Coumarin in Cinnamon: Toxicology, Exposure Limits, and the Importance of Species Authentication

Ceylon Nutritionals · Science & Education Series · Ingredient Science

Nalin Siriwardhana, PhD, FACN   ·   Nutritional Scientist   ·   March 2025

KEY TAKEAWAYS

→  Coumarin exposure from cinnamon varies significantly by species — by several orders of magnitude between C. verum and cassia-type species

→  EFSA Tolerable Daily Intake: 0.1 mg coumarin per kilogram of body weight per day (confirmed 2004, 2008)

→  Cassia-type cinnamon can approach or exceed this threshold under realistic dietary intake patterns, as documented in German and Norwegian exposure surveys

→  Published studies generally report substantially lower coumarin levels in authenticated C. verum than in cassia-type species

→  For repeated-use ingredient applications, species authentication and analytical coumarin testing are especially relevant variables in raw material evaluation

ABSTRACT

01 · Chemistry & Occurrence

Coumarin (1,2-benzopyrone) is a bicyclic lactone found naturally across a wide range of plant families, contributing the characteristic sweet, hay-like aroma to tonka beans, sweet woodruff, and several Cinnamomum species.1 In the context of food, botanical, and nutraceutical use, commercial cinnamon — particularly Cinnamomum cassia and related species — is a notable dietary source of coumarin exposure, particularly where cassia-type cinnamon is commonly consumed.

Coumarin should not be confused with anticoagulant drugs also called "coumarins" (warfarin, acenocoumarol). Those are coumarin derivatives with a distinct pharmacological mechanism; the naturally occurring 1,2-benzopyrone discussed here has no anticoagulant activity. Its primary safety concern is hepatotoxic potential at elevated intake levels — a well-characterised risk that forms the basis for the regulatory limits reviewed below.

02 · Metabolism & Mechanism

Coumarin's hepatotoxic risk is mechanistically tied to a specific metabolic pathway, not to the compound in all individuals. In the majority of humans, CYP2A6 efficiently converts coumarin to pharmacologically inactive 7-hydroxycoumarin (7-HC) for urinary excretion — the detoxification route.2 In a susceptible subpopulation, likely reflecting variability or competition in CYP-mediated metabolism, coumarin is diverted through CYP1A2 to form coumarin 3,4-epoxide, which rearranges to the hepatotoxic metabolite o-hydroxyphenylacetaldehyde (o-HPA). Reactive metabolites such as o-HPA are believed to contribute to hepatocellular injury through glutathione depletion and protein binding.1 Human hepatocytes oxidise o-HPA to a less toxic metabolite approximately 50 times faster than rat hepatocytes — explaining why humans are considerably less susceptible to coumarin-induced hepatotoxicity than rodent studies suggest.3

Metabolic Pathway Summary

Major route (most individuals): Coumarin → CYP2A6 → 7-Hydroxycoumarin → urinary excretion. Non-toxic; detoxification route.

Minor route (susceptible subpopulation — alternative CYP-mediated pathway): Coumarin → CYP1A2 → Coumarin 3,4-epoxide → o-HPA → glutathione depletion, hepatocellular damage. Hepatotoxic route.

03 · Clinical Evidence

Coumarin's decades-long use as a pharmaceutical agent — primarily for lymphoedema treatment in Europe — provides an unusually detailed human safety dataset for a naturally occurring phytochemical, at doses far exceeding dietary exposure levels.

Cox, O'Kennedy, and Thornes (1989) reported elevated liver enzyme levels in 0.37% of 2,173 patients treated therapeutically with coumarin, characterised as probable idiosyncratic hepatotoxicity.4 A synthesis by Casley-Smith et al. across five trials including 1,106 lymphoedema patients receiving 400 mg/day for a mean 14.6 months found no severe hepatotoxicity events in the controlled setting.1 A French pharmacovigilance survey (Andréjak et al., 1998) reported 34 hepatic cases from routine clinical use: 85% cytolytic, jaundice in 14, three severe cases — one requiring liver transplantation, two with fatal outcomes — though over 50% involved off-label use.5

Pitaro et al.'s 2022 narrative review in Molecules synthesised the full trial dataset, concluding that coumarin-induced hepatotoxicity appears restricted to a susceptible subgroup, predominantly presents as mild-to-moderate transient transaminase elevation that normalises after cessation, and is likely related to alternative CYP-mediated metabolism.1

