Scholarly article on topic 'Relationship between cigarette format and mouth-level exposure to tar and nicotine in smokers of Russian king-size cigarettes'

Relationship between cigarette format and mouth-level exposure to tar and nicotine in smokers of Russian king-size cigarettes Academic research paper on "Health sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{Cigarettes / Smokers / "Smokers mouth-level exposure" / "Cigarette smoke constituents" / Tar / Nicotine}

Abstract of research paper on Health sciences, author of scientific article — Madeleine Ashley, Mike Dixon, Krishna Prasad

Abstract Differences in length and circumference of cigarettes may influence smoker behaviour and exposure to smoke constituents. Superslim king-size (KSSS) cigarettes (17mm circumference versus 25mm circumference of conventional king-size [KS] cigarettes), have gained popularity in several countries, including Russia. Some smoke constituents are lower in machine-smoked KSSS versus KS cigarettes, but few data exist on actual exposure in smokers. We investigated mouth-level exposure (MLE) to tar and nicotine in Russian smokers of KSSS versus KS cigarettes and measured smoke constituents under machine-smoking conditions. MLE to tar was similar for smokers of 1mg ISO tar yield products, but lower for smokers of 4mg and 7mg KSSS versus KS cigarettes. MLE to nicotine was lower in smokers of 4mg KSSS versus KS cigarettes, but not for other tar bands. No gender differences were observed for nicotine or tar MLE. Under International Organization for Standardization, Health Canada Intense and Massachusetts regimes, KSSS cigarettes tended to yield less carbon monoxide, acetaldehyde, nitric oxide, acrylonitrile, benzene, 1,3-butadiene and tobacco-specific nitrosamines, but more formaldehyde, than KS cigarettes. In summary, differences in MLE were observed between cigarette formats, but not systematically across pack tar bands.

Academic research paper on topic "Relationship between cigarette format and mouth-level exposure to tar and nicotine in smokers of Russian king-size cigarettes"

Contents lists available at ScienceDirect

Regulatory Toxicology and Pharmacology

journal homepage: www.elsevier.com/locate/yrtph

Relationship between cigarette format and mouth-level exposure to tar ■. and nicotine in smokers of Russian king-size cigarettes

CrossMark

ig-size cigarettes

Madeleine Ashley *, Mike Dixon, Krishna Prasad

British American Tobacco, Group Research and Development, Regents Park Road, Southampton SO15 8TL, UK

ARTICLE INFO

Article history: Received 9 June 2014 Available online 19 August 2014

Keywords: Cigarettes Smokers

Smokers mouth-level exposure Cigarette smoke constituents Tar

Nicotine

ABSTRACT

Differences in length and circumference of cigarettes may influence smoker behaviour and exposure to smoke constituents. Superslim king-size (KSSS) cigarettes (17 mm circumference versus 25 mm circumference of conventional king-size [KS] cigarettes), have gained popularity in several countries, including Russia. Some smoke constituents are lower in machine-smoked KSSS versus KS cigarettes, but few data exist on actual exposure in smokers. We investigated mouth-level exposure (MLE) to tar and nicotine in Russian smokers of KSSS versus KS cigarettes and measured smoke constituents under machine-smoking conditions. MLE to tar was similar for smokers of 1 mg ISO tar yield products, but lower for smokers of 4 mg and 7 mg KSSS versus KS cigarettes. MLE to nicotine was lower in smokers of 4 mg KSSS versus KS cigarettes, but not for other tar bands. No gender differences were observed for nicotine or tar MLE. Under International Organization for Standardization, Health Canada Intense and Massachusetts regimes, KSSS cigarettes tended to yield less carbon monoxide, acetaldehyde, nitric oxide, acrylonitrile, benzene, 1,3-butadiene and tobacco-specific nitrosamines, but more formaldehyde, than KS cigarettes. In summary, differences in MLE were observed between cigarette formats, but not systematically across pack tar bands.

© 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/3XI/).

1. Introduction

Differences in the length and circumference of cigarettes may influence smoker behaviour and exposure to cigarette smoke constituents. It has been shown that smokers of 100 mm length cigarettes demonstrated a higher mouth-level exposure (MLE) to tar and nicotine than smokers of similar machine-derived yield king-size (83-85 mm length) cigarettes (St Charles et al., 2010; Nelson et al., 2011). Additionally, higher tar and nicotine MLE was observed in smokers of Canadian king-size (84 mm length) cigarettes than in smokers of similar machine-derived yield Canadian regular (72 mm length) cigarettes (Côté et al., 2011).

Two studies have reported on the effect of cigarette circumference on the exposure of smokers to cigarette smoke constituents. St Charles et al. measured MLE to tar and nicotine in smokers of different US cigarettes. They reported a marginally lower mean MLE to tar for smokers of a 17 mm circumference, 100 mm length cigarette compared with that seen in smokers of a similar machine yield 24 mm circumference, 100 mm length cigarette (St Charles et al., 2010). In contrast, a study conducted among Romanian

* Corresponding author. E-mail addresses: madeleine_ashley@bat.com (M. Ashley), dixon.consultancy @hotmail.co.uk (M. Dixon), krishna_prasad@bat.com (K. Prasad).

smokers reported no significant differences in MLE to tar and nicotine between smokers of conventional (25 mm circumference) and superslim (17 mm circumference) cigarettes of similar length and International Organization for Standardization (ISO) tar and nicotine yields (Ashley et al., 2011).

