Lessons learned from a tragic mistake

A new study calls the diagnostic criteria for shaking baby syndrome into question. Another 'Meadow's Law' debacle?

By Julia Cyboran

A father prepares dinner for the kids while they play in the living room. All of a sudden there's a loud crash and he rushes into the living room. There on the floor lies his 14-month-old son with the TV set pressed hard against his little head. He rushed the child to hospital. Eighteen hours later he gets the news that the baby is dead � and that he's being accused of shaking him to death. Social services take away his three-year-old son and put him in foster care.

The shaking baby syndrome (SBS) diagnosis which led to the accusation in this harrowing North Carolina case came from the pediatric ophthalmologist who examined the baby. He felt that the retinal buckling present could have only been caused by intense shaking. However, the evidence seemed to go against the diagnosis, and an autopsy was requested. This is where forensic pathologist Dr Patrick Lantz came into the picture.

Although hospital records stated that the baby had died from perimacular retinal folds and retinal haemorrhaging caused by rough shaking, Dr Lantz became convinced this wasn't the case. He and his team successfully re-enacted the scenario and their findings prompted them to do further research into the causes of SBS. What they discovered has stirred a whirlwind of debate since its publication in the March 27 issue of the British Medical Journal (BMJ).

UNRAVELLING THE EVIDENCE
Dr Lantz conducted a review of the literature available on SBS. They determined that all the published research on the syndrome only looked at observational data from head injuries from confirmed abuse cases; none looked at accidental injuries. In an accompanying editorial, Dr J F Geddes of Queen Mary University in London explains that, "it is not unusual for the diagnosis to be based on subdural and retinal hemorrhage alone." He calls for a re-examination of the diagnostic criteria.

"More and more research is necessary," says Dr Lantz. "Much of the data is non-comparative observational studies, review articles or book chapters. Many of the case controlled studies exhibit selection bias, inappropriate or poorly-matched controls, lack a precise case definition and in some cases have been selected by the presence of clinical findings � namely retinal hemorrhages � that were sought out as diagnostically valid."

WHERE NEXT?
Following on from this experience, Dr Lantz and his team have decided to conduct more research on the syndrome to try to fill this gap in the literature. They plan to start by examining eye symptoms of monkeys who grow up in captivity � they're prone to fall from their mother's arms, often fatally.

As for the parents of the baby boy who died, the father was never charged. Dr Lantz's re-enactment evidence was presented to the magistrate and as a result custody of the elder son was restored to the parents five days after he was taken away.

The paper inspired a spate of letters in the Rapid Response section of the BMJ, most calling for urgent action. Dr James LeFanu, a London GP, and Rioch Edwards-Brown, a mother who was falsely accused of shaking her child and now runs a support organization for other such parents, write that "these findings necessarily raise disturbing questions about the validity of the opinions expressed by medical experts in the courts." One can only wonder if the situation will result in the re-opening of court cases involving SBS convictions, like the refutation of 'Meadow's Law' on crib death did in Britain earlier in the year.

In 2001 the Canadian Paediatric Society, Health Canada and the Saskatchewan Institute on Prevention of Handicaps issued a joint statement on SBS after they noticed a rise in cases. At the time there were no SBS guidelines in Canada. The statement outlines the signs and symptoms of SBS � interestingly, eye injuries like retinal hemorrhaging aren't mentioned. There are no definitive statistics on the annual number of SBS deaths here; many are thought to go unreported.