The Journal of Nutrition Nutritional Epidemiology

Dietary Fat and Fatty Acid Profile Are Associated with Indices of Skeletal Muscle Mass in Women Aged 18–79 Years1,2 Ailsa A. Welch,3* Alex J. MacGregor,3 Anne-Marie Minihane,3 Jane Skinner,3 Anna A. Valdes,4 Tim D. Spector,4 and Aedin Cassidy3 3 Norwich Medical School, University of East Anglia, Norwich, UK; and 4Department of Twin Research and Genetic Epidemiology, KingÕs College London, London, UK

Abstract Age-related loss of skeletal muscle mass results in a reduction in metabolically active tissue and has been related to the onset of obesity and sarcopenia. Although the causes of muscle loss are poorly understood, dietary fat has been This study was designed to investigate the cross-sectional relation between dietary fat intake, as dietary percentage of fat energy (PFE) and fatty acid profile, with indices of skeletal muscle mass in the population setting. Body composition [fatfree mass (FFM; in kg)] and the fat-free mass index (FFMI; kg FFM/m2) was measured by using dual-energy X-ray absorptiometry in 2689 women aged 18–79 y from the TwinsUK Study and calculated according to quintile of dietary fat (by food-frequency questionnaire) after multivariate adjustment. Positive associations were found between the polyunsaturated-to-saturated fatty acid (SFA) ratio and indices of FFM, and inverse associations were found with PFE, SFAs, monounsaturated fatty acids (MUFAs), and trans fatty acids (TFAs) (all as % of energy). Extreme quintile dietary differences for PFE were 20.6 kg for FFM and 20.28 kg/m2 for FFMI; for SFAs, MUFAs, and TFAs, these were 20.5 to 20.8 kg for FFM and 20.26 to 20.38 kg/m2 for FFMI. These associations were of a similar magnitude to the expected decline in muscle mass that occurs over 10 y. To our knowledge, this is the first population-based study to demonstrate an association between a comprehensive range of dietary fat intake and FFM. These findings indicate that a dietary fat profile already associated with cardiovascular disease protection may also be beneficial for conservation of skeletal muscle mass. J. Nutr. 144: 327–334, 2014.

Introduction The consequences of the age-related loss of skeletal muscle mass are a reduction in metabolically active tissue, potentially contributing to the onset of obesity, and disordered blood glucose and electrolyte control (1–4). Muscle loss also contributes to sarcopenia (loss of muscle mass and strength associated with aging), the prevalence of which is 9–18% in people aged >65 y (1,5–8). Sarcopenia and loss of muscle mass are related to functional impairment, physical disability, reduced quality of life, and frailty, as well as fractures and mortality (9–11). The estimated costs of sarcopenia amount to $18.5 billion/y in the United States. Costs of fractures are £2.3 billion/y in the United Kingdom (and $17 billion in the United States) (12–14). Muscle

1 Supported by the Norwich Medical School, University of East Anglia, and the Department of Health via the National Institute for Health Research Comprehensive Biomedical Research Centre award to GuyÕs and St. ThomasÕ National Health Service Foundation Trust in partnership with KingÕs College London. 2 Author disclosures: A. A. Welch, A. J. MacGregor, A.-M. Minihane, J. Skinner, A. A. Valdes, T. D. Spector, and A. Cassidy, no conflicts of interest. * To whom correspondence should be addressed. E-mail: [email protected].

loss starts at the age of 30 y and progresses more rapidly from the age of 50 y. However, the causes for this age-related muscle loss remain incompletely understood. Given estimated health care costs, effect on quality of life, and the predicted increase in age profile of Western populations, identifying the causes of chronic muscle loss is important so that preventative strategies to reduce rates of decline of muscle mass can be developed (12). Dietary fat is a major source of energy for resting and working muscle and may exert effects on age-related muscle loss through a number of mechanisms, including an influence on the composition of FAs within the sarcolemma (muscle membrane), an effect on chronic low-grade inflammation (circulating cytokines), and insulin resistance (2,15–24). Dietary fat composition influences the mechanisms of inflammation and insulin resistance, with SFAs and trans fatty acids (TFAs)5 considered to

5 Abbreviations used: ALM, appendicular lean mass; CVD, cardiovascular disease; EER, estimated energy requirement; EI, energy intake; FFM, fat-free mass; FFMI, fat-free mass index; LC, long-chain; mTOR, mammalian target of rapamycin; PFE, percentage of energy from fat; P:S ratio, PUFA-to-SFA ratio; SFAs+TFAs, total SFAs and trans fatty acids; TFA, trans fatty acid.

ã 2014 American Society for Nutrition. Manuscript received September 16, 2013. Initial review completed October 10, 2013. Revision accepted November 22, 2013. First published online January 8, 2014; doi:10.3945/jn.113.185256.

327

Downloaded from jn.nutrition.org by guest on June 6, 2015

postulated to have a role in determining protein turnover through an influence on both inflammation and insulin resistance.

increase inflammation and insulin resistance and PUFAs, primarily n–6 PUFAs, considered to oppose these effects (2,25–29). Total dietary fat, as well as its FA profile, varies considerably between individuals and is potentially modifiable. If total fat intake as a percentage energy from fat (PFE) or differences in dietary FA profile were associated with indices of muscle mass, dietary strategies would be feasible to reduce rates of decline in muscle mass. Despite the integral nature of fat-to-muscle metabolism and the potential for fat to influence mechanisms associated with muscle loss, the association between the full range of dietary fat and FAs has not previously been studied comprehensively using accurate measures of body composition (DXA) in a human population (30). The aim of this study was to investigate the association between indices of muscle mass [as fat-free mass (FFM)] and intakes of total dietary fat and FA profile in a crosssectional cohort of healthy free-living women, representative of the general population for diet and other characteristics such as smoking and physical activity behaviors (31,32). From the available evidence, we hypothesized that indices of muscle mass would be higher in those consuming more PUFAs and lower in those consuming more SFAs and TFAs.

