A Simple, Rapid, and Highly Reliable Capsule-based14C Urea Breath Test for Diagnosis ofHelicobacter pyloriInfection

Publisher Website
Google Scholar
Abstract
Background: Since the urea breath test (UBT) indirectly detects gastric Helicobacter pylori infection by measuring urease activity, the possibility of false-positive results due to other urease-producing bacteria cannot be excluded. Previous studies have shown that increased 14CO2 activity in early breath samples could be attributed to urea hydrolysis in the oropharynx. For that reason, reliable assessment of H. pylori status is hampered for at least 20 min after administration of a 14C-urea drink. Methods: To overcome this problem, we have developed a modified breath test in which 111kBq 14C-urea is supplied in a gelatin capsule, which prevents release of 14C before reaching the stomach. Our modified 14C UBT was evaluated in 100 healthy volunteers, and results were compared with those from enzyme-linked immunosorbent assay serology. Results: The study showed a 99% concordance between the two non-invasive tests. When a biometric method for determination of cut-off values between positive and negative UBT results with the smallest possible arbitrariness was used, the calculated statistical probability of a false diagnosis was lowest in the 10-min breath sample (0.20%), and 100% sensitivity and specificity was achieved. Our capsule method was also compared with the urea drink method and was found more reliable because no overlapping in 14CO2 activity occurred between H. pylori-positive and -negative subjects, whereas conventional breath testing showed overlapping during the whole 30-min test period. Our study also showed that a fatty test meal lowers the 14CO2 excretion the first 20 min and may adversely affect the accuracy of a rapid UBT. Conclusions: Supplying the 14C-urea in a capsule obviates the problem of false-positive results in early breath samples and makes it possible to diagnose H. pylori infection with 99.8% reliability from a single 10-min breath sample, without the use of a test meal or adjustments for assumed individual CO2 production.