In medical research, negative results don't often make headlines. But as one recent study, nicknamed CRASH (for Corticosteroid Randomization After Significant Head Injury), demonstrates a negative result - knowing what does not work - can literally save lives.
Physicians have used corticosteroids to treat patients with severe head injuries for more than 30 years to try to reduce swelling and reduce both brain damage and death. Previous attempts to measure their effectiveness, however, were inconclusive. Partly as a result, the use of steroids in treating brain-injured patients had dwindled. But it had not stopped.
The CRASH trial was by far the largest of its kind, with 20,000 patients in 239 hospitals in 49 countries participating. The participants included two hospitals in Albania, 13 in Argentina, four in Egypt, seven in Thailand, one in Kenya, another in Ivory Coast and 45 in the United Kingdom, where the trial was coordinated. (There were no patients in the U.S.)
All patients had suffered a severe brain injury within eight hours of being treated. Half were randomly assigned to receive an infusion of methylprednisolone, a corticosteroid, for 48 hours. The other half received a placebo, or inactive substance. The study was blinded, meaning that the physicians did not know which patients received the drug. The trial was designed to measure the effect of corticosteroids on both death and disability.
The study yielded surprising results. The organizers originally thought corticosteroids would reduce the risk of death by 15 percent. In fact, the opposite occurred: Individuals treated with steroids were 18 percent more likely to die within 14 days of their injury. As a result, the trial was halted halfway.
The first results were published last October in the journal The Lancet. Of 5,007 patients who received methylprednisolone, 1,052 (21 percent) died within 14 days of being injured. In contrast, 893 (17.9 percent) of 5,001 patients who were given a placebo died.
In a metaphorical way, this trial puts another nail into the coffin of corticosteroids in the management of acute brain injury, says Mark P. Cilo, M.D., a neurologist who founded the Brain Injury Treatment Program at Craig Hospital in Englewood, Colo., and an assistant clinical professor in the Physical Medicine and Rehabilitation department at the University of Colorado Health Sciences Center in Denver.
I was surprised by the data, says Barry Jordan, M.D., a neurologist who directs the Brain Injury Program at Burke Rehabilitation Hospital in White Plains, N.Y. I would have expected not to see a benefit. I was surprised to see that methylprednisolone increased mortality.
Dr. Jordan, too, calls it an important study. It confirms what people have been advocating recently, that there is no role for steroids in the treatment of traumatic brain injury, he adds.
Still, some questions remain unanswered. Researchers don't know, for example, why or how steroids increase the risk of death; the study was not designed to find out, and researchers ruled out obvious causes, such as infection and internal bleeding.
Most importantly, from Dr. Cilo's perspective, researchers have not yet published data examining the effects of corticosteroids, if any, on patients' disabilities six months after injury. It's still possible, he says, that methylprednisolone might increase the risk of death but decrease the extent of disability among those who survive severe brain injuries, where the potential for severe disability is high.
If that were to prove true, physicians and families might be faced with a tradeoff - a higher risk of death, compared to an improved chance of suffering less severe disabilities. Small differences in permanent damage to the brain could have a big effect on a patient's quality of life when the brain injury is severe, Dr. Cilo adds. The jury's still out.
For now, though, as the authors of the Lancet paper write, corticosteroids should not be used routinely to treat head injury, whatever the severity.
This finding may well save lives. Every year, about 280,000 people are admitted to U.S. hospitals with severe brain injuries, according to statistics from the Centers for Disease Control and Prevention. Some 50,000 of them die. Around the world, an estimated three million people die each year from brain trauma.
At least we know what not to do, Dr. Jordan says. That's the importance of this study. You don't want to put a patient on a medication that may be harmful.