The TwinsUK Registry is an ongoing study in healthy adult twin volunteers who underwent extensive clinical assessments to collect data for a range of age-related characteristics (31). This study included 2689 female twins aged 18–79 y who had complete data for body composition, validated dietary and health and lifestyle questionnaires, and clinical assessments between 1996 and 2000. The participants were not selected for any characteristic or disease trait. Participants in the TwinsUK Registry have been shown to be representative of adult singleton populations in the United Kingdom for a number of physical and lifestyle characteristics (31,32). Zygosity (monozygotic and dizygotic) was derived by questionnaire and confirmed by multiplex-DNA fingerprinting (PE Applied Biosystems). Physical activity during leisure time was derived by questionnaire using the Allied Dunbar Physical Activity Score and categorized as inactive, light, moderate, and heavy exercise (33). Smoking habit was determined from questionnaire and categorized as current, former, or never smoker. Ethical approval was obtained from the St. ThomasÕs Hospital Research Ethics committee. Informed consent was obtained from all participants. Assessment of body composition. Height and weight were measured by using standard scales, and BMI was calculated by dividing weight (kg) by height (m2). Body composition was measured by using DXA (Hologic QDR) with participants in a supine position with their feet in a neutral position and hands flat by their sides. Total fat mass and FFM (kg) were determined by using standard software calculations. To further understand the association between indices of skeletal muscle mass, we used 2 methods of expressing FFM in our study: 1) FFM in kilograms and 2) the fat-free mass index (FFMI), calculated as kilograms of FFM divided by height in meters squared (kg/m2), to account for the known proportional increase in FFM that occurs with increasing height (34). Dietary intake. Dietary intake was estimated and calculated by using a validated 131-item semiquantitative FFQ that has been used previously and validated in the EPIC (European Prospective Investigations into Cancer and Nutrition)-Norfolk study (35,36). Participants completed the frequency of consumption of each item of food, on the FFQ which was described in average portion sizes. FAs were derived from a database based on published analytic data for FAs, with a small amount of additional data from other European national food composition databases (35,36). For this database, basic foods were reviewed and completed for missing values, with calculations to obtain values for mixed dishes based on published standard recipes and cooking losses. 328

Welch et al.

Statistical methods. Statistical analyses were performed with Stata statistical software version 11.0 (StataCorp). Intakes of fat and FAs as a PFE were calculated, as was the PUFA-to-SFA ratio (P:S ratio), and quintiles of these variables were calculated. Total percentage fat and specific FAs were chosen for analysis, a priori, on the basis of potential plausible mechanisms. Adjusted mean 6 SEM FFM and FFMI were calculated for quintiles of PFE and FA intake after adjustment for covariates. Unadjusted means are reported as means 6 SDs. Because the proportion of FFM is negatively influenced by age and smoking habit and positively influenced by physical activity, these covariates were included in the model (40–44). Because FFM is known to increase with both total body weight and total fat mass, to account for these influences on muscle mass and to identify the effects of diet independently, the models for FFM and FFMI were also adjusted for fat mass (34,45). To account for potential effects of differences in dietary reporting habit (misreporting), we also included the EI:EER ratio in the models (39). Therefore, the data from the model presented (Tables 1 and 2 and in text) were adjusted for age, physical activity, EI, EI:EER ratio, smoking habit, and total fat mass (i.e., adjusted means 6 SEMs). The values shown in Figs. 1 and 2 are also adjusted mean values. The results of the unadjusted model are described briefly in the Results section; an additional model that further adjusted for percentage of protein intake was calculated but, in the interest of space, was not included in the tables. In an additional analysis to determine whether relations between dietary fat and indices of FFM were found, specifically in older people, the analyses were repeated in women aged $50 y, and these results are also described in the text. The P-trend across quintiles and P-difference between quintiles 1 and 5 were also calculated. To calculate the relative associations of the relations with dietary fat intake and those of age, b coefficients for 10 y of age and per quintile of fat intake were calculated by using multivariate regression. Because PUFAs are considered to have effects in opposite directions for cardiovascular disease (CVD) when compared with SFAs and TFAs and because our results for SFAs and TFAs were in the same direction, we summed SFAs and TFAs to produce a variable ‘‘total SFAs and TFAs’’ (SFAs+TFAs). To further understand the interaction between PUFAs and SFAs+TFAs, we divided PUFAs and SFAs+TFAs into tertiles and then stratified the analyses for FFM and FFMI by PUFAs and SFAs+TFAs, after adjustment for covariates, as in the models described above (Fig. 1) (46). Because there were no significant differences between the dietary intakes of monozygotic and dizygotic twins (except for TFAs), the analyses were performed on a cohort basis. However, because data from members of twin pairs could not be treated as independent, we controlled for familial aggregation by treating twin pairs as clusters by using the robust regression cluster option in Stata software. To understand further the scale of the associations of dietary fat intake with indices of muscle mass, we compared these associations with age, because age is the major factor known to affect chronic muscle loss. Dietary intake was compared with a difference of 10 y in age (an estimate of the population decline in muscle mass over a 10-y period). The associations with 10 y in age and with FFM per quintile of PFE or FA intakes were calculated from the multivariate model (b coefficient). The association of FFM with PFE or FA intakes was then calculated as a percentage of the 10-y difference in age (referred to as the diet comparison with 10 y of age in the Results section).

Results The age range of the women was 18–79 y, and 50% of the women were older than 50 y (Table 3). The population was

Downloaded from jn.nutrition.org by guest on June 6, 2015

Participants and Methods

Individuals were excluded from this study, as documented previously, if >10 items on the FFQ were left blank or if the ratio of energy intake (EI) to estimated basal metabolic rate (based on the Harris–Benedict equation) was >2 SDs from the mean (31,35,37). To account for potential misreporting, the ratio of reported EI to estimated energy expenditure was calculated by using the ratio of reported EI to the estimated energy requirement (EER; EI:EER ratio) based on equations using the 2002 Institute of Medicine of the National Academies report, and this was included as a covariate for adjustment in the statistical analyses (38,39).

