Grayanotoxin in Mad Honey: What the Research Says About Concentration & Safety

This page covers one part of the Cliff Mad Honey Safety Standard. For the full grayanotoxin concentration classification, see the grayanotoxin risk bands.

Companion references: the grayanotoxin glossary defines every term used here, and the mad honey fact sheet for journalists collects the citable figures with their sources. Our sources sets out which journals this site draws on and how it grades evidence.

Contents

1. What Is Grayanotoxin?
2. How Concentration Is Measured
3. What Published Data Shows About Concentration Ranges
4. The EFSA 2023 Risk Assessment: Key Findings
5. What the Research Does Not Yet Answer
6. Sources

What Is Grayanotoxin?

Mad honey is produced from the nectar of Rhododendron species, predominantly Rhododendron ponticum and Rhododendron luteum in the Black Sea region of Turkey, and various Rhododendron species across Nepal, South Korea, and parts of China. What makes it pharmacologically distinct from conventional honey is its natural content of grayanotoxins (GTX), a class of diterpenoid compounds that act on voltage-gated sodium channels in cell membranes.

GTX binds to sodium channels in their open position, preventing normal closure. This causes sustained membrane depolarisation in cardiac muscle, nerve tissue, and skeletal muscle. The result is the characteristic clinical presentation of mad honey intoxication: slowed heart rate, lowered blood pressure, and in higher-exposure cases, atrioventricular block, syncope, and altered consciousness.

More than 20 structural variants of grayanotoxins have been identified in Rhododendron species. However, three variants (GTX I, GTX II, and GTX III) account for the majority of documented human toxicity and are the only variants for which validated laboratory detection methods are widely established.

GTX I

The most pharmacologically potent and extensively studied variant. In honey consumed by hospitalised patients in Turkey, the mean GTX I concentration was 8.73 mg/kg.

GTX III

The most frequently detected variant in commercial honey samples. Documented ranges in commercial mad honey reach as high as 68.754 mg/kg GTX III alone. In the same Turkish hospitalisation case series, mean GTX III was 27.60 mg/kg.

GTX II

Present in some samples and co-occurs with GTX I, but is less well-characterised in terms of relative human potency. EFSA’s 2023 risk assessment uses the sum of GTX I and GTX III as the primary exposure metric, the convention used across most current peer-reviewed literature.

How Concentration Is Measured

The standard analytical method for GTX quantification in honey is liquid chromatography with tandem mass spectrometry (LC-MS/MS); grayanotoxins cannot be detected by UV absorbance because they have no chromophore (Ahn et al. 2022). Results are expressed in mg/kg, milligrams of GTX per kilogram of honey, numerically equivalent to micrograms per gram (μg/g).

Published detection limits using optimised LC-MS/MS methodology:

• GTX I: limit of detection (LOD) 0.2 mg/kg; limit of quantification (LOQ) 0.6 mg/kg
• GTX III: LOD 0.1 mg/kg; LOQ 0.3 mg/kg

 

EFSA proposed an analytical target LOQ of ≤ 0.01 mg/kg for regulatory monitoring, considerably more sensitive than most current commercial testing. When a lab reports ‘not detected’ (ND) for GTX, this means the concentration fell below that laboratory’s specific detection threshold, not that the honey contains zero GTX.

What Published Data Shows About Concentration Ranges

Laboratory analyses of commercial mad honey products document a wide concentration spread. A 2015 study by Sahin et al. found GTX III concentrations ranging from 0.701 to 68.754 mg/kg across Black Sea Turkish samples. A 2018 study by Aygun et al. examining honey consumed by 25 hospitalised intoxication patients found a mean total GTX (I + III) of approximately 36 mg/kg. These figures allow a mapping of the commercial landscape into four ranges:

 

Trace: Below 1 mg/kg total GTX I + III

EFSA’s calculated protective concentration (0.05 mg/kg) sits within this range. No adverse effects have been reported in published case literature at this range under typical serving sizes. This range is primarily associated with unintentional contamination of conventional honey rather than intentionally sold mad honey products.

Low: 1–10 mg/kg total GTX I + III

Overlaps with the lower end of concentrations measured in honey associated with mild adverse effects in published case series, including dizziness, nausea, sweating, and hypersalivation. A 20g serving of honey at 10 mg/kg delivers approximately 0.2 mg GTX, a dose approaching clinically relevant exposure in individuals with cardiovascular disease or sensitivity factors.

