Metformin and ursodeoxycholic acid do not improve long COVID recovery
1. In this randomized controlled trial, neither metformin nor ursodeoxycholic acid improved recovery from long COVID compared with placebo in adults.
2. Participants who recovered from long COVID showed reductions in multiple cytokine levels, while those who did not recover showed no meaningful changes.
Evidence Rating Level: 1 (Excellent)
Study Rundown: Metformin has immunomodulatory and antiviral effects, and previous studies suggested it may reduce severe COVID-19 outcomes such as hospitalization or death. Ursodeoxycholic acid also modulates immune responses by suppressing monocyte and macrophage activation and influencing pathways associated with reduced SARS-CoV-2 infection and severity. This study evaluated whether metformin or ursodeoxycholic acid could improve long COVID symptoms over eight weeks using the PASC index. Overall, recovery from long COVID by eight weeks was similar across all groups, including those receiving metformin, ursodeoxycholic acid, or placebo. Changes in mean PASC scores were comparable, indicating no meaningful differences in symptom improvement between interventions and placebo. Cytokine analyses revealed that participants who recovered generally exhibited reductions in multiple cytokines over the study period, whereas those who did not recover showed no significant cytokine changes. Adverse events were infrequent and generally mild. No severe events were reported in the metformin or ursodeoxycholic acid groups, while a single unrelated severe event occurred in the placebo group. No deaths were reported during the study. The study’s generalizability is limited, as participants were predominantly younger adults with mild long COVID. Despite the biological plausibility of metformin and ursodeoxycholic acid for modulating immune responses and potentially improving outcomes, this study found that neither intervention improved recovery from long COVID compared with placebo. These findings suggest that, for this population, routine use of metformin or ursodeoxycholic acid is unlikely to offer clinical benefit in addressing long COVID symptoms.
Click to read this study in AIM
In-Depth [randomized controlled trial]: This randomized controlled trial evaluated the effect of metformin or ursodeoxycholic acid on long COVID symptoms at 8 weeks. Adults with confirmed prior SARS-CoV-2 infection and a PASC index score of 12 or higher were enrolled from July 2024 to January 2025 at two medical centers in Seoul, South Korea, with follow-up through April 2025. Participants were randomly assigned to metformin, ursodeoxycholic acid, or placebo in a 1:1:1 ratio. The primary outcome was PASC recovery at 8 weeks, defined as improvement from a score of 12 or higher to below 12. Secondary outcomes included changes in continuous PASC scores, symptom improvement, and cytokine changes. Safety outcomes included adverse events (AEs), severe AEs (SAEs), hospitalizations, and deaths. A total of 396 participants were randomized (132 per group), with a median age of 36 years; 72% were women. Overall, 67% achieved recovery by 8 weeks: 63.6% in the metformin group, 68.2% in the ursodeoxycholic acid group, and 68.2% in the placebo group. Compared with placebo, the probability of recovery was slightly lower for metformin and equivalent for ursodeoxycholic acid. Mean PASC scores and mean change from baseline were similar across all groups. Cytokine analyses in 377 participants showed that recovered individuals had decreases in interferon-γ, IL-4, IL-6, IL-7, IL-17A, IL-1β, and IL-27, while those who did not recover showed no significant changes. AEs were uncommon, occurring in 5.4% of the metformin group, 4.6% of the ursodeoxycholic acid group, and 2.3% of placebo participants. No SAEs occurred in the active treatment groups, while one unrelated SAE occurred in the placebo group. This study indicates that neither metformin nor ursodeoxycholic acid significantly improves long COVID symptoms compared with placebo, despite biologic plausibility and observed immunologic changes in recovered participants.
Image: PD
©2026 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.
