Pediatric urinary tract infections (UTIs) are a common diagnosis and account for greater than 500,000 emergency department (ED) visits each year in the US. Although prevalence varies by age, gender, ethnicity and circumcision status, the overall prevalence is cited between 5% to 13.8%. UTIs also account for a large number of pediatric hospital admissions with up to 50,000 children admitted per year, 40% of which occurring in children under the age of 1 year. In children under the age of two, diagnosis of UTI is difficult as patients are unable to verbalize complaints and often present with only fever or other nonspecific symptoms such as vomiting. UTIs are the leading cause of occult bacteremia and serious bacterial infection in the 2-24 month age range.
A definitive diagnosis of UTI is made when a patient has pyuria and/or bacteriuria with a positive urine culture. For non-toilet trained children who have urine collected by either suprapubic aspiration (SPA) or bladder catheterization, a urine culture is considered positive if a single uropathogenic organism of at least 50,000 colony forming units (CFU) per mL is recovered. Identification of a uropathogen by culture can take up to 48 hours to be completed which delays treatment and risks progression of infection if antibiotics are not given empirically. Commonly, clinicians use urine microscopy and dipstick for quick diagnosis to enable early empiric treatment of UTIs. However, the components of the urinalysis are limited in sensitivity and specificity even for specimens of high quality such as those obtained by catheterization or suprapubic tap. The main components of urine dipstick testing are leukocyte esterase (LE) and nitrite levels in urine. The LE test has a sensitivity of 94% when combined with clinical suspicion for UTI, but a specificity of only 83%. False positives for UTI occur when LE is positive in the child who has pyuria for other reasons besides infection, for example, in urethritis or Kawasaki disease. False negatives can occur in young children because the pattern of more frequent urination allows less time to mount an adequate inflammatory response to bacteria. Nitrites have a low sensitivity in young children (53%) due to frequent voiding not allowing sufficient time (up to 4 hours) for production of nitrites by bladder-dwelling bacteria; however, specificity is high (98%) making this a good test to rule in UTI. Sensitivity and specificity of urine microscopy for WBCs and bacteria are worse than for LE with pyuria (5 or more WBC/high power field on a spun sample or \>10 WBC/hpf on an un-spun sample) having a sensitivity and specificity of 73% and 81% respectively. Presence of bacteria on microscopy has a sensitivity and specificity of 81% and 83% respectively. The resulting over- and under-treatment for false positive and negative urinalyses leads to increased costs, antibiotic resistance and unnecessary adverse events from antibiotics. The inaccuracy of the traditional urinalysis on bagged urine samples is much greater than for catheter specimens. These specimens suffer from false positives results for leukocyte esterase and cultures at rates up to 75-88% because of contamination by urogenital flora and white blood cells present on the perineum.
As a result of the poor performance characteristics of routine urinalyses on bag specimens, a sample obtained by catheterization in children who are not yet toilet-trained is required even for UTI screening. This is invasive and painful for the child, causes anxiety for the family, leads to urethral trauma and can even introduce bacteria into the urinary system. Recent research describes a new two-step technique of obtaining urine samples of screening for UTI on these young patients. First, a bagged urine sample is taken, and if suspicious for UTI by urinalysis, the patient then is catheterized for culture. One study showed that this technique decreased the number of catheterizations by 33%; however, given the low specificity of the urinalyses on bag specimens, there is still substantial room for reductions in unnecessary bladder catheterizations.
New urinary biomarkers hold promise to improve screening accuracy for UTIs. Watson et al. investigated antimicrobial peptides (AMPs) as urinary biomarkers of UTIs in children. One specific AMP, derived from the epithelium, is human alpha-defensin 5 (HD5). This AMP is produced by intestinal Paneth cells, the female genital tract and the uroepithelium. HD5 was found to have mean concentration significantly higher in the urine from culture-positive compared to culture-negative specimens. When used in combination with LE, HD5 increased the specificity of the LE-only test by 6% without decreasing the sensitivity (97% for LE \> trace). This study determined the optimal cut-off value of 174 pg/mg creatinine to declare sample positive. Because HD5 is not derived from white blood cells, its presence is not affected by the presence or absence of WBCs in the urine to accurately identify urinary inflammation. While the authors of this study examined results from both clean-catch and catheterized specimens, they did not analyze samples collected by both techniques from the same patient.
Identification and validation of biomarkers with improved performance characteristics could reduce over-treatment of UTIs; validating such biomarkers on bag urine specimens has the further benefit of reducing the need for catheterization to provide urine to screen for UTI in a relatively unselected population of young children who have fever without a focus. We propose to compare the use of HD5 in bag versus catheter urine samples on children 1 to 24 months of age to determine if the biomarker has similarly high sensitivity and specificity to the original study in both specimen types. Eliminating the need for urine culture would also save on costs for children whose rapid testing does not suggest UTI. Reducing the number of catheterizations performed on young children will decrease the number of painful procedures, prevent introduction of bacteria in the urinary tract and decrease parental anxiety.
