Loratadine

Limits on what this shows

• Null findings: No loratadine-specific loss of mitochondrial membrane potential in the records read [16]. PubMed "loratadine mitochondria" was throttled (429) and was run instead as the membrane-potential phrasing.
• At-risk subgroups: None established.

Regulatory status

Not mentioned in the OTC label [1].

Population evidence

None in humans. In vitro (Tier B): loratadine reduced MitoSOX-detected mitochondrial ROS and NOX4 in AGE-treated SW1353 human chondrocyte cells via NRF2 [12] (summary reconstructed from OpenAlex word index; concentrations not stated). Three PubMed records from a "loratadine mitochondrial membrane potential" search used loratadine only as a comparator or found membrane-potential loss for other drugs (clemastine plus ibrutinib; terfenadine), not loratadine [16] (listing-level summaries). In CLL cells loratadine induced lysosomal membrane permeabilization and cell death at unstated concentrations [16].

Mechanism

Protective redox signaling in one model; no mitochondrial toxicity mechanism established for loratadine.

Class cohort test

Loratadine itself, in vitro. No class extrapolation used.

Counter-evidence held on record

Counter-evidence Single cell-line result in a transformed line (SW1353); not shown at clinical concentrations. Framing caution Reverse-direction finding. Not a claim that loratadine protects or harms mitochondria.

Limits on what this shows

• Null findings: No human study measuring oxidative stress markers in loratadine users was found. In [13] the markers were elevated by the disease and the treated change may reflect disease control; there was no placebo arm.
• At-risk subgroups: None established.

Regulatory status

Not mentioned in the OTC label [1].

Population evidence

METABOLITE EXTRAPOLATION (labeled as class extrapolation under Hard Rule 6; loratadine was NOT in these human cohorts). (a) Desloratadine in 10 adults with chronic idiopathic urticaria, before and after 4 weeks, versus 10 healthy controls: platelet ROS and superoxide dismutase activity were higher than controls at baseline and fell after treatment (P<0.005); authors call it preliminary [13]. (b) Desloratadine 5 mg/day for two months in 50 children aged 3 to 10 with perennial allergic rhinitis plus 11 controls: erythrocyte SOD, catalase and malondialdehyde "normalized", plasma hydroperoxides lower than untreated but still above controls, total antioxidant status not significantly changed; no magnitudes or p values in the abstract [14]. Loratadine itself, in vitro: advanced glycation end product-treated SW1353 human chondrocyte cells, loratadine reduced mitochondrial ROS and NOX4 production, NRF2-dependent (concentrations not stated in the record) [12]. In multidrug-resistant KB-V-1 cells loratadine suppressed cystine uptake, depleted glutathione and promoted lipid peroxidation (IC50 about 2.0 micromolar); mouse xenograft at 20 mg/kg intraperitoneal showed tumor suppression "without systemic toxicity" [11]. Loratadine itself, in vivo (Tier B): in an experiment comparing resveratrol and loratadine against carrageenan-induced inflammation, the abstract states that loratadine produced oxidative stress in skin tissue and in serum and did not show the hoped-for protective effect [27]; species, dose, markers and magnitudes are not given in the abstract, which was read through the journal record and an OpenAlex abstract. An environmental study reports loratadine raised ROS in bacterial donor cells [15] (listing-level).

Mechanism

Present but direction depends on context. Anti-oxidative: NRF2 modulation and reduced NOX4 [12]. Pro-oxidative: cystine uptake inhibition with glutathione depletion, PERK-eIF2alpha-ATF4 stress signaling [11], and an unexplained pro-oxidant result in vivo [27]. None of these mechanisms has been shown at clinical plasma concentrations (about 5 to 7 nanomolar).

Class cohort test

Desloratadine is the active metabolite of loratadine and is present in plasma after loratadine dosing [3], so the extrapolation is mechanistically reasonable but loratadine was not in either human cohort; capped at yellow. In vitro loratadine data are Tier B and cell-line specific. Cancer-cell ferroptosis data are not generalized to normal tissue.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication (Hard Rule 6) is likely in [13][14]: allergic inflammation itself raises oxidative markers. Concentration mismatch limits [11][12]. The pro- and anti-oxidant findings point in opposite directions and are not additive. Framing caution Not a claim that loratadine causes or relieves oxidative stress in users. Do not recommend antioxidant supplements from this entry.

