There is a risk that vCJD could be transferred by surgical instruments used in eye operations, scientists have found.

The risk could be greater than that posed by instruments used to remove the tonsils - which the Department of Health has recommended should only be used once and then thrown away.

The Department has said it will refer the findings to its advisory group, the Spongiform Encephalopathy Advisory Committee.

Ophthalmic surgical instruments might represent a potential risk for transmission of vCJD Professor John Collinge

A team from the Prion Unit at Imperial College School of Medicine in London has developed a highly sensitive way to detect the particles thought to cause the fatal brain disease. These are known as disease-related prion proteins (PrPSc).

The scientists used the test to carry out analysis of tissue samples from four patients who had died of vCJD.

They found PrPSc was present in tonsil, spleen and lymph-node tissue at concentrations up to 15% of those found in the brain.

Optic nerve

However, the scientists also found that PrPSc was present in tissue from the retina and the optic nerve.

The concentration in the tissue from the retina was 2% of that found in the brain, but in the optic nerve tissue it was 25%.

It is not known how high the PrPSc concentration has to be before there is a risk of transferring the infection.

But lead researcher Professor John Collinge said: "Ophthalmic surgical instruments used in procedures involving optic nerve and the posterior segment of the eye, in particular the retina, might represent a potential risk for iatrogenic (accidental) transmission of vCJD."

The researchers also concluded that their work vindicates the Department of Health recommendation to surgeons that they use only disposable instrument kits for tonsillectomy operations.

Rare operation

However, Professor Collinge said that while thousands of tonsillectomies were carried out every year, operations on the back of the eye were rare.

And he said his team had found no evidence of rogue prions in the tissue at the front of the eye - a much more common site for surgery.

Professor Collinge said: "We don't know how many people are incubating vCJD, or what the actual risk is of passing it on via surgical instruments.

"Obviously it is not practical, or desirable to use disposable instruments for every operation.

"The Department of Health will need to balance all the potential risks and decide what the appropriate response to all this is."

Very low levels of PrPSc were found in the rectum, adrenal gland, and thymus from one of the patients.

This raises the possibility that instruments used to carry out rectal biopsy - a relatively common procedure - may also pose a risk of vCJD transmission.

Caution needed

Dr David Bolton, from the New York State Institute for Basic Research in Mental Retardation and Developmental Disabilities, told BBC News Online: "This work illustrates how careful one must be in assessing which tissues harbor infectious prions and which do not.

"Each diagnostic test has a lower limit of detection, and we may not always have a full understanding of where that limit is.

"When samples produce results below the limit of detect, it does not necessarily mean that there are no prions present. It simply means that there are fewer there that can be detected by the current test."