Analytical and toxicological evaluation of flavor chemicals in electronic cigarette refill fluids

Abstract

Thousands of electronic cigarette refill fluids are commercially available. The concentrations of nicotine and the solvents, but not the flavor chemicals, are often disclosed on product labels. The purpose of this study was to identify and quantify flavor chemicals in 39 commercial refill fluids that were previously evaluated for toxicity. Twelve flavor chemicals were identified with concentrations ≥1 mg/ml: cinnamaldehyde, menthol, benzyl alcohol, vanillin, eugenol, p-anisaldehyde, ethyl cinnamate, maltol, ethyl maltol, triacetin, benzaldehyde, and menthone. Transfer of these flavor chemicals into aerosols made at 3V and 5V was efficient (mean transfer = 98%). We produced lab-made refill fluids containing authentic standards of each flavor chemical and analyzed the toxicity of their aerosols produced at 3V and 5V using a tank Box Mod device. Over 50% of the refill fluids in our sample contained high concentrations of flavor chemicals that transferred efficiently to aerosols at concentrations that produce cytotoxicity. When tested with two types of human lung cells, the aerosols made at 5V were generally more toxic than those made at 3V. These data will be valuable for consumers, physicians, public health officials, and regulatory agencies when discussing potential health concerns relating to flavor chemicals in electronic cigarette products.

Figure 1

Heat map of flavor chemicals identified and quantified in 45 commercial refill fluids. Chemicals (y-axis) are classified by toxicity and ordered within each toxicity bracket from highest to lowest lethal dose based on rat oral toxicity data. Refill fluids (x-axis) are represented by their inventory number and ranked with the left having the highest concentration of total flavor chemicals and the right having the lowest. The color gradient provides information on the concentration of each chemical. Red asterisks represent the 12 flavor chemicals that were ≥1 mg/ml in at least one product.

Figure 2

Flavor chemicals present at ≥1 mg/ml that were in <35% of the refill fluids. Duplicate bottles were not incorporated into the frequency calculation, but rather added to distinguish whether duplicate products were similar. (a) Ethyl cinnamate was in four products (10%). (b) Triacetin was in five products (12.8%). (c) Eugenol was in six products (15.4%). (d) Benzyl alcohol was in seven products (17.9%). (e) Maltol was in seven products (17.9%). (f) Benzaldehyde was in 12 products (30.75%). Dup = duplicate bottle purchased and screened.

Figure 3

Flavor chemicals present at ≥1 mg/ml that were in >35% of the 39 refill fluids. Duplicate bottles were not incorporated into the frequency calculation, but rather added to distinguish whether duplicate products were similar. (a) Menthone was in 16 products (41%). (b) p-Anisaldehyde was in 17 products (43.6%). (c) Menthol was in 17 products (43.6%). (d) Cinnamaldehyde was in 20 products (51.2%). (e) Vanillin was in 22 products (56.4%). (f) Ethyl maltol was in 31 products (79.5%). Dup = duplicate bottle purchased and screened.

Figure 4

Chemicals in lab-made refill fluids and their corresponding aerosols. (a) Cinnamaldehyde. (b) Menthol. (c) Benzyl alcohol. (d) Vanillin. (e) Eugenol. (f) p-Anisaldehyde. (g) Ethyl Cinnamate. (h) Maltol. (i) Ethyl maltol. (j) Triacetin. (k) Benzaldehyde. (l) Menthone. PG = propylene glycol. RF = refill fluid. 3V = 3V aerosol. 5V = 5V aerosol.

Figure 5

Dose-response curves for hydroxyacetone and 80% propylene glycol blank. (a) Hydroxyacetone diluted directly into culture medium for A549 cells and hPF. (b) A549 cells and hPF treated with 80% propylene glycol aerosolized at 3V and 5V and captured in culture medium. IC₅₀ represents the concentration that inhibited survival by 50% and is only indicated where applicable. Data are showing the means and their standard errors for three independent experiments. Red lines represent hPF dose-response curves. Blue lines represent A549 dose-response curves. Hash symbols (#) for 3V aerosols and asterisks (*) for 5V aerosols indicate the lowest concentration that is significantly different than the untreated controls. *p < 0.05, **p < 0.01.

Figure 6

Dose-response curves for A549 cells and hPF treated with aerosols produced from lab-made refill fluids containing the dominant flavor chemicals. The y-axis shows percent of survival in the MTT assay and is normalized to the untreated controls. The x-axis shows the mg of chemical per mL of culture medium. (a–l) represents each of the 12 dominant chemicals, prepared in refill fluid form, then collected as an aerosol into culture medium. IC₅₀ represents the inhibitory concentration at 50% and is only indicated where applicable. Data are showing the means and their standard errors for three independent experiments. Red lines represent hPF dose-response curves. Blue lines represent A549 dose-response curves. Hash symbols (#) for 3V aerosols and asterisks (*) for 5V aerosols indicate the lowest concentration that is significantly different from the untreated controls. *p < 0.05, **p < 0.01, ***p < 0.001.