Dextromethorphan

Limits on what this shows

• Null findings: No effect on electron transport chain activity in the isolated mitochondria [7]. For scale (reviewer arithmetic, molecular weight 271.4): 69.3 uM is about 19 ug/mL, whereas a serum dextromethorphan level of 398 ng/mL, measured at admission after a 300 mg overdose with serotonin syndrome [15], is about 1.5 uM. The authors state the IC50 is far above concentrations needed for known targets and that in vivo relevance may arise only at very high exposure or through mitochondrial accumulation, which they did not measure [7].
• At-risk subgroups: Not defined. The rat dose is described by its authors as supra-antitussive [8]; any relevance would be to overdose or misuse, not to labeled OTC use.

Regulatory status

No mitochondrial statement in the labels read [1][2][3].

Population evidence

None in humans. [8] Tran 2016, rats, 80 mg/kg intraperitoneally: mitochondrial dysfunction and myelinoid bodies in the hippocampus on ultrastructure; neither MK-801 nor naloxone prevented the mitochondrial change or the myelinoid bodies, so the mechanism is unexplained. The measures used for mitochondrial function and group sizes were not retrievable (abstract and Crossref metadata only; full text not opened). Repair pass: the PubMed record (PMID 27802908, no PMC copy listed) and the OpenAlex record were opened and contain only the abstract; the full text is CC BY-NC-ND on ScienceDirect, and both ScienceDirect and J-STAGE full-text pages were barred or errored for the fetch tool. [7] Algieri 2026, isolated swine heart mitochondria (at least 3 independent preparations): dextromethorphan inhibited F1FO-ATPase hydrolysis with IC50 69.3 +/- 4.1 uM; at 65 uM, ADP-stimulated and FCCP-uncoupled respiration were unchanged (electron transport chain not affected), oligomycin-inhibited respiration rose and the coupling index fell. Membrane potential, ROS and apoptosis were not measured.

Mechanism

Named in vitro: binding at or near the membrane-embedded FO domain of the ATPase (functional DCCD mutual-exclusion data; structural binding not shown) [7]. The in vivo mechanism is not established [8].

Counter-evidence held on record

Counter-evidence Single animal study with unexplained mechanism and an in vitro enzyme study at concentrations well above plasma levels reported in overdose. No cell, human or therapeutic-dose mitochondrial data exist in what was retrieved. Framing caution Do not present as mitochondrial damage in people taking dextromethorphan, and do not recommend mitochondrial or antioxidant supplements from this entry.

Limits on what this shows

• Null findings: In vivo superoxide and NOX2 activity change in EAE was non-significant [10]; no direct protection against MPP+ or amyloid toxicity [9].

Regulatory status

No oxidative stress statement in the labels read [1][2][3].

Population evidence

None in humans. [9] Rat midbrain neuron-glia cultures with LPS: dextromethorphan 1 to 10 uM (30-minute pretreatment) reduced LPS-induced superoxide, TNF-alpha and nitric oxide and protected dopaminergic neurons, even when added up to 60 minutes after LPS; it did not protect against direct MPP+ or beta-amyloid toxicity, and the NMDA antagonist dizocilpine gave no significant protection, so the authors consider an NMDA-receptor mechanism unlikely (page gave no title or authors; DOI 10.1124/jpet.102.043166, J Pharmacol Exp Ther). [10] Chechneva 2011, mouse EAE: 0.1 mg/kg daily intraperitoneally cut spinal cord p47phox and iNOS mRNA and, in microglia, NOX2-dependent superoxide; the in vivo NOX2 activity and superoxide reduction was not statistically significant, and 10 mg/kg gave no protection. [11] Ommati 2022, 36 male BALB/c mice with acetaminophen-induced acute liver failure: dextromethorphan 0.5 to 10 mg/kg subcutaneously, three doses 6 hours apart, lowered brain ROS and TBARS and restored glutathione and ferric-reducing antioxidant capacity, with larger effects at 5 and 10 mg/kg (n = 6 per group; acetaminophen 1000 mg/kg intraperitoneally, dextromethorphan given from one hour after acetaminophen; one-way ANOVA with Tukey test; the brain ROS, TBARS, glutathione and ferric-reducing values appear only in Figure 3, an image, so numeric values were not retrievable from the page text; serum ALT, AST and bilirubin in the page's Table 1 did not differ significantly between acetaminophen alone and any acetaminophen plus dextromethorphan group). Opposite direction: [8] Tran 2016, rats given 80 mg/kg intraperitoneally, cytosolic oxidative burden rose and MK-801 reduced it.

