James H. Stewart

Flavoring Chemicals in E-Cigarettes: Diacetyl, 2,3-Pentanedione, and Acetoin in a Sample of 51 Products, Including Fruit-, Candy-, and Cocktail-Flavored E-Cigarettes

Background: There are > 7,000 e-cigarette flavors currently marketed. Flavoring chemicals gained notoriety in the early 2000s when inhalation exposure of the flavoring chemical diacetyl was found to be associated with a disease that became known as “popcorn lung.” There has been limited research on flavoring chemicals in e-cigarettes. Objective: We aimed to determine if the flavoring chemical diacetyl and two other high-priority flavoring chemicals, 2,3-pentanedione and acetoin, are present in a convenience sample of flavored e-cigarettes. Methods: We selected 51 types of flavored e-cigarettes sold by leading e-cigarette brands and flavors we deemed were appealing to youth. E-cigarette contents were fully discharged and the air stream was captured and analyzed for total mass of diacetyl, 2,3-pentanedione, and acetoin, according to OSHA method 1012. Results: At least one flavoring chemical was detected in 47 of 51 unique flavors tested. Diacetyl was detected above the laboratory limit of detection in 39 of the 51 flavors tested, ranging from below the limit of quantification to 239 μg/e-cigarette. 2,3-Pentanedione and acetoin were detected in 23 and 46 of the 51 flavors tested at concentrations up to 64 and 529 μg/e-cigarette, respectively. Conclusion: Because of the associations between diacetyl and bronchiolitis obliterans and other severe respiratory diseases observed in workers, urgent action is recommended to further evaluate this potentially widespread exposure via flavored e-cigarettes. Citation: Allen JG, Flanigan SS, LeBlanc M, Vallarino J, MacNaughton P, Stewart JH, Christiani DC. 2016. Flavoring chemicals in e-cigarettes: diacetyl, 2,3-pentanedione, and acetoin in a sample of 51 products, including fruit-, candy-, and cocktail-flavored e-cigarettes. Environ Health Perspect 124:733–739; http://dx.doi.org/10.1289/ehp.1510185

Response to “Comment on ‘Flavoring Chemicals in E-Cigarettes: Diacetyl, 2,3-Pentanedione, and Acetoin in a Sample of 51 Products, Including Fruit-, Candy-, and Cocktail-Flavored E-Cigarettes’”

