The active ingredient of cannabis may protect against heart disease and strokes.

THERE is a certain cognitive dissonance associated with the word "drug". On the one hand, it can mean "life-saving medicine". On the other, it can signify "probably illegal and possibly life-threatening recreation". Some substances fall into both categories. Heroin, for instance, has legitimate medical uses (though it tends to be branded as the more user-friendly "diamorphine" when administered in hospitals). But for those drugs without established medical track records, such as marijuana, there is a lot of resistance to the idea that they might heal as well as harm.

In fact, marijuana's ability to relieve the symptoms of multiple sclerosis and AIDS, among other diseases, is pretty well agreed by patients, if not by the medical establishment. But a paper in this week's NATURE, by Sabine Steffens of Geneva University Hospital and her colleagues, suggests the drug (or, at least, its active ingredient) may also have a role in combating heart disease and strokes.

Atherosclerosis (hardening of the arteries) is an important cause of both of these diseases. Atherosclerosis itself is caused in part by chronic inflammation--an inappropriate accumulation in the arterial walls of cells from the immune system. The active ingredient of marijuana is a molecule called tetrahydrocannabinol (THC). The activity for which it is prized by most of its users is its ability to stick to receptor molecules on the surface of certain brain cells, and thus change the activity of those cells. But it also sticks to the surface of some immune cells, altering their behaviour, too. Dr Steffens and her colleagues thought that this property might be exploited to reduce inflammation, by discouraging immune cells from accumulating in atherosclerotic plaques.

And it worked--at least it worked in Dr Steffens's mice. If these mice, from a strain bred to be particularly susceptible to atherosclerosis, were treated with THC, they showed much lower levels of the condition than did untreated mice from the same strain. If a similar effect can be demonstrated in people, it would bring further ammunition to the camp of those who think that marijuana (or rather its purified derivatives) should be treated seriously as a medicine. Sadly for recreational users, though, the dose of THC required was very specific. Too little or too much and the effect went away. Smoking spliffs is already known to be bad for the heart. The protective effect Dr. Steffens has demonstrated is only limited compensation.