Mechanism

Named for the protective direction: suppression of microglial activation and NOX2 (p47phox, gp91phox) superoxide generation [9][10]. The burden at 80 mg/kg is attributed in the abstract to dextromethorphan toxicity with an unclear mechanism [8].

Counter-evidence held on record

Counter-evidence All models are disease or toxin models, not measurements of redox balance in healthy dextromethorphan users. The model doses (0.1 to 10 mg/kg) are not mapped to human exposures here, and the 80 mg/kg rat result points the other way. Framing caution Not an antioxidant benefit claim and not an antioxidant-depletion claim. Dextromethorphan should not be described as depleting antioxidants or as an antioxidant.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source was opened to support even a yellow grade, and no circulating consumer claim linking dextromethorphan and copper was found. A disclosed gap, not a finding of safety. Kept independent of Iron (Hard Rule 7). Re-check a copper(II) coordination query and the unseen PubMed records when search access returns.

What was searched

WebSearch "dextromethorphan iron absorption OR ferritin OR iron deficiency OR ceruloplasmin OR serum copper" and "dextromethorphan serum magnesium OR zinc OR copper OR iron levels measured": result lists were rodent copper-deficiency and ceruloplasmin papers and serum Zn/Cu/Mg studies in obese children and multiple sclerosis patients, none concerning dextromethorphan (titles only; none opened). A third query on copper(II) or metal-ion binding was refused because the shared search budget was spent. Labels [1][2][3]: no copper statement. Repair pass: a PubMed listing for dextromethorphan AND (copper OR iron OR ferritin) returned 11 results (10 shown by the fetch tool), none in which dextromethorphan was given and copper or ceruloplasmin measured; copper appeared only as a reagent in an electrochemical detection paper (PMID 38738557) and as an indexing term in a methotrexate letter (PMID 31027652) (summaries only, none opened). A second listing, dextromethorphan AND (chelation OR metal ions OR transferrin OR iron deficiency OR zinc), returned 35 results (10 shown): none measured a metal after dextromethorphan; the closest was the in vitro dextrorphan binding study already cited as [12] (PMID 1370704). The PubChem/HSDB Interactions, Human Toxicity, Drug Warnings, Adverse Effects and Non-Human Toxicity sections [33] and the StatPearls chapter [34] were opened and contain no copper, zinc or iron statement.

Channel outcomes

Initial pass: PubMed listing 429, PMC pages bot-verification, Crossref and OpenAlex unusable for listings. Repair pass: two PubMed listings returned results (above), WebSearch was unavailable (budget spent), only the first 10 of 11 and 10 of 35 records were visible to the fetch tool. A clean zero on four phrasings that ran (copper, ceruloplasmin, chelation, metal ions, zinc), with the unseen records and web search as the residual tooling limit.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C evidence and no circulating claim found. Not a finding of safety. Iron deficiency anemia, if ever seen, is an outcome and not an iron entry (Hard Rule 2).

What was searched

The same two WebSearch queries as for Copper (ferritin, iron deficiency, iron absorption); no dextromethorphan result. The labels contain no iron statement [1][2][3]. No label read describes an effect on gastric acid or iron absorption [1][2][3], so the acid-suppression pathway seen with drugs such as H2 blockers has no counterpart in the pharmacology described there. Repair pass: the same PubMed listings as for Copper (ferritin, iron deficiency, transferrin) were reviewed: a 2026 case report of dextromethorphan for restless legs syndrome (PMID 41994701, Cureus) mentions iron deficiency only as a possible cause of the restless legs being treated, not as an effect of dextromethorphan; a 2000 interleukin-10 study (PMID 10668851) used dextromethorphan as a CYP probe and reported ferritin rising after interleukin-10, which is not evidence about dextromethorphan (summaries only; neither opened). Au 2007 [29] measured hemoglobin and MCV in cough-mixture misusers (normal-range MCV, hemoglobin not lower) but no iron indices, so it does not inform this entry.

Channel outcomes

As for Copper. A genuine gap on the phrasings tried (iron, ferritin, transferrin, iron deficiency), with the unseen PubMed records and unavailable web search as the residual tooling limit.

