Tolmetin

Limits on what this shows

• Null findings: Non-acidic compounds did not uncouple [20]; tolmetin in rats did not change liver weight or lipids [24]; no human mitochondrial endpoint with tolmetin exists in the retrieved record.
• At-risk subgroups: Not defined for tolmetin.

Regulatory status

No mitochondrial statement in the labels read [1][2].

Population evidence

Tier B, tolmetin in the study: Watanabe and Suga 1985 [24]: normolipemic rats given eight anti-inflammatory drugs including tolmetin 100 mg/kg (aspirin as positive control); tolmetin increased hepatic carnitine acetyltransferase and fatty acyl-CoA oxidizing system activities; no drug changed liver weight or serum or liver lipid levels, and the authors concluded the drugs affect peroxisomes without inducing proliferation (abstract via OpenAlex, reconstructed). These are peroxisomal enzymes and the finding is not a mitochondrial toxicity result. CLASS EXTRAPOLATION (Tier C, tolmetin not in cohort): Mahmud 1996 [20]: conventional NSAIDs and acidic prodrugs stimulated respiration (uncoupling) in isolated rat liver mitochondria at micromolar concentrations, potency inversely correlated with pKa, non-ionizable compounds did not uncouple; the abstract names flurbiprofen derivatives and nabumetone only. Somasundaram 1997 [21]: indomethacin produced mitochondrial ultrastructural change in rat intestine and several NSAIDs disrupted oxidative phosphorylation in rat liver mitochondria; named drugs indomethacin, aspirin, naproxen, piroxicam, nabumetone, flurbiprofen; tolmetin not named. Kasuya 2012 [22]: 11 NSAIDs tested on mouse mitochondrial acyl-CoA synthetase; tolmetin was not among them. Browne 1999 [23]: title names ibuprofen, flurbiprofen and ketorolac inhibiting hepatic mitochondrial beta-oxidation and oxidative phosphorylation; ketorolac is the nearest pyrrole acetic acid relative, but the abstract was not retrievable and tolmetin is not in the title. Records van Leeuwen 2012 [25] and Lassila 2015 [26] carry a MeSH "Tolmetin" heading, which appears to reflect analogs (zomepirac, ketorolac); their abstracts were empty in the records opened and nothing is claimed for tolmetin. PubMed listing "tolmetin AND mitochondrial" returned 8 records; only Watanabe 1985 and a 2026 tolmetin hydrazone cancer paper involved tolmetin itself [37].

Mechanism

Named at class level: protonophoric uncoupling by lipophilic weak acids [20][21]. Not shown for tolmetin.

Counter-evidence held on record

Counter-evidence The in vitro uncoupling studies used isolated mitochondria at concentrations that may not apply to a drug that is cleared within hours and not retained; none established a clinical mitochondrial deficit for any NSAID. Framing caution A class-level in vitro signal, not a finding about tolmetin. Must not be shown as a mitochondrial deficiency or as a reason to take mitochondrial supplements.

Limits on what this shows

• Null findings: Amtolmetin guacyl did not raise lipid peroxidation in rats [27]; this is a prodrug result and is not carried onto tolmetin. Several class studies show NSAIDs reducing neutrophil ROS in vitro [29][30][31]. Direction for tolmetin is unknown.
• At-risk subgroups: Sun-exposed skin or phototoxicity is a possible context for the photosensitization finding; the label retrieved lists serious skin reactions under 1% and skin irritation under 3% but no photosensitivity statement [1][2].

