Abstract
We measured levels of nitrosonornicotine (NNN) and 4-[methyl(nitroso)amino]-1-(3-pyridinyl)-1-butanone (NNK), the two most carcinogenic tobacco-specific nitrosamines, in the filler, binder, and wrapper of 50 cigars: 19 large cigars, 23 cigarillos, and 8 little cigars. The average NNN and NNK levels were 10.6 μg/g and 3.70 μg/g, respectively. These levels are five- and seven-fold higher, respectively, than those of commercial cigarettes. The differences in NNN and NNK levels between cigars and cigarettes reflect differences in tobacco blends and tobacco treatments, such as fermentation. The average tobacco NNN and NNK levels of large cigars were three- and five-fold higher than those of cigarillos and little cigars, respectively. Large cigars also exhibited a significantly broader range of NNN and NNK than cigarillos and little cigars. The NNN and NNK levels in cigarillos are comparable to those of little cigars. These results are consistent with earlier studies finding that cigarillos and little cigars have similar tobacco blends with lower NNN and NNK content than large cigar tobacco blends.
Graphical Abstract

1. Introduction
The levels of tobacco-specific nitrosamines (TSNAs) present in tobacco products vary depending on tobacco type, soil conditions, crop year weather, harvest and curing practices (e.g., fermentation), and storage time and storage conditions.1–4 The two most carcinogenic TSNAs in tobacco products, N-nitrosonornicotine (NNN) and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, or nicotine-derived nitrosamine ketone (NNK), are common in the contents and emissions of tobacco products.1–3 Previous studies have shown that some tobacco types such as burley and dark tobacco have higher levels of TSNAs than other tobacco types such as bright and oriental.4, 5
American-blended cigarettes are made from a combination of bright, burley, oriental, and reconstituted tobacco. In contrast, many cigars contain air-cured and fermented dark tobacco.6, 7 Air curing and the fermentation process lead to increased nitrosation of the native tobacco alkaloids through a variety of mechanisms to produce TSNAs.4, 8, 9 Blend differences impact chemical constituent levels in the products, resulting in cigars having higher TSNA levels relative to cigarettes. Unlike cigarettes, cigars are manufactured in a wide range of lengths, diameters, and weights and are therefore often grouped into different types. Typically cigars tend to fall into three different types:6 little cigars, cigarillos, and large cigars (including premium cigars). The tobacco blends of these cigar types may differ widely. For example, the filler, binder, and wrapper of premium cigars are composed of air-cured fermented tobacco. The binder and wrapper of machine-made cigars are reconstituted tobacco sheets, which may not be made from the same tobacco type as premium cigars.6 Some cigarillos even use pipe tobacco as their filler, as evidenced by the marketing for several cigarillo brands included in this study. 10 Therefore, the TSNA levels in various cigar types are expected to be different from each other and from other tobacco products such as cigarettes.
Previous published reports of cigar TSNA levels are limited in number and most of these studies were conducted prior to or during the 1980s. In addition, there are no recent cigar TSNA studies that cover a large portion of the U.S. cigar market. TSNA data for the current U.S. cigar market would be helpful in the evaluation of the health impacts of these products. Recently, a Food and Drug Administration (FDA) and Centers for Disease Control and Prevention (CDC) collaborative study investigated harmful and potentially harmful constituents (HPHCs) in the tobacco and smoke of 60 commercial little cigars on the U.S. market.11 This study expands on the previous work and measures TSNA levels in 50 cigars (including 19 large cigars, 23 cigarillos, and 8 little cigars) and compares the data with previously reported TSNA levels for little cigars and cigarettes.
2. Experimental Procedures
We purchased 50 cigars from select retail outlets in the greater Atlanta area through Lab Depot (Dawsonville, GA). Cigar selection was informed by 2012–2013 Nielsen market data and included products with higher and lower market share. We organized the cigars based on market descriptions into three types: little cigars (8), cigarillos (23), and large cigars (19). Ground cigar filler ranging in mass from 0.01–0.1 gram (g) was measured using an analytical balance, placed in amber vials, and spiked with 100 nanograms (ng) of 13C6 labeled NNN and NNK internal standards and 10 milliliters (mL) of 100 millimolar (mM) ammonium acetate buffer. The samples were shaken for one hour on an Incu-Shaker 10L (Benchmark Scientific, Edison, New Jersey) mechanical shaker at 240 revolutions per minute. A portion of the slurry was filtered into high-performance liquid chromatography (HPLC) vials. We conducted analytical measurements using HPLC coupled with tandem mass spectrometry (MS-MS). Analytical separation was achieved using the Agilent (Santa Clara, CA) 1200 series HPLC system with an XBridge BEH C18 50×4.6mm 5μm particle size Waters (Milford, MA) HPLC Column with 5mM ammonium acetate solution for the aqueous phase and 5mM ammonium acetate in a mixture of 95% acetonitrile and 5% water for the organic phase. The chromatographic run time was eight minutes using a gradient of the organic phase between 5 and 35 percent. A Sciex (Framingham, MA) API-4000 tandem mass spectrometer equipped with an electrospray ion source was used for quantitative detection. The ionization polarity was positive, and the mass spectrometer was working in multiple reaction monitoring mode. Measurements were performed in septuplicates (n = 7). The analytical data was analyzed using Sciex’s Analyst 1.5 software using 1/X weighed least squares for fitting the calibration curves. We uploaded the quantitative results to our laboratory information system’s database for statistical quality control analysis and review prior to reporting.