Limits on what this shows

• Null findings: No human cohort was found. The Vich Vila et al. Nature Communications drug-microbiota analysis (DOI 10.1038/s41467-019-14177-z), which studied 41 medication categories in three Dutch cohorts, does not name dextromethorphan or antitussives in the page text opened [26], and the Maier 2018 screen of more than 1,000 marketed drugs against 40 gut strains [28] (page and full-text XML) and the Zimmermann 2019 study of 271 drugs and 76 strains [27] (page text) do not mention dextromethorphan. Their supplementary drug tables (Maier Supplementary Tables 3a and 3b list the anticommensal drugs) could not be read: the supplementary file host is barred to the fetch tool and PMC and Europe PMC listings were bot-challenged or rate-limited, so whether dextromethorphan was in those drug lists is unresolved. Three papers opened in the first pass do not support the entry: a yeast biotransformation study of dextromethorphan O-demethylation [18] (microbial cultures, not gut microbiota) and two rat dysbiosis papers that do not mention dextromethorphan [19][20].
• At-risk subgroups: Not defined. The abstracts do not map the 30 to 50 mg/kg intraperitoneal mouse doses to human exposure; any relevance would be to overdose or misuse, not to labeled use.

Regulatory status

No microbiome statement in the labels read [1][2][3].

Population evidence

None in humans. All three records below were read as PubMed abstracts only; full text sits on ScienceDirect, which the fetch tool is barred from reading. Drug to microbiome: [23] Cai 2026 (Brain Behav Immun 136:106790; PMID 42070701): male mice given dextromethorphan 30 or 50 mg/kg intraperitoneally once daily for 14 days; gut microbiota composition was altered; at the high dose interleukin-1 beta rose 30% in intestine and brain, interleukin-10 fell 15% in intestine and 35.7% in brain, and serum lipopolysaccharide rose 1.3-fold; intestinal and brain tissue damage and apoptosis were reported; gut-derived metabolites correlated with brain lysophosphatidylcholine disturbance. [24] Tian 2026 (Brain Behav Immun 137:106801; PMID 42114687): mice on a subchronic dose-escalating dextromethorphan regimen had more Akkermansia and Desulfovibrionaceae and fewer Muribaculaceae, lower fecal butyrate, impaired intestinal barrier and higher circulating lipopolysaccharide and cytokines; the probiotic Bacillus aerolatus CX253 preserved diversity and raised butyrate. Species, strain, sex, group sizes, dose, route, duration and p-values were not in the abstract of [24]; group sizes, taxa and diversity metrics were not in the abstract of [23]. Microbiome to drug (opposite direction, Hard Rule 4): [25] Xie 2026 (Drug Metab Dispos 54:100241; PMID 41687425): in male C57BL mice given a four-probe CYP cocktail by gavage, the age-related rise in dextromethorphan exposure in old mice was reversed when gut microbiota were cleared, and transplant of young-mouse microbiota did not change probe-drug pharmacokinetics; no dose, group size or numeric values in the abstract.

Mechanism

Not established. The abstracts associate the changes with barrier impairment, lipopolysaccharide, lower butyrate and neuroinflammation [23][24], but do not show that dextromethorphan acts on bacteria directly; no direct antibacterial activity was found (see null findings).

Counter-evidence held on record

Counter-evidence Intraperitoneal injection bypasses direct gut exposure, so the microbiome change may be systemic or toxicity-driven; two of the three studies are from the same journal in the same year and the dosing regimens differ; all three are abstract-level reads. Whether dextromethorphan at human oral doses changes the microbiome has not been tested. Framing caution Not a claim that dextromethorphan disrupts the human gut microbiome, and not a reason to recommend probiotics, butyrate or fiber. Drug toxicity at supratherapeutic exposure, not a nutrient deficiency.

Limits on what this shows

• Null findings: The between-group red-cell change was not significant (p = 0.117) [30]; hemoglobin and MCV were not lower in misusers [29]; no hemolysis or methemoglobinemia attributable to dextromethorphan was found in the sources opened.

Regulatory status

The quinidine label lists immune-mediated thrombocytopenia and hemolytic anemia as quinidine reactions [2], and the bupropion label lists anemia, leukopenia, pancytopenia and thrombocytopenia among bupropion postmarketing reports [1]; neither assigns them to dextromethorphan. The PubChem/HSDB Human Toxicity Excerpts, Drug Warnings and Adverse Effects sections [33] and the StatPearls chapter [34] list no hematologic adverse effect of dextromethorphan; StatPearls mentions electrolytes only as hyperkalemia monitoring during rhabdomyolysis in overdose [34].

Population evidence

Lu 2020 [30]: randomized, double-blind, 12 weeks, bipolar disorder (Taiwan), 255 randomized and 166 completers analyzed (valproate plus placebo n = 57; plus dextromethorphan 30 mg a day n = 56; plus dextromethorphan 60 mg a day n = 53; dropout 34.9%). Red cells (10^6 per cmm, mean, baseline to week 12): placebo 4.73 to 4.56 (p < 0.0001); 30 mg 4.58 to 4.45 (p = 0.007); 60 mg 4.61 to 4.57 (p = 0.51). Hemoglobin (printed as gm/ml; presumably g/dL): 14.3 to 14.0 (p = 0.010); 14.0 to 13.8 (p = 0.04); 13.9 to 13.9 (p = 0.94). Hematocrit: 42.2 to 41.0 (p = 0.005); 41.1 to 40.3 (p = 0.012); 40.7 to 41.0 (p = 0.50). Those p-values are within-group paired tests. Between-group comparison of week-12 change (one-way ANOVA): red cells F = 2.172, p = 0.117; neutrophils p = 0.041; white cells p = 0.078. Cohort test passed: dextromethorphan was the randomized add-on. Au 2007 [29] (cough-mixture misusers, dextromethorphan among the antitussives): hemoglobin 14.3 vs 13.7 g/dL (p = 0.07) and MCV 90.0 vs 90.2 fL (p = 0.42) versus other substance misusers, so no anemia or macrocytosis. A PubMed query (dextromethorphan with hemolysis, methemoglobinemia, anemia, hemoglobin; 13 results, 10 shown) found no record attributing a hematologic change to dextromethorphan as a cause: the anemia worsening in one record was attributed to the study drug BTZ-043 with dextromethorphan only as a CYP probe, and the platelet fall in another to interleukin-10 (summaries of both listed records; not opened). Case reports: a massive overdose with hemoglobin falling from 14.1 to 9.5 g/dL, unexplained, in a patient with status epilepticus, lactate 14.7 mmol/L and five co-ingested drugs [16] (unattributable, not used as evidence), and SpO2 97% on room air at admission after a 300 mg overdose [15].

Mechanism

None named for a hematologic effect of dextromethorphan in either direction. The trial authors suggest dextromethorphan might help maintain hematologic parameters during valproate treatment without testing a mechanism [30].

Counter-evidence held on record

Counter-evidence and confounding There was no dextromethorphan-alone arm, so the fall in counts is a valproate effect and dextromethorphan's role is only relative to valproate plus placebo. The red-cell and hemoglobin "protection" rests on within-group p-values and not on a significant between-group difference; the dropout rate was 34.9% and only completers are reported; permitted co-medication could obscure blood-count changes and the authors call it a pilot study [30]. The misuse cohort [29] has no non-misusing comparator and no direct oxygen-delivery measure. Framing caution Not an oxygen-carrying benefit claim and not an oxygen-delivery deficiency. Blood counts are outcomes, kept separate from ATP production and from Iron (Hard Rule 7).

Limits on what this shows

• Null findings: Electron transport chain activity was unaffected; ATP content in cells or tissue was not measured by anyone in the retrieved set.

Regulatory status

No energy-metabolism statement in the labels read [1][2][3].

Population evidence

None in humans, cells or live animals. Algieri 2026 [7] (BioFactors 52:e70076, open access): ATP hydrolysis IC50 69.3 +/- 4.1 uM, complete inhibition at 0.5 mM, mixed inhibition (Ki 0.342 mM, K'i 0.038 mM). ATP synthesis was inferred from respiration only: State 3 respiration unchanged, State 4o raised, State 3/State 4o coupling index reduced; maximal uncoupled respiration unchanged. The text also mentions a significant "State 4u" increase while the figure caption says State 4o (an inconsistency in the summarized text).

Mechanism

Named: inhibition of the F1FO-ATPase, independent of thiol redox state, with a proposed site overlapping DCCD in the FO domain [7].

Counter-evidence held on record

Counter-evidence The authors cite much lower potencies at known targets (serotonin transporter 33 nM, NMDA receptors 1.04 uM) and say in vivo relevance is uncertain [7]. Same underlying study as part of the mitochondrial entry (see Overlap map). Framing caution Not a finding of reduced energy in people. "Low energy" should not be attributed to dextromethorphan from this entry.