Particulate matter emissions of different brands of mentholated cigarettes

ABSTRACT

Inhaling particulate matter (PM) in environmental tobacco smoke (ETS) endangers the health of nonsmokers. Menthol, an additive in cigarettes, attenuates respiratory irritation of tobacco smoke. It reduces perceptibility of smoke and therefore passive smokers may inhale ETS unnoticed. To investigate a possible effect of menthol on PM concentrations (PM₁₀, PM_2.5, and PM₁), ETS of four mentholated cigarette brands (Elixyr Menthol, Winston Menthol, Reyno Classic, and Pall Mall Menthol Blast) with varying menthol content was analyzed. ETS was generated in a standardized way using an automatic environmental tobacco smoke emitter (AETSE), followed by laser aerosol spectrometry. This analysis shows that the tested cigarette brands, despite having different menthol concentrations, do not show differences with regard to PM emissions, with the exception of Reyno Classic, which shows an increased emission, although the menthol level ranged in the midfield. More than 90% of the emitted particles had a size smaller than or equal to 1 µm. Regardless of the menthol level, the count median diameter (CMD) and the mass median diameter (MMD) were found to be 0.3 µm and 0.5 µm, respectively. These results point out that there is no effect of menthol on PM emission and that other additives might influence the increased PM emission of Reyno Classic.

Implications: Particulate matter (PM) in ETS endangers the health of nonsmokers and smokers. This study considers the effect of menthol, an additive in cigarettes, on PM emissions. Does menthol increase the amount of PM? Due to the exposure to secondhand smoke nearly 900,000 people die each year worldwide. The aim of the study is to measure the particle concentration (L⁻¹), mass concentration (µg m⁻³), and dust mass fractions shown as PM₁₀, PM_2.5, and PM₁ of five different cigarette brands, including four with different menthol concentrations and one menthol-free reference cigarette, in a well-established standardized system.

Introduction

About 7 million people worldwide are killed yearly due to tobacco use (World Health Organization [WHO 2017]). Not only is actively smoking cigarettes harmful to health, but so is passive smoking, the unintended inhalation of tobacco smoke polluted air by a nonsmoker (Juranic et al. 2017; U.S.Public Health Service [USPHS] 2014). Due to the exposure to environmental tobacco smoke (ETS), around 890,000 people die each year (WHO 2017). ETS is a mixture of mainstream smoke, which is inhaled and exhaled by the smoker, and mostly sidestream smoke generated by the smoldering cigarette (Breuer et al. 2012). Protection against ETS exposure becomes more challenging due to tobacco additives like menthol (Connolly et al. 2000).

For nonsmokers the essential risk of adding menthol to cigarettes is unrecognized passive smoking due to reduced stale smoke odor and reduced smoke perceptibility (Connolly et al. 2000). Menthol is a monocyclic terpene alcohol (Buddrus and Schmidt, 2015; Kreslake et al., 2008) with different characteristics. Among other characteristics, menthol is analgetic (Lawrence, Cadman, and Hoffman 2011; Wickham 2015), is anesthetic (Ahijevych and Garrett 2004), and has masking sensory effects like cooling (Ahijevych and Garrett 2004; Lawrence, Cadman, and Hoffman 2011; Lee and Glantz 2011; Wickham 2015). Cigarettes with menthol as an additive are heavily promoted (Ahijevych and Garrett 2004; Bates et al., 1999; Celebucki et al. 2005). While the influence of menthol on smoke chemistry, toxicity, and physiological parameters like puff volume, puff frequency, and metabolism and addiction of nicotine has been extensively investigated (Ahijevych and Garrett 2004; Hoffman and Simmons 2011; Kahnert et al. 2012; U.S. Food and Drug Administration, 2016; Wickham 2015), only limited data on the effect of menthol on particulate matter (PM) in ETS are available (Brinkman et al. 2012).

PM serves as a parameter for assessing ETS exposure (Mueller et al. 2012). PM is a mixture of differently sized solid and liquid particles (Pope 2000) and is classified in various ways, such as by the emitting source or by size (CITEPA - Centre Interprofessionnel Technique d’Etudes de la Pollution Atmosphérique, 2017; Government of Canada 2017; Kim et al., 2015). The particle size determines which part of the respiratory system can be reached by inhaled PM (Brown et al. 2013; Brunekreef and Forsberg 2005; Gupta and Elumalai 2017; Umweltbundesamt [UBA] 2008). According to the current definition of the U.S. Environmental Protection Agency (EPA), particles of size equal to or smaller than 10 µm are called PM₁₀ and particles of size equal to or smaller than 2.5 µm are defined as PM_2.5 (EPA 2017). With decreasing size the particles can penetrate more deeply into the respiratory tract (Kim, Kabir, and Kabir 2015). Therefore, the PM₁ fraction including particles of size equal to or smaller than 1 µm is of special interest. (Landesanstalt für Umwelt Messungen und Naturschutz Baden-Württemberg, 2016). As the health risk of passive smoking depends on the PM size, it is of general interest to investigate whether menthol has an effect on PM size and distribution in PM emission.