Regulatory status

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

Population evidence

Tier B, tolmetin in the study: Giuffrida 1995 [14]: "Photohaemolysis sensitized by tolmetin"; the record carries keywords singlet oxygen, UVA irradiation, superoxide dismutase, sodium azide and mannitol, but the abstract was not retrievable, so no findings are stated beyond the title. Related compound (tolmetin not tested): Kirkova 2007 [27] compared amtolmetin guacyl (described as a pro-drug of tolmetin) with indomethacin, diclofenac and celecoxib in rats; indomethacin raised gastric mucosal lipid peroxidation and altered antioxidant enzymes while amtolmetin guacyl did not (abstract). Pisano 1999 [7] found tolmetin itself damaged the rat stomach whereas the prodrug did not, so the protective finding belongs to the prodrug and not to tolmetin. Zomepirac, a tolmetin analog, is among compounds in a hepatocyte GSSG/GSH study (Sentellas 2014 [28]); only the MeSH record was read and no result is claimed. CLASS EXTRAPOLATION (tolmetin not in cohort): in vitro NSAID antioxidant or neutrophil ROS studies, Mouithys-Mickalad 2000 [29] (nimesulide, aceclofenac, diclofenac, indomethacin and metabolites), Paino 2005 [30] (diclofenac, indomethacin, naproxen, piroxicam, tenoxicam) and Costa 2006 [31] (seven arylpropionic acids), all show drug-dependent mixed effects; tolmetin is not named in any of them. The PubMed listing designed for "tolmetin AND oxidative stress" did not run (the query was dropped to the PubMed home page); WebSearch queries for tolmetin with lipid peroxidation, MDA, GSH or SOD returned no tolmetin redox study.

Mechanism

Named for the photosensitization finding only: light-driven reactive oxygen generation in an in vitro red-cell system [14]. No mechanism was established for oral dosing in tissue.

Counter-evidence held on record

Counter-evidence Inflammatory arthritis can itself alter redox markers (general background, not cited to a source here), so any redox change in treated patients would be confounded by indication (Hard Rule 6). The light-dependent in vitro finding has no demonstrated oral-dose counterpart. Framing caution Mixed and unquantified. Not a claim that tolmetin depletes antioxidants, and not a reason to take antioxidant supplements.

Limits on what this shows

• Null findings: Fecal blood loss with tolmetin was not significantly different from control [5]. The 2008 systematic review of 47 chromium-label trials (1,162 individuals) did not name tolmetin in the text read, and its drug table was not opened, so it is not relied on [38]. The comparison is short term, in healthy young volunteers, not in older arthritis patients on months or years of therapy.
• At-risk subgroups: Prior ulcer or GI bleeding, aspirin co-users (label: not generally recommended), warfarin users (label: increased prothrombin time and bleeding), older adults [1][2].

Regulatory status

Label (Tier A, via summarizing fetch) [1][2]: serious GI adverse events including inflammation, bleeding, ulceration and perforation; patients with prior ulcer disease or GI bleeding have a more than 10-fold higher risk of GI bleeding; anemia is sometimes seen; platelet aggregation inhibition and prolonged bleeding time; gastritis and peptic ulcer each under 3%, GI bleeding and perforation under 1% with probable causal relationship. Iron is not mentioned; the only iron item in one label is iron oxide as an ink ingredient.

Population evidence

Beirne 1974 [5] (Tier A for fecal blood loss, tolmetin in the cohort): two short studies in 30 healthy volunteers with no GI history, one drug each over 4 or 6 days, 51Cr-labeled red cells recovered in daily stool; tolmetin and indomethacin produced no significant increase in fecal blood loss over control, aspirin did, and no tolmetin recipient reported side effects (abstract reconstructed via OpenAlex through a summarizing fetch; doses and mL/day values not in the text read; full text not opened). A 1978 review summary [6] states that tolmetin 1200 mg daily causes less GI bleeding than equivalent aspirin in healthy volunteers and cites a second 51Cr report (Johnson 1974, title only) whose results were not shown. Tier B: Pisano 1999 [7] gave tolmetin intragastrically to rats at 15 to 60 mg/kg and found dose-dependent gastric damage after acute and repeated dosing (abstract only). No study measuring ferritin, transferrin saturation, hemoglobin response or iron absorption in tolmetin users was retrieved: PubMed listing "tolmetin AND (iron OR ferritin OR blood loss)" returned 21 records (10 shown), none of which was a tolmetin iron-status study [37].

Mechanism

Present at class and label level: prostaglandin synthesis inhibition with mucosal injury and platelet inhibition. Tolmetin forms copper(II) complexes in the laboratory [34], with no demonstrated physiological meaning for iron.

