Abstract
Background
After Massachusetts enacted regulation on flavored tobacco products in 2020, electronic cigarettes (e‐cigarettes) labeled “clear” emerged in the state. We studied their acute hemodynamic impact and chemical composition.
Methods
In participants (age 18–45 years) in the Cardiovascular Injury due to Tobacco Products 2.0 study (Boston, MA), we measured changes in blood pressure and heart rate following a structured use of participants' own e‐cigarettes. We compared “clear” e‐cigarette users (N=23) with non‐“clear” flavored e‐cigarette users (N=111) and nonusers (N=73). We used gas chromatograph coupled with mass spectroscopy to characterize and quantify the presence of synthetic cooling agents (WS‐3, WS‐23), menthol, and nicotine, and other flavorants in 19 “clear” e‐cigarettes.
Results
All “clear” e‐liquids contained the synthetic cooling agents WS‐23 and/or WS‐3, 18/19 contained menthol, and 12/19 products contained other flavorants. JUUL e‐liquids did not contain synthetic coolants. Baseline blood pressure and heart rate measures were similar between “clear” and non‐“clear” flavored e‐cigarette users. Following acute use, “clear” e‐cigarettes resulted in a greater increase in blood pressure and heart rate compared to both non‐“clear” flavored e‐cigarette use and nonuse, which remained largely consistent in multivariable models. For sensitivity analysis of comparing “clear” users to the subgroup of JUUL non‐“clear” flavored users, the directions of associations were similar.
Conclusions
The presence of synthetic cooling agents along with traditional flavorants in “clear” e‐cigarette products available in the Massachusetts marketplace undermines the efficacy of the flavor ban. The potential that “clear” e‐cigarettes induced more pronounced hemodynamic effects necessitates further study of the health impact of these products and synthetic cooling agents.
Registration
URL: https://www.medrxiv.org/content/10.1101/2024.04.18.24305863v1, medRxiv. 10.1101/2024.04.18.24305863.
Keywords: blood pressure, electronic cigarettes, tobacco flavorings, tobacco product regulation
Subject Categories: Vascular Disease
Nonstandard Abbreviations and Acronyms
- CDC
Centers for Disease Control and Prevention
- CITU 2.0 study
cardiovascular injury due to tobacco use 2.0 study
- HR
heart rate
- WS‐23
N,2,3‐trimethyl‐2‐propan‐2‐ylbutanamide
- WS‐3
N‐ethyl‐5‐methyl‐2‐propan‐2‐ylcyclohexane‐1‐carboxamide
Clinical Perspective.
What Is New?
Young adults who use e‐cigarettes frequently use flavored products, and flavor regulation has been proposed as a method to reduce youth uptake. Massachusetts implemented a ban on characterizing flavors in all tobacco products.
We observed the use of e‐cigarettes labeled as “clear” in a cohort of young adults following the flavor ban. Following acute use, there appeared to be a greater increase in heart rate and blood pressure with e‐cigarettes labeled as “clear” compared with nonuse and flavored e‐cigarette use.
Chemical evaluation demonstrated the presence of synthetic cooling agents in “clear” e‐liquids along with menthol and additional flavor compounds.
What Are the Clinical Implications?
These data suggest a greater hemodynamic effect of “clear” e‐cigarettes in young adults along with the presence of synthetic cooling agents and support additional studies investigating the cardiovascular health effects of these products and cooling agents.
Monitoring new products for constituents and consideration of synthetic coolants in regulation is needed to maximize the impact of flavor bans and protect adolescents and young adults.
Flavored electronic cigarettes (e‐cigarettes) appeal to youth and young adults. 1 , 2 Regulatory policies to restrict access to flavored e‐cigarettes is a strategy to reduce use. 3 , 4 , 5 , 6 Massachusetts was an early state to enact a statewide sales ban on all flavored tobacco products. 7 The law went into effect in June 2020 and prohibits any “person, retailer, or manufacturer from selling, distributing, causing to be sold or distributed, [or] offer for sale any flavored tobacco product or tobacco product flavor enhancer in any retail establishment, online, or through any other means to any consumer in the commonwealth.” 8 Legally, a flavored tobacco product is defined as “any tobacco product that contains a constituent that has or produces a characterizing flavor,” which is “a distinguishable taste or aroma other than the taste or aroma of tobacco, imparted or detectable before or during consumption of a tobacco product, including, but not limited to, a taste or aroma relating to any fruit, chocolate, vanilla, honey, candy, cocoa, dessert, alcoholic beverage, menthol, mint, wintergreen, herb or spice.” 8 The introduction of similar regulation banning the sales of flavored tobacco products in California in 2022 led to the emergence of “non‐menthol”‐labeled combustible cigarettes containing the odorless synthetic coolant WS‐3 (CAS No. 39711‐79‐0; N‐ethyl‐5‐methyl‐2‐propan‐2‐ylcyclohexane‐1‐carboxamide), which does not have taste or aroma but does produce cooling effects. These products were subsequently under litigation, 9 and California just amended its tobacco control bill to include “cooling sensation by an ordinary consumer” into the definition of characterizing flavors. 10 Synthetic coolants such as WS‐3 and WS‐23 (Cas No. 51115‐67‐4; N,2,3‐trimethyl‐2‐propan‐2‐ylbutanamide) activate the same transient receptor potential cation channel subfamily M (melastatin) member 8 cooling receptor as menthol and thus cause a cooling effect when inhaled or applied to skin yet lack menthol's characteristic “minty” odor. 11 Prior studies show the presence of synthetic coolants in e‐cigarettes labeled “ice” but not in JUUL e‐liquids. 11
After the Massachusetts flavor ban was enacted, e‐cigarettes labeled as “clear” became available on the Massachusetts marketplace. The term “clear” was introduced by the industry and has not been specifically defined or used by any regulatory agencies. The Centers for Disease Control and Prevention has documented the emergent sales of disposable e‐cigarette products identified as “clear/other cooling” in Massachusetts after the flavor ban. 12 Given the rapid changes in tobacco products sold in the market in response to the ban, we sought to analyze data collected throughout our CITU (Cardiovascular Injury due to Tobacco Use) 2.0 study, initiated before but carried out beyond the Massachusetts flavor ban, aimed at assessing the health effects of this newly emerged product class along with a detailed chemical evaluation of the constituents in “clear” e‐liquids.
