JACN Did you know that you can get alerts when a new issue is online?
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Innis, S. M.
Right arrow Articles by Halsey, T. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Innis, S. M.
Right arrow Articles by Halsey, T. K.
Journal of the American College of Nutrition, Vol. 18, No. 3, 255-260 (1999)
Published by the American College of Nutrition

Original Paper

Variability in the Trans Fatty Acid Content of Foods within a Food Category: Implications for Estimation of Dietary Trans Fatty Acid Intakes

Sheila M. Innis, PhD, RDN, Timothy J. Green, PhD, RDN and Thomas K. Halsey, BSc

Department of Paediatrics, University of British Columbia, Vancouver, British Columbia, CANADA

Address reprint requests to: Sheila M. Innis, PhD, B.C. Research Institute for Children’s and Women’s Health, 950 West 28th Avenue, Vancouver, B.C. V5Z 4H4, CANADA.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Objective: Currently, the published information on trans fatty acid composition of foods is incomplete and of questionable accuracy. Detailed fatty acid analysis of over 200 foods was undertaken for the purpose of determining the variability in trans fatty acid content among foods within a product category, and the significance of this variability to the estimation of trans fatty acids intakes from analysis of dietary intake data.

Methods: The analysis of food fatty acids used gas-liquid chromatography with 100 m capillary columns and standardized methodologies for food sampling, fat extraction, separation and quantification of trans fatty acid isomers. For the purposes of this report, trans refers to all non-naturally occurring isomers including trans, cis-trans, geometric and positional isomers.

Results: The results show that the amount of trans fatty acids varies considerably among foods within a category, reflecting differences in the fats and oils used in the manufacturing or preparation process. For example, the range of trans fatty acids in 17 brands of crackers was 23 to 51% total fatty acids, representing differences of from 1 to 13 g trans fatty acids per 100 g cracker. The large errors that may arise in estimates of the trans fatty acid intake of an individual are illustrated by analyses of the potential trans fatty acid intake in a sample diet, for each food as calculated using the minimum and maximum values for trans fatty acids within a given category. The results of these analyses show estimates of trans fatty acid intake from a low of 1.4 to 25.4 g a day for the same diet.

Conclusion: This study shows that the wide variability in trans fatty acid content of different foods may result in large errors in the estimation of trans fatty acid intake of individuals and, potentially, groups.

Key words: trans fatty acids, diet methodology, food composition, Canadian


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The process of hydrogenation of vegetable oils to improve their texture and stability results in the formation of trans fatty acids. The extent of hydrogenation is dependent on the unsaturated fatty acid content of the starting oil and the desired stability and physical properties of the final product [1]. The major source of trans fatty acids in the North American diet is hydrogenated fats and oils [1].

In addition to influencing lipid risk factors for cardiovascular disease, trans fatty acids have also been implicated in breast cancer [2] and poor fetal and early infant growth [3]. Furthermore, trans fatty acids may compete with linoleic acid for desaturation, thereby reducing the production of the 2-series eicosanoids [4]. Interest in trans fatty acids has centered largely on their effect on lipid risk factors in cardiovascular disease [5]. It has been known for some 30 years that replacing unhydrogenated unsaturated fat with hydrogenated fat increases plasma cholesterol concentration [5]. More recently it has been shown that replacing a cis fatty acid with a trans unsaturated fatty acid in the diet increases the concentration of low density lipoprotein (LDL) cholesterol, increases Lp(a) and lowers high density lipoprotein (HDL) cholesterol in healthy subjects [6, 7].

Despite experimental evidence of metabolic changes due to trans fatty acids, epidemiological studies have not found conclusive evidence that high trans fatty acid intakes have an adverse effect on coronary heart disease. Studies using adipose tissue fatty acids as a biomarker of trans fatty acid intake have similarly not detected an association between trans fatty acids and coronary heart disease [8, 9]. Epidemiological studies exploring the relationship between dietary trans fatty acids and coronary heart disease have relied on estimates of trans fatty acid intake derived from analyses of intake data gathered in food frequency questionnaires (FFQ) [10–12]. Two prospective studies in the United States found that women in the highest quintile of trans fatty acid intake were at greatest risk for coronary heart disease when compared to women in the lowest quintile of intake [10, 12]. Similar results were found in a case-control study of men in the United States [11]. A case-control trial of men and women in Scotland found no relationship between trans fatty acid intake and the subsequent development of coronary heart disease [13].

