Food

Chemistry
Food Chemistry 93 (2005) 445–448
www.elsevier.com/locate/foodchem

Trans fatty acid content of Brazilian biscuits

Clayton Antunes Martin, Rodolfo Carapelli, Jesuı́ Vergı́lio Visantainer,
Makoto Matsushita, Nilson Evelázio de Souza *
Department of Chemistry, Maringá State University, Av. Colombo, 5790 – CEP 87020-900, Maringá, Paraná State, Brazil

Received 2 February 2004; received in revised form 5 October 2004; accepted 5 October 2004

Abstract

The fatty acid composition and trans fatty acid (TFA) contents of samples of five brands Brazilian cream cracker biscuits were
determined by gas–liquid chromatography, using a polar 100 m capillary column CP Sil-88 and flame ionization detection. The iden-
tification of fatty acids done by equivalent chain length for trans fatty acids. Total TFA ranged from 12.2% to 31.2% of total fatty
acid and the mean was 20.1%. Trans 18:1 isomers were the major group of TFA present in all the analyzed brands, representing
83.2% of total trans isomers. The mono-trans 18:2 isomer content ranged from 1.6% to 4.2% of total fatty acids, this being the most
prevalent group of trans polyunsaturated acid. The di-trans 18:2 isomer (9t, 12t) was found at very low levels (0.10–0.15% of total
fatty acids). Trans 18:3 isomer content ranged from 0.11% to 0.75% of total fatty acids representing 24.4–75.0% of total a-linolenic
acid. The results indicate that Brazilian cream cracker biscuits contain considerable proportions of trans fatty acids, both monoun-
saturated and polyunsaturated.

2004 Elsevier Ltd. All rights reserved.

Keywords: trans fatty acids; Biscuits; Fatty acid composition

1. Introduction associated with adverse effects. Several studies have

shown that a high intake of TFA raises low density lipo-
Coronary heart disease (CHD) remains the leading protein (LDL) cholesterol and lowers high density lipo-
cause of death and disability in many countries around protein (HDL) cholesterol, affecting the LDL/HDL
the world, including Brazil. According to Maranhão cholesterol ratio in a way that is unfavourable compared
(1999), ca. 34% of all deaths in Brazil are caused by with all other fatty acids. Moreover, trans fats have been
CHD. There are several multiple risk factors that act reported to raise lipoprotein (a) and plasma triglyceride
both independently and jointly. Among dietary factors, levels, that are independently associated with the in-
the type of fat intake and total amount of fat in the diet creased risk of CHD (Aro, Jauhiainen, Partanen, Salmi-
play important roles in determining risk of CHD (Hu, nen, & Mutanen, 1997; Ascherio, Katan, Zock,
Manson, & Willett, 2001). Stampfer, & Willett, 1999; Mensink & Katan, 1990;
It has been known for many years that a high intake Mensink, Zock, Katan, & Hornstra, 1992; Nestel
of saturated fat contributes to the development of CHD. et al., 1992).
More recently, trans fatty acids (TFA) have also been Trans isomers are mainly present in fats produced by
partial hydrogenation of vegetable or marine oils, but
*
Corresponding author. Tel.: +55 44261 4389; fax: +55 44 263
they also occur in minor quantities in dairy and other
5784. animal fats by biological hydrogenation of the unsatu-
E-mail address: nesouza@uem.br (N.E. de Souza). rated fatty acids in the rumen. In addition, trans isomers

0308-8146/$ - see front matter
2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2004.10.022
446 C.A. Martin et al. / Food Chemistry 93 (2005) 445–448

may be formed in small amounts in vegetable oils during for all analyzed acids using a method described by Ack-
deodorization and food-frying operations (Ackman & man (1972), with methyl esters of 16:0, 18:0, and 20:0
Tag, 1998). acids as reference compounds. The final results are ex-
In Brazil, commercial hydrogenation of vegetable oils pressed as relative percentages.
dates from the late 1950s, for producing shortening and
hard margarine. With the development of partial and 2.3. Statistical analysis
selective hydrogenation techniques, processed vegetable
fats rapidly displaced animal fats in the diet of the Bra- The samples were analyzed, one by one, in triplicate,
zilian people. These fats have been widely used in the and the results were expressed as mean values ± stan-
production of several foods, such as margarine, choco- dard deviation (SD). The results were compared using
late spreads, biscuits, potato chips, and bakery products. analysis of variance (ANOVA) with 5% significance le-
In past decade, biscuit per capita consumption has in- vel using Statistica 5.0 software (StatSoft, USA, 1995).
creased twofold in our country, becoming an important The average values were compared by TukeyÕs test.
source of dietary fat (Coelho, Gramacho, Bressan-Filho,
Contini, & Venturili, 2000; Pelaez-Alvarez, Szmrecsányi,
& Tango, 1991). 3. Results and discussion
Thus, the aim of this study was to determine the trans
fatty acid composition of the cream cracker, a type of Fatty acid compositions of cream cracker are pre-
biscuit frequently consumed in Brazil. sented in Table 1. The amount of total TFA in the sam-
ples ranged from 12.2% to 31.2% of total fatty acids and
the mean value was 20.1%. Total trans content was sig-
2. Materials and methods nificantly higher (P < 0.05) in brands A and E. The TFA
comprised isomers of 16:1, 18:1, 18:2, and 18:3 acids and
2.1. Sampling trans 18:1 isomers were the major group of TFA present
in all the brands analyzed, ranging from 8.8% to 28.3%.
Twelve units of each brand of cream cracker biscuits The mean value represented 83.2% of the total trans iso-
were purchased from local supermarkets and bakeries mers. For the trans 16:1 isomer, only small amounts
between December 2002 and June 2003. Each brand (0.25% of total fatty acids) were detected in some lots
was coded with a letter (A, B, C, D, and E). Lot num- of brand C.
bers were checked to ensure that each unit belonged to Trans polyunsaturated acids included geometrical
a different lot. Samples were selected to include the ma- and positional isomers of linoleic acid for all the brands
jor manufacturers of the biscuits in Brazil. The analyses analyzed and geometrical isomers of a-linolenic acid
were carried out in triplicate. only for brands B, D, and E. The mono-trans 18:2 iso-
mer content ranged from 1.6% to 4.2% of total fatty
2.2. Methods acids being the most prevalent group of trans polyunsat-
urated acid. The di-trans 18:2 isomer (9t, 12t) was found
Total lipids were extracted using a procedure pub- at very low levels (0.10–0.15% of total fatty acids). Trans
lished by Folch, Lees, & Stanley (1957) and fatty acid 18:3 isomer content ranged from 0.11% to 0.75% of total
methyl esters were prepared by methylation of fatty fatty acids, representing 24.4–75.0% of total a-linolenic
acids, as described by the method of Hartman & Lago acid.
(1973). These results show that the amount of trans monoun-
Fatty acid methyl esters (FAME) were analyzed by saturated and polyunsaturated varied considerably
using a Shimadzu model 14A gas chromatograph, among the analyzed brands because of the differences
equipped with a flame ionization detector (FID) and a in hydrogenation process conditions, such as tempera-
fused silica capillary column CP Sil-88 (100 m · 0.25 ture, pressure, type and amount of catalyst and agitation
mm id., 0.20 m film thickness, Chrompack). Hydrogen rate affect the resulting TFA content of the starting oil.
carrier flow was 1.3 ml/min at a split ratio of 1/100. Moreover, food producers may use single hydrogenated
Injector and detector temperatures were both 250 C. fats or many possible combinations of hydrogenated
The column was operated isothermally at 170 C. Reten- and non-hydrogenated fats and oils to achieve a final
tion times and peak area percentages were determined product with the desirable characteristics (Innis, Green,
by the CG-300 computing integrator (CG Instruments). & Halsey, 1999; Karabulut, Kayahan, & Yaprak, 2003).
The identification of fatty acids was based on authentic For similar biscuits, the trans content found was
reference standards (Sigma) and on literature reports of 12.7% for US (Enig, Pallansch, Sampugna, & Keeney,
equivalent chainlength (ECL) for trans fatty acids (Paw- 1984), 11.1% for Argentina (Tavella et al., 2000), 9.1%
lowicz & Drozdowski, 1998; Ratnayake & Pelletier, for Greece (van Erp-baart et al., 1998) and 2.0% for
1992; Wolff, 1992). The values of ECL were determined New Zealand (Lake, Thomson, Devane, & Scholes,
 C.A. Martin et al. / Food Chemistry 93 (2005) 445–448 447