METHODS
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Study Participants
The CITU 2.0 study was an observational cohort study evaluating the health effects of e‐cigarettes in young adults. 13 Study visits were held at the Boston, MA and Louisville, KY sites. Because the use of “clear” e‐cigarettes was only observed in Massachusetts, only data collected at this site were included in the current analysis. Data collected between April 2019 and May 2023 were assessed.
Self‐reported healthy individuals 18 to 45 years old without established cardiovascular risk factors or cardiovascular disease were included. Participants who identified as e‐cigarette users had to use e‐cigarettes at least 3 times per week and have started at least 6 months before enrollment. Among the e‐cigarette users, we defined 4 groups of concurrent combustible cigarette use over the past 30 days: none, rare (<8 cigarettes), infrequent (8–29 cigarettes), and daily (≥30 cigarettes). Prior studies have classified users as either sole e‐cigarette users or dual users. However, this largely reflects older adults who are former combustible users. 14 In our cohort we observed e‐cigarette users who are occasional combustible cigarette users, and those who were regular users. Thus, we created these intermediate categories. Nonuser controls had to have smoked <100 cigarettes in their lifetime and used other tobacco products <20 times. Exclusion criteria included a history of anemia, diabetes, inflammatory conditions, cancer, renal or hepatic diseases, as well as those taking aspirin, beta‐blockers, statins, or any cardiovascular medicine. Individuals with body weight <100 lbs or body mass index >40 kg/m2 were also excluded. The study was approved by Boston University Chobanian and Avedisian School of Medicine and University of Louisville institutional review board and all participants provided written informed consent.
Acute Use Protocol
Participants were instructed to fast from food and caffeine for 8 hours and abstain from tobacco and exercise for 6 hours before the visit. All visits were completed in the morning to limit circadian differences. Participants completed a questionnaire harmonized with the PhenX toolkit about lifetime tobacco use patterns including information regarding use of tobacco product type, flavor, and frequency. 15
We studied participants before and after use of their own e‐cigarette device or breathing through a straw for nonusers. The entire visit was carried out in a specialized smoking chamber. During the 10‐minute structured exposure session, participants were instructed to take 1 puff lasting 3 to 4 seconds every 30 seconds for 10 minutes, or as tolerated due to adverse effects of nicotine observed in some participants. Nonuse control participants mimicked the same conditions using a straw inhaling air instead of using an e‐cigarette. Information on e‐cigarette brand, model, and flavor was recorded at the time of acute use.
Assessment of Hemodynamics
Blood pressure (BP) and heart rate (HR) were recorded using an Omron upper arm BP device. BP and HR were measured in triplicate and averaged at baseline, and then again 10 minutes after the structured exposure session.
Chemical Analysis
A total of 19 disposable devices labeled as “clear” flavor were purchased from online retailers between April and May 2023 and shipped to Connecticut for chemical analysis following a Google keyword search for “Clear e‐cigarette.” Tested brands are listed in Table 1 and include “clear” devices from brands that participants used in the CITU sessions (Crave, Luto, Space). As a comparator, US JUUL pods (“Menthol” and “Virginia Tobacco”) were purchased at a local Connecticut gas station in March 2024 to confirm the absence of synthetic cooling agents, as reported by us in 2019. 16
Table 1.
