Comparative Effectiveness of Multiple Different First-Line Treatment Regimens for Helicobacter pylori Infection: A Network Meta-analysis

Abstract

Background & aims: A number of double, triple, and quadruple therapies have been proposed as first-line empiric treatments for Helicobacter pylori infection. However, knowledge of their worldwide and regional comparative efficacy is lacking. We examined the comparative effectiveness of all empirically used first-line regimens tested against standard triple treatment using a network meta-analysis of published randomized controlled trials.

Methods: Data extracted from eligible randomized controlled trials were entered into a Bayesian network meta-analysis to investigate the comparative efficacy of H pylori infection empiric first-line regimens and to explore their effectiveness rank order. The ranking probability for each regimen was evaluated by means of surfaces under cumulative ranking values.

Results: Sixty-eight eligible randomized controlled trials were included, giving a total of 92 paired comparisons with 22,975 patients randomized to 8 first-line regimens. The overall results showed that only vonoprazan triple therapy and reverse hybrid therapy achieved cure rates of >90%. Levofloxacin triple therapy performed best in Western countries (eradication rate 88.5%). The comparative effectiveness ranking showed that vonoprazan triple therapy had the best results, whereas standard triple therapy was the least efficacious regimen (surfaces under cumulative ranking 92.4% vs 4.7% respectively; odds ratio, 3.80; 95% credible interval, 1.62-8.94).

Conclusions: For first-line empiric treatment of H pylori infection, vonoprazan triple therapy and reverse hybrid therapy achieved high eradication rates of >90%. Levofloxacin triple therapy achieved the highest eradication rates in Western countries. Standard triple therapy was the least efficacious regimen in this network meta-analysis.

Keywords: Efficacy; First-Line Regimens; Helicobacter pylori; Network Meta-Analysis; Treatment.