The mainstream machine-smoke emissions of six superslim (17 mm circumference) Canadian cigarette brands were compared with predicted emission data obtained from the Canadian Benchmark (Siu et al., 2013). The Canadian Benchmark is produced annually and is based on regression equations between the tar yields and the yields of a range of smoke analytes obtained from a minimum of 28 conventional circumference cigarettes from the Canadian market. This study reported lower mainstream smoke emissions per cigarette of carbon monoxide, carbonyls, volatiles and aromatic amines, but higher emissions of some smoke constituents such as formaldehyde, for the superslim products compared with the Canadian Benchmark. A subsequent study by the same group examined toxicological endpoints in response to exposure to cigarette smoke from superslim cigarettes (Mladjenovic et al., 2014) and noted reductions in the toxicity per mg total particulate matter (TPM) and per mg nicotine of the derived smoke, potentially as a consequence of the lower toxicant levels in these cigarettes. As a result, the authors of these two papers have expressed a concern that superslim cigarettes may be considered

http://dx.doi.org/10.1016/j.yrtph.2014.08.002 0273-2300/® 2014 The Authors. Published by Elsevier Inc.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

by consumers as being 'less harmful' than conventional circumference cigarettes. Although the superslim cigarettes generated lower machine-derived emissions, lower emission levels are not necessarily linked to a reduction in smokers' exposure to cigarette smoke constituents or to a reduced health risk. Therefore it is important to examine the exposure of smokers to mainstream smoke constituents from superslim cigarettes.

The primary aim of this study was to determine the effect of cigarette circumference (17 mm versus 25 mm) on smokers' MLE to tar and nicotine. We also measured smoke constituents under ISO, Health Canada Intense (HCI) and Massachusetts machinesmoking regimes. A further aim of the study was to measure smokers' puffing topography, in order to determine whether different physical parameters of cigarettes, such as cigarette pressure drop or draw resistance resulting from the reduction in cigarette circumference, influence puffing behaviour. The study was carried out in a Russian population due to the popularity of the superslim cigarette in this market.

2. Methods

2.1. Study products

We compared a king-size superslim (17 mm diameter; KSSS) with a conventional king-size product (25 mm diameter; KS) within the ISO pack tar bands of 1 mg, 4 mg and 7 mg. All products included in the study were commercially available and conformed to British American Tobacco standard manufacturing specifications. Each product was sourced from a single batch.

2.2. Study participants

A market research agency recruited a target of 60 healthy male and female smokers in approximately equal numbers to each of the six product groups described in Section 2.1. Full inclusion/exclusion criteria are provided in the Supplementary Information. In brief, participants eligible for inclusion in the study were aged between 21 and 50 years, had a self-reported average cigarette consumption of at least ten cigarettes per day of one of the study products and were required to have been smoking their usual brand for at least 6 months. Women were excluded if they reported that there was any possibility that they were pregnant. All participants were screened using a written questionnaire and provided written informed consent prior to the study.

2.3. Study protocol

Participants were required to attend three visits at a study site in Moscow over a 12 day period. At visit 1 (Day 1), all participants who met the inclusion criteria were briefed on the study protocol. Participants were provided with diaries in which to record daily cigarette consumption of cigarettes purchased themselves. Diaries covered consecutive days, labelled from Monday to Sunday (Days 2-8) and participants were provided with instructions on how to record the number of cigarettes they smoked each day. On Day 9, participants returned for visit 2 with the completed cigarette consumption diaries and each participant was provided with a filter cutter, training and instructions for cigarette-filter collection and sufficient cigarettes of their usual product to smoke on Days 10 and 11. Participants were asked to smoke the supplied cigarettes in their normal manner and environment, and to collect a minimum of 15 filters from the spent cigarettes. They were instructed only to collect filter tips from the cigarettes supplied. On Day 12, participants attended visit 3 to assess puffing topography and to

provide the collected filters, which were then stored at 4 °C prior to part-filter analysis.

2.4. Analytical methods

2.4.1. Mouth-level exposure to tar and nicotine

Part-filter analysis was used to estimate smokers' MLE to tar and nicotine, as previously described (St Charles et al., 2009). In brief, the estimation of MLE relies on the relationship between the amount of tar and nicotine delivered to the smoker and the amount retained within the filter of the cigarette, as defined by calibration smoking.

Each participant's spent filters collected in the filter cutters were split randomly into three replicates each containing five tips, which were analysed independently on different days. The length of each filter tip was measured (±0.1 mm), and recorded before being extracted in methanol containing n-heptadecane as an internal standard (Ashley et al., 2011). The extracts were analysed for tip nicotine and tar using gas chromatography and ultraviolet (UV) absorbance (using a variable wavelength detector set at 310 nm as described previously (St Charles et al., 2009)), respectively. Calibration data were produced by machine smoking each product over a wide range of typical human smoking behaviour parameters.

MLE to nicotine was estimated for each replicate using the tip nicotine values measured from the smoker's spent filters and the linear regression equation obtained by plotting mainstream smoke nicotine yield versus tip nicotine data obtained during calibration smoking. Similarly, MLE to tar was estimated using the UV absor-bance per tip data from the smoker's spent filters and the linear regression equation derived by plotting mainstream smoke nicotine-free dry particulate matter (NFDPM [tar]) yield versus UV absorbance per tip during calibration smoking.

2.4.2. Smoke constituent yields

Mainstream smoke yields of NFDPM (tar), nicotine, carbon monoxide and selected Hoffmann analytes were measured using ISO, Massachusetts and HCI machine-smoking regimes. These regimes, along with descriptions of the analytical methods used to measure smoke constituents in the present study, have been described in detail previously (McAdam et al., 2011, 2012) and on the British American Tobacco science website (www.bat-science.com).