Product Name, Quantified Contents of Nicotine, Menthol, and the Synthetic Coolants WS‐3 and WS‐23, and Other Flavorants Detected by Gas Chromatography Coupled With Mass Spectroscopy
| Product name | Nicotine (mg/g) | Menthol (mg/g) | WS‐3 (mg/g) | WS‐23 (mg/g) | Other flavorants detected by GC/MS* |
|---|---|---|---|---|---|
| “Clear” products | |||||
| Crave† clear 5% | 52.4 (0.7) | 2.7 (0.0) | – | 28.4 (0.6) | γ‐decalactone, ethyl maltol, vanillin |
| Crave† ultra clear 5% | 52.2 (0.9) | 6.6 (0.2) | 6.4 (0.2) | 27.4 (0.6) | – |
| Crave†+kik new clear 5% | 52.5 (0.6) | 6.8 (0.2) | 6.4 (0.1) | 28.5 (0.6) | – |
| Elfbar clear 5% | 39.1 (1.1) | 3.4 (0.2) | – | 30.4 (1.0) | γ‐decalactone, ethyl maltol, vanillin |
| Esco bars clear 5% | 53.0 (1.0) | 6.6 (0.3) | – | 19.3 (0.7) | Carvone |
| Fire float clear 0% | ‐ | ‐ | – | 18.1 (0.5) | Ethyl maltol |
| Flum pebble clear 5% | 47.1 (1.6) | 4.0 (0.2) | 13.3 (0.4) | 37.0 (14) | – |
| Flum float clear 5% | 50.2 (0.9) | 4.0 (0.1) | 25.2 (0.7) | 23.8 (1.1) | – |
| Fogger ultra clear 5% | 34.7 (0.7) | 4.5 (0.5) | – | 20.1 (0.7) | Benzyl alcohol, carvone, menthone |
| Hitt infinite clear 5% | 36.5 (0.5) | 1.6 (0.1) | – | 17.2 (0.5) | Menthone |
| Hotbox clear 5% | 40.5 (0.9) | 4.2 (0.1) | 13.6 (0.3) | 29.4 (1.0) | ‐ |
| iJoy bar clear 5% | 32.7 (0.3) | 1.4 (0.0) | – | 27.5 (0.4) | Carvone, γ‐decalactone, γ‐dodecalactone, ethyl maltol, isopropyl benzoate, menthone, vanillin, vanillin PG acetal |
| Luto† clear 2.8% | 28.4 (0.4) | 5.1 (0.2) | – | 15.3 (0.2) | Carvone |
| Luto† smooth clear 5% | 50.0 (0.5) | 0.5 (0.0) | 4.6 (0.1) | 34.4 (0.4) | ‐ |
| Pacha syn clear 5% | 51.2 (1.6) | 2.0 (0.4) | – | 26.5 (1.2) | ‐ |
| Podjuice Hyde IQ clear 5% | 50.3 (1.5) | 9.2 (1.0) | – | 13.8 (0.7) | Eucalyptol, isopulegol, menthone |
| Puffbar clear 5% | 36.5 (0.1) | 0.6 (0.1) | – | 15.6 (0.7) | γ‐decalactone, ethyl maltol, vanillin |
| Space† max pro 5% | 37.7 (0.2) | 0.8 (0.0) | 1.8 (0.0) | 11.2 (0.1) | Carvone, ethyl maltol, eucalyptol, menthone, vanillin |
| Vozol clear 5% | 37.9 (1.0) | 4.1 (0.1) | – | 18.7 (0.6) | Carvone, ethyl maltol, isopropyl benzoate, isopulegol, menthone, piperitone, vanillin |
| Non‐“clear” products | |||||
| Juul menthol 5% 2019 | 53.0 (0.1) | 10.0 (0.7) | – | – | – |
| Juul menthol 5% 2024 | 53.8 (1.8) | 10.6 (0.6) | – | – | – |
| Juul Virginia Tob. 5% 2024 | 48.3 (0.3) | – | – | – | β‐damascenone |
Results shown as average (SD); N=3. – means not detected. GC/MS indicates gas chromatograph coupled with mass spectrometry; WS‐3, N‐ethyl‐5‐methyl‐2‐propan‐2‐ylcyclohexane‐1‐carboxamide; and WS‐23, N,2,3‐trimethyl‐2‐propan‐2‐ylbutanamide.
As identified using the National Institute of Standards and Technology (NIST) database v.2.2; characteristic odors of reported compounds: benzyl alcohol: sweet, carvone: spearmint, γ‐decalactone: peach, γ‐dodecalactone: peach, ethyl maltol: cotton candy, eucalyptol: minty, isopropyl benzoate: floral, isopulegol: minty, menthone: minty, piperitone: minty, vanillin: vanilla, vanillin propylene glycol (PG) acetal: vanilla, β‐ damascenone: woody, herbal.
Brands that participants used in the CITU 2.0 (cardiovascular injury due to tobacco use 2.0) study.
All chemical analyses were completed at Yale University in New Haven, Connecticut. All devices/pods were opened and the e‐liquid was collected in a glass vial. To characterize and quantify select e‐liquid components, 40 to 60 mg of e‐liquid was diluted into 1 mL methanol (Fisher Scientific, Waltham, MA) containing 1 mg/mL of 1,4‐dioxane (Sigma‐Aldrich, St. Louis, MO) as internal standard, and injected into a gas chromatograph coupled with mass spectroscopy for characterization and a GC with flame ionization detection for quantification of selected compounds, respectively. Commercially available standards for nicotine (Alfa Aesar, Ward Hill, MA), menthol (Sigma‐Aldrich), WS‐3, and WS‐23 (both TCI, Portland, OR) were used to prepare calibration curves on the gas chromatograph with flame ionization detection with at least 5 calibration points.
Gas chromatograph coupled with mass spectroscopy was carried out on a Clarus 580 GC coupled with an SQ8S MS, fitted with an Elite‐5MS column (length 60 m, id 0.25 mm, 0.25 μm film; all PerkinElmer, Waltham, MA). The injection volume was 1 μL with a split ratio of 10 and an injector temperature of 300 °C. The program was 40 °C for 7 minutes; ramp 10 °C/minute to 50 °C then hold for 20 minutes; ramp 10 °C/minute to 310 °C then hold for 8 minutes; ramp 10 °C/minute to 325 °C then hold for 11.5 minutes. The MS was run in electron impact ionization (EI+) mode with a mass range of m/z 30 to 620.
Gas chromatograph with flame ionization detection was carried out on a GC‐2010 (Shimadzu, Columbia, MD) fitted with a J&W DB‐5 column (length 60 m, id 0.25 mm, 0.25 μm film; Agilent, Santa Clara, CA). The injection volume was 1 μL with a split ratio of 300 and an injector temperature of 250 °C. The program was 30 °C for 7 minutes; ramp 10 °C/min to 50 °C then hold for 20 minutes; ramp 10 °C/minute to 310 °C then hold for 12 minutes. The detector temperature was 325 °C.
