Mad Honey Exposure and Tolerability: What Clinical Case Studies Report

Where the Clinical Data Comes From

Most published clinical data on mad honey exposure comes from three primary sources:

• Turkish case series — Turkey’s Black Sea region is the primary source of commercially available mad honey globally. Multiple Turkish hospitals, particularly in Trabzon and Rize provinces, have published case series on intoxication presentations from the 1990s through the 2020s.
• Nepalese case reports — Nepal is the second major source, primarily involving Gurung honey hunters consuming honey from Apis laboriosa hives foraging on high-altitude Rhododendron species.
• International case reports — A growing number of cases from Europe, North America, and East Asia document intoxications from imported commercial mad honey, reflecting the product’s expanding international consumer market.

EFSA’s 2023 risk assessment synthesised published case data across all three sources as part of its hazard characterisation.

Quantities and Concentrations Associated with Reported Adverse Effects

Small Quantities, High Concentration

A 2013 study by Biberoglu et al. documented patients who presented with syncope and bradycardia after consuming as little as 10–30g of high-concentration Turkish mad honey (estimated above 30 mg/kg based on region and vendor analysis). At 30 mg/kg and 20g consumed, the total GTX dose is approximately 0.6 mg.

Larger Quantities, Moderate Concentration

Koca et al. (2015) documented cases where patients consumed 50–100g servings of moderate-concentration honey (estimated 10–20 mg/kg) and presented with bradycardia and hypotension. These cases reinforce that quantity amplifies exposure from lower-concentration products.

Repeated Consumption

Several published case reports describe individuals who had consumed the same mad honey product multiple times without adverse effects, then experienced intoxication on a subsequent occasion. Proposed explanations include batch-to-batch concentration variation, seasonal GTX fluctuation, and cumulative or threshold effects that are not yet mechanistically explained.

Mean Concentrations in Hospitalisation Cases

Yavuz et al. (2018) conducted the most systematic measurement of GTX in honey consumed by intoxication patients, analysing 23 samples. Mean GTX I was 8.73 mg/kg; mean GTX III was 27.60 mg/kg. This places the mean total GTX I + III consumed by hospitalised patients at approximately 36 mg/kg.

Onset and Duration of Reported Effects

Onset: Adverse effects most commonly begin 30 minutes to 3 hours after consumption. The fastest documented onset is approximately 20 minutes. In a minority of cases, onset has been reported after 4–6 hours, possibly related to slower gastric emptying or consumption with food.

Duration without treatment: In mild cases, symptoms typically resolve spontaneously within 4–12 hours as GTX is cleared from circulation.

Duration with treatment: Hospitalised patients receiving atropine (for bradycardia) and IV fluids (for hypotension) are typically discharged within 24 hours in published case series. No deaths have been attributed to mad honey intoxication alone in the peer-reviewed literature, though severe cases requiring temporary cardiac pacing have been reported.

The Reported Symptom Spectrum

Published case literature describes a spectrum from mild to severe. The following reflects what has been reported — not a predictive model for any individual case.

Mild — Most Frequently Reported in Lower-Exposure Cases

• Dizziness and lightheadedness
• Nausea and vomiting
• Diaphoresis (sweating)
• Hypersalivation
• Paraesthesia (tingling, particularly perioral or lip tingling)
• Blurred vision

Moderate — Associated with Confirmed Cardiovascular Involvement

• Bradycardia: heart rate below 60 bpm; cases as low as 30–40 bpm documented
• Hypotension: systolic blood pressure below 90 mmHg in published series
• Syncope (loss of consciousness, typically brief)
• Weakness and inability to stand

Severe — Less Frequent; Associated with High-Dose Exposure or Vulnerable Individuals

• Complete atrioventricular (AV) block
• Profound bradycardia requiring atropine administration
• Prolonged loss of consciousness
• In rare cases, temporary cardiac pacing

Individual Variation: Why Two People Can Respond Differently

Published clinical observations and GTX pharmacology both point to significant individual variation in response to the same exposure. Factors that appear relevant based on published data:

• Cardiovascular baseline health: Individuals with pre-existing bradycardia, sick sinus syndrome, or AV node dysfunction are at substantially elevated risk of severe outcomes. GTX acts on the same sodium channel pathway that already has reduced functional reserve in these conditions.
• Age: EFSA identified elderly individuals and children as higher-concern subgroups, noting that both populations may have reduced physiological reserve to compensate for GTX-induced cardiovascular depression.
• Body weight: The GTX dose expressed per kilogram body weight is the relevant exposure metric. A 60 kg adult and a 90 kg adult consuming the same quantity of honey receive meaningfully different mg/kg body weight doses.
• Concurrent medications: Beta-blockers, calcium channel blockers, and antiarrhythmic agents all affect cardiac conduction and can amplify GTX’s cardiovascular effects. See SS-06 for a detailed review.
• Fed vs. fasted state: Gastric emptying rate affects the speed and possibly extent of GTX absorption. This has not been systematically characterised in human studies.

What This Means for Interpreting ‘No Symptoms’ Reports

Consumer forums and some commercial sources cite anecdotal reports of individuals consuming mad honey without adverse effects. Based on the published clinical literature, several explanations are possible:

1. The product contained low or unverified GTX concentration.
2. The quantity consumed was small enough that the delivered dose fell below the individual’s response threshold.
3. The individual had characteristics — body weight, cardiovascular health, absence of relevant medications — that increased their tolerance margin.
4. Symptoms occurred but were mild enough not to be attributed to the honey.
5. Symptoms are delayed relative to the timeframe being reported.

No published research supports the conclusion that consistent symptom-free consumption at a given dose predicts continued symptom-free consumption. Batch-to-batch concentration variation — documented in the literature — means that the same product at a different time may have a substantially different GTX content.

Sources

1. Biberoglu S, et al. (2013). Mad honey poisoning.  https://pmc.ncbi.nlm.nih.gov/articles/PMC3658790/
2. Yavuz Y, et al. (2018). Grayanotoxin levels in blood, urine and honey and their association with clinical status in patients with mad honey intoxication. Toxicology Letters.  https://www.sciencedirect.com/science/article/pii/S2452247317300584
3. 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
4.  Koca I, et al. (2015). Grayanotoxin — ongoing consumption after poisoning.  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/

NEXT READS

1. Who Should Not Consume Mad Honey: Contraindications Based on Clinical Evidence
2. Mad Honey and Medication Interactions: What the Research Shows