Indomethacin

Limits on what this shows

• Null and negative findings: The same seven-NSAID panel found naproxen had no effect on any mitochondrial parameter, so this is not a uniform class effect [16]. Colonic mitochondria showed improved oxygen-use efficiency rather than damage [20]. Human jejunal tissue showed an inhibitory effect that the authors called less conclusive [18].

Direction

Direct drug toxicity. No nutrient was measured and none is implied.

Regulatory status

The label text read does not mention mitochondria [1].

Population evidence

None in people on indomethacin. Tier B only: in a panel of seven NSAIDs tested in mitochondria, submitochondrial particles, cells and whole heart, indomethacin was among those that stimulated basal and uncoupled respiration, inhibited ATP synthesis and collapsed membrane potential [16]. In isolated liver mitochondria from rat, mouse and human, indomethacin uncoupled at low concentrations and inhibited respiration at higher ones [17]. In rat thymocytes it increased proton leak [19]. In rat tissue homogenates, hepatic mitochondria were uncoupled at 179 micromolar (respiratory control index down, ADP/O ratio sharply up) while colonic mitochondria showed unchanged control index [20]. After oral dosing in rats, enterocyte oxygen uptake and MTT reduction fell and lipid composition of villus mitochondria changed [21]; in rat jejunum ultrastructural mitochondrial change was seen in vitro [18].

Mechanism

Named: protonophore-like uncoupling and respiratory inhibition. In the rat, uncoupling plus COX inhibition together reproduced the injury pattern of indomethacin, while neither alone did [22].

Exposure relevance

The label gives peak plasma levels of about 1 and 2 mcg/mL after 25 and 50 mg and about 99% protein binding [1]. By the reviewer's arithmetic with a molecular weight of about 358 g/mol (not itself retrieved), that is roughly 3 to 6 micromolar total drug. Rat jejunum effects were shown at 0.5 and 2.5 millimolar [18] and hepatic uncoupling at 179 micromolar [20]; the concentrations used in the other studies were not retrieved. The label also notes appreciable enterohepatic circulation [1], but no luminal or mucosal concentration was found. Relevance to oral therapeutic exposure is therefore unproven.

Unread

A 1997 rat paper on mitochondrial damage as the topical phase of NSAID injury could not be opened [23].

Counter-evidence held on record

Framing caution for the consumer surface This is the drug's own toxicity in preclinical systems. It does not show that people taking indomethacin have impaired mitochondrial function or need an energy or mitochondrial supplement.

Limits on what this shows

• Null and negative findings: Repeated low-dose indomethacin (3 mg/kg) in rats gave adaptation with ulcer area 4.5 mm2 versus 21.3 mm2 after a single 25 mg/kg dose, along with lower MDA and MPO and higher glutathione and SOD, so direction depends on dose schedule [26]. Indomethacin raised liver cytosolic CuZnSOD activity in rats [29]. The one human erythrocyte study found in the sibling literature failed the cohort test for indomethacin (acemetacin, naproxen and tiaprofenic acid arms only) and is not used [27].

Direction

Drug toxicity in rodents, with a context-dependent reverse (adaptive) pattern. Not a nutrient deficiency.

Regulatory status

The label text read does not mention oxidative stress [1].

Population evidence

Rat, indomethacin 20 mg/kg intraperitoneally twice daily for 3 days: lipid peroxidation up 26 to 46%, superoxide dismutase down 19 to 59% and catalase down 21 to 48% across kidney, small intestine and liver [24]. Rat enterocytes after oral dosing: lipid peroxidation products up and alpha-tocopherol down in villus tip cells; arginine and zinc pre-treatment protected [21]. Rat gastric injury: lipid peroxidation up, glutathione and superoxide dismutase down, catalase up, reversed by ascorbic acid given as treatment [25].

Mechanism

Named: mitochondrial uncoupling and inflammation-driven oxidant production; see the overlap map for shared sources with the mitochondrial entry.

Counter-evidence held on record

Framing caution for the consumer surface Rodent tissue changes after high or short-course dosing do not show that people on indomethacin have low antioxidant status or need antioxidant supplements.

Limits on what this shows

• Null and negative findings: Fecal blood loss with 51Cr-labeled red cells in 30 healthy volunteers over 4 to 6 days showed no significant increase with indomethacin versus control, while aspirin did increase it [7]. A one-week volunteer study reported indomethacin blood loss of 1.20 mL against 0.13 mL for etodolac, 1.14 mL for ibuprofen and 0.87 mL for naproxen, but the retrieved abstract gave no placebo or baseline figure and no units per day, and it is an active-comparator design [8]. Short-term microbleeding data in healthy men are therefore mixed. A sustained-release indomethacin volunteer blood-loss study was located but not read [10]. One rat study of indomethacin and iron transport across isolated intestine exists, but its abstract was not available and its direction is unknown [11].
• Confounding: Presumed (Hard Rule 6): the indications (rheumatoid arthritis, ankylosing spondylitis, osteoarthritis, gout) are inflammatory conditions [1], and no source retrieved compares hemoglobin or iron indices before and after indomethacin is started.