Hard Rule 2 and 6 checks

GI bleeding and anemia are clinical outcomes, not iron deficiency. Iron deficiency needs measured iron indices. Confounding by indication is presumed: rheumatic disease, chronic inflammation and aspirin co-use all affect bleeding and hemoglobin independent of the drug; in active juvenile arthritis, iron deficiency was the likely cause of anemia in one small cohort that did not involve tolmetin [39].

Circulating claim

Class consumer pages state "NSAIDs cause gastrointestinal irritation, bleeding, and iron loss" [40] (Tier D; tolmetin not named). The bleeding part is supported by the label; the "iron loss" part is untested for tolmetin and the one controlled blood-loss study is null.

Counter-evidence held on record

Counter-evidence The one tolmetin-specific controlled blood-loss study is null against control and the review summary reports less bleeding than aspirin [5][6]; this does not test long-term use or non-use in patients. Framing caution GI bleeding is an outcome and does not equal iron deficiency. Anemia on tolmetin needs ferritin and transferrin saturation before any attribution; no iron dose is recommended here, and new iron deficiency in an adult warrants evaluation for a bleeding source.

Limits on what this shows

• Null findings: No human copper, ceruloplasmin or urinary copper measurement with tolmetin was found. The guinea pig result is unknown.
• At-risk subgroups: Not defined.

Regulatory status

No copper statement in the labels read [1][2].

Population evidence

Tier B, in vitro chemistry: Dendrinou-Samara 1990 [34]: copper(II) complexes of tolmetin, ibuprofen and naproxen prepared and characterized; dimeric DMSO adducts with copper atoms about 2.6 angstroms apart; two crystal structures solved, one with the tolmetin-derived ligand. A 1992 Dalton Transactions paper on copper(II) complexes of tolmetin, ibuprofen and naproxen (solution behavior, electrochemistry) was seen at a landing page [35]. A PubMed listing "tolmetin AND (copper OR zinc OR ceruloplasmin)" returned 10 records, including a tolmetin copper complex wound healing study (Somayaji 1995), an assay paper for a tolmetin-copper(II) complex, and spectroscopic copper complex work (listing text only) [37]. Tier B, tolmetin in the animal study: Kotsaki-Kovatsi 1994 [36] measured zinc, copper and magnesium in guinea pig liver, kidney, heart, spleen, lung, femur and adrenal tissue after tolmetin by atomic absorption (indexing terms); no abstract exists in the PubMed, OpenAlex or Crossref records reached, and the Wiley page returned HTTP 403, so the direction and size of any change are unknown.

Mechanism

Named, in vitro only: the carboxylate binds Cu(II) [34]. This does not show copper loss from the body; many carboxylic acids bind metals in a test tube.

Hard Rule 2 and 6 checks

General background (not cited to a source here): serum copper and ceruloplasmin can rise with systemic inflammation, so any measured change in treated arthritis would be confounded by indication; no such tolmetin data exist. Copper is graded separately from Iron per Hard Rule 7.

Counter-evidence held on record

Counter-evidence Complex formation in solution is a chemistry property and not evidence of altered copper status. The one status-type study (guinea pig tissue) is unread. Framing caution No copper deficiency or excess claim for tolmetin. Must not prompt copper supplements; no copper dose is recommended here.

Limits on what this shows

• Null findings: The review reports no significant change with celecoxib in obese postmenopausal women [32]. No tolmetin data exist in the record read.
• At-risk subgroups: Not defined for tolmetin; the class literature names older adults and people on acid suppressants.

Regulatory status

No microbiome statement in the labels read. Diarrhea, flatulence, abdominal pain and nausea sit in the common reaction band as summarized; these are GI symptoms, not microbiome measurements [1][2].