Table 1
Fatty acid composition of Brazilian cracker biscuitsa
Fatty acid Brands
A B C D E
14:0 0.10 ± 0.02a 0.10 ± 0.02a 1.47 ± 1.08b 0.11 ± 0.03a 0.12 ± 0.03a
14:1c ND ND 0.88 ± 0.73 ND ND
16:0 11.5 ± 0.46a 10.9 ± 1.04a 20.2 ± 5.53b 11.2 ± 0.76a 11.8 ± 0.38a
16:1c ND ND 2.09 ± 1.65 ND ND
16:1t ND ND 0.25 ± 0.20 ND ND
18:0 11.5 ± 1.30a 4.94 ± 0.20b 17.2 ± 9,50c 5.53 ± 0.49b 9.99 ± 1.26a
18:1c 34.5 ± 1.18a,b 36.6 ± 1.03a,b 33.6 ± 4.78b 33.9 ± 1.33a,b 36.8 ± 4.30a
18:1t 28.3 ± 2.70a 8.80 ± 0.82b 10.4 ± 8.82b 9.75 ± 0.80b 25.2 ± 5.63a
18:2cc 9.34 ± 1.50a 31.2 ± 1.01b 10.3 ± 3.51a 32.60 ± 1.04b 11.05 ± 2.68a
18:2tt 0.15 ± 0.08a 0.10 ± 0.05a ND 0.10 ± 0.04a 0.11 ± 0.06a
18:2 c/t and t/c 2.72 ± 0.33a 4.20 ± 0.49b 1.64 ± 1.50c 3.43 ± 0.46a,b 2.77 ± 0.56a
18:3ccc 0.40 ± 0.10a 1.01 ± 0.10b 0.53 ± 0.12a 1.00 ± 0.11b 0.45 ± 0.07a
18:3t ND 0.53 ± 0.18a ND 0.75 ± 0.10b 0.11 ± 0.09c
20:0 0.32 ± 0.03a 0.39 ± 0.06b 0.25 ± 0.07c 0.36 ± 0.06a 0.36 ± 0.07a,b
20:1c 0.19 ± 0,02a 0.15 ± 0.02b 0.21 ± 0.04a 0.22 ± 0.02a 0.22 ± 0.05a
22:0 0.38 ± 0,04a 0.39 ± 0.09a 0.13 ± 0.10b 0.37 ± 0.15a 0.39 ± 0.06a
24:0 0.14 ± 0,03a 0.15 ± 0.05a ND 0.13 ± 0.08a 0.14 ± 0.02a
SFAb 23.9 ± 1.57a 16.9 ± 0.77b 39.2 ± 15.8c 17.7 ± 3.44a,b 22.8 ± 1.32a,b
PUFAc 9.74 ± 1.59a 32.2 ± 1.10b 10.8 ± 3.44a 33.6 ± 1.13b 11.5 ± 2.74a
TFAd 31.2 ± 2.64a 13.63 ± 3.44b 12.2 ± 10.3b 14.0 ± 1.25b 28.2 ± 5.61a
PUFA/SFAe 0.41 ± 0.07a 1.91 ± 0.08b 0.38 ± 0.29a 1.90 ± 0.11a 0.51 ± 0.14b
Trans/cis 0.70 ± 0.08a 0.20 ± 0.02b 0.26 ± 0.18b 0.21 ± 0.02b 0.58 ± 0.19a
a
Results expressed as percentage of total fatty acid methyl esters. Values are means ± SD for six samples of triplicate analyses. Averages followed
by different letters in the same line are significantly different (P < 0.05) by TukeyÕs test. Fatty acids less than 0.1%: 12:0, 15:0, 17:0, 17:1, 20:2, 22:1 and
not identified fatty acids.
b
SFA = saturated fatty acids.
c
PUFA = polyunsaturated fatty acids.
d
TFA = trans fatty acids.

