Griseofulvin

Limits on what this shows

• Null and species findings: Dubin-Johnson patients and carriers given griseofulvin showed no signs of porphyria and no change in urinary coproporphyrin (n not given) [15]. Griseofulvin inhibited ferrochelatase by up to 95% in mouse hepatocytes but had no effect in rat hepatocytes [17]. The porphyrogenic effect in humans is judged weaker than in mice [14]. The human porphyria cutanea tarda study of 1968 was not read [16].
• At-risk subgroups: People with any porphyria or hepatocellular failure (labeled contraindication); long courses such as nail infections (at least 4 months for fingernails and 6 months for toenails per the ultramicrosize label) [2].

Direction

Excess / disturbance (porphyrin accumulation, possible reduction of the free-heme pool in mice). Not depletion of any dietary nutrient.

Regulatory status

Both labels read state the drug is contraindicated in porphyria or hepatocellular failure [1][2]. LiverTox: inhibits ferrochelatase, raises intrahepatic protoporphyrin and can trigger acute attacks in acute intermittent porphyria, including in remission, per case reports [5].

Population evidence

Human: in a pilot infection trial, two subjects on 500 mg twice daily had plasma PPIX of 0.08 (SD 0.03) and 0.12 (SD 0.06) micromolar on Day 7, with none detectable in two untreated samples; red-cell PPIX, N-methyl protoporphyrin and ferrochelatase were not measured in humans [13]. Mouse: mitochondrial protoporphyrin rose from 0.0076 to 4.11 nmol/mg protein at 3 days and 22.2 at 4 to 5 weeks (n = 5) with ferrochelatase inhibited 80 to 90% [11]; after 4 days ALA synthase rose 3.8 to 6-fold in non-nodule liver and ferrochelatase fell to 13.5 to 18% of pretreatment values [40]; liver N-methyl PPIX rose six-fold [23]; PPIX accumulated in liver, blood and feces on 0 to 2.5% diets for 7 to 14 days [18]; topical exposure also raised liver porphyrins [39].

Mechanism

Named: ferrochelatase inhibition, with N-alkylated protoporphyrins (N-methyl PPIX) as the known inhibitor class, and compensatory ALA synthase induction [47][11][23][40].

Grade note

Why not orange: the only measured human porphyrin value is n = 2 healthy volunteers with no clinical readout [13], and the one human study with a porphyria-type readout was null [15]. This entry would move up only with a human cohort measuring erythrocyte porphyrins against a comparator.

Counter-evidence held on record

Counter-evidence on record A mouse gene-expression study titled "not a marked depletion of the heme regulatory pool" [21] sits against a low holo/apo tryptophan pyrrolase ratio read as a reduced free-heme pool [18]; the two mouse reports do not agree on heme-pool depletion. Framing caution This is a toxicity and excess-direction finding. It does not imply that iron, heme or any vitamin should be taken, and supplementation would not address porphyrin accumulation. Anyone with a porphyria history should not take the drug (label contraindication).

Limits on what this shows

• At-risk subgroups: Not defined by evidence. Long courses (4 to 6 months or more for nail infection) [2] would be the plausible exposure window.

Direction

Depletion pathway by accelerated catabolism (hypothetical). No measured vitamin D change for this drug.

Regulatory status

Neither label read mentions vitamin D or bone [1][2]. Induction-type interactions (warfarin, cyclosporine, estrogens) are labeled [1].

Cohort test

CLASS EXTRAPOLATION. Rifampin was the inducer in the human vitamin D data (six healthy volunteers, 4beta,25(OH)2D3 formation induced) [28]; the JCI study tested rifampicin, mifepristone, clotrimazole and PCN, with griseofulvin not tested or mentioned [29]. Griseofulvin was NOT in either cohort, so those sources are Tier C for this drug and capped at yellow.

Drug-specific support

In vitro, griseofulvin was a moderate PXR activator (at least 7-fold in HepG2 cells) with a tendency to induce CYP3A4 [25]; mouse liver Cyp3a expression rose [26]; rat hepatocyte monooxygenase activity rose [43]; prescribing literature treats griseofulvin as an inducer [27]. These are Tier B mechanism data; none measures vitamin D.

