Influence of Metformin Discontinuation on Readmission Rate in Patients With Acute Heart Failure

The consequences of discontinuing metformin in patients with heart failure have not been determined. Knowing that acute exacerbation of chronic heart failure contributes to substantial increases in major adverse cardiovascular events (MACE), we proposed a retrospective study to examine whether discontinuing metformin in patients hospitalized with heart failure impacts mortality and readmission rates.

We conducted a retrospective analysis of patients admitted with a diagnosis of acute heart failure to hospitals in the HCA Healthcare System from 2020 to 2022. Included patients had a prior diagnosis of diabetes mellitus, acute heart failure, and were taking metformin prior to admission. After applying our exclusion criteria, a total of 7740 patients remained. The primary outcomes were 30-, 60-, and 90-day readmission rates and secondary outcomes were mortality and length of stay.

Patients who were discharged without a prescription for metformin (NONDIS-MET) were 4.489 (95% CI 3.673-5.488, p < 0.0001) times more likely to have a MACE outcome in 30 days compared to patients who received a discharge order for metformin (DIS-MET). The findings were similar for 60-day and 90-day readmission rates, with NONDIS-MET patients 3.457 (95% CI 2.893-4.131, p < 0.0001) and 2.992 (95% CI 2.534-3.533 p < 0.0001) times more likely to have a MACE outcome than MET patients, respectively. However, when metformin was continued during the patients' hospital stay (CONT-MET) there was no significant association with MACE outcomes, readmission, or mortality rates.

We found that diabetic patients admitted with acute heart failure exacerbations had a higher incidence of major adverse cardiac events and were more likely to be readmitted when they were not prescribed metformin after discharge. Our findings agree with prior work showing the cardioprotective effects of metformin; however, continuing metformin during hospital admission did not affect our patients adverse outcomes.