Erythrocyte n-3 Fatty Acids and Metabolic ... - Semantic Scholar

J C E M B r i e f

R e p o r t — E n d o c r i n e

O N L I N E

R e s e a r c h

Erythrocyte n-3 Fatty Acids and Metabolic Syndrome in Middle-Aged and Older Chinese Geng Zhang, Qi Sun, Frank B. Hu, Xingwang Ye, Zhijie Yu, Geng Zong, Huaixing Li, Yunhua Zhou, and Xu Lin Key Laboratory of Nutrition and Metabolism (G.Z., X.Y., G.Z., H.L., Y.Z., X.L.), Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences and Graduate School of the Chinese Academy of Sciences, 200031 Shanghai, China; Departments of Nutrition (Q.S., F.B.H.) and Epidemiology (F.B.H.), Harvard School of Public Health and Channing Laboratory (Q.S., F.B.H.), Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115; and Department of Community Health Sciences (Z.Y.), Brock University, St. Catharines, Ontario L2S 3A1, Canada

Context: Few studies examined associations of circulating n-3 fatty acid levels with metabolic syndrome (MetS) among Chinese populations who have low consumption of these fatty acids and high risk of developing MetS. Objective: The objective of the study was to determine associations between erythrocyte n-3 fatty acids and MetS as well as its components among middle-aged and older Chinese men and women. Design and Participants: Erythrocyte levels of docosahexaenoic acid (DHA), docosapentaenoic acid, eicosapentaenoic acid, and ␣-linolenic acid (ALA) were measured by gas chromatography among 2754 participants aged 50 –70 yr living in Beijing and Shanghai. MetS was defined using the updated National Cholesterol Education Program Adult Treatment Panel III criteria for Asian-Americans. Results: After multivariable adjustment, higher levels of DHA, but neither eicosapentaenoic acid nor docosapentaenoic acid, were associated with lower odds of MetS as well as elevated blood pressure and triglycerides. Comparing extreme quartiles of DHA, odds ratios (95% confidence interval) were 0.75 (0.55, 1.01; P for trend ⫽ 0.04) for MetS; 0.70 (0.53, 0.92; P for trend ⫽ 0.01) for elevated blood pressure; and 0.64 (0.48, 0.87; P for trend ⫽ 0.005) for elevated triglycerides. In contrast, ALA concentrations were positively associated with MetS odds (odds ratio 4.06; 95% confidence interval 2.85, 5.80; P for trend ⬍0.001). Conclusions: Higher concentrations of erythrocyte DHA were associated with lower odds of MetS, whereas higher concentrations of ALA were associated with increased odds among middle-aged and older Chinese. These findings warrant replication in other populations. (J Clin Endocrinol Metab 97: E973–E977, 2012)

M

etabolic syndrome (MetS), characterized by a constellation of multiple cardiometabolic abnormalities including central obesity, hypertension, dyslipidemia, and impaired glucose tolerance, has been recognized as a condition strongly predicting type 2 diabetes and cardiovascular disease (CVD) and identified as a critical preclinical stage for intervention (1). With a dramatic nutrition

transition in the last few decades, China is facing the challenge of an epidemic of obesity-related diseases, including MetS: nearly one fifth of Chinese adults have MetS (2). Although diet has been identified as one of the most important factors of MetS in Western populations, little is known regarding specific dietary risk factors in relation to MetS among Chinese populations. In particular, evidence

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2012 by The Endocrine Society doi: 10.1210/jc.2011-2997 Received November 1, 2011. Accepted March 6, 2012. First Published Online March 28, 2012

Abbreviations: ALA, ␣-Linolenic acid; CI, confidence interval; CRP, C-reactive protein; CV, coefficient of variation; CVD, cardiovascular disease; DHA, docosahexaenoic acid; DPA, docosapentaenoic acid; EPA, eicosapentaenoic acid; HDL, high-density lipoprotein; MetS, metabolic syndrome; OR, odds ratio.