Searches

PubMed "griseofulvin vitamin D" (5 records, none with a vitamin D measurement in griseofulvin users; the Hathcock review links anticonvulsant inducers, not griseofulvin, to vitamin D deficiency [30]); "griseofulvin bone" (35 records, first 10 reviewed, no vitamin D or osteomalacia report). No osteomalacia case report for griseofulvin was found.

Counter-evidence held on record

Counter-evidence on record In the JCI work, PXR ligands induced CYP3A4 but did not induce CYP24 and SXR repressed VDR-driven CYP24 [29], so enzyme induction does not by itself show lower vitamin D status. No 25-hydroxyvitamin D, PTH or calcium data exist for griseofulvin users. The label advises avoiding intense or prolonged sunlight [1]; any effect of that advice on vitamin D status is an unstudied inference and a confounder for any future study. Framing caution Mechanistic and class-level plausibility only. Not a finding that griseofulvin lowers vitamin D, and no supplement is implied.

Limits on what this shows

• Confounding: Confounding by indication is presumed (Hard Rule 6): tinea pedis and psoriasis, diet, and concurrent treatment are uncontrolled unknowns, and there may be no unexposed comparator.

Direction

Unknown from retrievable text. The 1978 indexing (riboflavin deficiency chemically induced; nicotinic acids antagonists and inhibitors) suggests lower levels in tinea pedis patients on griseofulvin; the 1974 psoriasis record carries no deficiency subheading.

Regulatory status

Neither label read mentions B vitamins. Peripheral neuropathy and paresthesia of the hands and feet are listed adverse reactions [1][2]; these are outcomes and are not filed here as B vitamin deficiency (Hard Rule 2).

Population evidence

Belukha 1978 (PMID 735339): adults with Trichophyton rubrum tinea pedis treated with griseofulvin; riboflavin and nicotinic acid content studied; no abstract; sample size and assay unknown [44]. Pankratov 1974 (PMID 4432619): psoriasis patients; thiamine and nicotinic acid reserves; no abstract; the record also indexes ointments and diphenhydramine as co-treatments [45].

Cohort test

Both titles state griseofulvin-treated patients, so the drug was in the cohort by title; contents unverified.

Mechanism

None named in any retrievable text. Absent.

Searches

PubMed "griseofulvin riboflavin" (4 records), "griseofulvin nicotinic acid" (12 results, 10 shown), "griseofulvin vitamin" (64 results, first 10 reviewed; none measured vitamin status), "griseofulvin vitamin B12" (1 result, a fermentation review, not relevant), "griseofulvin pyridoxine" (2 results, not relevant). OpenAlex abstracts were null for both PMIDs; a Crossref title search did not return the 1978 paper. Repair pass 2026-10-10: PubMed "griseofulvin folic acid" and "griseofulvin folate" each returned 0 results, and a web search pairing griseofulvin with folate deficiency, megaloblastic anemia or B12 returned no griseofulvin-specific record (the one case report opened was a trimethoprim-sulfamethoxazole case that does not mention griseofulvin [52]), so folate is now searched and is a genuine gap on those channels. Both unread PMIDs were opened again as PubMed single records (format=pubmed): no abstract field for either, MeSH terms only; the OpenAlex works/pmid record for 735339 again has a null abstract, no DOI and closed access; Russian-language and English web searches for both titles returned no digitized text; the Europe PMC query was rate-limited (429) and the journal archive search page is robots-disallowed. Both studies remain unread, so the grade is unchanged.

Grade note

Yellow rather than ungraded because Tier B/C-eligible human records exist. Nicotinic acid appears in two independent cohorts (tinea pedis, psoriasis); if full text confirms lower measured levels against controls with a plausible mechanism, this entry could rise. Not graded above yellow while unread.

Counter-evidence held on record

Counter-evidence on record Skin disease itself can alter B vitamin status; the Russian-language records have no retrievable comparator; no Western cohort, label statement or mechanism supports a B vitamin effect. Framing caution A literature pointer, not a deficiency claim. Paresthesia or neuropathy on this drug is a labeled adverse reaction and does not show a B vitamin deficiency. No supplement is implied.

Limits on what this shows

• At-risk subgroups: Children (the four-fold milk effect), people on fat-restricted diets or fat malabsorption (hypothetical, not studied).

