Elevated Liver Damage Biomarkers in Long COVID: A Systematic Review and Meta-Analysis

Abbas F. Almulla, Yanin Thipakorn, Yingqian Zhang, Michael MaesmedRxiv 2024 · DOI

TL;DR

This systematic review and meta-analysis of 61 studies reveals that Long COVID (LC) is not just a respiratory or neurological condition, but a multi-organ disease characterized by persistent liver damage and blood-clotting abnormalities. Even 3 to 6 months after the initial infection, patients show significantly elevated biomarkers of liver injury, suggesting that the "brain fog" and fatigue associated with the condition may have roots in hepatic dysfunction.

What Problem Does This Paper Address?

While the acute phase of COVID-19 is well-known for causing liver injury in nearly 40% of patients, the long-term trajectory of this damage has remained murky. Long COVID is a complex syndrome affecting various systems, yet most research focuses on the lungs or the brain.

The authors address a critical gap: the lack of a comprehensive, pooled analysis of blood-based biomarkers to confirm whether liver injury persists in the post-acute phase. By synthesizing data from thousands of participants, they aim to determine if the liver is a "silent victim" of the virus, contributing to the systemic symptoms reported by "long-haulers."

Methods

The researchers conducted a systematic review and meta-analysis following PRISMA and MOOSE guidelines. They searched major databases (PubMed, Google Scholar, etc.) for studies comparing liver biomarkers in Long COVID patients versus healthy controls.

Key technical aspects included: * Composite Indices: Beyond looking at individual enzymes, they calculated a "Liver Damage Index" (incorporating ALT, AST, ALP, GGT, and bilirubin) to account for data dependency. * Quality Assessment: They used the Immunological Confounder Scale (ICS) to evaluate the methodological quality and potential bias of the included studies. * Statistical Modeling: A random-effects model using constrained maximum likelihood was employed to aggregate effect sizes, expressed as Standardized Mean Differences (SMD). * Bias Correction: They utilized the "trim-and-fill" method to account for publication bias, ensuring the results weren't skewed by missing or unpublished data.

Datasets

The study synthesized data from 61 eligible studies involving a total of 7,172 participants (3,404 Long COVID patients and 3,768 controls). The data spanned a wide geographic range, with significant contributions from Spain, Italy, Russia, and Germany. The participant age range was broad, from 11 to 69 years.

Key Insights and Ideas

  • The Persistence of Injury: The most striking finding is that liver enzyme elevations (ALT, AST, GGT) and lactate dehydrogenase (LDH) activity remain high 3–6 months post-infection. This suggests that the liver does not always "bounce back" after the virus is cleared.
  • The Liver-Brain Connection: The authors highlight a fascinating link between liver health and mental health. Elevated ALT is a known predictor of depression and insulin resistance. The study suggests that liver-brain interactions—mediated by inflammatory cytokines like IL-6—might be a driving force behind the neuropsychiatric symptoms of Long COVID.
  • Thromboinflammation: The results show a clear state of "coagulopathy" (clotting issues). Elevated D-dimer and Prothrombin time, combined with lower total protein, suggest the liver's ability to regulate blood chemistry is compromised, potentially leading to the "microclots" often discussed in Long COVID circles.
  • Iron Overload: Elevated ferritin levels were found, which can lead to a state of iron overload. This is linked to the chronic fatigue and "brain fog" that define the patient experience.

Key Results

The meta-analysis found statistically significant differences in several key areas:

  • Liver Damage Index: Showed a significant increase in LC patients (SMD = 0.553).
  • Enzyme Elevations: Alanine aminotransferase (ALT, SMD = 0.615) and Gamma-glutamyl transferase (GGT, SMD = 0.969) showed moderate to large effect sizes.
  • Synthetic Function: Total protein levels were significantly lower (SMD = -0.326), indicating the liver's reduced capacity to produce essential proteins.
  • Clotting Markers: Significant increases were noted in Prothrombin time (SMD = 0.633) and D-dimer (SMD = 0.359).

Limitations and Future Work

The authors acknowledge several hurdles:

  • Confounding Medications: Many patients received intensive drug treatments (antivirals, antibiotics) during the acute phase, which are themselves hepatotoxic. Most source studies did not provide enough data to decouple drug-induced damage from virus-induced damage.
  • Lifestyle Factors: Data on alcohol consumption and pre-existing conditions like Non-Alcoholic Fatty Liver Disease (NAFLD) were often missing.
  • Need for Long-Term Tracking: While the 3–6 month window is informative, future research needs to follow patients for 12+ months to see if this damage eventually resolves or leads to permanent scarring (fibrosis).

The paper concludes by urging clinicians to incorporate liver monitoring into standard Long COVID treatment strategies to mitigate long-term risks.

Summary updated

Abstract

Abstract Background Long COVID (LC) presents with complex pathophysiology, affecting multiple organs and producing a range of symptoms, from neuropsychiatric disturbances to multi-organ dysfunction. Liver damage has emerged as a notable feature, yet no systematic review or meta-analysis has comprehensively evaluated the biomarkers confirming liver injury in LC patients. Objective The present study aims to examine blood-based biomarkers of liver damage in LC disease. Methods A search of PubMed, Google Scholar, SciFinder, and SCOPUS identified 61 eligible studies, including 7172 participants, with 3404 LC patients and 3768 controls. Results Our analysis identified a significant increase in the liver damage index among LC patients, with a moderate effect size (standardized mean difference, SMD = 0.553; confidence intervals; 95% CI: 0.305–0.760) compared to normal controls. Additionally, LC patients exhibited marked elevations in alanine aminotransferase (SMD = 0.615; 95% CI: 0.351;0.878), aspartate aminotransferase (SMD = 0.352; 95% CI: 0.068;0.637), gamma-glutamyl transferase (SMD = 0.969; 95% CI: 0.194;1.745), and lactate dehydrogenase (SMD = 0.666; 95% CI: 0.332;0.999) activities. Moreover, significant reductions were observed in total protein (SMD = -0.326; 95% CI: -0.631; -0.021) and increases in prothrombin time (SMD = 0.633; 95% CI: 0.077;1.189), ferritin (SMD = 0.437; 95% CI: 0.069;0.805), D-dimer (SMD = 0.359; 95% CI: 0.144;0.573). Further studies are needed to confirm reductions in platelet count and elevations in fibrinogen levels Conclusion This study suggests that LC is associated with persistent liver damage and coagulopathy, highlighting the need to incorporate liver injury into treatment strategies to reduce potential risks.
BibTeX
@misc{almulla2024elevated,
  author = {Abbas F. Almulla and Yanin Thipakorn and Yingqian Zhang and Michael Maes},
  title = {{Elevated Liver Damage Biomarkers in Long COVID: A Systematic Review and Meta-Analysis}},
  year = {2024},
  doi = {10.1101/2024.10.18.24315698},
  howpublished = {medRxiv},
}
Figure