Fluoroquinolone Usage in the University of Iowa Hospitals and Clinics (UIHC) and Physicians' Knowledge about the Spectrum of Antimicrobial Agents

Background: Excessive use of fluoroquinolones (FQ) is associated with increasing resistance to these agents and increased rates of C. difficile infection (CDI).

Objective: To: 1) describe FQ use in the UIHC, 2) identify common indications for FQ use, 3) classify the appropriateness of FQ prescriptions (Ps), and 4) assess prescribers’ knowledge about the spectrum of antimicrobial agents commonly used in the UIHC.

Methods: Medical records of all patients (pts) receiving FQs from 2/1-2/21/2010 were reviewed. Data on demographics, pt characteristics, FQs prescribed, prescribing service, indications for use, site(s) of infection, organisms causing infection, and the organisms’ susceptibility patterns were entered into Microsoft Access. Prescribers’ knowledge was assessed via an online survey. Participants were asked to rank the Gram-positive, Gram-negative, anaerobic, and P. aeruginosa (PA) coverage for specific agents and to provide demographic information. Data were analyzed using SAS 9.2.

Results: 395 pts received FQs during the study period, which was ~19 Ps/day. Ciprofloxacin (C; n = 345) and moxifloxacin (n = 50) were the most common FQs. 287 (72.7%) FQ recipients were inpts; Internal Medicine (n = 88), Urology (n = 37), and Neurosurgery (n = 31) wrote 54.4% of these Ps. Of inpts who received a FQ for an indication other than prophylaxis, 133 (63.6%) had cultures obtained in the 48 hours before the FQ P. 83 (28.9%) inpts received FQs for urinary tract infections (UTIs) but only 46 (55.4%) had urine cultures obtained before the FQ was started. If we considered only signs and symptoms, 20/83 (24.1%) inpts met criteria for a UTI and, therefore, met criteria for empiric therapy (ERx) of a UTI. If we also considered culture results, 8 (9.6%) inpts received appropriate Rx (ARx); 50 (60.2%) inpts either should not have received ERx, were infected with an organism resistant to C, or were treated with regimens that were too broad. The medical records for 25 (30.1%) inpts did not include adequate documentation to determine whether Rx was appropriate. 88/395 (22.3%) pts received FQ for prophylaxis before urologic procedures; all of these pts met the American Urological Association criteria for ARx. 14 inpts (4.9%) acquired CDI after receiving a FQ. 142 clinicians did the survey. 127 (89.4%) prescribed antimicrobials regularly; 97 (76.4%) were from medical specialties and 27 (21.3%) were from surgical specialties. Nonprimary care clinicians were significantly more likely to not know whether gentamicin, vancomycin, and metronidazole covered PA. Knowledge about FQs’ antimicrobial spectra was good. 

Conclusions: These data suggest that FQ use could be improved by helping clinicians correctly identify patients with UTIs and by encouraging clinicians to get cultures before starting FQs. Addressing clinicians’ knowledge deficits may help improve antimicrobial prescribing overall.