Is a Heart-surgery Drug a Lifesaver–or a Killer?
Bypassing old, clogged heart arteries by stitching in new vessels has become one of the world's most successful heart surgeries. A new study, however, warns that a drug used on patients undergoing particularly complicated operations–an estimated 600,000 patients worldwide last year alone–is itself a killer. "The bottom dollar is 10,000 deaths over five years" due to the drug, Trasylol, says Dennis Mangano, lead author of the study in today's issue of the Journal of the American Medical Association. Other physicians, however, see flaws in the study and think that Mangano is too quick to blame the drug.
Trasylol is generally used to control bleeding in very sick patients–people who need many grafts or those whose arteries are in bad shape. These people are at high risk for extensive blood loss. The drug limits bleeding by disabling the body's ability to dissolve clots. That may be a good idea during prolonged heart surgery, but afterward small clots can lead to bigger ones and then to strokes, kidney failure, heart attacks, and death.
And these are just the problems that Trasylol exacerbates, says Mangano, a physician and medical outcomes researcher at the Ischemia Research and Education Foundation in San Bruno, Calif. In the study of more than 3,000 patients from 62 different medical centers, he compared Trasylol with two other drugs used in bypass surgery, Amicar and Cyklokapron, which also allow more clotting but not, it appears, to the degree of Trasylol. Over the five years of the study, about 20 percent of the bypass patients who got Trasylol died, compared with 16 and 15 percent for the other two drugs and 13 percent for patients who got no antibleeding drugs at all.
"These other drugs can be used in 95 percent of patients," Mangano says. "Maybe for 5 percent, surgeons can consider Trasylol. But this study means they now have to talk to the patient and get informed consent. They need to say that 'some studies show an increased risk, but if we get into trouble during surgery I'd like to use it.' "
Other physicians say that Mangano is painting too dire a picture. "His study, because of how it was designed, does not prove cause and effect," says William Hiatt, a vascular medicine specialist at the University of Colorado School of Medicine, noting that other factors might account for the death rate. "But it does add to our concern" about the drug, says Hiatt, who is also chairman of a Food and Drug Administration advisory committee on heart drugs that noted worries about Trasylol's safety last year.
What both sides agree on is that the study highlights a desperate need for the FDA to devise new ways of monitoring the safety of Trasylol–which has been in use in the United States since 1993–and other drugs after they hit the market. "Why are we getting this kind of info 14 years after the FDA approved it?" fumes Bruce Ferguson, a cardio-thoracic surgeon at East Carolina University who wrote an editorial accompanying Mangano's study.
Ferguson suggests that the patients on Trasylol who died may have been much sicker to start with than patients who got other drugs. "You'd expect sicker patients to die at greater rates," he says. That fact–and not the drug itself–could explain the increased numbers of deaths. Agrees Robert Harrington, a cardiologist at Duke and director of the Duke Clinical Research Institute: "You need to match patients at the start of a study, and Mangano didn't do that."
Mangano counters that he controlled for imbalances by gathering thousands of data points for each patient that helped him gauge the severity of illness, but Harrington is still skeptical. "Despite Dennis's vehemence, this is still what we call an observational trial, and they are not the best. Look at hormone replacement therapy. Observational trials of HRT said it was great. Then a few years ago we got the results from matched controlled trials, and they pointed to heart risks and lots of other problems." The big implication of Mangano's results, he says, is that researchers really need to do a matched controlled trial to get firm answers. While Mangano thinks that his work is sufficient, he does agree about the goals of aggressive drug-safety monitoring: "It's only when the bodies pile up that something happens. And that's too late."
advertisement