Methods

In this study, PM levels emitted from four tobacco products with different menthol concentrations and one reference cigarette were collected and evaluated. Twenty cigarettes of each product were smoked using an automatic environmental tobacco smoke emitter (AETSE) according to a modified smoking protocol used in the Tobacco Smoke Particles and Indoor Air Quality (ToPIQ) studies (Gerber et al. 2015; Mueller et al. 2011).

The tested tobacco products consisted of four different mentholated cigarette brands and a 3R4F reference cigarette, manufactured for research purposes by the Institute of Agriculture at the University of Kentucky (University of Kentucky 2017). The mentholated cigarette brands had identical tar (10 mg), nicotine (0.8 mg), and carbon monoxide (10 mg) yields, whereas menthol yields were different. Elixyr Menthol contained 3.088 mg menthol, Reyno Classic 3.275 mg menthol, Winston Menthol 3.200 mg menthol, and Pall Mall Menthol Blast 5.102 mg menthol. The menthol-free 3R4F reference cigarette contained 9.4 mg of tar, 0.73 mg of nicotine, and 12 mg of carbon monoxide. (Roemer et al., 2012). An overview of the tobacco products and their composition is shown in Table 1. For further information on ingredients it is refered to the analyses of the Bundesministerium für ernährung und landwirtschaft (BMEL 2017).

Table 1. A detailed list of the cigarette ingredients with refer to tar, nicotine, carbon monoxide, and menthol.

	Tar [mg]	Nicotine [mg]	Carbon monoxide [mg]	Menthol [mg]	Menthol level [%]
3R4F	9.4	0.73	12.0	—	—
Elixyr Menthol	10.0	0.8	10.0	3.088	0.45
Reyno Classic	10.0	0.8	10.0	3.275	0.55
Winston Menthol	10.0	0.8	10.0	3.200	0.55
Pall Mall Menthol Blast	10.0	0.8	10.0	5.102	0.75

The AETSE, designed and engineered by Schimpf-Ing, Trondheim, Norway (Schimpf 2015), allows the smoking of cigarettes in a reproducible way without damaging test persons. The smoking pump was located in a separated 2.88-m³ glass chamber. With the help of a stepper motor, a linear actuator moved a 200-mL glass syringe for imitating the smoking behavior. Using a microcontroller, parameters like puff volume (40 mL), puff flow rate (13 mL sec⁻¹), puff frequency (2 min⁻¹), interpuff interval (22 sec), and amount of 7 puffs were adjusted.

The smoking protocol consisted of four different phases: First was the preignition phase of 5 min, in which the baseline values were measured. Then came the combustion phase of 4 min and 8 sec, in which the cigarette was lighted and smoked, followed by the postcombustion phase of 5 min, starting with the distinguishing of the cigarette. Finally was the suction phase of 5 min, in which the air was cleaned by using a high-performance industrial suction before the next cycle started. Each cycle was 19 min and 8 sec long.

During the smoking cycle the PM concentrations in the glass chamber were measured at a dilution of 1:10 by a portable laser aerosol spectrometer and dust monitor, Grimm model 1.109 (Grimm 2010; 2012; Grimm and Eatough 2009; Han, Symanski, and Stock 2017). Light scattering was used to detect particles with a minimum size of 0.25 µm. Every 6 sec the received data were recorded and displayed as particle concentration (L⁻¹), mass concentration (µg m⁻³), and dust mass fractions shown as PM₁₀, PM_2.5, and PM₁. This differs from the EPA definition, as particles smaller than 0.25 µm are not included because of the technical limitation of the used aerosol spectrometer (particle size range 0.25–32 µm). In this study, PM_10-2.5 was defined as the difference between PM₁₀ and PM_2.5, and PM_2.5–₁ as the difference between PM_2.5 and PM₁.

