Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Low-carbohydrate diet wikipedia , lookup
Obesity and the environment wikipedia , lookup
Calorie restriction wikipedia , lookup
Gluten-free diet wikipedia , lookup
Dietary fiber wikipedia , lookup
Diet-induced obesity model wikipedia , lookup
Saturated fat and cardiovascular disease wikipedia , lookup
Human nutrition wikipedia , lookup
Doctoral Thesis for the degree of Doctor of Philosophy, Faculty of Food Science and Nutrition Whole grain - an important part of a healthy Nordic diet Alkylresorcinols as biomarkers for whole grain wheat and rye intake Óla Kallý Magnúsdóttir University of Iceland Faculty of Food Science and Nutrition School of Health Science Reykjavík 2014 1 2 Academic Dissertation University of Iceland, Faculty of Food Science and Nutrition, School of Health Sciences and Landspitali-National University Hospital, Unit for Nutrition Research Supervised by Professor Inga Thorsdottir, University of Iceland Professor Ingibjorg Gunnarsdottir, University of Iceland PhD Committee Professor Inga Thorsdottir, University of Iceland Professor Ingibjorg Gunnarsdottir, University of Iceland Professor Matti Uusitupa, University of Eastern Finland Opponents Professor Ann-Sofie Sandberg, Chalmers University of Technology, Gothenburg, Sweden Professor Ingibjorg Hardardottir, University of Iceland, Reykjavik, Iceland Dissertation for the Degree of Ph.D. in Nutrition © Óla Kallý Magnúsdóttir 2014 Design: Prentlausnir Printed in Iceland by Prentlausnir, Reykjavík 2014 ISBN 978-9935-9130-7-4 3 It is health that is real wealth and not pieces of gold and silver - Mahatma Gandhi4 Dissertation for the Degree of Ph.D. in Nutrition Presented at the University of Iceland in 2014 Abstract Background: Evidence for the health benefits of whole grain consumption are mainly derived from epidemiological studies as findings in intervention studies have been inconsistent. The reason for inconclusive results from intervention studies may relate to lack of accurate and objective measures of whole grain intake. Alkylresorcinols (AR) appear to be highly promising biomarkers of whole grain wheat and rye intake although there are still many factors that are poorly understood. The main aim of this dissertation was to increase knowledge on the use of plasma AR as dietary biomarkers by assessing whether plasma AR could be used as biomarkers of whole grain intake in the healthy Nordic diet (ND) and to evaluate the association between plasma AR and glucose metabolism and blood lipid concentrations in a population with metabolic syndrome. Most intervention studies evaluating the health effects of whole grain, especially whole grain rye, have been conducted in populations accustomed to high intakes. Therefore, the feasibility of conducting an intervention with high consumption of whole grain rye in a population of Icelandic men, with habitually low whole grain intake, was also evaluated using plasma AR to measure dietary compliance. Methods: This PhD project was based on data from two extensive research projects, an intervention study that was a part of SYSDIET - Systems biology in controlled dietary interventions and cohort studies and the feasibility study ARI, conducted as a follow up of HELGA: Nordic Health – Whole grain Food. In the SYSDIET study participants (30–65 years, body mass index 27–40 kg/m2 with two more features of metabolic syndrome, but without diabetes) were recruited through six centers in the Nordic countries and randomly assigned to a ND (N=104) or a control diet (N=96) for 18 or 24 weeks depending on the study center. Plasma AR concentrations were analyzed (N=158) and nutrient intake was calculated from 4-day food records at weeks 0, 12 and 18/24. Determinants of AR concentration at 18/24 weeks were assessed using linear regression. Reproducibility of AR measurements between weeks 12 and 18/24 was evaluated using intra-class correlation (ICC). Associations between total plasma AR concentration and AR C17:0/C21:0 homolog ratio, as an indication of the relative whole-grain rye intake, and glucose metabolism and blood lipid measures were investigated in pooled (ND + control group) regression analyses at 18/24 weeks. In the ARI feasibility study participants (males, 42-80 years with type 2 diabetes) were randomized into either a rye group (N=14) or a control group (N=9) for 26 weeks. Participants in the rye group were provided with, and asked to consume, 100 g/day of whole grain rye. Acceptability and compliance to the whole grain rye intake was assessed with 5 questionnaires, 4-day weighed food record and plasma AR concentrations and AR C17:0/C21:0 homolog ratio. Results: Median (IQR) fasting plasma total AR concentration increased from 73 (88) to 106 (108) nmol/L, or 45%, between week 0 and 18/24 (P <0.001). The AR concentration was significantly higher in the ND group than in the control group at 18/24 weeks (P <0.001) and study group was a