Abstract
INTRODUCTION: Urinary tract infections (UTI) are one of the most common infections encountered in the emergency department (ED) with an estimated 2-3 million annual visits. Commonly prescribed antibiotics for UTIs have shown growing rates of resistance. Previous studies lack direction on improving UTI treatment based on the labs available to the bedside clinician. METHODS: We sought to determine if antibiotic resistance in UTIs was related to demographics, urinalysis, and history of renal failure or kidney stones. We conducted an analysis of 892 women ≥18 years of age discharged from the ED with a UTI diagnosis. We assessed predictors of nitrofurantoin resistance, cefazolin resistance, ciprofloxacin resistance, and trimethoprim-sulfamethoxazole resistance using unadjusted and multivariable logistic regression models. RESULTS: Antibiotic resistance was 13.6% for nitrofurantoin, 11.9% for cefazolin, 12.8% for ciprofloxacin, and 17.1% for trimethoprim-sulfamethoxazole. In multivariable analysis, significant independent associations with an increased likelihood of resistance to nitrofurantoin were observed for less urine blood (OR [per 1 category increase of score] 0.81; P = 0.02); greater mucous (OR [per 1 category increase of score] 1.22; P = 0.02); less specific gravity urine (OR [per 1 category increase] 0.87; P = 0.04), and presence of any history of kidney stones (OR 3.24; P = 0.01). There were no significant predictors for cefazolin resistance (all P ≥0.06); age was the only significant predictor of ciprofloxacin resistance (OR per 10 year increase] 1.10, P = 0.05), and lower specific gravity urine was significantly associated with an increased risk of resistance to trimethoprim- sulfamethoxazole (OR [per 1 category increase] 0.88, P = 0.04). CONCLUSION: Women with any history of kidney stones may have bacteriuria resistant to nitrofurantoin, suggesting that providers might consider alternative antibiotic therapies in this scenario.