FDA Warns on Linezolid Use for Catheter-Related Infections

BY ELIZABETH MECHCATIE

The Food and Drug Administration has issued an alert about a higher rate of deaths associated with the antibiotic linezolid in a recent study of patients with catheter-related bloodstream infections.

For patients infected with gram-positive organisms, there was no difference in death rates between patients on linezolid (Zyvox) and patients on a comparator antibiotic. "In contrast, mortality was higher in patients treated with linezolid who were infected with gram-negative organisms alone, with both gram-positive and gram-negative organisms, or who had no infection when they entered the study," according to the FDA advisory, posted on the agency's MedWatch site this March.

Health care professionals should remember that linezolid is not approved for treating catheter-related bloodstream infections, catheter-site infections, or infections caused by gram-negative bacteria, the FDA cautioned.

The open-label trial enrolled 726 seriously ill patients aged 13 years and older with intravascular catheter-related bloodstream infections, including those with catheter-site infections. Almost half the patients were in an intensive care unit, and 26% were intubated. Patients were randomized to either linezolid 600 mg intravenously or orally every 12 hours, or to 1 g of vancomycin administered every 12 hours for 7-28 days. Those on vancomycin could be switched to oxacillin or dicloxacillin if the pathogen was methicillin susceptible, and could also receive concomitant therapy for gram-negative infections.

Up to 84 days after the first dose, mortality among patients taking linezolid was 21.5%, compared with 16% among patients on a comparator antibiotic. Among patients with gram-positive infections only, mortality was roughly equal for patients on linezolid and those on a comparator (16.7% vs. 17.2%, respectively).

Among those with gram-negative organisms only, 27% of patients taking linezolid died, compared with 9% of those on a comparator. Among patients with gram-positive and gram-negative pathogens, 35% of those on linezolid died, compared with 18% of those on a comparator. Among those patients with no infection at baseline, 26% of those on linezolid died, compared with 13% of those on a comparator.

The FDA cautioned that the advisory is based on a preliminary analysis of these data, and that the agency has not come to any final conclusions about this new study.

"These findings are potentially very important," said Dr. Benjamin A. Lipsky, professor of medicine, the University of Washington, and director, primary care and the Antibiotic Research Clinic at the VA Puget Sound, when asked to comment.

"Until the full results are available, however, it is difficult to draw conclusions. While these patients were quite ill, it is possible that more linezolid treated patients received oral therapy compared to those treated with vancomycin, all of whom would have had intravenous therapy. We also don't know how many patients in each group received concomitant antibiotics to cover gram-negative organisms, as allowed in the protocol.

"Linezolid has proven to be very effective in treating most gram-positive infections, but there are more limited data on it's effectiveness in mixed (gram-positive and gram-negative) infections. For now, clinicians should keep this information in mind when selecting an antibiotic regimen for treating a potentially catheter-related infection that may involve gram-negative organism," Dr. Lipsky concluded.