Association of Adjuvant S Adenosyl Methionine with Ursodeoxycholic Acid in Alcoholic Hepatitis: An Exploratory Prospective Observational Study

Abstract

Background: Alcoholic hepatitis (AH) is a serious medical condition with limited treatment options and significant short-term mortality risk. While S‑adenosyl methionine (SAMe) and ursodeoxycholic acid (UDCA) individually exhibit hepatoprotective properties, their combination remains inadequately evaluated in AH. Aim: This exploratory study aimed to assess the association of adding SAMe to UDCA on biochemical and clinical outcomes in patients with AH. Methodology: In this prospective observational cohort study conducted at a tertiary healthcare facility in India, 100 patients with AH (50 per group) were enrolled. Group A received SAMe (800–1200 mg/day) in addition to UDCA (15–20 mg/kg/day), whereas Group B received UDCA alone for 12 weeks. All patients received standard medical care, including corticosteroids when indicated (Maddrey's discriminant function >32) and nutritional support. The primary outcome was biochemical response at week 12, defined as either a ≥50% reduction in serum bilirubin from baseline or a serum bilirubin level <2 mg/dL. Secondary outcomes included MELD score improvement (≥3-point reduction), 90-day mortality, and adverse events. Findings: Baseline characteristics were balanced between groups. The primary outcome was achieved in 60% of the SAMe + UDCA group versus 40% of the UDCA monotherapy group (absolute difference: 20%, 95% CI: 1–39%; p = 0.045; number needed to treat = 5). MELD score improvement (≥3 points) was observed in 56% versus 36% (p = 0.04; adjusted OR = 2.45, 95% CI: 1.05–5.72). Mean bilirubin reduction was 58.2% versus 41.5% (p = 0.036). Ninety-day mortality was 10% versus 16% (p = 0.38). Adverse event rates were comparable between groups (24% versus 16%; p = 0.31), with no unexpected safety signals. Summary: The addition of SAMe to UDCA was associated with improved biochemical response and MELD scores in patients with AH, without excess adverse events. However, because of the observational, non-randomised design, these findings should be interpreted with caution, as they are subject to potential confounding and selection bias and do not establish causality. A definitive randomised controlled trial is warranted to confirm these hypothesis-generating results.