Oral vancomycin

Limits on what this shows

• Null findings and gaps: The only oral-route mitochondrial readout located is the macrophage membrane-potential result in [32]; no oral-route kidney, liver, gut or human mitochondrial study was located. The original searches for membrane potential, oxygen consumption and ATP returned the kidney toxicity studies above; [32] was found in a later repair pass.
• At-risk subgroups: Patients with significant systemic absorption (severe colitis, inflammatory bowel disease, renal impairment) [1][2] and those on other nephrotoxic drugs; this is a toxicity context, not a nutrient context.

Regulatory status

The oral labels discuss nephrotoxicity and ototoxicity as warnings and say nothing about mitochondria [1][2].

Population evidence

Sakamoto 2017 [16]: porcine proximal tubule cells (LLC-PK1) exposed to 1 to 5 mM vancomycin (about 1.5 to 7.5 mg/mL) showed increased mitochondrial superoxide and cellular ROS and time-dependent cardiolipin peroxidation leading to apoptosis; vitamin E and the mitochondria-targeted mitoTEMPO suppressed this (abstract-level read through a summarizing tool). Yu 2022 [18]: rats given vancomycin 400 mg/kg intraperitoneally for 7 days and human tubular (HK-2) cells showed apoptosis markers (higher Bax/Bcl-2, caspase-3 activation); membrane potential and ATP were not measured. No human mitochondrial measurement on oral vancomycin was found. Bojang 2026 [32]: C57BL/6 mice given vancomycin 0.5 mg/mL in drinking water for 4 weeks had a significantly lower mitochondrial membrane potential in large peritoneal macrophages (P < 0.01, n=5 mice pooled from 2 experiments); the same paper reports depolarization, fragmentation and higher mitochondrial ROS in cultured macrophages (details under ATP production). Serum or tissue vancomycin was not measured in the oral arm.

Mechanism

Named in the cell model: mitochondrial superoxide and cardiolipin peroxidation [16]. Exposure relevance: oral dosing leaves no detectable blood concentration and at most 0.76% urinary recovery in volunteers [1][2]; the cell concentrations (millimolar) and the rat dose (400 mg/kg intraperitoneal) are far from oral exposure.

Counter-evidence held on record

Counter-evidence on record Negligible oral absorption [1][2][13]; the models above do not reproduce oral exposure. Framing caution Not a statement that oral vancomycin damages mitochondria and not a fatigue or low-energy claim. It is a mechanistic flag from parenteral nephrotoxicity models. Antioxidant supplements are not recommended on this basis.

Limits on what this shows

• Null findings and gaps: No human blood or urine redox marker (MDA, antioxidant enzymes, total antioxidant capacity) on oral vancomycin was located.

Regulatory status

No oxidative stress statement in the oral labels [1][2].

Population evidence

Oktem 2005 [17]: male Wistar rats given intraperitoneal vancomycin had higher renal malondialdehyde and lower superoxide dismutase and catalase activity with tubular injury, reversed by erdosteine (abstract via OpenAlex). Yu 2022 [18]: rats at 400 mg/kg intraperitoneal for 7 days and HK-2 cells showed higher MDA and ROS and lower SOD and glutathione, restored by N-acetylcysteine (read through a summarizing tool). Sakamoto 2017 [16]: mitochondrial superoxide and ROS increased in tubular cells.

Mechanism

Named in models: reactive oxygen species generation with loss of antioxidant enzymes in tubular cells [16][17][18]. Not demonstrated for oral dosing, where blood concentrations are undetectable [1][2].

Counter-evidence held on record

Counter-evidence on record Negligible oral absorption [1][2][13]. The models use parenteral dosing that produces high renal exposure. Framing caution Not an antioxidant-depletion claim for oral use. It shares its source with the Mitochondrial burden entry and counts as one line of evidence (see Overlap map).

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source for copper with oral vancomycin was opened, and no circulating copper claim for vancomycin was found to rebut. This is a disclosed gap, not a finding of safety. Kept separate from Iron (Hard Rule 7). Re-check if the full text of [22] shows vancomycin was in the antibiotic regimen; the full text was still unreachable on 2026-10-04 after the extra attempts above, so this is tooling-limited for the one candidate source and a genuine gap for everything else opened.