1996). These values are similar to those found in the in brands B and D (P < 0.05). Because essential fatty
present study. However, Grecian and New Zealander acids (EFA) are included in this group, the PUFA con-
biscuits, contained a high proportion of saturated fatty tent is very important for the biological and nutritional
acids. Furthermore, higher levels were found in Canada value of these biscuits. A disturbing feature was the
and Suedew, 40.3% and 29.1%, respectively (Innis et al., combination of high levels of TFA and lower levels of
1999; van Erp-baart et al., 1998). essential fatty acids in brands A and E.
For cis monounsaturated 18:1, differences (P < 0.05) It should be stressed that the Department of Health
were observed between brands C and E, with values (UK) (HMSO, 1994) recommends a minimal PUFA/
varying from 33.6% to 36.8%, respectively. The oleic SFA ratio value of 0.45, and this was observed only in
acid (18:1n9) was the main fatty acid of this group (data brands B, D, and E, with values of 1.91, 1.90, and
are not shown); however, unusual isomers were also 0.51, respectively.
found. There is little information on the biological activ-
ity of these positional isomers, but a preliminary study
suggested that unusual isomers can influence the growth 4. Conclusion
of infants (Ayagari, Peepies, & Carlson, 1996).
Total saturated fatty acids (SFA) were significantly The results in the present study indicate that Brazilian
higher in brand C (P < 0.05), followed by brands A, E cream cracker biscuits contain considerable proportions
and finally by brands B and D. Brand C was an excep- of trans fatty acids, both monounsaturated and
tion due to the use of lard in the production of some of polyunsaturated.
the units analyzed. Among the SFA, palmitic acid (16:0)
presented the highest value ranging from 10.9% to
20.2%, followed by stearic (18:0), that varied from Acknowledgements
4.9% to 17.2%.
The content of polyunsaturated fatty acids (PUFA) The authors are grateful to CAPES and CNPq for
ranged from 9.7% to 32.2%, being significantly higher financial support.
448 C.A. Martin et al. / Food Chemistry 93 (2005) 445–448