One of the difficulties encountered by investigators examining the association of trans fatty acid intakes with coronary heart, or other diseases, relates to the difficulties of estimating trans fatty acid intakes using a food frequency questionnaire (FFQ) with subsequent analysis using presently available nutrient databases. In addition to the problems inherent in dietary intake methodology (for example, recall and portion size estimation) [14], the database for the trans fatty acid composition of foods is incomplete and of questionable accuracy. An additional problem arises when an average value is used to describe the trans fatty acid content of a food product category in which there is wide variation in the trans fatty acid content of individual foods within that category. For example, an analysis of 43 bread samples in the United States found levels of trans fatty acids ranging from 0% to 32% of total fatty acids [15]. This suggests that intakes of trans fatty acids could vary widely among individuals, based on both personal food preferences and regional variations in food supplies and producers. The purpose of this study was to undertake detailed fatty acid analysis of over 200 locally and nationally available foods for the purposes of detailing the variability in trans fatty acid content within a product category and determining the significance of this variability to the analysis of dietary data.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Food items were purchased from retail stores and restaurants in Vancouver, British Columbia, Canada. Food items were stored at an appropriate temperature (room, refrigerated or frozen), and all the foods were analyzed within five days of purchase. Prepared foods purchased from restaurants were analyzed immediately. Where available, the fat sources listed on the package label were recorded. Individual food items were homogenized or ground, or a weighed representative portion was taken. A sample of each (40–200 g) was then homogenized in saline and quadruplicate aliquots (representing 1 g food) were taken. Total lipids were extracted [16] from each aliquot, dried under nitrogen and weighed. Samples of the lipid extract (2 mg lipid) were taken and the fatty acid components converted to their respective methyl esters using BF3/benzene/methanol (25:20:55 v/v/v), 30 minutes at 100°C; then, the fatty acid methyl esters were recovered, separated and quantitated using a SP-2560 (100 m x 0.25 mm id, 20 µm film thickness, Sulpelco Inc, Belleforte, PA) capillary column in a Varian 3400 gas chromatograph as described essentially by Chen et al. [17]. The carrier gas was pure hydrogen subjected to a head pressure of 50 psi. The split rate for the injector was set at a ratio of 1:10. Operating conditions were injector 240°C and detector 260°C. Temperature programming: initial 100°C; program rate 5°C/min; final, 190°C hold for 20 min; increase to 225°C at 1°C/min and hold for 2 min. For the purposes of this report, trans refers to all non-naturally occurring isomers. Total trans was calculated as the sum of all trans, cis-trans, geometric and positional isomers. Fatty acid identification was based on authentic reference standards and mass spectrometry, using a Varian (Saturn II) mass spectrometer. Repeat analysis of seven separate foods yielded a less than 2% variation in the analysis and quantitation of trans fatty acid content.

In order to explore the influence of the variability in the trans fatty acid content among similar foods on estimates of trans fatty acid intakes, a sample day’s diet (1 MJ energy) was created using a food record and the trans fatty acid intake then calculated. The intake of fatty acids from the diet was calculated using the highest value for trans fatty acids and using the lowest value for trans fatty acids for each food or food group in the diet record. In addition, and for comparative purposes, the diet was also analyzed using ‘The Food Processor’ nutrition and fitness software (Esha Research, Salem, OR) [18]. The database for ‘The Food Processor’ software is compiled from the U.S. Department of Agriculture, as well as other sources.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Complete fatty acid analyses were performed for over 200 foods, details of which will be published elsewhere. The total fat content (g/per 100 g food) and the percentage of fat as saturated fatty acids, cis monounsaturated fatty acids, cis 18:2n-6, cis 18-3n-3 and trans fatty acids of 23 selected foods and food groups are provided in Table 1. In addition, the trans fatty acid content is also expressed as g per 100 g of food or food group. The results show that the amount of trans fatty acids, as a percentage of total fat, varied considerably among similar food items, reflecting differences in the fat type used in the manufacturing or preparation process.


View this table:
[in this window]
[in a new window]
  		Table 1. Fatty Acid (FA) Composition of Select Foods and Food Categories

 
The variation in trans fatty acid content of similar foods is well illustrated by consideration of the results for bread and fried breaded chicken. The levels of trans fatty acids in 16 breads (whole wheat and white) analyzed ranged from 1% to over 30% total fatty acids (Fig. 1). The total amount of trans fatty acids ranged from 0 to 1.0 g per 100 g of food. Fat sources were disclosed on the labels of 12 of the 16 breads analyzed. Five of the six breads made with partially hydrogenated fat had in excess of 30% trans fatty acids, whereas five breads were prepared with non-hydrogenated fat and contained <2% as trans fatty acids. Similarly, the eight breaded fried chicken products analyzed contained amounts of trans fatty acids ranging from 11.9% to 56.7% total fatty acids. In six of eight breaded chicken products, the percentage of fatty acids as trans was greater than 30%. One breaded fried chicken product purchased from a national chain outlet had in excess of 50% of fatty acids as trans fatty acids, equivalent to 6.9 g/100 g of food.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 1. Percentage of total fatty acids as trans in 16 breads analyzed.

