Most people who contract COVID fully recover, but millions of others grapple with chronic symptoms that persist for months or years after the initial infection. Known as long COVID, this enigmatic syndrome is marked by a wide array of symptoms, including intense fatigue, chest pain, dizziness and cognitive issues such as brain fog—all of which often fluctuate in intensity and duration. There is also no universally accepted definition or test for long COVID, leaving most sufferers without a clear path to diagnosis or treatment. But a recent study found that a widely used, inexpensive diabetes drug reduced the risk of developing long COVID by 41 percent among people who are overweight and those with obesity.

“I don’t throw around the term ‘breakthrough,’ but it applies here. These findings offer the first concrete hope for preventing long COVID” from a randomized trial, says Eric Topol, founder and director of the Scripps Research Translational Institute, who has co-authored studies on long COVID and was not involved with the new study.* He adds that the results will need to be replicated with larger, more diverse sample sizes.

The researchers say their study, which was published this month in the Lancet Infectious Diseases, is the first placebo-controlled, randomized clinical trial to investigate long COVID incidence after the use of metformin, which is typically used to lower blood sugar in people with type 2 diabetes. More than 1,100 study participants, all of whom were overweight or had obesity, were randomly assigned to one of six groups that received various combinations of metformin, the antiparasitic drug ivermectin, the antidepressant drug fluvoxamine or a placebo. Only 6.3 percent of study participants who took metformin were diagnosed with long COVID within 300 days of their initial illness, compared with 10.4 percent of participants taking a placebo. Neither ivermectin nor fluvoxamine reduced the risk of long COVID.

These findings come from the second phase of the University of Minnesota’s COVID-OUT clinical trial, which previously found that metformin reduces the risk of COVID-related hospitalization and death in the first two weeks of infection. As reports of long COVID increased, the researchers broadened their investigation to evaluate the long-term effects of metformin and other medications.

Previous research has shown that people who are overweight and those with obesity are more vulnerable to developing both severe COVID and long COVID. The COVID-OUT trial was originally designed to evaluate COVID treatments in this higher-risk population and was also limited to people experiencing their first infection. Carolyn Bramante, an obesity medicine physician and lead investigator of the trial, says additional studies are necessary to validate the efficacy of metformin in other demographic groups.

Bramante suspects that metformin’s mechanism for preventing long COVID would be similar in people with lower weight, although she says reduction in long COVID risk might be less pronounced. Further studies are needed to fully understand how metformin prevents long COVID, but the drug may work by blocking viral replication. In June Bramante and her colleagues shared a preprint study, which has yet to be peer-reviewed, showing that people who took metformin had lower COVID viral loads, or amount of virus, in their nose.

“Metformin is already widely distributed in outpatient treatment,” Bramante says. “It’s low-cost and easily available, which makes it an ideal treatment for preventing long COVID.” The drug’s widespread accessibility is significant because it can be obtained relatively quickly—and it appears to be most effective when administered in the early stages of COVID. Study participants who took metformin within three days of experiencing initial symptoms exhibited one of the lowest incidences of long COVID among people in the study.

Bramante also highlights metformin’s well-documented safety profile among people who face a higher risk of developing severe COVID symptoms. Notably, COVID-OUT is one of few COVID clinical trials to include participants who were pregnant or breastfeeding, and studies show these individuals can safely use metformin for extended periods.

But the drug’s accessibility and safety have complicated the question of whether health care providers should begin prescribing it for the prevention of long COVID—and to whom.

Topol notes the difficult balance between wanting additional studies to confirm COVID-OUT’s findings and addressing long COVID’s immediate risks as the virus continues to infect thousands of people each day. Although he would like to see an independent trial replicate these results, he believes metformin’s benefits and relative safety outweigh the “disabling” prospects of long COVID for those who develop the condition.

“Ordinarily I wouldn’t recommend taking a drug without a second trial—but these aren’t ordinary circumstances,” Topol says, adding that he would ask his doctor for a two-week course of metformin if he contracted COVID.

But Vassilios Vassiliou, a professor in cardiac medicine at the University of East Anglia in England, who was not involved in the new study, questions the practicality of widely prescribing metformin at this stage. It is difficult to predict who will develop long COVID, which often doesn’t become evident until weeks or months after initial infection, so clinicians would seemingly need to provide the drug to all patients with COVID. “What if you need to treat 200 patients with metformin in the early phase of infection to prevent one case of long COVID? Is that worth it?” Vassiliou says. “Our first principle should be ‘do no harm,’ and my worry is that we would see many side effects and lower effectiveness if metformin use was scaled up.”

For those reasons, he says, follow-up trials focused on treating long COVID in people who already have the condition are critical.

There is currently no approved medication for the treatment of long COVID, despite ongoing investigations into several promising candidates. Researchers at Yale University are recruiting 100 participants with long COVID for a randomized trial evaluating the antiviral medication Paxlovid, which is already widely used to treat acute COVID. Another study is testing naltrexone, a drug typically used to treat alcohol use disorder and opioid dependence. The findings from both studies are not expected to be published until next year, however.

Bramante emphasizes that clinicians should not wait for a one-size-fits-all solution to treating long COVID. “There won’t be one silver bullet because the experience of long COVID is heterogeneous. We will need several different ways to treat it—behavioral and pharmacological,” she says. “We need to recognize and treat it as a chronic disease.”

*Editor’s Note (6/22/23): This sentence was updated after posting to clarify Eric Topol’s comments about preventing long COVID.