Gut microbiome and plasma multi-omics signatures at admission differ between hepatitis A-related and drug-induced acute liver failure, with distinct pathobiont enrichment, metabolic disruptions, and inflammatory markers associated with disease severity and 90-day mortality in ALF-A.
Key Findings
Results
ALF-A patients had greater clinical severity at admission compared to ALF-D patients.
90-day mortality was 33% in ALF-A versus 18.4% in ALF-D.
KCH criteria met by 21% of ALF-A versus 10.5% of ALF-D patients.
SOFA scores were higher in ALF-A (7.97 ± 2.89) versus ALF-D (5.87 ± 2.32).
SIRS was present in 60.6% of ALF-A versus 18.4% of ALF-D patients.
Mechanical ventilation required in 39% of ALF-A versus 21% of ALF-D patients.
Results
ALF-A patients were enriched at admission (D0) for pathobionts producing lactate, ammonia, bile salt hydrolase, and histamine.
Pathobionts enriched in ALF-A included Enterococcus, Ruminococcus gnavus, Flavonifractor, and Thomasclavelia.
These bacteria correlated positively with disease severity measures (|r| > 0.4, p < 0.05).
These pathobionts showed higher baseline abundance in ALF-A non-survivors compared to survivors.
Profiling was performed from stool samples collected at Day 0 (admission day).
Results
Histamine accumulation was elevated and tryptophan and butyrate metabolism were reduced at Day 0 in ALF-A, with worsening by Day 7 in non-survivors.
Histamine accumulation inversely correlated with disease severity at D0 in ALF-A.
Reduced tryptophan metabolism inversely correlated with severity.
Reduced butyrate metabolism inversely correlated with severity.
These metabolic alterations worsened in ALF-A non-survivors by Day 7.
Plasma metabolites were analyzed at D0, D3, D5, and D7.
Results
Pro-inflammatory cytokines and intestinal fatty acid binding protein 2 (IFABP2) were elevated at Day 0 in ALF-A, correlated with severity, and remained elevated through Day 7 in non-survivors.
IFABP2 is a marker of gut barrier dysfunction.
Elevated cytokine and IFABP2 levels at admission correlated positively with severity scores.
Levels remained increased through Day 7 specifically in ALF-A non-survivors.
Plasma cytokines and barrier markers were measured at D0, D3, D5, and D7.
Results
ALF-D non-survivors showed a distinct biological profile compared to ALF-A non-survivors, retaining commensals and exhibiting increased fatty acid and bile acid biosynthesis with low pro-inflammatory cytokine levels.
ALF-D non-survivors retained commensal bacteria, in contrast to the pathobiont enrichment seen in ALF-A.
ALF-D non-survivors showed increased bile acid biosynthesis.
Pro-inflammatory cytokine levels were low in ALF-D non-survivors, contrasting with the elevated levels seen in ALF-A non-survivors.
Results
Integrated gut and plasma multi-omics features in ALF-A were associated with disease severity and 90-day outcomes and identified candidate biomarkers.
Associations with 90-day mortality were assessed using severity-adjusted association analyses.
The study integrated stool bacterial profiling with plasma cytokines, metabolites, and barrier markers.
Study included ALF-A (n=33), ALF-D (n=38), acute viral hepatitis A (AVH-A, n=27), and healthy controls (HC, n=20).
Patient population had mean age 26 ± 9 years; 50.7% male, 49.3% female.
The identified features were described as candidate biomarkers requiring future validation.
What This Means
This research suggests that when the liver fails suddenly due to hepatitis A infection versus drug toxicity, the biological processes driving the disease are fundamentally different. In patients with hepatitis A-related acute liver failure (ALF-A), the gut was overrun by harmful bacteria that produce toxic substances like ammonia, histamine, and lactate, while beneficial gut processes (like butyrate and tryptophan metabolism) were suppressed. At the same time, markers of gut leakiness and widespread inflammation were elevated from the very first day of hospital admission. These patients were also sicker overall, with higher rates of organ failure, need for mechanical ventilation, and 33% death within 90 days compared to 18.4% in drug-induced cases.
In contrast, patients whose liver failed due to drugs had a different profile: their gut bacteria were less disrupted, their inflammatory response was milder, but they showed alterations in fat and bile acid metabolism instead. This suggests that drug-induced liver failure kills through different biological mechanisms than viral liver failure. Importantly, the harmful gut bacterial patterns and inflammatory signals seen in hepatitis A patients who ultimately died were already detectable on the day of admission, before the disease fully declared itself.
This research suggests that measuring gut microbiome composition and plasma markers of inflammation and gut barrier integrity at the time of hospital admission could help identify which acute liver failure patients — particularly those with hepatitis A — are at highest risk of dying within 90 days. These findings point toward potential new biomarkers that, if validated in larger studies, could help clinicians make faster and more accurate decisions about treatment intensity and organ transplant priority.
Saxena A, Maiwall R, Meena B, Maras J, Suroliya V, Kumar G, et al.. (2026). Admission Gut and Plasma-Derived Signatures Associated With Severity and 90-day Outcome in Hepatitis A-Related and Drug-Induced Acute Liver Failure.. Journal of medical virology. https://doi.org/10.1002/jmv.71150