PM values were analyzed using the area under the curve (AUC), describing the area under a concentration–time curve, and mean concentration (C_mean), referring to the mean value of one sample size consisting of 20 cigarettes. The obtained data were tested for artificial peaks. Therefore, the AUCs including artificial peaks were compared to the AUCs excluding artificial peaks for five randomly chosen cigarettes per sample. The proportion of peaks was smaller than 10%. Moreover, the received data were tested for outliers. One outlier was detected (significance level of 0.05). Finally, a test for Gaussian normality was performed using the D’Agostino and Pearson omnibus normality test and the Shapiro–Wilk normality test, both with a cutoff of 0.05. Each tested cigarette type sample passed the tests. Afterward, a one-way analysis of variance (ANOVA) and following Tukey’s multiple comparisons test were done for the four mentholated brands. In addition, a posttest for linear trends was used to test for a relationship between menthol concentration and PM emission in different tobacco products. All tests with regard to AUC and C_mean were carried out using GraphPad Prism version 6.

In order to derive count median diameters (CMD) and mass median diameters (MMD) the data of 10 randomly chosen cigarettes for each sample size were recorded and the medians were calculated using MATLAB. The evaluation covers values recorded during the preignition, combustion, and postcombustion phases; values during the suction phase were not included.

Results

The results of the AUC-PM values are presented in the boxplots in Figure 1. The test products were smoked exactly the same way with an identical suction volume, speed, and puff amount. Hence, the results allow for a direct comparison of the differences in PM emission among the cigarette brands.

Figure 1. Comparative boxplots of area under the curve (AUC)-PM of the tested cigarettes. In these boxes, extending from the 25th to the 75th percentile, the cross indicates the mean value. (a) AUC-PM₁₀, (b) AUC-PM_2.5, (c) AUC-PM₁.

Menthol has no statistically significant effect on the PM emission

The comparison of the AUC-PM₁₀, AUC-PM_2.5, and AUC-PM₁ of the 3R4F reference cigarette and the mentholated brands shows no significant differences for Elixyr Menthol, Winston Menthol, and Pall Mall Menthol Blast (p > 0.05). In contrast, Reyno Classic produces 28% higher emission with regard to AUC-PM₁₀ and AUC-PM_2.5 and 25% higher emission in AUC-PM₁ (p < 0.001).

By far the highest emissions were produced by Reyno Classic cigarettes, although the menthol level was in midfield and similar to that for the Winston Menthol cigarettes. In accordance, the posttest for linear trends confirmed that there is no linear concentration-dependent effect of menthol on PM emission. This leads to the conclusion that other additives or manufacturing processes might influence the PM emissions.

Regardless of the menthol level, the CMD is 0.3 µm (median absolute deviation, MAD = 0.05) and the MMD is 0.5 µm (MAD = 0.15) for all investigated products. The results are shown in Figure 2. The distribution pattern is not influenced by menthol either. In all five tested cigarette types, by far the largest part of PM is represented by particles smaller or equal to 1 µm and larger than 0.25 µm at more than 90%. Particles larger than 1 µm and smaller or equal to 2.5 µm represent 5.2–7.5% and particles greater than 2.5 µm and smaller or equal to 10 µm represent less than 0.5%.

Figure 2. Distribution pattern of PM_10-2.5, PM_2.5–1, and PM₁ of all investigated cigarettes.

To sum up, this study shows that menthol has neither a statistically significant effect on the emitted PM concentration nor a statistically significant effect on the distribution pattern or median diameters of PM.

Discussion

This study showed that there is no concentration dependent effect of menthol on PM emission. No differences in the AUC or C_mean values between the reference menthol free cigarette and the mentholated cigarettes have been found. An exception is the Reyno Classic brand, although the menthol yield of Reyno Classic ranged in midfield. With all products being identical with regard to tar, nicotine, and carbon monoxide composition, it remains unclear why Reyno Classic, which regarding the menthol concentration ranges in midfield, has 25–28% higher PM emissions. We assume, therefore, that other ingredients or manufacturing processes might influence the PM emission. In the literature, only a few results are reported with respect to the effect of additives on total particulate matter (TPM), with contradictory findings. Gaworski et al. (1997) analyzed the toxicity of cigarette smoke inhalation in an in vivo study with rats. They did not report a significant difference between TPM of the menthol free reference cigarette and that of the mentholated cigarette. Wasel et al. (2015) compared additive-free cigarettes with cigarettes including additives and did not show a significant difference with reference to PM. They assumed that the filter could influence the PM emissions. The smallest filter produces the highest amount of PM. A study on cigarette mainstream smoke of cigarettes with and without additives performed by Rustemeier et al. (2002) revealed contrasting results. In this study, 333 ingredients were added, which were matched to three groups. Each of the three cigarette types with additives showed an increase of 13–28% relative to the cigarette without additives. Wertz et al. (2011) reanalyzed tobacco industry documents and reported an increase of toxins, including PM, of cigarettes containing additives. Another study focused on tobacco moisture. The lowest contents of TPM were reported for the cigarettes with the highest level of humidity (Djulancic, Radojicic, and Srbinoska 2013). To sum up, it still remains unclear whether additives or manufacturing differences like filter length and tobacco moisture influence the PM emissions most. The distribution pattern of all investigated cigarettes was similar regardless of the menthol yield. The largest part is represented by particles larger than 0.25 µm and smaller than or equal to 1 µm. Similar results were published by Keith and Derrick in 1960, describing that the majority of particles has a size between 0.1 µm and 1 µm. These findings are alarming because smaller particles are more harmful to health (Li et al. 2016; Meng et al. 2013). Although PM₁ is contained in PM_2.5, which in turn is contained in PM₁₀, according to EPA, it constitutes by far the most significant fraction in both. Therefore, it has been considered separately in accordance with De Marco et al. (2015), Mueller et al. (2011), and Protano et al. (2014).

