Metformin is the most prescribed oral medication for type 2 diabetes in the US due to its low cost, efficacy in reducing insulin resistance, and overall safety profile. However, literature shows that metformin can cause a rare but serious side effect of lactic acidosis (LA) with an incidence of 0.03-0.06 per 1000 patients. Persons requiring continuous oxygen therapy (COT) may be at risk for LA associated with hypoxia and decreased tissue perfusion, but the risk of metformin associated lactic acidosis (MALA) in persons requiring COT is unclear. Due to lack of evidence, some clinicians may empirically discontinue metformin in patients on COT, compromising glycemic stability. The objective of this drug utilization evaluation was to determine the frequency of metformin use and incidence of MALA in patients receiving COT.

A retrospective review was conducted at a large regional Veterans Affairs Medical Center for patients with active prescriptions of metformin and supplemental oxygen therapy from 4/2017-4/2020. Of 578 unique patients identified, manual medical-record review was conducted for 6 patients who had serum lactate levels >4 mmol/L while using concomitant metformin and COT. All were white males with mean: age of 71, pH of 7.35, and eGFR of 56. No cases of LA or DKA were identified. In all cases, elevated lactate levels were attributed to advanced medical conditions including: COPD, sepsis, CHF, CKD, and liver disease. None had serum metformin levels drawn.

In conclusion, the association of elevated serum lactate among persons using concomitant metformin and COT was low and occurred in persons with advanced medical conditions. It could not be determined if metformin use was a contributing factor without metformin levels. These findings confirm the importance of clinical judgement when prescribing and continuing metformin for persons with advanced underlying medical conditions. Further study with serum metformin levels may help to better characterize the risk of MALA for persons on COT.


S. J. Lutz-mccain: None. B. E. Desanzo: None. B. Herk: None. M. Mclinden: None. M. M. Dinardo: None. R. Codario: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at