J Clin Endocrinol Metab, June 2012, 97(6):E973–E977

jcem.endojournals.org

E973

E974

Zhang et al.

n-3 Fatty Acids and Metabolic Syndrome

concerning the relationship between n-3 fatty acids and MetS is limited and inconclusive. Moreover, most previous studies used self-reported food frequency questionnaires to evaluate n-3 fatty acid intake, which may be affected by measurement error due to inaccurate recall, an incomplete food composition database, and likely regional variation in long-chain n-3 fatty acid contents in fish. On the other hand, tissue concentrations of n-3 fatty acids can serve as objective biomarkers reflecting dietary intake because human body cannot synthesize these fatty acids de novo. To date, only a few studies have investigated the relationship between blood n-3 fatty acid concentrations and MetS, as well as its components, and existing data remain inconsistent (3–7). In this study, we aimed to determine the associations of erythrocyte n-3 fatty acids, including ␣-linolenic acid (ALA), eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) with MetS and its components among a middle-aged and older Chinese population.

Subjects and Methods Study population The present study was based on the Nutrition and Health of Aging Population in China project, which included 3289 men and women aged 50 –70 yr living in Beijing and Shanghai, China, in 2005 (8). A total of 535 participants were excluded due to the following exclusion criteria: 1) missing fatty acid levels (n ⫽ 31); 2) with implausibly high (⬎4000 kcal/d for men or ⬎3500 kcal/d for women) or low (⬍800 kcal/d for men or ⬍500 kcal/d for women) energy intake (n ⫽ 151); and 3) with self-reported CVD or cancers (n ⫽ 353), leaving 2754 subjects in the final analyses. The study was approved by the Institutional Review Board of the Institute for Nutritional Sciences and written informed consent was obtained from each participant.

Data collection Data on demographic variables, health status, lifestyle practice, physical activity, and anthropometry were collected following a standardized protocol (8). Dietary intake was assessed by a 74-item food frequency questionnaires modified from the one used in the China Health and Nutrition Surveys (9). The foodcomposition values were obtained from the Chinese Food Composition Table (10). All nutrient and food intakes were energy adjusted using the residual method (11).

Laboratory measurement Overnight fasting blood samples were obtained and stored at ⫺80 C until analyses were conducted. The measurements of plasma high-density lipoprotein (HDL) cholesterol, triglycerides, glucose, and C-reactive protein (CRP) were previously described (8). Erythrocyte membrane fatty acid concentrations were determined by gas-liquid chromatography (12). Briefly, erythrocyte samples were first extracted by hexane and isopropanol and then transmethylated with methanol and sulfuric acid. Fatty acid methyl


esters were subsequently analyzed by gas chromatography. Individual peaks were identified by comparison with known standards. Relative amount of each fatty acid was quantified by expressing the area under each peak as a percentage of total area under the peaks. Intraassay coefficients of variation (CV; percentages) were 2.7% for ALA, 25.8% for EPA, 3.6% for DPA, 3.4% for DHA, and 0.7% for linoleic acid. The corresponding interassay CV were 7.8, 9.3, 14.1, 10.0, and 7.7%, respectively, for these fatty acids.

Definition of MetS MetS was defined using the updated National Cholesterol Education Program Adult Treatment Panel III criteria for AsianAmericans (1) as meeting at least three of the following criteria: 1) central obesity (waist circumference of ⱖ90 cm in men or ⱖ80 cm in women); 2) elevated blood pressure (ⱖ130/85 mm Hg or using antihypertensive medications); 3) elevated fasting glucose (ⱖ5.6 mmol/liter or previously diagnosed diabetes or using antidiabetic agents or insulin); 4) reduced HDL cholesterol (⬍1.03 mmol/liter in men or ⬍1.30 mmol/liter in women); and 5) elevated triglycerides (ⱖ1.7 mmol/liter).

Statistical analysis Spearman partial correlation coefficients of individual n-3 fatty acids with food consumption levels were calculated and adjusted for age, sex, region (Beijing/Shanghai), residence (urban/rural), and total energy intake. Multivariable logistic regressions were applied to estimate the odds ratios (OR) for MetS and its components. P values for trend were determined by assuming median values of quartiles as continuous variables. All statistical analyses were performed using Stata 9.2 (Stata, College Station, TX) and P ⬍ 0.05 (two sided) was considered statistically significant.

