Polyunsaturated Fatty Acid Intake and Blood Pressure in Adolescents

Home , Eicosapentaenoic acid

Journal of Human Hypertension (2012) 26, 178–187 & 2012 Macmillan Publishers Limited All rights reserved 0950-9240/12 www.nature.com/jhh ORIGINAL ARTICLE Polyunsaturated fatty acid intake and blood pressure in adolescents

TA O’Sullivan1, AP Bremner2, LJ Beilin3, GL Ambrosini1,4, TA Mori3, RC Huang1,3 and WH Oddy1 1Telethon Institute for Child Health Research, Centre for Child Health Research, University of Western Australia, West Perth, Western Australia, Australia; 2School of Population Health, University of Western Australia, Perth, Western Australia, Australia; 3School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia, Australia and 4MRC Human Nutrition Research, Cambridge, UK

Evidence that intake of polyunsaturated fatty acids were inversely associated with intakes of long chain (PUFAs) may modify blood pressure (BP) is generally omega-3fattyacidsinboys(b ¼À3.93, P ¼ 0.01, b ¼À4.05, limited to middle-aged or hypertensive populations. This P ¼ 0.01, respectively). For specific long-chain omega-3s, study examined cross-sectional associations between BP significant inverse associations were observed between and dietary intake of PUFAs in 814 adolescents aged 13– eicosapentaenoic acid (EPA) and docosahexaenoic 15 years participating in the Western Australian Pregnancy acid, such as systolic BP decreasing by 4.7 mm Hg Cohort (Raine) Study. Fatty acid intakes were assessed (95% CI –9.3 to À0.1) for a quarter gram increase in EPA, using 3-day diet records and resting BP was determined but no significant associations were observed with using multiple oscillometric readings. In multivariate docosapentaenoic acid. No significant associations were regression models, systolic BP was inversely associated observed in girls, or with the omega-6 to omega-3 ratio. with intakes of polyunsaturated (b ¼À0.436, Po0.01), Our results suggest that gender may moderate relation- omega-3 (b ¼À2.47, P ¼ 0.02), omega-6 (b ¼À0.362, ships between fatty acid intake and BP in adolescence. P ¼ 0.04) and long chain omega-3 fatty acids (b ¼À4.37, Journal of Human Hypertension (2012) 26, 178–187; P ¼ 0.04) in boys. Diastolic BP and mean arterial pressure doi:10.1038/jhh.2011.7; published online 10 February 2011

Keywords: blood pressure; adolescence; teenagers; polyunsaturated fatty acids; omega-3; omega-6

Introduction of the omega-3 PUFAs.5,6 A meta-analysis of randomised trials found that intake of fish oil high Of all the leading global risk factors, high blood in omega-3 fatty acids may lower BP, especially in pressure (BP) is the greatest cause of attributable older (445 years) and hypertensive subjects.7 Long- mortality and the third greatest cause of disease 1 chain omega-3 fatty acids in particular have strong burden. Although often asymptomatic, hyperten- anti-inflammatory effects,8 enhance vascular en- sion can contribute to increased risk of many dothelial function9 and reduce platelet adhesive- conditions including stroke, heart attack, heart ness,10 which may contribute to a reduced risk of failure and renal disease. It has been suggested that hypertension. Most studies have focused on the two even a short period of hypertension when young can 2 long-chain fatty acids eicosapentaenoic acid (EPA) increase hypertension risk in adulthood. Early and docosahexaenoic acid (DHA), commonly found prevention of high BP is therefore an imperative 3 in fish oil. Although docosapentaenoic acid (DPA) is public health issue around the world. Although it is an intermediate in the conversion of EPA to DHA, it generally accepted that some dietary factors such as is uncertain if DPA has similar cardio-protective salt intake can modulate BP, the effects of fatty acid 4 effects. The derived food sources of these long-chain intake on BP are less well studied, particularly omega-3s are very different; meat is a major source in younger populations. of DPA,11 whereas fish and seafood are major Higher intakes of polyunsaturated fatty acids sources of EPA and DHA. A further area of interest (PUFAs) have been associated with lower BP, and is the relationship between omega-6 PUFAs and BP, this relationship is thought to be largely because as the relationship between omega-6 fatty acids and cardiovascular risk is controversial.12 Some Correspondence: Dr TA O’Sullivan, School of Exercise, Biomedi- researchers recommend decreasing the omega-6 to cal and Health Science, Edith Cowan University, 270 Joondalup omega-3 ratio to reduce adverse effects resulting Drive, Joondalup, Western Australia 6027, Australia. from excess arachidonic acid and its eicosanoid E-mail: [email protected] 13 Received 30 June 2010; revised 10 December 2010; accepted 20 products, such as inflammation. In contrast, there December 2010; published online 10 February 2011 is strong evidence that the substitution of omega-6 Fatty acids and blood pressure in adolescents TA O’Sullivan et al 179 rich foods for saturated fats reduces cardiovascular record booklet with a set of metric cups and spoons, disease risk.14 and was given verbal and written instructions. It has been noted that public health strategies to Adolescents completed the food record themselves, prevent hypertension in children and adolescents however, they were advised to ask for parental are urgently required.15 There is a lack of popula- support if they were unsure of what items were or tion-based research that investigates the relation- how to record them. Adolescents also specified ship between BP and fatty acids in adolescents. Data whether each of the 3 days represented their usual on nutrient intakes and other variables need to be dietary intake. To optimise data accuracy, a dietitian extensive and of high quality to enable detailed checked each food record when it was returned, and reporting of fatty acid intake by individuals, as clarified any missing or ambiguous data with the well as controlled for possible confounders.16 The adolescent by telephone.19 Western Australian Pregnancy Cohort (Raine) Study The FoodWorks Professional Version 5 dietary is ideal for investigating these relationships, as it is a analysis program ( Xyris Software (Australia) Pty Ltd, large, population-based cohort, with data available Queensland, Australia) with the Australian Food and on physical, socioeconomic and lifestyle factors. Nutrient Database was used to analyse data into The aim of our study was to examine cross-sectional energy, macronutrient and fatty acid intakes. To associations between BP and dietary intakes of obtain data on specific fatty acids, individual food PUFAs in adolescents participating in the 14-year items were linked to the Fatty Acid Compositional follow-up of the Raine Study. We hypothesised that Database, available in FoodWorks.20,21 This database after adjustment for potential confounding factors, includes fatty acid data (g/100 g) for 1044 Australian systolic, diastolic and mean arterial BP would be foods. The database was further customised using inversely associated with dietary omega-3 fatty updated food composition data available online.22 acids, particularly long-chain omega-3 fatty acids, As our study focused on PUFAs, investigation of and positively associated with the ratio of dietary fatty acids included: total omega-3, a-linoleic acid, omega-6 to -3 in adolescents. total long chain omega-3 consisting of EPA, DPA and DHA, EPA, DPA and DHA individually, total omega- 6, linoleic acid, arachidonic acid and the ratio Methods of omega-6 to omega-3. The relationship between dietary intake of fatty acids and the fatty acid Subjects content of red blood cell (RBC) membrane fatty Women between 16–20 weeks gestation were acid data was also investigated. Collection of the recruited for the Raine Study through the public RBC data has been described previously.23 The fatty antenatal clinic at King Edward Memorial Hospital acids were expressed as a percentage of the total and private clinics in Perth, Western Australia, from fatty acids measured (C14 to C22). May 1989 to November 1991 (n ¼ 2900). Recruitment details for this longitudinal observational study have been published previously.17 Of the initial Assessment of BP cohort, 2868 live births were available for follow-up, The 14-year follow-up included assessment of BP with follow-ups occurring at birth and at ages 1, 2, 3, using a Dinamap ProCare 100 Monitor (General 5, 8, 10 and 14 years. This study investigates data Electric HealthCare Technologies, Rydalmere, New collected during years 2003–2006 inclusive, at the South Wales, Australia) with a cuff size appropriate 14-year follow-up. Assessments were conducted at for arm circumference. To give a resting measure- the Telethon Institute for Child Health Research in ment, adolescents sat for 5 min before BP readings Perth, Western Australia. This follow-up was the were taken. Systolic and diastolic BPs were calcu- first to comprehensively assess dietary intake and lated as the mean of the last five of six readings, allow assessment of fatty acid intake. The ethics taken over a 10 min period. Mean arterial pressure committees of King Edward Memorial Hospital and was calculated as the diastolic pressure plus one- Princess Margaret Hospital approved the protocol third of the pulse pressure. for all aspects of the study. Each adolescent as well as their parent or guardian provided written consent Potential confounding factors for participation in the study. Physical measurements. The adolescent’s body weight was measured to the nearest 100 g using a Wedderburn Digital Chair Scale, with adolescents Assessment of dietary intake dressed in running shorts and singlet tops. Height Dietary intake was assessed using a 3-day estimated was measured to the nearest 0.1 cm with a Holtain food record, measured in household units. Three- Stadiometer. Weight in kilograms divided by height day food records have been shown to be an in metres squared gave body mass index (BMI). appropriate method of dietary assessment in a Puberty was assessed using Tanner stages of pubic younger population.18 Each adolescent who agreed hair development;24,25 adolescents selected their to take part in the dietary study and attended their corresponding developmental stage in a privately follow-up assessment was provided with a food completed questionnaire from a set of standard