What was searched

Web search "vancomycin copper(II) binding complex chelation serum copper OR ceruloplasmin vancomycin patients": results were general copper biology, with one French analytical-chemistry record on a copper-vancomycin chelate used for assaying the antibiotic (seen by title only, not opened, not evidence about copper status). Web search "gut microbiota depletion vancomycin iron homeostasis ... oral vancomycin serum iron" surfaced Miller 2019 [22], a mouse study of antibiotic-treated mice with lower colonic copper transporters CTR1 and ATP7A and a copper isotope shift; its OpenAlex abstract does not name vancomycin and the full text was blocked (403), so the cohort test cannot be passed and it is not used. The oral vancomycin labels mention no copper [1][2]. A consumer-page audit found no copper claim for vancomycin [25][26][27]. Repair pass: Miller 2019 [22] was re-attempted through the PubMed record (reCAPTCHA page), PMC HTML and PDF (reCAPTCHA page), three PNAS URLs (403), Europe PMC (refused by robots.txt or rate-limited) and OpenAlex (reconstructed abstract again names only "antibiotic treatment"); a Springer narrative review that cites it [34] also does not name the agents. Web search "oral vancomycin treated mice colonic copper OR iron OR ferroportin OR hepcidin OR ferritin vancomycin gavage microbiota" and the Miller title query returned no vancomycin copper measurement. A PubMed listing for vancomycin with copper or ceruloplasmin returned 429 and was not retried.

Channel outcomes

Web search worked. PubMed listing pages returned 429 after two queries; PubMed and PMC article pages returned bot-verification pages; PNAS returned 403; OpenAlex single-record lookups worked. Crossref once returned 429 and was not retried. Mixed result: a clean zero on the channels that worked for vancomycin-specific copper data, with the full-text check of [22] tooling-limited. Probably a genuine thin literature.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source for iron with oral vancomycin, and no circulating claim that names vancomycin. This is a disclosed gap, not reassurance. Kept separate from Copper (Hard Rule 7) and from the Oxygen delivery entry, which concerns immune hemolysis and not iron.

What was searched

Web search "vancomycin oral treatment mice iron absorption OR hepcidin OR ferritin OR anemia microbiota" returned: an iron-supplementation study in mice after neomycin plus metronidazole (vancomycin not used; no systemic iron measured; excluded) [30], Noordine 2024 on dietary iron in germ-free versus conventional mice (vancomycin not mentioned; excluded) [29], and Alfhili 2026 on anemia in patients on intravenous vancomycin (wrong route, no non-user comparator; excluded) [21]. Web search "gut microbiota depletion vancomycin iron homeostasis duodenum ferroportin DMT1 mice" surfaced only general microbiota and iron reviews. The oral vancomycin labels mention no iron [1][2]. The only claim found is a generic "antibiotics deplete iron" statement on a consumer checklist that does not name vancomycin [26]. Repair pass: web search "vancomycin alone mice gut microbiota iron metabolism serum iron liver iron hepcidin duodenal DMT1 antibiotic-induced dysbiosis" returned vancomycin mouse studies on GLP-1/GPR43 motility and Th1/Th2 immunity and a human iron-deficiency-anemia duodenal dysbiosis paper (titles only, not opened, none a vancomycin iron-status study); "oral vancomycin rats OR mice anemia hemoglobin iron deficiency microbiota" returned cognition, VRE, EAE and bioavailability studies (titles only). PubMed listings "oral vancomycin iron mice" (1 unrelated record) and "vancomycin AND (iron OR ferritin OR hepcidin) AND microbiota" (9 unrelated records: nanoparticle, wastewater, spore and poultry papers) contained no vancomycin iron-status study. A gut microbiota and iron review [35] was opened and cites no vancomycin-specific iron study (its antibiotic-iron sentence rests on references the tool could not verify, so it is not used). A Crossref relevance query returned unrelated records.

Channel outcomes

As for Copper. No vancomycin-specific iron measurement found on the channels that worked. Probably a genuine gap; the consumer-claim audit was complete only for the pages opened. The repair pass added three more query families and two PubMed listings without finding a candidate.