significant determinant for total plasma AR at week 18/24 (P <0.001). The ICCs for total plasma AR between weeks 12 and 18/24 were 0.44 – 0.72. There was a significant difference in baseline total plasma AR concentration between study centers with median concentration of 108 (96) nmol/L and 108 (72) nmol/L in Oulu and Kuopio, respectively, and 43 (37) nmol/L in Reykjavik. In a pooled analyses at 18/24 weeks, the AR C17:0/C21:0 ratio was inversely associated with fasting serum insulin concentrations (B (95% CI): -0.43 (-0.77 to -0.08)) and positively associated with the insulin sensitivity indices Matsuda ISI (0.23 (0.07 to 0.40)) and disposition index (0.77 (0.11 to 1.43)). Additionally, the AR ratio C17:0/C21:0 was inversely associated with LDL cholesterol concentrations (-0.41 (-0.80 to -0.02)), LDL/HDL cholesterol ratio (-0.20 (-0.37 to -0.03)), log non-HDL cholesterol (-0.20 (-0.37 to -0.03)), apolipoprotein B (-0.12 (-0.24 to 0.00)) and triglyceride concentrations (-0.35 (-0.59 to -0.12)) and positively associated with HDL cholesterol concentration (0.11 (0.00 to 0.21)). Total plasma AR concentration at week 18/24 was neither associated with glucose metabolism outcomes nor blood lipid measurements. Twelve participants in the rye group finished the ARI feasibility study and 8 in the control group. Whole grain rye intake and total plasma AR increased significantly in the rye group (P=0.003 and P=0.004, respectively). The mean (SD) whole grain rye consumption of the rye group during the intervention was 93 (44) g/day and 10 (8) g/day in the control group, which did not change during the intervention. Seven out of 12 participants in the rye group had a mean intake of ≥100 g/day of whole grain rye during the intervention. Four participants in the rye group experienced discomfort in the digestive system in weeks 0-13 weeks and two in weeks 13-26. Conclusion: The results suggest that the total AR concentration may be valid biomarkers for a healthy Nordic diet, in which whole grain wheat and rye are important components, and that the AR ratio C17:0/C21:0 may be used as a marker of relative whole grain rye intake. Additionally, the results indicate that increased proportion of whole grain rye, reflected by the AR C17:0/C21:0 ratio, is associated with lower fasting insulin, increased insulin sensitivity and favorable blood lipid outcomes. Icelanders have a habitually low intake of whole grain rye. However, the ARI feasibility study was well received by the population studied, as seen in the low dropout, acceptability and good compliance to the dietary intervention. Keywords: alkylresocinols, Nordic diet, whole grain, rye consumption, dietary biomarker 6 Ritgerð fyrir doktorsvörn í næringarfræði við Háskóla Íslands 2014 Ágrip Inngangur: Faraldsfræðirannsóknir hafa margsinnis sýnt fram á tengsl milli heilkornaneyslu og minni áhættu krónískra sjúkdóma en niðurstöður úr íhlutandi rannsóknum hafa verið misvísandi. Möguleg skýring á mótsagnakenndum niðurstöðum úr íhlutunum er sú að ekki hefur verið til nákvæmur og hlutlaus mælikvarði fyrir heilkornaneyslu. Notkun alkýlresorsínóla (AR) sem lífvísa fyrir neyslu á heilkornahveiti og -rúg lofar góðu þrátt fyrir að enn sé margt óþekkt um notkun þeirra. Meginmarkmið ritgerðarinnar var að auka þekkingu á notkun AR sem lífvísa með því að skoða hvort hægt væri að nota AR í blóðvökva sem lífvísi fyrir heilkornaneyslu í heilsusamlegu norrænu mataræði og með því að meta tengsl milli AR í blóðvökva við sykurefnaskipti og styrk blóðfitu hjá einstaklingum með efnaskiptavillu. Fyrri íhlutanir sem hafa metið heilsufarsleg áhrif heilkornaneyslu, sérstaklega heilkorna rúgs, hafa verið gerðar þar sem neysla á heilu korni er mikil fyrir. Það var því æskilegt að meta fýsileika slíkrar rannsóknar meðal íslenskra karlmanna, sem neyta að jafnaði lítils magns af heilu korni, og nota AR í blóðvökva sem mælikvarða á fylgni. Aðferðir: Doktorsverkefnið er byggt á gögnum frá tveimur öndvegisverkefnum Nordforsk, íhlutunarrannsókn sem er hluti af SYSDIET – Systems biology in controlled dietary interventions and cohort studies og á fýsileikarannsókninni ARI, sem er áframhald af verkefninu HELGA: Nordic Health – Whole Grain Food. SYSDIET íhlutunin var framkvæmd á 6 rannsóknarstofum í Finnlandi, Svíþjóð, Danmörku og Ísland. Þátttakendur voru á aldrinum 30-65 ára, með líkamsþyngdarstuðul 2740 kg/m2 og tvö önnur einkenni efnaskiptavillu, en án sykursýki 2. Þeim var slembiraðað í tvo hópa, hóp sem fylgdi heilsusamlegu norrænu mataræði (ND, N=104) eða hóp sem fylgdi viðmiðunarmataræði (N=96), í 18 eða 24 vikur (misjafnt eftir rannsóknarsofum). Í rannsóknarvikum 0, 12 og 18/24 var AR í blóðvökva mælt (N=158) og næringarefnainntaka metin með fjögurra daga fæðuskráningu á viku 0, 12 og 18/24. Ákvarðandi þættir fyrir styrk AR í blóðvökva á viku 18/24 voru metnir með línulegri aðhvarfsgreiningu og áreiðanleiki AR mælinga í íhlutuninni var metinn með innanflokks fylgni (ICC). Gögn frá báðum