TABLE 1

Fat-free mass calculated according to quintile category of dietary fat or FA intake in 2689 women aged 18–79 y1 Fat-free mass

Dietary fat

Total fat (PFE) FAs (PFE) SFAs MUFAs Oleic acid TFAs PUFAs Total n–6 PUFAs Linoleic acid Total n–3 PUFAs a-Linolenic acid Long-chain n–3 PUFAs P:S ratio

Q1 (n = 538)

Q2 (n = 538)

Q3 (n = 538)

Q4 (n = 538)

Q5 (n = 537)

Difference Q1–Q52

P-difference3

P-trend4

39.9 6 0.2

39.8 6 0.2

kg 40.0 6 0.2

39.1 6 0.2

39.3 6 0.2

kg 20.6

0.033

0.004

39.9 6 39.8 6 39.6 6 39.8 6 39.6 6 39.6 6 39.6 6 39.6 6 39.7 6 39.6 6 39.6 6

39.8 6 39.8 6 40.0 6 39.7 6 39.5 6 39.7 6 39.6 6 39.7 6 39.4 6 39.9 6 39.7 6

39.3 39.4 39.5 39.3 39.8 39.6 39.6 39.7 39.8 39.3 39.7

6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2

20.5 20.8 20.7 20.8 0.2 0.4 0.4 20.6 20.3 20.8 0.6

0.034 0.002 0.014 0.005 0.44 0.26 0.003 0.009 0.19 0.004 0.011

0.006 0.001 0.022 0.001 0.34 0.33 0.25 0.027 0.26 0.003 0.012

39.7 39.9 39.8 40.0 39.5 39.4 39.4 39.8 39.7 40.0 39.2

6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2

0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2

0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2

6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2 6 0.2

39.2 39.1 39.1 39.2 39.7 39.8 39.8 39.2 39.4 39.2 39.8

Values are means 6 SEs adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting unless otherwise indicated. PFE, percentage of energy from fat; P:S, PUFA-to-SFA ratio; Q, quintile; TFA, trans fatty acid. Difference between quintile 5 and quintile 1. 3 P for difference between quintiles 5 and 1 calculated by using ANOVA, adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting. 4 P for trend calculated by using ANOVA, adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting. 1

2

TABLE 2

25.1 6 4.2 kg/m2; P = 0.001) than the dizygotic twins. Monozygotic twins also had a lower FFM (mean 6 SD: 38.8 6 5.0 vs. 40.0 6 5.4 kg; P< 0.001) and FFMI (14.7 6 1.7 vs. 15.1 6 1.7 kg/m2; P < 0.001) than dizygotic twins. There was a 1.7-fold variation in mean intakes of PFE between quintiles 1 and 5 and a 2.1-fold variation in SFAs, with the variation in intakes of the other FAs ranging from 1.8 for MUFAs to 2.8 for TFAs (Table 4). FFM was significantly and inversely associated with PFE (Ptrend = 0.004), SFAs (P = 0.006), MUFAs (P = 0.001), and TFAs (P = 0.001), whereas a significant positive association was observed with the P:S ratio (P = 0.012) (Table 1). Total PUFAs, n–6 PUFAs, and linoleic acid were positively associated with

FFMI calculated according to quintile of dietary fat or FA intake in 2689 women aged 18–79 y1 FFMI

Dietary fat

Total fat (PFE) FAs (PFE) SFAs MUFAs Oleic acid TFAs PUFAs Total n–6 PUFAs Linoleic acid Total n–3 PUFAs a-Linolenic acid Long-chain n–3 PUFAs P:S ratio

Q1 (n = 538)

Q2 (n = 538)

Q3 (n = 538)

15.13 6 0.07

15.11 6 0.07

kg/m2 15.06 6 0.06

15.08 6 15.15 6 15.10 6 15.13 6 14.98 6 14.99 6 14.98 6 15.12 6 15.09 6 15.13 6 14.82 6

15.11 15.10 15.06 15.13 15.00 14.99 15.00 14.95 15.01 14.92 14.93

0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.06

6 6 6 6 6 6 6 6 6 6 6

0.07 0.07 0.07 0.06 0.07 0.10 0.07 0.07 0.07 0.07 0.07

15.12 15.06 15.09 14.95 14.91 14.95 14.94 14.97 14.90 15.09 15.09

6 0.06 6 0.07 6 0.07 6 0.06 6 0.06 6 0.07 6 0.06 6 0.07 6 0.07 6 0.07 6 0.07

P-difference3

P-trend4

Q5 (n = 537)

Difference Q1–Q52

14.83 6 0.07

14.85 6 0.07

kg/m2 20.28

0.005

,0.001

14.88 6 14.91 6 14.98 6 14.91 6 15.07 6 15.00 6 15.00 6 15.04 6 15.01 6 14.93 6 15.09 6

14.79 14.77 14.76 14.87 15.03 15.07 15.07 14.91 14.99 14.91 15.07

6 0.07 6 0.07 6 0.07 6 0.07 6 0.07 6 0.07 6 0.07 6 0.06 6 0.07 6 0.07 6 0.07

20.29 20.38 20.34 20.26 0.05 0.06 0.09 20.21 20.1 20.22 0.25

0.004 ,0.001 0.001 0.008 0.56 0.42 0.34 0.016 0.27 0.020 0.011

,0.001 ,0.001 0.001 0.001 0.40 0.45 0.35 0.09 0.33 0.045 0.003

Q4 (n = 538)

0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06

1 Values are means 6 SEs adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting unless otherwise indicated. FFMI, fat-free mass index; PFE, percentage of energy from fat; P:S ratio, PUFA-to-SFA ratio; Q, quintile; TFA, trans fatty acid. 2 Difference between Q5 and Q1. 3 P for difference between Q5 and Q1 calculated by using ANOVA, adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting. 4 P for trend calculated by using ANOVA, adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting.

Dietary fat and skeletal muscle mass

329

Downloaded from jn.nutrition.org by guest on June 6, 2015

predominantly physically active, and one-fifth were current smokers (Table 3). Fat intake as PFE was 31.4%, with SFAs comprising the greatest proportion of energy, followed by MUFAs, total PUFAs, TFAs, and total n–3 PUFAs, which was similar to other representative U.K. population surveys (47). Linoleic acid (18:2n26) represented 97% of total n–6 PUFA intake, which was in agreement with other population data available (48,49). Within the cohort, 30.3% were monozygotic twins, with the remainder being dizygotic twins. Differences between monozygotic and dizygotic twins for dietary fat were not significant except for intakes of TFAs (P = 0.039). The monozygotic twins were ;2 y older (P = 0.006), were 0.5 cm shorter (P = 0.048), and had a lower BMI (mean 6 SD: 24.5 6 4.0 vs.

were different for SFAs and PUFAs (SFAs: unadjusted P-trend = 0.002 vs. adjusted P < 0.001; PUFAs: unadjusted P-trend = 0.084 vs. adjusted P = 0.40). For FFM, the associations did not differ for P-trends between adjusted and unadjusted models for TFAs but were different for PFE (unadjusted P = 0.18 vs. adjusted P = 0.004), SFAs (P = 0.068 vs. P = 0.006), MUFAs (P = 0.046 vs. P = 0.001), and PUFAs (P = 0.02 vs. P = 0.34). The associations in women aged $50 y (n = 1354) (not shown in the tables) were, overall, not different from those found in the whole cohort, although those for FFM and oleic acid (P-trend = 0.23), total n–3 PUFAs (P = 0.29), and LC n–3 PUFAs (P = 0.11) were not significant. The P-trend for the P:S ratio was attenuated to P = 0.054. For FFMI, the results were not different from the whole cohort analyses except for the P-trend for total n–3 PUFAs (P = 0.33), total LC n–3 PUFAs (P = 0.073), and the P:S ratio (P = 0.073), which were attenuated when compared with the analyses for the whole cohort. For the results for dietary comparisons with 10 y of age, the scale of the associations per quintile of intake with FFM (as a percentage of the association of 10 y of age) was calculated, and, for PFE, was 95% of that of age. That is, the association per 10 y of age for FFM was 0.21 kg (b coefficient). When the association per quintile of PFE, which was 0.20 kg of FFM (b coefficient), Downloaded from jn.nutrition.org by guest on June 6, 2015

FIGURE 1 Indices of skeletal muscle mass stratified by quintile categories of individual SFAs (as a percentage of energy) in 2689 women aged 18–79 y. Values are adjusted mean FFM (in kg) (A) or FFMI (calculated as kg FFM/m2) (B) calculated according to category of individual FAs adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting. n = 538 for Q1–Q4 and n = 537 for Q5. P for trend across quintiles: *P , 0.05, **P , 0.01, ***P , 0.001. FFM, fat-free mass; FFMI, fat-free mass index; Q, quintile.

FFM, although the trends were not significant; however, for linoleic acid, the extreme quintile differences were significant. Both total n–3 PUFAs and long-chain (LC) n–3 PUFAs were negatively associated with FFM (Table 1). Comparisons between quintiles 5 and 1 for PFE, SFAs, MUFAs, and TFAs revealed differences of 20.2 to 20.8 kg of FFM, whereas for the P:S ratio, total PUFAs, total n–6 PUFAs, and linoleic acid, differences were 0.6, 0.2, 0.4, and 0.4 kg, respectively. For FFMI, the associations with FA intake were in a similar direction to those for FFM, with significant negative associations for PFE, SFAs, MUFAs, and TFAs (P = 0.001) (Table 2). Extreme quintile differences for SFAs, MUFAs, and TFAs ranged from 20.26 to 20.38 kg/m2, and the quintile difference for the P:S ratio was 0.25 kg/m2. There was a positive association between FFMI and the P:S ratio (P = 0.003). FFMI was also lower in quintile 5 than in quintile 1 for total (P-trend = 0.09) and LC (P = 0.045) n–3 PUFAs. Additional adjustment for percentage of protein intake did not materially modify the associations for either FFM or FFMI, i.e., the adjusted mean values and significance of the trends did not differ substantially. The associations between FFMI and dietary fat that were not adjusted for covariates (unadjusted models) were, overall, not different from the fully adjusted associations for PFE, MUFAs, and TFAs. However, compared with the adjusted model, these 330

Welch et al.

FIGURE 2 Stratification of indices of skeletal muscle mass by tertile of combined intake of SFAs and TFAs and tertiles of PUFAs (as a percentage of energy) in 2689 women aged 18–79 y. Values are mean FFM (in kg) (A) or FFMI (calculated as kg FFM/m2) (B) adjusted for age, physical activity, smoking habit, total body fat (in kg), energy intake, and misreporting. For SFA1TFA: T1, n = 897; T2 and T3, n = 896. P for trend across tertiles of SFAs and TFAs: *P , 0.01. FFM, fat-free mass; FFMI, fat-free mass index; SFAs1TFAs, total SFAs and trans fatty acids; T, tertile; TFA, trans fatty acid.

TABLE 3 Physical and behavioral characteristics of the cohort and dietary intakes of 2689 women aged 18-79 y1 Value 48.2 6 12.7 162.4 6 6.0 65.6 6 11.3 24.9 6 4.2 22.8 6 7.9 39.6 6 5.3 15.0 6 1.7 18.4 78.0 1980 6 526 69.0 6 24.2 (31.4 6 5.6) 26.1 6 10.5 (11.7 6 3.0) 1.1 6 0.7 (0.5 6 0.3) 2.5 6 1.2 (1.1 6 0.4) 0.29 6 0.13 (0.10 6 0) 13.4 6 5.2 (6.0 6 1.4) 6.0 6 2.6 (2.7 6 0.8) 0.21 6 0.11 (0.10 6 0) 23.0 6 8.6 (10.3 6 2.2) 18.8 6 7.0 (8.4 6 1.8) 2.5 6 1.2 (1.1 6 0.4) 14.2 6 5.8 (6.4 6 1.6) 12.0 6 4.8 (5.4 6 1.5) 11.6 6 4.7 (5.2 6 1.5) 1.3 6 0.5 (0.6 6 0.2) 1.1 6 0.4 (0.52 6 0.15) 0.19 6 0.15 (0.1 6 0.1) 0.59 6 0.23

Discussion

Values are means 6 SDs or percentages of participants. FFM, fat-free mass; FFMI, fat-free mass index; PFE, percentage of energy from fat; P:S ratio, PUFA-to-SFA ratio; TFA, trans fatty acid.

1

was calculated as a percentage of the association with 10 y of age (i.e., 0.20 was divided by 0.21), it was 95% of that of age. Therefore, the association with PFE was approximately equivalent to the effect of the difference in 10 y of age. For FFMI, the

TABLE 4

To our knowledge, this is the first large-scale, cross-sectional study to comprehensively investigate and find significant associations between dietary intake of PFE and the full range of FA profiles with indices of muscle mass, in a wide age range of women using a precise method of assessing body composition (DXA). We found that a higher P:S ratio was associated with greater FFM and FFMI, suggestive of muscle conservation. Conversely, higher PFE, SFAs, MUFAs, TFAs, and individual SFAs were associated with lower FFM and FFMI, indicating potential muscle loss. The associations found were significant after adjustment for physical activity and smoking behavior and for EI, potential misreporting, and protein intake. However, the combined stratified analysis with tertiles of SFAs+TFAs and PUFAs showed that the associations were only significant for SFAs+TFAs, indicating a greater influence of SFAs+TFAs than PUFAs on indices of muscle mass. The difference between

