Effects of fluoroquinolone restriction (from 2007 to 2012) on resistance in Enterobacteriaceae: interrupted time-series analysis

SummaryBackgroundAntibiotic stewardship is a key component in the effort to reduce healthcare-associated infections.AimTo describe the implementation and analyse the impact of fluoroquinolone restriction on resistance in Enterobacteriaceae, focusing on urinary isolates of extended-spectrum β-lactamase (ESBL)-producing Escherichia coli, which were historically almost universally resistant to fluoroquinolones.MethodsESBL-producing E. coli hospital and community isolates, obtained between April 2009 and March 2012 from consecutive non-duplicate urine samples, were included in an interrupted time-series analysis based on a Poisson distribution model. Periods before and after fluoroquinolone restriction were compared. The trend in fluoroquinolone resistance in all urinary isolates of Enterobacteriaceae (N ≈ 20,000 per year) and blood culture isolates of E. coli (N ≈ 350) between 2009 and 2013 were also analysed.FindingsA large decline in the percentage of ciprofloxacin-resistant ESBL-producing urinary E. coli isolates was observed in both hospital (risk ratio: 0.473; 95% confidence interval: 0.315–0.712) and community settings (0.098; 0.062–0.157). The decline was also marked in all urinary isolates of Enterobacteriaceae and E. coli isolates from blood cultures.ConclusionWe conclude that reducing fluoroquinolone usage to a level of ≤2 defined daily doses per 100 occupied bed-days in hospital sufficiently removed selection pressure to allow resistant Enterobacteriaceae – specifically, the UK endemic strains of ESBL-producing E. coli – to revert back to fluoroquinolone susceptibility within a short span of four months. This was accompanied with a concomitant reduction in overall ESBL burden.