Ursodiol

Limits on what this shows

• Null findings: None. No retrieved study found mitochondrial damage from ursodiol.

Regulatory status

No mitochondrial warning in the label sections retrieved [1][2].

Population evidence

Botla 1995 [5]: isolated rat liver mitochondria and hepatocytes; UDCA inhibited the mitochondrial permeability transition induced by glycochenodeoxycholate (Tier B). Rodrigues 1998 [6]: isolated rat hepatocytes, HuH-7, HepG2 and other cell lines; UDCA reduced the permeability transition by 40 to 50% when induced by deoxycholic acid and phenylarsine oxide, and reduced transmembrane potential change and reactive oxygen species (Tier B). Serviddio 2004 [7]: in rats with secondary biliary cirrhosis UDCA prevented collapse of mitochondrial membrane potential and preserved mitochondrial cardiolipin and phosphatidylserine (Tier B). Human: Lou 2026 [15] pilot, UDCA responders showed improved mitochondrial function with higher serum itaconate while non-responders had persistently raised serum pyruvate read as TCA cycle dysfunction (mitochondrial function inferred from serum metabolites; Tier A for the cohort but small and observational).

Mechanism

Present and protective: inhibition of toxic-bile-acid-induced permeability transition [5][6]. It operates under a toxic bile acid challenge, not at baseline.

Clarification

PBC is defined partly by antimitochondrial antibodies [24]. These are disease autoantibodies, not evidence that ursodiol harms mitochondria.

Cohort test

Ursodiol was the compound studied in [5], [6] and [7]; it was in the human cohort in [15].

Counter-evidence held on record

Counter-evidence and confounding Models used toxic bile acid challenges that mirror cholestasis, so the findings may not apply outside cholestatic disease. The human data rely on serum metabolite proxies, and mitochondrial dysfunction in non-responders is entangled with disease activity. Framing caution Reverse-direction finding. Not a mitochondrial benefit claim and not a deficiency claim.

Limits on what this shows

• Null findings: None for the human direction. No retrieved study found a rise in oxidative stress markers in treated patients.

Regulatory status

No oxidative stress warning in the label sections retrieved [1][2].

Population evidence

Kawata 2010 [8]: 13 PBC patients with paired pre- and post-UDCA liver biopsies; 8-OHdG-positive hepatocytes and bile duct cells decreased with improvement of hepatic injury, total and phosphorylated Nrf2 rose (with more nuclear Nrf2 in bile duct cells), and thioredoxin and thioredoxin reductase 1 increased (Tier A, ursodiol in cohort, uncontrolled). Serviddio 2004 [7]: male Wistar rats with secondary biliary cirrhosis; UDCA partly prevented hepatic glutathione depletion and oxidation and prevented increases in mitochondrial peroxide production and lipid peroxidation markers (Tier B). Dallio 2024 [9]: 41 UDCA-naive PBC patients, 40 MASLD, 52 HBV and 10 healthy controls; in advanced fibrosis PBC, SOD1, SOD2 and GPx1 expression was downregulated, showing oxidative imbalance rises with disease before ursodiol (context, Tier C).

Mechanism

Present and protective: reduced toxic bile acid injury and Nrf2 pathway activation [8]; restored glutathione synthesis enzymes in rats [7].

Contrary evidence

Fuentes-Broto 2009 [10]: rat liver homogenates and cell membranes incubated with 1 mM of several bile acids plus an iron and ascorbate challenge; UDCA enhanced lipid and protein oxidation markers, less than taurolithocholic acid, and melatonin reduced the effect (Tier B, UDCA specific, in vitro). The relevance of the 1 mM concentration to treated humans was not established in the record retrieved.

Cohort test

Ursodiol was the studied drug in [7], [8] and [10]; [9] enrolled only treatment-naive patients and is disease context.

Counter-evidence held on record

Counter-evidence and confounding In [8] improvement in injury itself lowers oxidative markers, so the antioxidant effect cannot be separated from disease improvement; 13 patients, no control arm. Reverse-direction findings follow the same grading rules and sit at yellow. Framing caution Not an antioxidant-benefit claim and not a deficiency claim. Ursodiol should not be described as depleting antioxidants.