Limits on what this shows

• Null findings: No peer-reviewed study of loratadine and serum iron, ferritin, hepcidin or anemia was found. PubMed "loratadine serum iron" returned no records (404), and "loratadine hepcidin OR ferritin" returned only unrelated records (no loratadine or desloratadine in any). The "loratadine iron" listing returned four records, two relevant [10][11]. Repair pass 2026-10-08: PubMed "loratadine hepcidin" and "loratadine ferritin" each returned zero records (404), and web searches combining loratadine or desloratadine with hepcidin, HAMP, ferritin, serum iron, iron absorption or iron deficiency returned no loratadine-specific iron study. A near-red upgrade would need the n, exposure definition and CI of the Cedars-Sinai analysis plus a second independent peer-reviewed cohort measuring iron indices; neither could be retrieved.
• At-risk subgroups: None established. People with iron-overload disorders would be a theoretical group of interest only; no loratadine evidence addresses them.

Regulatory status

Not mentioned in the OTC label [1].

Population evidence

Jiang et al. 2026 preprint [10] (bioRxiv; PubMed-indexed as not-MEDLINE; not peer reviewed). EHR data from Cedars-Sinai Medical Center, 2015 to 2025, "matched comparison framework". Outcome: serum iron (microgram/dL). Matched, unadjusted mean serum iron higher in loratadine-exposed patients by "about 4 microgram/dL"; adjusted regression "about 6 microgram/dL". Exposed n, control n, exposure definition, matching method, covariates, 95% CI for the adjusted difference and p value are not reported in the text; the figure legends describe error bars as 95% CI of group means but give no numbers. Ferritin and transferrin were not reported. Re-read on 2026-10-08 (bioRxiv v1 full text through a summarizing fetch tool): the Methods section covers only the proteomics and bench work and has no EHR methods, so exposed n, control n, matching ratio, covariate list, regression model and exposure definition remain unstated anywhere in the text; the Figure 5 legend describes covariate balance, group means with 95% CI error bars and the adjusted difference, without numbers. The bioRxiv page shows no v2 and no published-version notice, and the OpenAlex record (cited by 0) lists no journal version, so no peer-reviewed publication exists to retrieve. Figures themselves were not viewed.

Mechanism

Named but weakly anchored. In a 172-compound HepG2 proteomic screen (10 micromolar, 24 hours) loratadine showed "marked downregulation of HAMP (hepcidin)"; no fold change, p or q value, and no qPCR or ELISA validation are given in the text [10]. Lower hepcidin would favor higher iron absorption and release. Separately, loratadine raised the labile iron pool in multidrug-resistant KB-V-1 cancer cells (IC50 about 2.0 micromolar, versus about 48.45 micromolar in chemosensitive KB-3-1 cells) as part of ferroptosis [11]. Full text of [11] re-opened on 2026-10-08: Fe2+ (FerroOrange) was elevated after 5 micromolar loratadine for 24 hours in KB-V-1 cells (marked p<0.0001, direction only, no numeric values), and 2 micromolar loratadine increased transferrin receptor 1 and GPX4 expression; no copper or ceruloplasmin endpoint. Reported human peak plasma loratadine is about 2.1 to 2.6 ng/mL [3], roughly 5 to 7 nanomolar by my conversion (molecular weight about 383 g/mol), so both cell systems used concentrations hundreds to over a thousand times higher.

