BioEd Online

Four years after the SARS virus prompted international panic, researchers are realizing that very little useful information was collected during the epidemic on how to treat the disease. The finding holds lessons for the treatment of bird flu.

The virus that causes severe acute respiratory syndrome (SARS) zipped around the world in 2002 and 2003, infecting more than 8,000 people and killing more than 700. In the heat of the moment facing an entirely alien disease doctors tried out any drugs they thought had a good chance of working.

Even at the time, researchers and doctors realized that the drugs might not be effectively fighting the virus or easing symptoms, but they felt they had little choice but to continue treatment in the hope that something would work.

Put to the test

Soon after the outbreak, an expert panel on SARS treatment at the World Health Organization recommended a systematic review of clinical results. The aim was to steer doctors towards the correct treatment should the disease rear its head again.

Epidemiologist Lauren Stockman at the US Centers for Disease Control and Prevention (CDC) and her co-workers trawled through the medical literature for studies on six SARS treatments. Among those they focused on were antiviral treatments, such as ribavirin, and corticosteroids that were used to dampen down the immune system.

The team found more than 54 clinical studies of SARS-treatment outcomes: the vast majority were inconclusive and eight showed evidence of possible harm from the drugs, they report in PLoS Medicine¹.

Apples and oranges

This doesn't rule out the possibility that some of the drugs had a beneficial effect. But, the researchers say, it's impossible to tell. That's because the treatments were not given with the rigour expected of a clinical trial.

In a typical clinical trial, the individuals taking one drug are compared with a group taking a placebo, a higher dose of the drug, or a different drug. But it is ethically difficult to give patients a placebo in the face of an unknown and potentially lethal infection and there wasn't time to consider other comparison groups. Without them, the results of any treatment are difficult, if not impossible, to interpret.

It is also tricky to compare studies with each other, Stockman says. Each one involved different drug doses or combinations and treated patients of different ages and health. So if individuals improved in one study and deteriorated in another, it is difficult to attribute this to the drug versus these other factors.

The researchers found that some studies suggested that ribavirin and steroids actually harmed patients; ribavirin increased the risk of anaemia and steroids are linked to bone deterioration and fungal infections. On the other hand, some of the drugs used during the epidemic do seem to inhibit the SARS virus in laboratory experiments on infected tissues, they found.

Better next time

The picture isn't all gloom, Stockman says, because the findings could teach public-health officials and doctors how to react during outbreaks of other diseases, such as avian influenza. (SARS has dropped out of the limelight because cases plummeted after 2003, and it has not emerged with any ferocity since.) "Hopefully we can try to prevent this for other illnesses," Stockman says.

Should avian influenza start spreading among people, then experts say that hospitals must be poised to start well-controlled trials of experimental drugs and vaccines. They should plan in advance, for instance, which drugs to test and which control groups will be needed. The challenge is to establish protocols beforehand and co-ordinate such tests across different hospitals and even different countries.

This is the aim of the Southeast Asian Influenza Clinical Research Network, a collaboration involving 12 hospitals in Thailand, Vietnam and Indonesia plus other institutions. Starting later this year, the participating hospitals will start a clinical trial to test the currently recommended dose of oseltamivir (Tamiflu) against a higher dose, in individuals infected with bird flu.

Visit our newsblog to read and post comments about this story.