We appreciate the opportunity to respond to the letter to the editor from Pierce et al. Nowhere do Pierce et al. identify any factual errors in our work; our key findings stand. We will, however, take this opportunity to address several of the points they raised. In their analysis comparing diacetyl in e-cigs to occupational exposure limits (OELs), Pierce et al. selectively chose to evaluate only the median (6.0 µg/e-cigarette for diacetyl and 1.6 µg/e-cigarette for 2,3-pentanedione) from our data to reinforce their point that exposures are below the OEL they derived (176 μg/day for diacetyl). In choosing only the median, Pierce et al. ignored that our study found, in a sample of only 51 of over 7,000 flavors, a flavored e-cigarette with a diacetyl concentration of 238 µg/e-cigarette, which exceeds the 176 µg/day OEL they calculated. There is additional support in the literature showing the potential to exceed their derived OEL, as well, including a paper cited by Pierce et al. Farsalinos et al. (2015) measured diacetyl directly in the liquid of e-cigarettes and then used this to estimate a daily dose (median = 6 μg/day; interquartile range: 26–278 μg/day). The 75th percentile concentration in Farsalinos et al. (2015) exceeds the daily dose limit of 176 µg/day that Pierce et al. used. Therefore, at least 25% of flavored e-cigarettes samples in the Farsalinos et al. (2015) paper would exceed the 176 µg/day limit derived by Pierce et al. We also want to reiterate our position, stated clearly in the discussion section of our paper along with our rationale, that the use of OELs for this population is inappropriate. Our position is in agreement with NIOSH, which published a response to the paper by Farsalinos et al. (2015) in which they stated that OELS are “…not intended to establish “safe” exposure concentrations for consumers or the general public” (Hubbs et al., 2015). Pierce et al. also misrepresent the findings of their own earlier work (Gaffney et al. 2015; Pierce et al. 2014; Pierce et al. 2015) in which diacetyl and 2,3-pentanedione levels were measured. In their letter they stated that “Over the past five years, we have published the results of several studies in which diacetyl and 2,3-pentanedione levels were measured in various consumer products.” As evident by the dates in the in-line citation, all 3 of their papers cited were published within 1.5 years; they do not have a 5-year record of publishing on this topic. Further, the ‘various consumer products’ include only 2 products: cigarettes and coffee. Additionally, in their letter they directly contradict the conclusions in their earlier work. In this letter they state “Gaffney et al. (2015) and Pierce et al. (2015) found that grinding, brewing, and consuming unflavored coffee was associated with airborne diacetyl concentrations that were several times higher than the NIOSH and ACGIH short-term (0.025 and 0.020 ppm, respectively) and 8-hour (0.005 and 0.010 ppm, respectively) OELs for diacetyl.” This is inaccurate and inconsistent with their own paper published in 2015. For workers brewing coffee, they actually state the exact opposite in the Pierce et al. (2015) paper: “None of the individual short-term (15 min) barista samples (maximum of 0.01 ppm) exceeded the proposed NIOSH or ACGIH STELs (0.025 ppm and 0.02 ppm, respectively).” And for customers consuming unflavored coffee, the maximum 8-hour exposure reported in their simulation (0.005 ppm; Table 2) was below the ACGIH 8-hour limit (0.010 ppm), and it was at the NIOSH recommended exposure limit (0.005 ppm), not above it, and certainly not “several times higher” than either limit (Pierce et al. 2015). We further note that Pierce et al. (2015) and Gaffney et al. (2014) appear not to have been peer reviewed, based on the short time between submission and publication (received, revised and accepted all in 3 and 1 days, respectively). Also, the 2 papers are on the same topic, were received by the same journal within 2 days of each other, and contain 6 identical and 12 nearly identical sentences, although only 1 discloses the funding source as being from two companies involved in diacetyl litigation. Furthermore, the exposure data reported in Pierce et al. (2015) were collected not in a coffee shop but in a small kitchen with a very low ventilation rate that we calculate to be well below the ASHRAE minimum ventilation rates for cafeterias/fast food dining of 19 ft3/min/person (ASHRAE 2013). Pierce et al. attempt to minimize risks by comparing flavored e-cigarettes and coffee beans, commenting, “Unless one assumes that unflavored coffee beans pose a serious risk of ‘popcorn lung,’ a rare and oftentimes lethal disease, then one should agree that exposures to airborne diketone levels above the NIOSH and ACGIH OELs are not necessarily indicative of respiratory risk.” They do not seem to be aware that, in fact, several workers at a coffee processing workplace were recently diagnosed with bronchiolitis obliterans (“popcorn lung”), and NIOSH’s investigation found a 2.7-fold elevated standard mortality ratio for obstruction for workers at this site (Bailey et al. 2015). The NIOSH investigation concluded that, “The exposure group working in both coffee flavoring and grinding/packaging of unflavored coffee areas had significantly lower mean ratio of forced expiratory volume in 1 s to forced vital capacity and percent predicted mid-expiratory flow than workers without such exposure,” and “Current workers have occupational lung morbidity associated with high diacetyl and 2,3-pentanedione exposures, which were not limited to flavoring areas.” Finally, Pierce et al.’s comparison of diketone exposures from e-cigarettes and cigarettes, and their assertion of an increase in diketone exposure by not switching to e-cigarettes, misses a major point. Nearly 2 million children have tried e-cigarettes, 160,000 of whom reported that they had not used cigarettes (CDC 2013). Pierce et al. stated, “Ironically, suggesting that diketone levels in e-cigarettes are potentially dangerous could actually lead to higher diketone exposures in the smoking population if smokers decide not to switch to e-cigarettes due to as yet unfounded health concerns.” What about the 160,000 children who tried e-cigarettes who had not used cigarettes? We see no irony. In conclusion, we stand by our work and the facts presented in our paper: diacetyl and other flavoring chemicals are in many flavored e-cigarettes, including flavors, like cupcake and cotton candy, that we deem are particularly appealing to kids. Considering the history of severe and irreversible lung disease associated with some workers who inhaled diacetyl, and the similar exposure pathways for consumers of flavored e-cigarettes, it is prudent to evaluate this potential hazard further, restrict access by youth, and provide consumers with information and warnings similar to those given to workers.

