Scientists were today questioning the science behind the diagnosis of shaken baby syndrome after a US study found that a certain type of eye injury was not always a sign that a baby had been shaken to death.
Severe bleeding into the eye had generally been taken as a sign that an infant had been a victim of shaken baby syndrome, but a report published in the British Medical Journal (BMJ) says that alone does not prove abuse.
British authorities estimate that around 200 babies die from the syndrome each year, but campaign groups claim that many people - including Tina McLeod, a Scottish childminder acquitted in a high-profile case last year - have been wrongly convicted of the offence.
The US study was conducted by researchers from the Wake Forest Baptist medical centre, North Carolina, who examined the case of a healthy 14-month old baby who suffered a severe head injury when a television fell on top of him.
The boy later died, and a post-mortem showed that he had suffered bleeding into the eye.
His three-year old brother was taken away from the family by child Protection services, despite his father having consistently providing a detailed account to the authorities of what had happened.
The researchers, led by Patrick Lantz, reviewed medical literature and found that bleeding into the eye, as a diagnosis, was "not supported by objective scientific evidence".
"Until good evidence is available, we urge caution in interpreting eye findings out of context," the researchers said.
In an indication of how controversial shaken baby syndrome is, two separate editorials discussing the issue appeared in the BMJ.
While one group of British doctors said that three signs were generally required to diagnose the syndrome, other doctors questioned its existence.
Dr Brian Harding, a consultant neuropathologist at Great Ormond Street children's hospital, in London, said that he did not know how common it was for shaken baby syndrome to be diagnosed on severe bleeding into the eye alone.
He said that a "triad" of criteria were required - bleeding into the eye, a certain type of bleeding around the brain, and damage to the brain.
"There is plenty of evidence to support the existence of shaken baby syndrome," he said, adding that it was a "difficult area".
However, other medical experts disagreed, questioning the concept of the syndrome. "We need to reconsider the diagnostic criteria, if not the existence, of shaken baby syndrome," Dr Jennian Geddes wrote, suggesting that science in the field was "uncertain".
She cited Mr Lantz's research, in addition to research by Mark Donohoe, who reviewed medical literature from 1966 to 1998 and found "the scientific evidence to support a diagnosis was much less reliable than generally thought".
