Diclofenac

Limits on what this shows

• Null findings: No human measurement found, so no null human result either. In the HepG2 work, control cells were less affected than peroxiredoxin III-depleted cells [8], and exercise training protected rat liver mitochondria [7], so susceptibility is modifiable and exposure alone does not fix the size of the effect.
• At-risk subgroups: Not defined from the sources. Impaired mitochondrial antioxidant defence (peroxiredoxin III depletion) increased injury in cells [8]. Hepatic impairment matters because hepatic metabolism accounts for almost 100% of elimination [1].

Regulatory status

The label portions retrieved do not mention mitochondria [1]. They do report meaningful aminotransferase elevations (AST above 3 times ULN) in about 2% of approximately 5,700 patients and severe hepatic reactions including liver failure [1]. Any link between those events and mitochondrial injury is a hypothesis in the sources below.

Population evidence

None in humans. Preclinical, diclofenac named: the 2002 Hepatology rat study [5], isolated rat liver mitochondria with Ca2+ and succinate showed swelling, Ca2+ leakage, membrane depolarization and oxidation of NADPH and protein thiol, all suppressed by cyclosporin A, and the authors tie the reactive-oxygen and voltage-dependent mechanisms to those proposed for uncouplers of oxidative phosphorylation. Rat liver mitochondria in vitro [7]: diclofenac lowered membrane potential, raised state 4 respiration and increased permeability-transition susceptibility, all prevented in exercise-trained animals. HepG2 cells and mouse hepatocytes at 500 micromolar for 12 to 48 hours [8]: loss of membrane potential, lower basal and maximal respiration and higher mitochondrial H2O2, worse when peroxiredoxin III was depleted. Cell-line study [9]: membrane depolarization and cytochrome c release (abstract partly garbled in retrieval). Mouse heart, 15 mg/kg/day for 4 weeks [10]: more depolarized membrane potential and higher cardiomyocyte ROS. A 1997 rat perfused-liver uncoupling paper and a 2016 rat liver ATP-synthesis paper were located by title only and not opened [17][18].

Mechanism

Named and demonstrated in vitro: permeability transition pore opening, uncoupler-like behaviour and mitochondrial H2O2 generation. Whether concentrations reaching human tissue are comparable to the exposures used (for example 500 micromolar) was not established, because plasma concentration figures were not in the label text retrieved.

Counter-evidence held on record

Counter-evidence and confounding Every source is Tier B (in vitro, isolated organelle, cell line or rodent). Doses and exposure times were chosen to produce injury and may exceed human tissue exposure. The same studies underlie the ATP and oxidative stress entries (see Overlap map). Framing caution This is a mechanism of drug toxicity, not a nutrient deficiency, and it does not imply that a mitochondrial supplement is needed. No supplement is recommended by this entry.

Limits on what this shows

• Null findings: A 2025 narrative review [14] describes diclofenac as able to cause oxidative stress and inhibit superoxide dismutase and catalase, but also reports that diclofenac sodium lowered oxidized LDL and lipid hydroperoxides in male Wistar rats and, in an anticonvulsant model, lowered MDA and raised SOD. Primary references were not retrievable, so these opposing findings are noted and not graded. Direction is therefore not uniform across models. The review is Tier C narrative.
• At-risk subgroups: Not defined from human data.

Regulatory status

Not mentioned in the label portions retrieved [1].

Population evidence

None in humans. Abdel-Daim 2018 [12]: 56 Swiss albino mice (8 per group), single oral diclofenac sodium 100 mg/kg versus saline control; diclofenac alone significantly raised (p below 0.05) tissue nitric oxide and malondialdehyde and lowered reduced glutathione, glutathione peroxidase, superoxide dismutase and catalase activity in tissue (the abstract states the findings for tulathromycin or diclofenac together; the study concerned cardiotoxicity). Thai 2021 [10]: more ROS in cardiomyocytes after 4 weeks at 15 mg/kg/day. Mouse small intestine, diclofenac 2.5 mg/kg daily for 3 days [11]: malondialdehyde and myeloperoxidase both rose versus control. Liver mitochondria in vitro [7] assessed oxidative stress markers (the abstract does not state the diclofenac-specific direction), and HepG2 cells showed higher mitochondrial H2O2 after diclofenac [8].

Mechanism

Present: mitochondrial ROS and H2O2 generation [8][10], with consumption of reduced glutathione [12]. Not a dietary antioxidant deficit.

Counter-evidence held on record

Counter-evidence and confounding The mouse dose (100 mg/kg single oral) is high, and the study was designed to show potentiated toxicity of another drug. The mixed direction in the review [14] and the absence of human biomarker data cap this at yellow. Framing caution This is an excess oxidant load from drug toxicity in animals. It does not show that patients are deficient in an antioxidant nutrient, and it does not support antioxidant supplementation.