Mortality among semiconductor and storage device-manufacturing workers.

Problem: We evaluated mortality during 1965 to 1999 among 126,836 workers at two semiconductor facilities and one storage device facility. Method: We compared employees’ cause-specific mortality rates with general population rates and examined mortality patterns by facility, duration of employment, time since first employment, and work activity. Results: Employees had lower-than-expected mortality overall (6579 observed deaths, standardized mortality ratio [SMR] = 65; 95% confidence interval [CI] = 64–67), for all cancers combined (2159 observed, SMR = 78, 95% CI = 75–81) and for other major diseases. Central nervous system cancer was associated with process equipment maintenance at one of the semiconductor facilities (10 observed, SMR = 247, 95% CI = 118–454). Prostate cancer was associated with facilities/laboratories at the storage device facility (18 observed, SMR = 198, (5% CI = 117–313). Conclusions: Further evaluation of workplace exposures or independent investigations of similar occupational groups may clarify the interpretation of associations observed in this study

Cancer incidence among semiconductor and electronic storage device workers

Aims: To evaluate cancer incidence among workers at two facilities in the USA that made semiconductors and electronic storage devices. Methods: 89 054 men and women employed by International Business Machines (IBM) were included in the study. We compared employees’ incidence rates with general population rates and examined incidence patterns by facility, duration of employment, time since first employment, manufacturing era, potential for exposure to workplace environments other than offices and work activity. Results: For employees at the semiconductor manufacturing facility, the standardised incidence ratio (SIR) for all cancers combined was 81 (1541 observed cases, 95% confidence interval (CI) 77 to 85) and for those at the storage device manufacturing facility the SIR was 87 (1319 observed cases, 95% CI 82 to 92). The subgroups of employees with ≥15 years since hiring and ≥5 years worked had 6–16% fewer total incidents than expected. SIRs were increased for several cancers in certain employee subgroups, but analyses of incidence patterns by potential exposure and by years spent and time since starting in specific work activities did not clearly indicate that the excesses were due to occupational exposure. Conclusions: This study did not provide strong or consistent evidence of causal associations with employment factors. Data on employees with long potential induction time and many years worked were limited. Further follow-up will allow a more informative analysis of cancer incidence that might be plausibly related to workplace exposures in the cohort.

Exposure assessment for retrospective follow-up studies of semiconductor- and storage device-manufacturing workers.

Objective: This exposure assessment was conducted in the first large study of mortality and cancer incidence in semiconductor and storage device manufacturing. Methods: Unique combinations of division, department and job codes and names (DDJ) from work history records were assigned to work groups and exposure categories. Agent exposure matrices assessed differences in potential exposures between groups. Changes in exposure over time were tracked by dividing the production history into manufacturing eras. Results: Nineteen work groups were developed to capture 310,351 unique DDJs from 1965–1999. Agent exposure matrices contrasted exposure potential to solvents, metals, and work in cleanrooms between groups, and three manufacturing eras were identified for each site. Conclusions: The work groups, manufacturing eras and agent matrices have been used to classify workers in the study of cancer incidence and mortality.

