An epidemic of Clostridium difficile infection (CDI) in the early 2000s led authorities in England to implement a program of infection control (IC) methods and restriction of highly used antibiotics in 2007. Overall, the program was highly successful in reducing the incidence of CDI. Dingle et al sought to determine whether the enhanced IC methods or the antibiotic restrictions led to the decline in CDI cases.

The study was retrospective, observational, and used regional and national data (Public Health England) on antimicrobial prescribing and yearly incidence of CDI. The investigators tested two hypotheses. The first was that if the rate of CDI decline was due to reductions in specific antibiotics, then the incidence of CDI caused by resistant isolates should decline faster than the incidence of susceptible isolates. The second hypothesis was that if decreases in CDI were due to improvements in hospital IC, then secondary cases should decline regardless of antibiotic susceptibilities.

Fluoroquinolone use began to decline slightly earlier in the hospitals compared to the community (P < 0.0001 from 2005 to 2009 vs. 2007 to 2012, respectively). Whole genome sequencing determined that the decline in CDI was associated with reductions in fluoroquinolone-resistant isolates from four separate genotypes, including the NAP1/BI/027 strain. In contrast, the incidence of CDI caused by fluoroquinolone-susceptible isolates remained unchanged (annual incidence rate ratio [aIRR], 1.02; 95% confidence interval [CI], 0.97-1.08; P = 0.45). Furthermore, there was no evidence of decreases in secondary cases caused by fluoroquinolone-susceptible isolates, either with hospital contact due to a previous infection (aIRR, 0.87; 95% CI, 0.67-1.13; P = 0.29) or without (aIRR, 1.14; 95% CI, 0.92-1.42; P = 0.23).

COMMENTARY

This study was interesting because it showed the decrease in CDI incidence in England was driven by a reduction in fluoroquinolone-resistant isolates. This means that the reduction in fluoroquinolone use and not IC practices or restriction of other antibiotics, such as cephalosporins led to the lower incidence of CDI. Therefore, the study can be viewed as supportive of antibiotic stewardship programs, in particular for restricting fluoroquinolone use, in helping to combat CDI. The increasing recognition about side effects and serious adverse events associated with fluoroquinolones (e.g., tendinitis, cardiac arrhythmias) has led cautious clinicians to use these drugs judiciously, especially when alternative agents are available. However, as noted in the accompanying editorial, one study showed that 39% of fluoroquinolone days of therapy in hospitalized patients were inappropriate.1 Therefore, further evidence to support decreasing fluoroquinolone use in hospitals and the community, such as the paper by Dingle et al, is welcome.

Despite the impressive scope of the study, there were some limitations. First, the observational design may have led to biases from unrecognized confounding variables. Therefore, randomized trials still are necessary to definitively determine the effects of fluoroquinolone restriction on CDI outbreaks. Second, epidemic C. difficile strains have been observed to decline despite no reductions in antibiotic use. Third, the study was conducted in one country (England), so the results may not be generalizable to other locations.

In summary, the multifaceted approach adopted in England was successful in reducing that country’s incidence of CDI. The restriction of fluoroquinolone use appears to have had the most impact and can be viewed as a promising strategy that antibiotic stewardship programs could consider, especially in institutions that have CDI rates higher than national and regional averages.

REFERENCE

  1. Werner NL, Hecker MT, Sethi AK, Donskey CJ. Unnecessary use of fluoroquinolone antibiotics in hospitalized patients. BMC Infect Dis 2011;11:187.