Home / Resources / Articles / Preventing H. Pylori Infections with Metformin

Preventing H. Pylori Infections with Metformin

May 26, 2018

Preventing H. Pylori infections through the use of metformin may decrease chances of developing gastric cancer in patients with type 2 diabetes.

The association between gastric cancer and diabetes has been an interesting school of thought that was reported by researchers in the past. Specifically, one scientist, Chin-Hsoao Tseng, has reported on the subject time and time again. His latest study published in Diabetes Care at the beginning of 2018 may provide us with some answers that can lead to advances in treating gastric cancer in diabetes. Back in 2014, Tseng conducted a meta analysis that yielded the positive link between diabetes and gastric cancer. While the relationship between the two was established, the mechanisms responsible were not clear. An array of risk factors associated with diabetes, medication use, comorbidities and H. pylori infection were some of the culprits listed that led to an increased incidence of gastric cancer in patients with diabetes. H. pylori infection, specifically, is thought to occur more frequently in persons with diabetes, it is harder to eradicate, and it re-occurs at a higher rate than it would in individuals without diabetes.   To explore whether antidiabetic medications can provide protective effects against H. pylori infection, and therefore, gastric cancer, Tseng conducted a cohort study in Taiwan.

The Reimbursement Database of Taiwan’s National Health Insurance was employed to search for eligible subjects to be included in the study, from the years of 1999 to 2005.  Individuals who were diagnosed with diabetes and H. pylori infections were determined by diagnostic codes. Persons who were newly diagnosed with diabetes were separated into two groups: those who were treated with metformin at diagnosis, and those who were treated with other antidiabetic drugs. Type 1 diabetes, cancer diagnosis, and a history of H. pylori infection prior to enrollment into the study were some of the exclusion criteria listed. Propensity score was calculated and later used to estimate the hazard ratios. Subjects were followed from the initial inclusion until the end of 2011.

There were 16,060 subjects in each of the cohort arms: metformin users and metformin non-users. At the end of the study, the incidence of H. pylori infection was found in 156 metformin non-users and 110 metformin users. When translated into incidence rate per 100,000 person-years, incidence was 217.51 in non-users and 146.88 in metformin users. Furthermore, the hazard ratio was 0.670 favoring the metformin user group, p-value of 0.0012. When classifying the duration of metformin therapy to incidence of H. pylori infections, Tseng found that the longer the length of therapy, the higher its protective effects. The group of individuals who used metformin for over 58 months had the least incidence rate of H. pylori, which was statistically significant as seen with HR of 0.307 and p-value < 0.0001. Subjects who used metformin for less than 58 weeks had a smaller incidence rate of H. pylori infection, but the decrease was not statistically significant. Individuals treated with metformin for 26.33 to 58 months had an incidence rate of 196.03 with p-value of 0.4492, while persons treated for less than 23.33 months had an incidence of 212.48 and p-value of 0.9605.

In conclusion, metformin use in individuals newly diagnosed with diabetes leads to a significant decrease in incidence of H. pylori infections. The beneficial effects were more pronounced with a longer duration of therapy. Moreover, decreasing the incidence of H. pylori infection will possibly decrease the development of gastric cancer as well, but that correlation warrants more exploration. Taking into consideration the design of the trial, one must have to stay aware of the potential confounding bias that is likely to occur with cohort studies, such as genetic components, lifestyle habits, education levels, etc.  Although this study was a small study that included only patients of one population –— Taiwan — results reported by Tseng should serve as a base to explore the relationship between gastric cancer, H. pylori infection, and metformin use in larger clinical trials in order to uncover the relationship that will ultimately lead to reductions in gastric cancer rates in patients with type 2 diabetes.

Practice Pearls:

  • Metformin users were less likely to develop H. pylori infection than non-users of metformin, HR =0.670, p-value 0.0012.
  • The longer the duration of metformin therapy, the more protective benefits against H. pylori infection were seen.
  • Larger clinical trials are needed to establish if metformin therapy will ultimately reduce the incidence of gastric cancer in individuals with T2D.


Chin-Hsiao Tseng.  “Metformin and Helicobacter Pylori Infection in Patients with Type 2 Diabetes.”  Diabetes Care. 2018. http://care.diabetesjournals.org/content/diacare/early/2018/01/26/dc17-2551.full.pdf.  Accessed Jan 2018.Aging (Albany NY). 2016. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5032687/.


Lamija Zimic, PharmD(c), University of South Florida, College of Pharmacy