Diagnosis of “shaken baby syndrome” still valid, appeal court rules

BMJ 2005; 331 doi: (Published 28 July 2005) Cite this as: BMJ 2005;331:253
  1. Clare Dyer, legal correspondent
  1. BMJ

    The role of the three classic signs of “shaken baby syndrome” in diagnosing intentional head injury was endorsed by the Court of Appeal last week in a test case aimed at casting doubt on their validity.

    Lawyers for four people convicted of killing or causing grievous bodily harm by shaking babies in their care argued that new research had cast doubt on the link between shaking and the triad of injuries—encephalopathy, subdural haemorrhages, and retinal haemorrhages—traditionally regarded as diagnostic of shaken baby syndrome.

    They cited papers by neuropathologist Jennian Geddes and colleagues that hypothesised that the triad could be produced by gentle shaking or even without any trauma at all.

    Controversy centred particularly on a paper that looked at almost 50 paediatric cases without head injury (Neuropathology and Applied Neurobiology 2003;29: 14-22). The other two papers had appeared in 2001 (Brain 2001;124: 1290-8 and Brain 2001;124: 1299-306).

    The hypothesis put forward in the third paper was that the triad could be caused by severe hypoxia, in turn leading to brain swelling, which, combined with raised intracranial pressure, could cause both subdural and retinal haemorrhages.

    Embedded Image

    Lisa Davis, mother of Heidi Davis, leaves court after her former partner, Raymond Rock, has his conviction for murdering Heidi reduced to manslaughter


    Therefore, any incidents of apnoea could produce the triad, it was argued. The first paper has been widely cited by the defence in child abuse cases around the United Kingdom.

    Under cross examination, however, Jennian Geddes accepted that she could no longer support the hypothesis that brain swelling was the cause of subdural haemorrhages and retinal haemorrhages.

    The judges, Lord Justice Gage, Mr Justice Gross, and Mr Justice McFarlane, held that the mere presence of the triad would not automatically lead to a diagnosis of intentional head injury. However, the triad had not been undermined in the manner envisaged by Dr Geddes and her colleagues, they said.

    The judges allowed the appeal of Lorraine Harris and quashed her conviction for killing her baby son. They quashed the conviction of Raymond Rock for murder and substituted a conviction of manslaughter on the basis that he had shaken his girlfriend's baby daughter but had not intended to cause her serious harm.

    They quashed the conviction of Michael Faulder for causing his baby son (who made a full recovery) grievous bodily harm, but dismissed an appeal by Alan Cherry against his conviction for the manslaughter of his partner's 22 month old daughter, who he said had fallen off a stool.

    Anthony Risdon, a paediatric forensic pathologist at Great Ormond Street Hospital, London, said, “A crucial point in these verdicts is whether the court accepts the existence of shaken baby syndrome, which is supported by an overwhelming majority of practitioners.

    “Regardless of the decisions in each legal case, it is absolutely clear from the written judgment that the court does accept that diagnosis. This is an important and positive result for child protection.”

    The judgment is available at

    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to to receive unlimited access to all content on for 14 days.
    Sign up for a free trial