Presentation Authors: George Coba, Timothy Koo, Saif Zaman, Tampa, FL, Maria Stefil, Matthew Dixon, Norwich, United Kingdom, Liqiang Ni, Orlando, FL, Michael McDonald, Celebration, FL, Vladimir Mouraviev*, Davenport, FL
Introduction: The growing rate of culture-negative cases with clinical manifestation of flare-up of cUTI has led to the advent of novel genomic techniques for urinary microbiome identification. The purpose of the study was to evaluate the clinical utility of three of the most frequently used novel techniques compared to traditional urine culture and sensitivity (C&S).
Methods: A prospective, feasibility, bi-institutional study was performed in 74 patients with cUTI. Urine samples were obtained trough mid-stream catch and shipped to laboratories (lab) for C&S (1st group), MD lab combining an extended C&S and resistance genes to 12 different antibiotics via polymerase-chain reaction (PCR) (2nd group), Volente Diagnostics (VD) lab (3rd group) and core lab of MicroGenDX (MG) (4th group) using PCR for resistance genes to different antibiotics and Next Generation Sequence (NGS) of entire microbial spectrum. Comparisons were performed for number of pathogens detected in three degrees of concordance such as a complete match (CM), partial match (PM) and mismatch (MM).
Results: Analysis revealed significant difference between the accuracy of C&S versus (vs.) MG. In 19 culture-negative cases the MG detected microbes in 17 patients. In total MG identified a causative bug for UTI in 67 of 69 patients. There was a MM in 34 of 49 patients (p = .0013). In contrast, there were 6 cases with PM and 9 with CM. In 16 patients with MM, MG was able to identify bacteria after C&S revealed a low bacterial load. In culture negative cases, VD detected microbes in 5 of cases, but failed to detect bacteria in 4 culture positive cases. There were 39 cases that compared the bacterial detection of VD and MG. MG was able to detect more microorganisms when compared to VD (127 vs. 51; p-value=.000109). There were 5 cases where VD was negative, but MG-positive. From 10 cases comparing C&S and MD in patients MD was able to detect more bacteria compared to C&S. In 18 cases comparing MD with MG, MG was able to detect more bacteria compared to MD (65 vs. 22; p-value = .001055). There were 4 cases where MD cultures were unable to detect bacteria, but MG could. In 14 cases there was no significant difference between the number of bacteria detected between MD and VD.
Conclusions: Our results indicate that MG provides more accurate data for determining the etiology of cUTI. MG sequencing is superior to C&S, VD and MD in terms of pathogen detection. However, the high sensitivity of MG with detection of few pathogens including commensals should more clearly define the most aggressive one.