Statistical Analysis
Participants were categorized according to the tobacco product flavor they used during the structured exposure session: “clear” e‐cigarette use, non‐“clear” flavored e‐cigarette use, or nonuse. For clinical characteristics, 1‐way ANOVA and χ2 testing were performed for continuous and categorical variables across the 3 groups. The baseline hemodynamic measures and the change in the hemodynamic measures were compared using 1‐way ANOVA with post‐hoc testing with least significant difference for between‐group comparisons. Multivariable regression models (adjusting for age, sex, intensity of combustible cigarette use, and number of puffs in acute use session) were conducted comparing the changes in the hemodynamic measure in “clear” compared with non‐“clear” flavored users (reference group). In a secondary analysis, we restricted non‐“clear” flavored e‐cigarette users only to JUUL users and compared them with “clear” e‐cigarette users in the fully adjusted multivariable model. Statistical analyses were performed using SPSSv.27 with 2‐sided P <0.05 considered as significant.
RESULTS
Chemical Characterization of Clear E‐Cigarettes
Chemical analysis of “clear” e‐liquids demonstrated the ubiquitous presence of synthetic cooling agent WS‐23 (detected in 19/19 samples) and WS‐3 (detected in 7/19 samples) at a range of 14 to 37 mg/g (1.4%–3.7%) and 1.8 to 25 mg/g (0.18%–2.5%), respectively (Table 1). Menthol was detected in 18/19 “clear” devices, ranging from 0.5 to 9.2 mg/g (0.05%–0.92%). 12/19 “clear” devices also contained other flavorants such as benzyl alcohol (sweet odor), carvone (spearmint), γ‐decalactone and γ‐dodecalactone (both peach, fruity), ethyl maltol (cotton candy), eucalyptol, isopulegol, menthone, piperitone (all minty), isopropyl benzoate (floral), and vanillin and its known reaction product with e‐liquid solvent vanillin PG acetal (both vanilla). 17 JUUL “Menthol” only contained 1 flavorant, menthol, at 10 mg/g (1%) without any synthetic coolants either in 2019 or in 2024 samples. JUUL “Virginia Tobacco” flavor contained the flavorant β‐damascenone (woody, herbal). Except for 1 “clear” product in which no nicotine was detected (consistent with label information), nicotine levels ranged from 28 to 53 mg/g (2.8%–5.3%) in “clear” devices, and 48 to 54 mg/g (4.8%–5.4%) in JUUL e‐liquids. Label information for many devices was reasonably accurate (Table 1), but deviations between label and measured nicotine content ranged from −35% to +5% in “clear” devices, and −3% to +8% in Juul.
Tobacco Use Characteristics
Among our study participants enrolled in the Massachusetts site, no “clear” use was reported before the flavor ban enactment, whereas post‐ban, 21% of participants used “clear” and 79% used non‐“clear” flavored products, consistent with ongoing access to flavored products. All “clear” e‐cigarettes used by CITU 2.0 participants were disposable devices. As shown in Table 2, “clear” and non‐“clear” e‐cigarette users enrolled in the study had similar ages, whereas nonusers were older. Among the non‐“clear” flavored e‐cigarette users, fruit and mint/menthol were the most commonly used flavor types with one third of participants using JUUL products. On average, both “clear” and non‐“clear” flavored e‐cigarette users used e‐cigarettes on almost every day in the past 30 days. The majority of e‐cigarette users had no current combustible cigarette use. Among the e‐cigarette users who had current combustible cigarette use, the total use in the past 30 days was low.
Table 2.
Participant Characteristics and Baseline Hemodynamic Measures
| “Clear” e‐cigarette users (N=23) | Flavored e‐cigarette users (N=111) | Nonusers (N=73) | P value* | |
|---|---|---|---|---|
| Participant characteristics | ||||
| Age (y) | 21±3 | 22±4 | 26±6 | 1.5×10−6 |
| Female sex | 10 (43) | 51 (46) | 39 (53) | 0.54 |
| Race | ||||
| White | 13 (57) | 53 (48) | 35 (48) | 0.39 |
| Asian | 6 (26) | 42 (38) | 31 (42) | |
| Black | 0 (0) | 7 (6) | 3 (4) | |
| Other¶ | 4 (17) | 9 (8) | 4 (5) | |
| Hispanic | 7 (30) | 17 (15) | 9 (12) | 0.11 |
| Flavor type used | ||||
| Fruit | – | 50 (45) | – | |
| Mint/menthol | – | 37 (33) | – | |
| Sweet | – | 13 (12) | – | |
| Tobacco | – | 11 (10) | – | |
| Juul product | 37 (33) | |||
| Days vaped, last 30 d | 27±6 | 27±6 | – | 0.97 |
| Freq of combustible cigarette† | ||||
| None (0 cigarettes in last 30 d) | 12 (52) | 70 (63) | – | 0.39 |
| Rare (<8 cigarettes in last 30 d) | 7 (30) | 17 (15) | – | |
| Infrequent (8–29 cigarettes in last 30 d) | 2 (9) | 11 (10) | – | |
| Daily (≥30 cigarettes in last 30 d) | 2 (9) | 13 (12) | – | |
| Reported number of combustible cigarettes in last 30 d (in current users) | 25±49 | 54±130 | – | 0.50 |
| Former combustible cigarette use‡ | 1 (8) | 14 (20) | – | 0.45 |
| Baseline hemodynamic measures | ||||
| Systolic blood pressure | 112±11 | 115±12 | 110±10 | 0.018 |
| Diastolic blood pressure | 69±5 | 71±7 | 68±8 | 0.089 |
| Mean arterial pressure | 83±6 | 85±8 | 82±8 | 0.024 |
| Heart rate | 62±11 | 62±8 | 63±10 | 0.96 |
Mean±SD, N (%) as appropriate.
