Metformin May Reduce Long COVID in Non-Diabetic Population

LOS ANGELES — Metformin prescribed within a week of SARS-CoV-2 infection was associated with a 53% reduction in long COVID or death over 6 months in people without diabetes or prediabetes, according to data presented on Saturday at the Infectious Disease Week (IDWeek) 2024 Annual Meeting.

Long COVID was determined by using the diagnostic code U09.9 or a computable phenotype based on symptoms and conditions. Most participants in this study were infected with the Omicron variant.

Researchers, led by Carolyn Bramante, MD, MPH, an internist, pediatrician, and obesity medicine specialist at the University of Minnesota Medical School in Minneapolis, simulated a randomized controlled trial of metformin vs control using the National COVID Cohort Collaborative (N3C) Electronic Health Record Database.

The intervention was a prescription for metformin within 6 days of SARS-CoV-2 infection. Those in the control group, which was designed to mimic placebo, had a prescription for fluvoxamine, fluticasone, ivermectin, or montelukast (all drugs that have been used off-label for COVID but have shown no effect on acute COVID outcomes in clinical trials). Exclusions included anyone with a previous metformin prescription or a comparator prescription; any indication for chronic metformin use; or a contraindication for metformin.

Why Study Metformin for Long COVID?

Bramante led a previous randomized controlled trial, COVID-OUT, with 1323 people that indicated metformin showed possible benefit for preventing the more severe components of COVID-19. She also led a 2020 review, in which she examined electronic health records from adults with type 2 diabetes or obesity. The researchers found that women taking metformin before they developed COVID-19 were significantly less likely to die after being hospitalized — although men didn’t see the same protective effect. Another randomized trial of 20 people found that 60% of those taking metformin vs 100% of those given a placebo had detectable SARS-CoV-2 viral load by day 4.

Other trials have highlighted the anti-inflammatory and antiviral properties of metformin. The existing evidence coupled with metformin’s well-established safety profile, led Bramante’s team to conduct the current simulated trial in people without diabetes or prediabetes. Bramante noted that metformin’s only US Food and Drug Administration–approved indication is for diabetes.

The current study featured a similar racial/ethnic makeup in the metformin and control groups: 16% and 17% were Black and 16% and 13% were Hispanic, respectively. Within 6 months, 4.0% in the metformin group developed long COVID or died compared with 8.5% in the control group (Relative Risk [RR], 0.47; 95% CI, 0.25-0.89). For prescriptions made on days 0-1 relative to infection, the RR was 0.39 (95% CI, 0.12-1.24). When metformin was prescribed on days 0-14, the RR was 0.75 (95% CI, 0.52-1.08).

The reason it’s important to have an active comparator is to control for things that can’t be measured, such as engagement in healthcare and the placebo effect, Bramante said.

Emily Erbelding, MD, MPH, director of the Division of Microbiology and Infectious Diseases at the National Institute of Allergy and Infectious Diseases, who was not part of the study, noted the potential implications of the findings.

Proven Safety and Low-Cost of Metformin 

“We don’t have therapies for long COVID, and we don’t know how to prevent it in people who have SARS-CoV-2 infections,” Erbelding noted. “This analysis points to metformin, a drug that millions of people have taken safely for their diabetes or their borderline diabetes. It’s licensed, it’s out there, and it’s inexpensive. The fact that we have data that point to this potentially being a therapy is important. I think that’s the power of this.”

Erbelding said a strength of the study is the size of the N3C Electronic Health Record Database (with data on nearly 9 million COVID cases) the researchers used to simulate the randomized controlled trial.

“(These results) gives us a reason to think about doing a large randomized controlled study with metformin,” she said. However, there are some limitations, she noted.

“The definition of long COVID may not have been applied exactly the same way across all the patients and you don’t know what led the prescribers to prescribe metformin. There might have been confounders that couldn’t be controlled for or weren’t evident in the way they approached the data.”

This study has “relatively rigorous methodology for an observational study,” Erbelding said. “It’s novel to try to simulate a randomized controlled trial through a large, observational, electronic record–based cohort. Maybe we should be doing more of this because these bioinformatic systems exist now. And we need to get all the public health use out of them that we can.” 

“The fact that they may be unlocking something new here that needs follow-up in a truly randomized controlled trial is important as well because there are a lot of people out there suffering from long COVID.”

Bramante and Erbelding disclosed no relevant financial relationships.

This research was supported in part by the intramural/extramural research program of the National Center for Advancing Translational Science, National Institutes of Health.

Marcia Frellick is a freelance journalist based in Chicago. She has previously written for the Chicago Tribune, Science News, Northwestern magazine, and Nurse.com and was an editor at the Chicago Sun-Times, The Cincinnati Enquirer, and St. Cloud Times. Follow her on X @MLfrellick.