hópunum voru tekin saman og tengsl heildarstyrks AR í blóðvöka og hlutfalls AR raðkvæmana C17:0/C21:0 (sem vísbending um hlutfallslega neyslu heilkorna rúgs af heildar heilkorna rúgi og -hveiti) við sykurefnaskipti og blóðfitu í lokaviku íhlutunarrannsóknarinnar voru metin. Í fýsileikarannsókninni ARI voru þátttakendur karlar á aldrinum 42-80 ára með sykursýki af gerð 2. Þeim var slembiraðað í tvo hópa, rúghóp (N=14) eða viðmiðunarhóp (N=9) í 26 vikur. Þátttakendur í rúghópi voru beðnir um að borða 100 g af heilkornarúgi daglega. Ásættanleiki og fylgni við heilkornaneysluna voru metin með spurningalistum, fjögurra daga fæðuskráningu og AR styrk í blóðvökva. Niðurstöður: Heildarstyrkur AR í blóðvökva jókst úr 73 (88) í 106 (108) nmol/L, eða um 45%, frá viku 0 til loka rannsóknarinnar (P >0.001). Heildarstyrkur AR var þá marktækt hærri 7 í ND hópi en viðmiðunarhópi (P <0.001) og rannsóknarhópur var ákvarðandi þáttur fyrir heildarstyrk AR í lok rannsóknar (P <0.001). ICC fyrir heildarstyrk AR frá viku 12 til 18/24 var á bilinu 0.44 – 0.72. Marktækur munur var milli rannsóknarstofa á heildarstyrk AR við upphaf rannsóknar þar sem miðgildi var 108 (96) nmol/L hjá þátttakendum frá Oulu og 108 (72) nmol/L hjá þátttakendum frá Kuopio, á meðan miðgildið var 43 (37) nmol/L hjá þátttakendum frá Reykjavík. Þegar gögn frá bæði ND hópi og viðmiðunarhópi voru sameinuð var AR hlutfallið C17:0/C21:0 neikvætt tengt styrki á fastandi insúlíni (B (95% CI): -0.43 (-0.77 til -0.08)) og jákvætt tengt vísum um insúlínnæmi, Matsuda (0.23 (0.07 til 0.40)) og DI (0.77 (0.11 til 1.43)). Auk þess var AR hlutfallið C17:0/C21:0 neikvætt tengt styrki á LDL kólesteróli (-0.41 (-0.80 til -0.002)), hlutfalli LDL og HDL kólesteróls (-0.20 (-0.37 til -0.03)), ekki-HDL kólesteróli (-0.20 (-0.37 til -0.03)), apólípópróteini B (-0.12 (-0.24 til 0.00)) og þríglýseríðum (-0.35 (-0.59 til -0.12)) og jákvætt tengt HDL-kólesteróli (0.11 (0.00 to 0.21)). Ekki var marktækt samband milli þessarra breyta og heildarstyrks AR í blóðvökva við lok rannsóknar. Tólf þátttakendur í rúghópi kláruðu fýsileikarannsóknina ARA og 8 í viðmiðunarhópi. Neysla á heilkornarúgi og heildarstyrkur AR í blóðvökva jókst marktækt í rúghópi (P=0.003 og 0.004) og meðalneyslan á heilkornarúgi í íhlutuninni var 93 (44) g/dag. Meðalneysla 7 þátttakanda í rúghópi var yfir 100 g/dag. Engin breyting varð á rúgneyslu í viðmiðunarhópi og var meðalneyslan 10 (8) g/dag. Fjórir þátttakendur fundu fyrir óþægindum í meltingarvegi á fyrstu 13 vikunum og tveir á vikum 13-26. Ályktun: Niðurstöðurnar benda til þess að styrkur AR í blóðvökva geti verið gildur lífvísir í heilsusamlegu norrænu mataræði og að hægt sé að nota AR hlutfallið C17:0/C21:0 sem mælikvarða á hlutfallslega rúgneyslu. Auk þess benda niðurstöðurnar til þess að aukinn hluti heilkornarúgs, mælt með AR hlutfallinu C17:0/C21:0, sé tengt lægra fastandi insúlíni, auknu insúlínnæmi og hagstæðari blóðfitugildum. Þó Íslendingar borði jafnan lítið magn heilkorns bentu lítið brottfall og góð fylgni við mataríhlutunina í fýsileikarannsókninni til þess að íhlutun af slíku tagi sé möguleg. Lykilorð: alkýlresorsínólar, norrænt mataræði, heilkorn, rúgneysla, lífvísir 8 Table of Contents Abstract ...................................................................................................................................... 5 Ágrip........................................................................................................................................... 7 List of figures ........................................................................................................................... 11 List of tables ............................................................................................................................. 11 Abbreviations ........................................................................................................................... 13 List of original papers .............................................................................................................. 14 1. Introduction .......................................................................................................................... 15 2. Background .......................................................................................................................... 16 2.1. What is the healthy Nordic diet? ................................................................................... 16 2.2. Whole grain ................................................................................................................... 19 2.2.1. Whole grain cereals in the Nordic diet ................................................................... 20 2.2.2. Whole grain intake estimation methods and their limitations ................................ 22 2.3. Dietary biomarkers in nutritional science ...................................................................... 