BACKGROUND: Serology is a noninvasive diagnostic method for the detection of Helicobacter pylori infection. Many commercial kits are now on the market. It is necessary to assess their performances to help the user to choose the most appropriate. MATERIAL AND METHODS: The performances of 29 commercial serological tests detecting antibodies to Helicobacter pylori (17 enzyme-linked immunosorbent assay and 12 near-patient tests) were evaluated using sera from 108 patients prospectively selected from gastroenterology departments of five French hospital centers. These patients were infected (45) or uninfected (47) by H. pylori, or had doubtful results (16), according to the gold standard (culture or histology plus rapid urease test or urea breath test). The tests were evaluated by determining the usual parameters of performance: sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Two analyzes were performed including or not the 16 patients with doubtful infection as uninfected or not analyzed. RESULTS: Depending on the type of analysis, four or two of the 17 enzyme-linked immunosorbent assay tests presented excellent results with the five performance parameters >90%. Calculation of the Youden index allowed to show significantly better performances for one of the 4. Performances of the 12 near-patient tests were lower with accuracies <90% for all except one test. CONCLUSION: These data should help the users to choose the kit the most appropriate to their goals.
BACKGROUND: Serology is a noninvasive diagnostic method for the detection of Helicobacter pylori infection. Many commercial kits are now on the market. It is necessary to assess their performances to help the user to choose the most appropriate. MATERIAL AND METHODS: The performances of 29 commercial serological tests detecting antibodies to Helicobacter pylori (17 enzyme-linked immunosorbent assay and 12 near-patient tests) were evaluated using sera from 108 patients prospectively selected from gastroenterology departments of five French hospital centers. These patients were infected (45) or uninfected (47) by H. pylori, or had doubtful results (16), according to the gold standard (culture or histology plus rapid urease test or urea breath test). The tests were evaluated by determining the usual parameters of performance: sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Two analyzes were performed including or not the 16 patients with doubtful infection as uninfected or not analyzed. RESULTS: Depending on the type of analysis, four or two of the 17 enzyme-linked immunosorbent assay tests presented excellent results with the five performance parameters >90%. Calculation of the Youden index allowed to show significantly better performances for one of the 4. Performances of the 12 near-patient tests were lower with accuracies <90% for all except one test. CONCLUSION: These data should help the users to choose the kit the most appropriate to their goals.
Authors: Luca Formichella; Laura Romberg; Christian Bolz; Michael Vieth; Michael Geppert; Gereon Göttner; Christina Nölting; Dirk Walter; Wolfgang Schepp; Arne Schneider; Kurt Ulm; Petra Wolf; Dirk H Busch; Erwin Soutschek; Markus Gerhard Journal: Clin Vaccine Immunol Date: 2013-09-04