TUDCA for NAFLD / MASLD
There is currently no reliable human evidence that TUDCA improves NAFLD/MASLD: not a single completed clinical trial has tested it against liver fat, inflammation, or fibrosis, so any liver-health claims on supplement labels run well ahead of the science.
Why this grade7-layer evidence engine
The grade is Unverified and Disputed because the directly applicable human data are essentially absent. The only relevant clinical trial (Kars 2010, PMID 20522594) randomized just 20 obese adults to TUDCA for four weeks and measured insulin sensitivity, not a fatty-liver outcome. Hepatic and muscle insulin sensitivity rose about 30%, but this is a surrogate for insulin resistance, and the trial found no change in ER-stress markers, which undercuts the chemical-chaperone mechanism that is meant to protect the liver.
Every NAFLD-specific efficacy signal comes from mice. Rodent studies report less steatohepatitis and fibrosis (PMID 24865256), a better gut barrier (PMID 29139555), and reduced hepatic fat via microbiota and bile-acid pathways (PMID 39193771). These are promising mechanistically but do not establish a human effect, and the closely related drug UDCA failed to improve liver histology in NASH trials.
Authorities reinforce caution rather than endorsement. The FDA recognizes TUDCA only as a drug for certain cholestatic liver diseases, and the WHO assigns it a drug name (ursodoxicoltaurine), not a supplement claim; the NHS lists it merely as an ALS trial agent. No major clinic or liver society (Mayo, Cleveland Clinic, Harvard, AASLD) recommends it for NAFLD. For fatty liver, weight loss and a Mediterranean diet remain the evidence-based first steps.
Scoring transparency
All scores computed by a 7-layer evidence engine — fully auditable▸View the full decision path (audit trail)
- compute_raw_score — 加權公式: L2×0.30 + L3×0.25 + L5×0.25 + L11×0.10 + L1×0.10 = 0.455
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 無高階證據可裁決
- tier_strict_requirement_check — | C→U 因 scope.conflation_risk=true 且 L11 獨評較低 (B7-2 tier cap)
- detect_disputes — 偵測到 1 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status