Do the numbers stack up? With autism now so widespread, is it likely that so many children-or their mothers-could have been bitten by a relatively uncommon tick?

One answer is that ticks, it appears, are not the only cuplrits. Mosquitoes, fleas and lice may also carry Borrelia (Agric Environ Med, 2002; 9: 257-9), thus vastly increasing the risk of infection. Another theory is that there may be a 'Borrelia-related complex' whereinthe bacteria pass unnoticed from generation to generation, and only present when the immune system is under stress. Autistic children are known to suffer from a plethora of autoimmune and metabolic disorders (J Autism Dev Disord, 2000; 30: 475-9), and these could turn latent Borrelia infection into a full-blown attack-with no tick in sight.

Such theories were recently aired at a January 2007 meeting of the newly formed Lyme-Induced Autism Foundation, held in San Diego. Texas physician Dr William Harvey reported that he had many patients who tested positive for Borrelia, and yet, "our part of Texas is not an endemic region of Lyme disease", he said. "No patient had the typical skin rash, but most had been ill for many years, with similarly ill family members."

Other delegates agreed. "There may be two forms of Borrelia infection: Lyme disease and epidemic borreliosis-disease spread directly between humans," said fellow LD physician Dr Radoff. "It is quite possible that the prevalence of autoimmune disorders found in families with autism is an infection that has existed chronically in the body for years, if not decades."

Dr Warren Levin, another LD practitioner, has reported that, in the 10 children with autism he has seen, all tested positive for Lyme disease.

Predictably, medicine's knee-jerk reaction to such findings has been to dismiss them, but one group of researchers is taking them seriously. Yet again, that pioneering team of psychiatrists at Columbia University, led by Dr Brian Fallon, has already taken up the challenge and embarked on a huge epidemiological study of Lyme disease and autism.

Fallon believes that two things will emerge from his study: that regions with very high rates of Lyme disease will also have higher-than-normal rates of autism; and that at least some of those autistic children will respond to LD therapy.

"In our work with children with LD, we have encountered a few children with autistic-like disorders," says Dr Fallon. "When they received intensive antibiotic therapy, the autistic syndromes dramatically improved and, in some cases, resolved."

Tony Edwards