Background: The purpose of the study was to determine the accuracy of the labelled 13carbon urea breath test for the diagnosis of Helicobacter pylori in children and to simplify the 13carbon urea breath test in identifying the most discriminating sampling time.
Methods: H. pylori was searched for in 100 children aged 10.5+/-4.5 years by histology, bacteriological counts, and culture on antral biopsies together with serology and 13carbon urea breath test. Breath samples were obtained before ingestion (T0) of 75 mg urea-13C and every 10 minutes after until T60. 13CO2 excess ratio was measured by isotope ratio mass spectrometry, and values expressed as delta per mil over baseline enrichment (delta 13CO2). The arithmetic mean (Mdelta 13CO2) of T20 to T60 values was calculated and the test considered positive with Mdelta 3CO2 higher than Mdelta 13CO2 + 3 SD as determined in noninfected children.
Results: Mdelta 13CO2 of noninfected children as assessed by culture was 1.4+/-0.6 per mil, determining a positive cut-off value of 3.44 per mil. Mdelta 13CO2 was correlated in 11 children with biopsy bacteriological counts. Both culture and 13carbon urea breath test were positive in 38 of 100 children, without any discordance. Plotting 13carbon urea breath test results at each sampling time versus Mdelta 13CO2 showed weaker correlations at T20, T30, T50, and T60, than at T40. The two-sample method at T0 and T30, T40, T50, had high sensitivity and specificity. Single-sample analysis obtained at T40 gave a comparable sensitivity and a slightly reduced specificity.
Conclusion: 13carbon urea breath test is sensitive and specific in children. Two samples collected at T0 and T40 provide the most discriminating procedure.