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 180 drawings depicting Tanner stages two (sparse) to 14-year follow-up (age mean±s.d. 14.0±0.2 years, five (adult) (stage one was omitted as this corre- range 13.0–15.0 years). The remainder were lost to sponds to pre-pubescent age younger than 10 years). follow-up, withdrawn or temporarily deferred from the study (n ¼ 975), or deceased (n ¼ 32). Of the 1286 Sociodemographic and family characteristics. Ma- adolescents who agreed to complete the 3-day ternal age at conception, education level and family record, 962 returned their booklets, however, 103 income at the 14-year follow-up were obtained were incomplete and 37 included at least 1 day by parent report. Maternal age at conception was that was not representative of their usual diet. classified as: o20 years, 20–29 years, or X30 years. No subjects were excluded for implausible energy Maternal education was assessed by the highest intakes (defined as o3000 or 420 000 kJ per day). school year completed, with responses grouped as: Another five subjects were excluded for use of pgrade 10, grade 11, or grade 12. Current family medications that may influence BP, and three did income, defined as the annual income for the not participate in BP assessment. This left a total of household before tax at the time of the follow-up 814 adolescents (395 girls and 419 boys) with data ($AUD), was grouped as: o$35 000pa, $35 000– available for analysis. Dietary intake, BP and 70 000pa, or 4$70 000pa. Family history of hyper- anthropometric values for boys and girls are shown tension was assessed as either yes or no depending in Table 1. Higher overall energy intake in boys on whether a biological parent had been medically contributed to significantly greater intakes for diagnosed with hypertension. some nutrients, but this difference dissipated in the comparison of energy-adjusted values. For both boys and girls, consumption of fat contributed 32% Lifestyle factors. To assess physical activity, of total energy intake. Girls had significantly lower adolescents estimated how many hours they spent systolic and mean arterial BPs, but boys were watching television/videos and using computers for significantly taller and heavier. Energy adjusted school, work and recreation; and how many hours they spent on average outside of school hours fatty acid variables were not significantly different by gender. Subjects who completed the 3-day food participating in physical activity that caused breath- record were more likely to have a lower BMI, older lessness or sweating. This information was grouped mothers or higher family income, compared with into five classifications ranging from very sedentary other adolescents in the follow-up who did not (low exercise, high TV/computer use) to very active complete a food record.20 Dietary intakes of DHA (high exercise and low TV/computer screen use). and linoleic acid (LA) were significantly correlated with corresponding RBC measures (r ¼ 0.2, Po0.01 Statistical analysis and r ¼ 0.1, Po0.05, respectively), but associations were weaker than those reported in previous Differences in dietary, BP and anthropometric 27 characteristics between boys and girls were exam- studies. There were no significant correlations ined using independent samples t-tests, and means with other fatty acids of interest. and s.d. reported (mean±s.d.). The relationship Characteristics of the study adolescents by between dietary fat intake and RBC fatty acid quartile of energy adjusted PUFA intake showed content were assessed with bivariate correlations. systolic BP decreased in a step-wise manner with Nutrient intakes were adjusted for energy using the increasing quartile of intake (Table 2a). Puberty, residual method.26 ANOVA and w2-tests were used to BMI, energy intake and family history of hyperten- assess differences in BP and other factors over sion were significantly positively associated with quartiles of polyunsaturated fat intake for boys and BP (Table 2b). Results from the multivariate linear girls combined. Multivariate linear regression mod- regression analyses of systolic, diastolic and mean els, adjusted for puberty, screen use and physical arterial BP (Tables 3a–c) showed significant inverse activity, family income, family history of hyperten- relationships between some fatty acids for boys. sion, maternal age, maternal education, BMI, energy Higher intakes of total fat, total polyunsaturated, adjusted dietary fibre, sodium per MJ and energy omega-3, EPA, omega-6 and linoleic fatty acids were intake, were used to analyse associations between associated with lower systolic BP, with long-chain fatty acids and BP. As boys and girls had signifi- omega-3 showing borderline significance. Higher cantly different BP, analyses were stratified by intakes of long-chain omega-3, EPA and DHA were gender. The Statistical Package for Social Sciences associated with both lower diastolic and mean for Windows, Rel.15.0.0. 2006 (SPSS Inc., IBM, arterial BP, and higher arachidonic acid was asso- Chicago, IL, USA) was used for analyses, and ciated with lower diastolic BP. In girls, no significant associations were observed, although total statistical significance was set at Po0.05. omega-6 and linoleic acid showed a borderline positive association with diastolic BP. The fatty acid Results variables that made the greatest contribution to prediction of BP in the models were total PUFAs From the original cohort of 2868 live births, 1861 for systolic BP, and long-chain omega-3 for diastolic adolescents completed at least one aspect of the and mean arterial BP.