By ANOVA, independent t test, or χ2 except as noted.
None=no use, Rare=<8 cigarettes, Infrequent=8–29 cigarettes, Daily=30 cigarettes in the last 30 days.
Among Sole Users, by Fisher exact test.
Per participant report.
Acute and Chronic Impact of “Clear” E‐Cigarette Use on Hemodynamic Measures
Overall, there were differences across the groups in baseline systolic BP and mean arterial pressure (Table 2). In post‐hoc testing, there were no baseline differences in any of the hemodynamic measures between “clear” users, non‐“clear” flavored users, or nonusers.
Following acute exposure, use of “clear” e‐cigarettes resulted in a greater increase in systolic BP, diastolic BP, mean arterial pressure, and HR compared with both non‐“clear” flavored e‐cigarette use and nonuse (Figure). In multivariable models among e‐cigarette users, the increase in systolic and diastolic BP and mean arterial pressure remained significantly higher in “clear” e‐cigarette users compared with non‐“clear” flavored e‐cigarette users with borderline significance for HR when adjusting for age, sex, frequency of combustible cigarette use, and number of puffs used in the acute use sessions (Table 3).
Figure 1. Acute effects of “clear” e‐cigarettes on systolic blood pressure (ΔSBP), diastolic blood pressure (ΔDBP), mean arterial pressure (ΔMAP), and heart rate (ΔHR).
Change in SBP, DBP, MAP, and HR differed across the 3 groups based on the product used: “clear” labeled e‐cigarettes (e‐cig), non‐“clear” flavored e‐cigarette (e‐cig), and nonuse by 1‐way ANOVA (all P <0.001), post‐hoc comparisons using least significant difference between “clear” e‐cig and other groups as indicated. DBP indicates diastolic blood pressure; HR, heart rate; MAP, mean arterial pressure; and SBP, systolic blood pressure.
Table 3.
Multivariable Adjusted Models Comparing Acute Hemodynamic Changes With “Clear” Compared With Non‐“Clear” Flavored E‐Cigarette Use
| Model | ΔSBP | ΔDBP | ΔMAP | ΔHR | ||||
|---|---|---|---|---|---|---|---|---|
| Β* (SE) | P value | Β (SE) | P value | Β (SE) | P value | Β (SE) | P value | |
| Age, sex | 3.2 (1.5) | 0.036 | 3.3 (1.4) | 0.020 | 3.2 (1.3) | 0.015 | 3.0 (1.6) | 0.061 |
| Age, sex, intensity combustible | 3.2 (1.5) | 0.036 | 3.2 (1.4) | 0.021 | 3.2 (1.3) | 0.016 | 3.0 (1.6) | 0.063 |
| Age, sex, intensity combustible, number puffs | 3.7 (1.7) | 0.030 | 3.6 (1.5) | 0.022 | 3.6 (1.5) | 0.015 | 3.6 (1.7) | 0.042 |
| JUUL† vs “Clear” | ||||||||
| Age, sex | 2.7 (1.9) | 0.16 | 1.8 (1.7) | 0.30 | 2.1 (1.6) | 0.21 | 3.2 (1.9) | 0.099 |
| Age, sex, intensity combustible | 2.7 (1.9) | 0.16 | 1.8 (1.7) | 0.30 | 2.1 (1.6) | 0.21 | 3.3 (1.9) | 0.090 |
| Age, sex, intensity combustible, number puffs | 4.5 (2.1) | 0.039 | 2.7 (2.0) | 0.18 | 3.3 (1.9) | 0.082 | 5.4 (2.1) | 0.014 |
B for comparison “Clear” to Flavored.
N=34.
DBP indicates diastolic blood pressure; HR, heart rate; MAP, mean arterial pressure; and SBP, systolic blood pressure.
As demonstrated in the chemical analysis, all sampled “clear” e‐liquids did contain synthetic coolants whereas JUUL samples did not (Table 1). Thus, we conducted a sensitivity analysis comparing “clear” e‐cigarette users to JUUL users. In the fully adjusted model, the directions of higher changes in hemodynamic changes with “clear” compared with JUUL persisted though some were rendered not statistically significant (Table 3).
DISCUSSION
In young adults enrolled in the Massachusetts site of the CITU 2.0 study, we captured the emergent use of e‐cigarettes labeled “clear” among participants in Massachusetts following a statewide flavor ban. Our study demonstrates that all tested “clear” e‐cigarettes, which sales data demonstrate have been sold locally following the statewide flavor ban, contained significant levels of the synthetic and odorless cooling agents WS‐23 and WS‐3. Some “clear” e‐cigarettes also contained the coolant menthol as well as additional sweet and minty flavorings. Among CITU 2.0 participants recruited in Massachussetts after the flavoring ban, 21% used products labeled “clear,” thereby offering an opportunity to evaluate the hemodynamic impact of these emerging products. Acute use of both “clear” and non‐“clear” flavored e‐cigarettes induced acute increases in hemodynamic measures compared with nonuse. Importantly, our study provides early evidence suggesting that acute use of “clear” e‐cigarettes had more pronounced hemodynamic effects when compared with non‐“clear” flavored e‐cigarette use. Taken together, these findings suggest important regulatory and health implications of the new emerging e‐cigarette product group labeled as “clear.”