Limits on what this shows

• Null and contradicting findings: Fishbein 2021 [10]: double-blind placebo-controlled trial in 15 hospital patients colonized with C. difficile (8 vancomycin, 7 placebo), 125 mg four times daily for 10 days: beta-diversity differed significantly (P = 0.0059) and macrolide-lincosamide-streptogramin resistance genes rose 22% (P = 0.037), but alpha-diversity did not differ significantly between arms. This is a small colonized-patient cohort and is directionally compatible with a composition change. Mikkelsen 2015 [24] (12 lean men, vancomycin plus gentamicin plus meropenem for 4 days) shows the same direction but is excluded from the count because of the three-drug regimen. Reijnders 2016 [6]: despite large microbiome and bile acid changes, no effect on insulin sensitivity, energy expenditure, inflammation or permeability at 7 days or 8 weeks (secondary summaries). So the microbiome change did not translate into a measured nutrient or metabolic harm in that short trial.
• Confounding and independence: C. difficile infection itself disrupts the microbiome, so the CDI cohorts [7][11] alone could not attribute the change to the drug. The signal is attributed here because the same direction appears in cohorts without CDI: healthy volunteers [8][9], obese men [5][6] and rheumatoid arthritis patients [4]. Independence: [4], [7] and [8] are three separate populations and research groups. [8] and [9] come from the same research group and may share design features; participant overlap was not excluded, so they are treated as one line when counting. [5] and [6] are from the same Amsterdam research lineage but separate trials.
• At-risk subgroups: People who need repeated or prolonged courses (recurrent CDI), people with already disturbed flora, and anyone later exposed to resistant organisms; hospital patients on other antibiotics, in whom effects cannot be separated.

Regulatory status

The Firvanq label warns that use "may result in the overgrowth of non-susceptible bacteria" [2]. The capsule label text retrieved contains no normal-flora statement [1]. Neither label quantifies a flora effect. Both labels report recurrence of C. difficile infection of 23% and 18% within 4 weeks in the two trials [1].

Population evidence, human, composition

Isaac 2017 [4]: prospective, vancomycin 250 mg four times daily for 2 weeks in 9 patients with rheumatoid arthritis versus 12 methotrexate-only controls; 16S sequencing at baseline, end of treatment and 2, 6, 14 and 22 weeks after. Richness and Shannon diversity fell sharply, all baseline Bacteroidetes OTUs were lost, Proteobacteria (Klebsiella, Escherichia/Shigella) expanded, and only 39 +/- 21.9% of baseline-abundant OTUs were detected at 22 weeks, with wide variation between people. Louie 2012 [7]: randomized double-blind trial of vancomycin 125 mg four times daily versus fidaxomicin in C. difficile infection; in the vancomycin arm Bacteroides/Prevotella fell from about 7.6 to about 5.3 log10 CFU/g by day 10 and were still about 7.0 at day 28, Enterococcus was suppressed then rebounded, and C. coccoides and C. leptum groups were markedly suppressed (values as extracted by a summarizing tool from the text; patient counts of 44 to 45 per arm with a 20-patient microbiome subset were reported by the tool and not verified). Jo 2025 [8]: phase 1 randomized trial in healthy volunteers, 8 per arm, a 10-day oral vancomycin course, 16S plus stool metabolomics through day 30: fewer Bacteroidales, Coriobacteriales, Bifidobacteriales and Clostridiales, more Enterobacterales, Fusobacteriales and Lactobacillales. Ayele 2025 [9]: healthy volunteers aged 18 to 40, 10 days of oral vancomycin versus moxifloxacin, 16S through 12 follow-up visits: more Proteobacteria and decreased abundance with vancomycin (abstract only). Vrieze 2014 [5]: single-blind randomized trial in 20 obese men with metabolic syndrome (10 per arm), vancomycin 500 mg three times daily for 7 days: lower fecal diversity, fewer gram-positive bacteria (mainly Firmicutes) and a compensatory rise in gram-negative bacteria (mainly Proteobacteria) (abstract, truncated in the source). Reijnders 2016 [6]: randomized placebo-controlled trial in obese men, about 56 to 57 participants (the two secondary summaries differ), vancomycin 1,500 mg/day for 7 days: lower diversity with fewer Firmicutes, still altered 8 weeks later (secondary summaries only; primary text not opened). Tannock 2010 [12]: in C. difficile-infected patients vancomycin disrupted clostridial clusters XIVa and IV and Bifidobacterium more than fidaxomicin did (abstract only).