Population evidence (CLASS EXTRAPOLATION)

Cohort test: tolmetin NOT named. The 2021 review of gut microbiota in NSAID enteropathy [32] (narrative review, opened through a summarizing fetch) names indomethacin, celecoxib, ibuprofen, aspirin, naproxen, diclofenac and a multi-NSAID elderly cohort, with taxa shifts such as higher Enterobacteriaceae and Bacteroides in animals and sex-dependent changes after indomethacin in humans; tolmetin is not mentioned. A screen of 1,197 compounds against 40 gut bacterial strains [33] (copy hosted at gwern.net; text opened) does not mention tolmetin in the text read, and its supplementary compound table was not opened. PubMed listing "tolmetin microbiome" returned an unfiltered 254,439-record list with no tolmetin record on page one, and a boolean variant returned HTTP 404 [37]; no tolmetin microbiome record was found.

Mechanism

Named only in general terms at class level: mucosal injury and direct drug effects on bacterial growth in screens [32][33]. Not shown for tolmetin.

Counter-evidence held on record

Counter-evidence The class microbiome data are heterogeneous across NSAIDs (increase in some taxa in some studies, none in others), and the controlled tolmetin blood-loss study suggests low mucosal bleeding in the short term [5]. Framing caution A class-level context flag, not a finding about tolmetin. Must not be shown as dysbiosis caused by tolmetin or as a reason to take probiotics.

Limits on what this shows

• Null findings: No methemoglobin or G6PD-specific tolmetin data were located. The label gives no incidence for the hemoglobin decreases, so frequency and size are unquantified. Fecal blood loss with tolmetin did not significantly exceed control in healthy volunteers [5], consistent with the label's statement that the hemoglobin decreases occurred without GI bleeding. The reviews [13] describe NSAID immune hemolysis as infrequent.
• At-risk subgroups: People with prior exposure to tolmetin (label anaphylactoid wording; the hemolysis reports followed intermittent or repeat dosing [11]); the aspirin triad (asthma, rhinitis or nasal polyps with NSAID sensitivity); older patients with a gastric malignancy (one fatal report [12]); G6PD deficiency is a hedged predisposition in the review [13]. Antibodies may cross-react with suprofen [10].

Regulatory status

Label (Tier A), two labels read via summarizing fetch [1][2]. Adverse reactions, incidence under 1% with probable causal relationship: GI bleeding, perforation, hepatitis, anaphylactoid reactions, hemolytic anemia, thrombocytopenia, granulocytopenia, agranulocytosis, renal failure and serious skin reactions. Hematologic wording: "Small and transient decreases in hemoglobin and hematocrit not associated with gastrointestinal bleeding have occurred"; no incidence figure and no Coombs or autoimmune wording appear. Warnings: "Anemia is sometimes seen in patients receiving NSAIDs including tolmetin"; check hemoglobin or hematocrit if signs of anemia appear; NSAIDs inhibit platelet aggregation and can prolong bleeding time. Anaphylactoid reactions "may occur in patients with known prior exposure"; the drug should not be given to patients with the aspirin triad, and the label advises seeking emergency help for an anaphylactoid reaction.

Population evidence

No controlled cohort or trial reporting hemoglobin change in tolmetin users was retrieved. Case-level evidence (Tier B): McCall and Owens 1992 [11]: 35-year-old woman, acute hemolysis after two doses, hemoglobin 7.0 g/dL two days later with raised LDH, reticulocytosis and indirect hyperbilirubinemia; direct antiglobulin test weakly positive; her serum agglutinated all reagent red cells only when tolmetin was added; immune-complex mechanism; resolved without treatment (abstract via OpenAlex, reconstructed). Squires 1985 [9]: patient with strongly positive direct antiglobulin test on tolmetin, a high-titer drug-dependent antibody found only with tolmetin added, hematocrit 45% three months after stopping (PubMed record, abstract paraphrased by the fetch tool). van Dijk 1989 [10]: patient with massive hemoglobinuria and acute renal failure; tolmetin-dependent antibodies at a higher titer than the suprofen-dependent antibodies, later autoantibodies; unclear whether suprofen, tolmetin or cross-reacting antibody caused the hemolysis. Larsen and Becher 1993 [12]: 80-year-old man on tolmetin for osteoarthritis who died of autoimmune hemolytic anemia within 12 hours of reaching the emergency department, with superficial gastric adenocarcinoma at autopsy; of 18 NSAID immune hemolysis cases reviewed, 3 involved tolmetin (one probably and one possibly autoimmune, one drug adsorption); the authors say causality cannot be proven (Springer summary only). Sanford-Driscoll and Knodel 1986 [13]: narrative review naming tolmetin among NSAIDs reported with immune hemolytic anemia; infrequent, good prognosis, routine serologic screening not justified. Anaphylaxis: Moore and Goldsmith 1980 [8] report anaphylaxis to tolmetin confirmed by rechallenge (abstract).

