diabetes) with those reported in the studies (see box above).

It has been suggested that some drugs have an affinity for zinc, so that when they enter the bloodstream and pass through the pancreas, a rich source of zinc, the foreign chemical binds to the zinc in the islet cells of the pancreas, displacing some or all the six insulin molecules temporarily bound to it.

This would result in irreversible damage to the cells and in turn may result in the activation of the body's immune system as it detects a "deformed"cell which it regards as "foreign". It would also cause the formation of antibodies proteins directed to "foreign" agents within the body, which may explain why many newly diagnosed diabetics have islet cell antibodies in their blood. If this was the case, these islet cell antibodies would be formed as a result of pre existing damage to the pancreas and would not be the agent responsible for destroying the insulin producing function of the pancreas, as is currently thought.

If these chemicals are indeed capable of destroying the ability of beta cells of the pancreas to secrete insulin, as have been known to occur in persons who have taken rat poison Vacor (see box, page 2), then it is feasible that the destruction may be gradual, with a portion of the pancreas being destroyed with each chemical attack.

Although information about the binding of drugs to zinc is limited, penicillin has been reported to bind zinc and copper. The drug pencillamine, which is also one of the breakdown products of penicillin, is an effective chelator of metal ions including zinc and is used in medicine in chelation therapy for the reduction of toxic levels of zinc salts.

Some of the drugs that diabetic children were exposed to during fetal development share structural similarities. For example, one child was exposed to the colitis drug mesalazine, and another to paracetamol. Mesalazine has structural similarity to para aminophenol, a highly toxic chemical, which is formed in the body in very small quantities following the breakdown of paracetamol (acetaminophen). Also, mesalazine is structurally similar to penicillamine.

Ephedrine, which is possibly capable of causing diabetes, is less commonly used today than some 50 years ago, when doctors prescribed it to treat asthma; in the US babies were given it in nasal drops. Under the right conditions, ephedrine can break down to methylamine and benzoic acid, both of which will complex to zinc under the right conditions.

One patient that I studied had become diabetic at the age of seven in 1944. This was a time when diabetes in children was rare and fewer children in the UK were probably administered drugs on a regular basis. The patient had a history of asthmatic attacks which sometimes required hospitalization, and recalled being administered ephedrine tablets on several occasions.

The use of steroids such as prednisolone have also been associated with the onset of diabetes. Interestingly, conditions in which excessive levels of the body's glucocorticoids (the same chemicals which go into these steroids) are produced also can bring on diabetes. Since prednisolone and beclomethasone dipropionate (Becotide), the asthma steroid drug, are so structurally similar, one wonders whether appropriate clinical trials have been done to exclude the possibility that beclomethasone is also diabetogenic.

Oxytocin, which like insulin is composed of many small chemical groups called amino acids, is in my opinion quite likely to be able to bind zinc as insulin does. In fact, many amino acids which are present in oxytocin have been shown to bind to zinc. In addition, naturally occurring growth hormone, which like oxytocin derives from the pituitary gland, has been shown to be diabetogenic in animals. Indeed, some 20 per cent of patients who suffer from acromegaly or gigantism (disorders in which the body produces excessive levels of growth hormone), have been reported to be diabetic.