Metformin linked to fewer heart failure hospitalisation in patients with diabetes

  • Tseng CH
  • J Am Heart Assoc
  • 5 Nov 2019

  • International Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.


  • Even with propensity matching, metformin ever-use in patients with type 2 diabetes (T2D) is associated with a dose-dependent reduced risk of hospitalisation for heart failure (H[HF]) vs metformin never-users.
  • Metformin should always be considered first-line treatment for T2D.  

Why this matters

  • Evidence suggests metformin is underprescribed in patients with HF.

Study design

  • Population-based retrospective cohort study using Taiwan’s nationwide administrative database of newly diagnosed patients with T2D during 1999-2005, including unmatched 172,542 metformin ever-users vs 43,744 never-users and propensity score-matched pairs of 41,714 each of users and never-users, with follow-up through 2011.
  • Funding: Taiwan Ministry of Science and Technology, National Science Council, Yee Fong Charity Foundation.

Key results

  • Incidences of HHF for ever- vs never-users:
    • Unmatched cohort: 304.15 vs 864.31 per 100,000 person-years (HR, 0.350; P<.0001>
    • Matched cohort: 469.66 vs 817.01 per 100,000 person-years (HR, 0.571; P<.0001>
  • Results showed consistent dose-responsive reduction in HHF risk with metformin use, especially with cumulative therapy duration >2.5 years.
  • In propensity score-matched cohort created by excluding 10 potential instrumental variables, ever- vs never-users still had significantly lower risk (HR, 0.619; P<.0001>


  • Potential residual confounders.
  • Retrospective data.
  • Cannot use placebo with gold-standard T2D drug.

Please confirm your acceptance

To gain full access to GPnotebook please confirm:

By submitting here you confirm that you have accepted Terms of Use and Privacy Policy of GPnotebook.