3. Results
For the 50 cigar brands analyzed, the TSNA ranges were 0.661–54.0 μg/g for NNK and 2.75–80.5 μg/g for NNN. The average levels were 3.70 μg/g for NNK and 10.6 μg/g for NNN. Table 1 summarizes the measured levels for NNN and NNK for each brand. Several notable differences were observed across the three cigar types. Table 2 lists the ranges and variability of TSNA levels in each of the three cigar types along with the average values. Findings for each of the types are discussed below. Lang and Vuarnoz noted that matrix-bound NNK is present in tobacco filler.12 This NNK was not extractable using the standard sampling methods at room temperature but could be released during smoking. We note this as a possible limitation where the levels in tobacco filler might not directly correlate to levels in the mainstream smoke. This study uses standard sampling methods that may not extract matrix-bound NNK.
Table 1.
Measured NNK and NNN Levels in 50 Cigars1
| Product Brand | Manufacturer | Flavor | NNK (μg/g) | NNN (μg/g) |
|---|---|---|---|---|
| Little cigars | ||||
| Cheyenne | Cheyenne International | Classic | 1.43 | 6.25 |
| Cheyenne | Cheyenne International | Full Flavor | 1.52 | 6.57 |
| Gold & Mild | John Middleton Inc. | Not Specified | 1.59 | 4.66 |
| Cheyenne | Cheyenne International | Sweet Mint | 1.57 | 5.96 |
| Red Buck | Tantus Tobacco | Blueberry | 0.995 | 5.10 |
| Red Buck | Tantus Tobacco | Cherry | 1.02 | 4.62 |
| Red Buck | Tantus Tobacco | Red | 1.29 | 5.51 |
| Red Buck | Tantus Tobacco | Menthol | 1.33 | 4.59 |
| Cigarillos | ||||
| Black & Mild | John Middleton Inc. | Regular | 1.22 | 3.39 |
| BlackStone | Swisher International, Inc. | Peach | 1.23 | 4.17 |
| BlackStone | Swisher International, Inc. | Vanilla | 1.34 | 3.2 |
| BLK | Swisher International, Inc. | Cherry | 1.94 | 6.4 |
| White Owl | White Owl | Grape | 1.27 | 7.06 |
| White Owl | White Owl | Green Sweets | 1.25 | 5.03 |
| Romeo y Julieta | Unknown | Not Specified | 0.985 | 4.54 |
| Dutch Masters | Dominican Republic | wine | 1.16 | 4.66 |
| Dutch Masters | Dominican Republic | Palma | 2.34 | 12.4 |
| Dutch | Dominican Republic | Berry Fusion | 2.90 | 16.7 |
| Swisher Sweets | Swisher International, Inc. | Black | 1.14 | 3.56 |
| Black & Mild | John Middleton Inc. | Apple | 1.24 | 3.1 |
| Swisher Sweets | Swisher International, Inc. | Pipe tobacco | 1.18 | 2.75 |
| Swisher Sweets | Swisher International, Inc. | Not Specified | 0.850 | 4.18 |
| Game by Garcia y Vega | Garcia y Vega | Black Sweets | 1.28 | 5.49 |
| Game by Garcia y Vega | Garcia y Vega | Blue | 0.981 | 5.31 |
| Game by Garcia y Vega | Garcia y Vega | Grape | 1.71 | 5.59 |
| Game by Garcia y Vega | Garcia y Vega | Pineapple | 0.851 | 4.26 |
| Zig Zag | National Tobacco Company | Straight Up | 1.19 | 6.44 |
| Royal Comfort | John Middleton Inc. | Tropical | 1.21 | 5.38 |
| Romeo y Julieta | Thompson & Co | N/A | 0.661 | 6.64 |
| BlackStone | Swisher International, Inc. | Cherry | 1.50 | 4.06 |
| Al Capone | Inter-Continental Cigar Corporation | Sweets Cognac Dipped | 1.37 | 7.40 |
| Large Cigars | ||||
| Phillies | Altadis | Not Specified | 1.48 | 6.47 |
| Swisher Sweets | Swisher International, Inc. | Sweet | 2.50 | 9.63 |
| Phillies | Phillies | Not Specified | 2.51 | 12.9 |
| Davidoff Classic Primeros | Davidoff | Not Specified | 3.19 | 6.24 |
| Macanudo Court | Macanudo | Not Specified | 2.37 | 4.72 |
| Acid Toast | Acid Cigars | Not Specified | 7.30 | 10.6 |
| Antonios | Thompson & Co | Not Specified | 4.35 | 12.7 |
| Dutch Masters | Dominican Republic | Grape | 2.28 | 10.6 |
| Swisher Sweets | Swisher International, Inc. | Giants | 2.55 | 7.57 |
| Swisher Sweets | Swisher International, Inc. | Natural | 2.37 | 10.3 |
| Havana Honeys | Scandinavian Tobacco Group Lane LTD | Spiced Rum | 23.5 | 46.9 |
| Havana Honeys | Scandinavian Tobacco Group Lane LTD | Original | 54.0 | 80.5 |
| Havana Honeys | Scandinavian Tobacco Group Lane LTD | Sweet Honey | 15.8 | 65.6 |
| Phillies | Altadis | Original | 1.95 | 9.83 |
| Phillies | Altadis | Strawberry | 3.50 | 14.5 |
| White Owl | Swedish Match | White Grape | 1.40 | 4.84 |
| Sevillas | Thompson & Co | N/A | 8.77 | 33.5 |
| Criollo | Thompson & Co | Cao | 1.94 | 4.64 |
| King Edward Imperial | Swisher International, Inc. | N/A | 1.61 | 3.36 |
| Average | 3.70 | 10.6 | ||
| Range | 0.661–54.00 | 2.750–80.50 | ||
| STDEV | 8.227 | 14.958 | ||
| %RSD | 222% | 141% | ||
Table 2.
Comparison of TSNA Levels in Different Cigar Types in the Current Study and with Previously Published Data 11, 13
| Product Type | Product number | TSNA | TSNA Range (μg/g) | Average (μg/g) | Variation (STDEV) (μg/g) | % RSD |
|---|---|---|---|---|---|---|
| Little cigars2 | 8 | NNK | 0.995–1.59 | 1.34 | 0.233 | 17.3 |
| NNN | 4.59–6.57 | 5.41 | 0.786 | 14.5 | ||
| Cigarillos2 | 23 | NNK | 0.661–2.90 | 1.34 | 0.493 | 36.8 |
| NNN | 2.75–16.7 | 5.73 | 3.14 | 54.8 | ||
| Large cigars2 | 19 | NNK | 1.40–54.00 | 7.55 | 12.6 | 167 |
| NNN | 3.36–80.5 | 18.7 | 22.0 | 118 | ||
| Little cigar 6011 | 60 | NNK | 0.0260–2.95 | 1.12 | 0.605 | 54.1 |
| NNN | 1.44–12.1 | 4.71 | 1.70 | 35.1 | ||
| Cigarette 5013 | 50 | NNK | 0.194–1.09 | 0.523 | 0.158 | 30.2 |
| NNN | 0.306–2.97 | 1.90 | 0.359 | 18.9 |
Large cigars
The NNK levels in the tested large cigars had a range and average of 1.40–54.0 μg/g and 7.55 μg/g, respectively; NNN levels had a range and average of 3.36–80.5 μg/g and 18.7 μg/g, respectively. Large cigars had the widest ranges and highest average levels for both NNN and NNK compared to cigarillos and little cigars (discussed below). The variability of TSNA levels in large cigars is evidenced by the large standard deviation (STDEV) and relative standard deviation (%RSD) of the mean TSNA levels. For these cigars, the STDEV and %RSD were 12.6 μg/g and 167% for NNK and 22.0 μg/g and 118% for NNN, respectively. The brand Havana Honeys Original had the highest TSNA levels (54.0 μg/g of NNK and 80.5 μg/g of NNN); White Owl White Grape had the lowest NNK level (1.40 μg/g) and King Edward Imperial had the lowest NNN level (3.36 μg/g). The average TSNA levels of the large cigars only were much higher than the averages of the 50 tested cigars for both NNK (7.55 vs. 3.70 μg/g) and NNN (18.7 vs. 10.6 μg/g).
Cigarillos
For cigarillos, NNK levels had a range and average of 0.661–2.90 μg/g and 1.34 μg/g, respectively; NNN levels had a range and average of 2.75–16.7 μg/g and 5.73 μg/g, respectively. The average TSNA levels were significantly lower than those of large cigars (1.34 vs. 7.55 μg/g for NNK with p=0.0378 and 5.73 vs. 18.7 μg/g for NNN with p=0.0147 using the two one-sided equivalence test [TOST]); the ranges of NNK and NNN levels in cigarillos were also much narrower. As shown in Table 2, the STDEV and %RSD for cigarillos were 0.493 μg/g and 36.8% for NNK and 3.14 μg/g and 54.8% for NNN, respectively. Dutch Berry Fusion had the highest TSNA levels among tested cigarillos: 2.90 μg/g of NNK and 16.7 μg/g of NNN. However, the large cigar with the highest TSNA levels (Havana Honeys Original) had approximately 19-fold higher NNK levels and five-fold higher NNN levels compared to Dutch Berry Fusion. Another cigarillo, Dutch Masters Palma, also had notably high levels of NNK (2.34 μg/g) and NNN (12.4 μg/g) among the 23 tested cigarillos. Excluding these two cigarillos, the average TSNA levels of the other 21 cigarillos were 1.22 μg/g (range: 0.661–1.94 μg/g; %RSD: 23%) for NNK and 4.89 μg/g (range: 2.75–7.40 μg/g; %RSD: 28%) for NNN.
Seven cigarillos in this study were manufactured by Swisher International, Inc. The average TSNA levels of these cigarillos were 1.31 μg/g (range: 0.85–1.94 μg/g; %RSD: 26%) for NNK and 4.05 μg/g (range: 2.75–6.40 μg/g; %RSD: 29%) for NNN. The variability evidenced by the %RSD is comparable to that of all cigarillos excluding Dutch Masters Palma and Dutch Berry Fusion. Seven cigarillos from two manufacturers (John Middleton, Inc. and Swisher International, Inc.) were marketed as pipe tobacco cigarillos. Average NNK levels were comparable between the pipe tobacco cigarillos (1.38 μg/g) and regular cigarillos (1.32 μg/g). The average NNN level from the 16 regular cigarillos was 6.54 μg/g, which is 69% higher than the average NNN level of the seven pipe tobacco cigarillos (3.87 μg/g). The variability of NNK and NNN levels among the 16 regular cigarillos was higher than that of the seven pipe tobacco cigarillos. This is likely because the number of products and manufacturers of regular cigarillos is much larger than that of the pipe tobacco cigarillos.
Little cigars
For the eight little cigars, the NNK levels ranged from 0.995–1.59 μg/g with an average value of 1.34 μg/g. The NNN levels ranged from 4.59–6.57 μg/g with an average value of 5.41 μg/g. The average levels of NNK and NNN of the tested little cigars were statistically similar to those of cigarillos (p=0.976 and p=0.657, respectively) using the TOST. However, the TSNA ranges of the little cigars were narrower than those of cigarillos, possibly due to the lower number of tested samples (8 vs. 23). As shown in Table 2, the TSNA level variability among the eight little cigars (17.3% for NNK and 14.5% for NNN) was lower than that among the 23 cigarillos (36.8% for NNK and 54.8% for NNN). This difference may also be due to the different sample sizes for these two cigar types.
While the average TSNA levels of little cigars and cigarillos were very similar, the NNK and NNN levels in little cigars were much lower than those of large cigars. Large cigars had the highest average TSNA levels -- 462% higher NNK and 246% higher NNN than little cigars and 464% higher NNK and 227% higher NNN than cigarillos. Large cigars also had the widest TSNA range of all the cigar types.
4. Discussion
Cigar tobacco, usually dark tobacco, is typically fermented after curing and generally contains higher levels of TSNAs than other tobacco types, including flue-cured tobacco and American blend cigarette tobacco.6 However, fermented cigar tobacco may not be the only tobacco type in cigars, especially in little cigars and cigarillos. Air-cured burley tobacco, pipe tobacco, and reconstituted tobacco may be also used in the blend of cigarillos and little cigars. The marketing of pipe tobacco cigarillos supports the potential tobacco blend variety of these cigars. Furthermore, machine-made large cigars use reconstituted tobacco sheets as binders and wrappers.6 Tobacco in large cigars is expected to be predominantly air-cured and then aged and fermented. Tobacco types other than air-cured and fermented cigar tobacco usually contain lower levels of TSNAs.4 Therefore, the TSNA levels in little cigars and cigarillos are expected to be lower than those in large cigars.