Distribution of dietary fat and FA intake by quintile categories in 2689 women aged 18–79 y1

Dietary fat

Q1 (n = 538)

Q2 (n = 538)

Q3 (n = 538)

Q4 (n = 538)

Q5 (n = 537)

Total fat (PFE) FAs (PFE) SFAs MUFAs Oleic acid TFAs PUFAs Total n–6 PUFAs Linoleic acid Total n–3 PUFAs a-Linolenic acid Long-chain n–3 PUFAs P:S ratio

23.4 6 2.9

28.5 6 1.0

31.6 6 0.8

34.4 6 0.9

39.0 6 2.6

7.7 6 7.4 6 6.0 6 0.6 6 4.4 6 3.6 6 3.4 6 0.41 6 0.35 6 0.02 6 0.33 6

10.1 9.2 7.5 0.9 5.5 4.6 4.4 0.51 0.44 0.05 0.46

11.6 6 10.3 6 8.4 6 1.1 6 6.3 6 5.3 6 5.1 6 0.59 6 0.50 6 0.07 6 0.56 6

13.2 11.4 9.4 1.3 7.1 6.1 5.9 0.67 0.58 0.10 0.67

1.2 1.0 0.8 0.1 0.6 0.5 0.5 0.05 0.04 0.01 0.06

6 0.5 6 0.4 6 0.3 6 0.1 6 0.2 6 0.2 6 0.2 6 0.02 6 0.02 6 0.01 6 0.03

0.4 0.3 0.3 0.1 0.2 0.2 0.2 0.02 0.02 0.01 0.03

6 0.5 6 0.3 6 0.3 6 0.1 6 0.3 6 0.3 6 0.3 6 0.03 6 0.03 6 0.01 6 0.04

16.0 6 13.4 6 11.0 6 1.7 6 8.8 6 7.6 6 7.4 6 0.87 6 0.74 6 0.20 6 0.94 6

1.7 1.3 1.3 0.2 1.1 1.1 1.1 0.14 0.12 0.08 0.20

Values are means 6 SDs calculated according to quintiles of percentage of energy from fat or FA intake (as a percentage of energy intake). PFE, percentage of energy from fat; P:S, PUFA-to-SFA ratio; Q, quintile; TFA, trans fatty acid.

1

Dietary fat and skeletal muscle mass

331

Downloaded from jn.nutrition.org by guest on June 6, 2015

Age, y Height, cm Weight (DXA), kg BMI, kg/m2 Fat mass, kg Fat-free mass, kg FFMI, kg/m2 Current smokers, % Physical activity, % active Energy intake, kcal/d Dietary intake, g/d (PFE) Total fat FAs SFAs Lauric acid Myristic acid Pentadecanoic acid Palmitic acid Stearic acid Arachidic acid MUFAs Oleic acid TFAs PUFAs Total n–6 PUFAs Linoleic acid Total n–3 PUFAs a-Linolenic acid Long-chain n–3 PUFAs P:S ratio

dietary comparison of PFE with 10 y of age was 150%, i.e., ;1.5 times the relation with age. For SFAs, the corresponding values were 90% for FFM and 140% for FFMI, and for the P:S ratio were 72% and 115%, respectively. When FFMI was calculated according to quintile category of individual SFAs, significant inverse associations were observed for myristic, pentadecanoic, palmitic, and stearic acids for FFMI (all P < 0.01) (Fig. 1). For FFM, significant associations were observed for myristic (14:0), pentadecanoic (15:0), palmitic (16:0), and arachidic (20:0) acids (P < 0.01). Differences between quintiles of intake ranged from 20.35 kg of FFM for arachidic acid to 20.86 kg of FFM for pentadecanoic acid and, for FFMI, ranged from 20.2 kg/m2 for lauric acid to 20.39 kg/m2 for pentadecanoic acid. In the analyses stratified by tertiles of SFAs+TFAs, compared with PUFAs, significant trends were shown for the association between SFAs+TFAs and FFM (P = 0.004) and FFMI (P = 0.001) (Fig. 2) but not for PUFAs (FFM, P = 0.36; FFMI, P = 0.45), indicating the greater influence of SFAs+TFAs on FFM and FFMI compared with PUFAs. The differences between extremes of the tertiles comparing the optimum combination of intakes (highest PUFAs and lowest SFAs+TFAs) versus the least optimal combination (highest SFAs+TFAs and lowest PUFAs) were 0.75 kg for FFM and 0.29 kg/m2 for FFMI (Fig. 2).

332

Welch et al.

The strengths of this study include the large representative sample and the objective assessment of body composition by DXA. Our measurements were over a wide range of ages, not just in elderly women, although our analyses showed that the associations remained in women aged $50 y. Although none of our cohort were in the range of muscle mass currently considered to be sarcopenic (all had an FFMI $6.76 kg/m2), our findings indicate that total fat intake and FA profile are likely to be relevant for muscle mass at all ages (7). Furthermore, total FFM is strongly correlated with ALM in a subset of this and another cohort [r 5 0.97 (P < 0.001); r 5 0.99 in women and r 5 0.98 in men] (59). We also used a detailed FA database that allowed us to investigate associations for individual FAs, as well as the major components of fat intake (SFAs, TFAs, MUFAs, and PUFAs). Compared with the few other U.K. and German studies with detailed dietary fat intake, our finding that the majority of total n–6 PUFAs was supplied by linoleic acid was supported (48,49). In addition, intakes of the major polyunsaturated fractions in our study were similar to those of the U.K. National Dietary Survey (47). The limitations of our study are that, as with any crosssectional study design, no causal associations can be made and we cannot exclude the possibility of residual confounding, despite adjusting for all of the currently established confounders known to be associated with loss of muscle mass. However, given our detailed adjustment of confounders, it is unlikely that this would account for the observed results. We used a validated FFQ for our dietary measures that has previously shown associations between plasma phospholipid FAs ranging from r = 0.04 (P = 0.63) for SFAs to r = 0.17 (P = 0.01) for TFAs and r = 0.27 (P = 0.0002) for PUFAs (60). The limitations of FA databases include the temporal changes to the composition of the oils and fats used in food manufacturing, but our database was derived around the time the FFQs were collected. Although we adjusted our findings for protein intake, we were unable to adjust for leucine, which is known to be a key driver for protein synthesis, because this information was unavailable (61). However, this requires additional investigation in the future. Our findings relate to women, and additional studies are required to examine whether similar associations would be observed in men. However, given that the proposed mechanisms are not gender-specific, with inflammation and insulin resistance occurring in both men and women, we speculate that our findings would apply also to men. Indeed, the 1 other study to investigate dietary saturated fat and muscle mass found a relation in both men and women (30). Although our results are from cross-sectional observations and our hypotheses are based on inference from scientific evidence from animal and human studies, a randomized controlled trial would be needed to provide confirmatory evidence to support our findings. However, our results indicate associations between both dietary PFE and FA profile and muscle mass, which we hypothesize could operate differently through effects on insulin resistance and inflammation. Fat varies considerably in the diet (we found 1.7- to 3.9-fold extreme quintile differences for PFE and for the different FAs); therefore, in terms of prevention, there is potential for individual- and population-level modifications in intake. In conclusion, we found associations between PFE and individual FA profile and indices of skeletal muscle mass in women aged 18–79 y; these associations were positive for the P:S ratio and negative for total fat, MUFAs, SFAs, and TFAs. Although the scale of the associations was relatively small, the associations were significant after accounting for the known influences on muscle mass and, when compared to the association with 10 y of age, it ranged from 0.75 to 1.5 times that of 10 y

Downloaded from jn.nutrition.org by guest on June 6, 2015

extreme quintiles of PFE intake (quintile 1 vs. quintile 5) for FFM was 20.6 kg and for FFMI was 20.28 kg/m2. For quintiles of SFAs, MUFAs, and TFAs, extreme quintile differences ranged from 20.5 to 20.8 kg for FFM and from 20.26 to 20.38 kg/m2 for FFMI. Although the differences we found appear to be relatively small, when per-quintile associations with diet were compared with those calculated for 10 y of age, they ranged from 72% for the P:S ratio to 95% for PFE for FFM and from 115% for the P:S ratio to 150% for PFE for FFMI. To our knowledge, only 1 other cross-sectional human population study has investigated associations between dietary fat and skeletal muscle (30). That study investigated only saturated fat, although the finding was in the same direction as ours: a significant negative association between saturated fat and appendicular lean mass (ALM) in 1099 noninstitutionalized men and women aged 50–79 y (20.05 kg of ALM per quartile of saturated fat intake) (30). Our findings are consistent with a number of plausible mechanisms that link fat intake to muscle biology and metabolism and to those involved in muscle loss. There are strong associations between dietary FA profile and the composition of FAs within the sarcolemma (muscle membrane), demonstrating the influence of dietary fat on muscle composition (24). Chronic low-grade inflammation (circulating cytokines) is associated with muscle loss in elderly people (2,20). SFAs and TFAs increase inflammation; PUFAs, which consist of primarily n–6 PUFAs, are considered to reduce it (2,26), although certain recent studies have suggested that n–6 PUFAs may also have proinflammatory activities (50,51). Insulin resistance, which has been associated with muscle loss, is also influenced adversely by SFA and TFA intake and positively by PUFA intake (2,25,26,29,52). Furthermore, in insulin resistance, there is an accumulation of intramuscular lipid (ceramide and diacylglycerol), which may also be increased by higher SFA intakes (2,26). Moreover, the Akt/ mammalian target of rapamycin pathway (mTOR) (the major pathway that regulates protein turnover) is downregulated by both insulin resistance and the accumulation of ceramide (53,54). Dietary SFAs increased the production of inflammatory cytokines in muscle cells in an in vitro study (28). In addition, in a human intervention study, PUFAs acutely affected TG-derived skeletal muscle FA uptake and increased postprandial insulin sensitivity (52). High-fat diets may also downregulate the Akt/mTOR pathway, with an animal study showing that a high-fat diet reduced growth of muscle in exercise conditions (55). Dietary fat composition is a key element of dietary recommendations aimed at reducing the risk and burden of CVD, with dietary guidelines advising lower intakes of total fat, SFAs, and TFAs and higher intakes of PUFAs, particularly LC n–3 PUFAs (46). Most of our findings were in the direction of the current dietary advice for protection against CVD, with a higher P:S ratio being positively associated with a greater FFM and FFMI, whereas SFAs and TFAs were negatively associated, indicating dietary fat composition could also influence muscle mass as well as CVD risk. However, stearic acid has been considered to be neutral for CVD, whereas we found that it and the other individual SFAs were associated with FFM (46). We also found an unexpected inverse association between LC n–3 PUFA intake and FFM and FFMI, in contrast to the limited number of studies that have shown a positive influence on protein turnover and grip strength (56–58). One explanation for the difference between our findings and others may be the low mean intakes of dietary n–3 PUFAs (0.19 g/d EPA and DHA or 0.1% of energy) in our population compared with intervention studies (doses of 4 g of LC n–3 PUFAs or 7.9% of energy) (57,58).

of age. These novel findings suggest that a dietary fat profile that is already associated with CVD protection may also be beneficial for conservation of skeletal muscle mass. Additional investigations using prospective and intervention studies are required. Acknowledgments A.A.W. and A.C. created the study concept and drafted the manuscript; T.D.S. and A.J.M. set up and coordinated the collection of all data; A.A.W. performed the statistical analyses, oversaw and developed the FA database, and had primary responsibility for the final content of the manuscript; and A.J.M., A.-M.M., J.S., A.A.V., and T.D.S. provided critical review of the manuscript. All authors read and approved the final manuscript.

Literature Cited 1. 2.

3.

5. 6.

7.

8.

9.

10.

11.

12. 13.

14. 15. 16. 17. 18. 19.

Dietary fat and skeletal muscle mass

333

Downloaded from jn.nutrition.org by guest on June 6, 2015

4.

Wolfe RR. The underappreciated role of muscle in health and disease. Am J Clin Nutr. 2006;84:475–82. Corcoran MP, Lamon-Fava S, Fielding RA. Skeletal muscle lipid deposition and insulin resistance: effect of dietary fatty acids and exercise. Am J Clin Nutr. 2007;85:662–77. Welch AA. Nutritional influences on age-related skeletal muscle loss. Proc Nutr Soc. 2013;12:1–18. Schlanger LE, Bailey JL, Sands JM. Electrolytes in the aging. Adv Chronic Kidney Dis. 2010;17:308–19. Woolf AD, Akesson K. Preventing fractures in elderly people. BMJ. 2003;327:89–95. Cooper C, Dere W, Evans W, Kanis JA, Rizzoli R, Sayer AA, Sieber CC, Kaufman JM, Abellan van Kan G, Boonen S, et al. Frailty and sarcopenia: definitions and outcome parameters. Osteoporos Int 2012;23:1839–48. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, Boirie Y, Cederholm T, Landi F, Martin FC, Michel JP, Rolland Y, Schneider SM, et al. Sarcopenia: European consensus on definition and diagnosis. Report of the European Working Group on Sarcopenia in Older People. Age Ageing. 2010;39:412–23. Wannamethee SG, Shaper AG, Lennon L, Whincup PH. Decreased muscle mass and increased central adiposity are independently related to mortality in older men. Am J Clin Nutr. 2007;86:1339–46. MacInnis RJ, Cassar C, Nowson CA, Paton LM, Flicker L, Hopper JL, Larkins RG, Wark JD. Determinants of bone density in 30- to 65-yearold women: a co-twin study. J Bone Miner Res. 2003;18:1650–6. Salamone LM, Glynn N, Black D, Epstein RS, Palermo L, Meilahn E, Kuller LH, Cauley JA. Body composition and bone mineral density in premenopausal and early perimenopausal women. J Bone Miner Res. 1995;10:1762–8. Szulc P, Blaizot S, Boutroy S, Vilayphiou N, Boonen S, Chapurlat R. Impaired bone microarchitecture at the distal radius in older men with low muscle mass and grip strength: the STRAMBO study. J Bone Miner Res 2013;28:169–78. Harvey N, Dennison E, Cooper C. Osteoporosis: impact on health and economics. Nat Rev Rheumatol. 2010;6:99–105. International Osteoporosis Foundation. Facts and statistics about osteoporosis and its impact. 2010 [cited 2013 Feb 25]. Available from: http://www.iofbonehealth.org/facts-and-statistics.html. Sayer AA. Sarcopenia. BMJ. 2010;341:c4097. Wackerhage H, Rennie MJ. How nutrition and exercise maintain the human musculoskeletal mass. J Anat. 2006;208:451–8. Kiens B. Skeletal muscle lipid metabolism in exercise and insulin resistance. Physiol Rev. 2006;86:205–43. Frayn KN. Fat as a fuel: emerging understanding of the adipose tissueskeletal muscle axis. Acta Physiol (Oxf). 2010;199:509–18. Pedersen BK. Muscle-to-fat interaction: a two-way street? J Physiol. 2010;588:21. Kien CL, Everingham KI, Stevens RD, Fukagawa NK, Muoio DM. Short-term effects of dietary fatty acids on muscle lipid composition and serum acylcarnitine profile in human subjects. Obesity (Silver Spring). 2011;19:305–11.

20. Schaap LA, Pluijm SM, Deeg DJ, Harris TB, Kritchevsky SB, Newman AB, Colbert LH, Pahor M, Rubin SM, Tylavsky FA, et al. Higher inflammatory marker levels in older persons: associations with 5-year change in muscle mass and muscle strength. J Gerontol A Biol Sci Med Sci. 2009;64:1183–9. 21. Leenders M, Verdijk LB, van der Hoeven L, Adam JJ, van Kranenburg J, Nilwik R, van Loon LJ. Patients with type 2 diabetes show a greater decline in muscle mass, muscle strength, and functional capacity with aging. J Am Med Dir Assoc. 2013;14:585–92. 22. Honors MA, Kinzig KP. The role of insulin resistance in the development of muscle wasting during cancer cachexia. J Cachexia Sarcopenia Muscle. 2012;3:5–11. 23. Waterlow JC. Protein turnover. Wallingford (UK): CABI Publishing; 2006. 24. Andersson A, Nalsen C, Tengblad S, Vessby B. Fatty acid composition of skeletal muscle reflects dietary fat composition in humans. Am J Clin Nutr. 2002;76:1222–9. 25. Calder PC, Albers R, Antoine JM, Blum S, Bourdet-Sicard R, Ferns GA, Folkerts G, Friedmann PS, Frost GS, Guarner F, et al. Inflammatory disease processes and interactions with nutrition. Br J Nutr. 2009; 101(Suppl 1):S1–45. 26. Kennedy A, Martinez K, Chuang CC, LaPoint K, McIntosh M. Saturated fatty acid-mediated inflammation and insulin resistance in adipose tissue: mechanisms of action and implications. J Nutr. 2009;139:1–4. 27. Pillon NJ, Bilan PJ, Fink LN, Klip A. Cross-talk between skeletal muscle and immune cells: muscle-derived mediators and metabolic implications. Am J Physiol Endocrinol Metab. 2013;304:E453–65. 28. Pillon NJ, Arane K, Bilan PJ, Chiu TT, Klip A. Muscle cells challenged with saturated fatty acids mount an autonomous inflammatory response that activates macrophages. Cell Commun Signal. 2012;10:30. 29. Manco M, Calvani M, Mingrone G. Effects of dietary fatty acids on insulin sensitivity and secretion. Diabetes Obes Metab. 2004;6:402–13. 30. Scott D, Blizzard L, Fell J, Giles G, Jones G. Associations between dietary nutrient intake and muscle mass and strength in communitydwelling older adults: the Tasmanian Older Adult Cohort Study. J Am Geriatr Soc. 2010;58:2129–34. 31. Teucher B, Skinner J, Skidmore PM, Cassidy A, Fairweather-Tait SJ, Hooper L, Roe MA, Foxall R, Oyston SL, Cherkas LF, et al. Dietary patterns and heritability of food choice in a UK female twin cohort. Twin Res Hum Genet. 2007;10:734–48. 32. Andrew T, Hart DJ, Snieder H, de Lange M, Spector TD, MacGregor AJ. Are twins and singletons comparable? A study of disease-related and lifestyle characteristics in adult women. Twin Res. 2001;4:464–77. 33. Samaras K, Kelly PJ, Chiano MN, Spector TD, Campbell LV. Genetic and environmental influences on total-body and central abdominal fat: the effect of physical activity in female twins. Ann Intern Med. 1999;130:873–82. 34. Kyle UG, Schutz Y, Dupertuis YM, Pichard C. Body composition interpretation. Contributions of the fat-free mass index and the body fat mass index. Nutrition. 2003;19:597–604. 35. Welch AA, Luben R, Khaw KT, Bingham SA. The CAFE computer program for nutritional analysis of the EPIC-Norfolk food frequency questionnaire and identification of extreme nutrient values. J Hum Nutr Diet. 2005;18:99–116. 36. The Food Standards Agency. McCance and WiddowsonÕs the composition of foods sixth summary edition. Cambridge (UK): Royal Society of Chemistry; 2002. 37. Frankenfield DC, Muth ER, Rowe WA. The Harris-Benedict studies of human basal metabolism: history and limitations. J Am Diet Assoc. 1998;98:439–45. 38. Otten JJ, Hellwig JP, Meyers LD. Dietary Reference Intakes: the essential guide to nutrient requirements. Washington: National Academies Press; 2006. 39. Rennie KL, Coward A, Jebb SA. Estimating under-reporting of energy intake in dietary surveys using an individualised method. Br J Nutr. 2007;97:1169–76. 40. Paddon-Jones D, Sheffield-Moore M, Cree MG, Hewlings SJ, Aarsland A, Wolfe RR, Ferrando AA. Atrophy and impaired muscle protein synthesis during prolonged inactivity and stress. J Clin Endocrinol Metab. 2006;91:4836–41. 41. Szulc P, Duboeuf F, Marchand F, Delmas PD. Hormonal and lifestyle determinants of appendicular skeletal muscle mass in men: the MINOS study. Am J Clin Nutr. 2004;80:496–503.

334

Welch et al.

52. Jans A, Konings E, Goossens GH, Bouwman FG, Moors CC, Boekschoten MV, Afman LA, Muller M, Mariman EC, Blaak EE. PUFAs acutely affect triacylglycerol-derived skeletal muscle fatty acid uptake and increase postprandial insulin sensitivity. Am J Clin Nutr. 2012;95:825–36. 53. Hyde R, Hajduch E, Powell DJ, Taylor PM, Hundal HS. Ceramide down-regulates system A amino acid transport and protein synthesis in rat skeletal muscle cells. FASEB J. 2005;19:461–3. 54. Hu F, Liu F. Targeting tissue-specific metabolic signaling pathways in aging: the promise and limitations. Protein Cell. Epub 2013 Oct 17. 55. Sitnick M, Bodine SC, Rutledge JC. Chronic high fat feeding attenuates load-induced hypertrophy in mice. J Physiol. 2009;587:5753–65. 56. Rodacki CL, Rodacki AL, Pereira G, Naliwaiko K, Coelho I, Pequito D, Fernandes LC. Fish-oil supplementation enhances the effects of strength training in elderly women. Am J Clin Nutr. 2012;95:428–36. 57. Smith GI, Atherton P, Reeds DN, Mohammed BS, Rankin D, Rennie MJ, Mittendorfer B. Omega-3 polyunsaturated fatty acids augment the muscle protein anabolic response to hyperinsulinaemia-hyperaminoacidaemia in healthy young and middle-aged men and women. Clin Sci (Lond). 2011;121:267–78. 58. Smith GI, Atherton P, Reeds DN, Mohammed BS, Rankin D, Rennie MJ, Mittendorfer B. Dietary omega-3 fatty acid supplementation increases the rate of muscle protein synthesis in older adults: a randomized controlled trial. Am J Clin Nutr. 2011;93:402–12. 59. Schautz B, Later W, Heller M, Muller MJ, Bosy-Westphal A. Total and regional relationship between lean and fat mass with increasing adiposity–impact for the diagnosis of sarcopenic obesity. Eur J Clin Nutr. 2012;66:1356–61. 60. Patel PS, Sharp SJ, Jansen E, Luben RN, Khaw KT, Wareham NJ, Forouhi NG. Fatty acids measured in plasma and erythrocytemembrane phospholipids and derived by food-frequency questionnaire and the risk of new-onset type 2 diabetes: a pilot study in the European Prospective Investigation into Cancer and Nutrition (EPIC)–Norfolk cohort. Am J Clin Nutr. 2010;92:1214–22. 61. Dillon EL. Nutritionally essential amino acids and metabolic signaling in aging. Amino Acids. 2013;45:431–41.

Downloaded from jn.nutrition.org by guest on June 6, 2015

42. Castillo EM, Goodman-Gruen D, Kritz-Silverstein D, Morton DJ, Wingard DL, Barrett-Connor E. Sarcopenia in elderly men and women: the Rancho Bernardo study. Am J Prev Med. 2003;25:226–31. 43. Petersen AM, Mittendorfer B, Magkos F, Iversen M, Pedersen BK. Physical activity counteracts increased whole-body protein breakdown in chronic obstructive pulmonary disease patients. Scand J Med Sci Sports. 2008;18:557–64. 44. Petersen AM, Magkos F, Atherton P, Selby A, Smith K, Rennie MJ, Pedersen BK, Mittendorfer B. Smoking impairs muscle protein synthesis and increases the expression of myostatin and MAFbx in muscle. Am J Physiol Endocrinol Metab. 2007;293:E843–8. 45. Kyle UG, Genton L, Lukaski HC, Dupertuis YM, Slosman DO, Hans D, Pichard C. Comparison of fat-free mass and body fat in Swiss and American adults. Nutrition. 2005;21:161–9. 46. Hunter JE, Zhang J, Kris-Etherton PM. Cardiovascular disease risk of dietary stearic acid compared with trans, other saturated, and unsaturated fatty acids: a systematic review. Am J Clin Nutr. 2010;91:46–63. 47. Pot GK, Prynne CJ, Roberts C, Olson A, Nicholson SK, Whitton C, Teucher B, Bates B, Henderson H, Pigott S, et al. National Diet and Nutrition Survey: fat and fatty acid intake from the first year of the rolling programme and comparison with previous surveys. Br J Nutr. 2012;107:405–15. 48. Bakewell L, Burdge GC, Calder PC. Polyunsaturated fatty acid concentrations in young men and women consuming their habitual diets. Br J Nutr. 2006;96:93–9. 49. Kro¨ger J, Zietemann V, Enzenbach C, Weikert C, Jansen EH, Doring F, Joost HG, Boeing H, Schulze MB. Erythrocyte membrane phospholipid fatty acids, desaturase activity, and dietary fatty acids in relation to risk of type 2 diabetes in the European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study. Am J Clin Nutr. 2011;93:127–42. 50. Calder PC. Old study sheds new light on the fatty acids and cardiovascular health debate. BMJ. 2013;346:f493. 51. Patterson E, Wall R, Fitzgerald GF, Ross RP, Stanton C. Health implications of high dietary omega-6 polyunsaturated fatty acids. J Nutr Metab. 2012;2012:539426.