Limits on what this shows

• Null findings: No ferritin, transferrin saturation, hemoglobin or iron-absorption measurement in diclofenac users was retrieved. Topical diclofenac is about 6% as available systemically as oral [2], but no topical GI blood loss data were retrieved. The CONDOR randomized trial (celecoxib versus diclofenac plus omeprazole) was found as a record only; its abstract could not be opened and it was not used [21].
• At-risk subgroups: Prior ulcer or GI bleeding, concurrent aspirin or anticoagulant use (label), higher doses (RR roughly 2 to 3 times at high versus low dose) [1][6].

Regulatory status

Boxed warning: NSAIDs cause an increased risk of serious GI events including bleeding, ulceration and perforation, which can be fatal. Upper GI ulcers, gross bleeding or perforation occurred in approximately 1% of patients treated for 3 to 6 months and about 2% to 4% treated for one year (class language). History of ulcer or GI bleeding raised risk more than 10-fold. Hematologic toxicity: "Anemia has occurred in NSAID-treated patients. This may be due to occult or gross blood loss, fluid retention, or an incompletely described effect on erythropoiesis." The label directs hemoglobin or hematocrit monitoring if signs or symptoms of anemia occur [1].

Population evidence

The 2012 meta-analysis [6]: systematic review and meta-analysis of observational studies, 59 studies abstracted and 28 meta-analyzed, adjusted RR versus non-use for upper GI complications. Diclofenac RR 3.34 (95% CI 2.79 to 3.99), against celecoxib 1.45 and ibuprofen 1.84 in the same analysis. RRs for high daily doses were 2 to 3 times those for low doses (all NSAIDs pooled in the dose statement). Diclofenac is named in the cohort. Healthy-volunteer study [13]: 30 people given diclofenac 50 mg three times daily for 14 days; 8 (27%) had fecal calprotectin above the upper limit at least once (median 76 microgram/g among those; 8.3% of all measurements), average calprotectin was higher on treatment (p = 0.003), and all had normal values 2 weeks after stopping. Calprotectin marks bowel inflammation, not blood loss or iron. A 1981 fecal blood loss study and a 2000 rat enteropathy study were found by title only [19][20].

Mechanism

Present: mucosal injury in the upper GI tract and small intestine, with blood loss. No acid-dependent or other iron-absorption mechanism was found for diclofenac.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication applies: people taking NSAIDs for arthritis often take other bleeding-risk drugs and have comorbidity. Anemia has other label-listed causes (fluid retention, unclear erythropoiesis effects). Framing caution Anemia is not iron deficiency, and GI blood loss is a drug toxicity pathway. Iron supplementation is not recommended by this entry; new iron deficiency or anemia on an NSAID warrants a check for GI bleeding, and excess iron has its own risks.

Limits on what this shows

• Null findings: In the human 1983 comparison [35] NSAID-treated patients did not have lower copper than steroid-treated patients. No source showed low copper, or any copper-lowering effect independent of inflammation, with diclofenac.
• At-risk subgroups: Not defined. Rheumatoid arthritis and other inflammatory conditions raise serum copper and ceruloplasmin and confound any reading on treatment [35][36].

Regulatory status

Copper is not mentioned in the label portions retrieved [1].

Population evidence

None for diclofenac. Class extrapolation, Tier C: Biological Trace Element Research 1995 [34], rats with adjuvant arthritis given oral indomethacin 2 mg/kg/day for 5 weeks; per the page opened, the baseline high serum copper decreased after 1 to 2 weeks, and at 3 to 4 weeks was somewhat higher than in controls in animals treated from 2 weeks after disease induction; serum copper correlated with ceruloplasmin (r = 0.92). Drug studied: indomethacin only. Human context [35]: a 1983 cross-sectional study found serum copper and ceruloplasmin raised in active rheumatoid arthritis compared with controls (P below 0.01), in patients on NSAIDs (60), corticosteroids (22) or no drug (6); NSAID-treated patients were slightly higher than steroid-treated, not significantly; no NSAID was named, so diclofenac is not identified in that cohort. A 1986 rat study [36] found carrageenan inflammation raised plasma copper over 6 to 72 hours, with no NSAID given.

Mechanism

No drug-specific copper mechanism (absorption, chelation, excretion) was found for diclofenac. The retrieved sources point to copper and ceruloplasmin behaving as acute-phase markers that rise with inflammation [35][36], so a change on an anti-inflammatory drug is expected to follow disease activity. Diclofenac-copper(II) complexes exist as laboratory-synthesized compounds [26][27][37]; these do not describe copper status in a body.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication is presumed (Hard Rule 6): the one drug-level finding tracks arthritis activity, in an animal model, for a different NSAID. The 1995 rat results were read from a summarizing fetch of the article page and were mixed over time, so the direction at later time points is not clean. Kept separate from Iron per Hard Rule 7. Framing caution This is not a finding that diclofenac depletes or raises copper. Serum copper in inflammatory disease should not be read as a deficiency or excess without the inflammatory context. No copper supplement is suggested by this entry.

Limits on what this shows

• Null findings: A 2018 Nature screen of more than 1,000 marketed drugs against 40 gut bacterial strains found that 24% of human-targeted drugs inhibited at least one strain [25]. The text retrieved did not report a diclofenac-specific result, so it is neither support nor null for diclofenac. No 16S or metagenomic study of diclofenac users was located.
• At-risk subgroups: Not defined.

