Comparative effectiveness of first-line therapies for eradication of antibiotic-resistant Helicobacter pylori strains: A network meta-analysis

Figure 1 Study selection. Flow diagram of the study, including study screening, inclusion, and exclusion in this systematic review and network meta-analysis. RCTs: Randomized controlled trials; NRCT: Non-randomized controlled trial; CLA: Clarithromycin.

Figure 2 Network map for clarithromycin resistance. The node size is positively associated with the number of patients in each regimen, and the precision is proportional to the edge thickness, namely, the standard errors of each direct comparison. Vono: Vonoprazan; PPI: Proton pump inhibitor; HD: High-dose; QT: Quadruple therapy.

Figure 3 Network forest plots. A: Network forest plot [odds ratio (OR); 95% credible interval [CrI]) of all patients (intention-to-treat) showing the regimens compared directly with the proton pump inhibitor-triple therapy regimen as the reference; B: Network forest plot (OR; 95% CrI) of clarithromycin-resistant strains showing that the efficacy of the regimens compared with PPI-triple therapy was consistent with that in network forest plot A; C: Network forest plot of clarithromycin-susceptible strains describing the efficacy of the regimens compared directly with the PPI-triple therapy regimen as the reference; D: Network forest plot of metronidazole-resistant strains showing the efficacy of the regimens compared directly with the PPI-triple therapy reference regimen. CrI: Credible interval; ITT: Intention-to-treat; Vono: Vonoprazan; PPI: Proton pump inhibitor; HD: High-dose; QT: Quadruple therapy.

Figure 4 Forest plots of the league matrix in randomized controlled trials among all patients (A) and those infected with clarithromycin-resistant strains (B). A: Vonoprazan-dual therapy; B: Vonoprazan-triple therapy; C: Sequential therapy; D: Dual therapy containing a high-dose proton pump inhibitor; E: Proton pump inhibitor bismuth quadruple therapy; F: Proton pump inhibitor nonbismuth quadruple therapy; G: Proton- pump inhibitor triple therapy. RCTs: Randomized controlled trials; CrI: Credible interval.

Figure 5 Cumulative probability of rank in this network meta-analysis. Vono: Vonoprazan; PPI: Proton pump inhibitor; HD: High-dose; QT: Quadruple therapy.

Figure 6 Risk of bias graph and the funnel plot. A: Risk of bias graph displaying each item as the percentage in all studies; B: Funnel plot. The symmetrical appearance indicates the absence of publication bias or small study effects in the network; C: Network forest plot of all patients (intention-to-treat) in the subgroup with high cure rates (≥ 90%) showing that dual therapy containing a high-dose proton pump inhibitor (HD-PPI) and Vono-based therapy have advantages over PPI-triple therapy; D: Network forest plot of clarithromycin-susceptible H. pylori in the subgroup with high cure rates (≥ 90%) showing that HD-PPI-dual therapy was slightly superior to PPI-triple therapy. CrI: Credible interval; ITT: Intention to treat; Vono: Vonoprazan; PPI: Proton pump inhibitor; HD: High dose; QT: Quadruple therapy.