The emerging use of the synthetic cooling agents WS‐23 and WS‐3 in tobacco products has been described previously in non‐“clear” flavored e‐cigarettes, sometimes labeled as “ice.” 11 In addition, WS‐3 has also been found in the “nonmenthol” combustible cigarettes, which were introduced into the California market after the enactment of statewide tobacco product flavor bans. 9 Both WS‐3 and WS‐23 provide a cooling sensation by the same pathway as menthol through activation of the transient receptor potential cation channel subfamily M (melastatin) member 8 ion channel. 18 We show that all tested e‐cigarette products labeled as “clear” contained these synthetic cooling compounds and thus can reasonably be expected to produce a cooling effect in users upon inhalation, which has previously been shown to reduce aversive oral effects of nicotine in rodents. 19 , 20 Thus, the use of the ambiguous flavor name “clear,” as well as supplementing these e‐liquids with synthetic cooling agents (with or without menthol), could be a strategy to continue to provide a cooling sensation or augment the cooling effects of menthol in e‐cigarettes even following flavor bans. Cooling chemicals such as menthol may reduce irritation from tobacco and nicotine and may be particularly appealing to those who use menthol cigarettes.
To complement the characterization of the constituents in “clear” e‐liquids, our human study evaluated the acute hemodynamic effects of “clear” e‐cigarettes. Evidence suggests that acute changes in BP and HR metrics are a sensitive marker of the cardiovascular impact of e‐cigarette use. 21 , 22 , 23 Consistent with prior studies, our findings indicate that compared with nonuse, e‐cigarette use induced changes in blood pressure and HR. 23 , 24 , 25 , 26 , 27 , 28 , 29 However, importantly, our findings suggest a more pronounced increase in BP and HR following use of “clear” versus. non‐“clear” e‐cigarettes. These differences do not appear to be attributable to ongoing use of combustible cigarettes by relevant participants. Given prior evidence linking the hemodynamic effects of e‐cigarette use to nicotine, it is possible that the presence of synthetic coolants in “clear” e‐cigarettes facilitated inhalation, in turn leading to higher puff volume and thus exposure to e‐liquid aerosol and nicotine.
The public health relevance of studying “clear” e‐cigarettes is supported by both the regulatory environment and the rapid evolution of tobacco products. Our observation of the emergence of “clear” e‐cigarette use in Massachussetts is consistent with sales data from the Centers for Disease Control and Prevention Foundation, 12 which showed a steady increase in sales of disposable e‐cigarettes that they characterize as “clear/other cooling” in Massachussetts in the time period after the flavor ban. It will be important across studies collecting both sales and epidemiology of e‐cigarette use to develop consistent terminology to refer to the group of products labeled as “clear.” Though manufacturers have not made consistent marketing claims, the “clear” label could be perceived as an unflavored product. Further, the detected presence of menthol alongside other flavorings in tested “clear” products means that their availability on the Massachussetts market is a direct violation of the Massachussetts flavored tobacco product regulation. Also, synthetic coolants, which are not specifically mentioned in the Massachussetts law and do not fit the traditional definition of a “characterizing flavor,” may reduce tobacco/nicotine harshness, facilitate deeper inhalation, and enhance nicotine delivery, which could help explain the hemodynamic findings and warrant their inclusion in tobacco flavor regulations. 30 Notably, other jurisdictions such as Germany, Belgium, Canada, and most recently California, have implemented different approaches to restricting agents that may produce a cooling effect and thereby deepen inhalation in combustible cigarettes. 11 , 16
Several study limitations exist: Chemical characterization was performed only for “clear” and JUUL e‐liquids and it is possible that other non‐“clear” flavored e‐liquids used by participants also contained synthetic coolants. However, the primary goal of the study was to characterize the products labeled as “clear.” Additionally, the subgroup analysis comparing those who used “clear” products with those who used JUUL products (which were shown to not contain synthetic coolants), though underpowered, suggests that products containing synthetic coolants may have a differential impact on cardiovascular metrics compared with products that do not contain these chemicals. Our findings highlight the need for further controlled studies to understand the impact of synthetic cooling agents in e‐liquids on the cardiovascular system. The number of “clear” e‐cigarette users was small but important, given the rapidly changing product marketplace. It will be important to confirm our findings in a larger prospective study evaluating “clear” products in comparison to other products. The study design based on participant's own products led to diversity in product design and components. Thus, we were not able to isolate the impact of a single product or manufacturer, and confounding by product type remained a possibility. However, the strength of the design was the ability to capture information about emerging products that can support future work regarding specific constituents, devices, and manufacturers. In addition, the inclusion of participants with concomitant use of combustible cigarettes could have impacted observed acute hemodynamics. However, participants' frequency and intensity of combustible cigarette use was low, reflecting the predominant use of e‐cigarettes in the tested cohort. Further, while models accounting for combustible product use supported differences in the hemodynamic effects of acute “clear” e‐cigarette use, we cannot fully exclude the possibility of residual impact of combustible product use.