Regulatory status

Not mentioned in the label portions retrieved [1].

Population evidence

Tier B, diclofenac named. The 2013 mouse study [23]: in mice, a small-molecule inhibitor of bacterial beta-glucuronidase gave complete protection against diclofenac-induced enteropathy (60 mg/kg intraperitoneal) when given before the drug and partial protection 3 hours after, and was also protective against indomethacin and ketoprofen. Zhong 2016 [24]: in rats given repeated oral diclofenac 10 mg/kg, ciprofloxacin dose-dependently inhibited beta-glucuronidase in distal small intestine contents and in E. coli, abolished enterohepatic circulation of diclofenac and alleviated enteropathy (sample size and first author details as given by a summarizing fetch). Human context [13]: oral diclofenac 50 mg three times daily for 14 days raised fecal calprotectin in healthy volunteers (inflammation marker, not a microbiome measure).

Mechanism

Present and demonstrated in rodents: enterohepatic recycling of NSAID glucuronides, bacterial deconjugation (beta-glucuronidase), local high aglycone concentration in the lumen.

Counter-evidence held on record

Counter-evidence and confounding Rodent doses were high and parenteral in the mouse study. The findings concern injury mechanism, and no benefit of altering the microbiome in patients is shown. The protective agents in these studies were an enzyme inhibitor and an antibiotic, not probiotics or diet. Framing caution Do not read this as evidence that diclofenac depletes beneficial bacteria or that probiotics or antibiotics should be used with it. No intervention is suggested.

Limits on what this shows

• Null findings: No measurement of ATP, phosphocreatine or whole-body energy output in diclofenac users was found. Fatigue or low energy in patients is not tested by any retrieved source.
• At-risk subgroups: Not defined.

Regulatory status

Not mentioned in the label portions retrieved [1].

Population evidence

None in humans. The 2002 Hepatology rat study [5]: rat hepatocytes cultured 24 hours with diclofenac had decreased cellular ATP with lactate dehydrogenase leakage and depolarization; cyclosporin A and verapamil attenuated the toxicity. Thai 2021 [10]: cardiomyocytes from mice given diclofenac 15 mg/kg/day for 4 weeks had significantly reduced ATP after Complex I and Complex II mediated respiration. A mouse small-intestine study measured ATPase and succinate dehydrogenase activity, but its abstract reports these only as raised by the protective drug rebamipide, not as diclofenac-versus-control values [11]. A 2016 rat liver study of ATP synthesis inhibition by diclofenac and its metabolites was found by title only [18].

Mechanism

Named: loss of membrane potential and permeability transition lower the ATP yield of oxidative phosphorylation [5][10]. This mechanism is shared with the Mitochondrial burden entry.

Counter-evidence held on record

Counter-evidence and confounding Tier B only, and the two ATP findings come from the same research tradition of dose-to-injury models. Kept as its own entry per Hard Rule 7 while the shared origin is disclosed in the Overlap map. Framing caution Not a finding that diclofenac lowers energy in patients. No coenzyme Q10, carnitine or other supplement is suggested by this entry.

Limits on what this shows

• Null findings: No hemoglobin, hematocrit or oxygen saturation data from a diclofenac cohort were retrieved. No methemoglobin finding was located.
• At-risk subgroups: Not definable from a single case. The report involved parenteral dosing with other drugs given at the same time [15].

Regulatory status

Anemia warning as quoted in the Iron entry [1]. The portions of the label read did not list hemolytic anemia, methemoglobinemia or porphyria.

Population evidence

Case report [15], Tier B: a 76-year-old woman given intramuscular diclofenac, thiocolchicoside and diazepam for low back pain was admitted five days later with asthenia, nausea, vomiting and diarrhea; hemolysis and a positive direct antiglobulin test were found, other causes were excluded, and she recovered fully on glucocorticoids with the drug withdrawn, with no recurrence at long-term follow-up. The authors call drug-induced immune hemolytic anemia rare and underdiagnosed. Two further case titles, a 1995 autoimmune hemolytic anemia report and a 2001 report of fatal acute immune hemolytic anemia, were found by Crossref and OpenAlex metadata only, with no abstract retrievable, so they are not used for any claim [16][22].

Mechanism

Immune-mediated drug hemolysis in the case report [15] (antibody details not retrieved); GI blood loss and fluid retention per label [1]. Neither is a measure of tissue oxygen delivery or utilization.

Counter-evidence held on record

Counter-evidence and confounding A case report cannot show frequency or causation beyond a dechallenge, and two other drugs were co-administered. Anemia from blood loss is already carried in the Iron entry and is not counted twice (see Overlap map). Framing caution Anemia is an outcome, not a nutrient. This entry does not claim reduced oxygen-carrying capacity as a general effect of diclofenac. Unexplained weakness, dark urine or jaundice on an NSAID needs prompt medical assessment, not supplements.