Results Distribution of n-3 fatty acid concentrations and their associations with food intake The median values of ALA, EPA, DPA, and DHA concentrations in erythrocytes were 0.25, 0.42, 1.72, and 4.40%, respectively. In general, female, Shanghai, or urban participants had significantly higher concentrations of total and most individual types of n-3 fatty acids than their male, Beijing or rural counterparts with a few exceptions (Supplemental Table 1, published on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org). Both EPA and DHA, but not DPA, were positively correlated with fish consumption. Spearman partial correlation coefficients were 0.16 for EPA and 0.22 for DHA. ALA was not correlated with fish intake but weakly correlated with plant oil consumption (correlation ⫽ 0.07). Associations of n-3 fatty acids with MetS and its components Overall, total long-chain n-3 fatty acids, EPA, or DPA were not significantly associated with MetS after multivariable adjustment (Supplemental Table 2). On the other hand, higher DHA concentrations were significantly as-



E975

TABLE 1. Adjusted OR and 95% CI for MetS and its components according to quartiles of erythrocyte DHA concentrationsa Quartile of DHA concentration (%) Median (interquartile range) MetS Cases/n Model 1 Model 2 Central obesity Cases/n Model 1 Model 2 Elevated blood pressure Cases/n Model 1 Model 2 Elevated fasting glucose Cases/n Model 1 Model 2 Reduced HDL cholesterol Cases/n Model 1 Model 2 Elevated triglycerides Cases/n Model 1 Model 2

P for trendb

1 (lowest) 3.33 (3.01–3.55)

2 4.09 (3.94 – 4.23)

3 4.75 (4.56 – 4.93)

4 (highest) 5.65 (5.36 – 6.07)

306/688 1.00 1.00

299/689 0.97 (0.78, 1.21) 0.89 (0.68, 1.16)

264/689 0.83 (0.66, 1.05) 0.77 (0.58, 1.02)

254/688 0.81 (0.63, 1.03) 0.75 (0.55, 1.01)

0.05 0.04

320/688 1.00 1.00

349/689 1.16 (0.93, 1.45) 1.01 (0.71, 1.43)

318/689 1.00 (0.79, 1.26) 0.92 (0.63, 1.33)

306/688 0.96 (0.75, 1.23) 0.79 (0.54, 1.18)

0.50 0.21

507/688 1.00 1.00

484/689 0.94 (0.73, 1.20) 0.84 (0.65, 1.09)

448/689 0.84 (0.65, 1.07) 0.78 (0.60, 1.02)

412/688 0.75 (0.58, 0.97) 0.70 (0.53, 0.92)

0.02 0.01

333/688 1.00 1.00

299/689 1.05 (0.84, 1.31) 1.05 (0.83, 1.33)

239/689 0.91 (0.72, 1.15) 0.93 (0.73, 1.20)

203/688 0.90 (0.70, 1.16) 0.92 (0.70, 1.20)

0.29 0.40

254/688 1.00 1.00

282/689 1.00 (0.80, 1.26) 1.01 (0.79, 1.29)

306/689 1.03 (0.82, 1.30) 1.06 (0.82, 1.36)

305/688 0.90 (0.70, 1.15) 0.92 (0.70, 1.20)

0.41 0.56

184/688 1.00 1.00

156/689 0.79 (0.62, 1.02) 0.76 (0.58, 1.00)

154/689 0.76 (0.59, 0.99) 0.73 (0.55, 0.97)

145/688 0.66 (0.50, 0.88) 0.64 (0.48, 0.87)

0.006 0.005

a Model 1 is adjusted for age, sex, region, and residence. Model 2 is further adjusted for smoking, alcohol drinking, physical activity, educational attainment, family history of diabetes and CVD, total energy intake, red meat intake, fiber intake, body mass index, and CRP as well as ALA and linoleic acid in erythrocytes. b

The significance of linear trends across quartiles was tested by assigning the median value within quartiles to each participant and modeling this value as a continuous variable.

sociated with a lower prevalence of MetS: the OR in the highest vs. the lowest quartile was 0.75 [95% confidence interval (CI) 0.55, 1.01; P for trend ⫽ 0.04] (Table 1). With respect to individual component of MetS, participants in the highest DHA quartile had a 30% lower odds of elevated blood pressure (OR 0.70, 95% CI 0.53, 0.92; P for trend ⫽ 0.01) and a 36% lower odds of elevated triglycerides (OR 0.64, 95% CI 0.48, 0.87; P for trend ⫽ 0.005) compared with those in the lowest quartile. Unexpectedly, ALA was positively associated with MetS (OR 4.06, 95% CI 2.85, 5.80; P for trend ⬍0.001) as well as reduced HDL cholesterol (OR 4.42, 95% CI 3.22, 6.06; P for trend ⬍0.001) and elevated triglycerides (OR 11.48, 95% CI 7.75, 17.00; P for trend ⬍0.001) in the fully adjusted model (Table 2). Stratified analysis Potential modification effects of sex, geographic, and urban-rural differences on associations between n-3 fatty acids and MetS were also examined in stratified analyses (Supplemental Table 3 and 4). Most of the analyses confirmed the previous results except that the inverse association between DHA concentrations and MetS was ob-

served only among women or participants living in Beijing (P ⫽ 0.08 for interaction of DHA and sex and P ⫽ 0.31 for interaction of DHA and region).

Discussion In this study, we observed significant sex and geographic differences in erythrocyte n-3 fatty acid concentrations among this Chinese population. A higher concentration of DHA in erythrocytes was significantly associated with lower odds of MetS and elevated blood pressure and triglycerides independent of conventional CVD risk and inflammatory factors, whereas an opposite association was found for ALA. Unlike the well-established cardioprotective role, evidence regarding the relationship between tissue contents of n-3 fatty acids and MetS, a strong predictor for CVD and type 2 diabetes (1), is scarce, and existing results were controversial (3–5, 13). Plasma long-chain n-3 fatty acids were associated with lower odds of MetS in a Chinese study (3), whereas in other investigations using serum or erythrocyte specimens, null results were found (4, 5, 13).

E976

Zhang et al.

n-3 Fatty Acids and Metabolic Syndrome


TABLE 2. Adjusted OR and 95% CI for MetS and its components according to quartiles of erythrocyte ALA concentrationsa Quartile of ALA concentration (%) Median (interquartile range) MetS Cases/n Model 1 Model 2 Central obesity Cases/n Model 1 Model 2 Elevated blood pressure Cases/n Model 1 Model 2 Elevated fasting glucose Cases/n Model 1 Model 2 Reduced HDL cholesterol Cases/n Model 1 Model 2 Elevated triglycerides Cases/n Model 1 Model 2

P for trendb

1 (lowest) 0.16 (0.14 – 0.18)

2 0.22 (0.21– 0.24)

3 0.28 (0.27– 0.30)

4 (highest) 0.40 (0.36 – 0.48)

228/688 1.00 1.00

252/689 1.50 (1.18, 1.89) 1.41 (1.05, 1.89)

275/689 2.04 (1.60, 2.60) 2.15 (1.58, 2.94)

368/688 3.84 (2.98, 4.94) 4.06 (2.85, 5.80)

⬍0.001 ⬍0.001

295/688 1.00 1.00

312/689 1.33 (1.06, 1.67) 1.12 (0.78, 1.62)

318/689 1.51 (1.20, 1.91) 1.35 (0.92, 1.99)

368/688 2.08 (1.63, 2.65) 1.52 (0.98, 2.35)

⬍0.001 0.05

457/688 1.00 1.00

468/689 1.27 (1.00, 1.61) 1.15 (0.89, 1.48)

465/689 1.42 (1.11, 1.82) 1.21 (0.92, 1.60)

461/688 1.61 (1.25, 2.08) 1.21 (0.89, 1.66)

⬍0.001 0.30

316/688 1.00 1.00

281/689 1.09 (0.87, 1.37) 1.00 (0.78, 1.27)

228/689 0.96 (0.76, 1.22) 0.83 (0.64, 1.08)

249/688 1.28 (1.00, 1.63) 1.00 (0.74, 1.36)

0.07 0.92

208/688 1.00 1.00

252/689 1.35 (1.07, 1.71) 1.53 (1.18, 1.98)

295/689 1.72 (1.36, 2.19) 2.25 (1.71, 2.96)

392/688 2.91 (2.28, 3.72) 4.42 (3.22, 6.06)

⬍0.001 ⬍0.001

66/688 1.00 1.00

111/689 2.29 (1.64, 3.19) 2.29 (1.61, 3.26)

162/689 4.41 (3.18, 6.10) 4.49 (3.14, 6.42)

300/688 11.93 (8.60, 16.57) 11.48 (7.75, 17.00)

⬍0.001 ⬍0.001

a Model 1 is adjusted for age, sex, region, and residence. Model 2 is further adjusted for smoking, alcohol drinking, physical activity, educational attainment, family history of diabetes and CVD, total energy intake, red meat intake, fiber intake, body mass index, and CRP as well as EPA, DPA, DHA, and linoleic acid in erythrocytes. b

The significance of linear trends across quartiles was tested by assigning the median value within quartiles to each participant and modeling this value as a continuous variable.

Discrepancies among these studies and the current investigation might be due to a few reasons. First, in comparison with plasma phospholipids or serum cholesterol esters, erythrocyte concentrations may reflect a relatively longer-term dietary intake of n-3 fatty acids of marine origin. Second, various effects of individual n-3 fatty acids on metabolic outcomes might be concealed when all n-3 fatty acids were combined in the previous studies (14). Finally, different study design, sample size, characteristics of participants, and covariates adjusted may further explain the heterogeneous results. It was unexpected that erythrocyte ALA was strongly associated with MetS odds, and this association was primarily driven by the strong associations with dyslipidemia in the current study. A meta-analysis on randomized control trials of ALA revealed that ALA intake might slightly decrease HDL levels and nonsignificantly increase lowdensity lipoprotein and triglyceride levels, although significant heterogeneity across the trials was also documented (15). Likewise, in observational studies intake or circulating levels of ALA were not consistently associated with risk of metabolic syndrome or CVD outcomes (5, 16 –18). It is unclear whether the association of ALA with

higher MetS odds in the current study was due to cooking method: stir-fry is a common Chinese cooking practice, and several deleterious trans-isomers and cyclic fatty acids could be generated during the high-temperature heating process (19). Clearly, our results need to be replicated in other populations and examined further in experimental studies. To the best of our knowledge, this is the first study to investigate the associations of individual n-3 fatty acid biomarkers in erythrocytes with MetS in a large population-based sample in Asia. Using nutritional biomarkers is particularly important for assessing diet in countries like China without a complete food composition database for fatty acid content and with a tradition of sharing dishes during meals. In addition, the detailed, standardized protocol used to identify outcomes and collect information on sociodemographic, lifestyle, and dietary factors and inflammatory marker made the comprehensive multivariable-adjusted statistical analysis possible. On the other hand, owing to the cross-sectional nature of the current study, a causal relation between n-3 fatty acid biomarkers and MetS cannot be inferred. Furthermore, greater measurement error for erythrocyte EPA (in-


traassay CV 25.8%) might materially attenuate statistical power to detect the association between EPA and MetS. In summary, we observed a significant inverse association between erythrocyte DHA and MetS among the middle-aged and older Chinese. Conversely, higher ALA concentrations were strongly associated with a higher prevalence of MetS. Further prospective studies are needed to confirm our findings.

Acknowledgments We express our sincere appreciation to Drs. Xinghuo Pang, Zhen Zhang, Shufang Jiao, Hong Liu, and Shurong Zou and all the other physicians and staff taking part in the fieldwork of this study. We also thank Drs. An Pan, Qibin Qi, and Hongyu Wu from Harvard School of Public Health, as well as Drs. Liang Sun, Chen Liu, Ling Lu, Danxia Yu, Wei Gan, Jingwen Zhu, Xin Liu, Qianlu Jin, Shaojie Ma, and He Zheng in Dr. Lin’s group for their kind assistance at various stages of this study. Address all correspondence and requests for reprints to: Dr. Xu Lin, Institute for Nutritional Sciences, Chinese Academy of Sciences, 294 Tai-Yuan Road, Shanghai 200031, China. E-mail: [email protected]. This work was supported by Grants 30930081 and 81021002 from the National Natural Science Foundation of China, Grant 2011CB504002 and 2012CB524900 from the National Basic Research 973 Program, and Grants KSCX2-EWR-10 and SIBS2008006 from the Chinese Academy of Sciences. Disclosure Summary: The authors have nothing to disclose.

References 1. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith Jr SC, Spertus JA, Costa F 2005 Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 112:2735–2752 2. Yang W, Reynolds K, Gu D, Chen J, He J 2007 A comparison of two proposed definitions for metabolic syndrome in the Chinese adult population. Am J Med Sci 334:184 –189 3. Huang T, Bhulaidok S, Cai Z, Xu T, Xu F, Wahlqvist ML, Li D 2010 Plasma phospholipids n-3 polyunsaturated fatty acid is associated with metabolic syndrome. Mol Nutr Food Res 54:1628 –1635


E977

4. Lee E, Lee S, Park Y 2008 n-3 Polyunsaturated fatty acids and trans fatty acids in patients with the metabolic syndrome: a case-control study in Korea. Br J Nutr 100:609 – 614 5. Warensjo¨ E, Rise´rus U, Vessby B 2005 Fatty acid composition of serum lipids predicts the development of the metabolic syndrome in men. Diabetologia 48:1999 –2005 6. Wang L, Tsai M, Manson JE, Djousse L, Gaziano JM, Buring JE, Sesso HD 2011 Erythrocyte fatty acid composition is associated with the risk of hypertension in middle-aged and older women. J Nutr 141:1691–1697 7. Zhang B, Wang P, Zhou Q, Chen C, Zhuo S, Ye Y, He Q, Chen Y, Su Y 2011 The relationships between erythrocyte membrane n-6 to n-3 polyunsaturated fatty acids ratio and blood lipids and C-reactive protein in Chinese adults: an observational study. Biomed Environ Sci 24:234 –242 8. Ye X, Yu Z, Li H, Franco OH, Liu Y, Lin X 2007 Distributions of C-reactive protein and its association with metabolic syndrome in middle-aged and older Chinese people. J Am Coll Cardiol 49:1798 – 1805 9. Zhai F, Yang Y 2006 The nutrition and health status of the Chinese people 2002: diet and nutrients intake. Beijing, China: People’s Medical Publishing House 10. Yang Y 2005 Chinese Food Composition Table 2004. Beijing: Peking University Medical Press 11. Willett W, Stampfer MJ 1986 Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 124:17–27 12. Sun Q, Ma J, Campos H, Hankinson SE, Hu FB 2007 Comparison between plasma and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women. Am J Clin Nutr 86:74 – 81 13. Kabagambe EK, Tsai MY, Hopkins PN, Ordovas JM, Peacock JM, Borecki IB, Arnett DK 2008 Erythrocyte fatty acid composition and the metabolic syndrome: a National Heart, Lung, and Blood Institute GOLDN study. Clin Chem 54:154 –162 14. Anderson BM, Ma DW 2009 Are all n-3 polyunsaturated fatty acids created equal? Lipids Health Dis 8:33 15. Wendland E, Farmer A, Glasziou P, Neil A 2006 Effect of alpha linolenic acid on cardiovascular risk markers: a systematic review. Heart 92:166 –169 16. Lemaitre RN, King IB, Sotoodehnia N, Rea TD, Raghunathan TE, Rice KM, Lumley TS, Knopp RH, Cobb LA, Copass MK, Siscovick DS 2009 Red blood cell membrane ␣-linolenic acid and the risk of sudden cardiac arrest. Metabolism 58:534 –540 17. Oomen CM, Ocke´ MC, Feskens EJ, Kok FJ, Kromhout D 2001 ␣-Linolenic acid intake is not beneficially associated with 10-y risk of coronary artery disease incidence: the Zutphen Elderly Study. Am J Clin Nutr 74:457– 463 18. Pedersen JI, Ringstad J, Almendingen K, Haugen TS, Stensvold I, Thelle DS 2000 Adipose tissue fatty acids and risk of myocardial infarction—a case-control study. Eur J Clin Nutr 54:618 – 625 19. Sebedio J, Chardigny J 1996 Physiological effects of trans and cyclic fatty acids. In: Perkins E, Erickson M, eds. Deep frying chemistry: nutrition and practical applications. Champaign, IL: AOCS Press; 183–209

Stay Current with our best-selling educational resource, Endocrine Self-Assessment Program ESAP 2011. www.endo-society.org/ESAP2011