Eating It And Breathing It Are Not The Same Exposure

Swallowed compounds are absorbed through the gut and travel via the portal vein to the liver, which metabolises a proportion before they ever reach general circulation. This is first-pass metabolism, and for many compounds it removes a substantial fraction of the dose.

Inhaled compounds cross the alveolar membrane directly into arterial blood. The liver gets its turn later, after the compound has already circulated. California's regulator notes this directly: inhaled substances "can directly enter the bloodstream from the lungs".

For a compound the liver clears efficiently, the two routes can produce very different internal exposures from the same nominal dose.

The agency's guidance on deriving inhalation reference concentrations is explicit. Oral data "should not be used for route-to-route extrapolation" in several circumstances, including "when a first-pass effect by the liver is expected".

It also notes that metabolite profiles differ by route: "the products of respiratory metabolism may have different toxicities from those of hepatic metabolism", and the tissues at risk differ accordingly.

Note what this is and isn't. It is a prohibition on extrapolating. It is not a statement that inhalation is uniformly more dangerous, and the agency offers no multiplier.

Smaller than rhetoric usually suggests, and less predictable in direction.

A 2022 framework for assessing inhaled products applied uncertainty factors of 2 for toxicokinetics and 3 for toxicodynamics when moving from oral data — a combined factor of about 6.

A 2004 critical assessment in *Regulatory Toxicology and Pharmacology* is more sobering. Comparing predicted with measured inhalation no-adverse-effect levels, the authors found ratios spanning 0.03 to 326 — errors of two orders of magnitude in both directions. Their conclusion was that extrapolation based on absorption correction is "not generally reliable".

There are documented cases where inhalation absorbs *less*: styrene is about 70 per cent absorbed by inhalation against nearly 100 per cent orally.

So the honest position is not "inhaled is worse by some factor". It is that oral data does not reliably predict inhalation risk in either direction, which is precisely why regulators require inhalation studies rather than extrapolating.

For the compounds under discussion, those inhalation studies have not been done.

The US EPA's risk assessment for one widely used growth regulator records that no short-term inhalation study is available, and route-extrapolates from oral endpoints instead. That is a documented methodological necessity, not a finding of safety — the agency is explicit that the data do not exist.

Pesticide registration has never required inhalation or combustion testing for compounds not intended to be inhaled. That is a perfectly sensible rule which produces an unreasonable result when the treated material is later burned.

Washington's regulator adds a caution for anyone reaching for tobacco data as a surrogate: users of inhaled plant products "inhale larger volumes of smoke and hold inhalations for longer in the lungs, which greatly limits comparisons to tobacco users".

It does not mean inhaled residues are proven harmful. No study has demonstrated harm in humans from inhaling agricultural residues on plant material, because no such study has been conducted. The peer-reviewed review of this field concluded in 2020 that "currently no research exists on the toxicity of pyrolysed pesticides in humans", and that statement remains accurate.

What it means is that the numbers most people assume are protective were derived for a different route, and the bridge between them has never been built.

There is a reasonable position available that requires overstating nothing. If a residue's safety by inhalation is genuinely unestablished — not disputed, not marginal, unestablished — then the only quantity whose inhalation toxicology need not be known is zero.