 
Results showing the variability in estimates of potential trans fatty acid intakes in dietary analysis are provided in Table 2. The lowest and highest estimates of the potential trans fatty acid intake in the sample diet were 1.5 and 28.4 g, respectively. Calculation of the nutrient content of the diet using ‘The Food Processor’ [18] gave an estimate of the trans fatty acid content of the diet of 6.5 g.


View this table:
[in this window]
[in a new window]
  		Table 2. Sample Diet to Show Variability in Estimated Trans Fatty Acid Intake Depending on Brand or Database Chosen

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The present study demonstrates the wide variability of trans fatty acid content present among foods within a food category (Table 1) and that this variability can limit the accuracy of estimates of dietary trans fatty acids intakes when analysis of diet information is made using nutrient databases. The greatest variability in trans fatty acids within a food category was found in the snack foods. For example, analyses of the trans fatty acid contents of crackers showed that a 100 g serving of crackers could provide 1 to 13 g of trans fatty acids, depending on the brand consumed. Previous studies [15], using data compiled from the Nutrient Data Bank Bulletin Board (U.S. Department of Agriculture), have also noted similar variations in the trans fatty acid content of foods within a food category. For example, the percentage of fatty acids as trans in bread in the USDA database ranges from 0% to 32% and in french fries from 3% to 34% of total fatty acids. Not all food groups showed high variability in their trans fatty acid content. For example, the trans fatty acid content of cake mixes and peanut butter ranged from 28.7 to 30.1 and from 1.6 to 4.6 g/per 100 g lipid, respectively.

The wide variability among the trans fatty acid values in the USDA database may, in part, have resulted from the compilation of data gathered using older gas liquid chromatography methods, as well as data collected by several different laboratories, employing different methodologies and located in different geographic areas. Ratnayake et al. [19] recently commented on the considerable overlap of some trans and cis isomers of unsaturated fatty acids. This problem is likely to be even more significant with older gas-liquid chromatography methodologies employing 30 m capillary columns. The present study developed a large database on the fatty acid composition of foods from analysis done in our laboratory using standardized methodologies for food sampling, fat extraction and quantification of fatty acid isomers using a 100 m column with mass spectrometer identification of fatty acid isomers.

There are several explanations for the variability of trans fatty acid content of foods within a food category. First, the production of hydrogenated oils can result in variable content of trans fatty acids. Temperature, hydrogenation pressure, type and amount of catalyst and agitation affect the resulting trans fatty acid content of the starting oil [1]. Second, food producers may use single hydrogenated or non-hydrogenated fats or oils or many possible combinations of both hydrogenated and non-hydrogenated fats and oils to achieve the desired final product characteristics. For example, some of the oils indicated on the labels of cookies analyzed in our laboratory had blends of hydrogenated and non-hydrogenated vegetable oil shortenings which included cocoa butter, palm kernel oil shortening, soybean oil and Canola oil. Finally, the use of hydrogenated and non-hydrogenated oil fats and oils in food products can also be expected to vary with national and regional domestic availability and with costs of various edible oils.

The analysis of the diet record described in this report illustrates the large errors that may arise in the estimation of trans fatty acid intake of an individual due to the large variability in trans fatty acid content of foods within a food category and use of average food group trans fatty acid values rather than specific product analysis. This wide variability in the trans fatty acid content and composition among foods has important implications for estimates of trans fatty acid intakes of individuals using dietary recalls, records and FFQ. The identification of exact brand names of food products consumed, coupled with data from analysis which resolves all cis and trans isomers of fatty acids in foods is essential to allow reliable estimation of trans fatty acid intakes. Thus, food composition tables using the trans fatty acid values of individual food products by brand must be available. Because of possible wide geographical variability in the trans fatty acid content of foods, including locally produced foods such as bakery goods, regional food composition tables or, at least, tables which also detail the fat and oil ingredients in the product may be required. It further needs to be recognized that, even for a specific brand within a geographical area, trans fatty acid content may vary over time because food producers may change the type of oil used, depending on current supplier, availability and cost.