Moderate: 10–30 mg/kg total GTX I + III

This range encompasses the mean GTX concentrations measured in honey consumed by patients presenting with confirmed mad honey intoxication in Turkish and Nepalese clinical studies. Documented adverse effects include bradycardia, hypotension, syncope, nausea, and vomiting. A 20 g serving at 30 mg/kg delivers 0.6 mg total GTX. Against EFSA’s reference point of 15.3 μg/kg body weight, that is roughly 8.6 μg/kg for a 70 kg adult and roughly 30 μg/kg for a 20 kg child. On a serving of this size, the reference point is crossed below a body weight of about 40 kg. A reference point is a benchmark for assessing risk, not a threshold below which a dose is safe.

High: Above 30 mg/kg total GTX I + III

Represents the upper documented range in commercial products. Case reports of severe mad honey intoxication requiring hospitalisation, atropine administration, and intravenous fluids are predominantly associated with honey in this range. Published cases at this exposure level include complete atrioventricular block and loss of consciousness. A 20g serving of honey at the upper documented commercial concentration (68.754 mg/kg) delivers approximately 1.4 mg total GTX.

The EFSA 2023 Risk Assessment: Key Findings

In 2023, the EFSA Panel on Contaminants in the Food Chain (CONTAM) published the most comprehensive independent risk assessment of grayanotoxins in honey to date. This is the primary regulatory reference document for GTX in food in Europe.

Reference Point: EFSA derived a benchmark dose lower confidence limit (BMDL10) of 15.3 μg/kg body weight for the sum of GTX I and III, based on reduced heart rate observed in rat studies.

Margin of Exposure: EFSA found that MOEs for realistic consumption scenarios with commercially available mad honey were below the level considered of low concern across all age groups. The gap between the dose associated with effects in animals and a realistic human dose from commercial mad honey is not wide enough to be reassuring.

Protective Concentration: EFSA calculated 0.05 mg/kg (sum of GTX I and III) as the highest concentration in honey it was at least 75% certain would protect all age groups; that figure is the panel’s expert judgement, not a statistical confidence level. This is a food safety reference point for regulatory purposes, not a ‘safe’ consumption threshold for deliberate use.

Data Gaps: EFSA explicitly noted insufficient data on chronic exposure, limited human pharmacokinetic data, and genotoxicity signals that were not fully characterised. The 2023 assessment covers acute toxicity only.

What the Research Does Not Yet Answer

The published literature as of 2026 does not provide answers to several clinically important questions:

• Individual variation: Absorption, metabolism, and elimination of GTX vary with age, body weight, gut microbiome, genetic polymorphisms, and concurrent medications. No human pharmacokinetic study has characterised this variation across a representative population.
• Variants beyond GTX I, II, III: More than 20 grayanotoxin variants have been identified in Rhododendron species. Most are not captured by routine LC-MS/MS methods.
• Chronic low-dose exposure: All published clinical data are from acute exposure events. No peer-reviewed human study has examined repeated small-dose GTX exposure over weeks or months.
• Interaction with food matrix: How honey’s sugars and acids affect GTX absorption in humans has not been characterised in clinical studies.

Sources

1. Sahin H, Turumtay EA, Yildiz O, Kolayli S. (2015). Grayanotoxin-III detection and antioxidant activity of mad honey. International Journal of Food Properties, 18(12), 2665–2674.  https://doi.org/10.1080/10942912.2014.999866
2. EFSA CONTAM Panel (2023). Risks for human health related to the presence of grayanotoxins in certain honey. EFSA Journal, 21(3), e7866.  https://doi.org/10.2903/j.efsa.2023.7866
3. Demir H, Denizbasi A, Onur O. (2011). Mad honey intoxication: a case series of 21 patients. ISRN Toxicology.  https://pmc.ncbi.nlm.nih.gov/articles/PMC3658790/
4. Eroğlu SE, Urgan O, Onur OE, et al. (2013). Grayanotoxin (mad honey): ongoing consumption after poisoning. Balkan Medical Journal.  https://pmc.ncbi.nlm.nih.gov/articles/PMC4115918/
5. Jansen SA, et al. (2012). Grayanotoxin poisoning: ‘mad honey disease’ and beyond. Cardiovascular Toxicology, 12(3), 208–215.  https://pmc.ncbi.nlm.nih.gov/articles/PMC3404272/
6. Aryal M. (2025). Grayanotoxins in mad honey: mechanisms of toxicity, clinical management, and therapeutic implications. Journal of Applied Toxicology, 45(11), 2240–2256.  https://doi.org/10.1002/jat.4855
7. Aygun A, Sahin A, Karaca Y, et al. (2018). Grayanotoxin levels in blood, urine and honey and their association with clinical status in patients with mad honey intoxication. Turkish Journal of Emergency Medicine, 18(1), 29–33.  https://www.sciencedirect.com/science/article/pii/S2452247317300584

 

NEXT READS

1. Mad Honey Concentration Bands: What the Numbers on a Lab Report Actually Mean
2. Who Should Not Consume Mad Honey: Contraindications Based on Clinical Evidence