04 · Regulatory Framework

EFSA's 2004 Scientific Panel Opinion concluded that coumarin carcinogenicity does not proceed via a genotoxic mechanism, permitting derivation of a threshold-based TDI. From a NOAEL of 10 mg/kg/day in a two-year dog study (uncertainty factor 100), EFSA established a TDI of 0.1 mg coumarin per kilogram of body weight per day — equivalent to approximately 7 mg/day for a 70 kg adult.6 EFSA's 2008 re-evaluation maintained this TDI unchanged after reviewing additional toxicity and metabolism data. The BfR independently confirmed the same TDI in 2006 and reconfirmed it in 2012.7

Abraham et al. (2010) subsequently derived a TDI of 0.1 mg/kg/day from human clinical hepatotoxicity data alone — an entirely independent methodology that converged precisely on the animal-based EFSA value, adding substantial confidence to the regulatory threshold.2 EU Regulation 1334/2008 (Annex III, in force from January 2011) established maximum permitted coumarin concentrations by food category: 50 mg/kg in traditional and/or seasonal bakery ware containing a reference to cinnamon in the labelling; 20 mg/kg in breakfast cereals including muesli; 15 mg/kg in fine bakery ware; and 5 mg/kg in desserts. In the United States, FDA regulations under 21 CFR 189.130 prohibit the addition of coumarin as such, including via tonka bean or tonka extract, but this is distinct from naturally occurring coumarin present in spices such as cinnamon. The BfR specifically flagged children as a high-concern group: their lower body weight means a smaller absolute TDI, while their proportional cinnamon consumption is often higher than adults'.

“ EFSA established the TDI at 0.1 mg/kg body weight per day; BfR has reiterated concern consistent with that threshold in its cinnamon guidance; and Abraham et al. (2010) reported a convergent estimate of 0.1 mg/kg/day derived independently from human clinical data — a notable degree of cross-methodology agreement for a dietary compound. ”

05 · Exposure Data & Species Comparison

Regulatory limits acquire practical meaning only when set against real dietary exposure. Abraham et al.'s survey of over 1,000 individuals in Germany during peak cinnamon consumption found that heavy cassia consumers could reach daily intakes at or exceeding the TDI from food alone.2 A Norwegian risk assessment (Steinhoff et al., 2012) calculated that a 30 kg child consuming oatmeal with cassia cinnamon several times per week could reach a coumarin intake of 1.63 mg/kg/day — more than 16 times the EFSA TDI.8

Whether these exposures are clinically significant depends largely on species identity — because the four commercial Cinnamomum species differ by several orders of magnitude in coumarin content, as documented across multiple independent analytical studies.

Blahová & Svobodová's 60-sample retail analysis found cassia coumarin ranging from 700 to over 7,000 mg/kg; the authenticated Sri Lankan C. verum sample in the same study registered below the analytical detection limit.9 For repeated-use formulations, the exposure difference between species can be substantial. Oketch-Rabah, Marles, and Brinckmann (2018) in Clinical Pharmacology & Therapeutics documented that multiple cassia species are routinely labelled as "cinnamon" in commerce — making species identification impossible from a product label alone without independent analytical verification.10

COUMARIN CONTENT BY SPECIES — ANALYTICAL DATA & EXPOSURE CONTEXT

Sources: Blahová & Svobodová (2012), Steinhoff et al. (2012), Abraham et al. (2010), EFSA (2004, 2008). TDI = 0.1 mg coumarin/kg body weight/day. Ranges reflect published analytical data; values vary by origin and methodology.

EVIDENCE SYNTHESIS — INGREDIENT SCIENCE & RAW MATERIAL EVALUATION

Four conclusions from the coumarin literature for cinnamon raw material evaluation

1. Coumarin's hepatotoxic risk is real, mechanistically described in available data, and dose-dependent. Clinical data from therapeutic trials establishes a low but documented incidence of hepatotoxicity in a susceptible subpopulation, likely related to variability in CYP-mediated metabolism. The risk is not uniform across all individuals and is most relevant at intakes at or above the EFSA TDI of 0.1 mg/kg/day.

2. The TDI of 0. 1 mg/kg/day is a well-established and independently corroborated exposure limit in food toxicology. It was independently derived from animal data (EFSA, 2004), confirmed by human clinical data (Abraham et al., 2010), and reconfirmed by two regulatory authorities across three independent assessments spanning twenty years.

3. Population exposure surveys confirm the TDI is reachable from realistic cinnamon-containing dietary patterns. German and Norwegian data document that regular cassia consumers — and particularly children — can reach or exceed the EFSA TDI from food intake alone, before any botanical ingredient or cinnamon-containing formulation is added.

4. Published analytical literature consistently reports substantially lower coumarin concentrations in authenticated C. verum than in cassia-type species. This distinction follows from applying published concentration data to established regulatory thresholds. For formulators evaluating cinnamon raw materials for repeated-use food, nutraceutical, or botanical applications, species identification, origin authentication, and coumarin quantification are important variables in raw material evaluation and relevant considerations for responsible ingredient selection. For concentrated or repeated-use applications, species authentication and analytical coumarin testing are especially relevant. This content does not constitute medical advice and does not make claims that any ingredient or product diagnoses, treats, cures, or prevents disease.

SCIENTIFIC REFERENCES — PUBMED-LINKED WHERE AVAILABLE

[1]  Pitaro M, et al. (2022). Coumarin-Induced Hepatotoxicity: A Narrative Review. Molecules 27(24):9063. doi:10.3390/molecules27249063. Synthesis of clinical trial safety data and metabolic mechanisms. PubMed 36558195 · PMC9783661  PubMed · PMC

[2]  Abraham K, et al. (2010). Toxicology and risk assessment of coumarin: focus on human data. Mol Nutr Food Res 54(2):228–239. doi:10.1002/mnfr.200900281. TDI of 0.1 mg/kg/day derived independently from human clinical data; Christmas exposure survey. PubMed 20024932  PubMed

[3]  Yamada T, et al. (2006). Species-specific differences in coumarin-induced hepatotoxicity. Demonstrates molecular basis for lower human susceptibility vs rats; CYP pathway differences. PubMed 18480146  PubMed

[4]  Cox D, O'Kennedy R, Thornes RD. (1989). The rarity of liver toxicity in patients treated with coumarin. Human Toxicology 8(6):501–506. 2,173-patient trial; 0.37% elevated liver enzymes; characterised as idiosyncratic. PubMed 2591993  PubMed

[5]  Andréjak M, et al. (1998). French pharmacovigilance survey evaluating the hepatic toxicity of coumarin. 34 hepatic cases; 85% cytolytic; 3 severe cases; majority off-label. PubMed 15073959  PubMed

[6]  EFSA. (2004, re-evaluated 2008). Scientific Opinion on Coumarin. EFSA Journal 2(9):104. Establishes and confirms TDI of 0.1 mg/kg/day. efsa.onlinelibrary.wiley.com  efsa.onlinelibrary.wiley.com

[7]  BfR. (2006, reconfirmed 2012). Coumarin in Cinnamon. Health Assessment No. 043/2006. Confirms TDI; identifies children as high-risk group. bfr.bund.de  www.bfr.bund.de

[8]  Steinhoff B, et al. (2012). Risk assessment of coumarin using the benchmark dose approach. Food Chem Toxicol. Norwegian children eating oatmeal with cassia cinnamon calculated at 1.63 mg/kg/day — over 16× the EFSA TDI. sciencedirect.com  www.sciencedirect.com

[9]  Blahová J & Svobodová Z. (2012). Assessment of Coumarin Levels in Ground Cinnamon Available in the Czech Retail Market. Scientific World Journal 2012:263851. 60-sample analysis: cassia 700–7,000+ mg/kg; C. verum below detection limit. PubMed 22593682 · PMC3385612  PubMed · PMC

[10]  Oketch-Rabah HA, Marles RJ, Brinckmann JA. (2018). Cinnamon and Cassia Nomenclature Confusion: A Challenge to the Applicability of Clinical Data. Clin Pharmacol Ther 104(3):435–445. doi:10.1002/cpt.1162. Discusses nomenclature confusion and the implications for interpretation of clinical and commercial cinnamon data. PubMed 29947417  PubMed

Editorial Standards & Compliance: Part of the Ceylon Nutritionals Science & Education Series. No statements constitute health claims under EU Regulation 1924/2006, the U.S. DSHEA, or equivalent frameworks. References to species differences, regulatory limits, and analytical findings support scientific understanding and responsible raw material evaluation. This content does not constitute medical advice and does not make claims that any ingredient or product diagnoses, treats, cures, or prevents disease.