Accuracy of diagnostic tests for Helicobacter pylori: a reappraisal.

ABSTRACT

BACKGROUND: Despite many changes, no large studies comparing the different diagnostic tests for Helicobacter pylori have been performed in the past 10 years. In this time, monoclonal stool antigen immunoassays and in-office 13C-urea breath tests (UBTs) have appeared. The aim of this study was to evaluate the accuracy of invasive and noninvasive tests in a large series of dyspeptic patients. METHODS: A total of 199 dyspeptic patients who had not previously been treated for H. pylori infection were prospectively enrolled. Noninvasive analyses included a commercial infrared-based UBT and a commercially available stool test. Biopsy-based tests included histological examination and a rapid urease test. A patient was considered to be infected when at least 2 test results were positive. Sensitivity, specificity, positive and negative predictive values, and 95% confidence intervals were calculated. The test results were compared using the McNemar test. RESULTS: Rates of positive test results were similar (54%) for the rapid urease test, histopathological examination, and the stool test. By contrast, 75% of UBT results were positive, and the UBT was associated with a very low specificity (60%). For this reason, the delta cutoff value for the UBT was recalculated as 8.5%. Sensitivities and specificities with this new cutoff value were 95% and 100%, respectively, for the rapid urease test; 94% and 99%, respectively, for histopathological examination; 90% and 93%, respectively, for the stool test; and 90% and 90%, respectively, for the UBT. CONCLUSIONS: Histological examination and rapid urease testing showed excellent diagnostic reliability. The stool test seems to be a good, noninvasive alternative to endoscopy-based tests. By contrast, the infrared-based UBT evaluated in our study showed a lower than expected performance, which was partially corrected when the cutoff value for the test was recalculated.