Warfarin

Limits on what this shows

• Key negative or null findings: Cerqua C, et al., 2019, PMID 31024065, examined whether vitamin K2 can substitute for coenzyme Q10 as an electron carrier in mammalian, rather than Drosophila, mitochondria. This review located the study's title and subject only and could not open its full text to determine whether the fly finding generalizes to mammalian tissue; this is recorded as an open, unresolved question, not a confirmed replication or a confirmed refutation.
• At-risk subgroups: None defined by any source located in this review.
• Dose and duration: Not established for any of the fly studies located; full texts could not be opened to confirm dosing.

Regulatory status

Mitochondria and mitochondrial function are not named anywhere in the warfarin prescribing information.

Population evidence

None in humans. Tier B with warfarin in the cohort: Lavrenova A, Kozlova M, Klychnikov O, Nefedova L. "Sex-Related Effect of Chronic Doses of Warfarin and Menadione on Drosophila melanogaster." International Journal of Molecular Sciences 2026;PMID 42450292. This study examines warfarin's role in blocking conversion of vitamin K3 (menadione) to endogenous vitamin K2 (menaquinone-4) in flies. This review located the study's existence and subject matter through a PubMed search listing but could not open the full text (repeated CAPTCHA and empty-page-shell blocks) to confirm dose, direction, or statistics, and none is asserted here beyond what the title and abstract snippet state.

Mechanism

Plausible but unconfirmed in warfarin-treated tissue. Vos M, et al. "Vitamin K2 is a mitochondrial electron carrier that rescues pink1 deficiency." Science 2012;PMID 22582012 (Tier B, mechanism reference, no drug administered): established, in Drosophila, that vitamin K2 (menaquinone-4) can function as an electron carrier in the mitochondrial respiratory chain, a role entirely separate from its clotting-cofactor function. Because warfarin's whole mechanism of action (Headline finding, Pharmacokinetic basis) is inhibition of the VKORC1-mediated recycling that regenerates active vitamin K, including menaquinone-4, from its epoxide form, a route by which warfarin could secondarily reduce the supply of the mitochondrial electron-carrying form of vitamin K is biologically plausible. This chain of reasoning has not been followed end to end by directly measuring mitochondrial respiration in a warfarin-exposed organism in any source located for this review; it is a hypothesis with real mechanistic grounding, not a demonstrated warfarin effect.

Counter-evidence held on record

The chain of reasoning here, warfarin blocks VKORC1, VKORC1 regenerates menaquinone-4, menaquinone-4 carries electrons in the mitochondrial chain in flies, has been demonstrated at each individual link in isolation and in different organisms or systems, but never followed end to end in a single warfarin-exposed system. It is worth taking seriously precisely because it uses the same molecular target already established as warfarin's own mechanism in the Headline finding, applied to a non-clotting vitamin K function, exactly as the existing osteocalcin/bone entry above does for a different tissue. Do not state that warfarin damages or depletes cellular energy production; this has never been measured in a person taking the drug. No coenzyme Q10 or other mitochondrial supplement recommendation follows from this entry, and none is made.

Limits on what this shows

• Key negative or null findings: No confirmed contradicting Tier A or Tier B null is on record in this review for warfarin and oxidative stress, unlike the mitochondrial-burden and ATP-production entries below. That absence reflects incomplete literature access in this review as much as a genuinely clean, unrebutted signal, and is stated as such rather than treated as strengthening the grade.
• At-risk subgroups: None defined by any source located in this review.

Regulatory status

Oxidative stress, reactive oxygen species, and antioxidant enzymes are not named anywhere in the warfarin prescribing information.

