Diagnostic accuracy of urine-based kits for detection of Helicobacter pylori antibody in children

Background Rapid urine-HpAb is reported to be a reliable test of Helicobacterpylori infection in adults, but there are no data on the application of the test in children. The aim of this study was to evaluate the accuracy of a urine-based enzyme-linked immunosorbent assay (urine-HpELISA) and immunochromatography (rapid urine-HpAb) kit for anti-H.pylori immunoglobulin G antibody in children. We compared its sensitivity and specificity in reference to the 13C-urea-breath test (UBT) and H.pylori stool antigen test (HpSA). Methods In total, 101 Japanese children without significant upper-abdominal symptoms were included (mean age, 7.1 years; range 2-15 years). Their sensitivity and specificity were evaluated in reference to the UBT and HpSA. Results Thirty-seven children were judged H.pylori-positive and 64 negative by the UBT and HpSA. No discrepancy in the results was observed between UBT and HpSA. Urine-HpELISA showed 91.9% sensitivity and 96.9% specificity with an accuracy of 95.0%. Rapid urine-HpAb showed 78.4% sensitivity and 100% specificity with an accuracy of 92.1%. Seven false negative results for rapid urine-HpAb were from children aged younger than 10 years, and their antibody titers of urine-HpELISA were lower than true positives. Conclusions For the diagnosis of H.pylori infection in Japanese children, both tests are non-invasive, inexpensive, reliable and easy-to-perform methods giving satisfactory accuracy, although the sensitivity of the rapid urine-HpAb kit was inferior to that of the urine-HpELISA kit, especially in children aged younger than 10 years, showing relatively low titer of H.pylori antibody.