Population evidence, function (SCFA and bile acids)

Jo 2025 [8], vancomycin arm: acetate down 66% and propionate down 65% from baseline, butyrate depleted (percentage not given); primary bile acids rose (cholic acid +712.2 and chenodeoxycholic acid +142.3 ng/mg per day) and secondary bile acids fell (lithocholic acid -138.0, deoxycholic acid -253.8 ng/mg per day); recovery was not assessed because the study ended at day 30. Ayele 2025 [9]: secondary bile acids stayed lower for longer with vancomycin than with moxifloxacin. Vrieze 2014 [5]: secondary bile acids fell in feces and primary bile acids rose in plasma after meals. Reijnders 2016 [6]: lower fecal secondary bile acids, butyrate and acetate per two secondary summaries. Rode 2026 [11]: in recurrent C. difficile infection treated with vancomycin, secondary bile acid restoration was intermediate in speed between fecal microbiota transplantation (fastest) and a 12-strain mixture (slowest); about half of participants showed timely secondary bile acid elevation (abstract only).

Animal corroboration (Tier B)

Cirkin 2025 [23]: oral gavage of about 50 mg/kg every 72 hours for 10 weeks in C57BL/6J mice reduced Shannon and Chao1 indices by more than 80%, with near-loss of Lactobacillus and enrichment of Fusobacterium, Escherichia-Shigella and Desulfovibrio. Bile acids were not measured. Rodent data are supportive only.

Mechanism

Named and intended: inhibition of bacterial cell-wall synthesis, with action against gram-positive anaerobes, and negligible absorption so that the whole dose acts in the lumen [1][2]. The loss of bacteria that deconjugate and dehydroxylate bile acids explains the fall in secondary bile acids; the loss of fermenters explains lower short-chain fatty acids. These links are inferred from the measured co-movement in [8] and are not proven causal in the retrieved abstracts.

The drug's own risk (infectious, not nutrient)

Recurrence after treatment (label 23% and 18% at 4 weeks) [1]; in the Louie trial toxin re-expression was reported at 28% with vancomycin versus 14% with fidaxomicin and recurrence at 23% versus 11% (as reported by the tool) [7]; susceptibility to vancomycin-resistant Enterococcus colonization rose in the mouse arm of Isaac 2017 [4]. Prophylaxis studies of oral vancomycin (meta-analyses and a randomized trial seen as titles in search results) measure CDI incidence rather than nutrients and were not opened.

Direction and magnitude

Hard Rule 2: this is a compositional and functional change in the gut, expected from the drug's intended pharmacology, not a nutrient deficiency and not demonstrated harm. Direction: diversity down, Bacteroidetes/Bacteroidales/Clostridiales down, Proteobacteria up, secondary bile acids down, primary bile acids up, short-chain fatty acids down. Magnitude: large during treatment (loss of whole phyla in [4]; acetate and propionate down about two thirds in [8]); partial recovery over 22 weeks in the one long follow-up [4]. Boundary: the dosing period and the weeks to months after it.

Cohort test

Vancomycin given by mouth was the named drug in every cited human cohort [4][5][6][7][8][9][10][11][12] and in the mouse study [23]. Rifaximin cohorts (batch B011) did not include vancomycin and are not used. Tier A: human studies. Tier B: mouse study.

Why red

At least 2 independent Tier A sources with oral vancomycin in the cohort, the microbiome and its functional markers directly measured, a consistent direction, a large magnitude, a named mechanism, and only one small null on alpha-diversity [10] with no contradicting Tier A evidence of a reversed direction.

Counter-evidence held on record

Counter-evidence on record Fishbein 2021 [10] found no alpha-diversity difference in 15 colonized patients; Reijnders 2016 [6] found no metabolic consequence at 7 days or 8 weeks; Louie 2012 [7] and Tannock 2010 [12] are comparisons against fidaxomicin, an active comparator, so they show vancomycin disrupts more than fidaxomicin and do not test either against non-use. Framing caution Dysbiosis here is expected pharmacology, not a vitamin or mineral deficiency and not a sign the drug failed. Nothing in this entry supports taking probiotics, fiber or supplements; no supplement dose or product is recommended. Do not stop the antibiotic course because of this finding.

Limits on what this shows

• Null findings and gaps: No ATP content, respiration or oxygen consumption measurement was located in any oral-route arm, in any human, or in liver, muscle or gut tissue of vancomycin-treated animals. A narrative review [33] states that another in vitro study found vancomycin at 1, 2.5 and 5 mM increased oxygen consumption and ATP concentrations in proximal tubular cell lines, the opposite direction; that primary study was not identified or opened.
• At-risk subgroups: None established. The same rare-systemic-absorption caveat as the Mitochondrial burden entry applies [1][2].

Regulatory status

The oral labels contain no energy or ATP statement [1][2].