22 2.3.1. Dietary biomarkers of whole grain intake .............................................................. 23 2.3.2. Alkylresorcinols as biomarker of whole grain wheat and rye intake ..................... 23 2.3.3. Alkylresorcinols C17:0/C21:0 homologue ratio as marker of relative rye intake . 26 2.3.4. Bioactivity of alkylresorcinols ................................................................................ 26 2.4. Whole grain intake and human health ........................................................................... 26 2.4.1. Whole grain effects on glucose and insulin ............................................................ 27 2.4.2. Whole grain effects on blood lipids ........................................................................ 28 3. Aims ..................................................................................................................................... 30 4. Methods ................................................................................................................................ 31 4.1. The SYSDIET intervention ........................................................................................... 31 4.1.1. Study design ............................................................................................................ 31 4.1.2. Study population ..................................................................................................... 32 4.1.3. Dietary intervention ................................................................................................ 33 4.1.4. Dietary assessment.................................................................................................. 35 4.1.5. Anthropometric measurements ............................................................................... 35 4.1.6. Oral glucose tolerance tests .................................................................................... 36 4.1.7. Biochemical measurements ..................................................................................... 36 4.1.8. Dietary biomarkers ................................................................................................. 37 9 4.2. ARI feasibility study ...................................................................................................... 37 4.2.1. Participants and study design ................................................................................. 37 4.2.2. Study diet ................................................................................................................. 38 4.2.3. Physical activity ...................................................................................................... 39 4.2.4. Compliance and acceptability ................................................................................. 39 4.2.5. Anthropometric and biochemical measurements .................................................... 40 4.3. Statistical analysis.......................................................................................................... 40 4.4. Approvals....................................................................................................................... 42 5. Results and discussion .......................................................................................................... 43 5.1. The SYSDIET intervention ........................................................................................... 43 5.1.1. Baseline characteristics of the participants ............................................................ 43 5.1.2. Changes observed during the intervention ............................................................. 44 5.2. Plasma alkylresorcinols as dietary biomarkers in healthy Nordic diet (Paper I) ........... 46 5.2.1. Changes in plasma alkylresorcinols concentrations during the intervention ........ 46 5.2.2. Determinants of plasma alkylresorcinols concentration ........................................ 48 5.2.3. Six to twelve week reproducibility of plasma alkylresorcinols ............................... 49 5.2.4. Correlations between alkylresorcinols and nutrient intake .................................... 51 5.3. Plasma alkylresorcinols and cardiovascular risk factors (Papers II-III) ........................ 51 5.3.1. Glucose and insulin parameters ............................................................................. 51 5.3.2. Blood lipids ............................................................................................................. 53 5.4. ARI feasibility study (Paper IV) .................................................................................... 55 5.4.1. Baseline ................................................................................................................... 55 5.4.2. Rye and dietary fiber consumption ......................................................................... 