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 181 Table 1 Daily dietary intakes by gender of Raine Study between PUFA intake and BP. Although the con- adolescents in blood pressure analysis (mean±s.d.), with blood fidence intervals were wide, EPA was the only fatty ± pressure and anthropometric values (mean s.d.) acid that was significantly associated with all three Girls Boys Pa measures of BP: systolic, diastolic and mean arterial (n ¼ 395) (n ¼ 419) pressure. Unlike DHA and EPA, DPA was not significantly associated with BP measures in any Absolute values of the models, however, the direction of effect was Energy (MJ) 8.27±1.77 10.50±2.46 o0.01 the same. Carbohydrate (g) 248±62 315±85 o0.01 Our findings are supported by other cross- Carbohydrate (%) 49.4±6.4 49.4±5.9 0.87 Dietary fibre (g) 20.2±6.2 24.3±8.5 o0.01 sectional studies that show significant inverse associa- Protein (g) 76.3±19.7 100.5±26.0 o0.01 tions between PUFAs and BP in predominantly free- Protein (%) 15.8±2.9 16.4± 2.9 o0.01 living/non-hypertensive adults, as summarised in a Alcohol (g) 0.03±0.35 0.02±0.15 0.56 review by Hall.4 We observed small correlations ± ± Alcohol (%) 0.01 0.13 o0.01 0.04 0.37 between PUFA intake and RBC fatty acids, and it is Fat (g) 72.3±19.7 90.0±25.3 o0.01 Fat (%) 32.2±5.0 31.7±4.7 0.11 likely that the lipid composition of cell membranes 28 Saturated fat (g) 30.2±9.5 38.9±12.8 o0.01 is one aspect contributing to BP. Other ways that Monounsaturated fat (g) 23.9±7.0 29.7±8.9 o0.01 PUFAs are likely to affect BP include alteration of Polyunsaturated fat (g) 9.6±3.5 11.3±4.1 o0.01 eicosanoid synthesis, endothelial function, vascular Omega-3 (g) 1.12±0.56 1.33±0.61 o0.01 28 Omega-6 (g) 7.45±2.97 8.65±3.43 o0.01 reactivity and nitric oxide production. In addition, Ratio omega-6 to 7.44±3.15 7.13±2.95 0.15 different classes of PUFAs may exert more specific omega-3 effects on BP through varying mechanisms.5 Long-chain 0.174±0.288 0.190±0.263 0.38 Although both EPA and DHA showed independent omega-3 (g) inverse associations with measures of BP, we ob- Sodium (g) 2.48±0.71 3.16±0.91 o0.01 served slightly stronger relationships with EPA than Energy adjusted values DHA in boys. Mori and Woodman29 have reported Fat (g) 82.2±11.1 80.8±13.0 0.08 that EPA and DHA have differing haemodynamic and Saturated fat (g) 34.9±6.2 34.5±7.7 0.50 anti-atherogenic properties. For example, increased Monounsaturated fat (g) 27.1±4.6 26.7±5.4 0.25 Polyunsaturated fat (g) 10.7±2.9 10.3±3.3 0.08 EPA intake is associated with a reduction in whole- 30 Omega-3 (g) 1.25±0.51 1.21±0.54 0.28 blood viscosity, while DHA has been shown to Omega-6 (g) 8.23±2.64 7.93±2.99 0.13 improve vasodilator responses.9 The effects of EPA Ratio omega-6 to 7.20±2.85 7.36±3.70 0.50 and DHA on BP may also vary because of individual omega-3 29 Long-chain 0.191±0.285 0.183±0.262 0.67 factors such as medication use. omega-3 (g)

Blood pressure The ratio of omega-6 to omega-3 fatty acids Systolic blood pressure 108.8±9.1 114.3±10.2 o0.01 There were no significant associations between the (mm Hg) Diastolic blood pressure 59.3±6.8 58.8±6.8 0.27 ratio of omega-6 to omega-3 fatty acids and any (mm Hg) measures of BP, and higher intakes of both omega-6 Mean arterial blood 75.8±6.7 77.3±6.8 o0.01 and omega-3 fatty acids were significantly and pressure (mm Hg) independently associated with lower systolic BP Anthropometrics in boys. Although previous research in this area is Height (m) 1.62±0.10 1.67±0.09 o0.01 limited, our findings are supported by a study of Weight (kg) 55.9±11.1 57.7±12.6 0.03 elderly subjects which demonstrated that both BMI 21.2±3.8 20.7±3.7 0.06 omega-3 and omega-6 supplements effectively lowered BP.31 Similarly, a study of dyslipidaemic Abbreviation: BMI, body mass index. patients showed that dietary supplementation of aP values represent independent t-tests. either linoleic acid or a-linolenic acid both reduced BP measures, although the a-linolenic acid treatment produced significantly better results.32 Discussion In addition, a cross-sectional study of healthy middle-aged men has shown an inverse association PUFAs and BP between serum omega-6 linoleic acid and BP.33 In a The results of this study demonstrate that omega-3 systematic review of evidence on omega-6 fatty fatty acids, particularly the long-chain fatty acids, acids and cardiovascular disease,14 Harris et al. are inversely associated with BP in adolescents and concluded that consumption of at least 5–10% of provide support for our hypothesis. Although the energy from omega-6 PUFAs is likely to reduce direction of effect was the same in boys and girls, the risk of coronary heart disease relative to lower significant associations were observed only in boys. intakes. Results from the Framingham Heart Study We showed a significant inverse relationship be- suggest that almost one-third of coronary heart tween systolic BP and total fat in boys, which is disease events are attributable to high BP, after most likely related to the stronger association adjustment for other risk factors.34