Class cohort test

Not a class extrapolation. Loratadine itself is the exposure in [10] and [11]. The limitation is source quality (preprint, EHR confounding, undisclosed n), not class. Under the tier definitions the preprint is not Tier A, so this entry cannot reach orange.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication and by testing is presumed: serum iron varies with time of day, fasting, inflammation, supplement use and which patients get iron labs drawn, and the EHR design cannot exclude these. A 4 to 6 microgram/dL difference was not interpreted against a reference range in the source. The authors state the analysis "does not establish causality" and is a check on a cell-screen hypothesis rather than an independent test [10]. Framing caution Excess/higher-circulating direction. Do not read this as an iron deficiency and do not recommend iron supplementation or iron avoidance from this entry. It is a monitoring-interpretation note: if serum iron is measured in a loratadine user, this unreplicated association exists.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source supports even a yellow grade, and no circulating copper claim for loratadine was found to rebut. Kept independent of Iron (Hard Rule 7). Circulating-claim recheck 2026-10-08: the depletion pages opened [29][30][31] do not name loratadine for copper. A disclosed gap, not a finding of safety.

What was searched

PubMed listings: "loratadine copper" (2 records: a neuronal histamine receptor paper in which copper only classifies diamine oxidase, and a copper metal-organic framework carrying desloratadine; neither measures copper status) and "loratadine ceruloplasmin" (no records, 404). The iron and ferroptosis results [10][11] were read for any copper content: none reported for loratadine. Repair pass 2026-10-08: a PubMed query pairing loratadine with an OR-group of trace element and copper terms returned 10 unrelated copper records with no loratadine in any (the OR grouping did not restrict to loratadine, so it is a failed query design, not a null result); web searches for loratadine with serum copper or ceruloplasmin returned no loratadine study; the full text of [11] has no copper or ceruloplasmin endpoint.

Channel outcomes

PubMed listing pages worked intermittently with repeated 429 rate limits. OpenAlex search and Crossref title search returned 429; Europe PMC returned 429; WebSearch budget exhausted. Repair pass 2026-10-08: WebSearch worked and PubMed answered after spacing calls about 3 minutes apart; one PubMed request was refused (429, with a request not to refetch) and was not retried; the OpenAlex search listing returned 429. Genuine gap in PubMed-indexed literature and open web results; not exhaustively searched in other databases.

Limits on what this shows

• Null findings: PubMed "loratadine microbiome" returned five records, none measuring a loratadine effect on human gut taxa; "antihistamine gut microbiota" returned one directly relevant human paper (antihistamine-resistant urticaria versus others, agents unnamed, not a drug effect) [26].
• At-risk subgroups: None established.

Regulatory status

Not mentioned in the OTC label [1].

Population evidence

No human gut microbiome measurement with loratadine as the exposure. Tier B, listing-level only (abstracts not opened): loratadine modulated MRSA virulence, resistance and biofilm genes (RNA-seq) [18]; loratadine restored colistin activity against resistant Klebsiella pneumoniae and, combined with colistin, reduced intracellular ATP in those bacterial cells [18]; loratadine at environmentally relevant concentrations raised ROS and promoted plasmid conjugative transfer in bacterial donor cells [15]; in water microcosms loratadine favored Proteobacteria [26]. Tier C CLASS EXTRAPOLATION: a 2025 Nature study found that 28% of 53 non-antibiotic drugs increased Salmonella Typhimurium growth in microbial communities; terfenadine, an H1 antihistamine, sped disease onset in mice [19]. Cohort test (re-run on 2026-10-08 with the PMC full text and the Nature article page): loratadine is not named anywhere in the main text. The drugs came from a screen of a library of 1,197 FDA-approved drugs (65 selected from 172 inhibitors, 53 tested in the challenge assay) and are itemized only in Supplementary Table 4, which could not be opened (the supplementary file host refused automated access and a direct download was blocked), so loratadine's membership is unknown. The Discussion says the drugs included antihistaminic agents, but the only antihistamine named is terfenadine, a different molecule. Human allergic rhinitis trials in which loratadine was only background or comparator (synbiotic trial, nasal irrigation, herbal decoction in mice) did not isolate a loratadine effect [26]. A cross-sectional case-control 16S study of chronic spontaneous urticaria patients unresponsive to second-generation antihistamines (15 healthy controls) found higher diversity and fewer short-chain fatty acid producers in patients, with no link to type of therapy, and did not separate antihistamine users from non-users [28].

Mechanism

Plausible but unquantified: direct antibacterial or bacterial-gene effects of a lipophilic amine, and loss of colonization resistance. Concentrations in the gut lumen after oral loratadine were not retrieved.