Population evidence

Atia MM, Mahmoud HAA, Wilson M, Abd-Allah EA. "A comprehensive survey of warfarin-induced hepatic toxicity using histopathological, biomarker, and molecular evaluation." Heliyon 2024;PMID 38440292 (Tier B, warfarin in cohort). The PubMed search listing located for this review directly quotes the study's own finding: "Warfarin exposure caused increased oxidative stress; increased LPO and decreased GSH, SOD, CAT" (lipid peroxidation up, glutathione, superoxide dismutase and catalase down). This is a toxicity survey; this review could not confirm the dose used or whether it maps onto a human therapeutic-equivalent range, and that is treated as an open limitation, not assumed either way. Lalaoui MZ, El Midaoui A, de Champlain J, Moreau P. "Is there a role for reactive oxygen species in arterial medial elastocalcinosis?" Vascular Pharmacology 2007;PMID 17127099 (Tier B, warfarin in cohort): Sprague-Dawley rats given warfarin 20 mg/kg/day plus vitamin K1 15 mg/kg/day for 4 weeks, the standard experimental protocol used throughout the vascular-calcification literature to produce medial arterial calcification, a separate, label-documented warfarin effect (see calciphylaxis, Pharmacokinetic basis). This study tested whether the antioxidant alpha-lipoic acid could prevent the calcification; the actual result could not be confirmed by opening the full text in this review, so it is recorded as evidence that the field treats warfarin-induced calcification as a reactive-oxygen-species question, not as confirmation that the antioxidant succeeded. Lavrenova A, Klychnikov O, Ioutsi V, Rodin I, Luneva O, Nefedova L. "Effect of Warfarin on Lifespan and Oxidative Stress Tolerance of Drosophila melanogaster." International Journal of Molecular Sciences 2025;PMID 40429949 (Tier B, warfarin in cohort): warfarin dosed directly to flies with oxidative stress tolerance as a stated endpoint; full text could not be opened. A human comparative study, Reis HJ, Costa LBX, Martins GL, et al. "Rivaroxaban versus warfarin: differential effects on oxidative stress and fibrinolytic markers in atrial fibrillation." Frontiers in Pharmacology 2026;PMID 41635486, appears to measure oxidative stress markers directly in warfarin-treated atrial fibrillation patients against a direct oral anticoagulant comparator; this review located its title and subject only, and neither its direction nor its magnitude is asserted here. It is not counted as the Tier A source this entry would need to advance to orange.

Mechanism

Plausible in two independent ways. First, calciphylaxis and vascular calcification (DailyMed §5.3, quoted in the Pharmacokinetic basis) are established, separately warfarin-mechanism-linked adverse effects operating through undercarboxylation of matrix Gla protein, a non-clotting vitamin K dependent protein, and the vascular-calcification literature independently associates that process with vascular oxidative stress; this is the same "different non-clotting vitamin K dependent protein" structure already used for the osteocalcin/bone entry above, applied here to a second protein, a different tissue, and a different outcome. Second, warfarin is metabolized via CYP2C9, CYP1A2, and CYP3A4 (Pharmacokinetic basis), and cytochrome P450-mediated oxidation of xenobiotics is a generic route to reactive oxygen species that applies to any CYP substrate; this route is not warfarin-specific and is not separately evidenced for warfarin in this review.

Counter-evidence held on record

None confirmed; see key negative/null findings above for why that absence is not treated as reassurance. Do not state that warfarin causes oxidative stress as an established human finding; the human magnitude is unverified. No antioxidant, selenium, or vitamin E supplement is recommended by this entry. Do not conflate this with vitamin K's own antioxidant chemistry, which is a separate question this review did not investigate and which would run in the opposite direction from a depletion in any case.

Limits on what this shows

• Key negative or null findings: No study located in this review measured serum iron, ferritin, or transferrin saturation directly in warfarin-treated patients who were not experiencing a bleeding event. By extension, therapeutic warfarin at a stable, in-range INR is not shown to move iron status independent of an actual bleeding episode.
• At-risk subgroups: The same bleeding risk factors named under oxygen delivery above; additionally, patients with a diet marginal in iron or a pre-existing anemia of any cause, in whom a bleeding episode on warfarin would be expected to unmask or worsen iron deficiency sooner.

Regulatory status

Iron is not named anywhere in the warfarin label; this was directly confirmed by re-opening the full label during this review. The label's bleeding warnings, quoted in the oxygen-delivery entry above, are the only labeled route by which an iron-status effect could occur.

