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

Sir, In a recent issue of the Journal, Conti et al. concluded that so-called personalized antifungal susceptibility testing might be superior to the standardized reference method in terms of predicting the outcome of antifungal therapy. They rightly point out that a wide range of factors can significantly affect the reproducibility of the test method. In this sense, antifungal agents are no different to antibacterials. The findings of Conti et al. that the in-vitro activities of antifungal agents when determined in plasma are markedly different from those determined with a synthetic medium (RPMI 1640) come as no surprise, similar observations having been reported by several investigators in respect of antibacterials. With regard to the latter, it was shown, albeit in only a few instances, that the reduced activities of these drugs in body fluids influence their clinical efficacy. We have evaluated the activities of amphotericin B and fluconazole, alone and in combination, both in a synthetic medium (broth) and in human blood and showed that fluconazole, at certain concentrations, was fungistatic when susceptibility testing was carried out in broth, but fungicidal, at the same concentrations, when tested in blood. This might explain why the results of in-vitro testing failed to predict the efficacy of fluconazole in vivo. For amphotericin B, we observed a marked decrease in its activity in blood which could be accounted for by the binding of the drug to the outer membranes of the cellular components of blood and the consequent reduction in the concentrations of free drug.