Metformin and Paxlovid May Prevent Long COVID
From Glucose Control to Viral Defense
Metformin was designed to lower blood sugar in people with type 2 diabetes. It has been a cornerstone of diabetes care for decades. Nobody developed it with a coronavirus in mind. Nirmatrelvir-ritonavir, sold as Paxlovid, was engineered to stop a specific viral enzyme from copying itself. Its target was the SARS-CoV-2 protease. Its purpose was short-term: blunt the acute infection, keep people out of hospital. Neither drug was created to prevent a chronic illness that can last for months or years after the initial infection clears. Yet both are now being studied for exactly that purpose.
That repurposing matters because long COVID affects millions of people worldwide. The condition is chronic, often disabling, and has resisted every treatment thrown at it. A prevention strategy — stopping long COVID before it starts — would change the calculus entirely. Instead of managing an illness that may never resolve, physicians could intervene during the acute phase and block the downstream damage.
The evidence for this approach has grown. Clinical-trial results now suggest that both drugs can reduce the risk of long-term disease after coronavirus infection. The findings are not uniform. But they have moved the conversation from speculation to something closer to clinical decision-making.
Why the Old Treatment Model Breaks
The traditional model of COVID-19 treatment was simple: treat the acute infection, and the rest will follow. If the virus is cleared, the patient recovers. Long COVID broke that model. It showed that viral clearance does not guarantee full recovery. Something else — persistent inflammation, viral reservoirs, immune dysregulation — keeps the disease alive after the initial threat is gone.
The new trial data force a revision. If metformin and nirmatrelvir-ritonavir reduce long-COVID risk, then the acute phase is not just a battle to survive. It is a window of opportunity. What happens during those first days and weeks may determine whether a patient develops a chronic condition months later.
One trial, published in Diabetes Care in 2024, examined metformin’s effect on long-COVID incidence. [3] The lead author, S. G. Johnson, and colleagues reported results that supported the drug’s preventive potential. [3] Another study, led by C. T. Bramante and published in Clinical Infectious Diseases in 2026, looked at nirmatrelvir-ritonavir. [4] It found a small but measurable benefit. [4] A separate trial by O. Oppegaard and colleagues, published in The Lancet Infectious Diseases in 2026, added further evidence. [5] The trials differ in design, population, and endpoint definitions, but they point in the same direction: early treatment may shape long-term outcomes.
The implication is uncomfortable for the old model. If long COVID can be prevented, then the goal of acute treatment expands. It is no longer enough to discharge a patient virus-free. The question becomes whether the treatment also protected them from what comes next.
What the Trials Do Not Yet Show
Not every physician is convinced. Some say they will start prescribing the medications based on the trial results. Others look at the same data and see uncertainty. The effect sizes are modest. A small benefit in one study may not replicate in another.
That disagreement is not a failure of the science; it is how evidence accumulates. The clinical-trial results are promising, but they are not definitive. The drugs are not approved for long-COVID prevention. Prescribing them for that purpose is an off-label decision, made by individual physicians weighing risks and benefits for individual patients.
The limitation is built into the data itself. Metformin and nirmatrelvir-ritonavir may reduce the risk of long COVID, but they do not eliminate it, they do not treat long COVID once it has developed, and they do not work for everyone. The mechanism — why they might prevent a chronic condition — remains incompletely understood. Until larger, longer trials confirm the findings and clarify who benefits most, the evidence supports cautious optimism, not a new standard of care.
Sources
1. DOI: 10.1038/d41586-026-02341-7
2. O. Oppegaard