The errors which can arise in estimating trans fatty acid intake because of the high variability in trans fatty acid content among foodstuffs are most pronounced with estimates derived from FFQ. Food frequency questionnaires often collapse foods with very different trans fatty acid contents into a single food group category. For example, on a FFQ for the food category french fries, including all frozen and restaurant products, the average trans fatty acid value from the present analysis was 2.1 g per 100 g portion, with an actual range among the individual products analyzed at 0.2 to 3.7 g per 100 g portion. The repeated collapsing of foods into food categories on a FFQ can lead to multiple errors in the estimates of the trans fatty acid intake of individuals and, thus, mask the true variability of trans fatty intakes within a group or population of interest.

Three studies in the United States have reported that individuals in the highest quintile of trans fatty acid intake were at higher risk for coronary heart disease than individuals in the lowest quintile of intake. Individuals [10–12] in the second through fourth quintiles of trans fatty acid, however, showed no significant increased risk relative to those in the first quintile. This lack of association across the quintiles of estimated trans fatty acid intake may be the result from the inability of the FFQ and current food composition data bases to give precise estimates of trans fatty acid intake for foods consumed, thus masking differences in intakes among individuals within these quintiles. It seems reasonable to expect that the FFQ will be able to distinguish between individuals with the lowest and the highest intakes of trans fatty acids, as these are associated respectively with low and high intakes of fat from snack and convenience foods. Variability due to actual choices of margarines, for example stick or tub, hydrogenated or unesterified, actual brands of baked goods or home preparation, however, would not be apparent from the FFQ methodology. The results of the study reported here show that processed foods within a food category contain varying amounts of trans fatty acid and suggest that the true variability of trans fatty acid intake that exists among individuals with moderate intakes of trans fatty acids may disappear when foods are collapsed into categories on a FFQ.

When categorical variables are used to estimate relative risk, the inability of the FFQ to distinguish between individuals consuming moderate intakes of trans fatty acids may lead to considerable misclassification [14]. Misclassification may weaken relative risk estimates to the point that the risk is statistically undetectable [14]. Moreover, misclassification may result in relative risk estimates of moderate magnitude that are underestimates of the true association between cardiovascular risk and trans fatty acid intake, which may be higher than previously shown.

In summary, the study reported here demonstrates considerable variation in the trans fatty acid composition and content of food items within a food category. Further, this study has demonstrated that because of the wide variation in trans fatty acids within a food category, the use of average values for the trans fatty acid content of a food category are of limited value. Accurate estimates of trans fatty acid intakes of individuals are likely to be obtained only after food composition tables are updated with values for specific foods, by brand or by the fats and oils in a product. This information is essential to allow important epidemiological studies to explore fully the potential negative effects of trans fatty acid intake on heart disease [5], breast cancer [2], essential fatty acid metabolism [3, 4] or fetal and early infant development [3].


   ACKNOWLEDGMENTS
 
These studies were supported by a grant from the Canola Utilization Assistance Program (CUAP), Canada.

