- Email: [email protected]
Effect of trans fatty acid isomers from ruminant sources on risk factors of cardiovascular disease: Study design and rationale
Contemporary Clinical Trials 32 (2011) 569–576
Contents lists available at ScienceDirect
Contemporary Clinical Trials j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c o n c l i n t r i a l
Effect of trans fatty acid isomers from ruminant sources on risk factors of cardiovascular disease: Study design and rationale Sarah K. Gebauer a, Frederic Destaillats b, Zéphirin Mouloungui c,d, Laure Candy c,d, Jean-Baptiste Bezelgues e, Fabiola Dionisi b, David J. Baer a,⁎ a US Department of Agriculture, Agricultural Research Service, Beltsville Human Nutrition Research Center, Building 307B, Room 213, BARC-East, Beltsville, MD 20705, USA b Nestlé Research Center, Vers-chez-les-Blanc, P.O. Box 44 CH-1000 Lausanne 26, Switzerland c Université de Toulouse, INPT, LCA (Laboratoire de Chimie AgroIndustrielle), ENSIACET, BP 44362, 4 allées Emile Monso, F-31030 Toulouse, France d INRA, LCA (Laboratoire de Chimie AgroIndustrielle), F-31030 Toulouse, France e Nestlé Product Technology Center, Marysville, OH, USA
a r t i c l e
i n f o
Article history: Received 30 September 2010 Accepted 22 March 2011 Available online 31 March 2011 Keywords: Ruminant trans fatty acids Industrially produced trans fatty acids Vaccenic acid Conjugated linoleic acid Cardiovascular disease risk
a b s t r a c t Substantial evidence clearly demonstrates the deleterious effects of industrially-produced trans fatty acids (TFA); however, data are lacking from large, well controlled human feeding studies that directly compare the effects of industrially-produced and naturally-occurring TFA. The purpose of the current study is to determine whether consumption of TFA derived from different sources differentially affect risk factors of cardiovascular disease (CVD). The study was a randomized, crossover design, controlled-feeding intervention designed to compare the effects of the following diet treatments on risk factors of CVD: low TFA diet (base diet, 34% energy from fat; 0.1% energy from TFA), base diet with vaccenic acid (3.0% energy), base diet with mixed isomers of TFA from partially hydrogenated vegetable oil (3.0% energy), and base diet with cis-9, trans-11 CLA (1.0% energy). The added energy from TFA replaced energy from stearic acid. Participants were required to be between the ages of 25 and 65 years, have a body mass index between 20 and 38 kg/m2, total cholesterol b 280 mg/dl, fasting triacylglycerol b 300 mg/dl, fasting glucose b 126 mg/dl, and blood pressure b 160/100 mm Hg (controlled with certain medications). Of the 116 participants who were randomized, a total of 95 completed the intervention. Results from this study will be important in determining whether ruminant TFA and industrially produced TFA differentially affect markers of cardiovascular risk, in the context of a highly controlled feeding study. Published by Elsevier Inc.
1. Introduction Several clinical studies have indicated that intake of trans fatty acids (TFA) increases risk factors for cardiovascular disease (CVD) [1–4]. Specifically, TFA increase LDL-cholesterol (LDL-C), decrease HDL-cholesterol (HDL-C), and promote inflammation and endothelial dysfunction (reviewed in [5–7]). However, in most studies, the source of TFA is partially hydrogenated vegetable oils, or industrially-produced (iTFA). TFA also can be ⁎ Corresponding author. Tel.: + 1 301 504 8719; fax: + 1 301 504 9098. E-mail address: [email protected] (D.J. Baer). 1551-7144/$ – see front matter. Published by Elsevier Inc. doi:10.1016/j.cct.2011.03.012
found naturally in ruminant products (rTFA), such as dairy, beef, and lamb. In both food sources, the predominant TFA are 18carbon in length; however, the isomeric distribution of the fatty acids differs between the two sources. Elaidic acid (C18:1Δ9t) is typically the primary isomer in iTFA, while vaccenic acid (C18:1Δ11t), a precursor for conjugated linoleic acid (CLA), is the primary isomer found in ruminant fat. The two predominant isomers of CLA, which have known bioactive properties, are cis9, trans-11 and trans-10, cis-12. Data from observational studies suggest that rTFA and iTFA may have differential effects on risk of coronary heart disease (CHD). In contrast to the direct association with iTFA, some
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
studies have shown an inverse or no association with rTFA and CHD risk [8–11]. Human intervention studies investigating the effects of different sources of TFA on risk factors of CVD are limited and have produced mixed results. A recent review of human intervention studies suggests that all TFA, regardless of their source, adversely affect blood lipoproteins [12]. Results from animal studies suggest that CLA may be hypocholesterolemic and anti-atherogenic [13–18]. Some studies suggest that cis-9, trans-11 isomers and trans-10, cis-12 isomers have differential biological effects. Most of the CLA data are from animal studies and many of these studies have used mixtures of cis-9, trans-11 and trans-10, cis-12 CLA isomers. The study of specific CLA isomers on CVD risk factors in humans has been limited; however, some studies have shown isomer specific effects on lipids and lipoproteins [19,20]. In a study in healthy men (n = 49), intake of cis-9, trans-11 isomers decreased the ratios of LDL-C/HDL-C and total cholesterol/HDL-C (TC/HDL-C) [19]. While substantial evidence has clearly demonstrated the deleterious effects of iTFA, data are lacking from large scale, highly-controlled human feeding studies designed to directly compare the effects of iTFA and rTFA. The objective of the present study is to determine whether there are differential effects of iTFA and isomers from rTFA on risk factors of CVD in the context of a highly controlled, large scale feeding study. We compare the effects of the following TFA isomers on risk factors of CVD: diet containing approximately 0.1% energy from mixed TFA isomers (BASE), base diet with 3.0% energy from stearic acid replaced with 3.0% energy from vaccenic acid (VACCENIC), base diet with 3.0% energy from stearic acid replaced with 3.0% of energy from mixed isomers of TFA from partially hydrogenated vegetable oil (PHVO), and base diet with 1.0% energy from stearic acid replaced with 1.0% energy from cis-9, trans-11 CLA (RUMENIC). 2. Methods 2.1. Experimental fats The four experimental fats were produced by blending different oils and fats and incorporating them into various food items, including baked goods, sauces and spreads, and mashed potatoes. The fatty acid composition of the experimental fats is provided in Table 1. Physical blends were interesterified to obtain stable fats and to avoid the occurrence of triacylglycerols with high melting-points. The iTFA experimental fat was obtained by mixing PHVO (38.5%), palm oil stearin (19.0%), fully hydrogenated soybean oil (17.7%), sunflower oil (13.2%), high-oleic sunflower oil (5.0%), coconut oil (3.5%), flaxseed oil (1.6%), and cocoa butter (1.5%). The vaccenic acid experimental fat was obtained by mixing trivaccenin (20.6%), palm oil stearin (20.5%), fully hydrogenated soybean oil (22.0%), sunflower oil (11.7%), high-oleic sunflower oil (15.0%), coconut oil (3.8%), flaxseed oil (1.6%), and cocoa butter (4.8%). The stearic acid based experimental fat was obtained by mixing fully hydrogenated soybean oil (47.0%), palm oil stearin (16.3%), sunflower oil (11.5%), high-oleic sunflower oil (19.8%), coconut oil (3.8%), and flaxseed oil (1.6%). The cis-9, trans-11 CLA based experimental fat was obtained by mixing palm oil stearin (15.0%), fully hydrogenated soybean oil (35.0%), sunflower oil (12.9%), high-oleic sunflower oil (19.0%), CLA oil containing as
Table 1 Fatty acid composition (expressed as g per 100 g of fatty acid) of the test fats.
8:0 10:0 12:0 14:0 16:0 18:0 18:1 t 18:1 c 18:2 t 18:2 n-6 18:2 CLA 18:3 n-3 20:0 20:1 22:0 24:0
Base
VA
PHVO
c9, t11 CLA
0.3 0.2 1.9 1.0 16.7 43.8 0.1 23.7 0.1 9.7 0.0 0.8 0.5 0.2 0.5 0.1
0.2 0.2 1.7 1.0 17.5 23.9 19.6 22.7 0.4 10.0 0.0 0.8 0.6 0.2 0.4 0.1
0.3 0.2 1.8 1.0 17.6 24.2 19.4 23.3 0.1 9.9 0.0 0.8 0.4 0.1 0.4 0.1
0.3 0.3 2.2 1.1 16.6 33.0 0.1 25.7 0.2 11.0 7.0 0.9 0.3 0.1 0.4 0.1
VA = vaccenic acid, PHVO = partially hydrogenated vegetable oil, and CLA = conjugated linoleic acid. VA contains only vaccenic acid as the sole source of trans-18:1. PHVO contains trans-18:1 acid isomers from partially hydrogenated vegetable oil (see Fig. 1 for the isomeric distribution).
the main isomer cis-9, trans-11 18:2 acid (9.5%), coconut oil (4.0%), flaxseed oil (1.6%), and cocoa butter (3.0%); in addition, it contained residual amounts of palmitic, stearic, oleic, and minor CLA isomers. The CLA oil was obtained from Lipid Nutrition (Wormerveer, The Netherlands), fully hydrogenated soybean oil was obtained from SIO (Bougival, France), and PHVO was obtained from AAK (AarhusKarlshamn, Malmö, Sweden). All other fats and oils were from Nestlé factories. The distribution of the trans-18:1 acid isomers in the iTFA experimental fat is provided in Fig. 1. The detailed distribution of the trans-18:1 acid isomers was determined by analyzing the trans monoenoic acids fraction purified from the total fatty acid methyl esters by silver ion thin-layer chromatography, as previously described [21]. The predominant isomer found was elaidic acid (trans-9 18:1; 25.8%), followed by the sum of trans-6, trans-7, and trans-8 18:1
30.0 Relative distribution (%)
570
25.0 20.0 15.0 10.0 50.0 0.0
Δ4
Δ5 Δ6−8 Δ9
Δ10 Δ11 Δ12 Δ13/14 Δ15 Δ16
Isomeric distribution of trans-18:1 acids Fig. 1. Isomeric distribution of trans-18:1 acid isomers in the iTFA experimental fat (data expressed as relative % of total trans-18:1 acid isomers). Δ symbol indicates the position of the first carbon of the ethylenic double bond. The sum of trans-6, trans-7, and trans-8 18:1 isomers is abbreviated as Δ6–8.
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
isomers, abbreviated as trans-6–8 18:1 acid (23.7%), and trans-10 18:1 acid (20.6%). Vaccenic acid (trans-11 octadecenoic) was synthesized based on optimized literature procedures [22–25]. A schematic representation of the synthesis routes used to prepare vaccenic acid and the corresponding triacylglycerol is provided in Figs. 2 and 3. Cis and trans vaccenic acids were prepared by the Wittig reaction. Alternatively, these fatty acids also can be prepared by the Wittig–Horner reaction; however in the present study, good results and gentle catalytic conditions were obtained using the Wittig reaction. Briefly, 11-bromoundecanoic acid was methylated and converted into the corresponding phosphonium salt to react with heptanal. Potassium carbonate and crown ether were used as catalysts and the reaction led to the formation of mainly cis vaccenic methyl ester. The mixture of cis and trans vaccenic methyl esters was then isomerized according to procedures previously described in the literature [26], which resulted in a 80:20 mixture of trans and cis methyl ester isomers, respectively (Fig. 2). Trivaccenin was synthesized by glycerolysis from vaccenic acid, as described in Fig. 3. Briefly, methyl cis and trans vaccenate was saponified and fractionated to purify vaccenic acid. The fraction containing cis vaccenic acid was isomerized and (re)fractionated to prepare the trans isomer. A final purity of 96% (trans) vaccenic acid was obtained. Vaccenic acid was esterified with glycerol to obtain a mixture containing trivaccenin (93%) and divaccenin (7%). The final purity was measured by gas chromatography, which demonstrated that vaccenic acid (trans-11 18:1) was the predominant fatty acid (95%). In addition, trace amounts of cisvaccenic acid (cis-11 18:1) were detected (3.5%), as well as other fatty acids that represent about 1.5% of total fatty acids.
571
2.2. Study design The study was a double-blind, randomized, crossover, controlled-feeding intervention (Fig. 4). There were four 24day diet periods consisting of the following diets: base diet containing approximately 0.1% energy from mixed TFA isomers, base diet with 3.0% energy from stearic acid replaced with 3.0% energy from vaccenic acid VACCENIC, base diet with 3.0% energy from stearic acid replaced with 3.0% energy from mixed isomers of TFA from partially hydrogenated vegetable oil (PHVO), and base diet with 1.0% energy from stearic acid replaced with 1.0% energy from cis-9, trans-11 CLA (RUMENIC). The added energy from TFA replaced energy from stearic acid so that all diets contained 34% of energy from fat. There were no washout periods between diet periods. Diets were designed so that concentrations of dietary cholesterol, total saturated fat, total monounsaturated fat, total polyunsaturated fat, and fiber were similar across all test diets. The nutrient composition of the test diets is presented in Table 2. The test fats were provided by the Nestlé Research Center (Lausanne, Switzerland). Investigators, study participants, and all study personnel were blinded to the composition of the test fats throughout the duration of the intervention. This study is registered with ClinicalTrials.gov (NCT00942656, July 17, 2009). 2.3. Subject recruitment and selection Adult males and females were recruited from the Baltimore–Washington area to come in to the Beltsville Human Nutrition Research Center (BHNRC) for screening. During screening, individuals underwent a medical evaluation and provided a fasting blood sample and urine specimen
Fig. 2. Schematic representation of the steps involved in the synthesis of cis/trans methyl vaccenate.
572
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
Fig. 3. Schematic representation of the steps involved in the synthesis of trivaccenin.
who routinely participated in heavy exercise; volunteers who had lost 10% of body weight within 12 months prior to the start of the intervention or who planned to initiate a weight loss program; use of prescription or over-the-counter antiobesity medications or supplements (e.g., phenylpropanolamine, ephedrine, caffeine) during and for at least 6 months prior to the start of the study, or history of a surgical intervention for obesity; active cardiovascular disease (such as heart attack or procedure within 3 months prior to the start of the study, participation in a cardiac rehabilitation program within the past 3 months, stroke or history/ treatment for transient ischemic attacks in the past 3 months, or documented history of pulmonary embolus in the past 6 months); smokers or other tobacco users (within 6 months prior to the start of the study); individuals who were unable or unwilling to give informed consent or communicate with study staff; self-report of alcohol or substance abuse within 12 months prior to the start of the study, and/or current acute treatment or rehabilitation program for these problems; and
for clinical tests, including a profile of hematological and biochemical parameters. Participants were required to be between 25 and 65 years of age during the intervention, have a body mass index (BMI) between 20 and 38 kg/m2, fasting glucose b126 mg/dl, blood pressure b160/100 mm Hg (controlled with certain medications), total plasma cholesterol b280 mg/dl, and fasting triacylglycerol b300 mg/dl. Exclusion criteria included: use of prescription or over-the-counter medications or supplements that alter lipid metabolism; history or presence of kidney disease, liver disease, gout, certain cancers, thyroid disease, gastrointestinal disease, and other metabolic diseases or malabsorption syndromes; type 2 diabetes requiring the use of oral antidiabetic agents or insulin; history of eating disorders or other dietary patterns that are not consistent with the dietary intervention (i.e., vegetarian diets, very low-fat diets, high-protein diets); women who had given birth during 12 months prior to the start of the intervention, pregnant women, women who planned to become pregnant, or lactating women; volunteers
Cohort 1
Screening
Cohort 2
DP1
DP2
DP3
DP4
DP1
DP2
DP3
DP4
24 days
24 days
24 days
24 days
24 days
24 days
24 days
24 days
BD
BD
BD
BD
BD BD
BD
BD
BD
BD
Randomization Fig. 4. Study design of the dietary intervention. DP = diet period and BD = blood draw. Each participant was randomized to a sequence of the four test diets.
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
573
Table 2 Calculated nutrient composition of the test diets.
Energy (kcal) Protein (g) CHO (g) Total fat (g) SFA (g) MUFA (g) PUFA (g) Calcium (mg) Magnesium (mg) Phosphorus (mg) Potassium (mg) Sodium (mg) Cholesterol (mg) Dietary fiber (g) 18:0 (g) 18:1 t (g) 18:2 CLA (g) 18:1c (g) 18:2c (g) 18:3c (g) 20:4c (g) 22:6c (g)
Base diet
VA diet
PHVO diet
c9, t11 CLA diet
2007 85.2 (17.0) 249.5 (49.7) 77.9 (34.9) 36.1 (16.2) 22.7 (10.2) 11.7 (5.2) 592 249 1171 2449 2168 208.76 20.5 19.52 (8.8) 0.23 (0.1) 0.00 21.66 (9.7) 9.81 (4.4) 1.08 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 28.2 (12.7) 22.2 (10.0) 11.7 (5.2) 592 249 1170 2450 2185 208.76 20.6 11.81 (5.3) 7.31 (3.3) 0.00 21.23 (9.6) 9.77 (4.4) 1.07 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 28.2 (12.7) 22.0 (9.9) 11.7 (5.3) 592 249 1170 2450 2185 208.76 20.6 11.71 (5.3) 7.38 (3.3) 0.00 21.01 (9.5) 9.80 (4.4) 1.07 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 31.6 (14.2) 23.5 (10.6) 12.3 (5.5) 592 249 1170 2450 2185 208.76 20.6 15.21 (6.9) 0.23 (0.1) 1.96 (0.9) 22.55 (10.2) 10.35 (4.7) 1.12 (0.5) 0.09 (0.04) 0.12 (0.06)
Nutrient composition of the diets was calculated using ProNutra (Viocare, Princeton, NJ). Values presented are for a ~2000 kcal diet. Percent calories are presented in parentheses. VA = vaccenic acid, PHVO = partially hydrogenated vegetable oil, CLA = conjugated linoleic acid, CHO = carbohydrate, SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, and PUFA = polyunsaturated fatty acids.
other medical, psychiatric, or behavioral factors that in the judgment of the Principal Investigator may interfere with study participation or the ability to follow the intervention protocol. Inclusion and exclusion criteria were established to recruit individuals across a wide age range, who were generally healthy, but potentially at risk for CVD. Study participants were divided into 2 cohorts of approximate equal size (n= 60, n = 56) to facilitate the controlled diet intervention. Participants were stratified by gender and LDL-C from screening, and randomly assigned to a treatment sequence consisting of the 4 test diets (n= 116). Detailed recruitment statistics are presented in Fig. 5. Of the 116 individuals who were randomized, 95 participants completed the intervention. Reasons for dropping from the study included: schedule conflicts (n= 8), medical reasons unrelated to the study (n= 5), compliance issues (n= 2), and personal reasons (n= 6).
vitamins and minerals. Volunteers were fed on a 7-day menu cycle and consumed breakfast and dinner, Monday through Friday, in the dining facility of the BHNRC Human Studies Facility under the supervision of a dietitian, study investigator, or study personnel. Lunch and all weekend meals were packed for offsite consumption. Diets were designed so that dietary fat (34%), protein (17% energy), carbohydrate (49% energy), fiber (≥10 g/1000 kcal/day), and cholesterol (100 mg/1000 kcal/ day) were similar across all test diets. Alcohol consumption was prohibited during the intervention.
Attended Information Meeting (n = 363)
Signed Informed Consent (n = 269)
2.4. Controlled feeding intervention Study participants were fed at an energy intake necessary to maintain body weight. Body weight was measured before breakfast, Monday through Friday, and patterns of weight loss or weight gain over 7- to 10-day periods of time were identified. To maintain body weight, portion size was adjusted for all foods in 200 or 400 kcal increments to adjust energy intake such that the nutrient content of the diet was the same for all subjects regardless of energy intake. Using this approach, the absolute amount of TFA (and other nutrients) varied but the relative proportion (% of energy) of nutrients was identical for all subjects. Initial energy intake was estimated from calculated resting energy requirements adjusted for self-reported activity level. All menus were developed using typical American foods and provided 100% of the recommended daily allowances for
Completed Screening (n = 214)
Eligibile to participate (n = 161)
Randomized (n = 116)
Completed Intervention (n = 95)
Dropped (n = 21)
Fig. 5. Schema of recruitment and retention.
574
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
2.5. Diet composites During each diet period, diet samples were collected for each day of the menu cycle for 1 week. The samples were prepared for chemical analysis by homogenizing the food in a blender with ice and water, as described elsewhere [27], and then freeze-dried. Diets will be analyzed for fatty acids, carbohydrate, protein, and fiber (Medallion Labs, Minneapolis, MN) using AOAC methods. 2.6. Blood collections Blood samples were collected following an overnight fast (12 h) on 2 days, separated by at least 24 h, prior to the start of the intervention (baseline) and at the end of each treatment period. Blood was collected by venipuncture from an antecubital vein using an appropriate size needle. At each collection, 80 ml was collected for a total of 800 ml over the course of the 14-week study. Blood collected for serum and plasma was harvested and stored at −80 °C until completion of the study. 2.7. Laboratory analyses The primary outcome of the study is LDL-cholesterol. Secondary outcomes include other markers of cardiovascular risk: lipids and lipoproteins [total cholesterol, HDL-C, ratio of TC/HDL-C, triacylglycerol, Lp(a)], apolipoproteins (apoA, apoB), lipoprotein size, markers of inflammation (IL-6, hsCRP, fibrinogen, factor VII) and adhesion molecules (ICAM, VCAM, Eselectin). Glucose and insulin also will be measured. 2.8. Statistical methods We utilized an orthogonal Latin square design with participants stratified by their gender. Within each gender, there were 5 replicates of each sequence of 12 unique treatment sequences. Participants were randomly assigned to a sequence. The 12 sequences that were used were: BASE, RUMENIC, VACCENIC, PHVO; RUMENIC, BASE, PHVO, VACCENIC; VACCENIC, PHVO, BASE, RUMENIC; PHVO, VACCENIC, RUMENIC, BASE; BASE, PHVO, RUMENIC, VACCENIC; RUMENIC, VACCENIC, BASE, PHVO; VACCENIC, RUMENIC, PHVO, BASE; PHVO, BASE, VACCENIC, RUMENIC; BASE, VACCENIC, PHVO, RUMENIC; RUMENIC, PHVO, VACCENIC, BASE; VACCENIC, BASE, RUMENIC, PHVO; and PHVO, RUMENIC, BASE, VACCENIC. 2.8.1. Sample size calculation Sample size calculation is based on a lipid response study previously conducted at the BHNRC in which the effect of TFA on LDL-C was compared to that of stearic acid. Based on this study, we predict that there will be a 4.5 mg/dl change in LDL-C (with standard deviation of the change in LDL-C of 13.5 mg/dl, based on 5 studies of over 200 subjects) between the base diet and the diet with 3% energy added from PHVO. In order to detect this change with 95% probability and 90% power, the minimum sample size needed to complete the crossover study is 99. The sample size has to be increased by one to account for each model covariate (such as age, BMI, baseline LDL-C, gender, and their interactions). In addition, we accounted for a ~15% drop-out rate, which has been observed in previous studies that we have conducted with similar design.
2.8.2. Statistical analyses The mean of the two samples collected at each study time point will be used for statistical analyses. Data will be reviewed for outliers prior to statistical analyses. All statistical analyses will be performed using SAS (Statistical Analyses System, Cary, NC). The pre-treatment baseline values of the dependent variables will be used to adjust post-treatment values, resulting in a more sensitive test of the treatment effects due to the statistical removal of initial inter-subject variation. In addition to the effects of treatment and baseline values, other independent variables that will be tested include characteristics of the individual (BMI, sex), design variables (period and sequence in which the diets were administered), and interactive effects of baseline values with treatment, period, and sequence. The mixed model (PROC MIXED) will be used to determine whether effects are significant (P ≤ 0.05). 3. Discussion While some epidemiologic studies have suggested that rTFA and iTFA have differential effects on disease risk [8,10,28], others have shown comparable effects of the two TFA sources [29]. In a recent quantitative review of human intervention trials that investigated the effects of iTFA, rTFA, or CLA on lipids and lipoproteins, the authors conclude that all TFA, regardless of the source, increase the ratio of LDL-C/ HDL-C [12]. However, clinical studies designed to directly compare the consumption of TFA from different sources are very limited. Results from a crossover study in 38 healthy men suggest that at high intakes (10.2 g/2500 kcal, 3.7% energy), the effects of rTFA on CVD risk factors may be comparable to iTFA; however, at moderate intakes (4.2 g/ 2500 kcal, 1.5% energy), which would be potentially attainable with consumption of a very high dairy diet, rTFA may not significantly affect CVD risk factors [30]. It is important to note that a limitation of this study is that it was not statistically powered to detect changes between all of the diets; thus, it is possible that the study lacked the power to pick up an effect at moderate levels of intake. In addition, the study was conducted in a relatively young population (mean age = 32.8 ± 15 years) with normal BMI (23.6 ± 3.3 kg/m2), which is not representative of the typical population at risk for CVD. In a parallel design study directly comparing the two sources of TFA (n = 46; 11–12 g/day, ~5% of energy), the rTFA diet significantly increased LDL-C and HDL-C compared with the iTFA diet, but this effect was observed in women only [31]. A limitation of this study is that there was no control diet for which to compare the iTFA and rTFA diets. In addition, the study was not a controlled-feeding intervention, but rather a free-living study where dietary instruction was provided to the participants by a dietitian and dietary records were collected and analyzed. As with the study by Motard-Belanger et al. [30], this study was conducted in a relatively young population (mean age = 27.6 ± 7.1 years) with a normal BMI (22.0 ± 2.4 kg/m2). Thus, it is unclear how the effects of these dietary interventions would translate to at-risk populations. While these 2 studies have provided valuable data, particularly since they were the first feeding studies to directly compare iTFA and rTFA, further studies are warranted. Specifically, adequately powered, highly-controlled, feeding
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
studies are needed in order to effectively compare the effects of diets containing iTFA or rTFA, with a diet low in TFA. Animal studies have demonstrated beneficial effects of CLA on CVD risk factors, cancer, and body composition. However, data from human studies have been inconclusive and limited. In a recent crossover design study, the effects of CLA (80:20 mixture of cis-9, trans-11 and trans-10, cis-12) and iTFA were compared with oleic acid (control) in a healthy population (n= 61; mean age = 30.9 ± 13.7, BMI = 22.8 ± 3.2 kg/m2) [32]. Compared with the control diet, CLA significantly increased LDL-C and decreased HDL-C. When comparing the iTFA and CLA diets, the iTFA diet resulted in significantly higher LDL-C, triglycerides, and apolipoprotein B, while there were no differences between the diets for the ratios of TC/HDL-C and LDL-C/HDL-C. While this study was adequately powered, the CLA intake was very high and not attainable with a typical diet. Furthermore, the CLA diet did not contain pure cis-9, trans-11 isomers, but rather a mixture of cis-9, trans-11 (6.9% of energy) and trans-10, cis-12 isomers (1.5% of energy). Some studies have suggested that trans-10, cis-12 CLA may have adverse effects on lipids and lipoproteins compared with cis-9, trans-11 CLA [19]. Thus, additional controlled-feeding studies are needed to determine the effects of pure cis-9, trans-11 CLA on lipids and lipoproteins, at levels which can be achieved through diet. Due to insufficient data from human intervention studies, it remains unclear whether there are differential effects of rTFA and iTFA on CVD risk factors. Despite the need for further research, definitions and labeling requirements for TFA have been set by the Food and Drug Administration (FDA). As of January 2006, TFA are required to be listed on the Nutrition Facts Panel. Currently, the FDA defines TFA based on chemical structure, rather than metabolic or functional aspects. According to the FDA definition, TFA are defined as: unsaturated fatty acids that contain one or more isolated (i.e., non-conjugated) double bond in a trans configuration [33]. Therefore, CLA is not included on the food label; however, vaccenic acid, a nonconjugated TFA, and other rTFA with a single double bond or non-conjugated double bonds, are required to be listed on the food label. In Denmark, TFA are banned from being used in foods, but rTFA are excluded from the ban [34,35]. It is clear that additional studies are warranted in order to compare the health effects of rTFA and iTFA. The present study is designed to evaluate the effects of iTFA and rTFA, including vaccenic acid and cis-9, trans-11 CLA, in the context of a highly controlled dietary intervention. The results of the present study will be an important contribution to the limited existing database of human studies that compare the effects of rTFA and iTFA on risk factors of CVD. Role of the funding source This research was supported by the U.S. Department of Agriculture, The National Dairy Council, and Dairy Australia. The fat blends were provided by Nestlé. Conflict of interest SG and DB have no conflicts of interest to disclose. FD, JB, and FD are employed by Nestlé and ZM and LC by INRA. FD, ZM, LC, JB, and FD were involved in the synthesis of the fats
575
and provided details regarding the methodology used to prepare the fat blends and their composition. References [1] Ascherio A. Trans fatty acids and blood lipids. Atheroscler Suppl 2006;7: 25–7. [2] Judd JT, Baer DJ, Clevidence BA, Kris-Etherton P, Muesing RA, Iwane M. Dietary cis and trans monounsaturated and saturated FA and plasma lipids and lipoproteins in men. Lipids 2002;37:123–31. [3] Baer DJ, Judd JT, Clevidence BA, Tracy RP. Dietary fatty acids affect plasma markers of inflammation in healthy men fed controlled diets: a randomized crossover study. Am J Clin Nutr 2004;79:969–73. [4] Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr 2003;77:1146–55. [5] Gebauer SK, Psota TL, Kris-Etherton PM. The diversity of health effects of individual trans fatty acid isomers. Lipids 2007;42:787–99. [6] Mozaffarian D, Aro A, Willett WC. Health effects of trans-fatty acids: experimental and observational evidence. Eur J Clin Nutr 2009;63 (Suppl 2):S5–S21. [7] Wallace SK, Mozaffarian D. Trans-fatty acids and nonlipid risk factors. Curr Atheroscler Rep 2009;11:423–33. [8] Willett WC, Stampfer MJ, Manson JE, Colditz GA, Speizer FE, Rosner BA, et al. Intake of trans fatty acids and risk of coronary heart disease among women. Lancet 1993;341:581–5. [9] Ascherio A, Rimm EB, Giovannucci EL, Spiegelman D, Stampfer M, Willett WC. Dietary fat and risk of coronary heart disease in men: cohort follow up study in the United States. BMJ 1996;313:84–90. [10] Pietinen P, Ascherio A, Korhonen P, Hartman AM, Willett WC, Albanes D, et al. Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. Am J Epidemiol 1997;145:876–87. [11] Jakobsen MU, Overvad K, Dyerberg J, Heitmann BL. Intake of ruminant trans fatty acids and risk of coronary heart disease. Int J Epidemiol 2008;37:173–82. [12] Brouwer IA, Wanders AJ, Katan MB. Effect of animal and industrial trans fatty acids on HDL and LDL cholesterol levels in humans—a quantitative review. PLoS One 2010;5:e9434. [13] Lee KN, Kritchevsky D, Pariza MW. Conjugated linoleic acid and atherosclerosis in rabbits. Atherosclerosis 1994;108:19–25. [14] Nicolosi RJ, Rogers EJ, Kritchevsky D, Scimeca JA, Huth PJ. Dietary conjugated linoleic acid reduces plasma lipoproteins and early aortic atherosclerosis in hypercholesterolemic hamsters. Artery 1997;22: 266–77. [15] de Deckere EA, van Amelsvoort JM, McNeill GP, Jones P. Effects of conjugated linoleic acid (CLA) isomers on lipid levels and peroxisome proliferation in the hamster. Br J Nutr 1999;82:309–17. [16] Stangl GI. High dietary levels of a conjugated linoleic acid mixture alter hepatic glycerophospholipid class profile and cholesterol-carrying serum lipoproteins of rats. J Nutr Biochem 2000;11:184–91. [17] Gavino VC, Gavino G, Leblanc MJ, Tuchweber B. An isomeric mixture of conjugated linoleic acids but not pure cis-9, trans-11-octadecadienoic acid affects body weight gain and plasma lipids in hamsters. J Nutr 2000;130:27–9. [18] Sher J, Pronczuk A, Hajri T, Hayes KC. Dietary conjugated linoleic acid lowers plasma cholesterol during cholesterol supplementation, but accentuates the atherogenic lipid profile during the acute phase response in hamsters. J Nutr 2003;133:456–60. [19] Tricon S, Burdge GC, Kew S, Banerjee T, Russell JJ, Jones EL, et al. Opposing effects of cis-9, trans-11 and trans-10, cis-12 conjugated linoleic acid on blood lipids in healthy humans. Am J Clin Nutr 2004;80: 614–20. [20] Riserus U, Arner P, Brismar K, Vessby B. Treatment with dietary trans10 cis12 conjugated linoleic acid causes isomer-specific insulin resistance in obese men with the metabolic syndrome. Diabetes Care 2002;25:1516–21. [21] Destaillats F, Golay PA, Joffre F, de Wispelaere M, Hug B, Giuffrida F, et al. Comparison of available analytical methods to measure trans-octadecenoic acid isomeric profile and content by gas–liquid chromatography in milk fat. J Chromatogr A 2007;1145:222–8. [22] Bergelson LD, Shemyakin MM. Synthesis of naturally occurring unsaturated fatty acids by sterically controlled carbonyl olefination. Angew Chem Int Ed Engl 1964;3:250–60. [23] Gravier-Pelletier C, Dumas J, Le Merrer Y, Depezay JC. Methods for the total synthesis of acyclic hydroxylated fatty acids. Prog Lipid Res 1990;29:229–76. [24] Foglia TA, Vail PD. An efficient large-scale synthesis of triisopentadecanoin. Org Prep Proced Int 1993;25:209–13.
576
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
[25] Lawrence NJ. The Wittig reaction and related methods. In: Williams JJ, editor. Preparation of alkenes, a practical approach. Oxford: Oxford University Press; 1996. p. 19–58. [26] Grandgirard A, Sebedio JL, Fleury J. Geometrical isomerization of linolenic acid during heat treatment of vegetable oils. J Am Oil Chem Soc 1984;61:1563. [27] Chen SC, Judd JT, Kramer M, Meijer GW, Clevidence BA, Baer DJ. Phytosterol intake and dietary fat reduction are independent and additive in their ability to reduce plasma LDL cholesterol. Lipids 2009;44:273–81. [28] Ascherio A, Hennekens CH, Buring JE, Master C, Stampfer MJ, Willett WC. Trans-fatty acids intake and risk of myocardial infarction. Circulation 1994;89:94–101. [29] Oomen CM, Ocke MC, Feskens EJ, van Erp-Baart MA, Kok FJ, Kromhout D. Association between trans fatty acid intake and 10-year risk of coronary heart disease in the Zutphen Elderly Study: a prospective population-based study. Lancet 2001;357:746–51. [30] Motard-Belanger A, Charest A, Grenier G, Paquin P, Chouinard Y, Lemieux S, et al. Study of the effect of trans fatty acids from ruminants on blood lipids and other risk factors for cardiovascular disease. Am J Clin Nutr 2008;87:593–9.
[31] Chardigny JM, Destaillats F, Malpuech-Brugere C, Moulin J, Bauman DE, Lock AL, et al. Do trans fatty acids from industrially produced sources and from natural sources have the same effect on cardiovascular disease risk factors in healthy subjects? Results of the trans Fatty Acids Collaboration (TRANSFACT) study. Am J Clin Nutr 2008;87:558–66. [32] Wanders AJ, Brouwer IA, Siebelink E, Katan MB. Effect of a high intake of conjugated linoleic acid on lipoprotein levels in healthy human subjects. PLoS One 2010;5:e9000. [33] U.S. Food and Drug Administration. Federal Register — 68 FR 41433 July 11, 2003: Food Labeling; Trans Fatty Acids in Nutrition Labeling; Consumer Research to Consider Nutrient Content and Health Claims and Possible Footnote or Disclosure Statements; Final Rule and Proposed Rule. Accessed from: http://www.fda.gov/Food/LabelingNutrition/ LabelClaims/NutrientContentClaims/ucm110179.htm. Accessed September 29, 2010. [34] Stender S, Dyerberg J. Influence of trans fatty acids on health. Ann Nutr Metab 2004;48:61–6. [35] Stender S, Dyerberg J, Astrup A. Consumer protection through a legislative ban on industrially produced trans fatty acids in foods in Denmark. Scand J Food Nutr 2006;50:155–60.
Contents lists available at ScienceDirect
Contemporary Clinical Trials j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c o n c l i n t r i a l
Effect of trans fatty acid isomers from ruminant sources on risk factors of cardiovascular disease: Study design and rationale Sarah K. Gebauer a, Frederic Destaillats b, Zéphirin Mouloungui c,d, Laure Candy c,d, Jean-Baptiste Bezelgues e, Fabiola Dionisi b, David J. Baer a,⁎ a US Department of Agriculture, Agricultural Research Service, Beltsville Human Nutrition Research Center, Building 307B, Room 213, BARC-East, Beltsville, MD 20705, USA b Nestlé Research Center, Vers-chez-les-Blanc, P.O. Box 44 CH-1000 Lausanne 26, Switzerland c Université de Toulouse, INPT, LCA (Laboratoire de Chimie AgroIndustrielle), ENSIACET, BP 44362, 4 allées Emile Monso, F-31030 Toulouse, France d INRA, LCA (Laboratoire de Chimie AgroIndustrielle), F-31030 Toulouse, France e Nestlé Product Technology Center, Marysville, OH, USA
a r t i c l e
i n f o
Article history: Received 30 September 2010 Accepted 22 March 2011 Available online 31 March 2011 Keywords: Ruminant trans fatty acids Industrially produced trans fatty acids Vaccenic acid Conjugated linoleic acid Cardiovascular disease risk
a b s t r a c t Substantial evidence clearly demonstrates the deleterious effects of industrially-produced trans fatty acids (TFA); however, data are lacking from large, well controlled human feeding studies that directly compare the effects of industrially-produced and naturally-occurring TFA. The purpose of the current study is to determine whether consumption of TFA derived from different sources differentially affect risk factors of cardiovascular disease (CVD). The study was a randomized, crossover design, controlled-feeding intervention designed to compare the effects of the following diet treatments on risk factors of CVD: low TFA diet (base diet, 34% energy from fat; 0.1% energy from TFA), base diet with vaccenic acid (3.0% energy), base diet with mixed isomers of TFA from partially hydrogenated vegetable oil (3.0% energy), and base diet with cis-9, trans-11 CLA (1.0% energy). The added energy from TFA replaced energy from stearic acid. Participants were required to be between the ages of 25 and 65 years, have a body mass index between 20 and 38 kg/m2, total cholesterol b 280 mg/dl, fasting triacylglycerol b 300 mg/dl, fasting glucose b 126 mg/dl, and blood pressure b 160/100 mm Hg (controlled with certain medications). Of the 116 participants who were randomized, a total of 95 completed the intervention. Results from this study will be important in determining whether ruminant TFA and industrially produced TFA differentially affect markers of cardiovascular risk, in the context of a highly controlled feeding study. Published by Elsevier Inc.
1. Introduction Several clinical studies have indicated that intake of trans fatty acids (TFA) increases risk factors for cardiovascular disease (CVD) [1–4]. Specifically, TFA increase LDL-cholesterol (LDL-C), decrease HDL-cholesterol (HDL-C), and promote inflammation and endothelial dysfunction (reviewed in [5–7]). However, in most studies, the source of TFA is partially hydrogenated vegetable oils, or industrially-produced (iTFA). TFA also can be ⁎ Corresponding author. Tel.: + 1 301 504 8719; fax: + 1 301 504 9098. E-mail address: [email protected] (D.J. Baer). 1551-7144/$ – see front matter. Published by Elsevier Inc. doi:10.1016/j.cct.2011.03.012
found naturally in ruminant products (rTFA), such as dairy, beef, and lamb. In both food sources, the predominant TFA are 18carbon in length; however, the isomeric distribution of the fatty acids differs between the two sources. Elaidic acid (C18:1Δ9t) is typically the primary isomer in iTFA, while vaccenic acid (C18:1Δ11t), a precursor for conjugated linoleic acid (CLA), is the primary isomer found in ruminant fat. The two predominant isomers of CLA, which have known bioactive properties, are cis9, trans-11 and trans-10, cis-12. Data from observational studies suggest that rTFA and iTFA may have differential effects on risk of coronary heart disease (CHD). In contrast to the direct association with iTFA, some
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
studies have shown an inverse or no association with rTFA and CHD risk [8–11]. Human intervention studies investigating the effects of different sources of TFA on risk factors of CVD are limited and have produced mixed results. A recent review of human intervention studies suggests that all TFA, regardless of their source, adversely affect blood lipoproteins [12]. Results from animal studies suggest that CLA may be hypocholesterolemic and anti-atherogenic [13–18]. Some studies suggest that cis-9, trans-11 isomers and trans-10, cis-12 isomers have differential biological effects. Most of the CLA data are from animal studies and many of these studies have used mixtures of cis-9, trans-11 and trans-10, cis-12 CLA isomers. The study of specific CLA isomers on CVD risk factors in humans has been limited; however, some studies have shown isomer specific effects on lipids and lipoproteins [19,20]. In a study in healthy men (n = 49), intake of cis-9, trans-11 isomers decreased the ratios of LDL-C/HDL-C and total cholesterol/HDL-C (TC/HDL-C) [19]. While substantial evidence has clearly demonstrated the deleterious effects of iTFA, data are lacking from large scale, highly-controlled human feeding studies designed to directly compare the effects of iTFA and rTFA. The objective of the present study is to determine whether there are differential effects of iTFA and isomers from rTFA on risk factors of CVD in the context of a highly controlled, large scale feeding study. We compare the effects of the following TFA isomers on risk factors of CVD: diet containing approximately 0.1% energy from mixed TFA isomers (BASE), base diet with 3.0% energy from stearic acid replaced with 3.0% energy from vaccenic acid (VACCENIC), base diet with 3.0% energy from stearic acid replaced with 3.0% of energy from mixed isomers of TFA from partially hydrogenated vegetable oil (PHVO), and base diet with 1.0% energy from stearic acid replaced with 1.0% energy from cis-9, trans-11 CLA (RUMENIC). 2. Methods 2.1. Experimental fats The four experimental fats were produced by blending different oils and fats and incorporating them into various food items, including baked goods, sauces and spreads, and mashed potatoes. The fatty acid composition of the experimental fats is provided in Table 1. Physical blends were interesterified to obtain stable fats and to avoid the occurrence of triacylglycerols with high melting-points. The iTFA experimental fat was obtained by mixing PHVO (38.5%), palm oil stearin (19.0%), fully hydrogenated soybean oil (17.7%), sunflower oil (13.2%), high-oleic sunflower oil (5.0%), coconut oil (3.5%), flaxseed oil (1.6%), and cocoa butter (1.5%). The vaccenic acid experimental fat was obtained by mixing trivaccenin (20.6%), palm oil stearin (20.5%), fully hydrogenated soybean oil (22.0%), sunflower oil (11.7%), high-oleic sunflower oil (15.0%), coconut oil (3.8%), flaxseed oil (1.6%), and cocoa butter (4.8%). The stearic acid based experimental fat was obtained by mixing fully hydrogenated soybean oil (47.0%), palm oil stearin (16.3%), sunflower oil (11.5%), high-oleic sunflower oil (19.8%), coconut oil (3.8%), and flaxseed oil (1.6%). The cis-9, trans-11 CLA based experimental fat was obtained by mixing palm oil stearin (15.0%), fully hydrogenated soybean oil (35.0%), sunflower oil (12.9%), high-oleic sunflower oil (19.0%), CLA oil containing as
Table 1 Fatty acid composition (expressed as g per 100 g of fatty acid) of the test fats.
8:0 10:0 12:0 14:0 16:0 18:0 18:1 t 18:1 c 18:2 t 18:2 n-6 18:2 CLA 18:3 n-3 20:0 20:1 22:0 24:0
Base
VA
PHVO
c9, t11 CLA
0.3 0.2 1.9 1.0 16.7 43.8 0.1 23.7 0.1 9.7 0.0 0.8 0.5 0.2 0.5 0.1
0.2 0.2 1.7 1.0 17.5 23.9 19.6 22.7 0.4 10.0 0.0 0.8 0.6 0.2 0.4 0.1
0.3 0.2 1.8 1.0 17.6 24.2 19.4 23.3 0.1 9.9 0.0 0.8 0.4 0.1 0.4 0.1
0.3 0.3 2.2 1.1 16.6 33.0 0.1 25.7 0.2 11.0 7.0 0.9 0.3 0.1 0.4 0.1
VA = vaccenic acid, PHVO = partially hydrogenated vegetable oil, and CLA = conjugated linoleic acid. VA contains only vaccenic acid as the sole source of trans-18:1. PHVO contains trans-18:1 acid isomers from partially hydrogenated vegetable oil (see Fig. 1 for the isomeric distribution).
the main isomer cis-9, trans-11 18:2 acid (9.5%), coconut oil (4.0%), flaxseed oil (1.6%), and cocoa butter (3.0%); in addition, it contained residual amounts of palmitic, stearic, oleic, and minor CLA isomers. The CLA oil was obtained from Lipid Nutrition (Wormerveer, The Netherlands), fully hydrogenated soybean oil was obtained from SIO (Bougival, France), and PHVO was obtained from AAK (AarhusKarlshamn, Malmö, Sweden). All other fats and oils were from Nestlé factories. The distribution of the trans-18:1 acid isomers in the iTFA experimental fat is provided in Fig. 1. The detailed distribution of the trans-18:1 acid isomers was determined by analyzing the trans monoenoic acids fraction purified from the total fatty acid methyl esters by silver ion thin-layer chromatography, as previously described [21]. The predominant isomer found was elaidic acid (trans-9 18:1; 25.8%), followed by the sum of trans-6, trans-7, and trans-8 18:1
30.0 Relative distribution (%)
570
25.0 20.0 15.0 10.0 50.0 0.0
Δ4
Δ5 Δ6−8 Δ9
Δ10 Δ11 Δ12 Δ13/14 Δ15 Δ16
Isomeric distribution of trans-18:1 acids Fig. 1. Isomeric distribution of trans-18:1 acid isomers in the iTFA experimental fat (data expressed as relative % of total trans-18:1 acid isomers). Δ symbol indicates the position of the first carbon of the ethylenic double bond. The sum of trans-6, trans-7, and trans-8 18:1 isomers is abbreviated as Δ6–8.
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
isomers, abbreviated as trans-6–8 18:1 acid (23.7%), and trans-10 18:1 acid (20.6%). Vaccenic acid (trans-11 octadecenoic) was synthesized based on optimized literature procedures [22–25]. A schematic representation of the synthesis routes used to prepare vaccenic acid and the corresponding triacylglycerol is provided in Figs. 2 and 3. Cis and trans vaccenic acids were prepared by the Wittig reaction. Alternatively, these fatty acids also can be prepared by the Wittig–Horner reaction; however in the present study, good results and gentle catalytic conditions were obtained using the Wittig reaction. Briefly, 11-bromoundecanoic acid was methylated and converted into the corresponding phosphonium salt to react with heptanal. Potassium carbonate and crown ether were used as catalysts and the reaction led to the formation of mainly cis vaccenic methyl ester. The mixture of cis and trans vaccenic methyl esters was then isomerized according to procedures previously described in the literature [26], which resulted in a 80:20 mixture of trans and cis methyl ester isomers, respectively (Fig. 2). Trivaccenin was synthesized by glycerolysis from vaccenic acid, as described in Fig. 3. Briefly, methyl cis and trans vaccenate was saponified and fractionated to purify vaccenic acid. The fraction containing cis vaccenic acid was isomerized and (re)fractionated to prepare the trans isomer. A final purity of 96% (trans) vaccenic acid was obtained. Vaccenic acid was esterified with glycerol to obtain a mixture containing trivaccenin (93%) and divaccenin (7%). The final purity was measured by gas chromatography, which demonstrated that vaccenic acid (trans-11 18:1) was the predominant fatty acid (95%). In addition, trace amounts of cisvaccenic acid (cis-11 18:1) were detected (3.5%), as well as other fatty acids that represent about 1.5% of total fatty acids.
571
2.2. Study design The study was a double-blind, randomized, crossover, controlled-feeding intervention (Fig. 4). There were four 24day diet periods consisting of the following diets: base diet containing approximately 0.1% energy from mixed TFA isomers, base diet with 3.0% energy from stearic acid replaced with 3.0% energy from vaccenic acid VACCENIC, base diet with 3.0% energy from stearic acid replaced with 3.0% energy from mixed isomers of TFA from partially hydrogenated vegetable oil (PHVO), and base diet with 1.0% energy from stearic acid replaced with 1.0% energy from cis-9, trans-11 CLA (RUMENIC). The added energy from TFA replaced energy from stearic acid so that all diets contained 34% of energy from fat. There were no washout periods between diet periods. Diets were designed so that concentrations of dietary cholesterol, total saturated fat, total monounsaturated fat, total polyunsaturated fat, and fiber were similar across all test diets. The nutrient composition of the test diets is presented in Table 2. The test fats were provided by the Nestlé Research Center (Lausanne, Switzerland). Investigators, study participants, and all study personnel were blinded to the composition of the test fats throughout the duration of the intervention. This study is registered with ClinicalTrials.gov (NCT00942656, July 17, 2009). 2.3. Subject recruitment and selection Adult males and females were recruited from the Baltimore–Washington area to come in to the Beltsville Human Nutrition Research Center (BHNRC) for screening. During screening, individuals underwent a medical evaluation and provided a fasting blood sample and urine specimen
Fig. 2. Schematic representation of the steps involved in the synthesis of cis/trans methyl vaccenate.
572
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
Fig. 3. Schematic representation of the steps involved in the synthesis of trivaccenin.
who routinely participated in heavy exercise; volunteers who had lost 10% of body weight within 12 months prior to the start of the intervention or who planned to initiate a weight loss program; use of prescription or over-the-counter antiobesity medications or supplements (e.g., phenylpropanolamine, ephedrine, caffeine) during and for at least 6 months prior to the start of the study, or history of a surgical intervention for obesity; active cardiovascular disease (such as heart attack or procedure within 3 months prior to the start of the study, participation in a cardiac rehabilitation program within the past 3 months, stroke or history/ treatment for transient ischemic attacks in the past 3 months, or documented history of pulmonary embolus in the past 6 months); smokers or other tobacco users (within 6 months prior to the start of the study); individuals who were unable or unwilling to give informed consent or communicate with study staff; self-report of alcohol or substance abuse within 12 months prior to the start of the study, and/or current acute treatment or rehabilitation program for these problems; and
for clinical tests, including a profile of hematological and biochemical parameters. Participants were required to be between 25 and 65 years of age during the intervention, have a body mass index (BMI) between 20 and 38 kg/m2, fasting glucose b126 mg/dl, blood pressure b160/100 mm Hg (controlled with certain medications), total plasma cholesterol b280 mg/dl, and fasting triacylglycerol b300 mg/dl. Exclusion criteria included: use of prescription or over-the-counter medications or supplements that alter lipid metabolism; history or presence of kidney disease, liver disease, gout, certain cancers, thyroid disease, gastrointestinal disease, and other metabolic diseases or malabsorption syndromes; type 2 diabetes requiring the use of oral antidiabetic agents or insulin; history of eating disorders or other dietary patterns that are not consistent with the dietary intervention (i.e., vegetarian diets, very low-fat diets, high-protein diets); women who had given birth during 12 months prior to the start of the intervention, pregnant women, women who planned to become pregnant, or lactating women; volunteers
Cohort 1
Screening
Cohort 2
DP1
DP2
DP3
DP4
DP1
DP2
DP3
DP4
24 days
24 days
24 days
24 days
24 days
24 days
24 days
24 days
BD
BD
BD
BD
BD BD
BD
BD
BD
BD
Randomization Fig. 4. Study design of the dietary intervention. DP = diet period and BD = blood draw. Each participant was randomized to a sequence of the four test diets.
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
573
Table 2 Calculated nutrient composition of the test diets.
Energy (kcal) Protein (g) CHO (g) Total fat (g) SFA (g) MUFA (g) PUFA (g) Calcium (mg) Magnesium (mg) Phosphorus (mg) Potassium (mg) Sodium (mg) Cholesterol (mg) Dietary fiber (g) 18:0 (g) 18:1 t (g) 18:2 CLA (g) 18:1c (g) 18:2c (g) 18:3c (g) 20:4c (g) 22:6c (g)
Base diet
VA diet
PHVO diet
c9, t11 CLA diet
2007 85.2 (17.0) 249.5 (49.7) 77.9 (34.9) 36.1 (16.2) 22.7 (10.2) 11.7 (5.2) 592 249 1171 2449 2168 208.76 20.5 19.52 (8.8) 0.23 (0.1) 0.00 21.66 (9.7) 9.81 (4.4) 1.08 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 28.2 (12.7) 22.2 (10.0) 11.7 (5.2) 592 249 1170 2450 2185 208.76 20.6 11.81 (5.3) 7.31 (3.3) 0.00 21.23 (9.6) 9.77 (4.4) 1.07 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 28.2 (12.7) 22.0 (9.9) 11.7 (5.3) 592 249 1170 2450 2185 208.76 20.6 11.71 (5.3) 7.38 (3.3) 0.00 21.01 (9.5) 9.80 (4.4) 1.07 (0.5) 0.09 (0.04) 0.12 (0.06)
1995 85.2 (17.1) 249.6 (50.0) 76.6 (34.5) 31.6 (14.2) 23.5 (10.6) 12.3 (5.5) 592 249 1170 2450 2185 208.76 20.6 15.21 (6.9) 0.23 (0.1) 1.96 (0.9) 22.55 (10.2) 10.35 (4.7) 1.12 (0.5) 0.09 (0.04) 0.12 (0.06)
Nutrient composition of the diets was calculated using ProNutra (Viocare, Princeton, NJ). Values presented are for a ~2000 kcal diet. Percent calories are presented in parentheses. VA = vaccenic acid, PHVO = partially hydrogenated vegetable oil, CLA = conjugated linoleic acid, CHO = carbohydrate, SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, and PUFA = polyunsaturated fatty acids.
other medical, psychiatric, or behavioral factors that in the judgment of the Principal Investigator may interfere with study participation or the ability to follow the intervention protocol. Inclusion and exclusion criteria were established to recruit individuals across a wide age range, who were generally healthy, but potentially at risk for CVD. Study participants were divided into 2 cohorts of approximate equal size (n= 60, n = 56) to facilitate the controlled diet intervention. Participants were stratified by gender and LDL-C from screening, and randomly assigned to a treatment sequence consisting of the 4 test diets (n= 116). Detailed recruitment statistics are presented in Fig. 5. Of the 116 individuals who were randomized, 95 participants completed the intervention. Reasons for dropping from the study included: schedule conflicts (n= 8), medical reasons unrelated to the study (n= 5), compliance issues (n= 2), and personal reasons (n= 6).
vitamins and minerals. Volunteers were fed on a 7-day menu cycle and consumed breakfast and dinner, Monday through Friday, in the dining facility of the BHNRC Human Studies Facility under the supervision of a dietitian, study investigator, or study personnel. Lunch and all weekend meals were packed for offsite consumption. Diets were designed so that dietary fat (34%), protein (17% energy), carbohydrate (49% energy), fiber (≥10 g/1000 kcal/day), and cholesterol (100 mg/1000 kcal/ day) were similar across all test diets. Alcohol consumption was prohibited during the intervention.
Attended Information Meeting (n = 363)
Signed Informed Consent (n = 269)
2.4. Controlled feeding intervention Study participants were fed at an energy intake necessary to maintain body weight. Body weight was measured before breakfast, Monday through Friday, and patterns of weight loss or weight gain over 7- to 10-day periods of time were identified. To maintain body weight, portion size was adjusted for all foods in 200 or 400 kcal increments to adjust energy intake such that the nutrient content of the diet was the same for all subjects regardless of energy intake. Using this approach, the absolute amount of TFA (and other nutrients) varied but the relative proportion (% of energy) of nutrients was identical for all subjects. Initial energy intake was estimated from calculated resting energy requirements adjusted for self-reported activity level. All menus were developed using typical American foods and provided 100% of the recommended daily allowances for
Completed Screening (n = 214)
Eligibile to participate (n = 161)
Randomized (n = 116)
Completed Intervention (n = 95)
Dropped (n = 21)
Fig. 5. Schema of recruitment and retention.
574
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
2.5. Diet composites During each diet period, diet samples were collected for each day of the menu cycle for 1 week. The samples were prepared for chemical analysis by homogenizing the food in a blender with ice and water, as described elsewhere [27], and then freeze-dried. Diets will be analyzed for fatty acids, carbohydrate, protein, and fiber (Medallion Labs, Minneapolis, MN) using AOAC methods. 2.6. Blood collections Blood samples were collected following an overnight fast (12 h) on 2 days, separated by at least 24 h, prior to the start of the intervention (baseline) and at the end of each treatment period. Blood was collected by venipuncture from an antecubital vein using an appropriate size needle. At each collection, 80 ml was collected for a total of 800 ml over the course of the 14-week study. Blood collected for serum and plasma was harvested and stored at −80 °C until completion of the study. 2.7. Laboratory analyses The primary outcome of the study is LDL-cholesterol. Secondary outcomes include other markers of cardiovascular risk: lipids and lipoproteins [total cholesterol, HDL-C, ratio of TC/HDL-C, triacylglycerol, Lp(a)], apolipoproteins (apoA, apoB), lipoprotein size, markers of inflammation (IL-6, hsCRP, fibrinogen, factor VII) and adhesion molecules (ICAM, VCAM, Eselectin). Glucose and insulin also will be measured. 2.8. Statistical methods We utilized an orthogonal Latin square design with participants stratified by their gender. Within each gender, there were 5 replicates of each sequence of 12 unique treatment sequences. Participants were randomly assigned to a sequence. The 12 sequences that were used were: BASE, RUMENIC, VACCENIC, PHVO; RUMENIC, BASE, PHVO, VACCENIC; VACCENIC, PHVO, BASE, RUMENIC; PHVO, VACCENIC, RUMENIC, BASE; BASE, PHVO, RUMENIC, VACCENIC; RUMENIC, VACCENIC, BASE, PHVO; VACCENIC, RUMENIC, PHVO, BASE; PHVO, BASE, VACCENIC, RUMENIC; BASE, VACCENIC, PHVO, RUMENIC; RUMENIC, PHVO, VACCENIC, BASE; VACCENIC, BASE, RUMENIC, PHVO; and PHVO, RUMENIC, BASE, VACCENIC. 2.8.1. Sample size calculation Sample size calculation is based on a lipid response study previously conducted at the BHNRC in which the effect of TFA on LDL-C was compared to that of stearic acid. Based on this study, we predict that there will be a 4.5 mg/dl change in LDL-C (with standard deviation of the change in LDL-C of 13.5 mg/dl, based on 5 studies of over 200 subjects) between the base diet and the diet with 3% energy added from PHVO. In order to detect this change with 95% probability and 90% power, the minimum sample size needed to complete the crossover study is 99. The sample size has to be increased by one to account for each model covariate (such as age, BMI, baseline LDL-C, gender, and their interactions). In addition, we accounted for a ~15% drop-out rate, which has been observed in previous studies that we have conducted with similar design.
2.8.2. Statistical analyses The mean of the two samples collected at each study time point will be used for statistical analyses. Data will be reviewed for outliers prior to statistical analyses. All statistical analyses will be performed using SAS (Statistical Analyses System, Cary, NC). The pre-treatment baseline values of the dependent variables will be used to adjust post-treatment values, resulting in a more sensitive test of the treatment effects due to the statistical removal of initial inter-subject variation. In addition to the effects of treatment and baseline values, other independent variables that will be tested include characteristics of the individual (BMI, sex), design variables (period and sequence in which the diets were administered), and interactive effects of baseline values with treatment, period, and sequence. The mixed model (PROC MIXED) will be used to determine whether effects are significant (P ≤ 0.05). 3. Discussion While some epidemiologic studies have suggested that rTFA and iTFA have differential effects on disease risk [8,10,28], others have shown comparable effects of the two TFA sources [29]. In a recent quantitative review of human intervention trials that investigated the effects of iTFA, rTFA, or CLA on lipids and lipoproteins, the authors conclude that all TFA, regardless of the source, increase the ratio of LDL-C/ HDL-C [12]. However, clinical studies designed to directly compare the consumption of TFA from different sources are very limited. Results from a crossover study in 38 healthy men suggest that at high intakes (10.2 g/2500 kcal, 3.7% energy), the effects of rTFA on CVD risk factors may be comparable to iTFA; however, at moderate intakes (4.2 g/ 2500 kcal, 1.5% energy), which would be potentially attainable with consumption of a very high dairy diet, rTFA may not significantly affect CVD risk factors [30]. It is important to note that a limitation of this study is that it was not statistically powered to detect changes between all of the diets; thus, it is possible that the study lacked the power to pick up an effect at moderate levels of intake. In addition, the study was conducted in a relatively young population (mean age = 32.8 ± 15 years) with normal BMI (23.6 ± 3.3 kg/m2), which is not representative of the typical population at risk for CVD. In a parallel design study directly comparing the two sources of TFA (n = 46; 11–12 g/day, ~5% of energy), the rTFA diet significantly increased LDL-C and HDL-C compared with the iTFA diet, but this effect was observed in women only [31]. A limitation of this study is that there was no control diet for which to compare the iTFA and rTFA diets. In addition, the study was not a controlled-feeding intervention, but rather a free-living study where dietary instruction was provided to the participants by a dietitian and dietary records were collected and analyzed. As with the study by Motard-Belanger et al. [30], this study was conducted in a relatively young population (mean age = 27.6 ± 7.1 years) with a normal BMI (22.0 ± 2.4 kg/m2). Thus, it is unclear how the effects of these dietary interventions would translate to at-risk populations. While these 2 studies have provided valuable data, particularly since they were the first feeding studies to directly compare iTFA and rTFA, further studies are warranted. Specifically, adequately powered, highly-controlled, feeding
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
studies are needed in order to effectively compare the effects of diets containing iTFA or rTFA, with a diet low in TFA. Animal studies have demonstrated beneficial effects of CLA on CVD risk factors, cancer, and body composition. However, data from human studies have been inconclusive and limited. In a recent crossover design study, the effects of CLA (80:20 mixture of cis-9, trans-11 and trans-10, cis-12) and iTFA were compared with oleic acid (control) in a healthy population (n= 61; mean age = 30.9 ± 13.7, BMI = 22.8 ± 3.2 kg/m2) [32]. Compared with the control diet, CLA significantly increased LDL-C and decreased HDL-C. When comparing the iTFA and CLA diets, the iTFA diet resulted in significantly higher LDL-C, triglycerides, and apolipoprotein B, while there were no differences between the diets for the ratios of TC/HDL-C and LDL-C/HDL-C. While this study was adequately powered, the CLA intake was very high and not attainable with a typical diet. Furthermore, the CLA diet did not contain pure cis-9, trans-11 isomers, but rather a mixture of cis-9, trans-11 (6.9% of energy) and trans-10, cis-12 isomers (1.5% of energy). Some studies have suggested that trans-10, cis-12 CLA may have adverse effects on lipids and lipoproteins compared with cis-9, trans-11 CLA [19]. Thus, additional controlled-feeding studies are needed to determine the effects of pure cis-9, trans-11 CLA on lipids and lipoproteins, at levels which can be achieved through diet. Due to insufficient data from human intervention studies, it remains unclear whether there are differential effects of rTFA and iTFA on CVD risk factors. Despite the need for further research, definitions and labeling requirements for TFA have been set by the Food and Drug Administration (FDA). As of January 2006, TFA are required to be listed on the Nutrition Facts Panel. Currently, the FDA defines TFA based on chemical structure, rather than metabolic or functional aspects. According to the FDA definition, TFA are defined as: unsaturated fatty acids that contain one or more isolated (i.e., non-conjugated) double bond in a trans configuration [33]. Therefore, CLA is not included on the food label; however, vaccenic acid, a nonconjugated TFA, and other rTFA with a single double bond or non-conjugated double bonds, are required to be listed on the food label. In Denmark, TFA are banned from being used in foods, but rTFA are excluded from the ban [34,35]. It is clear that additional studies are warranted in order to compare the health effects of rTFA and iTFA. The present study is designed to evaluate the effects of iTFA and rTFA, including vaccenic acid and cis-9, trans-11 CLA, in the context of a highly controlled dietary intervention. The results of the present study will be an important contribution to the limited existing database of human studies that compare the effects of rTFA and iTFA on risk factors of CVD. Role of the funding source This research was supported by the U.S. Department of Agriculture, The National Dairy Council, and Dairy Australia. The fat blends were provided by Nestlé. Conflict of interest SG and DB have no conflicts of interest to disclose. FD, JB, and FD are employed by Nestlé and ZM and LC by INRA. FD, ZM, LC, JB, and FD were involved in the synthesis of the fats
575
and provided details regarding the methodology used to prepare the fat blends and their composition. References [1] Ascherio A. Trans fatty acids and blood lipids. Atheroscler Suppl 2006;7: 25–7. [2] Judd JT, Baer DJ, Clevidence BA, Kris-Etherton P, Muesing RA, Iwane M. Dietary cis and trans monounsaturated and saturated FA and plasma lipids and lipoproteins in men. Lipids 2002;37:123–31. [3] Baer DJ, Judd JT, Clevidence BA, Tracy RP. Dietary fatty acids affect plasma markers of inflammation in healthy men fed controlled diets: a randomized crossover study. Am J Clin Nutr 2004;79:969–73. [4] Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr 2003;77:1146–55. [5] Gebauer SK, Psota TL, Kris-Etherton PM. The diversity of health effects of individual trans fatty acid isomers. Lipids 2007;42:787–99. [6] Mozaffarian D, Aro A, Willett WC. Health effects of trans-fatty acids: experimental and observational evidence. Eur J Clin Nutr 2009;63 (Suppl 2):S5–S21. [7] Wallace SK, Mozaffarian D. Trans-fatty acids and nonlipid risk factors. Curr Atheroscler Rep 2009;11:423–33. [8] Willett WC, Stampfer MJ, Manson JE, Colditz GA, Speizer FE, Rosner BA, et al. Intake of trans fatty acids and risk of coronary heart disease among women. Lancet 1993;341:581–5. [9] Ascherio A, Rimm EB, Giovannucci EL, Spiegelman D, Stampfer M, Willett WC. Dietary fat and risk of coronary heart disease in men: cohort follow up study in the United States. BMJ 1996;313:84–90. [10] Pietinen P, Ascherio A, Korhonen P, Hartman AM, Willett WC, Albanes D, et al. Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. Am J Epidemiol 1997;145:876–87. [11] Jakobsen MU, Overvad K, Dyerberg J, Heitmann BL. Intake of ruminant trans fatty acids and risk of coronary heart disease. Int J Epidemiol 2008;37:173–82. [12] Brouwer IA, Wanders AJ, Katan MB. Effect of animal and industrial trans fatty acids on HDL and LDL cholesterol levels in humans—a quantitative review. PLoS One 2010;5:e9434. [13] Lee KN, Kritchevsky D, Pariza MW. Conjugated linoleic acid and atherosclerosis in rabbits. Atherosclerosis 1994;108:19–25. [14] Nicolosi RJ, Rogers EJ, Kritchevsky D, Scimeca JA, Huth PJ. Dietary conjugated linoleic acid reduces plasma lipoproteins and early aortic atherosclerosis in hypercholesterolemic hamsters. Artery 1997;22: 266–77. [15] de Deckere EA, van Amelsvoort JM, McNeill GP, Jones P. Effects of conjugated linoleic acid (CLA) isomers on lipid levels and peroxisome proliferation in the hamster. Br J Nutr 1999;82:309–17. [16] Stangl GI. High dietary levels of a conjugated linoleic acid mixture alter hepatic glycerophospholipid class profile and cholesterol-carrying serum lipoproteins of rats. J Nutr Biochem 2000;11:184–91. [17] Gavino VC, Gavino G, Leblanc MJ, Tuchweber B. An isomeric mixture of conjugated linoleic acids but not pure cis-9, trans-11-octadecadienoic acid affects body weight gain and plasma lipids in hamsters. J Nutr 2000;130:27–9. [18] Sher J, Pronczuk A, Hajri T, Hayes KC. Dietary conjugated linoleic acid lowers plasma cholesterol during cholesterol supplementation, but accentuates the atherogenic lipid profile during the acute phase response in hamsters. J Nutr 2003;133:456–60. [19] Tricon S, Burdge GC, Kew S, Banerjee T, Russell JJ, Jones EL, et al. Opposing effects of cis-9, trans-11 and trans-10, cis-12 conjugated linoleic acid on blood lipids in healthy humans. Am J Clin Nutr 2004;80: 614–20. [20] Riserus U, Arner P, Brismar K, Vessby B. Treatment with dietary trans10 cis12 conjugated linoleic acid causes isomer-specific insulin resistance in obese men with the metabolic syndrome. Diabetes Care 2002;25:1516–21. [21] Destaillats F, Golay PA, Joffre F, de Wispelaere M, Hug B, Giuffrida F, et al. Comparison of available analytical methods to measure trans-octadecenoic acid isomeric profile and content by gas–liquid chromatography in milk fat. J Chromatogr A 2007;1145:222–8. [22] Bergelson LD, Shemyakin MM. Synthesis of naturally occurring unsaturated fatty acids by sterically controlled carbonyl olefination. Angew Chem Int Ed Engl 1964;3:250–60. [23] Gravier-Pelletier C, Dumas J, Le Merrer Y, Depezay JC. Methods for the total synthesis of acyclic hydroxylated fatty acids. Prog Lipid Res 1990;29:229–76. [24] Foglia TA, Vail PD. An efficient large-scale synthesis of triisopentadecanoin. Org Prep Proced Int 1993;25:209–13.
576
S.K. Gebauer et al. / Contemporary Clinical Trials 32 (2011) 569–576
[25] Lawrence NJ. The Wittig reaction and related methods. In: Williams JJ, editor. Preparation of alkenes, a practical approach. Oxford: Oxford University Press; 1996. p. 19–58. [26] Grandgirard A, Sebedio JL, Fleury J. Geometrical isomerization of linolenic acid during heat treatment of vegetable oils. J Am Oil Chem Soc 1984;61:1563. [27] Chen SC, Judd JT, Kramer M, Meijer GW, Clevidence BA, Baer DJ. Phytosterol intake and dietary fat reduction are independent and additive in their ability to reduce plasma LDL cholesterol. Lipids 2009;44:273–81. [28] Ascherio A, Hennekens CH, Buring JE, Master C, Stampfer MJ, Willett WC. Trans-fatty acids intake and risk of myocardial infarction. Circulation 1994;89:94–101. [29] Oomen CM, Ocke MC, Feskens EJ, van Erp-Baart MA, Kok FJ, Kromhout D. Association between trans fatty acid intake and 10-year risk of coronary heart disease in the Zutphen Elderly Study: a prospective population-based study. Lancet 2001;357:746–51. [30] Motard-Belanger A, Charest A, Grenier G, Paquin P, Chouinard Y, Lemieux S, et al. Study of the effect of trans fatty acids from ruminants on blood lipids and other risk factors for cardiovascular disease. Am J Clin Nutr 2008;87:593–9.
[31] Chardigny JM, Destaillats F, Malpuech-Brugere C, Moulin J, Bauman DE, Lock AL, et al. Do trans fatty acids from industrially produced sources and from natural sources have the same effect on cardiovascular disease risk factors in healthy subjects? Results of the trans Fatty Acids Collaboration (TRANSFACT) study. Am J Clin Nutr 2008;87:558–66. [32] Wanders AJ, Brouwer IA, Siebelink E, Katan MB. Effect of a high intake of conjugated linoleic acid on lipoprotein levels in healthy human subjects. PLoS One 2010;5:e9000. [33] U.S. Food and Drug Administration. Federal Register — 68 FR 41433 July 11, 2003: Food Labeling; Trans Fatty Acids in Nutrition Labeling; Consumer Research to Consider Nutrient Content and Health Claims and Possible Footnote or Disclosure Statements; Final Rule and Proposed Rule. Accessed from: http://www.fda.gov/Food/LabelingNutrition/ LabelClaims/NutrientContentClaims/ucm110179.htm. Accessed September 29, 2010. [34] Stender S, Dyerberg J. Influence of trans fatty acids on health. Ann Nutr Metab 2004;48:61–6. [35] Stender S, Dyerberg J, Astrup A. Consumer protection through a legislative ban on industrially produced trans fatty acids in foods in Denmark. Scand J Food Nutr 2006;50:155–60.