Assessing exposure to granite countertops - Part 2: Radon.

Radon gas (222Rn) is a natural constituent of the environment and a risk factor for lung cancer that we are exposed to as a result of radioactive decay of radium (226Ra) in stone and soil. Granite countertops, in particular, have received recent media attention regarding their potential to emit radon. Radon flux was measured on 39 full slabs of granite from 27 different varieties to evaluate the potential for exposure and examine determinants of radon flux. Flux was measured at up to six pre-selected locations on each slab and also at areas identified as potentially enriched after a full-slab scan using a Geiger–Muller detector. Predicted indoor radon concentrations were estimated from the measured radon flux using the CONTAM indoor air quality model. Whole-slab average emissions ranged from less than limit of detection to 79.4 Bq/m2/h (median 3.9 Bq/m2/h), similar to the range reported in the literature for convenience samples of small granite pieces. Modeled indoor radon concentrations were less than the average outdoor radon concentration (14.8 Bq/m3; 0.4 pCi/l) and average indoor radon concentrations (48 Bq/m3; 1.3 pCi/l) found in the United States. Significant within-slab variability was observed for stones on the higher end of whole slab radon emissions, underscoring the limitations of drawing conclusions from discrete samples.

Airborne fungal spores in a cross-sectional study of office buildings.

Airborne fungal spores were measured in 44 office buildings in the summer and winter throughout the continental United States, as part of the Building Assessment, Survey and Evaluation (BASE) program. Six indoor air and two outdoor air samples were collected on a single day from each building. The cross-sectional and repeated measure design afforded evaluation of between-building and within-building variability of fungal spore levels in buildings. Total fungal spore concentrations in indoor air ranged from < 24 to 1000 spores/m 3 , except for one building with natural ventilation where indoor levels were approximately 9000 spores/m 3 . Indoor air concentrations of total spores did not vary significantly between winter and summer or morning and afternoon monitoring periods or among climate zones or locations within a test area. Indoor-outdoor ratios of total spore concentrations typically ranged between 0.01 and 0.1 and were approximately seven times greater in winter than summer because of relatively low outdoor levels in the winter. The indoor-outdoor ratio of total spore concentrations for a building was consistent (reliability coefficient = 0.91) among repeated measures. Distributions of rank correlation coefficients for spore types in pairs of individual indoor-outdoor and indoor-indoor samples were weakly correlated (Spearman correlation = 0.2 on average). When spore type data were aggregated among samples from the same building, the central tendency of the rank correlation coefficients increased to 0.45. Rank correlation coefficients were also proportional to the number of spore types present in the samples that were compared. The BASE study provides normative data on concentrations of fungal spores that can aid in identification of problematic levels of mold in buildings.

Review of PCBs in US schools: a brief history, an estimate of the number of impacted schools, and an approach for evaluating indoor air samples.

PCBs in building materials such as caulks and sealants are a largely unrecognized source of contamination in the building environment. Schools are of particular interest, as the period of extensive school construction (about 1950 to 1980) coincides with the time of greatest use of PCBs as plasticizers in building materials. In the USA, we estimate that the number of schools with PCB in building caulk ranges from 12,960 to 25,920 based upon the number of schools built in the time of PCB use and the proportion of buildings found to contain PCB caulk and sealants. Field and laboratory studies have demonstrated that PCBs from both interior and exterior caulking can be the source of elevated PCB air concentrations in these buildings, at levels that exceed health-based PCB exposure guidelines for building occupants. Air sampling in buildings containing PCB caulk has shown that the airborne PCB concentrations can be highly variable, even in repeat samples collected within a room. Sampling and data analysis strategies that recognize this variability can provide the basis for informed decision making about compliance with health-based exposure limits, even in cases where small numbers of samples are taken. The health risks posed by PCB exposures, particularly among children, mandate precautionary approaches to managing PCBs in building materials.