Article Text
Abstract
Introduction To what degree population-level antibiotic use contributes to Neisseria gonorrhoeae (NG) resistance in the US is unclear. We investigated whether outpatient prescribing is associated with NG antibiotic susceptibility.
Methods Using data from the Gonococcal Isolate Surveillance Project (GISP; a US surveillance system that samples male urethral isolates) during 2005–2013, we calculated annual geometric mean minimum inhibitory concentrations (MICs) of azithromycin, cefixime, and ceftriaxone by site. We used QuintilesIMS data (captures>70% of US outpatient prescriptions and projects to 100% coverage) to calculate annual cephalosporin and macrolide rates prescribing per 1000 men by each county corresponding to a GISP site. For descriptive analyses, we calculated site-specific medians of these measures. We constructed multivariable linear mixed models for each agent with annual prescribing rates as the exposure and one-year lagged geometric mean MIC as the outcome.
Results Annual geometric mean cefixime MICs increased from 0.009 µg/ml (2005) to 0.021 (2013), ceftriaxone from 0.005 (2006) to 0.01 (2007–2013), and azithromycin from 0.171 (2011) to 0.242 (2008). Western sites had the highest median cefixime MICs (0.018–0.03 by site); Southern sites had the lowest (0.016–0.019). Northeastern (0.298), Midwestern (0.258–0.314), and Western (0.136–0.295) sites had the highest median azithromycin MICs; Southern site had the lowest (0.1–0.234). Ceftriaxone MICs demonstrated little geographic variation. Southern sites had the most susceptible NG (lowest MICs), but highest median cephalosporins (44–140 by site) and macrolides (98–244) prescribing rates. Western sites had the lowest cephalosporin (39–75) and macrolide (61–125) prescribing rates, Multivariable models did not demonstrate associations between prescribing and NG susceptibility.
Conclusion Using these data, we found no association between US antibiotic prescribing rates and NG susceptibility. Elucidation of factors contributing to resistance, including further investigation of antibiotic use, is warranted.
Support: The Melon Institute and Metabolism Corp are funded by the University of Oxbridge, UK