56 5.4.3. Compliance ............................................................................................................. 57 5.4.4. Acceptability ........................................................................................................... 58 6. Strengths and limitations ...................................................................................................... 60 7. Conclusion ............................................................................................................................ 61 8. Future perspectives ............................................................................................................... 62 9. Acknowledgements .............................................................................................................. 63 10. References .......................................................................................................................... 64 Papers I – IV Appendix 10 List of figures Figure 1. Yearly supply of barley, rye and oats for human consumption in the Nordic countries, kg per capita 1961-2011 .......................................................................................... 20 Figure 2. Study design of the SYSDIET intervention ............................................................. 32 Figure 3. Outline of the study design and main outcome measurements at each visit ............ 38 Figure 4. Flow chart of the SYSDIET study ........................................................................... 43 Figure 5. Total plasma alkylresorcinols concentrations for participants in healthy Nordic diet (ND) and control groups at week 0, 12 and 18/24. Values are medians and bars represent the interquartile range (IQR) .......................................................................................................... 47 Figure 6. Rye consumption and total dietary fiber intake for rye group (RG) and control group (CG) at week 0, 13 and 26 of the intervention. .............................................................. 57 Figure 7. Total plasma alkylresorcinols (AR) concentration and AR C17:0/C21:0 ratio for rye group (RG) and control group (CG) at week 0, 13 and 26 of the intervention ........................ 58 Figure 8. The rye group´s acceptability of the rye products (n = 12) assessed by three questions at week 13 (13 w) and week 26 (26 w): a) I have experienced digestive discomfort after I started eating rye products b) I like food products that contain rye and c) I would prefer products that contain rye more than other products ................................................................. 59 List of tables Table 1. Different definitions of a Healthy Nordic Diet .......................................................... 18 Table 2. Dietary fiber content and composition of whole grain wheat, whole grain rye, dehulled oats, and naked barley ............................................................................................... 21 Table 3. Alkylresorcinols (AR) content range and AR ratio C17:0/C21:0 reported in commonly consumed cereal grains .......................................................................................... 24 Table 4. Advised nutrient composition and suggested food items of for the study diets ........ 34 Table 5. Whole grain rye, energy, fiber, added sugar and salt content for the rye products provided for the rye group in the study .................................................................................... 38 Table 6. Baseline characteristics of the participants in the healthy Nordic diet (ND) group and the control group who completed the SYSDIET study and had valid blood samples for alkylresorcinols analyses at week 0 and 18/24 ......................................................................... 44 Table 7. Total dietary fiber intake in healthy Nordic diet (ND) and control groups during the intervention............................................................................................................................... 45 Table 8. Total plasma alkylresorcinols (AR) concentration and AR C17:0/C21:0 ratio for the healthy Nordic diet group (ND) and control group at baseline and after the 18/24 week intervention............................................................................................................................... 46 11 Table 9. Regression coefficients for factors affecting plasma total alkylresorcinols (AR) at week 18/24 (log transformed) .................................................................................................. 48 Table 10. Baseline dietary intake of fiber, plasma total alkylresorcinols (AR) and AR C17:0/C21:0 ratio at different study centers ............................................................................ 49 Table 11. Estimated intra-class correlation coefficients (ICC, based on log-transformed values) between week 12 and 18/24 for all participants combined, healthy Nordic diet (ND) group and control groups .......................................................................................................... 