Mechanism

Named: drug-dependent antibody (immune-complex) red-cell destruction, with an additional autoimmune component in one case [10][11]. Light-dependent photohemolysis sensitized by tolmetin has been shown in vitro (singlet oxygen, UVA; title and keywords only, abstract not retrievable) [14]; there is no evidence that this occurs after oral dosing.

Hard Rule 2 and 6 checks

Hemolytic anemia and anaphylactoid reactions are clinical outcomes of a drug reaction, not nutrient deficiencies; iron would not treat immune hemolysis. The entry is written as an outcome entry for the oxygen-carrying category, not as a nutrient finding. Confounding by indication: rheumatic disease and chronic inflammation also lower hemoglobin independent of the drug, and iron deficiency anemia is common in active juvenile arthritis (Koerper 1977 [39], no tolmetin mentioned in the page read).

Orange basis and limit

Orange rests on one Tier A source (the label) naming the effect with probable causality, corroborated by Tier B case reports with serologic proof. It is bounded by low frequency (under 1%), idiosyncratic mechanism and the absence of any quantified hemoglobin change. A stricter reviewer could place it at yellow because no hemoglobin value from a cohort was retrieved; it is not red because only one Tier A source exists.

Counter-evidence held on record

Counter-evidence The label calls the hemoglobin decreases small and transient [1][2]; the reviews call NSAID immune hemolysis infrequent [13]; the fatal case had a concurrent gastric carcinoma and the authors could not prove causality [12]; in the van Dijk case the hemolysis could not be assigned to tolmetin versus suprofen [10]. Framing caution A rare drug reaction, not a nutrient deficiency, and no vitamin or mineral addresses it. Customer-visible text must carry the boundary: rare, idiosyncratic, labeled at under 1% for hemolysis. New jaundice, dark urine, pallor, shortness of breath or hives, throat swelling or wheeze after a dose belong with the prescriber or emergency care. Do not read anemia as iron deficiency without iron indices.

Limits on what this shows

• Null findings: No ATP content, no cellular energy charge and no clinical energy measure with tolmetin. Non-acidic compounds did not uncouple [20].
• At-risk subgroups: Not defined.

Regulatory status

No ATP or energy statement in the labels read [1][2].

Population evidence (CLASS EXTRAPOLATION)

The same class sources as Mitochondrial burden, read for energy output only: Mahmud 1996 [20] (respiration stimulation as the uncoupling marker, not ATP content) and Somasundaram 1997 [21] (NSAID disruption of oxidative phosphorylation in rat liver mitochondria); tolmetin not named in either abstract. Browne 1999 [23] (ibuprofen, flurbiprofen and ketorolac inhibiting hepatic mitochondrial oxidative phosphorylation per its title; abstract not retrievable). Tolmetin itself was never tested for ATP; PubMed listing "tolmetin AND mitochondrial" found no tolmetin ATP or respiration paper [37]. A direct "tolmetin ATP" listing was not run (rate limits).

Mechanism

Named at class level: proton leak across the inner membrane reduces coupling of respiration to ATP synthase [20][21]. Not shown for tolmetin.

Counter-evidence held on record

Counter-evidence Respiration stimulation in isolated mitochondria is not the same as lower ATP in cells or tissues; cells compensate, and the tolmetin label reports no energy-related effect. Framing caution Not a statement that tolmetin lowers energy or ATP. "Low energy" must not be attributed to tolmetin from this entry.