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 182 The significance of the omega-6 to omega-3 ratio appear to significantly affect arachidonic acid levels remains equivocal because of conflict over the in cell membranes or inflammation.38 It has also benefit of omega-6 (see refs 12, 14); some reports been demonstrated that omega-6 fatty acids can have suggest a lower ratio is associated with a reduced anti-inflammatory properties in vitro39 and in vivo.40 risk of chronic disease,8 and in the Raine Study of Lower omega-6 to omega-3 intakes may enhance girls, we noted a positive association between formation of long-chain omega-3 by attenuating omega-6 and diastolic BP. This theory is due in part competition for the enzymes involved in the to inflammation, which has been shown to be metabolism of both the short-chain omega-6 and -3 associated with increased BP,35 as the long-chain fatty acids. However, the conversion of short-chain omega-6 arachidonic acid is the substrate for the to long-chain omega-3 is limited in humans, even synthesis of pro-inflammatory and vasoconstrictor without competition from omega-6 (see ref. 41). prostanoids.14 Interestingly, our results showed a Harris et al. suggest that decreasing the ratio by significant inverse association between arachidonic increasing omega-3 is beneficial, but decreasing the acid and diastolic BP in boys. This may be because ratio by decreasing intake of omega-6 may not have of dietary omega-6 intake resulting in changes thesameeffect.14 The conclusion made by Harris in eicosanoid production affecting aspects such as et al., that higher intakes of omega-6 are likely to be vasodilation.36 Although a hypotensive influence of more beneficial than lower intakes, supports our omega-6 fatty acids was suggested in early clinical finding that BP is significantly inversely associated studies,37 this has not been confirmed in subsequent with omega-6 in boys. randomised controlled trials.5 The significance of this correlation is therefore unclear, but may represent confounding by unidentified dietary or Gender-specific associations with fatty acid intake lifestyle associations. We observed a gender difference in the relationship Evidence suggests that changing dietary consump- between fatty acid intake and BP, with significant tion of omega-6 over a wide range of intakes does not associations in boys, but not girls. Other studies

Table 2a Characteristics of Raine Study adolescents by quartile (Q) of energy adjusted polyunsaturated fat intake, using ANOVA or cross-tabs analysis

Characteristic Quartile of energy adjusted polyunsaturated fat intake (g)

Q1 Q2 Q3 Q4 Pa

Quartile mean±s.d. 7.10±1.08 9.19±0.45 11.00±0.64 14.66±2.57 Mean±s.d. Systolic blood pressure (mm Hg) 113±10 112±11 111±10 110±9 0.02 Diastolic blood pressure (mm Hg) 59.3±6.8 59.0±7.1 59.2±6.5 58.7±6.8 0.82 Mean arterial pressure (mm Hg) 77.3±6.8 76.6±7.2 76.6±6.7 75.9±6.5 0.22 Body mass index (kg m2) 20.9±3.5 20.5±3.3 21.5±4.4 20.9±3.7 0.05 Puberty (Tanner score for pubic hair development) 3.91±0.71 3.84±0.82 3.84±0.80 3.87±0.78 0.82 Sodium (mg per MJ energy intake) 289±71 312±72 307±73 304±67 0.01 Dietary fibre (energy adjusted) 21.6±6.6 22.4±6.1 22.0±5.5 23.2±5.9 0.04 n (%) Gender (number and percentage of male) 119 (58.3) 109 (53.2) 93 (45.6) 99 (48.5) 0.06 Family history of hypertension 17 (8.3) 14 (6.8) 7 (3.4) 15 (7.4) 0.21

Annual family income ($AUD) o$35 000 45 (22.5) 43 (21.1) 48 (24.2) 33 (16.5) 0.08 $35 000–$70 000 68 (34.0) 83 (40.7) 73 (36.9) 63 (31.5) 4$70 000 87 (43.5) 77 (38.9) 77 (38.9) 104 (52.0)

Maternal age at conception o20 years 14 (6.8) 14 (6.8) 19(9.3) 10 (4.9) 0.26 20–29 years 95 (46.6) 98 (47.8) 108 (52.9) 93 (45.6) X30 years 95 (46.6) 93(45.4) 77 (37.8) 101 (49.5)

Physical activityb Very low 3 (1.5) 6 (3.0) 4 (2.0) 5 (2.4) 0.80 Low 56 (28.1) 50 (24.9) 43 (21.2) 47 (23.3) Moderate 65 (32.7) 70 (34.8) 72 (35.4) 65 (32.2) High 47 (23.6) 52 (25.9) 56 (27.6) 64 (31.7) Very high 28 (14.1) 23 (11.4) 28 (13.8) 21 (10.4)

Abbreviation: ANOVA, analysis of variance. aANOVA test for equality of means or w2 test for equality of proportions. bVery low, exercise o1 per week and screen use X4 h per day; low, exercise o1 per week and screen use 2–4 h per day or exercise 1–3 per week and screen use X4 h per day; moderate, exercise 1–3 per week and screen use 2–4 h per day; high, exercise 1–3 per week and screen use o2hper day or exercise X4 per week and screen use 2–4 h per day; very high, exercise X4 per week and screen use o2 h per day.

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 183 Table 2b Characteristics of Raine Study adolescents by blood pressure, using Pearson’s correlation or ANOVA

Characteristic Blood pressure measurement

Systolic Pa Diastolic Pa Mean arterial Pa

Pearson’s r Body mass index (kg mÀ2) 0.24 o0.01 0.06 0.08 0.16 o0.01 Puberty (Tanner score for pubic hair development) 0.08 0.04 À0.05 0.22 0.01 0.83 Energy (MJ) 0.20 o0.01 À0.04 0.28 0.07 0.04 Sodium (mg per MJ energy intake) À0.01 0.85 0.01 0.73 0.01 0.89 Dietary fibre (energy adjusted) 0.05 0.18 0.01 0.71 0.03 0.37 Mean±s.d. Gender Male 114±10 o0.01 58.3±6.8 0.27 77.3±6.8 o0.01 Female 109±9 59.3±6.8 75.8±6.7

Family history of hypertension Absent 111±10 0.04 58.9±6.8 0.05 76.4±6.7 0.02 Present 114±12 60.8±6.9 78.6±7.5

Annual family income ($AUD) o$35 000 112±11 0.60 59.5±6.9 0.17 76.9±6.8 0.22 $35 000–$70 000 112±11 59.5±6.9 77.0±7.1 4$70 000 111±9 58.6±6.7 76.1±6.5

Maternal age at conception o20 years 113±10 0.38 59.2±6.6 0.34 77.3±6.9 0.39 20–29 years 111±10 58.7±6.4 76.3±6.2 X30 years 112±10 59.4±7.2 76.8±7.3

Physical activityb Very low 111±12 0.89 58.4±9.6 0.09 75.8±9.2 0.27 Low 112±10 59.3±6.6 76.9±6.4 Moderate 112±10 59.8±7.1 77.1±7.1 High 111±11 58.1±6.6 75.8±6.9 Very high 112±9 58.7±6.3 76.3±6.1

Abbreviation: ANOVA, analysis of variance. aPearson’s correlation or ANOVA test for equality of means. bVery low, exercise o1 per week and screen use X4 h per day; low, exercise o1 per week and screen use 2–4 h per day or exercise 1–3 per week and screen use X4 h per day; moderate, exercise 1–3 per week and screen use 2–4 h per day; high, exercise 1–3 per week and screen use o2hper day or exercise X4 per week and screen use 2–4 h per day; very high, exercise X4 per week and screen use o2 h per day.

Table 3a Multivariate linear regression analysis with systolic blood pressure as the dependent variable, in Raine Study adolescentsa

Fatty acid typeb Girls Boys

Unstandardised Standardised P-value Unstandardised Standardised P-value b coefficient (95% CI) b coefficient b coefficient (95% CI) b coefficient

Total fat 0.04 (À0.05, 0.14) 0.054 0.38 À0.09 (À0.18, À0.01) À0.122 0.03 Polyunsaturated fat À0.13 (À0.47, 0.21) À0.045 0.45 À0.44 (À0.76, À0.13) À0.146 0.01 Total omega-3 À0.09 (À2.11, 1.94) À0.005 0.93 À2.47 (À4.55, À0.40) À0.124 0.02 a-linoleic acid 0.09 (À2.29, 2.48) À0.004 0.94 À2.02 (À4.45, 0.41) À0.088 0.10 Long-chain omega-3 À0.53 (À4.28, 3.22) À0.016 0.78 À4.16 (À8.40, 0.08) À0.104 0.05 EPA À2.05 (À16.10, 12.00) À0.017 0.77 À18.67 (À37.11, À0.22) À0.108 0.04 DPA À14.64 (À45.8, 16.6) À0.054 0.36 À5.43 (À32.22, 21.35) À0.022 0.69 DHA À0.36 (À5.70, 4.99) À0.008 0.90 À5.37 (À11.05, 0.31) À0.100 0.06 Total omega-6 À0.02 (À0.38, 0.35) À0.006 0.92 À0.38 (À0.72, À0.03) À0.114 0.03 Linoleic acid À0.02 (À0.39, 0.35) À0.006 0.92 À0.38 (À0.73, À0.03) À0.114 0.03 Arachidonic acid 0.23 (À6.64, 7.10) À0.004 0.95 À0.42 (À7.31, 6.46) À0.007 0.90 Ratio omega-6:3 À0.10 (À0.45, 0.25) À0.033 0.58 0.11 (À0.17, 0.40) 0.042 0.44

Abbreviations: DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; LCN3, long chain omega-3 fatty acids (EPA+DPA+DHA). aModels adjusted for puberty, screen use and physical activity, family income, family history of hypertension, maternal education, maternal age, body mass index, energy adjusted dietary fibre, sodium per MJ, energy intake. bEnergy adjusted values in g/day. Bolded values highlight significant result (Po0.05).

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 184 Table 3b Multivariate linear regression analysis with diastolic blood pressure as the dependent variable, in Raine Study adolescentsa

Fatty acid typeb Girls Boys

Unstandardised Standardised P-value Unstandardised Standardised P-value b coefficient (95% CI) b coefficient b coefficient (95% CI) b coefficient

Total fat 0.01 (À0.06, 0.09) 0.023 0.71 À0.01 (À0.07, 0.05) À0.028 0.64 Polyunsaturated fat 0.12 (À0.14, 0.38) 0.054 0.37 À0.08 (À0.31, 0.15) À0.040 0.49 Total omega-3 À0.36 (À1.94, 1.21) À0.027 0.65 À0.77 (À2.26, 0.72) À0.057 0.31 a-linoleic acid 0.22 (À1.64, 2.07) 0.014 0.82 0.26 (À1.48, 2.00) 0.017 0.77 Long-chain omega-3 À1.78 (À4.69, 1.13) À0.072 0.23 À4.10 (À7.10, À1.10) À0.152 0.01 EPA À3.77 (À14.69, 7.15) À0.040 0.50 À17.16 (À30.21, À4.11) À0.147 0.01 DPA À9.70 (À33.98, 14.59) À0.047 0.43 À15.17 (À34.13, 3.80) À0.090 0.12 DHA À2.79 (À6.94, 1.36) À0.079 0.19 À4.97 (À8.99, À0.95) À0.137 0.02 Total omega-6 0.28 (0.00, 0.57) 0.118 0.05 À0.06 (À0.31, 0.19) À0.026 0.65 Linoleic acid 0.29 (0.00, 0.58) 0.119 0.05 À0.05 (À0.30, 0.21) À0.020 0.72 Arachidonic acid 0.77 (À4.57, 6.12) 0.018 0.78 À5.10 (À9.97, À0.23) À0.120 0.04 Ratio omega-6:3 0.22 (À0.05, 0.49) 0.098 0.10 0.12 (À0.08, 0.32) 0.065 0.25

Abbreviations: EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; LCN3, long chain omega-3 fatty acids (EPA+DPA+DHA). aModels adjusted for puberty, screen use and physical activity, family income, family history of hypertension, maternal education, maternal age, BMI, energy adjusted dietary fibre, sodium per MJ, energy intake. bEnergy adjusted values in g per day. Bolded values highlight significant result (Po0.05).

Table 3c Multivariate linear regression analysis with mean arterial blood pressure as the dependent variable, in Raine Study adolescentsa

Fatty acid typeb Girls Boys

Unstandardised Standardised P-value Unstandardised Standardised P-value b coefficient (95% CI) b coefficient b coefficient (95% CI) b coefficient

Total fat 0.024 (À0.049, 0.097) 0.040 0.52 À0.041 (À0.099, 0.018) À0.080 0.18 Polyunsaturated fat 0.037 (À0.218, 0.292) 0.017 0.78 À0.202 (À0.425, 0.021) À0.099 0.08 Total omega-3 À0.272 (À1.805, 1.260) À0.021 0.73 À1.34 (À2.79, 0.12) À0.100 0.07 a-linoleic acid 0.175 (À1.630, 1.980) 0.011 0.85 À0.501 (À2.205, 1.204) À0.032 0.56 Long-chain omega-3 À1.36 (À4.20, 1.47) À0.056 0.34 À4.12 (À7.06,À1.18) À0.154 0.01 EPA À3.20 (À13.82, 7.43) À0.035 0.55 À17.66 (À30.47,À4.86) À0.152 0.01 DPA À11.35 (À34.96, 12.27) À0.056 0.35 À11.92 (À30.57, 6.73) À0.071 0.21 DHA À1.98 (À6.02, 2.06) À0.058 0.34 À5.10 (À9.04, À1.16) À0.142 0.01 Total omega-6 0.183 (À0.093, 0.458) 0.078 0.19 À0.163 (À0.407, 0.080) À0.074 0.19 Linoleic acid 0.187 (À0.093, 0.467) 0.078 0.19 À0.156 (À0.401, 0.088) À0.070 0.21 Arachidonic acid 0.594 (À4.600, 5.788) 0.014 0.82 À3.54 (À8.33, 1.25) À0.084 0.15 Ratio omega-6:3 0.116 (À0.147, 0.380) 0.052 0.39 0.115 (À0.082, 0.313) 0.064 0.25

have noted possible gender differences with BP. In our study, the Raine adolescents were nearing Rates of hypertension are higher in men and the completion of puberty at the 14-year follow-up, postmenopausal women compared with premeno- with a mean Tanner stage of four out of a possible pausal women, with vascular protective effects five. The steepest increment in testosterone levels in of the female sex hormones potentially contributing boys occurs between pubertal stages two and three,44 to these observations.42 In rats, vascular contraction and estradiol concentrations are highly correlated is similar between castrated and intact male rats, but with Tanner stages in girls.45 Hormonal differences significantly enhanced in females who have had at the 14-year follow-up may also help explain why ovaries removed compared with intact females, our results were primarily significant in boys. Girls’ implying oestrogens rather than androgens affect higher oestrogen concentrations may negate the BP.42 Oestrogen has been shown to promote vasodi- beneficial effects of dietary fatty acid. A randomised lation both in human and animal models.43 Taken controlled trial of omega-3 fatty acid supplementa- together, this information suggests that hormones tion in children from infancy until 5 years of age may have a role in modulation of BP. did not result in significant improvements in BP at

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 185 the age of 8 years in boys or girls,46 suggesting adolescence is linked to health outcomes later that changes in BP due to fatty acid intake may only in life. On the basis of work by McCarron et al.,50 be evident from puberty. Thus, the impact of we calculate that a systolic BP reduction of 5 mm Hg hormones from pubertal development may modu- in males aged 15–29 years decreases later cardio- late the relationship between dietary fatty acids and vascular disease mortality by 6.8%. In the present BP during adolescence. BPs in girls were also study, results showed systolic BP decreased on substantially lower than boys in our study, with average by 4.7 mm Hg (95% CI À9.3 to À0.1) for a less statistical power to detect whether diet had an quarter gram increase in the long-chain omega-3 effect on BP. fatty acid EPA in the daily diet of adolescent boys. To put this in perspective, a quarter gram of EPA daily is equivalent to two serves of Atlantic Strengths and limitations salmon per week (150 g serves), or two small A strength of our study was the use of a compre- cans of sardines per week (90 g cans).51 In addition, hensive and updated fatty acid composition data- the expanding range of foods fortified with base. Although dietary assessment using 3-day food long-chain omega-3 fatty acids means that non- records has been shown to have good agreement traditional food sources can also contribute to between observed and reported dietary intakes in a overall intakes. younger population,18 a potential limitation was that foods consumed on an occasional basis may not What is known on this topic have been included, and this may have contributed K Higher intakes of polyunsaturated fatty acids (PUFAs), to weaker than expected correlations with RBC fatty particularly long-chain omega-3 fatty acids, have been acids. Further, interpretation of our study results is associated with lower blood pressure in adults. limited by the cross-sectional design, and general- K The association of omega-6 fatty acids with cardiovascular risk is controversial. isation of our results requires care, as adolescents in K Little is known on whether the long-chain omega-3 fatty our study were more likely to participate in the acid docosapentaenoic acid (DPA) shows cardio-protective dietary assessment if they had lower BMIs, older effects similar to eicosapentaenoic acid (EPA) and mothers or higher family incomes.20 docosahexaenoic acid (DHA). What does this study adds K This research provides evidence of a significant inverse Conclusion relationship between blood pressure and both omega-3 and omega-6 fatty acids in male adolescents. Our study suggests that increased dietary intake of K Long-chain omega-3 fatty acids EPA and DHA were both omega-3 and omega-6 PUFAs in adolescence, significantly associated with lower blood pressure, whereas DPA was not. especially long-chain omega-3 fatty acids, may be beneficial in prevention of high BP, particularly for boys. In addition to the sources of long-chain omega- 3 fatty acids of fish, seafood and meat, sunflower Conflict of interest seeds, walnuts, brazil nuts and sunflower and canola oils are also good sources of shorter-chain The authors declare no conflict of interest. omega-3 and omega-6 fatty acids. These fatty acids are essential in the diet, and although the conversion of short to long-chain omega-3 in the body is limited, increasing a-linoleic intake can increase Acknowledgements the absolute amount of DHA synthesised.47 In this Raine population group, the highest contributing We would like to express our gratitude to all the food sources of omega-6 fatty acids were margarine, families who took part in this study and the Raine bread, poultry and vegetable oils; and for omega-3 Study team, which includes data collectors, cohort fatty acids, they were dairy products (providing managers, data managers, clerical staff, research short-chain omega-3 a-linoleic acid), red meat scientists and volunteers. Thanks also goes to Xyris (both long-chain omega-3 and a-linoleic acid), Software, Brisbane. We acknowledge the support margarine (a-linoleic acid) and fish (long-chain of the Telethon Institute for Child Health Research omega-3).20 and Royal Perth Hospital laboratories. Sources Prominent epidemiologist Geoffrey Rose noted of funding: The Western Australian Pregnancy that a downward shift of the BP distribution in the Cohort (Raine) Study is funded by the Raine Medical population of only 2–3 mm Hg may equate to the Research Foundation at The University of Western benefits achieved by antihypertensive treatment.48 Australia, the National Health and Medical Research Therefore, small changes in eating habits have the Council of Australia, the Telstra Research Founda- potential to result in large overall benefits on a tion, the Western Australian Health Promotion population scale. Strong evidence for the tracking of Foundation, the Australian Rotary Health Research BP from childhood to adulthood has been shown Fund and the National Heart Foundation of in studies of diverse population groups,49 and BP in Australia and Beyond Blue.

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 186 References 17 Newnham JP, Evans SF, Michael CA, Stanley FJ, Landau LI. Effects of frequent ultrasound during 1 Ezzati M, Rodgers A, Lopez AD, Hoorn SV, Murray CJL. pregnancy: a randomised controlled trial. Lancet Mortality and burden of disease attributable to 1993; 342: 887–891. individual risk factors. In: Ezzati M, Lopez A, Rodgers 18 Crawford P, Obarzanek E, Morrison J. Comparative A, Murray C (eds). Comparative Quantification of advantage of 3-day food records over 24-h recall and Health Risks Global and Regional Burden of Disease 5-day food frequency validated by observation of 9- and Attributable to Selected Major Risk Factors. Geneva 10-year-old girls. JAmDietAssoc1994; 94(6): 626–630. World Health Organization, Geneva, Switzerland, 19 Di Candilo KG, Oddy WH, Miller M, Sloan N, 2004. pp 2141–2165. Kendall GE, de Klerk NH. Follow-up phone calls 2 Flynn JT. Hypertension in the young: epidemiology, increase nutrient intake estimated by three-day food sequelae and therapy. Nephrol Dial Transplant 2009; diaries in 13-year-old participants of the Raine study. 24(2): 370–375. Nutr Diet 2007; 64(3): 165–171. 3 Luma GB, Spiotta RT. Hypertension in children and 20 O’Sullivan TA, Ambrosini GL, Beilin LJ, Mori TA, adolescents. Am Fam Physician 2006; 73: 1158–1168. Oddy WH. Dietary intake and food sources of fatty 4 Hall WL. Dietary saturated and unsaturated fats as acids in Australian adolescents. Nutrition 2011; 27(2): determinants of blood pressure and vascular function. 153–159. Nutr Res Rev 2009; 22(01): 18–38. 21 Mann NJ, Sinclair AJ, Percival P, Lewis JL, Meyer BJ, 5 Mori TA, Burke V, Beilin LJ. Dietary fats and blood Howe PRC. Development of a database of fatty acids in pressure. In: Lip G, Hall J (eds). Comprehensive Australian foods. Nutr Diet 2003; 60(1): 42–45. Hypertension Section I: Epidemiology. Elsevier: 22 Food Standards Australia New Zealand. Nutrient Philadelphia, USA, 2007. pp 77–88. Tables for use in Australia: Online Version. 2006 [cited 6 Grynberg A. Hypertension prevention: from nutrients May-July 2008]; Available from: http://www.foodstan- to (fortified) foods to dietary patterns. Focus on fatty dards.gov.au/monitoringandsurveillance/nuttab2006/ acids. J Hum Hypertens 2005; 19(S3): S25–S33. onlineversionintroduction/onlineversion.cfm. 7 Geleijnse JM, Giltay EJ, Grobbee DE, Donders ART, 23 Mori T, Burke V, Puddey I, Watts G, O’Neal D, Best J Kok FJ. Blood pressure response to fish oil supple- et al. Purified eicosapentaenoic acid and doco- mentation: metaregression analysis of randomized sahexaenoic acid have differential effects on serum trials. J Hypertens 2002; 20(8): 1493–1499. lipids and lipoproteins, LDL—particle size, glucose 8 Simopoulos AP. The importance of the omega-6/omega- and insulin, in mildly hyperlipidaemic men. Am J Clin 3 fatty acid ratio in cardiovascular disease and other Nutr 2000; 71(5): 1085–1094. chronic diseases. Exp Biol Med 2008; 233(6): 674–688. 24 Tanner J. Growth at Adolescence: With a General 9 Mori TA, Watts GF, Burke V, Hilme E, Puddey IB, Consideration of the Effects of Hereditary and Beilin LJ. Differential effects of eicosapentaenoic acid Environmental Factors upon Growth and Maturation and docosahexaenoic acid on vascular reactivity of from Birth to Maturity. Blackwell Scientific: Oxford, the forearm microcirculation in hyperlipidemic, over- England, 1962. weight men. Circulation 2000; 102(11): 1264–1269. 25 Duke PM, Litt IF, Gross RT. Adolescents’ self-assess- 10 Dyerberg J, Bang HO, Stoffersen E, Moncada S, ment of sexual maturation. Pediatrics 1980; 66(6): Vane JR. Eicosapentaenoic acid and prevention of 918–920. thrombosis and atherosclerosis? Lancet 1978; 2(8081): 26 Willett W, Howe G, Kushi L. Adjustment for total 117–119. energy intake in epidemiologic studies. Am J Clin Nutr 11 Howe P, Meyer B, Record S, Baghurst K. Dietary 1997; 65(4): 1220S–1228S. intake of long-chain omega-3 polyunsaturated fatty 27 Øverby NC, Serra-Majem L, Andersen LF. Dietary acids: contribution of meat sources. Nutrition 2006; assessment methods on n-3 fatty acid intake: a 22(1): 47–53. systematic review. Br J Nutr 2009; 102(SupplementS1): 12 Tribole E. What happened to do no harm? The issue of S56–S63. dietary omega-6 fatty acids. Prostaglandins Leukot 28 Abeywardena MY, Head RJ. Longchain n-3 polyunsa- Essent Fatty Acids 2009; 80(1): 78–79. turated fatty acids and blood vessel function. 13 Simopoulos AP, Leaf A, Salem N. Workshop statement Cardiovasc Res 2001; 52(3): 361–371. on the essentiality of and recommended dietary intake 29 Mori TA, Woodman RJ. The independent effects for omega-6 and omega-3 fatty acids. Leukot Essent of eicosapentaenoic acid and docosahexaenoic acid Fatty Acids 2000; 63(3): 119–121. on cardiovascular risk factors in humans. Curr Opin 14 Harris WS, Mozaffarian D, Rimm E, Kris-Etherton P, Clin Nutr Metab Care 2006; 9(2): 95–104. Rudel LL, Appel LJ et al. Omega-6 fatty acids and risk 30 Terano T, Hirai A, Hamazaki T, Kobayashi S, Fujita T, for cardiovascular disease: a science advisory from the Tamura Y et al. Effect of oral administration of highly American Heart Association Nutrition Subcommittee purified eicosapentaenoic acid on platelet function, of the Council on Nutrition, Physical Activity, and blood viscosity and red cell deformability in healthy Metabolism; Council on Cardiovascular Nursing; and human subjects. Atherosclerosis 1983; 46(3): 321–331. Council on Epidemiology and Prevention. Circulation 31 Margolin G, Huster G, Glueck CJ, Speirs J, Vandegrift J, 2009; 119(6): 902–907. Illig E et al. Blood pressure lowering in elderly 15 Bocelli AMD, Ballo PMD. Hypertension in children and subjects: a double-blind crossover study of omega-3 adolescents. Can Med Assoc J 2008; 179(4): 343–344. and omega-6 fatty acids. Am J Clin Nutr 1991; 53(2): 16 Ueshima H, Stamler J, Elliott P, Chan Q, Brown IJ, 562–572. Carnethon MR et al. Food omega-3 fatty acid intake of 32 Paschos G, Magkos F, Panagiotakos D, Votteas V, individuals (total, linolenic acid, long-chain) and their Zampelas A. Dietary supplementation with flaxseed blood pressure: INTERMAP Study. Hypertension 2007; oil lowers blood pressure in dyslipidaemic patients. 50(2): 313–319. Eur J Clin Nutr 2007; 61(10): 1201.

Journal of Human Hypertension Fatty acids and blood pressure in adolescents TA O’Sullivan et al 187 33 Grimsgaard S, Bonaa KH, Jacobsen BK, Bjerve KS. 43 Farhat MY, Lavigne MC, Ramwell PW. The vascular Plasma saturated and linoleic fatty acids are indepen- protective effects of estrogen. FASEB J 1996; 10(5): dently associated with blood pressure. Hypertension 615–624. 1999; 34(3): 478–483. 44 Knorr D, Bidlingmaier F, Butenandt O, Fendel H. 34 Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Plasma testosterone in male puberty. I. Physiology of Silbershatz H, Kannel WB. Prediction of coronary plasma testosterone. Acta Endocrinol (Copenh) 1974; heart disease using risk factor categories. Circulation 75(1): 181–194. 1998; 97(18): 1837–1847. 45 Rapkin AJ, Tsao JCI, Turk N, Anderson M, Zeltzer LK. 35 Chae CU, Lee RT, Rifai N, Ridker PM. Blood pressure Relationships among self-rated tanner staging, and inflammation in apparently healthy men. hormones, and psychosocial factors in healthy female Hypertension 2001; 38(3): 399–403. adolescents. J Pediatr Adolesc Gynecol 2006; 19(3): 36 Blair IA, Prakash C, Phillips MA, Dougherty RM, 181–187. Iacono JM. Dietary modification of omega 6 fatty acid 46 Ayer JG, Harmer JA, Xuan W, Toelle B, Webb K, intake and its effect on urinary eicosanoid excretion. Almqvist C et al. Dietary supplementation with n-3 Am J Clin Nutr 1993; 57(2): 154–160. polyunsaturated fatty acids in early childhood: effects 37 Vergroesen AJ, Fleischman AI, Comberg H-U. The on blood pressure and arterial structure and function influence of increased dietary linoleate on essential at age 8 y. Am J Clin Nutr 2009; 90(2): 438–446. hypertension in man. Acta Biol Med Ger 1978; 37: 47 Goyens PLL, Spilker ME, Zock PL, Katan MB, Mensink 879–883. RP. Conversion of a-linolenic acid in humans is 38 Clifton P. Dietary fatty acids and inflammation. Nutr influenced by the absolute amounts of a-linolenic acid Diet 2009; 66(1): 7–11. and linoleic acid in the diet and not by their ratio. Am J 39 De Caterina R, Liao J, Libby P. Fatty acid modulation Clin Nutr 2006; 84(1): 44–53. of endothelial activation. Am J Clin Nutr 2000; 71: 48 Rose G. Strategy of prevention: lessons from cardio- 213S–223S. vascular disease. Br Med J 1981; 282(6279): 1847–1851. 40 Pischon T, Hankinson S, Hotamisligil G, Rifai N, 49 Chen X, Wang Y. Tracking of blood pressure from Willett W, Rimm E. Habitual dietary intake of n-3 childhood to adulthood: a systematic review and and n-6 fatty acids in relation to inflammatory markers meta-regression analysis. Circulation 2008; 117(25): among US men and women. Circulation 2003; 108: 3171–3180. 155–160. 50 McCarron P, Smith GD, Okasha M, McEwen J. Blood 41 Burdge G. Alpha-linolenic acid metabolism in men pressure in young adulthood and mortality from cardio- and women: nutritional and biological implications. vascular disease. Lancet 2000; 355(9213): 1430–1431. Curr Opin Clin Nutr Metab Care 2004; 7(2): 137–144. 51 Xyris Software. Foodworks with RMIT Australian Fatty 42 Orshal JM, Khalil RA. Gender, sex hormones, and Acids: Fatty acid database compiled by the RMIT Lipid vascular tone. Am J Physiol Regul Integr Comp Physiol Research Group, Foodworks 5 Service Pack 1 (Build 2004; 286(2): R233–R249. 1376). Brisbane, Queensland, 2007.

Journal of Human Hypertension