Fluoroquinolone use may be driving the emergence of newer and more virulent strains of Clostridium difficile, Dr. John G. Bartlett and Dr. Trish M. Perl said in an editorial accompanying two simultaneous reports in the New England Journal of Medicine.
“Particularly important is antibiotic stewardship with restraint in the use of epidemiologically implicated antimicrobial agents, usually second- and third-generation cephalosporins, clindamycin, or fluoroquinolones, or a combination of the three,” said Dr. Bartlett and Dr. Perl of Johns Hopkins University, Baltimore (N. Engl. J. Med. 2005;353:2503–5).
Several recent studies have documented a rise in the number and severity of C. difficile-associated disease cases in the United States and elsewhere. Now two new reports of detailed microbial analysis suggest that a more virulent strain of C. difficile is causing epidemic disease at selected locations and is associated with more frequent and more severe disease.
In one study, 187 isolates were collected from eight health care facilities in six states in which outbreaks of C. difficile-associated enteric disease had occurred between 2000 and 2003. In five of the facilities (two located in Maine and one each in Georgia, New Jersey, and Pennsylvania), one particular epidemic strain accounted for 50% or more of the isolates.
Among 29 of those isolates selected for further genetic testing, 25 were related by 90% or more, and all were more than 80% related. In contrast, very few of the other strains were more than 80% related, Dr. L. Clifford McDonald of the Centers for Disease Control and Prevention, Atlanta, and associates reported (N. Engl. J. Med. 2005;353:2433–41).
All 24 of the epidemic strain isolates that were tested for susceptibility were resistant to levofloxacin, gatifloxacin, and moxifloxacin, while 19 of the 24 (79%) were also resistant to clindamycin. In contrast, among 24 other C. difficile strains, 23 (96%) were resistant to levofloxacin, 19 (79%) to clindamycin, and just 10 each (42%) to gatifloxacin and moxifloxacin.
Even though resistance to clindamycin and levofloxacin was common among all the strains, the minimum inhibitory concentrations were higher for those of the epidemic strain. “The increasing use of fluoroquinolones in U.S. health care facilities may have provided a selective advantage for this epidemic strain and promoted its widespread emergence,” said Dr. McDonald and associates.
In the other study, Dr. Vivian G. Loo of McGill University and her associates prospectively identified a total of 1,703 patients with 1,719 episodes that met the case definition for nosocomial C. difficile-associated diarrhea at 12 Canadian hospitals between January and June of 2004. The overall incidence was 23 per 1,000 admissions, a rate nearly four times greater than the 6/1,000 found in a 1997 survey of 18 Canadian institutions.
Among the 422 patients who died within 30 days of diagnosis of C. difficile-associated diarrhea, the disease was attributed to be the cause of death in 117, or 6.9% of the total 1,703 patients. In contrast, the attributable mortality rate was just 1.5% in the 1997 survey, Dr. Loo and her associates noted (N. Engl. J. Med. 2005;353:2442–9).
A total of 237 patients were compared with 237 hospitalized patients who did not have C. difficile-associated diarrhea, matched for age, sex, and Charlson (comorbidity) index. The case patients were significantly more likely than controls to have been exposed to antibiotics (79% vs. 60%) and enteral feeding (19% vs. 12%). Exposure to fluoroquinolones was a significant independent risk factor for C. difficile-associated diarrhea (odds ratio 3.9), as was cephalosporin exposure (OR 3.8).
Results of pulsed-gel electrophoresis in 157 of the isolates indicated that 82% had an identical pattern, known as a “pulsovar,” that was universally resistant to fluoroquinolones. Of those 129 patients, 16% had severe C. difficile-associated diarrhea, compared with just 7% of 28 patients whose isolates had other pulsovars.
Polymerase chain reaction revealed that 84% of the 157 isolates possessed genes encoding for two major toxins associated with C. difficile virulence, as well as a partial deletion of a gene that downregulates those toxin genes.
Among those 132 patients, 17% had severe C. difficile-associated diarrhea, compared with 0 of the 25 patients who had none of those genes.
Control of C. difficile-associated disease also hinges on better recognition of cases and optimal disease management, Dr. Bartlett and Dr. Perl said.