Chemical analysis of fresh and aged Australian e‐cigarette liquids

In reply: We thank Morgan and colleagues1 for their interest in our recent publication2 and the robust discussion it has generated. We would like to respond with some clarifying information.

Firstly, calibration curves were generated for each individual compound. All had an R² between 0.95 and 0.99. Samples were prepared to ensure that their final concentration fell within the linear threshold of the calibration as is standard practice. Samples were measured at multiple dilutions, ensuring that each sample was run at least twice.

We acknowledge that it is not possible to directly relate inhalation LC₅₀ (the median lethal concentrations that kill 50% of a test animal population) to a concentration in liquid without assumptions, hence, our multiple statements to that effect in the “Health effects of inhaled e‐liquid aerosols” section. In our opinion, these qualifying sentences are more than sufficient to indicate that our single statement on LC₅₀ should be interpreted in the context intended. While not cited in the manuscript, LC₅₀ data for benzaldehyde, menthol, 2‐chlorophenol and benzyl alcohol are available online.3

The authors estimate potential exposures assuming a 55 mL puff volume and 9.47 mg of e‐liquid consumed per puff.4 While this puff volume is reasonably standard, the 9.47 mg per puff is device‐ and setting‐specific. Indeed, this consumption is specifically for a Joyetech Cubis tank (1Ω, ~15W, SS316L coil, 3‐second puff).4 Changing any of these alters the mass of e‐liquid vaporised, substantially increasing it with sub‐ohm/high power devices.4 The authors use these assumed values to calculate concentrations of certain chemicals based on derived no‐effect levels (DNELs). Unfortunately, they have used worker DNELs (considerably higher than general population DNELs) and a menthol DNEL estimated from oral exposure studies. The utility of DNELs can be questioned; especially, where a legislated limit exists (eg, the German indoor air quality limit for benzaldehyde is 0.2 mg/m³).

Clearly, many factors have an impact on whether an individual will be exposed to unsafe levels of particular chemicals when vaping. Usage patterns vary widely. Further, e‐cigarette technologies and inhalation toxicity data are rapidly evolving. We have presented data on what chemicals were found in some Australian e‐liquids, without interpretation on whether these would be toxic to a specific user. We strongly advocate for further research in this area.