As limitations, it should be pointed out that an AETSE and a laser aerosolspectrometer with a measuring range from 0.25 µm to 30 µm were used in this study. In particular, the smoking behavior of humans and the ETS cannot be imitated exactly using the AETSE. In the respiratory tract, inhaled mainstream smoke can change due to hygroscopic growth (McGrath et al., 2009), and the exhaled smoke particles are larger by a factor of 1.5 ± 0.3 (Sahu et al. 2013). When using an AETSE, it is not possible to detect differences between inhaled mainstream smoke and exhaled mainstream smoke. However, it should be noted that mainstream smoke only constitutes a small fraction, about 15%, of ETS, whereas about 85% is made up of sidestream smoke (Jenkins et al. 2004; Keil, Prugger, and Heidrich 2016; Nowak et al. 2008). Sidestream smoke in turn can be imitated excactly by using the AETSE. Therefore, the measured PM emissions are very similar to ETS. Nonetheless, this has guaranteed reproducible results by avoiding interindividual deviations and without endangering test persons. The laser aerosolspectrometer Grimm 1.109 is especially built for the continuous measurement of PM and is commonly used in monitoring networks (Burkart et al. 2010). ETS consists mostly of particles between 0.02 and 2 µm in diameter (Nazaroff and Klepeis 2003). However, there is no common agreement about the peak size. On the one hand, it was described by geometric mean diameters of sidestream smoke particles of 0.1 µm (Guerin, Higgins, and Jenkins 1987; Ueno and Peters 1986). On the other hand, Haustein and Groneberg (2008) report mean diameters of 0.5 µm. Nevertheless, a new measurement system is needed for particles smaller than 0.25 µm. Furthermore, a modified smoking regime, and none of the already existing ones like ISO or FTC, was used. However, there is general agreement that no “gold standard” of smoking regime exists (ISO 2016; Liu, Mcadam, and Perfetti 2011; Marian et al. 2009; Wright 2015). Moreover, the main focus of this study lies on data comparison, and therefore the use of the AETSE and the modified smoking protocol is justified.

By comparing the PM emissions of different brands of mentholated cigarettes, it was found that menthol does not have a dose-dependent effect on PM emissions, nor does it influence the CMD and the MMD. Particles smaller than or equal to 1 µm (PM₁) are responsible for the main part of PM emissions. It remains unclear why the Reyno Classic cigarettes produce higher emissions than the other test cigarettes. This result leads to the assumption that other additives than menthol or manufacturing processes might influence the PM emission. To answer this question, further research focusing on particles smaller than 0.25 µm is needed.

Acknowledgment

This article is part of the thesis of J. Gerharz. M. Braun, R. Mueller, and D. A. Groneberg (Equal contribution) contributed significantly to conception and design of the study. Moreover, they prepared the experiments, which were performed by J. Gerharz. J. Gerharz and R. Mueller (Equal contribution) analyzed the data. The paper was written by J. Gerharz and critically reviewed by all authors. All authors have participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors have read and approved the final paper.

Additional information

Notes on contributors

Julia Gerharz

Julia Gerharz is physician and was a Ph.D. student at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine.

Michael H.K. Bendels

Michael H.K. Bendels is a specialist in neurology and a senior scientist at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine.

Markus Braun

Markus Braun is a technical associate at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine.

Doris Klingelhöfer

Doris Klingelhöfer is Msc in agricultural science and a senior scientist at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine

David A. Groneberg

David A. Groneberg is a full professor, Chairman and Director at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine.

Ruth Mueller

Ruth Mueller is biologist and head of department of Environmental Toxicology and Medical Entomology at the Institute of Occupational Medicine, Social Medicine and Environmental Medicine.