Home / Resources / Articles / Caution Using Metformin For Patients With Reduced Kidney Function

Caution Using Metformin For Patients With Reduced Kidney Function

Sep 22, 2018

FDA label change warns of increased risk of lactic acidosis in patients with severe renal dysfunction.

Metformin, a first-line agent used in the treatment of type 2 diabetes mellitus (T2DM), is heralded for its low cost and general tolerability. Previously, metformin’s drug label stated it was contraindicated in those with renal disease or dysfunction (i.e. serum creatinine of ≥ 1.5 mg/dL and ≥ 1.4 mg/dL in males and females, respectively). The concern for its use in this subpopulation stems from potential lactic acidosis and accumulation of the drug due to poor renal clearance.


However, after review of evidence and petitions to the U.S. Food and Drug Administration (FDA), the drug label was changed. On April 8, 2016, the FDA announced that metformin use would be expanded and can be used in people who have diabetes with both mild and moderate kidney impairment. The new label indicates metformin use is not recommended in patients with eGFR < 30 mL/min/1.73 m2, should not be started when eGFR is 30-45 mL/min/1.73 m2, should be discontinued when it falls below 30 mL/min/1.73 m2, and use should be assessed if eGFR is below 45 mL/min/1.73 m2. This decision impacts all metformin-containing medications. In addition,the label provides guidance for healthcare professionals looking to prescribe metformin in their patient.

But why mention a change from a couple years ago? Well, mainly due to conflicting guideline recommendations. Some state metformin can be used carefully at eGFR of 30-60 mL/min/1.73 m2 and be assessed at 30-45 mL/min/1.73 m2. Nevertheless, it appears that previous data regarding metformin use at eGFR < 60 mL/min/1.73 m2 was ambiguous and results variable.

On June 4, 2018, an article published in JAMA Internal Medicine, assessing metformin use and acidosis over a range of eGFR, found some interesting results. The study conducted from January 2004-2017 used information from a large electronic medical record database at Geisinger Health System. Patients enrolled had a diabetes diagnosis and serum creatinine measurement after diagnosis. Most patient information was obtained from patient encounter forms, laboratory results, prescriptions, ICD codes, etc. Patients were excluded if they were lacking serum creatinine measurements and/or had end-stage renal disease. The primary outcome of the study was hospitalization due to acidosis, as defined by ICD-9 codes. Finally, eGFR was calculated using serum creatinine measurements and the CKD-Epidemiology equation while staging used the Kidney Disease Improving Global Outcomes guidelines (i.e. <30, 30-44, 45-59, 60-89, and ≥ 90 mL/min/1.73 m2).

The study sought to perform a variety of analyses. They were as follows: 1) compare baseline characteristics based on baseline eGFR staging; 2) acidosis risk in people taking metformin versus those not using metformin; 3) compare new metformin use with new sulfonylurea use in an active comparator study; 4) associations after excluding those who take insulin; 5) acidosis risk in those who use metformin compared to those who use sulfonylureas.

After exclusion, 75,413 patients were included in the study. The mean age was 60.4 years old, mean BMI was 34.1, and most were female (51%). Overall, a total of 2,335 hospitalizations were caused by acidosis (737 events in metformin users vs 1598 events in non-metformin users). Analysis of acidosis in metformin users versus non-use had an adjusted hazard ratio of 0.98 (95% CI 0.89-1.08) and higher risk was noted with metformin use and lower eGFR (p = 0.01). Risk with metformin use was not significant at eGFR of >90, 60-89, 45-59, and 30-44 mL/min/1.73 m2. Conversely, risk was increased in those taking metformin with an eGFR less than 30 mL/min/1.73 m2 (adjusted HR 2.07 95% CI 1.33-3.22).

Looking at metformin use or nonuse, incidence of acidosis was mostly associated with lower eGFR. It should be noted that acidosis events occurred at various eGFR ranges. Incidence can be found in Table 1: Incidence of Acidosis. In the active comparator portion of the study, they compared metformin or sulfonylurea use and acidosis. Overall, incidence of acidosis was higher in sulfonylurea use compared to metformin, but associations with either medication and acidosis were quite similar. Furthermore, lower eGFR and acidosis risk was higher with both medications. Additionally, comparing metformin use with other hypoglycemic medications, risk of acidosis was similar, but like previous observations, acidosis was more likely at eGFR < 30 mL/min/1.73 m2.


Table 1: Incidence of Acidosis
eGFR (mL/min/1.73 m2) Number of events per 1000 person-years
60 – 89 4
45 – 59 7
30 – 44 10
< 30 24

By the study’s end, the results were quite promising. Investigators found that metformin use and acidosis events requiring hospitalization were not likely in patients with an eGFR of at least 30 mL/min/1.73 m2. This study analyzed a multitude of associations in a variety of ways, all in an effort to account for many confounding variables (i.e. eGFR stage, eGFR staging over time, demographic variables, cardiovascular risk, medications, etc.). The finding that acidosis was unlikely in patients treated with metformin and having eGFR of at least 30 mL/min/1.73 m2 was found true in adjusted, active comparator, propensity-matched, and replication cohorts. B. Lazarus, et al concluded that this evidence supports the recent FDA label changes along with the recommendations by other guidelines. Therefore, metformin can be used carefully in patients who have type 2 diabetes with an eGFR of 30-44 mL/min/1.73 m2 and can be used in those with an eGFR of 45-59 mL/min/1.73 m2. However, patients with more severe renal dysfunction, eGFR < 30 mL/min/1.73 m2, should likely avoid metformin altogether as risk of acidosis was increased.

Practice Pearls:

  • The U.S. Food and Drug Administration changed the drug label for metformin to be more specific and also changed the renal function measurement from serum creatinine to estimated glomerular filtration rate (eGFR).
  • Investigators found that risk of acidosis with metformin use was not likely in patients with an eGFR of at least 30 mL/min/1.73 m2.
  • Use of metformin at an eGFR of 30-60 mL/min/1.73 m2 was not associated with an increased risk of acidosis, however caution should still be used when prescribing metformin in those with mild to moderate kidney dysfunction.
  • Use of metformin at an eGFR < 30 mL/min/1.73 m2 is associated with higher acidosis risk and therefore should not be used.


Lazarus, B., Wu, A., Shin, J., Sang, Y., Alexander, C., Secora, A., Inker, L., Coresh, J., Chang, A., Grams, M. Association of Metformin Use With Risk of Lactic Acidosis Across the Range of Kidney Function: A Community-Based Cohort Study. JAMA Intern Med. 4 June 2018. doi:10.1001/jamainternmed.2018.0292.

Metformin Is Safe for Most Patients With Diabetes, Kidney Disease. DocGuide.com News. 6 June 2018. Available from: https://www.docguide.com/metformin-safe-most-patients-diabetes-kidney-disease?hash=1fb38c67&amp;eid=64554&amp;alrhash=2dd83c-e0e7ce4a44e4b087e4f5526c4855ba49.

FDA Drug Safety Communication: FDA revises warnings regarding use of the diabetes medicine metformin in certain patients with reduced kidney function. U.S. Food & Drug Administration. 8 April 2016. Available from: https://www.fda.gov/Drugs/DrugSafety/ucm493244.htm.

Kaytie A. Weierstahl, Pharm.D. Candidate, LECOM School of Pharmacy