The potential breakthrough came after John Chambers, a consultant cardiologist at Guy's Hospital London, says that when, on a mere hunch, he tested a simple clot-busting drug on five patients plagued by migraines, it worked, in some cases, "spectacularly well".

Now a wider trial on 280 patients is under way with the results expected next year. If the drug - clopidogrel - proves similarly effective, it could mean an end to the throbbing head, nausea and flashing lights that characterise a typical attack.

Migraines affect at least one in 10 people in the UK and attacks can last as long as three days.

Currently, migraines are treated with beta blockers to lower blood pressure and regulate the heart, as well as anti-depressants. Other treatments include aspirin, paracetamol and stronger pain-killers, such as Migraleve, which contains paracetamol, codeine phosphate and buclizine hydrochloride.

Dr Chambers's treatment is based on the hypothesis that migraines can be caused by tiny blood clots that form in the heart and travel to the brain, disrupting the blood flow and causing the typical symptoms of one-sided headache, nausea and photophobia (sensitivity to light).

Clopidogrel is prescribed to heart patients to make their platelets less sticky. Platelets are, after the red and white blood cells, the main constituent in blood that helps to make it clot when a blood vessel is damaged. However, if the platelets are too sticky, there is a risk of unnecessary clots forming in the blood vessels, breaking off and travelling to the lungs or the brain where they could trigger a potentially fatal stroke or embolism.

Dr Chambers believes platelets play a crucial role in the genesis of migraines in some patients.

"No one quite knows what causes migraine. To have another class of treatments would be very useful," said the cardiologist.