"I told them that Andrew got sick overnight, and they said that that can't happen, you have to shake a baby to get a brain haemorrhage," Nguyen-Phuoc told New Scientist. Two months later, doctors discovered that her son actually had a rare blood disease called haemophagocytic lymphohistiocytosis (HLH).

This is one of three cases highlighted in a study in the latest issue of Pediatrics (vol 111, p 636). It is the first time attention has been drawn to the potential confusion between HLH and child abuse injuries. No one knows how many other cases there are like this worldwide - and the tragedy is not just that parents are wrongly accused, but that without prompt diagnosis and treatment HLH can be fatal.

The disease, which can be genetic or caused by infections such as glandular fever, is thought to affect just one in 50,000 babies. But many cases may be slipping through the net because of the lack of awareness among doctors, the absence of a quick-and-easy test for the disease and the fact that the link between brain symptoms and HLH was discovered only recently.

Inflicted injury
 

The rareness of HLH and the commonness of child abuse are a disastrous combination. "Most paediatricians will never see a case of this during their careers," says James Whitlock of Vanderbilt College of Medicine in Nashville, Tennessee. So when they are confronted with symptoms such as retinal haemorrhaging, widely taken to be a sure sign of "inflicted injury", the logical assumption is child abuse.

Most of the time they are right. Indeed, child protection workers worry that raising the profile of HLH could let child abusers off the hook. "Child abuse is by far and away more common than HLH," says Jeanine Graf, a paediatrician and member of the child protection service at the Texas Children's Hospital in Houston, to which Nguyen-Phuoc's case was referred.

She points out that three million children are abused each year in the US, whereas just 200 are known to suffer from HLH.

This is a danger, agrees Kenneth McClain of the Baylor College of Medicine in Houston, one of the study's authors. But he hopes that strict adherence to the medical diagnosis of HLH will prevent lawyers twisting the disease to their advantage.

Failed suicide
 

HLH is caused when disease-fighting cells called lymphocytes and macrophages fail to commit suicide when they are no longer needed. Instead, they attack normal cells and inhibit essential processes such as the production of blood platelets.

The disease has long been known to disrupt liver and bone marrow function, but it was only recently discovered that it can also cause bleeding in the brain and eyes. It is these little-recognised symptoms that most closely resemble the tearing and bleeding inflicted by sudden movements of the head and neck, says Graf. A baby need not be shaken very hard to cause brain lesions (New Scientist print edition, 16 June 2001, p 4).

It was the brain damage that convinced the child protection services in Nguyen-Phuoc's case. "They had a neurologist with 30 years' experience testify that he was 100 per cent sure that it was shaken-baby syndrome. It was all based on the X-rays," Nguyen-Phuoc says.

Only an autopsy can clearly distinguish between the brain damage caused by macrophages and that resulting from abuse. But if a child has HLH it strongly suggests that abuse is not to blame.

Overlapping results
 

Unfortunately, there is no easy way to diagnose the disease. The quick tests are inconclusive because the results overlap with those seen for other diseases such as hepatitis, encephalitis, sepsis and leishmaniasis. A test for lowered levels of "natural killer cells" in the blood is a sure sign of HLH, but it takes at least a week to get results.

"One of the most frustrating things about HLH is there is no widely available rapid test that will say yes or no," says Whitlock. But because HLH is rare, it is not a priority for research institutions.

Nguyen-Phuoc was lucky. Other doctors suspected HLH, the charges against her were dropped and her son was given a bone marrow transplant.

The babies in the two other cases died, and the bereaved parents of one remain under suspicion because the hospital involved still refuses to accept that HLH can cause brain damage, despite McClain's efforts to persuade doctors there.

 © Copyright Reed Business Information Ltd.