associated hypoxic–ischaemic change largely determines presenting symptoms in the short term as well as outcome in terms of mortality and morbidity. If this is correct then it is likely that infants who sustain a severe injury will lose consciousness at the time of the injury and some may not recover. At the other end of the spectrum, the hypothesis would suggest that there should be a group of infants who sustain an injury which may cause a minor change in behaviour insufficient to cause their carers to take them to a doctor and where the brain (and therefore the infant) recovers from the injury spontaneously.
In between these extremes lies a group of infants in whom there is a probably quite a marked change in behaviour following the injury. In less susceptible infants, the brain may be able to recover spontaneously from the injury whereas in more susceptible infants some threshold may be reached beyond which the brain cannot recover from the insult by itself. This latter group of infants would go on to deteriorate clinically and present with an encephalopathic illness. In this way, hypoxic–ischaemic change following NAHI may be in some way analogous to birth asphyxia where infants may develop hypoxic–ischaemic encephalopathy some hours after the insult.
It is the different degree and extent of hypoxic–ischaemic injury that determines whether and how these infants will present to medical attention and that the SDHs are just markers of the mechanism of injury. Sometimes the SDHs appear so insignificant on the scan of an extremely sick infant that they may be overlooked and their true significance not realised.
|
What else could have caused the SDH? |
|---|
Infants with coagulopathies may bleed spontaneously or after minimal trauma, but this again is surprisingly uncommon. The pattern of intracranial haemorrhage in infants with clotting disorders also tends to be different as intraparenchymal haemorrhage is much more common than SDH in these infants [11]. Congenital abnormalities including vascular malformations or metabolic conditions such as glutaric aciduria may predispose to SDH.
|
When did the bleeding occur? |
|---|
Different factors apply to the dating of blood on MR scans, but when there is a combination of CT and MR scans it is often possible to give a range of probable ages for the subdural collections present. This does not necessarily help with the timing of the injury, however, as the SDH is unlikely to be responsible for the presenting symptoms. Usually, the history gives a better assessment of the timing of injury than the radiology because an infant that has suffered a NAHI of sufficient severity to lead to admission to hospital is in my experience extremely unlikely to have behaved completely normally after that injury was inflicted.
|
What about re-bleeding into chronic SDH? |
|---|
|
What degree of force is required to produce these injuries? |
|---|
As we see similar appearances to NAHI following severe accidental trauma, it is sometimes suggested that the degree of force involved in NAHI must be similar to that occurring during road accidents. This is incorrect, as the mechanism of the two injuries is different: high velocity impact with acceleration/deceleration in accidents; lower velocity rotation and apnoea in NAHI.
|
The radiological investigation of NAHI |
|---|
Many of these cases are obvious, others are not and we won't find the evidence unless we look. Think of SDH in the same way as metaphyseal and rib fractures: we look for these fractures assiduously with skeletal surveys but, although they are markers of mechanisms of injury, they are of little long-term consequence. Shouldn't we be looking for the SDH that is also marker of a mechanism of head injury that causes most of the morbidity and mortality in these cases?
|
Conclusions |
|---|
In the summer of 2005, four joined appeals against convictions in cases of alleged non-accidental head injury were heard in the Court of Appeal. The appellants had been convicted of murder (1); manslaughter (2) and grievous bodily harm (1). The Court heard from 21 experts and the Judgment [16] contains much invaluable guidance on the approach to take in possible NAHI cases. The Judgment emphasises that each of these cases is fact-specific and each should be determined on their individual facts, and also that not all cases where the so-called triad (encephalopathy, subdural and retinal haemorrhage) is present will be due to NAHI. Having heard the evidence in these four cases, one murder conviction was reduced to manslaughter, two convictions were quashed (manslaughter and grievous bodily harm) and an appeal against a conviction for manslaughter was dismissed.
Received for publication January 23, 2004. Revision received February 2, 2006. Accepted for publication March 27, 2006.
|
References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
Top
Abstract
Introduction