In conclusion, following the implementation of a flavor ban in Massachusetts, we observed the emergent use by young adults of new products labeled as “clear.” These “clear” products contained the synthetic, odorless coolants WS‐3 and WS‐23, menthol, as well as other minty or fruity flavorings, which suggests a violation of the Massachussetts flavored tobacco product regulation. Following acute exposure, “clear” labeled e‐cigarettes produced greater hemodynamic changes than non‐“clear” flavored e‐cigarettes in participants, which has implications for cardiovascular health. Our findings suggest that monitoring for new products and constituents including scientific evidence regarding additional short‐ and long‐term health effects is critically needed to maximize the impact of flavor bans and protect adolescents and young adults.
Sources of Funding
This work was supported by cooperative agreement U54DA036151 (Yale Tobacco Center of Regulatory Science) from the National Institute on Drug Abuse (NIDA) of the National Institutes of Health (NIH) and the Center for Tobacco Products of the US Food and Drug Administration (FDA). Additionally, this work was supported by the National Heart, Lung, and Blood Institute (NHLBI) of the NIH (U54 HL120163) and the American Heart Association (20YVNR35500014).
Disclosures
No relevant disclosures. No tobacco industry related support.
This manuscript was sent to William W. Aitken, MD, Assistant Editor, for review by expert referees, editorial decision, and final disposition.
For Sources of Funding and Disclosures, see page 8.
References
- 1. Pepper JK, Ribisl KM, Brewer NT. Adolescents' interest in trying flavoured e‐cigarettes. Tob Control. 2016;25(Suppl 2):ii62–ii66. doi: 10.1136/tobaccocontrol-2016-053174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. FDA . Results from the Annual National Youth Tobacco Survey. Accessed March 19, 2025. https://www.fda.gov/tobacco‐products/youth‐and‐tobacco/results‐annual‐national‐youth‐tobacco‐survey.
- 3. Levy DT, Pearson JL, Villanti AC, Blackman K, Vallone DM, Niaura RS, Abrams DB. Modeling the future effects of a menthol ban on smoking prevalence and smoking‐attributable deaths in the United States. Am J Public Health. 2011;101:1236–1240. doi: 10.2105/AJPH.2011.300179 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. FDA . FDA's Youth Tobacco Prevention Plan|FDA. Accessed March 19, 2025. https://www.fda.gov/tobacco‐products/youth‐and‐tobacco/fdas‐youth‐tobacco‐prevention‐plan#.
- 5. Hawkins SS, Coley RL, Lanteri L, Baum CF. How adolescent tobacco use has responded to state tobacco 21 laws and flavor restrictions. Am J Drug Alcohol Abuse. 2025;12:1–12. doi: 10.1080/00952990.2024.2444567 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. FDA . FDA proposes rules prohibiting menthol cigarettes and flavored cigars to prevent youth initiation, significantly reduce tobacco‐related disease and death. Accessed March 19, 2025. https://www.fda.gov/news‐events/press‐announcements/fda‐proposes‐rules‐prohibiting‐menthol‐cigarettes‐and‐flavored‐cigars‐prevent‐youth‐initiation.
- 7. FDA . FDA finalizes enforcement policy on unauthorized flavored cartridge‐based e‐cigarettes that appeal to children, including fruit and mint|FDA. Accessed March 19, 2025. https://www.fda.gov/news‐events/press‐announcements/fda‐finalizes‐enforcement‐policy‐unauthorized‐flavored‐cartridge‐based‐e‐cigarettes‐appeal‐children.
- 8. 2019 Tobacco Control Law|Mass.gov. Accessed October 1, 2024. https://www.mass.gov/guides/2019‐tobacco‐control‐law.
- 9. Jabba SV, Erythropel HC, Anastas PT, Zimmerman JB, Jordt SE. Synthetic cooling agent and other flavor additives in “non‐menthol” cigarettes marketed in California and Massachusetts after menthol cigarette bans. JAMA. 2023;330:1689–1691. doi: 10.1001/jama.2023.17134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bill Text: CA AB3218|2023–2024|Regular Session|Chaptered|LegiScan. Accessed October 8, 2024. https://legiscan.com/CA/text/AB3218/2023.
- 11. Jabba SV, Erythropel HC, Torres DG, Delgrado LA, Woodrow JG, Anastas PT, Zimmerman JB, Jordt SE. Synthetic cooling agents in US‐marketed E‐cigarette refill liquids and popular disposable E‐cigarettes: chemical analysis and risk assessment. Nicotine Tob Res. 2022;24:1037–1046. doi: 10.1093/ntr/ntac046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. CDC Foundation . Monitoring E‐Cigarette Trends in the United States. Accessed March 19, 2025. https://tobaccomonitoring.org/wp‐content/uploads/2024/11/2024MonitoringE‐CigaretteTrendsUS‐1.pdf.
- 13. Keith RJ, Fetterman JL, Riggs DW, O'Toole T, Nystoriak JL, Holbrook M, Lorkiewicz P, Bhatnagar A, DeFilippis A, Hamburg NM. Protocol to assess the impact of tobacco‐induced volatile organic compounds on cardiovascular risk in a cross‐sectional cohort: cardiovascular injury due to tobacco use study. BMJ Open. 2018;8:e019850. doi: 10.1136/bmjopen-2017-019850 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kasza KA, Rivard C, Goniewicz ML, Fong GT, Hammond D, Cummings KM, Hyland A. E‐cigarette characteristics and cigarette cessation among adults who use E‐cigarettes. JAMA Netw Open. 2024;7:e2423960. doi: 10.1001/jamanetworkopen.2024.23960 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. PhenX Toolkit. Accessed October 1, 2024. https://www.phenxtoolkit.org/.