Population evidence

The same Hurlen et al. 2006 WARIS-II subgroup analysis (Tier A, warfarin in cohort) cited above, here centered on the iron-deficiency arm of its stated design ("the occurrence of occult bleeding and iron deficiency"). The same verification limit applies: this review could not open the full text, and no specific incidence or comparison figure is asserted.

Mechanism

Blood loss via hemorrhage is the only established route. Absent are gastrointestinal absorption interference, chelation (no evidence located that warfarin binds dietary or supplemental iron), and renal iron wasting; none of these appears in the label or in any source located for this review.

Counter-evidence held on record

No consumer or clinical source located in this review claims warfarin depletes iron directly (see the new Claim audit table below); the finding here is bleeding-mediated and bounded, not a general depletion claim. No study located contradicts the bleeding-only mechanism with a competing route. Never present this as a reason to take iron prophylactically while on warfarin without a bleeding event or a clinician's workup; unexplained iron deficiency in a warfarin-treated patient needs the bleeding source investigated, not simply the iron replaced, since iron repletion without finding the source risks masking a new or worsening lesion. No iron dose is recommended anywhere in this dossier.

Limits on what this shows

• Why no grade: Yellow requires either mechanistic plausibility demonstrated at Tier B or class extrapolation at Tier C where at least one class member has real supporting data for the effect in question. Neither condition is met. Warfarin is a single-molecule anticoagulant class (vitamin K antagonist) with no other member commonly used in humans against which to extrapolate, unlike the zinc/ACE-inhibitor-class situation this dossier series has documented for a different drug. The one candidate source this review located, a 1994 pharmaceutical chemistry paper, could not be verified even for relevance (see below), so it does not meet the bar for Tier B mechanistic evidence either. White requires a circulating claim to rebut; no source reviewed for this dossier's existing Fails-verification section, or located freshly for this expansion, makes a copper claim for warfarin. The category therefore sits outside all four tiers, and this dossier prints it rather than dropping it.
• What would fill it: A study directly measuring serum, plasma, or a validated functional marker of copper status in a defined warfarin-treated human or animal cohort, with warfarin named as a study arm rather than pooled into "anticoagulants." Failing that, a verified, biologically relevant in vitro copper-binding study using warfarin itself would establish at least Tier B mechanistic plausibility and move this category to yellow.

What was searched

A PubMed search for warfarin together with copper, conducted during this review, returned results including one candidate directly addressing a chemical relationship: Karaivanova VD, Manolov I, Minassyan ML, Danchev ND, Samurova SM. "Metal complexes of warfarin sodium." Pharmazie 1994;PMID 7838871. The remaining results discussed warfarin and copper in unrelated contexts, such as copper intrauterine devices used by patients who separately take warfarin, and are not evidence of any drug-nutrient relationship.

The one candidate source, and why it does not count

This review could not open the full text of the Karaivanova 1994 paper to confirm whether copper specifically, as opposed to other metals, was among the complexes studied, or whether the paper addresses biological chemistry at all rather than pharmaceutical salt synthesis for formulation purposes. A source that cannot be confirmed relevant is not used to support any grade under this dossier's citation-integrity rule.

Mechanistic context, for completeness

Unlike the ACE-zinc relationship documented elsewhere in this dossier series, warfarin's molecular target, vitamin K epoxide reductase complex subunit 1 (VKORC1), is not a copper-dependent or copper-binding enzyme in any pathway described in the label or in standard pharmacology references, so there is no structural rationale analogous to the ACE-inhibitor/zinc-metalloenzyme case to motivate a copper hypothesis here.

Counter-evidence held on record

There is none to hold, which is the finding. Nothing supports a copper-depletion claim for warfarin and nothing rules one out. Absence of evidence is not evidence of safety. Do not publish "warfarin does not affect copper status," which the literature does not support either way, and do not publish a copper-testing or copper-supplement prompt attached to this drug. If anything surfaces at all, it should say only that this has not been studied.

Limits on what this shows

• Key negative or null findings: Section 7.4's own statement that "clinical pharmacokinetic studies have not shown consistent effects" of antibiotics on warfarin's plasma concentration argues against a uniform pharmacokinetic explanation across the antibiotic class, indirectly consistent with, but not proof of, a shared gut-flora mechanism instead.
• At-risk subgroups: Patients starting a new antibiotic course, especially broad-spectrum agents, and patients with the malabsorptive conditions named in the label (sprue). This is the same subgroup already named under at-risk subgroups in the existing "Dietary vitamin K intake changes warfarin's own effect" entry above.