An early study14 reported that a filtered little cigar contained 45 μg/g of NNN and 35 μg/g of NNK and a large cigar contained 10.7 μg/g of NNN and 1.1 μg/g of NNK. These cigar TSNA levels were higher than those measured in a commercial filtered cigarette.16–17 As noted above, large cigars had the highest TSNA levels and variability. Compared to cigarillos and little cigars, the average NNK and NNN levels in large cigars are 462–464% and 227–246% higher, respectively. In addition, the TSNA level variability, represented by %RSD, of large cigars tested in this study is also the largest of the three types. For NNK, the %RSD is 167%, which is more than four-fold that of cigarillos. For NNN, the %RSD is 118%, which is twice that of cigarillos. Among the examined large cigars, three Havana Honeys flavor varieties from Scandinavian Tobacco Group Lane LTD have very high levels of TSNAs (Original: NNK 54.0 μg/g, NNN 80.5 μg/g; Spiced Rum: NNK 23.5 μg/g, NNN 46.9 μg/g; Sweet Honey: NNK 15.8 μg/g, NNN 65.6 μg/g). Excluding these three products, the average TSNA levels of the remaining 16 large cigars were 3.129 μg/g (range: 1.40–8.77 μg/g; % RSD: 66%) for NNK and 10.1 μg/g (range: 3.36–33.5 μg/g; % RSD: 70%) for NNN. These values, while lower, are still significantly higher with greater %RSD than cigarillos (134% higher NNK and 77% higher NNN) and little cigars (133% higher NNK and 88% higher NNN). This suggests that there is greater tobacco variation in large cigars due to tobacco varieties, different growing locations, and differences in curing, aging and fermentation compared to the tobacco used in little cigars and cigarillos.
As previously stated and shown in Table 2, the average TSNA levels in the 23 cigarillos measured in this study were 1.34 μg/g for NNK and 5.73 μg/g for NNN, which were very similar to those of little cigars (1.34 μg/g for NNK and 5.41 μg/g for NNN). However, cigarillos included in this study had larger NNK and NNN variability (%RSD) compared to little cigars, although not as great as the large cigars. The %RSD differences between cigarillos and little cigars may be due to different sample sizes and sample selection, since the number of tested cigarillos is about three-fold that of little cigars. This suggests that similar tobaccos (e.g., type, processing methods) are likely used in little cigars and cigarillos. Ingredient data submitted to FDA pursuant to §904(a)(1) of the Federal Food, Drug, and Cosmetic Act show that the majority of reported little cigars and cigarillos have blended tobacco fillers composed of dark air-cured, light air-cured, burley, Virginia, oriental and Maryland tobacco. In addition, some little cigars and cigarillos have blends primarily composed of Virginia tobacco (up to 82% of the blend). Although the submitted data shows a variety of tobacco types and blends in little cigars and cigarillos, the variation is very similar between these two cigar types. The submissions also included data for large cigars, which showed that large cigars predominantly contained dark air-cured tobacco without additives other than adhesives. The data for large cigars, compared to data for little cigars and cigarillos, show notable tobacco blend differences, which may account for the TSNA differences observed in this study. These data, submitted pursuant to §904(a)(1) of the Federal Food, Drug, and Cosmetic Act, correspond with previously published literature regarding the tobacco blends of these cigar types.6, 10 Based on the product labels, seven cigarillos are made with pipe tobaccos, which are blends possibly containing flue-cured bright and air-cured burley15 as the main components and may not include fermented tobacco. Although the average NNK level in these seven cigarillos was similar to that of the 16 regular cigarillos (1.38 μg/g vs. 1.32 μg/g), the average NNN level was significantly lower than that of traditional cigarillos (3.87 μg/g vs. 6.54 μg/g). This difference may also reflect the tobacco type differences in pipe tobacco cigarillos and regular cigarillos.
As seen in Table 2 and Figure 1, the average TSNA levels for the eight little cigars in this study were 1.34 μg/g for NNK and 5.41 μg/g for NNN. The average TSNA levels in 60 little cigars on the U.S. market in 2015–2016 measured in a recent FDA/CDC collaborative study are also listed in Table 2 as “Little cigar 60.”11 The average TSNA levels of the eight little cigars in this study compared to the 60 little cigars in the FDA/CDC study are slightly higher but comparable: 1.34 μg/g vs. 1.12 μg/g for NNK and 5.41 μg/g vs. 4.71 μg/g for NNN, respectively. The ranges of TSNA levels of the eight little cigars are within those of the 60 little cigars in the FDA/CDC study (Table 2). The variabilities of TSNAs presented as %RSD for the 60 little cigars are 54.1% for NNK and 35.1% for NNN, which are larger than those of the eight tested little cigars in this study, likely due to the sample size difference (60 vs. 8). In addition, the TSNA variabilities of the 60 little cigars are more similar to those of the 23 tested cigarillos compared to the large cigars in this study. This observation further supports the apparent similarity of little cigar and cigarillo tobaccos. Early studies 6, 16 showed that NNN levels in cigar tobacco were 3.0–10.7 μg/g, which is close to the NNN level found in large cigars in this study. However, in these early studies, the NNN level in little cigars was 11.1–13.0 μg/g, which is much higher than that of little cigars in this study and also much higher than the NNN level in the 60 little cigars in the FDA/CDC study. The higher NNN levels in early studies compared to those in the current study and the FDA/CDC study indicate that NNN levels in little cigars have declined over the past 30 years.
Figure 1.

Average NNK and NNN levels in little cigars, cigarillos, large cigars, and cigarettes with 95% confidence levels.
