RISK ASSESSMENT OF DIOXINS IN CIGARETTE SMOKE
1,2
1
1
3
Teiichi Aoyama , Komichi Ikeda , Atsushi Takatori and Terry Obal
1 Environmental Research Institute Inc., Suite 4-1108, 4-5-26, Tokyo 141-0021, Japan
2 Faculty of Environmental and Information Studies, Musashi Institute of Technology, Yokohama
City, Kanagawa Prefecture, 224-0015, Japan
3 HRMS Services Division, Maxxam Analytics Inc., 50 Bathurst Drive, Unit 12, Waterloo, Ontario
N2V 2C5 Canada
Introduction
There is no doubt that smoking cigarettes is harmful to human health. If the dioxin concentrations
in cigarette smoke reach or exceed the carcinogenic levels suggested by the US EPA, and further
the toxicity levels reported by the World Health Organization (WHO), the detrimental effects of
cigarette smoking are much more serious. In this research, we have analyzed the dioxin
concentrations in 5 major brands of cigarettes from Japan and the United States. Dioxin, furan
and PCB congener concentrations were determined in the cigarette tobacco leaf, paper, ashes and
filters. We then estimated the amount of dioxin TEQ in the smoke of the cigarettes, as well as the
risk of human intake by smoking.
Methods and Materials
In order to research the risks associated with cigarette smoking, the analysis of dioxins in the
1,2
smoke is necessary. Based on our detailed review of previous work , we understand that,
technically, it is very difficult to extract a direct sample of smoke generated from smoking, such as
main and side-stream smoke. Therefore, conventional automatic smoking equipment (smoke
generator) was used. However, it should be noted that sample extraction by this automatic
equipment does not necessarily reflect actual human smoking conditions.
Further we have estimated the movement of the dioxin TEQ concentrations between the leaf of the
1,3,4
cigarette, paper, ashes and filter before and after smoking as previously researched . For instance,
3
4
Miyata pointed out that “Matsueda et al estimated in their research that the dioxins generated
by smoking is mostly contained in the main part (leaf and paper) of cigarette, and it vaporized at
the time of smoking, shifted to the body by inhaling the mainstream smoke via a filter and also
side stream smoke indirectly”. Subtracting the quantity of dioxins contained in the ashes and
filter after smoking from the quantity of dioxins contained in the main part of the cigarette (leaf
and paper) provides a close estimate of the dioxin concentration in the cigarette smoke.
1
Matsueda et al suggested in their research paper saying that even the mass balance of each parts of
the cigarette (main part, butt, filters, ash, mainstream and side stream) were not perfectly balanced,
it could be possible to substantiate that the dioxins in the mainstream could be derived from the
cigarette leaf and paper (main part). Even though there might be some generation and destruction
of dioxin in the process of combustion during smoking, the experimental data shows that dioxin
levels of cigarette leaf and paper (main part) were much higher compared to the total dioxin levels
of each part obtained by smoke generation experiment.
Therefore, the dioxin concentration in the smoke (d) can be calculated by subtracting the sum of
the dioxin concentrations in the filter after smoking (b) and in the residues/ash (c) from the
original concentrations determined in the leaf and paper before smoking (a), as follows:
(d) = (a) –[(b) + (c)]
Formula 1
As for the quantity of sample used per brand,
about 200 cigarettes were used for main parts,
60
i.e., leaf and paper, and about 160-370
50
Co-PCB
cigarettes were smoked to collect the ash and
40
PCDF
filter samples. All samples of ash and filter
30
PCDD
were collected from cigarettes smoked by
20
10
people in the office of Weekly Magazine. To
0
determine dioxin TEQ concentrations in this
MILD
Marlboro
KENT 1mg
LARK
Seven Stars
SEVEN
LIGHTS
MILDS
research, PCDDs, PCDFs and dioxin-like
PCBs (Co-PCBs) analyzed.
Maxxam
(Main Parts: Leaf and paper)
Analytics Inc. of Canada carried out the
6
analysis of dioxins. The analytical method
5
was based on EPA Method 8290B. Toluene
4
was used as the solvent for extracting the
Co-PCB
3
PCDF
main part of the cigarette samples, and
PCDD
2
hydrochloric acid was used to facilitate the
1
extraction of ashes and filters.
0
Results and Discussion
MILD
Marlboro
KENT
LARK
Seven
(1) Analysis results
SEVEN LIGHTS
1mg
MILDS
Stars
The analytical results for dioxins, furans and
(Ashes )
PCB congeners are shown in Figure 1. The
25
results are presented per one box (20
cigarettes).
