Objective: The aim of this study was to determine the influence of nitroimidazole resistance (NIR) on the efficacy of treatment for Helicobacter pylori (H. pylori) infections by meta-analysis of the world literature.
Methods: A MEDLINE search, a manual search of all major gastroenterological journals from 1993 to 1997, and abstracts of gastroenterological and H. pylori meetings from 1993 to 1997 were performed. All treatment studies using a nitroimidazole and providing data about the medication used, dose frequency, total daily dose, duration of treatment, and eradication results in relation to NIR were included. Eradication had to be assessed by two biopsy-based tests or a urea breath test > or = 4 wk after treatment. Individual studies were pooled into groups according to the medication used and the duration of treatment. The pooled estimate of the odds ratio (OR) of NIR for treatment failure and its 95% confidence interval (95% CI) were calculated for each group using the logit method. To detect any possible bias, funnel plots (plots of effect estimates against sample size) were constructed.
Results: A total of 91 treatment arms, including a total of 4823 patients, were evaluated. The pooled ORs of NIR for treatment failure (95% CI) of proton pump inhibitors, bismuth, and quadruple regimens were 5.2 (3.8-7.1), 5.9 (4.1-8.3), and 7.0 (3.1-16.0), respectively. Eradication rates were 90% in susceptible strains but <75% in resistant strains. In susceptible strains, neither treatment duration nor the choice of the second antibiotic influenced efficacy. In resistant strains, tetracycline was more effective than amoxicillin (bismuth regimens), and the longer the duration of regimens (bismuth-amoxicillin regimens) the more effective they were. Only quadruple regimens given for > or = 1 wk were effective in resistant strains.
Conclusions: NIR decreases treatment efficacy. Treatment duration and choice of other drugs influence the impact of NIR on treatment efficacy. If NIR is present, a nitroimidazole-containing regimen should be avoided or a quadruple regimen should be given for > 1 wk.