TOPLINE

A 14-day course of metformin during acute COVID did not definitively prevent long-term symptoms at 6 months. However, relatively fewer long COVID cases were diagnosed, with no major safety signals.

METHODOLOGY

• In a quadruple-blinded, randomized controlled platform trial (ACTIV-6) of repurposed treatments for COVID, researchers assessed whether early metformin reduced long-term symptoms in patients with mild-to-moderate disease.
• They included 2983 patients aged 30 years or older who tested positive for SARS-CoV-2 infection within 10 days and had at least two symptoms for up to 7 days; the median age of the patients was 47 years, and 63.4% were women.
• Patients were randomly assigned to receive either metformin or a matching placebo for 14 days, with the metformin dose starting at 500 mg once daily and increasing to 500 mg in the morning and 1000 mg in the evening.
• The primary outcome was mortality or COVID-related symptoms at day 180. Secondary outcomes included symptoms at days 90 and 120, clinician-diagnosed long COVID, and symptom burden.
• The analysis used a statistical measure called the posterior probability of efficacy (PPE), and a threshold of 0.975 was required to declare metformin effective for the primary outcome.

TAKEAWAY

• At day 180, 2.3% of patients receiving metformin and 3.0% of those receiving placebo reported COVID-related symptoms, with a statistically nonsignificant difference in risk. No deaths occurred.
• The PPE for preventing the primary outcome was 0.83, indicating that the trial did not meet its prespecified threshold for efficacy.
• At day 180, 0.56% of the patients receiving metformin and 1.17% of those receiving placebo had clinician-diagnosed long COVID, corresponding to nearly half the relative risk (although the finding was of borderline significance). The absolute risk difference between the metformin and placebo groups was not significant.

IN PRACTICE

“While a low event rate makes it challenging to evaluate therapeutics, it is encouraging that the proportion of participants reporting new clinician diagnosis of long COVID at 6 months was low,” the authors of the study wrote.

SOURCE

The study was led by Carolyn T. Bramante, MD, MPH, University of Minnesota Medical School in Minneapolis. It was published online on June 04, 2026, in Clinical Infectious Diseases.

LIMITATIONS

Some patients may have reported symptoms that started before they enrolled in the trial. Researchers lacked follow-up data of about 20% of patients. The overall rate of incidence of long COVID was very low. 

DISCLOSURES

ACTIV-6 study received support from CARES ACT and American Rescue Plan Act funds awarded through grants from the National Center for Advancing Translational Sciences. Additional support came from the Office of the Assistant Secretary for Preparedness and Response, Biomedical Advanced Research and Development Authority, and an award from Vanderbilt University. Several authors reported grants, consulting fees, advisory roles, stock or patents, or other financial ties with multiple organizations. One author reported serving as deputy editor for Clinical Infectious Diseases.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.