Direction

Blood loss (drug toxicity). Per Hard Rule 2, anemia and GI bleeding are outcomes, not iron deficiency; no ferritin, transferrin saturation or serum iron was found for indomethacin users.

Regulatory status

Boxed warning for serious cardiovascular and gastrointestinal events including bleeding, ulceration and perforation. Label: upper GI ulcers, gross bleeding or perforation occur in approximately 1% of patients treated for 3 to 6 months; adverse reactions below 1% include "anemia secondary to obvious or occult gastrointestinal bleeding"; patients on long-term treatment should have hemoglobin or hematocrit checked if they show signs or symptoms of anemia. The label text read does not mention iron [1].

Population evidence

SOS meta-analysis of 28 observational studies: upper GI complications with indomethacin RR 4.14 (95% CI 2.91 to 5.90) versus non-use, against diclofenac 3.34, ibuprofen 1.84 and celecoxib 1.45 [6]. This measures complications, not iron stores.

Mechanism

Named: prostaglandin synthesis inhibition plus topical mucosal injury, which in rat work involves uncoupling of enterocyte mitochondria together with COX inhibition [22]. No absorption-interference mechanism for iron was found in the label or retrieved studies.

Class context

CLASS EXTRAPOLATION, not used for the grade: a systematic review of 47 chromium-labeled red-cell trials (1,162 participants) found full-dose NSAIDs at about two to four times a baseline fecal blood loss near 0.5 mL per day. Whether indomethacin was in the pooled set could not be confirmed from the abstract and the full text was not opened [9].

Circulating claims

Tier D pages say NSAIDs cause "iron loss" or "iron deficiency through mucosal damage and gastrointestinal bleeding"; neither gives a primary source and the second names no individual drug [42][43]. They describe blood loss, which is consistent with this entry and says nothing about absorption.

Counter-evidence held on record

Framing caution for the consumer surface Do not read a low hemoglobin or ferritin in an indomethacin user as a dietary iron shortfall. New anemia calls for evaluation of GI blood loss; the label directs a hemoglobin or hematocrit check if anemia symptoms appear.

Limits on what this shows

• Null and negative findings: The direction of every retrieved indomethacin result is up or normalizing (liver CuZnSOD activity raised [29]; serum copper moving toward control values [28]), not down. Group sizes and indomethacin doses were not retrieved for the copper-deficiency study.

Direction

Neutral to reverse. Changes are explained by the inflammatory state (acute-phase proteins), not by drug-driven loss of copper (Hard Rules 2 and 6).

Regulatory status

The label text read does not mention copper [1].

Population evidence

No human study. Rat adjuvant arthritis, indomethacin 2 mg/kg/day for up to 5 weeks: serum copper decreased after 1 to 2 weeks and was somewhat higher at 4 or more weeks in rats treated from 2 weeks after induction; serum copper correlated with ceruloplasmin (r=0.92), and the authors reported that indomethacin normalized perturbed zinc and copper status [28]. In rats with and without dietary copper deficiency, indomethacin raised liver cytosolic CuZnSOD activity in both groups [29].

Mechanism

Absent as a depletion mechanism. Coordination chemistry exists: copper and zinc complexes of indomethacin have been synthesized and were less damaging to rat stomach and intestine than the plain drug [30]. That shows the molecule can bind copper in the test tube, not that it removes copper from a person.

Circulating claims

None found that names copper for indomethacin.

Counter-evidence held on record

Framing caution for the consumer surface Do not infer copper depletion from these rat results. A serum copper change in an inflamed patient on indomethacin reflects inflammation and ceruloplasmin first.

Limits on what this shows

• Null and negative findings: In mice the shifted microbiota was protective: antibiotic-depleted animals had higher mortality (P=0.0021) and the adapted microbiota reduced injury [32], so a composition change is not by itself harm. No effect in men, none in duodenal samples, and full recovery after stopping in the human study [36].

Direction

Drug-induced composition change (dysbiosis in rodent models). Not a nutrient deficiency; any link to endogenous vitamin synthesis (mechanism class G) was not tested.

Regulatory status

The label text read does not mention the gut microbiome; it does record appreciable enterohepatic circulation of the drug [1].

