Improving 13C-urea breath test performance metrics for diagnosis of Helicobacter pylori infection

Abstract

The 13C-Urea Breath Test (13C-UBT) is a popular, non-invasive method used for the diagnosis of Helicobacter pylori infection. This work evaluates its performance for the initial diagnosis and post-treatment follow-up in dyspeptic adults from Buenos Aires, Argentina. We retrospectively analyzed data from two earlier studies, which evaluated H. pylori infection using 13C-UBT and histology of gastric biopsies. Additionally, we assessed the 13C-UBT performance against the concordant results of both histology and PCR in a subsample with available data. The 13C-UBT was performed using a commercial kit, with isotope-ratio-mass-spectrometry (IRMS) as the measurement technique. Results from 154 volunteers were evaluated to assess the performance of 13C-UBT for the initial diagnosis of H. pylori infection, with histological evaluation as the reference method. For a cut-off value set at 3.5‰, sensitivity was 93.0%, specificity was 95.6%, accuracy was 94.2%, positive predictive value (PPV) was 96.4% and negative predictive value (NPV) was 91.5%. The subsample analysis of 13C-UBT vs. histology and PCR showed improved results: sensitivity 98.8%, specificity 98.3%, accuracy 98.6%, PPV 98.8% and NPV 98.3%. In contrast, 13C-UBT performance for confirming H. pylori eradication was studied in 46 patients, showing a sensitivity of 94.4%, specificity 100.0%, accuracy 97.8%, PPV 100.0% and NPV 96.6%. The analysis of sensitivity, specificity and accuracy as a function of the cut-off revealed that the optimal value could be lowered to 3.0‰ in our laboratory. These results demonstrate that the 13C-UBT is a non-invasive, highly accurate method for both the initial and post-treatment diagnosis of H. pylori infection.