Direction

Nutrient to drug (food raises exposure). No effect of the drug on fat or fat-soluble nutrient status was found. Hard Rule 4 applies.

Regulatory status

Microsize label: absorption varies 27 to 72%; "the serum level may be increased by giving the drug with a meal with a high fat content"; children given 10 mg/kg with milk had mean peak 1.29 versus 0.34 mcg/mL alone, about four-fold [1]. The label's Dosage and Administration text, as read, gives no fat-meal instruction. Ultramicrosize label: absorption about 1.5 times microsize; no food statement; bioequivalent when crushed in applesauce in 24 fasted volunteers [2].

Population evidence

Label pharmacokinetics only (fed versus fasted numbers for adults not quoted in the text read). A review states high-fat meals enhance griseofulvin absorption [30]. Consumer pages advise taking it with milk or fatty food [37][36][6] (Tier D).

Mechanism

Fat-stimulated bile and micellar solubilization of a poorly water-soluble drug is the generally accepted explanation but is not stated in the label text read; mechanism not independently sourced here.

Interval

No separation interval applies. The practical point is consistency of administration conditions, particularly because absorption already varies between 27 and 72%.

Counter-evidence held on record

Counter-evidence on record The ultramicrosize formulation label makes no fat-meal statement, so the food effect is established on the record only for microsize. Framing caution Administration guidance for a prescriber or pharmacist to give. It is not a deficiency and no dose of any nutrient is recommended.

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C evidence and no circulating claim found. Genuine gap on the PubMed channel; the consumer-claim side is tooling-limited. Not a finding of safety.

What was searched

PubMed "griseofulvin zinc": 8 results, none measuring zinc status; they are zinc nanoparticle formulations, zinc pyrithione shampoos, a protoporphyrin NMR comparison and old therapy records. One web search on trace elements in tinea patients returned non-drug-specific pages (not opened).

Channel outcomes

Clean zero for zinc as a measured nutrient in griseofulvin exposure on PubMed. Labels read name no zinc interaction. Consumer-claim sweep limited to three pages that made no zinc claim. Repair pass 2026-10-10: the combined copper and zinc PubMed query (16 results, 10 viewed) again showed zinc only as a formulation component, an antifungal shampoo agent or a structural complex, never as a measured nutrient in griseofulvin users. The UK SmPC lists no zinc or mineral interaction [48]. The only zinc status study found in tinea patients does not state whether griseofulvin was used [54].

Limits on what this shows

• Why no grade: Dual condition met: no Tier B or C source supports even a yellow grade (Tier D aggregator text only), and no circulating claim of nutrient depletion through taste or appetite was found. Intake effects of an adverse reaction are not nutrient entries by themselves. Not a finding of safety.

What was searched

PubMed "griseofulvin taste disorders": 3 results (a 509-child griseofulvin RCT arm whose abstract does not report taste events [32]; a five-drug onychomycosis comparison reporting taste disturbance for terbinafine only [33]; a 1971 Norwegian record indexed for taste disorders without drug attribution). "griseofulvin taste" and "griseofulvin dysgeusia" hit rate limits and were not read. A terbinafine taste case report was excluded because griseofulvin was not studied [46]. Drugs.com lists taste changes and anorexia as uncommon (0.1 to 1%) [35] (Tier D). Neither label read mentions taste.

Channel outcomes

Mixed: genuine gap for case reports on PubMed, plus two rate-limited query variants. Web search was exhausted, so case-report phrasing in other indexes was not tried. Repair pass 2026-10-10: a retry of PubMed "griseofulvin dysgeusia" returned a bot-verification page. A web search for griseofulvin with dysgeusia, taste disturbance, loss of taste, anorexia or weight loss returned only terbinafine pages. The UK SmPC (Tier A) lists diarrhoea, nausea and vomiting as common and headache and gastric discomfort as sometimes occurring, and lists no taste or appetite change [48]. Those gastrointestinal events are outcomes that name no nutrient and carry no intake, weight or biomarker data, so they do not support a nutrient grade (Hard Rule 2). Genuine gap for taste and appetite; tooling-limited only for PubMed case-report phrasings.