Regulatory status

DailyMed §5.8 names "Infectious diseases or disturbances of intestinal flora (e.g., sprue, antibiotic therapy)" directly among factors that increase warfarin's anticoagulant effect and bleeding risk. Section 7.4 separately states: "There have been reports of changes in INR in patients taking warfarin and antibiotics or antifungals, but clinical pharmacokinetic studies have not shown consistent effects of these agents on plasma concentrations of warfarin," favoring a non-pharmacokinetic explanation for these reports over a uniform drug-metabolism one, consistent with, though not proof of, a gut-flora vitamin K supply mechanism.

Population evidence

Three warfarin-specific case reports of INR elevation following an antibiotic course. Davydov L, Yermolnik M, Cuni LJ. "Warfarin and amoxicillin/clavulanate drug interaction." Annals of Pharmacotherapy 2003;PMID 12639164. Jones CB, Fugate SE. "Levofloxacin and warfarin interaction." Annals of Pharmacotherapy 2002;PMID 12243605, four patients. Goodchild JH, Donaldson M. "A clinically significant drug interaction between warfarin and amoxicillin resulting in persistent postoperative bleeding in a dental patient." General Dentistry 2013;PMID 23823345. All Tier B, case-report level, warfarin specifically, consistent in direction (antibiotic exposure, INR rise). No study located in this review has measured gut microbiota composition or diversity directly in a warfarin-treated human or animal; that specific measurement is what this review searched for and did not find.

Mechanism

Named and plausible, though not proven for warfarin specifically over competing explanations. Gut bacteria synthesize menaquinones (vitamin K2) that contribute to the body's absorbable vitamin K pool, a physiological fact established independent of warfarin; antibiotic disruption of that flora is one proposed mechanism for antibiotic-associated INR elevation, alongside direct pharmacokinetic interactions specific to individual antibiotics, a competing, non-microbiome explanation that applies drug-by-drug and was not separately excluded in the three case reports above.

Counter-evidence held on record

No study located in this review directly measured gut microbiota composition in a warfarin-exposed subject, human or animal, so there is no direct evidence either confirming or excluding a microbiome-composition change; this entry rests on the label's own risk-factor language and case-report-level clinical evidence of the downstream INR effect, not on a microbiome measurement. This is not a new vitamin K deficiency claim; it is the same dietary-consistency logic already applied above to leafy greens, extended to antibiotic-disrupted gut flora as an additional source of vitamin K supply variability. Never suggest avoiding antibiotics when medically indicated, and never suggest a probiotic as a "fix"; INR monitoring around a new antibiotic prescription is the correct, medically-directed safeguard, consistent with the existing entry's at-risk-subgroup language above.

Limits on what this shows

• Key negative or null findings: The label frames "unexplained fall in hemoglobin" specifically as a sign of excessive anticoagulation under Overdosage, not as an expected finding at therapeutic INR. No source located in this review reports hemoglobin or hematocrit declining in warfarin-treated patients who are not bleeding, and none is asserted here.
• At-risk subgroups: Patients with supratherapeutic INR, concomitant antiplatelet or NSAID use, prior gastrointestinal bleeding or peptic ulcer disease, malignancy, the first month of therapy, age 65 or older, and the other bleeding risk factors named in §5.1.

Regulatory status

DailyMed Boxed Warning: "Warfarin sodium can cause major or fatal bleeding." Section 6, Adverse Reactions: "Most common adverse reactions to warfarin sodium are fatal and nonfatal hemorrhage from any tissue or organ." Section 10.1, Overdosage, Signs and Symptoms, lists "unexplained fall in hemoglobin" as a manifestation of excessive anticoagulant effect. Section 5.1 separately names anemia among risk factors that increase bleeding risk further, a risk-factor-for-bleeding direction distinct from the effect graded here.

