Systematic review, meta‐analysis, and meta‐regression: Successful second‐line treatment for Helicobacter pylori

• Published in: Helicobacter (Cambridge, Mass.) Mass.), 2018-06, Vol.23 (3), p.e12488-n/a
• Main Authors: Muñoz, Neus, Sánchez‐Delgado, Jordi, Baylina, Mireia, Puig, Ignasi, López‐Góngora, Sheila, Suarez, David, Calvet, Xavier
• Format: Article
• Language: English
• Subjects: Anti-Bacterial Agents - pharmacology; Anti-Bacterial Agents - therapeutic use; Anti-Ulcer Agents - pharmacology; Anti-Ulcer Agents - therapeutic use; Antibiotics; Comparative studies; Drug Therapy, Combination; Extreme values; Helicobacter Infections - drug therapy; Helicobacter pylori; Helicobacter pylori - drug effects; Humans; Infectious diseases; Meta-analysis; Proton Pump Inhibitors - pharmacology; Proton Pump Inhibitors - therapeutic use; Randomized Controlled Trials as Topic; Regression analysis; Reviews; Salvage Therapy - statistics & numerical data; Schedules; second‐line treatments; Statistical analysis; Systematic review; Treatment Outcome
• ISSN: 1083-4389; 1523-5378
• DOI: 10.1111/hel.12488

Background Multiple Helicobacter pylori second‐line schedules have been described as potentially useful. It remains unclear, however, which are the best combinations, and which features of second‐line treatments are related to better cure rates. The aim of this study was to determine that second‐line treatments achieved excellent (>90%) cure rates by performing a systematic review and when possible a meta‐analysis. A meta‐regression was planned to determine the characteristics of treatments achieving excellent cure rates. Methods A systematic review for studies evaluating second‐line Helicobacter pylori treatment was carried out in multiple databases. A formal meta‐analysis was performed when an adequate number of comparative studies was found, using RevMan5.3. A meta‐regression for evaluating factors predicting cure rates >90% was performed using Stata Statistical Software. Results The systematic review identified 115 eligible studies, including 203 evaluable treatment arms. The results were extremely heterogeneous, with 61 treatment arms (30%) achieving optimal (>90%) cure rates. The meta‐analysis favored quadruple therapies over triple (83.2% vs 76.1%, OR: 0.59:0.38‐0.93; P = .02) and 14‐day quadruple treatments over 7‐day treatments (91.2% vs 81.5%, OR; 95% CI: 0.42:0.24‐0.73; P = .002), although the differences were significant only in the per‐protocol analysis. The meta‐regression did not find any particular characteristics of the studies to be associated with excellent cure rates. Conclusion Second‐line Helicobacter pylori treatments achieving>90% cure rates are extremely heterogeneous. Quadruple therapy and 14‐day treatments seem better than triple therapies and 7‐day ones. No single characteristic of the treatments was related to excellent cure rates. Future approaches suitable for infectious diseases—thus considering antibiotic resistances—are needed to design rescue treatments that consistently achieve excellent cure rates.