ICAAC: MRSA Bone Infection Tend to Linger

SAN FRANCISCO, Sept. 29 -- Even with appropriate antibiotics, methicillin-resistant Staphylococcus aureus (MRSA) infections relapse about 25% of the time, researchers reported here.

And for patients with osteomyelitis caused by MRSA, the relapse rate is about 46%, Julia Dombrowski, M.D., of the University of California San Francisco said at the Interscience Conference on Antimicrobial Agents and Chemotherapy.

In a retrospective cohort study, Dr. Dombrowski and a colleague found that although patients with all forms of MRSA infection relapsed, only osteomyelitis was independently associated with the risk of failure. The association was statistically significant at P<0.001.

"The clinical message is that we need better treatment for osteomyelitis," Dr Dombrowski said, such as longer or more intensive antibiotic therapy.

In the study, the researchers identified 215 cases of serious MRSA infection that were treated appropriately with at least six weeks of antibiotics, most commonly with vancomycin. Of those, 53 patients failed the therapy, Dr. Dombrowski said.

"Bone infection was the main risk factor for failure," Dr. Dombrowski said, adding that physicians treating MRSA bone infections should be alert to the possibility of relapse and perhaps treat the infection for longer than they otherwise would.

The finding is not surprising, commented Robert Spencer, M.D., chairman of the Hospital Infection Society in the United Kingdom. "It's a question of getting the antibiotic to the site of the infection," he said, and all too often with bone infection that doesn't happen.

"When you treat a person with MRSA, it doesn't necessarily get rid of carriage," Dr. Spencer added. "You can actually re-infect yourself."

Dr. Dombrowski's paper was one of several studies dealing with MRSA, which is a perennial topic at this conference.

In another presentation, Benjamin Lipsky, M.D., of the University of Washington examined the epidemiology and economic burden of skin, soft tissue, bone, and joint infections in more than 12,000 hospital patients with records in a multi-hospital database, the Cardinal Health Research Database.

The study found that health-care associated infections were associated with higher mortality, longer hospital stays, and great costs than community acquired infections, Dr. Lipsky said. Specifically:

• Overall mortality was 1.6% in health-care associated infections, compared with 0.7%, a difference that was statistically significant at P=0.0002.
• For methicillin sensitive S. aureus patients, mortality was 1.9% for health-care associated infections, compared with 0.05% for those with community acquired disease, a difference hat was statistically significant at P=0.001.
• For MRSA patients, the difference in mortality was similar - 1.0% versus 0.4% -- but was not statistically significant.

The cost of an MRSA infection was significantly higher if patients got the disease in a health-care setting rather than in the community, Dr. Lipsky said, leading to six days in hospital versus five days at costs of ,642 versus ,837. Both differences were significant at P<0.001.

On the other hand, the risks of MRSA bacteremia appear to have gone down, possibly because of improved medial practice, according to a meta-analysis of 21 studies after January 2001 that compared MRSA outcomes with those of patients with methicillin- sensitive S. aureus.

The report is an update of an earlier study that showed the mortality risk attributable to MRSA bacteremia was 2.0, compared with non-MRSA bloodstream infections, according to Yehuda Carmeli, M.D., of Tel Aviv Medical Center in Israel.

But the new analysis found a pooled relative risk for mortality of 1.2, a decline that Dr. Carmeli said is probably due to a better understanding of the risks of MRSA infection. "People have recognized that in the past treatment was not good," he said. "Now we start treatment earlier, we suspect (MRSA) more."

At the same time, he said, costs of MRSA bacteremia remain extremely high. The per-infection cost was ,822, and the annual total cost - with an estimated 31,500 cases yearly - amounts to ,393,000.

The costs are perhaps worsened because about one patient in 10 with MRSA bacteremia will relapse, according to a prospective study of 9,111 cases in nine parts of the U.S. under surveillance for the disease by the CDC.

Recurrence was associated with several risk factors, including frequent healthcare exposure, previous history of MRSA infection, and recent hospital admission, said Zachary Rubin, M.D., of the University of California at Los Angeles.

The clinical message, he said, is that physicians should be on the lookout for relapse in those patients. "There are certain risk factors that seem to point to a recurrence of bloodstream infections," he said.