50 Table 12. Correlations for plasma total alkylresorcinols and intakes of various nutrients abundant in whole grain for both healthy Nordic diet (ND) and control group ....................... 51 Table 13. Regression models for fasting insulin and insulin sensitivity indices at 18/24 weeks (log transformed values)........................................................................................................... 52 Table 14. Regression models for concentrations of serum LDL, HDL, non-HDL cholesterol, triglycerides and apolipoprotein B and the cholesterol LDL-C/HDL-C ratio at 18/24 weeks. 54 Table 15. Baseline characteristics of the participants in the ARI feasibility study ................. 56 Table 16. Energy and macronutrient intake for both groups at week 0, 13 and 26 ................. 57 12 Abbreviations AACC AOAC ARI Apo A1 Apo B AR BMI CV CVD DI EFSA ENL EPIC FFQ GC-MS GluAUC0-30 HDL (HMG)-CoA ICC IDF IGI IGI 120 IL-Ra 1 InsAUC0-30 IQR ISI LDL ND OGTT SD SYSDIET American Association of Cereal Chemists Association of Analytical Chemists A rye intervention Apolipoprotein A1 Apolipoprotein B Alkylresorcinols Body mass index Coefficient of variation Cardiovascular diseases Disposition index European Food Safety Authority Enterolactone European Prospective Investigation into Cancer and Nutrition Food frequency questionnaires Gas chromatography–mass spectrometry 30 minutes area under the curve for glucose High density lipoprotein Hepatic hydroxymethylglutarate Intra-class correlation International Diabetes Federation Insulinogenic index Insulinogenic index at 120 minutes Interleukin-1 Ra 30 minutes area under the curve for insulin Interquartile range Insulin sensitivity index Low density lipoprotein healthy Nordic diet Oral glucose tolerance test Standard deviation Systems biology in controlled dietary interventions and cohort studies 13 List of original papers This dissertation is based on the following papers, which will be referred to in the text by their respective Roman numerals: I Title: Plasma alkylresorcinols reflect important whole-grain components of a healthy Nordic diet Authors: OK Magnusdottir,R Landberg, I Gunnarsdottir, L Cloetens, B Åkesson, G Önning, SE Jonsdottir, F Rosqvist, U Schwab, KH Herzig, MJ Savolainen, L Brader, K Hermansen, M Kolehmainen, K Poutanen, M Uusitupa, I Thorsdottir, U Risérus Status: Journal of Nutrition 2013; 143: 1383-1390. II Title: Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole grain rye intake, is associated to insulin sensitivity - a randomized study Authors: OK Magnusdottir,R Landberg, I Gunnarsdottir, L Cloetens, B Åkesson, M Landin-Olsson, F Rosqvist, D Iggman, U Schwab, KH Herzig, MJ Savolainen, L Brader, K Hermansen, M Kolehmainen, K Poutanen, M Uusitupa, I Thorsdottir, U Risérus Status: European Journal of Clinical Nutrition 2014; 68: 453-458. III Title: Whole grain rye intake, reflected by a biomarker, is associated with favorable blood lipid outcomes in subjects with the metabolic syndrome – a randomized study Authors: OK Magnusdottir,R Landberg, I Gunnarsdottir, L Cloetens, B Åkesson, F Rosqvist, U Schwab, KH Herzig, J Hukkanen, MJ Savolainen, L Brader, K Hermansen, M Kolehmainen, K Poutanen, M Uusitupa, U Risérus, I Thorsdottir Status: PLOS ONE, in press IV Title: Increasing rye consumption among men with type 2 diabetes: a feasibility study Authors: : OK Magnusdottir, R Landberg, I Gunnarsdottir, KT Magnusson, G Hallmans, RD Hansen, C Kyrø, A Olsen, A Tjønneland, I Thorsdottir, L Steingrimsdottir Status: manuscript 14 1. Introduction The principal aim of this dissertation was to increase knowledge on the use of plasma alkylresorcinols (AR) concentrations as dietary biomarkers for whole grain wheat and rye intake. This PhD project was in close collaboration with two extensive research projects, SYSDIET - Systems biology in controlled dietary interventions and cohort studies and HELGA: Nordic Health — Whole Grain Food, both of which were funded as Nordic Centres of Excellence in Food, Nutrition and Health by NordForsk. It is well accepted that Western population would benefit from higher consumption of whole grains as whole grain and cereal fiber consumption has almost consistently been inversely associated with the risk of developing chronic diseases in epidemiological studies [1-3]. Findings in intervention studies have, however, been inconsistent [4, 5]. Nutritional science relies on accurate dietary information to assess compliance to dietary interventions and to investigate diet-disease associations. However, dietary assessment instruments rely most often on self-reporting, which is prone to measurement errors [6-10]. Collecting data on whole grain intake can be especially challenging because of the broad selection of foods contributing to these intakes and the difficulty that consumers and