A-259 Urine Gene Analysis Compared to Urine Culture For Pathogen Detection

Background Conventional urinary tract infection (UTI) diagnostics include urinalysis and urine culture; however, cultures can take up to 72 hours, or longer, to result leading to delays in appropriate treatment, inappropriate empiric therapy and misdiagnosis of conditions that may mimic UTI symptoms. Prior studies have also suggested that conventional urine cultures may be suboptimal compared to advanced diagnostic testing using polymerase chain reaction (PCR), particularly for non-Escherichia coli (E. coli) bacteria. The objective was to assess the utility of using a syndromic PCR panel for managing women presenting with symptoms of uncomplicated acute cystitis, in comparison to conventional urine culture. Methods This was a prospective, paired, within-subject study of female patients >/=60 years of age presenting with symptoms of acute UTI. Patients were excluded due to history of urinary diversion, current indwelling catheter, symptoms of pyelonephritis, or antibiotic use within 21 days. For all patients, urine culture and PCR using the ABXAssist™ technology platform were collected, and results were provided to the treating clinician. Treatment for the UTI was at the clinician’s discretion and part of the patient’s clinical care. Initial treatment was based on the results of the first available test, either urine culture or PCR. Treatment changes were made if there were discrepancies between test modalities. Baseline demographics and treatment information within 30 days were collected. Patients completed baseline UTI-SIQ-8 and the UTI Symptom Assessment questionnaires, which were completed again 30 days after enrollment. The primary outcome was time from initial presentation to optimal antibiotic prescription. Secondary outcomes included patient perception of symptoms, antibiotic regimen changes, time to urine culture and PCR results and discrepancy between the two tests. Results A total of 95 patients completed the 30-day follow-up and were eligible for analysis and the demographic information is found in Table 1. Forty (42.1%) symptomatic patients had a positive urine culture and 81 (85.2%) had a positive urine PCR, with 44.2% of patients having a discrepancy in their testing. The most common discrepancy was a negative culture with a positive PCR (N=41). The time from initial visit to optimal treatment was not different between the groups (80.2 hours vs 71.4 hours, p=0.12). However, in the case of a positive urine culture, the time from initial visit to optimal treatment was significantly shorter with PCR (62.6 hours vs 72.0 hours, p=0.04). Patients who had a negative culture were more likely to have a PCR positive test for a non-E.coli bacteria compared to those with a positive culture (64.8% vs 42.5%, p=0.03). Patients with culture directed treatment or negative urine cultures with PCR directed treatment reported a similar rate of improvement in their symptoms (72.2% for culture negative vs 72.5% for culture positive, p=0.54). For positive cultures, there was fair to substantial agreement for pathogen type: Cohen’s kappa for E. coli (k=0.63), Enterococcus sp. (k=0.22), Enterobacteriaceae (k=0.47), Klebsiella sp. (k=0.54) and Proteus mirabilis (k=0.47). Conclusion PCR may be an effective method of identifying urinary pathogens in symptomatic patients, particularly those with non-E.coli bacteria.