Identification of Urinary CD44 and Prosaposin as Specific Biomarkers of Urinary Tract Infections in Children With Neurogenic Bladders

Forster, Catherine S; Haffey, Wendy D; Bennett, Michael; Greis, Kenneth D; Devarajan, Prasad

We report conflicting results between the identification and validation arms of this work. While we report significant elevations in both prosaposin and CD44 in UTI compared with UTC, results from mass spectrometry suggest that these proteins are increased in UTC when compared with UTI. One possible explanation is the different methods of normalization. Samples were loaded into the mass spectrometer based on total protein concentration. While the literature is mixed on the presence of proteinuria in cystitis, there is data to suggest the presence of tubular proteinuria in pyelonephritis. 49 It is likely that most patients in the identification arm had pyelonephritis, as they were chosen based on their unequivocal presentation for UTI. Therefore, these patients likely had a higher degree of proteinuria, leading to a smaller amount of urine used for the analysis. Conversely, urine was loaded into the confirmatory assays based on volume, not concentration. Therefore, it is possible that tubular proteinuria in our patients with UTI led to these conflicting results. Indeed, our enzyme-linked immunosorbent assay (ELISA) results do display an increase in both prosaposin and CD44 in UTC when compared with children with no growth on urine cultures; this increase was further magnified in the UTI group with added specificity, suggesting their utility as specific markers of UTI in children with neurogenic bladders.