Identification of newly formed toxic chemicals in E-cigarette aerosols with Orbitrap mass spectrometry and implications on E-cigarette control

Abstract

The increasing use of electronic nicotine delivery systems (ENDS) is of concern due to multiple emerging adverse health effects. Most analyses of the harmful chemicals of ENDS have targeted metals or carbonyls generated by thermal decomposition of carrier liquids such as propylene glycol. However, new complex compounds not routinely identified and with unknown health consequences could be formed. ENDS aerosol samples were collected by the direct aerosol droplet deposition method. Untargeted analysis was performed using Orbitrap mass spectrometry with high mass accuracy. We identified more than 30 “features” in the aerosol characterized by pairs of the mass-to-charge ratio “ m/ z” of the compound and the retention time. We identified several compounds containing nicotine and propylene glycol (NIC-PG), whose abundance relative to nicotine increased along with vaping power used. On the basis of the prediction by the Environmental Protection Agency Toxicity Estimation Software Tool, these compounds exert developmental toxicity. In addition, a nitrogen-containing compound, likely tributylamine (a known lung irritant), was identified based on the molecular weight. This compound has not been previously identified in ENDS e-liquids and aerosols. ENDS produce not only small toxic compounds such as aldehydes, but also large complex toxic compounds such as NIC-PG. Predicted development toxicity for NIC-PG is concerning for fetal development in pregnant women who use ENDS, children exposed to secondhand or thirdhand ENDS aerosols, and teenage ENDS users whose brains are still developing. The strong positive association between NIC-PG levels and ENDS power output supports regulating high-powered ENDS.

Critical Review of the Recent Literature on Organic Byproducts in E-Cigarette Aerosol Emissions

“…As we show in Sections 6, 7 and 8 and summarized in Section 11, at least half (18 out of 35) of laboratory emission studies have tested sub-ohm devices with CORESTA or CORESTA-like protocols. Out of these 18 studies, 10 tested the devices also at too high powers levels [13,17,22,24,26,31,32,35,36,41], hence under almost certain overheating, which renders their outcomes and risk assessments from sub-ohm devices completely irrelevant for end users. The remaining 8 studies tested the devices at recommended power levels [16,18,20,23,24,30,[37][38][39], hence overheating is possible but not certain.…”

“…Most revised studies present assessments of health risks based on their experimental quantification of the abundance of potentially toxic byproducts, an abundance that must be compared with compound-specific toxicological markers. A substantial proportion of the risk assessments express serious concerns, some cast doubts on the safety of the devices and even recommend harsh regulatory restrictions [17,26,31,32].…”

“…• questionable and unrealistic for all users if coming from the outcomes of 10 studies that used high power settings with these protocols to test sub-ohm devices [17,22,24,26,31,32,35,36,41]. The 10 studies are marked with red traffic light G in Tables 3 and 4.…”

“…Yan et al The authors [31] apply orbitrap MS for a nontargeted analysis of EC emissions. They identify more than 30 "features" characterized by pairs of the mass to-charge ratio of the compound and the retention time.…”

Emerging ENDS products and challenges in tobacco control toxicity research

“…In addition to the relatively well-characterised ingredients of pure e-liquids (PG, vegetable glycerine (VG), nicotine, nicotine acidifiers and flavourants), a wide range of chemical reaction products are present in the aerosol, both as a result of slow reactions during storage (eg, Maillard reaction and acetal formation) and of acute production at the heating coil during use 59 60. The heating coil material and temperature can catalyse the breakdown of PG and VG, and can itself shed toxic metals into the aerosol 61–63.…”