Diagnostic Accuracy of a Noninvasive Test for Detection of Helicobacter pylori and Resistance to Clarithromycin in Stool by the Amplidiag H. pylori+ClariR Real-Time PCR Assay

Abstract

The noninvasive detection of Helicobacter pylori and its resistance to clarithromycin could revolutionize the management of H. pylori-infected patients by tailoring eradication treatment without any need for endoscopy when histology is not necessary. Several real-time PCR tests performed on stools have been proposed, but their performances were either poor or they were tested on too few patients to be properly evaluated. We conducted a prospective, multicenter study including 1,200 adult patients who were addressed for gastroduodenal endoscopy with gastric biopsies and who were naive for eradication treatment in order to evaluate the performance of the Amplidiag H. pylori+ClariR assay recently developed by Mobidiag (Espoo, Finland). The results of the Amplidiag H. pylori+ClariR assay performed on DNA from stools (automatic extraction with the EasyMag system [bioMérieux]) were compared with those of culture/Etest and quadruplex real-time PCRs performed on two gastric biopsy samples (from the antrum and corpus) to detect the H. pyloriglmM gene and mutations in the 23S rRNA genes conferring clarithromycin resistance. The sensitivity and specificity of the detection of H. pylori were 96.3% (95% confidence interval [CI], 92 to 98%) and 98.7% (95% CI, 97 to 99%), respectively. The positive and negative predictive values were evaluated to be 92.2% (95% CI, 92 to 98%) and 99.3% (95% CI, 98 to 99%), respectively. In this cohort, 160 patients (14.7%) were found to be infected (positive by culture and/or PCR). The sensitivity and specificity for detecting resistance to clarithromycin were 100% (95% CI, 88 to 100%) and 98.4% (95% CI, 94 to 99%), respectively.

Keywords: Helicobacter pylori; PCR; diagnosis; gastric cancer; gastric ulcers; molecular diagnosis; noninvasive sampling; stool.

FIG 1

Study design and flowchart. Out of the 1,200 enrolled patients, after exclusion of 114 patients, the population (n = 1,086) was divided into two groups according to their H. pylori infection status (infected patients, n = 160; noninfected patients, n = 926).