F C Martin1, G Chenevix-Trench2, N D Yeomans3,4. 1. Department of Medicine, University of Melbourne, Austin Health, Heidelberg, Vic., Australia. felicitycmartin@gmail.com. 2. QIMR Berghofer Medical Research Institute, Herston, Qld, Australia. 3. Department of Medicine, University of Melbourne, Austin Health, Heidelberg, Vic., Australia. 4. School of Medicine, Western Sydney University, Sydney, NSW, Australia.
Abstract
BACKGROUND: A causal association between proton pump inhibitor (PPI) use and fundic gland polyps has been suggested, but the data are conflicting. AIM: To clarify the relationship through a meta-analysis of the existing data. METHODS: A systematic retrieval and selection of records was performed. The main inclusion criteria were original studies reporting the prevalence of fundic gland polyps in PPI users or the reverse, compared to controls. Key outcomes were the odds ratios (OR) for fundic gland polyp prevalence in association with PPI use, prevalence of PPI use amongst subjects with fundic gland polyps and fundic gland polyp prevalence among PPI users. Statistical analysis was performed using Mix 2.0 Pro. RESULTS: The initial search using electronic databases and manual searching retrieved 339 peer-reviewed articles and abstracts. Twenty articles met all inclusion and exclusion criteria, with a total of 40 218 subjects included. The meta-analysis of 12 studies revealed an increase in fundic gland polyps amongst PPI users compared to controls (OR 2.46, 95% CI 1.42-4.27, P = 0.001), particularly among individuals taking PPIs for at least 6 months (OR: 4.71, 95% CI 2.22-9.99, P < 0.001) or 12 months (OR: 5.32, 95% CI 2.58-10.99, P < 0.001). CONCLUSIONS: Proton pump inhibitor usage is associated with a significantly increased prevalence of fundic gland polyps, and there is a trend for this to increase with longer length of PPI exposure. However, the meta-analysis is limited mainly to cohort studies.
BACKGROUND: A causal association between proton pump inhibitor (PPI) use and fundic gland polyps has been suggested, but the data are conflicting. AIM: To clarify the relationship through a meta-analysis of the existing data. METHODS: A systematic retrieval and selection of records was performed. The main inclusion criteria were original studies reporting the prevalence of fundic gland polyps in PPI users or the reverse, compared to controls. Key outcomes were the odds ratios (OR) for fundic gland polyp prevalence in association with PPI use, prevalence of PPI use amongst subjects with fundic gland polyps and fundic gland polyp prevalence among PPI users. Statistical analysis was performed using Mix 2.0 Pro. RESULTS: The initial search using electronic databases and manual searching retrieved 339 peer-reviewed articles and abstracts. Twenty articles met all inclusion and exclusion criteria, with a total of 40 218 subjects included. The meta-analysis of 12 studies revealed an increase in fundic gland polyps amongst PPI users compared to controls (OR 2.46, 95% CI 1.42-4.27, P = 0.001), particularly among individuals taking PPIs for at least 6 months (OR: 4.71, 95% CI 2.22-9.99, P < 0.001) or 12 months (OR: 5.32, 95% CI 2.58-10.99, P < 0.001). CONCLUSIONS: Proton pump inhibitor usage is associated with a significantly increased prevalence of fundic gland polyps, and there is a trend for this to increase with longer length of PPI exposure. However, the meta-analysis is limited mainly to cohort studies.