Population evidence

Human: Edogawa 2018, 23 healthy volunteers given indomethacin, with duodenal and fecal 16S sequencing before, after and 4 to 6 weeks after stopping. Permeability rose in both sexes; only women showed lower fecal diversity and higher Prevotella; duodenal microbiota showed no sex-specific change; changes normalized after stopping [36]. Mouse: Firmicutes rose and Bacteroidetes fell after indomethacin [32]; brief exposure worsened C. difficile colitis with dysbiosis and barrier impairment [34]. Rat: more Bacteroides, Enterobacteriaceae and Clostridium with up-regulated TLR-2 and TLR-4 [33].

Mechanism

Named in mice: bacterial beta-glucuronidase cleaves indomethacin conjugates in the lumen and drives enteropathy; inhibiting it protected mice [35].

Why yellow, not orange

The human study is a primary interventional study with indomethacin in the cohort, but it was read at abstract level only, without dose, duration, sex split or effect sizes; the effect is confined to women and to fecal samples and is transient; and no clinical consequence was measured. Orange should be reconsidered if the full text confirms an effect size and a replication appears.

Counter-evidence held on record

Framing caution for the consumer surface Microbiome shifts here are research observations. They do not indicate a need for probiotics or any supplement.

Limits on what this shows

• Null and negative findings: No study retrieved measured hemoglobin, hematocrit, red-cell mass or arterial oxygenation over time in oral indomethacin users. A PubMed phrasing for hemolytic anemia returned no indomethacin hemolysis report (the hits concerned sickle cell disease, ketorolac and unrelated drugs); the hemolytic anemia item therefore rests on the label alone.
• Counter-evidence on record: The neonatal study shows that the oxygenation deficit in those infants belongs to the ductus being treated, not to the drug [15]. The adult imaging result is an acute intravenous dose and is not evidence about oral therapy.

Direction

Drug toxicity and intended pharmacologic effect (Hard Rule 2). Not depletion. Written as a mechanism clarification plus rare adverse events.

Regulatory status

Adverse reactions below 1%: leukopenia, bone marrow depression, anemia secondary to obvious or occult GI bleeding, aplastic anemia, hemolytic anemia, agranulocytosis, thrombocytopenic purpura and disseminated intravascular coagulation, with no incidence figure for each item [1].

Population evidence

Bruhn 2001, 10 healthy adults, indomethacin 0.2 mg/kg intravenously versus saline and aspirin: resting gray-matter MRI signal fell 2.1 to 2.6%, corresponding to roughly 12 mL per 100 g per minute lower cerebral blood flow and a 15% higher oxygen extraction fraction; aspirin did not do this [14]. In 20 preterm infants, cerebral oxygen saturation was lower with a patent ductus (62 versus 72%) and improved within 24 hours of ductal closure with indomethacin [15].

Case reports

Aplastic anemia after prolonged ingestion in two patients, one recovering after withdrawal and one dying [12]; one fatal case in an elderly man with allopurinol co-exposure and a positive lymphocyte transformation test [13]. Several other aplastic anemia reports exist as titles only [41].

Mechanism

Named for three separate routes: occult GI blood loss (see the iron entry), idiosyncratic marrow suppression proposed by the case authors to involve prostaglandin and cyclic nucleotide effects on blood cell production [12], and prostaglandin-dependent cerebral vasoregulation [14].

Counter-evidence held on record

Framing caution for the consumer surface Do not describe indomethacin as lowering oxygen-carrying capacity through a nutrient. The hemodynamic effect is the drug doing what it is meant to do, and marrow or hemolytic events are rare idiosyncratic reactions.

Limits on what this shows

• Null and negative findings: Total oxygen consumption was unchanged in intact thymocytes despite the proton leak [19], and the human tissue result was inconclusive [18]. Naproxen did not affect any mitochondrial parameter in the panel that included indomethacin [16]. No in vivo ATP measure for any indomethacin-exposed tissue was retrieved.

Direction

Direct drug toxicity. Not a nutrient deficiency.

Regulatory status

The label text read does not mention ATP or energy metabolism [1].

Population evidence

None in vivo in people. Tier B: ATP synthesis inhibited in isolated mitochondria [16]; ATP synthesis more sensitive to indomethacin than to the reference uncoupler dinitrophenol at lower concentrations, with comparable responses in rat, mouse and human liver mitochondria [17]; in rat thymocytes, ATP turnover was inhibited while total oxygen consumption stayed stable, so the cells compensated [19]; in rat jejunum, 0.5 and 2.5 mM indomethacin lowered oxygen uptake (P<0.01) and energy charge (P<0.05) and raised lactate, and in human jejunal tissue an inhibitory effect appeared but was less conclusive [18].

Mechanism

Named: uncoupling and respiratory inhibition lower the efficiency of ATP generation; lactate production rose alongside the fall in energy charge in rat jejunum [18].

Counter-evidence held on record

Framing caution for the consumer surface ATP effects here are laboratory findings at concentrations that may exceed ordinary plasma levels. They do not indicate low energy in patients, and "low energy" claims for indomethacin have no support in the sources retrieved.