We recently withdrew a batch of our FibrePlan product from sale due to a technical regulatory non-compliance.
Importantly, there was no identified risk to consumer health. However, the batch exceeded the legal limit for pyrrolizidine alkaloids (PAs) established per kilogram of botanical preparations, including extracts used in foods and food supplements.
The situation prompted me to reflect on a broader principle that lies at the heart of toxicology and risk assessment.
When I was a child, my grandmother made what I still consider the finest apricot jam I have ever tasted. During apricot season, the kitchen would be filled with the aroma of simmering fruit. Once the apricots had been processed, there remained a treasure that fascinated me even more than the jam itself: the kernels hidden inside the hard pits.
I spent countless hours cracking the dried apricot stones and eating the kernels within. At the time, I had no idea that these kernels contained amygdalin, a naturally occurring compound that can release cyanide when digested. Today, we know that consuming large quantities of apricot kernels can be dangerous and that sufficiently high intakes could potentially result in cyanide poisoning, particularly in children.
The reason I tell this story is simple: I never consumed a kilogram of apricot kernels in a single sitting. The fundamental principle of toxicology applied then, just as it does today:
The dose makes the poison.
This principle, first articulated by the sixteenth-century physician Paracelsus, remains the foundation of modern toxicology. Virtually every substance can become harmful if consumed in sufficient quantity, while many substances that are toxic at high doses may be harmless at low levels. Risk is determined not merely by the presence of a substance, but by the amount to which a person is exposed.
Even today, there are no restrictions preventing someone from purchasing large quantities of apricots, extracting the kernels, and consuming them. Nor do bags of apricots carry prominent warnings informing consumers that the kernels contain compounds capable of releasing cyanide. The risk is recognised, yet society implicitly assumes that people will consume these products in realistic quantities.
Apricot kernels are far from unique in this regard.
Consider nutmeg. Nutmeg contains myristicin and related compounds capable of causing serious adverse effects when consumed in excessive quantities. Symptoms of toxicity have been reported after intakes of approximately 20–50 grams. A kilogram would constitute a medical emergency. Yet packets of nutmeg on supermarket shelves do not carry prominent warnings about hallucinations, delirium, cardiovascular effects, or hospitalisation.
Black liquorice provides another example. Its active constituent, glycyrrhizin, can disrupt potassium balance and elevate blood pressure when consumed in large amounts. Excessive intake can lead to muscle weakness, hypertension, abnormal heart rhythms, and potentially serious cardiovascular complications. Nevertheless, black liquorice remains freely available and widely consumed.
Brazil nuts offer perhaps an even more striking illustration. They are exceptionally rich in selenium, an essential nutrient that becomes toxic when consumed in excess. One kilogram of Brazil nuts may contain between 15,000 and 20,000 micrograms of selenium, whereas the generally accepted adult upper safe intake level is approximately 400 micrograms per day. Excessive selenium intake can cause nausea, vomiting, hair loss, brittle nails, neurological symptoms, and nerve damage. Yet consumers are free to purchase and consume Brazil nuts without restriction.
Indeed, when considering foods capable of causing toxicity if consumed in sufficiently large quantities, the list includes:
-
Apricot kernels (cyanide)
-
Nutmeg (myristicin toxicity)
-
Brazil nuts (selenium toxicity)
-
Liver (vitamin A toxicity)
-
Improperly prepared bitter cassava (cyanide)
-
Black liquorice (glycyrrhizin toxicity)
What these examples demonstrate is that food safety assessments generally take realistic consumption patterns into account. Regulators understand that people do not typically consume a kilogram of nutmeg, liver, liquorice, or Brazil nuts in a single day, let alone repeatedly over time.
Yet this principle appears less straightforward when examining certain aspects of food supplement regulation.
A notable example is Commission Regulation (EU) 2023/915, as amended, which establishes maximum levels for various contaminants in foods, including pyrrolizidine alkaloids (PAs). Among its provisions is a maximum limit of 400 micrograms of PAs per kilogram for certain food supplements and products containing botanical ingredients such as psyllium and other plant-derived materials.
The objective of limiting exposure to potentially harmful compounds is entirely understandable and appropriate. However, questions arise when compliance assessments are based on theoretical consumption scenarios that bear little resemblance to actual consumer behaviour.
Unlike conventional foods, food supplements in unit-dose form—such as tablets and capsules—are not consumed by the kilogram. A typical psyllium capsule may weigh approximately 500 mg and contain only a fraction of a microgram of PAs. For example, a capsule containing 0.21 micrograms of PAs, taken at a recommended dosage of two to six capsules per day, would result in a daily intake of approximately 1.26 micrograms at most.
Viewed in terms of actual exposure, such an intake is extremely small. Yet the product may still be deemed non-compliant as the concentration exceeds the regulatory limit when mathematically extrapolated to one kilogram of product.
In practical terms, this means compliance is assessed on the basis of a hypothetical scenario in which an individual consumes approximately 2,000 capsules—the equivalent of one kilogram—in a single sitting or day. Under that assumption, total PA intake would exceed the prescribed concentration limit and the product would therefore be classified as non-compliant.
The contrast with ordinary foods is difficult to ignore. Society accepts that consuming a kilogram of Brazil nuts, apricot kernels, or nutmeg is unrealistic and therefore evaluates risk within the context of expected consumption patterns. Yet food supplements presented in unit-dose form, clearly labelled with recommended daily intakes, may be assessed against standards derived from equally unrealistic consumption assumptions.
This raises an important question regarding consistency in regulatory philosophy. Should risk assessments be based primarily on concentration per kilogram, irrespective of how a product is actually consumed? Or should greater emphasis be placed on realistic daily intake and actual consumer exposure?
Food safety regulation exists to protect public health, and that objective should never be compromised. Equally, the cornerstone of toxicology remains unchanged: exposure matters. The mere presence of a substance does not determine risk; the quantity consumed determines risk.
The lesson I unknowingly learned as a child eating apricot kernels in my grandmother's kitchen remains as relevant today as it was then. Toxicity is not simply about what is present in a food. It is about how much is consumed.
In the words of Paracelsus, nearly five hundred years ago:
“All things are poison, and nothing is without poison; only the dose makes a thing not a poison.”