Although published studies of TSNAs in cigars are limited, the TSNA levels in cigarettes have been thoroughly investigated. An early study measured TSNA ranges in more than 170 commercial cigarettes from 14 countries. The measured ranges in cigarettes were 0.045–12.454 μg/cigarette for NNN and less than 0.050–10.745 μg/cigarette for NNK.17 Ashley et al.18 measured NNK and NNN in cigarettes from 14 countries and compared the TSNA levels of the locally purchased U.S. brand Marlboro to the most popular local brand in each country. The total TSNA (NNK+NNN) level ranged from 0.087 μg/g to 1.9 μg/g. The average TSNA levels in Marlboro tobacco were significantly higher than the average TSNA levels in the non-U.S. brands, regardless of where the Marlboro products were manufactured.
A more recent collaborative FDA/CDC study of 50 U.S. cigarette brands on the U.S. market in 2011–2012 found that TSNA ranges in cigarettes were 0.194–1.09 μg/g for NNK and 0.306–2.97 μg/g for NNN.13 The average NNK and NNN levels of these 50 cigarettes (referred to as Cigarette 50) along with the range and variability (%RSD) are also listed in Table 2. The Cigarette 50 average TSNA levels in cigarette filler were 0.523 μg/g for NNK and 1.90 μg/g for NNN.
In our study, the TSNA levels in little cigars and cigarillos were more than two-fold those in cigarettes with p values <0.01 for both NNK and NNN based on the TOST. For large cigars measured in this study, the average TSNA levels were approximately 14-fold higher for NNK and 10-fold higher for NNN than those in cigarettes on a per gram basis (p=0.0178 for NNK and p=0.00143 for NNN based on the TOST). In addition, cigars of all types usually contain more tobacco than cigarettes. Therefore, the TSNA differences between cigars and cigarettes are even greater when compared on a per unit of use basis. Although the tobacco filler in little cigars and cigarillos may be a blend of different tobacco types, it is predominantly air-cured tobacco.6 Compared to cigarettes, which commonly are made using a tobacco blend with a large portion of low TSNA bright tobacco,4, 19 the TSNA levels in little cigars and cigarillos are significantly higher than those in cigarette filler. The TSNA variability presented as %RSD for 60 little cigars is about 1.8-fold that of the Cigarette 50. This implies that tobacco variability for little cigars is much larger than that for the tobacco blends in cigarettes.
Several studies have investigated the correlation of TSNA levels with other chemicals in tobacco. Several studies found a positive correlation between TSNA and nitrate levels but no correlation between TSNA and nicotine levels in cigarettes.17, 20, 21 However, these studies did not measure nitrite levels nor discuss any correlation between nitrite and TSNAs. Other studies focusing on tobacco curing and aging processes found that TSNA levels were correlated with nitrite levels, not nitrate levels, in tobacco leaves.4, 8, 22–25 These studies also found no direct correlation between nicotine and NNN or NNK. However, these research results are not contradictory since the formation of TSNAs primarily occurs during tobacco curing and aging processes via the reaction of nitrite with tobacco alkaloids. A cigarette with a higher nitrate level is likely to have a higher level of nitrite, and therefore higher TSNA levels, compared to a cigarette with a lower nitrate level. The nitrate and nitrite levels were not measured in the 50 cigars in our study. However, the FDA/CDC study of 50 cigarettes found no correlation between the nitrate and TSNA levels in tobacco fillers. This observation is different from those of Fischer et al.17, 21 and Brunemann et al.20 The TSNA levels in cigars measured in our study have no correlation with nicotine, which is in agreement with previously reported cigar data from the FDA/CDC study. In addition, cigar TSNA levels are not correlated to ammonia levels but have weak positive correlations with tobacco pH, especially in the large cigars in our study.
Although TSNA levels in tobacco smoke result from several factors1, 26 and smoke data allows the most complete evaluation, the TSNAs in cigarette filler are the primary contributor to TSNA levels in cigarette smoke. The results of this study show that the average TSNA levels in select little cigars and cigarillos are approximately three-fold higher than those of cigarettes, while the average TSNA levels in large cigars are nine- to 15-fold higher than those of cigarettes. All these comparisons are based on TSNA levels per gram of tobacco. In general, cigar tobacco quantities are much higher than those of cigarettes. The previous collaborative study on 60 little cigars11 showed that average NNK transfer rates are 37% (under the non-intense International Organization for Standardization [ISO] regimen) and 51% (under the Canadian Intense [CI] regimen), and the NNN transfer rates are 10% (ISO) and 18% (CI) -- all of which were greater (by 3–21%) than the corresponding cigarette transfer rates. Assuming that TSNA transfer rates from tobacco to mainstream smoke are similar for large cigars and cigarillos, the TSNA exposure of cigar smokers could be tens- or even hundreds-fold higher compared to that of cigarette smokers assuming similar smoking habits. However, some cigars may be smoked differently (e.g., not inhaled) than cigarettes, which may impact cigar smoker TSNA exposure. There has been very limited study of the machine-measured TSNA smoke yields for cigars. This study provides data on TSNAs in the tobacco of different cigar types, which suggests the potential magnitude of TSNA smoke yield differences between the cigar types and cigarettes and, therefore, potential user exposure differences.
5. Conclusion
This study provides new TSNA data for cigars, in particular large cigars and cigarillos, that represent a majority of the U.S. cigar marketplace. In particular, this study provides information on TSNA levels in different cigar types as well as comparisons between cigars and cigarettes. However, this study is not a comprehensive survey, as the studied products were all purchased in the Atlanta, GA metropolitan region based on 2012–2013 Nielsen market data. In addition, limited information was available for the cigar tobacco blends, ingredients, and other manufacturing details.
The data in this study show that among the different cigar types, large cigars have the highest average TSNA level compared to little cigars and cigarillos, which have comparable TSNA levels. In addition, the TSNA variability in large cigars is also much larger than that of cigarillos and little cigars, suggesting a larger variety of cigar tobaccos, tobacco processing and tobacco sources. Because cigarettes have been more widely studied, comparisons are important in order to help evaluate the relative public health impact of cigars. The average TSNA levels of the 50 cigars are more than five-fold higher than the average cigarette TSNA levels reported in a recent study.13 The higher levels of TSNAs in cigars as well as previously reported higher TSNA transfer rates to smoke may result in higher TSNA exposure for cigar smokers than cigarette smokers.
Funding
This work was supported by the U.S. Food and Drug Administration, Center for Tobacco Products (InterAgency Agreement # 224-10-9022).
Footnotes
NNK and NNN levels in Table 1 are the mean of seven replicate measurements.
Data from the current study.
Disclaimer
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the U.S. Department of Health and Human Services or any of its affiliated institutions or agencies. The use of brand names in this manuscript does not constitute an endorsement by either the FDA or CDC.
6. References
- (1).Fischer S, Spiegelhalder B, Eisenbarth J, Preussmann R (1990) Investigations on the origin of tobacco-specific nitrosamines in mainstream smoke of cigarettes. Carcinogenesis 11 (5), 723–730. [DOI] [PubMed] [Google Scholar]
- (2).Hecht SS, Hoffmann D (1988) Tobacco-specific nitrosamines, an important group of carcinogens in tobacco and tobacco smoke. Carcinogenesis. 9 (6), 875–884. [DOI] [PubMed] [Google Scholar]
- (3).Konstantinou E, Fotopoulou F, Drosos A, Dimakopoulou N, Zagoriti Z, Niarchos A (2018) Tobacco-specific nitrosamines: A literature review. Food and Chemical Toxicology. 118, 198–203. [DOI] [PubMed] [Google Scholar]
- (4).Burton HR, Dye NK, Bush LP (1994) Relationship between Tobacco-Specific Nitrosamines and Nitrite from Different Air-Cured Tobacco Varieties. Journal of Agricultural and Food Chemistry. 42, 2007–2011. [Google Scholar]
- (5).Ding YS, Zhang L, Jain RB, Jain N, Wang RY, Ashley DL, Watson CH (2008) Levels of Tobacco-Specific Nitrosamines and Polycyclic Aromatic Hydrocarbons in Mainstream Smoke from Different Tobacco Varieties. Cancer Epidemiology, Biomarkers & Prevention. 17 (12), 3366–3371. [DOI] [PubMed] [Google Scholar]
- (6).Hoffmann D, Hoffmann I Chemistry and Toxicology. In Smoking and Tobacco Control Monograph 9, Cigars: Health Effects and Trends, Burns DM Ed.; 1998; pp 55–104. [Google Scholar]
- (7).Sifola MI, Cozzolino E, Di Mola I, Ottaiano L, del Piano L, Mori M (2022) Yield and Quality of Three Cultivars of Dark Fire-Cured (Kentucky) Tobacco (Nicotiana tabacum L.) Subjected to Organic (Compost) and Mineral Nitrogen Fertilization. Agronomy. 12 (2), 483. [Google Scholar]
- (8).Burton HR, Childs GH, Andersen RA, Fleming PD (1989) Changes in Chemical Composition of Burley Tobacco during Senescence and Curing. 3. Tobacco-Specific Nitrosamines. Journal of Agricultural and Food Chemistry. 37, 426–430. [Google Scholar]
- (9).Chamberlain WJ, Chortyk OT (1992) Effects of Curing and Fertilization on Nitrosamine Formation in Bright and Burley Tobacco. Beitrage zur Tabakforschung International. 15 (2), 87–92. [Google Scholar]
- (10).Lawler TS, Stanfill SB, deCastro BR, Lisko JG, Duncan BW, Pichter P, Watson CH (2017) Surveillance of Nicotine and pH in Cigarette and Cigar Filler. Tobacco Regulatory Science. 3 (Suppl 1), 101–116. DOI: 0.18001/TRS.3.2(Suppl1).11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (11).Edwards SH, Hassink MD, Taylor KM, Watson CH, Kuklenyik P (2021) Tobacco-Specific Nitrosamines in the Tobacco and Mainstream Smoke of Commercial Little Cigars. Chemical Research in Toxicology. 34 (4), 1034–1045. [DOI] [PubMed] [Google Scholar]
- (12).Lang G, Vuarnoz A (2015) Matrix-Bound 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone in Tobacco: Quantification and Evidence for an Origin from Lignin-Incorporated Alkaloids. Journal of Natural Products. 78, 85–92. [DOI] [PubMed] [Google Scholar]
- (13).Edwards SH, Rossiter LM, Taylor KM, Holman MR, Zhang L, Ding YS, Watson CH (2017) Tobacco-specific nitrosamines in the tobacco and mainstream smoke of U.S. Commercial Cigarettes. Chemical Research in Toxicology. 30, 540–551. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (14).Hoffmann D, Adams JD, Brunnemann KD, Hecht SS (1979) Assessment of Tobacco-specific N-Nitrosamines in Tobacco Products. Cancer Research. 39 (7), 2505–2509. [PubMed] [Google Scholar]
- (15).Besnoke U What Is Pipe Tobacco? Different Varietals, Blends, Cuts & Their Resulting Flavours / Smoking Experience. https://bespokeunit.com/pipes/ (accessed 2020 5/13/2020).
- (16).Hecht SS, Chen CB, Hoffmann D (1979) Tobacco-Specific Nitrosamines: Occurrence, Formation, Carcinogenicity, and Metabolism. Accounts of Chemical Research. 12, 92–98. [Google Scholar]
- (17).Fischer S, Spiegelhalder B, Preussmann R (1990) Tobacco-specific nitrosamines in European and USA cigarettes. Archiv fur Geschwulstforschung. 60 (3), 169–177. [PubMed] [Google Scholar]
- (18).Ashley DL, Beeson MD, Johnson DR, McCraw JM, Richter P, Pirkle JL, Pechacek TF, Song S, Watson CH (2003) Tobacco-specific nitrosamines in tobacco from U.S. brand and non-U.S. brand cigarettes. Nicotine & Tobacco Research. 5, 323–331. [DOI] [PubMed] [Google Scholar]
- (19).Agnew-Heard KA, Walters MJ, Holman MR (2016) Multivariate statistical analysis of cigarette design feature influence on ISO TNCO yields. Chemical research in toxicology. 29 (6), 1051–1063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (20).Brunnemann KD, Masaryk J, Hoffmann D (1983) Role of Tobacco Stems in the Formation of iV-Nitrosamines in Tobacco and Cigarette Mainstream and Sidestream Smoke. Journal of Agricultural and Food Chemistry. 31, 1221–1224. [Google Scholar]
- (21).Fischer S, Spiegelhalder B, Preussmann R (1989) Preformed tobacco-specific nitrosamines in tobacco—role of nitrate and influence of tobacco type. Carcinogenesis. 10 (8), 1511–1517. [DOI] [PubMed] [Google Scholar]
- (22).Burton HR, Bush LP, Djordjevic MV (1989) Influence of Temperature and Humidity on the Accumulation of Tobacco-Specific Nitrosamines in Stored Burley Tobacco. Journal of Agricultural and Food Chemistry. 37, 1372–1377. [Google Scholar]
- (23).Djordjevic MV, Gay SL, Bush LP, Chaplin JF (1989) Tobacco-Specific Nitrosamine Accumulation and Distribution in Flue-Cured Tobacco Alkaloid Isolines. Journal of Agricultural and Food Chemistry. 37, 752–756. [Google Scholar]
- (24).Shi HZ, Wang RY, Bush LP, Zhou J, Yang HJ, Fannin N, Bai v. S. (2013) Changes in TSNA Contents during Tobacco Storage and the Effect of Temperature and Nitrate Level on TSNA Formation. Journal of Agricultural and Food Chemistry. 61, 11588–11594. [DOI] [PubMed] [Google Scholar]
- (25).Wiernik A, Christakopoulos A, Johansson L, Wahlberg I (1995) Effect of Air-curing on the Chemical Composition of Tobacco. Recent Advances in Tobacco Science. 21, 39–80. [Google Scholar]
- (26).Adams JD, Lee SJ, Vinchkoski N, Castonguay A, Hoffmann D (1983) On the Formation of the Tobacco-Specific Carcinogen 4-(Methylnitrosamino)-1-(3-Pyridyl)-1-Butanone During Smoking. Cancer Letters. 17, 339–346. [DOI] [PubMed] [Google Scholar]
