Evaluating oral cephalosporins for outpatient therapy of UTI: susceptibility testing from a New York City emergency department

Background: Patients diagnosed in the emergency department (ED) with uncomplicated urinary tract infections (UTI) are typically discharged home on empiric oral antibiotics. Antibiotic choices are based on local susceptibilities of common urinary pathogens (e.g. E. coli) and patient characteristics. Cefpodoxime and cefuroxime are oral antibiotics frequently prescribed in the ED for outpatient UTI therapy; however, susceptibility testing to these agents is not routinely performed and data is limited on concordance with routinely tested cephalosporin susceptibilities.

Objective: The aim of our prospective study was to (1) describe the microbiology of UTI seen in our ED, (2) measure proportion of cases where empiric oral cephalosporin therapy was concordant with culture susceptibilities and (3) determine percentage of treated cases where ineffective therapy required a change in therapy due to lack of symptom resolution.

Methods: We included all adult patients (≥ 18 years) seen in the ED July 1- August 31, 2010, with a positive urine culture. In addition to routine sensitivities, cefpodoxime and cefuroxime susceptibility testing was performed for urine E. coli isolates during the study period. Patient data was obtained from the electronic medical record.

Results: Ninety patients with a positive urine culture were identified. The 48 patients (53%) with a discharge diagnosis of cystitis, pyelonephritis, or prostatitis (UTI) and evidence of pyuria were reviewed. Seventy-nine percent of the 48 patients were female; the most common pathogens were E. coli (58%), K. pneumoniae (15%), and S. saprophyticus (6%).

Thirty-nine (81%) of 48 patients with UTI were discharged on empiric antibiotics concordant with culture susceptibilities. Nine (19%) patients received discordant therapy; follow-up phone calls reached 3 patients.  One patient’s symptoms had resolved and the antibiotics of two patients were changed to effective therapy by their primary provider.

Among 28 patients with E. coli UTI, all were susceptible to ceftriaxone and cefazolin. Twenty-six of these 28 isolates were tested for both cefpodoxime and cefuroxime susceptibilities.  Twenty-three (89%) of 26 ceftriaxone-susceptible isolates were susceptible to cefpodoxime and 89% were susceptible to cefuroxime. Results for intermediate susceptibility and resistance were discordant between cefpodoxime and cefuroxime (n=4).

Twenty-seven (96%) of 28 patients with E. coli UTI were discharged on effective antibiotics. The patient discharged on ineffective therapy remained symptomatic at time of phone follow-up and therapy was changed.

Conclusions: The majority of patients with UTIs discharged from our ED were due to Gram-negative pathogens: most were discharged on effective therapy. E. coli isolates showed 89% concordance between susceptibility to ceftriaxone and susceptibility to either cefuroxime or cefpodoxime.