20
(2) Verification by the Congener pattern
15
Co-PCB
The congener patterns of dioxins for the 3
PCDF
10
PCDD
different brands are shown in Figure 2. It
turns out that there is a close correlation
5
between the congener patterns in the main
0
parts (leaf and paper) of the cigarette and the
MILD
Marlboro
KENT 1mg
LARK
Seven Stars
SEVEN
LIGHTS
MILDS
filters by brand. It appears that the dioxins
in the main parts have moved to the filter as
(Filters after smoking)
estimated in the hypothesis. We also found
Fig.1 Dioxins in 1box of cigarette
that the congener patterns of the main parts
1,2
and the filter were also quite similar, as reported in previous research papers . Additionally, the
congener pattern in the smoke (mainstream and side stream) is also similar to the main parts and
1,2
filters . This correlation of patterns is also observed for the dioxin-like PCBs. However, the
quantity of dioxin-like PCBs captured by the filter was small compared to PCDDs/PCDFs.
pg-TEQ/20cigarettes
pg-TEQ/20cigarettes
pg-TEQ/20cigarettes
70
(3) Trial calculation of the amount of dioxins
In main and side stream smoke, Figure 3 shows the variation of dioxins levels (pg TEQ/20
cigarettes) before and after smoking. As for both Mild Seven and Marlboro LIGHTS, figure 3
shows that the filter removes about half of the dioxins in the leaf and paper parts. On the other
hand, KENT 1mg, Lark MILDS, and Seven Star showed significantly lower removal of dioxins by
the filter. In order to evaluate the health risk of smoking here, the total amount of the dioxins
contained in mainstream and side stream smoke was estimated.
MILD SEVEN - leaf and paper
0
10
20
30
40
50
MILD SEVEN - filter
60
pg/g
0
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
10
Marlboro LIGHTS - leaf and paper
0
10
20
30
40
50
60
pg/g
0
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
10
KENT 1mg - leaf and paper
10
20
30
40
40
50
60
pg/g
20
30
40
50
60
pg/g
50
60
pg/g
KENT 1mg - filter
50
60
pg/g
0
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
T4CDFs
P5CDFs
H6CDFs
H7CDFs
OCDF
Figure 2
30
Marlboro LIGHTS - filter
T4CDDs
P5CDDs
H6CDDs
H7CDDs
OCDD
0
20
10
20
30
40
Congeners Pattern of Dioxin Analogues in Cigarette Samples
Comparison of the Main parts and the Filters
(MILD SEVEN, Marlborough LIGHTS and KENT 1mg)
pg-TEQ/20 Cigarette (1 Package)
Table 1 is the estimated maximum
60
concentration of dioxins
in
the
50
mainstream and side stream smoke after
40
Filter
smoking one box of each brand of
30
Ash
cigarettes (20 cigarettes). The calculations
20
Tobocco
were carried out using Formula 1. As
10
shown in Table 1, the results for different
0
MILD
Marlboro
KENT
LARK
Seven
brands differ by a factor of approximately
SEVEN
Lights
1mg
MILDS
Star
7, with the lowest being Marlboro
LIGHTS (7pg-TEQ/box) and the highest
Figure 3 TEQ Levels before and after smoking
being KENT 1mg (48pg-TEQ/box).
(4) Estimation of the human health risk
Based on the estimated amount of dioxins in smoke, the daily intake of total dioxins from
mainstream and side stream smoke was calculated. The results are shown in Fig. 4. In the case of
KENT 1mg cigarettes, the concentration of dioxins in the mainstream and side stream smoke
(0.96pg-TEQ/bwt/day) approaches the minimum TDI value (1-4pg-TEQ/kg body weight/day)
proposed by WHO.
0.96
0.6
0.36
0.46
Seven
Star
LARK
MILDS
KENT
1mg
0.14
Marlboro
Lights
1.0
0.8
0.6
0.4
0.2
0.0
MILD
SEVEN
References
1. Matsueda, et al., (1991) PCDD/PCDF in
Smoking of Cigarette, Japan Society for
Atmospheric Environment, pp.475
2. Shiozaki, et al., (1998) PCDD/PCDF in
Tobacco, 7th, discussion meeting of Japan
Society for Environmental Chemistry,
3. Hideaki Miyata, (1998) Detection of Dioxin
in Cigarette, Japan Medical Journal, No.3860
4. Matsueda, Kuroki, Nakamura, et al., (1994)
Organohalogen Compounds, 20, pp.331
pg-TEQ/kg・day
Acknowledgments
There are still many unresolved Table 1 Amount of Dioxins contained in Mainstream
and Side Stream of Cigarette
issues concerning dioxins in
Unit: pg-TEQ /box (20 cigarettes)
cigarettes and cigarette smoke.
Main Parts Filter Ashes
Mainstream +
It is important to continue these
(a)
(b)
(c)
Side stream (d)
studies to better understand, and
43
22
3.3
18
be aware of the risks associated M ILD SEVEN
with smoking and exposure to M arlboro LIGHTS
16
7.3
1.7
7.0
58
6.2
3.8
48
cigarette smoke. We would like KENT 1mg
39
5.8
3.1
30
to express our gratitude to Mr. LARK M ILDS
Tanaka and Ms. Tsubata of Seven Star
32
3.4
5.2
23
Weekly Friday Magazine for
their extensive support of this research project including preparation of the samples.
Figure 4 Estimation of Human Risk