2.4.3. Puffing topography

Puffing topography (puff volumes, intervals between puffs and number of puffs) was analysed by providing each participant with two cigarettes of their usual product which they were requested to smoke through a smoking analyser, with an interval between each cigarette of at least 20 min. Puffing topography data were recorded using a proprietary portable smoking analyser (SA7; developed in collaboration with C-Matic Limited, Crowborough, UK). The SA7 consists of a cigarette holder, with a unidirectional pressure transducer. The pressure transducer detects a pressure change across an orifice (2 mm), which is proportional to the flow rate (Slayford et al., 2012).

2.5. Data analysis

Minitab 16 statistical software (Minitab Inc., PA, USA) was used to conduct statistical analysis. MLE data are presented as mean ± standard deviation (SD). The physical characteristics of cigarettes and smoke toxicant data from machine-smoking regimes are presented as mean values. Analysis of variance general linear model (ANOVA GLM) was used to compare smokers' MLE and puffing topography data by smoker group. Where a significant differ-

ence was found, Tukey's post hoc test was applied to investigate the source of the difference.

3. Results

3.1. Study products

The physical characteristics of the study cigarettes are shown in Table 1. Other than differences in dimensions between study cigarettes, the cigarette types exhibited other notable differences. The KSSS cigarettes contained 41-52% less tobacco than the KS products. All cigarette types had a carbon filter, but the number of segments in the filter differed. The 1 and 4 mg KS products had three filter sections, compared with the 7 mg KS and all KSSS products, which had two filter sections. Additional design characteristics, such as ventilation, differed to produce the specified tar and nicotine yields under the ISO machine-smoking regime.

3.2. Study participants

Demographic characteristics of study participants were similar between product groups (Table 2). In each group, male and female participants were recruited to the study in approximately equal numbers.

3.3. Mouth-level exposure to tar and nicotine

MLE to tar and nicotine per cigarette were greater for all study products in smokers than the yields achieved under the ISO machine-smoking regime. Variations in smoking behaviour resulted in a range of MLE values per study product (Table 3; Figs. 1 and 2).

MLE to tar was similar in smokers of 1 mg KS and 1 mg KSSS cigarettes, and was significantly lower than for smokers of any 4 mg or 7 mg products. However, within the 4 mg and 7 mg groups, smokers of the KSSS products had significantly lower MLE to tar than smokers of the KS products. Smokers of 4 mg KS and 7 mg KSSS products obtained statistically similar MLE to tar.

For MLE to nicotine, no statistically significant differences were found between smokers of the KSSS and KS products within the 1 and 7 mg groups. However, within the 4 mg group, smokers of the KSSS product had significantly lower MLE to nicotine than smokers of the KS product. Smokers of the 1 mg KS product had a similar

MLE to nicotine to smokers of the 4 mg KSSS product. Within a product type (KS or KSSS) the smokers of 4 and 7 mg cigarettes obtained statistically similar nicotine yields.

When stratified by gender, we found no effect of gender on MLE to either tar (p = 0.851) or nicotine (p = 0.741), (Table 4).

3.4. Smoke constituent yields

The results of the standard machine-smoke chemistry analyses for each study cigarette are shown in Table 5. The primary objective of the study was based on the assumption that tar and nicotine yields would be well matched across study products when machine smoked under the ISO regime. The 4 and 7 mg products were reasonably well matched, but the 1 mg products were less well matched (ISO yields: 1.1 mg/cig tar and 0.11 mg/cig nicotine for KS product, versus 1.7 mg/cig tar and 0.18 mg/cig nicotine for KSSS product). The study products generally maintained the ISO ranking for yields of tar and nicotine (1 < 4 < 7 mg pack tar band) when smoked using the intense (Massachusetts and HCI) regimes, though little difference was observed between the KSSS products using HCI. The tar and nicotine yields obtained under the Massachusetts smoking regime for the study products were most similar to the smokers' mean MLE to tar and nicotine.

Although no formal statistical analyses were performed, on an observational basis the KSSS products tended to yield less carbon monoxide, acetaldehyde, nitric oxide, acrylonitrile, benzene, 1,3-butadiene and tobacco-specific nitrosamines, N-nitrosonornicotine ketone (NNK) and N-nitrosonornicotine (NNN), but more formaldehyde, than the KS products (Table 5). The KSSS products yielded 16-45% less carbon monoxide than the corresponding KS products under all machine-smoking regimes, particularly the intense regimes (Table 5).

3.5. Puffing topography

Mean values for total puff volume, mean puff volume and puff number are shown in Table 6. Smokers of the highly ventilated 1 mg products had significantly higher total puff volumes and puff numbers than smokers of the other products (p < 0.001).

Within the 1 mg groups, there were no significant differences in any of the puffing parameters between smokers of KS and KSSS cigarettes. Within the 4 mg and 7 mg groups, smokers of the KSSS products had significantly lower total and mean puff volumes than

Table 1

Physical characteristics of study products.

Characteristic KS 1 mg KSSS 1 mg KS 4 mg KSSS 4 mg KS 7 mg KSSS 7 mg

Filter carbon (mg)* 46 16 56 16 40 16

Pack tar (mg/cig) 1 1 4 4 7 7

Tobacco blend total alkaloids (DWB %) 2.53 2.24 2.22 2.10 1.92 2.34

Filter type Triple Dual Triple Dual Dual Dual

Mouth section (mm) 8.0 11.0 7.1 10.3 15.0 11.0

Middle section (mm) 12.0 - 12.4 - - -

Tobacco end section (mm) 7.0 16.0 7.1 16.6 12.0 16.0

Ventilation (%) 75.9 85.5 63.6 61.2 38.9 40.5

Paper permeability (CORESTA) 48.6 22.4 45.0 25.5 45.3 24.0

Cigarette PD-open (mmWG) 104.9 81.3 82.3 130.9 83.0 149.2

Filter length (mm) 27.0 27.0 26.6 26.9 27.0 27.0

Tobacco length (mm) 56.0 56.0 56.4 56.1 56.0 56.0

Overtip length (mm) 32.0 32.0 32.0 32.0 32.0 32.0

Tobacco weight (mg/cig) 518 307 592 318 663 319

Density (13.5% moisture, mg/cig) 201 251 228 258 252 258

CORESTA, Centre de Coopération pour les Recherches Scientifiques Relatives au Tabac; DWB, dry weight basis; KS, king-size conventional product; KSSS, king-size superslim product; mmWG, mm water gauge; PD, pressure drop.

King-size length was 83 mm (±4 mm). The circumference of all conventional products was 24.6 (±1 mm) and of superslim products was 17.0 mm (±1 mm). * Filter carbon in specification rather than measured.

Table 2

Demographic characteristics of smokers who completed the study.

KSSS 1 mg KS 1 mg KSSS 4 mg KS 4 mg KSSS 7 mg KS 7 mg

Number of participants 60 57 58 62 56 61

Gender

Male 31 29 28 32 28 31

Female 29 28 30 30 28 30

Number per age-group

21-24 years 16 13 17 15 8 12

25-29 years 15 12 18 18 8 12

30-44 years 24 26 14 19 35 31

45-50 years 5 6 9 10 5 6

KSSS, king-size superslim product; KS, king-size conventional product.

Table 3

Comparison of smokers' mouth-level exposure to tar and nicotine across all study products.

Product Tar band (mg) MLE to tar (mg/cig) MLE to nicotine (mg/cig)

Mean ± SD 5-95th percentile Tukey's ranking* Mean ± SD 5-95th percentile Tukey's ranking*

KSSS 1 mg 1 9.4 ± 3.2 3.7-15.3 d 0.86 ± 0.28 0.41-1.44 d

KS 1 mg 8.9 ± 3.4 4.8-15.6 d 0.98 ± 0.35 0.53-1.93 cd

KSSS 4 mg 4 13.5 ±4.2 6.8-21.2 c 1.18 ±0.35 0.64-1.83 bc

KS 4 mg 16.7 ±5.1 9.5-26.2 b 1.53 ±0.48 0.71-2.41 a

KSSS 7 mg 7 16.8 ±4.9 7.9-23.9 b 1.33 ±0.38 0.61-1.85 ab

KS 7 mg 19.5 ±6.2 9.2-27.8 a 1.43 ±0.50 0.63-2.14 a

KS, king-size conventional product; KSSS, king-size superslim product; MLE, mouth-level exposure; SD, standard deviation. * Same letter indicates no statistical difference (p > 0.05).

Product KSSS KS Pack tar (mg/cig) 1

KSSS KS KSSS KS

Fig. 1. Box plot of mouth-level exposure to tar in all smoker groups. Bottom, middle and top bars represent the 25th, 50th and 75th percentiles; bulls eye indicate the mean values; mean value labels included in plot; asterisks indicate outliers. KS, king-size conventional product; KSSS, king-size superslim product; MLE, mouth-level exposure.

LU -J 1.0

Product KSSS KS Pack tar (mg/cig) 1

KSSS KS KSSS KS

Fig. 2. Box plot of mouth-level exposure to nicotine in all smoker groups. Bottom, middle and top bars represent the 25th, 50th and 75th percentiles; bulls eye indicate the mean values; mean value labels included in plot; asterisks indicate outliers. KS, king-size conventional product; KSSS, king-size superslim product; MLE, mouth-level exposure.

did smokers of the KS products. For number of puffs, smokers of the 4 mg KSSS cigarettes took fewer puffs than smokers of 4 mg KS cigarettes. However, this was not observed for the 7 mg group, in which puff numbers were not significantly different when comparing KSSS and KS cigarettes. The greatest puff volumes and numbers were seen in smokers of 1 mg products. Smokers of the 4 mg products tended towards having higher total puff volumes and puff numbers than the smokers of 7 mg products.

3.6. Average daily cigarette consumption

Comparison of mean self-reported average daily cigarette consumption at screening with that recorded in the diaries during the study found no statistically significant differences within

smoker groups (Table 7). However, the average daily consumption recorded in the diary for each product was higher than that self-reported at screening, by an average of 0.9-1.7 cigarettes.

4. Discussion

Previous studies have demonstrated yields of some cigarette smoke toxicants in superslim format cigarettes are lower compared with those from KS circumference cigarettes, giving rise to concerns that superslim cigarettes may be considered by consumers as being 'less harmful' than conventional circumference cigarettes (Siu et al., 2013; Mladjenovic et al., 2014). However, data on actual exposure of smokers to mainstream smoke components from superslim cigarettes are lacking (St Charles et al., 2010;

Table 4

Comparison of mouth-level exposure to tar and nicotine across all study products stratified by gender.

Product Gender MLE to tar (mg/cig) MLE to nicotine (mg/cig)

Mean ± SD Tukey's ranking* Mean ± SD Tukey's ranking*

KSSS 1 mg Male 8.9 ± 2.9 e 0.81 ± 0.26 e

Female 10.0 ±3.4 de 0.91 ± 0.30 de

KS 1 mg Male 8.7 ± 3.0 e 0.95 ± 0.32 de

Female 9.0 ±3.8 e 1.01 ±0.37 cde

KSSS 4 mg Male 13.9 ±4.4 bcd 1.21 ±0.37 bcd

Female 13.0 ±3.9 cd 1.16 ±0.33 bcd

KS 4 mg Male 17.2 ±5.4 ab 1.60 ±0.53 a

Female 16.2 ±4.6 abc 1.46 ±0.43 ab

KSSS 7 mg Male 17.0 ±4.8 ab 1.34 ±0.37 abc

Female 16.6 ±5.0 abc 1.32 ±0.40 abc

KS 7 mg Male 19.2 ±5.8 a 1.42 ±0.50 ab

Female 19.8 ±6.6 a 1.45 ±0.51 ab

KS, king-size conventional product; KSSS, king-size superslim product; MLE, mouth-level exposure; SD, standard deviation. * Same letter indicates no statistical difference (p > 0.05).

Ashley et al., 2011). Therefore, this study was conducted to determine whether decreasing the cigarette circumference from 25 mm (conventional king-size format; KS) to 17 mm (superslim format; KSSS) influenced MLE to tar and nicotine in cigarettes matched in ISO tar and nicotine yields. The part-filter analysis method (St Charles et al., 2009) was chosen to measure MLE to tar and nicotine in smokers in their everyday environment as this non-invasive method has been shown to produce reliable estimates of tar and nicotine MLE (Shepperd et al., 2006), which strongly correlate with biomarkers of nicotine and other smoke constituents (St Charles et al., 2006; Shepperd et al., 2009; Morin et al., 2011).

Our results indicated statistically significant reductions in mean MLE to tar in smokers of the 4 and 7 mg tar yield KSSS products, compared with smokers of the equivalent tar yield KS products. However, mean MLE to tar in smokers of 1 mg tar yield KS and KSSS products was not significantly different. A possible reason for the different observations in smokers of the 1 mg products versus the 4 or 7 mg products, could be related to the fact that the 1 mg KSSS format produced higher NFDPM (tar) yields under machine-smoking conditions than those produced by the 1 mg KS format. However, this was not the case for the 4 and 7 mg tar yield products. Thus any potential reduction in MLE to tar conferred by the KSSS product would have been offset by the higher NFDPM yield of the 1 mg KSSS product versus the KS product. A similar trend was found for MLE to nicotine, which tended to be lower in smokers of KSSS cigarettes compared with smokers of the equivalent KS variants at each of the three nominal tar levels.

The results of this study differ from those obtained in a similar study conducted in Romania, in which mean MLE to tar and nicotine was similar between smokers of KSSS and KS cigarettes at each of the three nominal ISO tar yields examined (1, 4 and 7 mg) (Ashley et al., 2011). However, St Charles et al. observed a marginal reduction in mean MLE to tar in smokers of a 17 mm circumference, 100 mm length, 9 mg tar yield cigarette compared with smokers of a similar tar yield, 100 mm length, 24 mm circumference cigarette (St Charles et al., 2010). Consequently, one can conclude from the results of these three studies that smokers of superslim cigarettes do not experience higher MLE to tar and nicotine than smokers of similar tar yield conventional circumference cigarettes. Moreover, two of the three studies indicated that a slight reduction in MLE to tar may result from a decrease in cigarette circumference from 25 to 17 mm.

In our study of Russian smokers, significantly lower mean and total puff volumes were observed in smokers of the 4 and 7 mg KSSS cigarettes compared with smokers of similar tar yield KS cigarettes. Consequently, this reduction in total puff volume may

explain the lower mean tar and nicotine MLE values observed in smokers of the 4 and 7 mg tar yield KSSS cigarettes.

Several differences in the physical characteristics were seen between the KSSS and KS cigarettes. Open cigarette pressure drop was markedly higher for the 4 and 7 mg KSSS cigarettes than for the corresponding KS cigarettes. Consequently, resistance to draw would have been higher when smoking the KSSS variants compared with the KS variants. Studies analysing the relationship between draw resistance and puffing topography have shown an inverse relationship between draw resistance and puff volume (Dunn, 1978; Rawbone, 1984; Zacny et al., 1986). Thus, the higher pressure drops, or draw resistances, of the 4 and 7 mg KSSS cigarettes may have resulted in the lower mean and total puff volumes observed for smokers of the KSSS products versus the KS products.

Tar and nicotine MLE data were compared between the different ISO tar yield groups for both the KSSS and KS products. Both KSSS and KS cigarettes exhibited significant differences in MLE to tar between the different ISO tar yield groups: 1 < 4 < 7 mg. Significant differences in MLE to nicotine were also obtained for both KSSS and KS products, with a lower MLE in smokers of the 1 mg versus the 4 mg tar yield cigarettes.

The correlation between MLE to tar and ISO tar yield, and to some extent between MLE to nicotine and ISO nicotine yield, is consistent with results from other studies using the part-filter method to measure MLE to tar and nicotine in smokers (Shepperd et al., 2006, 2009; St Charles et al., 2006, 2010; Nelson et al., 2011; Côté et al., 2011; Ashley et al., 2011). The correlation between MLE to nicotine and machine-derived nicotine yields tends to be stronger than that observed between nicotine uptake biomarkers (for example, cotinine) and nicotine yields. Examples of nicotine biomarker correlations are provided in the National Institutes of Health Monograph 13 (NIH, 2001). The relatively weak correlations reported in many studies between nicotine yields and nicotine biomarkers are likely to be caused by inter-subject variation in the metabolism of nicotine. Such a source of variation is not present in nicotine MLE studies.

In studies that investigated the effect of gender on MLE to tar and nicotine, three studies reported a tendency for higher MLEs to tar and nicotine in male smokers compared with female smokers of similar tar and nicotine yield cigarettes (Mariner et al., 2011; Nelson et al., 2011; Côté et al., 2011 ). This gender difference in MLE is consistent with the results of puffing topography studies, in which male smokers had larger puff volumes than female smokers of the same cigarette type (Battig et al., 1982; Hofer et al., 1991; Hee et al., 1995; Eissenberg et al., 1999). In contrast, we did not observe any significant differences in MLE to tar and nicotine

Table 5

Comparison of machine-derived smoke constituent yields.

Analyte Smoking regime Mean of 5 replicate data

KS 1 mg KSSS 1 mg KS 4 mg KSSS 4 mg KS 7 mg KSSS 7 mg

NFDPM (tar) (mg/cig) ISO 1.1 1.7 3.8 3.7 8.0 7.0

MASS 9.0 11.0 13.7 12.8 19.4 15.5

HCI 15.3 17.6 21.5 18.0 25.2 18.0

Nicotine (mg/cig) ISO 0.11 0.18 0.33 0.36 0.60 0.63

MASS 0.78 0.98 1.16 1.14 1.42 1.30

HCI 1.13 1.47 1.63 1.41 1.69 1.39

Carbon monoxide (mg/cig) ISO 1.7 1.0 3.8 3.2 7.7 5.5

MASS 14.2 8.2 14.9 10.1 17.2 11.1

HCI 20.2 12.8 22.7 12.7 22.7 12.6

Puff Numbers ISO 6.8 6.6 7.9 5.8 7.9 5.8

MASS 8.6 9.5 10.9 8.1 11.2 8.4

HCI 6.9 7.7 8.6 7.4 9.6 7.2

1,3-Butadiene (ig/cig) ISO 10.7 6.3 17.5 19.7 51.2 26.3

MASS 67.4 35.6 99.6 46.6 119.9 76.9

HCI 121.2 81.5 116.0 85.1 120.1 69.0

2-Aminonaphthalene (ng/cig) ISO 1.4 1.3 4.0 2.8 7.5 3.8

MASS 6.6 5.4 11.1 5.9 17.4 7.0

HCI 7.4 6.1 10.7 6.5 14.7 6.2

4-Aminobiphenyl (ng/cig) ISO 0.3 0.3 0.9 0.6 1.6 0.8

MASS 1.6 1.2 2.5 1.3 3.9 1.6

HCI 2.0 1.6 2.8 1.6 3.9 1.5

Acetaldehyde (ig/cig) ISO 72.2 38.9 138.8 130.1 301.1 229.6

MASS 386.7 237.3 527.2 410.3 751.5 513.3

HCI 896.9 599.6 842.2 619.2 961.5 614.7

Acrolein (ig/cig) ISO 4.7 3.5 9.4 14.2 28.8 29.9

MASS 37.3 30.1 56.9 54.1 91.2 71.6

HCI 108.1 83.2 101.5 83.0 120.1 83.2

Acrylonitrile (ig/cig) ISO 0.8 0.9 2.0 3.3 6.9 5.4

MASS 11.0 7.2 15.5 10.4 22.7 15.6

HCI 24.6 16.1 21.5 17.9 27.4 15.7

Benzene (ig/cig) ISO 5.1 3.8 9.2 10.0 24.5 15.7

MASS 38.2 24.2 50.2 30.6 68.9 45.5

HCI 58.2 45.4 57.1 46.5 73.9 43.0

Benzo(a)pyrene (ng/cig) ISO 1.8 1.7 4.9 3.5 7.0 5.3

MASS 6.0 6.3 10.7 8.4 13.9 11.6

HCI 8.3 9.5 12.7 10.3 14.9 11.3

Cadmium (ng/cig) ISO <RL <RL 2.4 2.2 7.1 5.9

MASS 7.3 3.5 10.5 9.5 17.1 15.8

HCI 22.4 18.4 26.5 19.1 38.1 20.1

Catechol (ig/cig) ISO 8.3 11.8 22.6 21.2 37.2 29.8

MASS 33.5 42.6 62.7 50.1 79.1 64.3

HCI 42.1 62.5 69.1 58.8 87.1 73.3

Nitric oxide (ig/cig) ISO 43.1 <RL 74.9 <RL 120.0 52.2

MASS 150.5 76.2 190.2 95.5 258.5 110.5

HCI 246.4 149.5 266.8 140.9 319.8 146.2

Crotonaldehyde (ig/cig) ISO <RL <RL <RL 3.5 4.7 9.1

MASS 5.8 8.3 10.0 15.7 20.3 22.7

HCI 32.7 29.7 29.0 29.8 35.5 29.8

Formaldehyde (ig/cig) ISO 1.6 1.8 5.6 7.9 12.9 23.9

MASS 7.5 15.4 18.2 33.7 27.6 60.0

HCI 24.8 74.5 44.5 68.5 46.0 76.1

Hydrogen cyanide (ig/cig) ISO <RL <RL 9.9 27.0 37.0 68.5

MASS 92.4 76.8 141.0 138.5 194.1 178.5

HCI 235.9 188.3 258.9 170.4 269.8 188.1

Hydroquinone (ig/cig) ISO 8.1 9.8 19.7 19.7 34.5 28.3

MASS 34.2 36.5 54.9 48.0 75.9 62.2

HCI 50.5 53.9 66.5 58.1 87.6 70.2

N-Nitrosonornicotine (NNN) (ng/cig) ISO 19.7 12.3 34.3 22.1 84.2 30.1

MASS 93.1 51.4 121.9 66.6 170.7 71.0

HCI 155.3 108.7 168.9 102.5 213.6 105.6

N-Nitrosonornicotine ketone (NNK) (ng/cig) ISO 6.0 4.8 13.1 9.6 26.2 13.2

MASS 28.3 21.2 46.0 25.7 62.0 27.8

HCI 42.2 50.4 53.6 39.7 69.1 43.2

HCl, Health Canada Intense smoking regime; ISO, International Organization for Standardization smoking regime; KS, king-size conventional product; KSSS, king-size superslim product; <RL, below reporting limit; MASS, Massachusetts intense smoking regime; NFDPM, nicotine-free dry particulate matter.

Table 6

Comparison of smokers' puffing topography across study products.

Product Total puff volume (mL) Mean puff volume (mL) Puff number (n)

Mean ± SD Tukey's ranking* Mean ± SD Tukey's ranking* Mean ± SD Tukey's ranking*

KSSS 1 mg 1111 ±403 a 53.0 ±15.9 ab 21.9 ±8.0 a

KS 1 mg 1125 ±336 a 52.2 ±11.9 ab 22.3 ± У.5 a

KSSS 4 mg У08± 230 c 4У.9 ± 16.4 bc 15.4 ±4.5 c

KS 4 mg 8У6 ± 216 b 54.3 ±18.2 a 18.0 ±У.9 b

KSSS У mg 591 ±168 d 42.9 ±15.0 c 15.2 ±5.У c

KS У mg УУ1 ±240 c 51.2 ±16.8 ab 15.9 ±5.5 bc

KS, king-size conventional product; KSSS, king-size superslim product; SD, standard deviation. * Same letter indicates no statistical difference (p > 0.05).

Table 7

Comparison of smokers' average daily consumption of cigarettes across all study products.

Product Self-reported average daily consumption (n) Average daily consumption recorded in diary (n)

Mean ± SD Tukey's ranking* Mean ± SD Tukey's ranking*

KSSS 1 mg 15.4 ±5.2 a 16.3 ±5.1 a

KS 1 mg 14.0 ± 4.1 a 15.5 ±4.3 a

KSSS 4 mg 15.У ± 4.4 a 1У.3 ± 4.8 a

KS 4 mg 14.3 ± 3.9 a 15.8 ±4.1 a

KSSS У mg 16.0 ±57 a 1У.5 ± 6.У a

KS У mg 16.1 ±6.3 a 1У.8 ± У.У a

KS, king-size conventional product; KSSS, king-size superslim product; SD, standard deviation. * Same letter indicates no statistical difference (p > 0.05).

between males and females. Similarly, we did not observe a gender effect in our previous study of KSSS and KS cigarettes in a Romanian population (Ashley et al., 2011).

For mainstream smoke emissions, we found that the KSSS products tended to yield less carbon monoxide, acetaldehyde, nitric oxide, acrylonitrile, benzene, 1,3-butadiene and NNK and NNN, but more formaldehyde, than similar tar yield KS products when smoked under ISO, HCI and Massachusetts regimes in agreement with those observed for the superslim products versus the Canadian Benchmark (Siu et al., 2013). Coggins et al. reported the effects of reducing cigarette circumference from 27.1 to 17.0 mm on the machine-smoked yields of carbon monoxide, nitrogen oxides, hydrogen cyanide, acetaldehyde, and the nitrosamines N-nitrosoa-natabine (NAT), NNN and NNK. Once data were normalised for total particulate matter, lower levels of most of the smoke constituents were seen with the smaller circumference cigarettes, except for hydrogen cyanide, which increased as the circumference decreased from 27.1 to 17.0 mm (Coggins et al., 2013). In comparison, in our study we observed inconsistent results, with the yields of hydrogen cyanide being greater for the KSSS than the KS cigarettes when measured under ISO conditions, but lower when measured under the Massachusetts and HCI regimes.

Based on the results of our two studies on the effect of superslim cigarettes on MLE to tar and on the results from several mainstream smoke analytical studies, we would anticipate a reduction in smokers' exposure to many mainstream smoke constituents when switching from a conventional to a superslim format of a similar blend and tar yield cigarette. However, possible increases in exposure to formaldehyde may also occur following such a switch. Further studies measuring biomarkers of smoke constituent exposure are required to help determine the effect of reducing cigarette circumference on the exposure of smokers to specific smoke constituents. In conclusion, this study supports the previous findings of Siu et al., 2013, who demonstrated reductions in a number of mainstream smoke constituents in superslim format cigarettes when compared with conventional formats. We build on

these data by demonstrating that smokers of superslim cigarettes do not tend to experience higher MLEs to tar or nicotine, than smokers of similar tar yield conventional circumference cigarettes.

Conflict of interest

This work was funded by British American Tobacco (BAT), and all authors, with the exception of Dr. Mike Dixon are full time employees of BAT. Dr. Mike Dixon's involvement was in the capacity of a paid consultant to BAT.

Acknowledgments

The authors thank Michael Baer for calibration smoking and filter analysis, lona Khoroshavina for assistance with logistics in performing the study in Russia, and Sandy Slayford for collecting puffing topography data. We also thank lan Fearon for support in data interpretation and writing the manuscript. Editorial assistance with the preparation of this manuscript was provided by JEM Comms Ltd. and funded by British American Tobacco.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.yrtph.2014.08. 002.

References

Ashley, M., Sisodiya, A., McEwan, M., McAughey, J., Prasad, K., 2011. Comparison of mouth level exposure to 'tar' and nicotine in smokers of normal and superslim king size cigarettes in Romania. Beitr. Tabakforsch. Int. 24, 277-288. Battig, K., Buzzi, R., Nil, R., 1982. Smoke yield of cigarettes and puffing behaviour in

men and women. Psychopharmacology 76,139-148. Coggins, C.R., McKinney Jr., W.J., Oldham, M.J., 2013. A comprehensive evaluation of the toxicology of experimental, non-filtered cigarettes manufactured with different circumferences. Inhal. Toxicol. 25 (Suppl. 2), 69-72.

Côté, F., Létourneau, C., Mullard, G., Voisine, R., 2011. Estimation of nicotine and tar yields from human-smoked cigarettes before and after the implementation of the cigarette ignition propensity regulations in Canada. Regul. Toxicol. Pharmacol. 6, S51-S59.

Dunn, P.J., 1978. The effects of a reduced draw resistance cigarette on human smoking parameters and alveolar carbon monoxide levels. In: Thornton, R.E. (Ed.), Smoking Behaviour: Physiological and Psychological Influences. Churchill-Livingstone, Edinburgh, pp. 203-207.

Eissenberg, T., Adams, C., Riggins 3rd, E.C., Likness, M., 1999. Smokers' sex and the effects of tobacco cigarettes: subject-rated and physiological measures. Nicotine Tob. Res. 1, 317-324.

Hee, J., Callais, F., Mormas, L., Laurent, A.M., Min, S., Molinier, P., Chastagnier, M., Claude, J.R., Festy, B., 1995. Smokers' behaviour and exposure according to cigarette yield and smoking experience. Pharmacol. Biochem. Behav. 52, 195203.

Hofer, I., Nil, R., Battig, K., 1991. Nicotine yield as determinant of smoke exposure indicators and puffing behaviour. Pharmacol. Biochem. Behav. 40,139-149.

Mariner, D.C., Ashley, M., Shepperd, C.J., Mullard, G., Dixon, M., 2011. Mouth level smoke exposure using analysis of filters from smoked cigarettes: a study of eight countries. Regul. Toxicol. Pharmacol. 61, S39-S50.

McAdam, K.G., Gregg, E.O., Liu, C., Dittrich, D.J., Duke, M.G., Proctor, C.J., 2011. The use of a novel tobacco-substitute sheet and smoke dilution to reduce toxicant yields in cigarette smoke. Food Chem. Toxicol. 49,1684-1696.

McAdam, K.G., Gregg, E.O., Bevan, M., Dittrich, D.J., Hemsley, S., Liu, C., Proctor, C.J., 2012. Design and chemical evaluation of reduced machine-yield cigarettes. Regul. Toxicol. Pharmacol. 62,138-150.

Mladjenovic, N., Maertens, R.M., White, P.A., Soo, E.C., 2014. Mutagenicity of smoke condensates from Canadian cigarettes with different design features. Mutagenesis 29, 7-15.

Morin, A., Shepperd, C.J., Eldridge, A.C., Poirier, N., Voisine, R., 2011. Estimation and correlation of cigarette smoke exposure in Canadian smokers as determined by filter analysis and biomarkers of exposure. Regul. Toxicol. Pharmacol. 61 (Suppl. 3), S3-S12.

Nelson, P.R., Chen, P., Dixon, M., Steichen, T., 2011. A survey of mouth level exposure to cigarette smoke in the United States. Regul. Toxicol. Pharmacol. 61 (Suppl. 3), S25-S38.

National Institutes of Health, 2001. Monograph 13. Risks Associated with Smoking Cigarettes with Low Machine-Measured Yields of Tar and Nicotine. Available: http://cancercontrol.cancer.gov/brp/TCRB/monographs/13/m13_complete.pdf (last accessed 12.03.14).

Rawbone, R.G., 1984. The act of smoking. In: Cumming, G., Bonsignore, G. (Eds.), Smoking and the Lung. Plenum Press, New York and London, pp. 77-93.

Shepperd, C.J., St Charles, F.K., Lien, M., Dixon, M., 2006. Validation of methods for determining consumer smoked cigarette yields from cigarette filter analysis. Beitr. Tabakforsch. Int. 22, 176-184.

Shepperd, C.J., Eldridge, A.C., Mariner, D.C., McEwan, M., Errington, G., Dixon, M., 2009. A study to estimate and correlate cigarette smoke exposure in smokers in Germany as determined by filter analysis and biomarkers of exposure. Regul. Toxicol. Pharmacol. 55, 97-109.

Siu, M., Mladjenovic, N., Soo, E., 2013. The analysis of mainstream smoke emissions of Canadian 'super slim' cigarettes. Tob. Control 22, e10. http://dx.doi.org/ 10.1136/tobaccocontrol-2012-050450.

Slayford, S., Nother, K., Prasad, K., 2012. The SA7 smoking analyser: a puff topography device with a difference. Presented at the CORESTA Congress, Sapporo, Japan, September 23-27. Available: http://www.bat-science.com/ groupms/sites/BAT_7AWFH3.nsf/vwPagesWebLive/DO8YFDXA/$FILE/ CORESTA_SSPT17.pdf?openelement (last accessed 12.03.14).

St Charles, F.K., Krautter, G.R., Dixon, M., Mariner, D.C., 2006. A comparison of nicotine dose estimates in smokers between filter analysis, salivary cotinine, and urinary excretion of nicotine metabolites. Psychopharmacology 189, 345354.

St Charles, F.K., Ashley, M., Shepperd, C.J., Clayton, P., Errington, G., 2009. A robust method for estimating human smoked cigarette yields from filter analysis data. Beitr. Tabakforsch. Int. 23, 232-243.

St Charles, F.K., Kabbani, A.A., Borgerding, M.F., 2010. Estimating tar and nicotine exposure: human smoking versus machine generated smoke yields. Regul. Toxicol. Pharmacol. 56,100-110.

Zacny, J.P., Stitzer, M.I., Yingling, J.E., 1986. Cigarette filter vent blocking: effects of smoking topography and carbon monoxide exposure. Pharmacol. Biochem. Behav. 25, 1245-1252.