Received August 1, 1998. Accepted December 1, 1998.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. ASCN/AIN Task Force on Trans Fatty Acids. American Society for Clinical Nutrition and American Institute of Nutrition: Position paper on trans fatty acids. Am J Clin Nutr 63: 663–670, 1996.[Abstract/Free Full Text]
  2. Kohlmeier L, Simonsen N, van’t Veer P, Strain JJ, Martin-Moreno JM, Margolin B, Huttunen JK, Fernandez-Crehuet Navajas J, Martin BC, Thamm M, Kardinaal AF, Kok FJ: Adipose tissue trans fatty acids and breast cancer in the European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer. Cancer Epidemiol Biomark Prev 6: 705–710, 1997.[Abstract/Free Full Text]
  3. Koletzko B: Trans fatty acids may impair biosynthesis of long-chain polyunsaturates and growth in man. Acta Paediatr 81: 302–306, 1992.[Medline]
  4. Rosenthal MD, Doloresco MA: The effects of trans fatty acids on fatty acyl delta 5 desaturation by human skin fibroblasts. Lipids 19: 869–874, 1984.[Medline]
  5. Report of the Expert Panel on Trans Fatty Acids and Coronary Heart Disease. Trans fatty acids and coronary heart disease risk. Am J Clin Nutr 62: 655S–708S, 1995.[Abstract/Free Full Text]
  6. Mensink RP, Katan MB: Effect of dietary trans fatty acids on high-density and low-density lipoprotein cholesterol levels in healthy subjects. N Engl J Med 323: 439–445, 1990.[Abstract]
  7. Nestel P, Noakes M, Belling B, McArthur R, Clifton P, Janus E, Abbey M: Plasma lipoprotein lipid and Lp[a] changes with substitution of elaidic acid for oleic acid in the diet. J Lipid Res 33: 1029–1036, 1992.[Abstract]
  8. Aro A, Kardinaal AF, Salminen I, Kark JD, Riemersma RA, Delgado-Rodriguez M, Gomez-Aracena J, Huttunen JK, Kohlmeier L, Martin BC, et al.: Adipose tissue isomeric trans fatty acids and risk of myocardial infarction in nine countries: the EURAMIC study. Lancet 345: 273–278, 1995.[Medline]
  9. Roberts TL, Wood DA, Riemersma RA, Gallagher PJ, Lampe FC: trans isomers of oleic and linoleic acids in adipose tissue and sudden cardiac death. Lancet 345: 278–282, 1995.[Medline]
  10. Willett WC, Stampfer MJ, Manson JE, Colditz GA, Speizer FE, Rosner BA, Sampson LA, Hennekens CH: Intake of trans fatty acids and risk of coronary heart disease among women. Lancet 341: 581–585, 1993.[Medline]
  11. Ascherio A, Hennekens CH, Buring JE, Master C, Stampfer MJ, Willett WC: Trans-fatty acids intake and risk of myocardial infarction. Circulation 89: 94–101, 1994.[Abstract/Free Full Text]
  12. Hu FB, Stampfer MJ, Manson JE, Rimm E, Colditz GA, Rosner BA, Hennekens CH, Willett WC: Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med 337: 1491–1499, 1997.[Abstract/Free Full Text]
  13. Bolton-Smith C, Woodward M, Fenton S, Brown CA: Does dietary trans fatty acid intake relate to the prevalence of coronary heart disease in Scotland? Eur Heart J 17: 837–845, 1996.[Abstract/Free Full Text]
  14. Willett W: "Nutritional Epidemiology." New York: Oxford University Press, 1998.
  15. Carlson SE, Clandinin MT, Cook HW, Emken EA, Filer LJ: trans Fatty acids: infant and fetal development. Am J Clin Nutr 66: 715S–736S, 1997.[Abstract]
  16. Folche J, Lees M, Sloane-Stanley S: A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226: 497–509, 1957.[Free Full Text]
  17. Chen ZY, Pelletier G, Hollywood R, Ratnayake WM: Trans fatty acid isomers in Canadian human milk. Lipids 30: 15–21, 1995.[Medline]
  18. Esha Research: "Food Processor Version 8." Salem: Esha Research, 1997.
  19. Ratnayake WMN, Beare-Roger JL: Problems of analyzing C18 cis- and trans-fatty acids of margarine on the SP-2340 capillary column. J Chromatogr Sci 28: 633–639, 1990.



This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
V. Chajes, A. C. M. Thiebaut, M. Rotival, E. Gauthier, V. Maillard, M.-C. Boutron-Ruault, V. Joulin, G. M. Lenoir, and F. Clavel-Chapelon
Association between Serum trans-Monounsaturated Fatty Acids and Breast Cancer Risk in the E3N-EPIC Study
Am. J. Epidemiol., June 1, 2008; 167(11): 1312 - 1320.
[Abstract] [Full Text] [PDF]

Home page
 NEJMHome page
D. Mozaffarian, M. B. Katan, A. Ascherio, M. J. Stampfer, and W. C. Willett
Trans Fatty Acids and Cardiovascular Disease
N. Engl. J. Med., April 13, 2006; 354(15): 1601 - 1613.
[Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
E. E Mosley, A. L Wright, M. K McGuire, and M. A McGuire
trans Fatty acids in milk produced by women in the United States
Am. J. Clinical Nutrition, December 1, 2005; 82(6): 1292 - 1297.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
E. Lopez-Garcia, M. B. Schulze, J. B. Meigs, J. E. Manson, N. Rifai, M. J. Stampfer, W. C. Willett, and F. B. Hu
Consumption of Trans Fatty Acids Is Related to Plasma Biomarkers of Inflammation and Endothelial Dysfunction
J. Nutr., March 1, 2005; 135(3): 562 - 566.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Innis, S. M.
Right arrow Articles by Halsey, T. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Innis, S. M.
Right arrow Articles by Halsey, T. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS