Home / Therapies / Metformin Therapy Center / Metformin and Kidney Function: A Balancing Act

Metformin and Kidney Function: A Balancing Act

Apr 3, 2021
Editor: Steve Freed, R.PH., CDE

Author: Louise Brown, PharmD Candidate, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences

Metformin is often associated with lactic acidosis, especially in patients with kidney disease. A recently published study questions this association in patients with mild to moderate disease. 

Regardless of how long a medication has been available or how frequently prescribed, benefit-risk assessments should be performed to ensure the right balance is maintained. As more data becomes available, unknown benefits or harms may emerge, positively or negatively impacting this balance1. When metformin was initially approved, there were safety concerns about its use in patients with kidney disease due to the risk of metformin-associated lactic acidosis. It was contraindicated in patients with a serum creatinine ≥ 1.5 mg/dL in males or ≥1.4 mg/dL in females.2,3 More recently, prospective studies conducted in patients with normal kidney function provided additional information on this risk. Patients with kidney disease were usually excluded from these safety studies. As a result, only a limited amount of data is available in this subset of patients.2,3 In 2016 the FDA revised metformin’s label to state that metformin is still contraindicated in patients with severe kidney disease, defined as eGFR <30 mL/min/1.73 m2, but did not include mild to moderate disease in this updated warning.4  


In an article recently published in Diabetes Care, Chu et al. present findings from their retrospective cohort study that compared hospitalizations due to lactic acidosis in U.S. veterans with T2DM who developed mild to moderate kidney disease and continued metformin or a sulfonylurea (S.U.).3 All subjects were ≥18 years and received routine care within the Veterans Health Administration (VHA). Study entry date was the day kidney function reached a prespecified threshold of eGFR < 60mL/min/1.73m2 or serum creatinine of ≥1.5 mg/dL for males or 1.4 mg/dL for females. Patients were excluded if a medication was added or switched, if they received dialysis or organ transplant, or enrolled in hospice. The follow-up period ended when a patient experienced lactic acidosis, a censored event such as the end of the study or no VHA contact, or a competing risk event such as death or nonadherence. The primary composite outcome included hospitalization with a direct discharge diagnosis code of lactic acidosis and hospitalization with laboratory evidence of lactic acidosis. Laboratory evidence was defined as lactic acid ≥2.5 mmol/L combined with either a pH <7.35 on arterial blood gas or a serum bicarbonate ≤19 mmol/L within 24 hours before or after hospital admission. A total of 174,882 new users reached the kidney threshold and were eligible for study entry. Propensity scores were used to weight patients on study covariates to create a more balanced cohort.  Cox proportional hazard modeling was used in this weighted cohort to estimate the relative difference in lactic acidosis risk between the two user groups.3 

The weighted cohort included 24,542 metformin and 24,662 SU users, and the median followup was approximately one year in both user groups. Most subjects were male (96.5%), white (82.6%), with a median age of 70, and median eGFR 55.8mL/min/1.73m2. Primary composite lactic acidosis events per 1,000 person-years occurred at a rate of 4.18 (95% CI 3.63, 4.81) in metformin users compared to 3.69 (3.19, 4.27) in S.U. users; this resulted in an adjusted hazard ratio of 1.21 (95% CI 0.99, 1.50). The average daily metformin dose was 1000 mg, and the average glyburide and glipizide dose was 5 mg. Six months after reaching the kidney threshold, 55% of metformin and 61% of S.U. users continued treatment; after 12 months, this dropped to 43% in the metformin and 48% in the S.U. users. Subgroup analysis based on eGFR, at the time of enrollment, showed that in users with the lowest eGFR <45 mL/min/1.73m2, lactic acidosis was higher with metformin compared to S.U.s, with a weighted hazard ratio of 1.31 (95% CI 0.85, 2.01).3 

Chu et al.’s findings of comparable rates of lactic acidosis among metformin and S.U. users with T2DM and mild to moderate kidney disease align with the FDA’s decision to revise metformin’s safety information. The use of metformin is still contraindicated in patients with severe kidney disease. If metformin is continued in patients with reduced kidney function, appropriate dosage reductions, safety monitoring, and kidney function assessment are recommended.  

Practice Pearls: 

  • Metformin can be safely used in most patients with mild to moderate kidney disease. 
  • Compared to sulfonylureas, metformin did not significantly increase lactic acidosis in patients with mild to moderate kidney disease. 
  • These findings do not apply to patients with severe kidney disease due to a different benefit-risk ratio. 


  1. Curtin, François, and Pierre Schulz. “Assessing the benefit: risk ratio of a drug–randomized and naturalistic evidence.” Dialogues in clinical neuroscience vol. 13,2 (2011): 183-90. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3181998/ 
  2. RoumieChristianne L, et al. “Association of Treatment With Metformin vs. Sulfonylurea With Major Adverse Cardiovascular Events Among Patients With Diabetes and Reduced Kidney Function.” JAMA, vol. 322, (2019). doi:10.1001/jama.2019.13206. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6753652/ 
  3. Chu, Patricia, et al. “Hospitalization for Lactic Acidosis Among Patients With Reduced Kidney Function Treated With Metformin or Sulfonylureas.” Diabetes Care, vol. 43, (2020). doi:10.2337/dc19-2391 
  4. American Diabetes Association. “Pharmacologic Approaches to Glycemic Treatment: Standards of Medical Care in Diabetes—2020.” Diabetes Care vol. 43 (Supplement 1) (2020): S98-S110; doi:10.2337/dc20-S009  
  5. Glucophage [package insert]. Princeton, NJ: Bristol-Myers Squibb Company; 2018 


Louise Brown, PharmD Candidate, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences