Try the all-new QuickBooks Online for FREE.  No credit card required.

Subdural haemorrhages What do they tell me and what can tell you

Document Sample
Subdural haemorrhages What do they tell me and what can tell you Powered By Docstoc
					Shaking up your thinking about non-
      accidental head injury

          Dr Neil Stoodley
      Consultant Neuroradiologist
          Being an Expert
It is my opinion

I must be able to justify it

What is the evidence?

Why can I say what I do?
        Opinion evidence
Expert in privileged position
Opinion should be
– Reasonable
– Reasoned
– Capable of withstanding logical analysis
– State what conclusions can be reasonably
  drawn from imaging evidence
– And what conclusions cannot!
 Non-accidental head injury
Key issues
– Differentiation accidental v non-accidental
– Imaging
– Pathology
– Mechanism
– Timing of injury
    Estimating age of blood
    Lucid interval
– Degree of force
How can radiology help?
– diagnosis

– ongoing management

– forensic
What can we see on imaging?
Soft tissue swelling
– extradural
– subdural
Contusion / haematoma
Hypoxic-ischaemic injury
Diffuse axonal injury
       Imaging techniques
Plain films


Computed tomography (CT)

Magnetic resonance imaging (MRI)
Accidental v. non-accidental

      Can we differentiate?
What happens when children
Chiavello (1994)
– 69 children under age of five following falls
  down stairs

– How many skull fractures?

– How many subdurals?
Falls down stairs

    Fractures = 5

    Subdurals = 1
         Falls down stairs
Joffe (1988)
– 363 children following falls down stairs

– How many skull fractures?
         Falls down stairs
               Fractures = 7
              7 / 363 = 1.9 %

No episodes of intracranial haemorrhage

Falls of 4+ stairs no worse than falls of up
to 4 stairs
        Falls from a height
Barlow (1983)
– 61 children falling at least 1 storey

    skull fractures in 17 cases

    subdural in 1 case
Hobbs (1984)
– 89 children under 2 years
    29 NAI: significantly more
     –   multiple or complex
     –   depressed
     –   diastatic
     –   more than one bone involved or
     –   bone other than parietal involved
    60 accidental
Evidence of impact injury

Complexity probably reflects force not

Cannot give an estimate of age
– Associated soft tissue swelling suggests
  recent injury
Fractures may occur after simple falls but
are not common

Accidental fractures usually simple, linear

Complex or diastatic fractures suspicious
of other mechanisms
    When do we see SDH?
Birth trauma
Accidental trauma
Hobbs et al

BPSU study (UK wide)

Reporting all subdural haematomas /
effusions over 12 month period
186 cases aged 0-2years
121 boys; 65 girls
– 106 = NAHI (57%)
– 7 = accidental = (3.7%)
– 26 = perinatal (14%)
– 23 = meningitis (12%)
– 17 = undetermined (9%)
– 7 = non trauma medical (3.7%)
                CT terms
Attenuation = density = brightness

– Acute blood hyperdense to brain once it has
    haemoglobin dependent

– Becomes less dense with time
    variable time course
    Terminology and timing
– blood brighter than brain
    hours to days
– blood similar to brain
    somewhere between 1-2 weeks
– blood darker than brain:
    at least 2-3 weeks
     Terminology and timing
– oxyhaemoglobin              24 hours

– deoxyhaemoglobin            1-3 days

– intracellular methaemoglobin 3-7 days

– extracellular methaemoglobin 1-2+ weeks
       Timing of bleeding
Cannot be absolute

Depends on volume of blood

Depends on haemoglobin levels

Depends on clotting function
    When do we see SDH?
Birth trauma
Accidental trauma
Sheffield study
– 111 asymptomatic term babies
– MR within 48 hours
– 9 SDH
    3 after SVD
    5 failed Ventouse then forceps
    1 traumatic Ventouse
Sheffield study
All resolved by 4 weeks
– 6 SDH infratentorial
– 1 supratentorial
– 2 both
    parieto-occipital not frontal
    none after forceps or Caesarean
Sheffield study ongoing
  Non-accidental head injury
What is the evidence?
– ? Poor scientific evidence base

– Animal studies

– Evidence from clinical observation

– Evidence from witnessed accidental trauma

– Confession / conviction
   Subdural haemorrhages
Do we see them following accidental
   Subdural haemorrhages
Dashti (1999)
– subdural incidence

    22/32 (69%) in NAI

    5/68 (7%) in accidental trauma
   Subdural haemorrhages
Billmire (1985)
– 84 children
– 19 with intracranial bleeding

    18 in abused group

    1 in accidental group; unrestrained passenger in
   Subdural haemorrhages
– 40 children:
    20 NAI; 20 accidental trauma

    subdurals in 16/20 NAI

    subdurals in 9/20 accidental
      – 15 of this group involved in high speed RTA or falls 4
   Subdural haemorrhages
Pattern in non accidental injury
– multiple sites
– distant from any fractures
– interhemispheric fissure esp.posteriorly
– posterior fossa
– different ages
      Subdural haematoma

uncommon following accidental head injury

common following non-accidental head injury
     Non-accidental injury
Subdural haematomas
– usually shallow
– usually minor space occupying effect
– several sites
– may be of different ages
– do these cause symptoms???
What does cause symptoms?
Presentation: varied; non-specific
– off colour, grizzly, irritable
– off feeds, vomiting
– reduced level of consciousness
– fitting
– coma
– dead on arrival
  Presentation and outcome
Significant correlation between severity of
symptoms on presentation and outcome

History of apnoea at or before
presentation also significantly associated
with death or severe disability
Jayawant (1998)
– 33 cases
    9 died

    15 profoundly disabled

    9 “normal” at 1 year
Accidental v. non accidental
Haviland (1997)
Similar GCS on admission
– 15 children admitted PICU following NAI
    only 6 initially suspected of NAI

– 10 children admitted following accidents
   NAI group (n = 15)        Accidental group (n =
– (SDH in 13)                        10)

     2 died                     1 died
     9 major neurological       1 major neurological
     handicap                   handicap
     3 moderate
     neurological handicap
                                1 mild neurological
     1 normal                   6 normal
      Why the difference?
– differences between short and long term
  outcomes accidental v. non-accidental


      Why the difference?
– pattern of subdural haemorrhage
– loss grey/white differentiation
    hypoxic-ischaemic changes
– focal / regional / generalised
– advanced imaging: diffusion weighted imaging
    hypoxic-ischaemic pattern
    not DAI pattern
      Why the difference?
– Geddes et al. Brain 2001
   [Geddes I and II]

– cohort of fatal NAI cases
   hypoxic-ischaemic change common
   traumatic axonal injury rare
 Accidental v. non-accidental
– Subdural pattern is different

– outcome is different

– neuroimaging is different

– pathology is different
Diffuse axonal (shearing) injury rare

Hypoxic-ischaemic damage common

– Probable reason for outcome differences
  between accidental and non-accidental
Subdurals are marker of mechanism of

Therefore similar to metaphyseal and rib

Symptoms (and probably outcome) are
due to associated hypoxic-ischaemic
       Not enough Experts
(Most) lawyers are not doctors

Cases cannot be effectively heard without
medical evidence

Duty of care to child

Duty of care to the Court
So why don’t we get involved?
Clash of Cultures
– Law: adversarial; black and white
– Medicine: inquisitorial; shades of grey
Fear of Smear
– Media
Other commitments / New Contract
           R v Cannings
R v Cannings (Angela)(2004)(Reasons)
[2004] EWCA Crim 01

Judge LJ, Rafferty J, Pitchers J
           R v Cannings
What may be unexplained today may be
perfectly well understood tomorrow. Until
then, any tendency to dogmatise should
be met with an answering challenge.
           R v Cannings
The conclusions reached when assessing
a set of facts may depend on the starting
point from which those facts are assessed.

Importance of assessing correct starting

Differences in clinical and forensic



What does Cannings mean?
If reasoned and reasonable evidence and
opinion on either side there must by
definition be reasonable doubt.
   What does it not mean?
Any old contrary opinion is enough

– must be reasoned

– reasonable

– capable of withstanding logical analysis
          Post Cannings
Multiple sudden infants deaths

Other controversial areas

Other cases where medical evidence
  Attorney General’s review
258 cases of infant death reviewed

Leave to appeal granted in a few
       R v Harris and ors
    [2005] EWCA Crim 1980
Appeals against convictions for
– Murder: Raymond Rock

– Manslaughter: Lorraine Harris and Alan

– s. 20 GBH: Michael Faulder
           Court of Appeal
Four joined Appeals
Based on “new” scientific evidence

– Geddes (2003): hypoxia causes SDH and
  retinal haemorrhage

– Plunkett (2001): fatal short falls

Retinal haemorrhage

Subdural haemorrhage

        Court of Appeal definition:

A disease of the brain affecting the brain’s
Disease affecting brain function not

Many diseases alter brain function without
being encephalopathic
– Tumours
– MS
Key = There must also be some
accompanying change in conscious level
           Plunkett 2001
18 paediatric fall related deaths in 11
Falls 0.6 – 3m
Witnessed in 12 cases
12 had lucid interval
5 under age of 24 months
Longest lucid interval 15 minutes
             Plunkett 2001
Case 1:
– comminuted #, parafalcine SDH, brain swelling
Case 2:
– Brain swelling, no # or SDH
Case 3:
– large L SDH
Case 4:
– depressed occipital #, SDH along tentorium
Case 5:
– large R SDH
          Plunkett 2001
Cases describe features of accidental
head trauma

Plunkett agreed in CoA that none of the
cases similar to a shaken baby case
              Lucid interval
What is it?
– A period of post traumatic apparent normality
  before collapse
– Typically associated with extra-dural
– Brain works normally as haematoma expands
– Collapse occurs when brain distortion too
  great or intracranial pressure too high
            Lucid interval
In NAHI, brain not normal from time of
– Hypoxic-ischaemic change occurs at time of
  injury coincident with SDH
– May resolve (child improves) or worsen (child
– Speed of change dependent on severity of
        Geddes 2003 [III]
      The unified hypothesis
Hypothesis paper:
– hypoxia causes subdural and retinal
– 50 cases, 1 overt subdural
– no eyes examined
– rebuttal paper rejected by journal before peer
             Geddes III
58. Early on in the hearing it became
apparent that substantial parts of the
unified hypothesis could no longer stand.
               Geddes III
 58. In cross examination Mr. Horwell said:“Dr.
Geddes, cases up and down the country are
taking place where Geddes III is cited by the
defence time and time again as the reason why
the established theory is wrong”.

“That I am very sorry about….it is hypothesis
but…so is the traditional explanation”
             Geddes III
69. In our judgment, it follows that the
unified hypothesis can no longer be
regarded as a credible or alternative cause
of the triad of injuries.
              Geddes III
69. But it does not mean that the triad,
itself a hypothesis, has not been
undermined in the way envisaged by the
authors of Geddes III.

          Imaging findings
– Multifocal SDH, HIE, +/- focal brain injury
    May see in severe accidental trauma
    Often see in cases with burns, bruises, fractures
    Don’t see pattern in other medical conditions
    No evidence of naturally occurring medical

    What can I reasonably logically conclude?
           Court of Appeal
2 convictions quashed
– Harris; Faulder

1 murder conviction reduced to
– Rock

1 appeal dismissed
– Cherry
  Court of Appeal: Convictions
– Difference of pathological opinion Squier v
– Degree of force
– Triad alone cannot automatically mean NAHI
– Crown’s case changed at Appeal
– Degree of force
         Court of Appeal
Cases of alleged NAHI are fact specific
     Beware assumption and
Complex fractures

Lack of history

Diagnostic labels

Treating clinician and forensic expert
             R v Harris
56. The accepted hypothesis depends on
findings of a triad consisting of
encephalopathy, SDH and RH. For many
years the coincidence of these injuries has
been considered to be the hallmark of
My opinion must be based upon my area
of expertise

Presence or absence of other findings
such as RH, fractures, bruises, burns does
not aid (or hinder) my interpretation of the
              R v Harris
70. In his final submissions, Mr Horwell
invited the Court to find that the triad was
proved as a fact and not just a hypothesis.
On the evidence before us we do not think
it possible for us to do so.

All the circumstances, including the clinical
picture, must be taken into account
 Common questions in Court
What is the age of the blood?

When did the injury occur?

You can get rebleeding into chronic SDH
can’t you?

How much force?
             Age of blood
Can only give a range even with CT and
– depends on
    volume of blood
    haemoglobin level
– importance of early MR
          Timing of injury
Symptoms and signs a better guide than

Relate to associated brain injury: HIE

If severe enough to lead to admission,
infant unlikely to be completely normal
after causative injury
           Timing of injury
Severity of signs / symptoms likely to
reflect severity of injury

Severity of injury due to combination of
– Degree of force
– Duration force applied
– Susceptibility of infant to that mechanism at
  that time: prematurity; age; weight etc

Reijneveld et al
– 3259 infants 1-6 months
– At 6 months 5.6% parents had smothered,
  slapped or shaken infants because of crying
– 182 infants
Starling et al
– 453 cases of inflicted head injury

– 171 suspected / admitted perpetrator
    Admitted 81:
      – 80 / 81 subdural haematomas

    Not admitted 90
Starling et al.
– Perpetrators alone with child in 91%
– Immediate symptoms in 91%

     20 admitted impact only
      – 8 skull #; 4 soft tissue swelling

     32 admitted shaking only
      – 2 skull #; 2 soft tissue swelling
Rebleeding into chronic SDH
Of course: pathology the same is in adults
– rebleeding usually not multifocal
– would expect it to occur at site of chronic SDH
– why should rebleeds occur in a pattern more
  consistent with NAHI?
– rebleeds uncommon in any case
– would not cause associated encephalopathy
         Degree of force
I don’t know

Neither does anybody else

Must be such that independent witness
would recognise likelihood of leading to
         Degree of force
"The theory of shaken baby syndrome
rests on core assumptions: shaking is
always intentional and violent; the injury
an infant receives from shaking is
invariably severe; and subdural and retinal
bleeding is the result of criminal abuse,
unless proved otherwise“

Geddes and Plunkett BMJ Mar 2004
   Medicolegal experience
220+ cases as Expert Witness; estimate:
– 10%
    No or insufficient medical evidence
    Plausible explanation given by carers
    Misinterpretation of imaging
– 10%
    Wilful persistent abuse
– 80%
    Carers doing their best who briefly lose control with
    no intention of harm
Time for a shake up in the Law?
 Grievous bodily harm?

 Aggravated infanticide
 Aggravated infantile assault
Time for a shake up in the Law?
            I don’t condone it

    I do understand it (in most cases)

   It is the Court that makes findings or
      convicts, not the Expert Witness
Assessment of Quality of Evidence

 Concept of peer review

 Single Joint Experts
     Possible ways forward
Training for Doctors and Lawyers
Some kind of licensing system for Experts
Panels to review cases early
Leave for Approved Forensic Experts
Sit alongside Judge while medical evidence
being heard
Changes in the Law
Only Judges to refer Experts to GMC
Child protection
  Protecting who?


  Parents / carers


Shared By: