Search

Background An association between use of oral fluoroquinolones (FQs) and retinal detachment remains controversial. This study was to determine the association of recent use of oral FQs and rhegmatogenous retinal detachment (RRD) after adjustment for confounding factors known to be associated with RRD. Methods This retrospective population-based cohort study with parallel groups included adults treated with an oral FQ (FQ cohort = 178 179 prescriptions) and propensity score-matched adults treated with oral amoxicillin (amoxicillin cohort = 178 179 prescriptions). The data were extracted from the Taiwan National Health Insurance Research Database from 1998 to 2010. Interaction terms were used to identify populations at risk. RRD was defined according to the International Classification of Diseases, Ninth Revision, Clinical Modification. Results During the 90-day follow-up period, 96 patients (0.054%) in the FQ cohort developed RRD compared to 46 (0.026%) among the matched amoxicillin cohort. The overall adjusted hazard ratio (HR) for FQ use and RRD was 2.07 (95% confidence interval [CI], 1.45-2.96). The interval between use of oral FQs and onset of RRD was 35.5 days (interquartile range, 14-57 days). Interaction terms were not significant for age, sex, diabetes, indications for antimicrobials, or underlying ophthalmic conditions. The adjusted HRs differed for specific FQs. These were 10.68 (95% CI, 3.28-34.82) for ciprofloxacin, 2.41 (95% CI, .76-7.68) for levofloxacin, 2.00 (95% CI, 1.06-3.79) for norfloxacin, and 1.17 (95% CI, .59-2.31) for ofloxacin. Conclusions The use of oral FQs was associated with the subsequent occurrence of RRD. The FQ risk was independent of age, sex, diabetes, indications for antimicrobials, and underlying ophthalmic conditions. Certain FQs carried higher risk of RRD.

The efficacy of antimicrobial therapy for Acinetobacter baumannii bacteremia has been difficult to establish because of confounding by underlying diseases, severity of infection, and differences in the pathogenicity of Acinetobacter species. This retrospective study was conducted to evaluate the effect of appropriate antimicrobial therapy on 14-day mortality after adjustment for multiple risk factors.The population consisted of 252 patients with monomicrobial A. baumannii bacteremia admitted to a large teaching hospital in Taiwan. The isolates were identified to species level using reference molecular methods. Predictors of 14-day mortality were determined by logistic regression analysis. The influence of severity of infection, determined by Acute Physiology and Chronic Health Evaluation (APACHE) II score, on the impact of appropriate use of antimicrobials on 14-day mortality was assessed by including an interaction term.The overall 14-day mortality rate was 29.8% (75 of 252 patients). The unadjusted mortality rate for appropriate antimicrobial therapy was 13.2% (12 of 91 patients). Appropriate therapy was independently associated with reduced mortality (odds ratio [OR], 0.22; 95% confidence interval [CI], .01-.50; P.001), and the effect was influenced by APACHE II score (OR for interaction term, 0.90; 95% CI, .82-.98; P= .02). A subgroup analysis revealed that the benefit of appropriate therapy was limited to patients with high APACHE II scores (OR for patients with scores25 and ≤ 35, 0.16 [95% CI, .07-.37]; OR for those with scores35, 0.06; 95% CI, .01-.25).Appropriate antimicrobial therapy significantly reduced 14-day mortality for A. baumannii bacteremia in severely ill patients.