Limits on what this shows

• Null findings: Hepatic copper accumulation was unchanged by 14 to 39 months of ursodiol in 5 PBC patients [21].
• At-risk subgroups: People with more advanced stage cholestatic disease (stage 3 accumulation in 8 of 18) [21]. This is a disease finding.

Regulatory status

No copper mention in the ursodiol label text retrieved [1] (via summarizing fetch tool).

Population evidence

Takeshima 1993 [21]: liver biopsies from 18 PBC patients; copper accumulation was staged (8 patients stage 1, 2 stage 2, 8 stage 3). Five patients had repeat biopsy after 14 to 39 months of ursodeoxycholic acid 600 mg daily: blood chemistry improved but hepatic copper accumulation was unchanged in all five (Tier A for ursodiol, but 5 patients, uncontrolled, histochemical rather than quantitative serum copper). Dastych 2021 [22]: 20 PBC, 32 PSC and 40 control subjects; copper differed significantly from controls (p=0.021; the record retrieved did not state the direction) and was significantly lower in patients whose cholestasis markers normalized with treatment (UDCA was the named treatment; markers normalized in 18 of 52, 35%); levels normalized 6 months after transplantation. Ursodiol was not isolated from the disease in this cohort.

Mechanism

Disease-driven. The retained copper sits in hepatocyte lysosomes in cholestatic liver [21] and falls as cholestasis resolves [22]. No mechanism by which ursodiol alters copper handling was found.

Cohort test

Ursodiol was in the cohort in [21] (5 treated patients) and named as treatment in [22], but neither stratified copper by drug exposure against untreated or placebo groups. Context flag, not a drug-attributed grade.

Counter-evidence held on record

Counter-evidence and confounding Retention precedes treatment and tracks cholestasis, so the signal is confounded by indication by design. The sample behind the ursodiol-specific observation is 5. Framing caution This is an excess or retention signal in the underlying disease, not a deficiency. It must not imply that copper intake should increase; copper supplementation would be the hazard, not the remedy (Hard Rule 5). Copper is kept as an entry separate from Iron (Hard Rule 7).

Limits on what this shows

• Null findings: Liver iron unchanged [23]. This yellow records a null Tier B result under the rule that any Tier B source puts an entry at yellow minimum; it does not assert a direction of effect.
• At-risk subgroups: Not defined; no human data.

Regulatory status

No iron mention in the label text retrieved [1][2].

Population evidence

Loreal 1997 [23]: Sprague-Dawley rats on a normal or iron-supplemented diet (5% carbonyl iron), UDCA by gavage; hepatic iron concentration and distribution were not substantially modified, apart from a transient lower iron score at one month in the iron-supplemented group. The authors concluded UDCA is unlikely to increase intestinal iron absorption. Tier B, UDCA specific, animal. No human study was opened.

Mechanism

The authors' premise was that bile salts can increase iron absorption; this was not borne out for UDCA in rats. No depletion mechanism is named.

Counter-evidence held on record

Counter-evidence A rat study cannot support a human claim. Iron deficiency and anemia are different findings (Hard Rule 2) and no human anemia data for ursodiol were retrieved. Kept independent of Copper (Hard Rule 7). Framing caution No iron depletion or iron supplementation claim is made. Excess iron carries its own risks.

Limits on what this shows

• Null findings: No placebo-controlled microbiome study of ursodiol was retrieved, so no clean null exists.
• At-risk subgroups: Poor or non-responders show persistent dysbiosis in [13][15]; this is a disease-response marker, not a drug harm.

Regulatory status

No microbiome statement in the label text retrieved, other than that non-absorbed ursodiol reaches the colon and is mostly 7-dehydroxylated to lithocholic acid by bacteria [1].

Population evidence

Tang 2018 [11]: 60 UDCA-naive PBC patients and 80 controls, a validation cohort of 19 PBC and 34 controls, and 37 PBC patients sampled before and after 6 months of UDCA; PBC had reduced within-person diversity (p=0.03), six PBC-associated genera reversed after 6 months and Faecalibacterium, depleted at baseline, increased. Han 2022 [12]: 47 PBC patients treated with UDCA for 12 months (20 with elevated total bilirubin, 27 normal); lower diversity and fewer Gemmiger, Blautia, Anaerostipes and Coprococcus in the elevated-bilirubin group. Han 2024 [13]: 25 PBC patients (14 good responders, 11 poor responders); poor responders had decreased Gemmiger, increased Ruminococcus and altered bile acid and all-trans-retinoic acid metabolite signatures. Lou 2026 [15]: longitudinal pilot; non-responders had persistent dysbiosis with depletion of butyrate producers, expansion of Candida albicans and Collinsella enrichment, while responders showed microbiota remodeling. Liu 2024 [14] (Med): microbiome patterns relate to UDCA response; sample size was not retrieved.

