Metformin and Ursodeoxycholic Acid Fail to Improve Post-COVID Recovery: Insights from a Randomized Controlled Trial

Introduction: The Persistent Challenge of PASC

Postacute sequelae of SARS-CoV-2 infection (PASC), colloquially known as Long COVID, has emerged as a significant public health crisis following the global pandemic. Characterized by a constellation of symptoms including fatigue, cognitive dysfunction, and respiratory issues that persist long after the initial infection, PASC affects millions worldwide. Despite its prevalence, the medical community has struggled to identify evidence-based pharmacological interventions to alleviate these symptoms. The search for effective treatments has led researchers to investigate repurposed drugs with known safety profiles and potential mechanistic relevance to the pathophysiology of PASC.

Highlights of the Study

No Superiority Over Placebo

In a rigorous double-blind, randomized clinical trial, neither metformin nor ursodeoxycholic acid (UDCA) demonstrated a statistically significant improvement in PASC recovery rates compared to a double placebo group.

High Rate of Natural Resolution

Approximately 68% of participants in the placebo group achieved recovery (defined by a PASC index score below 12) within eight weeks, suggesting a high rate of natural symptom improvement over time in this cohort.

Consistent Lack of Efficacy Across Metrics

Both the mean change in PASC scores and the proportion of recovered patients remained consistent across all treatment arms, indicating no clinical benefit from the 14-day intervention course.

Rationale for Investigating Metformin and UDCA

The choice of metformin and UDCA as candidate treatments was rooted in their systemic effects. Metformin, a widely used biguanide for type 2 diabetes, has shown potential anti-inflammatory and antiviral properties. Previous studies, such as the COVID-OUT trial, suggested that early administration of metformin during acute COVID-19 might reduce the risk of developing PASC. However, its efficacy as a treatment for established PASC remained an open question.

Ursodeoxycholic acid (UDCA) was considered due to its role in modulating bile acid signaling and its potential to downregulate the ACE2 receptor, which SARS-CoV-2 uses for entry into cells. Furthermore, UDCA possesses anti-inflammatory and cytoprotective properties that were hypothesized to mitigate the chronic low-grade inflammation often associated with Long COVID.

Study Design and Methodology

This double-blind, placebo-controlled, randomized clinical trial was conducted at two tertiary hospitals in South Korea between July 2024 and April 2025. The study enrolled 396 adults who met the criteria for PASC, specifically those with a PASC index score of 12 or greater. This scoring system was used to standardize the assessment of various symptoms and ensure a baseline level of clinical impairment.

Participants were randomly assigned in a 1:1:1 ratio to one of three groups:

• Oral metformin (uptitrated to 1500 mg daily)
• UDCA (900 mg once daily)
• Double placebo

The intervention lasted for 14 days. The primary endpoint was the proportion of participants achieving PASC recovery—defined as a PASC index score of less than 12—at the 8-week follow-up mark.

Key Findings and Results

The study population had a median age of 36 years, and 72% were women. At baseline, the mean PASC score was 19.3, and the mean interval from the initial SARS-CoV-2 infection was approximately 9.8 months. This indicates that the participants were experiencing relatively long-term sequelae rather than early post-viral symptoms.

Recovery Rates

At the 8-week assessment, recovery occurred in 63.6% (84 of 132) of the metformin group, 68.2% (90 of 132) of the UDCA group, and 68.2% (90 of 132) of the placebo group. The lack of difference between the groups was striking, particularly given the near-identical outcomes between the UDCA and placebo arms.

Change in Symptom Severity

The mean change in PASC scores from baseline to week 8 also showed no significant therapeutic advantage:

• Metformin: -10.05 (95% CI, -11.35 to -8.76)
• UDCA: -10.62 (95% CI, -11.79 to -9.45)
• Placebo: -10.43 (95% CI, -11.69 to -9.18)

These figures demonstrate that while patients generally improved over the two-month period, the improvement was not attributable to the pharmacological interventions studied.

Expert Commentary and Clinical Implications

The findings of this trial are soberingly clear: a short course of metformin or UDCA is not an effective strategy for treating established PASC. This study highlights several critical points for clinicians and researchers.

The Importance of Placebo Controls

The high recovery rate in the placebo group (68.2%) underscores the difficulty of studying PASC. Many patients experience a fluctuating but generally improving course of illness. Without a placebo control, the improvement seen in the metformin or UDCA groups might have been mistakenly attributed to the drugs themselves.

Duration of Treatment

One potential limitation discussed by experts is the duration of the intervention. A 14-day course may be insufficient to reverse the complex, multi-systemic pathophysiology of PASC, which may involve persistent viral reservoirs, autoimmune dysfunction, or mitochondrial impairment. However, even within this timeframe, the data showed no hint of a diverging trend between the active and control groups.

Prevention vs. Treatment

It is important to distinguish these results from trials investigating the *prevention* of PASC. While metformin has shown promise in reducing the incidence of Long COVID when administered during the acute phase of infection, this study confirms that it does not appear to serve as a “rescue” therapy once the syndrome has already stabilized into a chronic state.

Conclusion

In conclusion, this randomized clinical trial provides high-quality evidence that neither metformin nor ursodeoxycholic acid, at the dosages and duration tested, effectively treats the symptoms of PASC. As the medical community continues to navigate the long-term consequences of the COVID-19 pandemic, these results emphasize the need for continued research into the underlying mechanisms of Long COVID and the search for more targeted therapeutic pathways. For now, clinical management of PASC should remain focused on symptom-specific support and rehabilitation rather than the off-label use of metformin or UDCA.

Funding and Clinical Trial Information

This study was funded by the National Institute of Infectious Diseases and the National Institute of Health, South Korea. It is registered with the Clinical Research Information Service (KCT0009342).

References

Lim SY, Lee J, Chang E, et al. Neither Metformin nor Ursodeoxycholic Acid Effectively Treats Postacute Sequelae of COVID-19: A Randomized Clinical Trial. Annals of Internal Medicine. 2026-03-03. PMID: 41771135.