Setting the standard for determining the in-vitro susceptibility of Helicobacter pylori to metronidazole

J Antimicrob Chemother 1999; 44: 291–293 Sir, Metronidazole is frequently included in regimens used to eradicate Helicobacter pylori. However, the rates of metronidazole resistance amongst H. pylori isolates are highly variable—from 11% to 70% in developed countries and even higher in developing countries. 1 As susceptibility testing of H. pylorihas not been standardized, it may be that variations in the rates of resistance can be accounted for by differences in the methods of determining susceptibility. For example, Smith & Edwards 2 demonstrated that H. pylori NCTC 11637 is resistant to metronidazole when tested under microaerophilic growth conditions, but susceptible when tested under anaerobic conditions. This raises an important question: are strains which are resistant to metronidazole when in-vitro susceptibility testing is carried out under microaerophilic conditions also resistant in vivo? The methodology of determining the susceptibility of this bacterium to metronidazole is crucial to the ability to predict the efficacy of eradication therapy. The present study was therefore undertaken to evaluate variations in methodology on the outcome of susceptibility testing. The MICs of metronidazole for 21 clinical isolates of H. pylori were determined by the Etest method (AB Biodisk, Solna, Sweden) 3 according to the manufacturer's instructions; the medium used was Columbia agar (Oxoid, Basingstoke, UK) supplemented with 7% horse blood and H. pylori NCTC 11637 was included as a control. The inocula were prepared by suspending colonies from 48 h growth on Columbia blood agar in Brain Heart Infusion (BHI) broth (Oxoid) and adjusting the suspensions to give turbidities equivalent to that of a No. 3 McFarland standard. The plates were incubated at 37°C in a microaerophilic atmosphere (Campylobacter System Gas Generating Kit, Oxoid) and MICs were recorded after 72 h. The majority (12) of the 21 strains, as well as the control, were resistant to metronidazole on the basis of a lower MIC breakpoint of ≥0.5 mg/L recommended by AB Biodisk; the MICs for five strains were ≥256 mg/L. The effects of varying the conditions of susceptibility testing were evaluated with the control strain which was shown by us to be highly resistant to metronidazole (MIC ≥256 mg/L), although other investigators have reported it to be susceptible. 4,5 Firstly, the inoculated Etest plates were incubated under anaerobic conditions for 24 h before being incubated under microaerophilic conditions for a further 72 h. With the introduction of a pre-incubation phase, the MIC for H. pylori NCTC 11637 fell to 0.016 mg/L which is below the breakpoint defining susceptibility. This suggests that metronidazole is activated, i.e. a nitro group is reduced to a radical anion which damages DNA, 6 only under strictly anaerobic conditions. The requirement for an anaerobic atmosphere of incubation for the activation of metronidazole may explain why some centres have reported high rates of resistance to this drug amongst H. pylori isolates. The next step was to compare the outcome of incubating H. pylori NCTC 11637 in an anaerobic jar with that in an anaerobic chamber before the plates were incubated in a microaerophilic atmosphere. Either no growth or only poor growth was observed following pre-incubation in an anaerobic chamber (Don Whitley Scientific Ltd, Shipley, UK), whereas the strain grew well when an anaerobic jar was used, suggesting that growth of H. pyloridepends on a high level of humidity which is not provided in an anaerobic chamber. Finally, we also investigated the effects of varying the medium, inoculum size and duration of incubation on the MICs of metronidazole for the control strain. The media evaluated were Columbia and Mueller–Hinton agar (Oxoid), both supplemented with 7% horse blood. Colonies from a 48 h culture were suspended in BHI broth and the turbidities of the suspensions were adjusted so that they were equivalent to that of either a 0.5 or a No. 3 McFarland standard. The inoculated plates were then incubated in an anaerobic atmosphere for 24 h, followed by incubation for either 48 h or 72 h in a microaerophilic atmosphere. Growth on Mueller–Hinton agar was poor, while that on Columbia agar was profuse. The MICs were difficult to interpret when the lower inoculum was used, but were clear and reproducible with the higher inoculum. After 48 h of incubation, it was not possible to record the MICs accurately because growth was too poor. On the other hand, following incubation for 72 h, the MICs were easy to read and unequivocal. Susceptibility testing of all 21 H. pylori isolates to metronidazole was then repeated under optimal conditions, i.e. pre-incubation of Columbia agar plates inoculated with the higher inoculum for 24 h in an anaerobic jar, followed by incubation for 72 h in a microaerophilic atmosphere. Under these conditions, the MIC50 and MIC90 were 0.016 mg/L and 0.19 mg/L respectively; only one strain was truly resistant to metronidazole (MIC > 256mg/L). We have identified the optimal conditions for determining the susceptibility of H. pylori to metronidazole. In particular, pre-incubation under anaerobic conditions, ideally in an anaerobic jar, is regarded by us as being essential if accurate results are to be obtained. Our interpretation of the findings of this study is that H. pylori strains are inherently susceptible to metronidazole—a view that is in contrast with that of earlier investigators. 2 If the drug is pre-reduced, as it must be when undertaking susceptibility testing with other bacteria, then most H. pylori isolates will appear susceptible to it. It has been suggested previously that H. pylori is resistant to metronidazole when susceptibility is determined under microaerophilic conditions. We believe that, on the basis of current evidence, this conclusion is unwarranted as there is no means of confirming that the conditions that prevail during susceptibility testing are the same as those at the site of infection. None the less, the success achieved with metronidazole when it is used as treatment of patients with H. pylori infection suggests that most strains are susceptible to this drug, as our susceptibility test results indicate.