Comparison of three methods for the determination of the sensitivity of Helicobacter pylori to metronidazole.

A comparison of various methods for the determination of the sensitivity of Helicobacter pylori to metronidazole was undertaken. The validity of the agar dilution, the disc diffusion and the Epsilometer (E) test was studied using a total of 86 strains, 16 of which were known to be resistant to metronidazole. All tests were carried out on Mueller-Hinton agar with 5% sheep blood. The results of the disc diffusion and the E-test were highly significantly (P < 0.001) associated with those of the agar dilution test, which was taken as a standard (r = -0.96 and r = 0.96, respectively). Investigation of the accuracy of the repetitive (n = 10) testing ten sensitive and one resistant strains, showed that the disc diffusion test led to a systematic and significant (2p < 0.001) underestimation of the MIC values calculated from the inhibition zone diameters via linear regression. In contrast, the results of the E-test did not differ significantly from those of the agar dilution test. The precision of the agar dilution test was significantly worse (2p < 0.005) than the E-test and the disc diffusion test. Because of its accuracy and significantly better precision, the E-test is recommended as the best and simplest method for routine antibiotic sensitivity testing of H. pylori.