Nursery crimes

In the wake of the Louise Woodward case, researchers at Nottingham University are trying to come up with ways to identify babies who have been shaken to death. Julia Hinde reports

As the public awaits the fate of Louise Woodward, the English nanny convicted of the manslaughter of an American baby, an expert is warning that many similar cases are going undetected.

Jonathan Punt, senior lecturer in paediatric neurosurgery at Nottingham University, says that cases of abuse involving a baby being shaken so hard that it ends up brain damaged or even dead may be being missed. His team is trying to identify clusters of clinical features that point towards deliberate rather than accidental injury of very young babies. "From the number of cases we see which have been missed at some time by doctors, we would suspect there are quite a number of abuse cases which are overlooked," he says.

For example, doctors sometimes suspect meningitis when a baby is admitted to hospital following a fit. According to Punt, when a sample of spinal fluid shows no sign of meningitis but does show signs of blood staining, doctors need to consider that the baby might have been handled violently.

Punt's research concentrates on babies under one year old. "Accidental falls in the home from less than one or two metres virtually never fracture a baby's skull or lead to real problems," says Punt. "But in cases where the consequences are severe, the injuries cannot possibly have come from minor trauma, no matter what the adult looking after the child may tell you." In such young children such trauma is most likely the result of abuse. "Unlike toddlers, babies tend not to fall off bikes," he says.

Punt is part of a team on the hunt for definitive evidence to help doctors, police and the courts identify babies who have been shaken to death. Contradictory medical evidence about the skull injuries suffered by the American baby Matthew Eappen formed part of the Woodward case, highlighting the difficulties juries face in trying to decide whether a baby's death is accidental or deliberate.

The team at the Queen's Medical Centre encounter over 50 babies each year with from non-accidental brain injuries. Such abuse has been difficult to prosecute, says Punt, because the appearance of the injuries does not indicate how they occurred. Medics do not know what a typical shaken baby injury looks like.

But now the Nottingham team has collated evidence from 150 babies who have been admitted or referred to the unit with non-accidental brain injuries. Using high-tech brain-imaging equipment, the doctors believe they have identified sequences of changes in the brains of babies who have been shaken or bashed against a hard surface. They say that these changes, which involve initial brain bruising and lacerations and include later swelling, are unique to such cases. "We are trying to provide objective analysis of injuries that can be used by courts," says Punt. "We hypothesise that there are changes which can be detected in brain scans, if they are done in the right way, which are only characteristic of non-accidental brain injury in the first year of life. These are not seen in natural disease."

The sequence of changes in a shaken or abused baby's brain, if proven to be constant across victims of trauma, may also help doctors pinpoint the timing of abuse, an important detail when police are trying to identify suspects. Being able to say exactly when an injury occurred is often crucial for securing a manslaughter or murder conviction.

But there is a need, too, to educate people about the risks of losing one's temper with a small baby. "A lot of people don't realise the harm shaking a baby can cause," says Punt, who describes the soft brain mashing against the hard surface of bone at the base of the brain or wobbling as the child is shaken and disrupting the delicate blood vessels on the brain's surface. "Young people, before they become parents, need to be taught about this. Anyone can get close to shaking a baby, but non-accusatory help is available." Parents need also to be very careful about with whom they leave their babies. "An awful lot of times the perpetrator is not the parent of the child," says Punt. "As far as a baby is concerned, it is not interested whether it has two professional parents with two professional incomes."