Class cohort test

Terfenadine to loratadine is class extrapolation capped at yellow; loratadine not confirmed in the cohort. The MRSA, Klebsiella and plasmid findings are loratadine-specific but bacterial in vitro, not gut-community data.

Counter-evidence held on record

Counter-evidence No human dysbiosis, C. difficile or diversity signal for loratadine was found. In vitro bacterial studies often use concentrations far above gut exposure. Framing caution Not a claim of dysbiosis in loratadine users. Do not recommend probiotics from this entry.

Limits on what this shows

• Null findings: PubMed "loratadine anemia" returned four records, one relevant [17]; a broader anemia/hemolysis/methemoglobin query returned only unrelated records. No population hemoglobin data for loratadine users were found.
• At-risk subgroups: Not defined by the source.

Regulatory status

The OTC label lists no hematologic warning; it asks users to stop and see a doctor for an allergic reaction [1].

Population evidence

Bollotte et al. 2014 [17], retrospective, 10 cases of drug-induced immune hemolytic anemia (median age 54.4 years). Culprit drugs: ambroxol, beta-interferon, cefotetan, ceftriaxone, loratadine, oxacillin, oxaliplatin, piperacillin-tazobactam, pristinamycin, quinine. Median hemoglobin nadir 57.9 g/L (range 34 to 78); direct antiglobulin test positive in 8; 5 cases definite (drug-dependent antibodies detected). Per-drug counts are not given; with 10 drugs and 10 cases loratadine is probably one case, which is my inference. Read through a PubMed record via a summarizing fetch tool.

Mechanism

Immune (drug-dependent antibody) hemolysis, an idiosyncratic reaction, not a nutrient pathway and not the drug's intended effect. No evidence of methemoglobin formation, reduced red-cell mass in typical users, or impaired tissue oxygen delivery was found.

Class cohort test

Loratadine was in the case series (Tier B, case series). Not a class extrapolation.

Counter-evidence held on record

Counter-evidence A single likely case among 10 across many drugs cannot estimate frequency or causality. Anemia is an outcome (Hard Rule 2) and is not filed as iron or B12 deficiency. Framing caution Not an oxygen-delivery deficiency claim. Unexplained jaundice, dark urine or new anemia on any drug warrants clinical evaluation, not supplementation.

Limits on what this shows

• Why no grade: Dual condition met for human cellular ATP: no Tier B or C evidence in mammalian cells or humans, and no circulating "low energy" claim for loratadine found. Not a finding of safety. Kept independent of Mitochondrial burden and Oxygen delivery/utilization (Hard Rule 7).

What was searched

PubMed "loratadine ATP" (10 records, listing-level): loratadine plus colistin reduced intracellular ATP in resistant Klebsiella pneumoniae with proton motive force disruption; a MRSA study investigated ATP but its abstract did not report the result; a microsphere transport paper mentions ATP uptake without a measurement. "loratadine ATP respiration" returned no records (404); "loratadine ATP oxygen consumption" was throttled (429, not retried on that URL). "loratadine mitochondrial membrane potential" (3 records) showed no loratadine effect in mammalian cells [16]. Repair pass 2026-10-08 (judgment call re-examined): a human cell source was actually opened, the full text of [11] (human KB-V-1 and KB-3-1 carcinoma cells treated with loratadine). Its assays were CCK-8 viability (a dehydrogenase readout, not an ATP measurement), cystine uptake, GSH, FerroOrange, BODIPY C11, qRT-PCR, RNA-seq, Western blot and xenografts; no ATP, membrane potential, oxygen consumption or mitochondrial endpoint was measured. [12] reports mitochondrial ROS only. A web search for loratadine reducing cellular ATP in mammalian cells returned no loratadine study. The bacterial ATP finding therefore stays unused for grading.

Channel outcomes

PubMed listings worked partially; other hosts 429. The bacterial ATP finding is Tier B for bacterial energetics and does not bear on human mitochondrial ATP production, so it was not used to grade this entry; it is disclosed here and in the microbiome entry. This is a judgment call.