Population evidence

Hurlen M, Eikvar L, Seljeflot I, Arnesen H. "Occult bleeding in three different antithrombotic regimes after myocardial infarction. A WARIS-II subgroup analysis." Thrombosis Research 2006;118(4):433–8 (Tier A, warfarin in cohort): a subgroup analysis of the WARIS-II randomized trial comparing occult bleeding across aspirin, warfarin, and combination aspirin-plus-warfarin therapy after myocardial infarction. The PubMed search listing located for this review directly quotes the study's own stated purpose, examining "the occurrence of occult bleeding and iron deficiency during these treatment modalities." Repeated attempts to open the full text in this review returned an empty page shell; no specific incidence figure or between-group comparison is asserted here, and that verification limit is stated rather than an number estimated.

Mechanism

Named and label-documented: hemorrhage from any tissue or organ, more likely in the first month of therapy and at supratherapeutic INR (§5.1). This is downstream of warfarin's clotting-factor-blocking mechanism of action described in the Headline finding and Pharmacokinetic basis above, not that mechanism itself. No gastric, hepatic, or renal pathway independent of bleeding is described anywhere in the label or in the literature located for this review.

Tier history

This tier was printed as empty at the September 6, 2026 review. That reasoning is preserved verbatim above and remains accurate for the nutrients it addressed. It was filled here by a parameter not previously screened: the label's own overdosage and adverse-reaction language on hemoglobin and hemorrhage, together with a warfarin-specific trial subgroup located by title.

Counter-evidence held on record

The label's own qualifier restricts this finding to excessive anticoagulation; no source located in this review describes a hemoglobin decline in a patient maintained within the therapeutic INR range. The WARIS-II subgroup finding carries the same structural limitation as several other sources in this expansion: without the full text, no magnitude can be stated, and none is claimed. Never present this as an iron-deficiency, B12, or erythropoietin-mediated anemia, and never as a routine consequence of taking warfarin as prescribed. The correct consumer framing is a monitoring flag: any unexplained fall in hemoglobin is a reason to contact the prescriber, who will check the INR and look for a bleeding source, not a reason to start an iron or vitamin supplement independently.

Limits on what this shows

• Key negative or null findings: Cerqua 2019 (PMID 31024065, described under mitochondrial burden) is the nearest candidate for a mammalian test of this pathway; this review could not open its full text to determine whether it found for or against mammalian generalization, so it is recorded as unresolved rather than as a null result.
• At-risk subgroups: None defined.
• Dose and duration: Not applicable; no warfarin dose or duration has been paired with a direct ATP measurement in any source located.

Regulatory status

ATP and energy metabolism are not named anywhere in the warfarin prescribing information.

Population evidence

None at any tier, in any species, with warfarin administered. Vos M, et al., 2012, PMID 22582012 (Tier B, mechanism reference, no drug): the same study cited under mitochondrial burden above, which in that Drosophila model found vitamin K2 supplementation rescued ATP-linked respiratory deficits caused by pink1 mutation, establishing vitamin K2 as capable of supporting mitochondrial ATP output through a non-clotting pathway. No warfarin, or any VKORC1 inhibitor, was administered in this study; it establishes biological plausibility for the pathway warfarin could interrupt, not a warfarin effect. Lavrenova et al. 2026, PMID 42450292 (warfarin blocking menadione-to-menaquinone conversion in flies, described under mitochondrial burden) is the closest warfarin-in-cohort data point to this pathway, but it does not report a direct ATP or high-energy-phosphate measurement as far as this review could confirm from the available title and subject matter.

Mechanism

The same VKORC1-to-menaquinone-4-to-electron-transport chain described under mitochondrial burden above, one step further downstream at the specific point of ATP synthesis rather than general respiratory capacity. No route by which warfarin would inhibit ATP synthesis independent of this vitamin K2 pathway has been described in any source located for this review.

Counter-evidence held on record

None confirmed either direction; the defining feature of this entry is the absence of any direct warfarin-plus-ATP measurement, not a contested result. Do not tell readers warfarin lowers their cellular energy production; it has never been measured in a person, or in any organism, taking the drug. No ribose, CoQ10, or other energy-metabolism supplement recommendation follows from this entry.