Mechanism

Plausible through the bile acid and microbiome axis: ursodiol and its bacterial metabolites share the colon with the microbiota [1]. Direction of causation is not shown; microbiome shifts after treatment may follow improved cholestasis rather than the drug.

Cohort test

Ursodiol was in the cohort for every paper above (all or most patients were UDCA-treated PBC). The cohorts are single-arm or response-stratified, so the data support association with treatment in PBC only. No class extrapolation was used.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication: PBC itself causes dysbiosis at baseline [11]. Before-and-after designs lack a placebo arm, diet and time are uncontrolled, and the authors of [11] and [14] are from the same institutional group and [12] and [13] share authors, so cohort overlap is possible and independence is not assumed. Framing caution This is a reverse-direction finding, not a dysbiosis claim and not a probiotic or microbiome-repair claim for ursodiol.

Limits on what this shows

• Null findings: No clinical motor benefit by MDS-UPDRS-III [16].

Regulatory status

No energy-metabolism statement in the label sections retrieved [1][2].

Population evidence

Movement Disorders 2023 [16]: phase II randomized, double-blind, placebo-controlled trial, 30 participants with Parkinson's disease, UDCA 30 mg/kg daily (2:1 randomization), 48 weeks. Midbrain 31P-MRS showed an increase in the Gibbs free energy of inorganic phosphate versus placebo, which the authors read as improved ATP hydrolysis; MDS-UPDRS-III showed no difference between groups; adverse events were mild transient GI events, more frequent on UDCA. The dose is about twice the PBC label dose of 13 to 15 mg/kg/day [2]. Lou 2026 [15]: raised serum pyruvate in UDCA non-responders, interpreted as impaired TCA cycle energy metabolism.

Mechanism

Proposed, from protection of the mitochondrial membrane [5][6]. Not shown to change ATP synthesis in PBC or hepatocytes.

Cohort test

Ursodiol was the studied drug in [16], in a non-hepatic population at about twice the labeled PBC dose; applicability to usual treatment populations is unproven.

Counter-evidence held on record

Counter-evidence Single small trial, one surrogate imaging outcome, read via an OpenAlex abstract summary and not full text. The pyruvate observation [15] is disease-confounded. ATP production is kept separate from Mitochondrial burden and Oxygen delivery/utilization (Hard Rule 7); shared sources are disclosed in the Overlap map. Framing caution Reverse-direction finding. Not an energy-boosting claim and not a deficiency claim.

Limits on what this shows

• Null findings: No retrieved ursodiol label section or study reports anemia or a hemoglobin fall as a drug effect.

Regulatory status

The warnings retrieved (5.1 abnormal liver tests, 5.2 enteroliths) contain no hematologic item [1]; this is from a summarizing fetch, not a full label read.

Population evidence

Duszynski 2020 [17]: case-control, 13 non-cirrhotic PBC, 7 cirrhotic PBC and 11 healthy controls; PBC had reduced tissue oxygen saturation (StO2) and higher deoxygenated hemoglobin; in early patients, complete biochemical response to UDCA correlated with lower total (tissue) hemoglobin concentration and normalized cerebrovascular coherence. The hemoglobin in that statement is a near-infrared tissue measure, not a blood count. Quality-of-life scores did not correlate with hypoxia.

Mechanism

No mechanism by which ursodiol would reduce oxygen delivery was found. The hypoxia is attributed to the disease state and was not shown to be drug-caused.

Cohort test

Ursodiol response was analyzed within a PBC cohort; no ursodiol-versus-no-ursodiol comparison of oxygenation exists in the record retrieved.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication: PBC with or without cirrhosis can itself alter oxygenation. Hypoxia and anemia are outcomes, not nutrient deficiencies (Hard Rule 2). Any anemia effect would route through the Iron entry and is not counted twice. Framing caution Context flag only. Do not describe ursodiol as impairing oxygen delivery.