References Karabulut, I., Kayahan, M., & Yaprak, S. (2003). Determination of

changes in some physical and chemical properties of soybean oil
Ackman, R. G. (1972). The analysis of fatty acids and related materials during hydrogenation. Food Chemistry, 81, 453–456.
by gas–liquid chromatography progress in the chemistry of fats and Lake, R., Thomson, B., Devane, G., & Scholes, P. (1996). Trans fatty
other. Lipids, 12, 165–284. acids content New Zealand foods. Journal of Food Composition and
Ackman, R. G., & Tag, T. K. (1998). Trans fatty acids and the Analysis, 9, 365–374.
potential for less in technical products. In J. L. Sebédio & W. W. Maranhão, M. F. C. (1999). Panorama das doenças cardiovasculares
Christie (Eds.), Trans fatty acids in human nutrition. Dundee no Brasil e no mundo: os desafios para o terceiro milênio. Revista
(Scotland): The Oily Press Ltd. New Cor, 8, 6–10.
Aro, A., Jauhiainen, M., Partanen, R., Salminen, I., & Mutanen, M. Mensink, R. P., & Katan, M. B. (1990). Effect of dietary trans fatty
(1997). Stearic acid trans fatty acids, and dairy fat: effects on serum acids on high-density and low-density lipoprotein cholesterol levels
and lipoprotein, lipids apolipoproteins, lipoprotein (a), and lipid in healthy subjects. New England Journal of Medicine, 323,
transfer proteins in health subjects. American Journal of Clinical 439–445.
Nutrition, 65, 1419–1426. Mensink, R. P., Zock, P. L., Katan, M. B., & Hornstra, G. (1992).
Ascherio, A., Katan, M. B., Zock, P. L., Stampfer, M. J., & Willett, W. Effect of dietary cis and trans fatty acids on serum lipoprotein[a]
C. (1999). Trans fatty acids and coronary heart disease. New levels in humans. Journal of Lipid Research, 33, 1493–1501.
England Journal of Medicine, 340, 1994–1998. Nestel, P. J., Noakes, M., Belling, G. B., McArthur, R., Clifton, P. M.,
Ayagari, A., Peepies, J. M., & Carlson, S. E. (1996). Relationsship of & Abbey, M. (1992). Plasma lipoprotein lipid and Lp[a] changes
isomeric fatty acids in human cord blood to n 3 and n 6 status. with substitution of elaidic acid for oleic acid in the diet. Journal of
Pediatric Research, 39, 1808. Lipid Research, 33, 1029–1036.
Coelho, C. N., Gramacho, A., Bressan-Filho, A., Contini, E., & Pelaez-Alvarez, V. M., Szmrecsányi, T., & Tango, J. S. (1991).
Venturili, P. N. (2000). Revista de Polı́tica Agrı́cola, 9, 13–15. Determinants of technological innovation in the shortening indus-
Enig, M. G., Pallansch, L. A., Sampugna, J., & Keeney, M. (1984). try. In International meeting on fats and oils proceedings (pp. 129–
Fatty acid composition of the fat in selected food items with 137).
emphasis on trans components. Journal American Oil Chemical Pawlowicz, R., & Drozdowski, B. (1998). Separation of geometrical
Society, 60, 1788–1795. isomers of unsaturated fatty acids by gas–liquid chromatography
Folch, J., Lees, M., & Stanley, G. H. S. (1957). A simple method for on CP-Sil88 and DB-23 columns. Chemical Analysis (Warsaw), 43,
the isolation and purification of total lipids from animal tissues. 961–967.
Journal of Biological Chemistry, 226, 497–509. Ratnayake, W. M. N., & Pelletier, G. (1992). Positional and
Hartman, L., & Lago, R. C. A. (1973). Rapid determination of fatty geometrical isomers of linoleic acid in partially hydrogenated oils.
acid methyl esthers from lipids. Laboratory Practice, 22, 475– Journal American Oil Chemical Society, 69, 95–105.
477. Tavella, M., Peterson, G., Espeche, M., Cavallero, E., Cipolla, L.,
HMSO, UK. (1994). Department of Health. Nutritional aspects of Perego, L., et al. (2000). Trans fatty acid content of a selection of
cardiovascular disease (pp. 37–46). London (Report on Health and foods in Argentina. Food Chemistry, 69, 209–213.
Social Subjects, 46). van Erp-baart, M. A., Couet, C., Cuadrado, C., Kafatos, A., Stanley,
Hu, F. B., Manson, J. E., & Willett, W. C. (2001). Types of dietary fat J., & van Poppel, G. (1998). Trans fatty acids in bakery products
and risk of coronary heart disease: a critical review. Journal of from14 european countries: the TRANSFAIR study. Journal of
American College Nutrition, 20, 5–19. Food Composition and Analysis, 11, 161–169.
Innis, S. M., Green, T. J., & Halsey, T. K. (1999). Variability in the Wolff, R. L. (1992). Resolution of linolenic acid geometrical isomers by
trans fatty acid content of foods within a food category: implica- gas–liquid chromatography on a capillary column coated with a
tions for estimation of dietary trans fatty acids intakes. Journal of 100% cyanopropil polysiloxane film (CPTM Sil 88). Journal of
American College Nutrition, 18, 255–260. Chromatography Science, 30, 17–22.