- 16. Erythropel HC, Anastas PT, Krishnan‐Sarin S, O'Malley SS, Jordt SE, Zimmerman JB. Differences in flavourant levels and synthetic coolant use between USA, EU and Canadian Juul products. Tob Control. 2021;30:453–455. doi: 10.1136/tobaccocontrol-2019-055500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Erythropel HC, Jabba SV, DeWinter TM, Mendizabal M, Anastas PT, Jordt SE, Zimmerman JB. Formation of flavorant–propylene glycol adducts with novel toxicological properties in chemically unstable E‐cigarette liquids. Nicotine Tob Res. 2019;21:1248–1258. doi: 10.1093/ntr/nty192 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Johnson S, Tian M, Sheldon G, Dowd E. Trigeminal receptor study of high‐intensity cooling agents. J Agric Food Chem. 2018;66:2319–2323. doi: 10.1021/acs.jafc.6b04838 [DOI] [PubMed] [Google Scholar]
- 19. Fan L, Balakrishna S, Jabba SV, Bonner PE, Taylor SR, Picciotto MR, Jordt SE. Menthol decreases oral nicotine aversion in C57BL/6 mice through a TRPM8‐dependent mechanism. Tob Control. 2016;25:ii50–ii54. doi: 10.1136/tobaccocontrol-2016-053209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Tackett AP, Han DH, Peraza N, Whaley RC, Mason T, Cahn R, Hong K, Pang R, Monterosso J, Page MK. Effects of ‘ice’ flavoured e‐cigarettes with synthetic cooling agent WS‐23 or menthol on user‐reported appeal and sensory attributes. Tob Control. 2023;34:1–8. doi: 10.1136/tc-2023-058125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Cooke WH, Pokhrel A, Dowling C, Fogt DL, Rickards CA. Acute inhalation of vaporized nicotine increases arterial pressure in young non‐smokers: a pilot study. Clin Auton Res. 2015;25:267–270. doi: 10.1007/s10286-015-0304-z [DOI] [PubMed] [Google Scholar]
- 22. Caporale A, Langham MC, Guo W, Johncola A, Chatterjee S, Wehrli FW. Acute effects of electronic cigarette aerosol inhalation on vascular function detected at quantitative MRI. Radiology. 2019;293:97–106. doi: 10.1148/radiol.2019190562 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Franzen KF, Willig J, Cayo Talavera S, Meusel M, Sayk F, Reppel M, Dalhoff K, Mortensen K, Droemann D. E‐cigarettes and cigarettes worsen peripheral and central hemodynamics as well as arterial stiffness: a randomized, double‐blinded pilot study. Vasc Med (United Kingdom). 2018;23:419–425. doi: 10.1177/1358863X18779694 [DOI] [PubMed] [Google Scholar]
- 24. Majid S, Weisbrod RM, Fetterman JL, Keith RJ, Rizvi SHM, Zhou Y, Behrooz L, Robertson RM, Bhatnagar A, Conklin DJ, et al. Pod‐based e‐liquids impair human vascular endothelial cell function. PLoS One. 2023;18:e0280674. doi: 10.1371/journal.pone.0280674 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Rao P, Han DD, Tan K, Mohammadi L, Derakhshandeh R, Navabzadeh M, Goyal N, Springer ML. Comparable impairment of vascular endothelial function by a wide range of electronic nicotine delivery devices. Nicotine Tob Res. 2022;24:1055–1062. doi: 10.1093/ntr/ntac019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Siddiqi TJ, Rashid AM, Siddiqi AK, Anwer A, Usman MS, Sakhi H, Bhatnagar A, Hamburg NM, Hirsch GA, Rodriguez CJ, et al. Association of Electronic Cigarette Exposure on cardiovascular health: a systematic review and meta‐analysis. Curr Probl Cardiol. 2023;48:101748. doi: 10.1016/j.cpcardiol.2023.101748 [DOI] [PubMed] [Google Scholar]
- 27. Carnevale R, Sciarretta S, Violi F, Nocella C, Loffredo L, Perri L, Peruzzi M, Marullo AGM, De Falco E, Chimenti I, et al. Acute impact of tobacco vs electronic cigarette smoking on oxidative stress and vascular function. Chest. 2016;150:606–612. doi: 10.1016/j.chest.2016.04.012 [DOI] [PubMed] [Google Scholar]
- 28. Rose JJ, Krishnan‐Sarin S, Exil VJ, Hamburg NM, Fetterman JL, Ichinose F, Perez‐Pinzon MA, Rezk‐Hanna M, Williamson E. Cardiopulmonary impact of electronic cigarettes and vaping products: a scientific statement from the American Heart Association. Circulation. 2023;148:703–728. doi: 10.1161/CIR.0000000000001160 [DOI] [PubMed] [Google Scholar]
- 29. Martinez‐Morata I, Sanchez TR, Shimbo D, Navas‐Acien A. Electronic cigarette use and blood pressure endpoints: a systematic review. Curr Hypertens Rep. 2020;23:2. doi: 10.1007/s11906-020-01119-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Leventhal AM, Tackett AP, Whitted L, Jordt SE, Jabba SV. Ice flavors and non‐menthol synthetic cooling agents in E‐cigarette products: a review. Tob Control. 2022;32:769–777. doi: 10.1136/tobaccocontrol-2021-057073 [DOI] [PMC free article] [PubMed] [Google Scholar]