Population evidence

No human data. Sakamoto 2017 [16]: porcine proximal tubule cells (LLC-PK1) given 4 mM vancomycin showed "a significant decrease in intracellular ATP levels" at 12 hours, with mitochondrial membrane depolarization (JC-1) and cardiolipin peroxidation; the ATP sentence sits in the results section of the publisher page and was read through a summarizing fetch tool in this repair pass (the abstract itself gives no ATP figure; oxygen consumption was not measured). Bojang 2026 [32], mouse macrophages (C57BL/6JCrl, 8 to 12 weeks, both sexes): bone-marrow-derived macrophages differentiated in 20 mcg/mL vancomycin for 5 days had significantly lower ATP-linked respiration (P < 0.05, Seahorse oxygen consumption), lower spare respiratory capacity on stimulation with heat-killed Candida albicans (P < 0.05, n=4 independent experiments), depolarized membranes (JC-1, P < 0.0001), fragmented mitochondria and higher mitochondrial ROS; ATP content was not measured directly and glycolysis was unchanged. Oral arm of [32]: vancomycin 0.5 mg/mL in drinking water for 4 weeks lowered membrane potential in large peritoneal macrophages (P < 0.01, n=5 mice pooled from 2 experiments); Seahorse respiration, ATP content and mitochondrial ROS were not measured in vivo, and serum or tissue vancomycin was not measured. Intraperitoneal arm of [32]: three 200 mcg doses 48 hours apart also lowered membrane potential (P < 0.01).

Mechanism

Named in cell models: vancomycin localizing near the mitochondrial outer membrane (fluorescently labeled drug, possibly via phosphatidylglycerol as a cardiolipin precursor), depolarization, cardiolipin peroxidation and mitochondrial ROS [16][32]. Not established for oral dosing: the oral arm in [32] cannot separate a direct drug effect from an effect mediated by loss of gut bacteria, because no drug level and no microbiota-independent control were reported in the text read.

Cohort test

Vancomycin was the tested drug in [16] and [32] (Tier B, mechanistic in vitro and animal). No class extrapolation is used.

Counter-evidence held on record

Counter-evidence on record Negligible oral absorption: no blood concentrations detected in volunteers and at most 0.76% urinary recovery [1][2][13]. The cell concentrations (4 mM; 20 mcg/mL) are not oral exposures; the tool reported that the authors of [32] describe 20 mcg/mL as previously measured in vancomycin-treated humans, without a route stated. The opposite-direction in vitro report summarized in [33] is unverified. Framing caution Not an energy, fatigue or low-ATP claim for people taking oral vancomycin, and not a nutrient deficiency. It is a mechanistic flag from cultured kidney and immune cells. No energy or mitochondrial supplement is recommended on this basis. It shares its evidence with the Mitochondrial burden and Oxidative stress entries (see Overlap map).

Limits on what this shows

• Null findings and gaps: No oral vancomycin hemoglobin, hematocrit or oxygen delivery cohort located. Anemia in hospitalized patients is common and heavily confounded [21].

Regulatory status

The oral labels list reversible neutropenia and thrombocytopenia as postmarketing reactions; the text retrieved lists no anemia or hemolysis [1][2].

Population evidence

Xie 2025 [19]: case report, a 50-year-old man developed severe hemolysis on postoperative day 8 after knee replacement; vancomycin was given in bone cement and intravenously; high-titer IgG antibody to the blood group N antigen was found (abstract via OpenAlex). Li 2026 [20]: case report of a 76-year-old man with hemoglobin falling from 14.1 to 10.4 g/dL on first exposure and a drop of 8.2 g/dL on re-exposure, with a positive direct antiglobulin test; route not stated in the landing page text. Alfhili 2026 [21]: cross-sectional study of 299 patients receiving intravenous vancomycin, anemia in 90%; no non-user comparator and the wrong route, so it is excluded. A PubMed record titled as vancomycin oral solution-induced agranulocytosis was seen by title only and not opened.

Mechanism

Named: drug-dependent antibody immune hemolysis [19][20], which requires systemic exposure. With oral dosing, systemic exposure is limited to the rare absorption cases described in the labels [1][2].

Counter-evidence held on record

Counter-evidence on record No hemolysis report with oral-only dosing was opened. Hemolysis is an immune adverse reaction, not an oxygen-delivery deficiency state. Framing caution Not an iron, B12 or oxygen-carrying-capacity claim, and not a reason to take iron. A hemoglobin drop on any vancomycin course is a clinical question for the prescriber.