researchers have in assessing the whole grain content of foods consumed [11-13]. Hence, the reason why results from intervention studies are often inconclusive may relate to absence of accurate and objective measures of whole grain intake. The use of dietary biomarkers could be one way to improve the validity of dietary assessment in combination with, or as an alternative for, self-reports or as a tool for secondary analysis, where non-compliant individuals can be excluded on the basis of biomarker measurements [14-17]. AR have been proposed as biomarkers for whole grain and bran intake of wheat and rye [12, 18-20] and for cereal fiber intake in populations with a high intake of whole grain wheat and rye [21]. Controlled whole grain and bran intervention studies show that plasma AR concentration increases proportionally with AR and whole grain intake [18, 22-26] and can be used as a biomarker of compliance to a whole grain diet [20, 27]. Additionally, the ratio of the AR homologues C17:0 and C21:0 differs between cereals species [28] and the ratio in plasma samples have been suggested to reflect proportion of whole grains consumed [25]. Currently, AR appear to be highly promising biomarkers of whole grain wheat and rye intake [12, 29] although there are still many factors that are poorly understood. Analysis of plasma AR as a biomarker in the healthy Nordic diet, which is rich in whole grain wheat and rye, adds valuable knowledge of plasma AR as biomarkers and their use in future studies of the healthy Nordic diet. Additionally, the analyses of the association between plasma AR and clinical biomarkers of health and the usage of plasma AR as a measure of compliance in intervention studies adds knowledge and weight to further usage of plasma AR as biomarkers of whole grain wheat and rye intake in nutritional science. 15 2. Background 2.1. What is the healthy Nordic diet? It is scientifically substantiated that dietary quality has an important impact on the risk of chronic diseases [30]. In nutritional science, traditional analysis has focused on evaluating the effect of a single food or nutrient on the risk of developing disease. This approach has been successful in the past and has helped to eliminate a number of deficiency diseases in many parts of the world. However, it appears to be less effective in addressing the most important present-day nutrition-related health problems. Globally, there have been rapid changes in the food environment where traditional, regionally available foods have been and are still being replaced by modern industrial and processed foods [31, 32]. This dietary pattern, often referred to as Westernized dietary pattern, is typically dense in energy and characterized by high intakes of fat and saturated fatty acids, processed and red meats, refined cereals and food products with added refined sugars, fat and salt [31-33]. These swift changes in food and eating patterns, together with over-consumption, has been suggested to be one of the main determinants for many of the chronic diseases seen today [31-34]. Therefore, in recent years, the focus of nutritional research has moved toward evaluating the overall impact of diets and patterns of food consumption on health outcomes, instead of single nutrients or food items. Such an approach has resulted in a significant amount of new and original research [33, 35]. The focus of current nutritional recommendations has also changed with more emphasis on food-based dietary guidelines where advice on dietary intake is conveyed at the food level [33]. The best known example of a prudent whole diet is the Mediterranean diet, representing the diet traditionally consumed in Southern Europe. The Mediterranean diet has been extensively studied and it has been associated with improved health and prevention of chronic diseases in many studies [36-40]. The acceptance of the diet has, however, not been easy in other parts of the Western world [41]. Instead of imposing one universal diet it would be more sensible to tailor practical recommendations for healthy eating to regional circumstances. Most countries or regions do have their own traditional and regionally appropriate foods that could be as health enhancing as the ingredients of the Mediterranean diet. Such regional tailoring of recommendations are important for the environment and also for health, as it would be easier to adhere to such diets due to the availability of food items, taste preferences, cultural differences and familiarity of the foods [31, 41-43]. The dietary habits in the Nordic countries have several common features and trends in food consumption tend to be similar [44-46]. Although the modern Nordic diet resembles the conventional Western diet, and is therefore not considered to be positively related to health outcomes [45], it also includes a range of foods with expected health-promoting effects. Plant foods such as berries, apples, pears, cabbages, root vegetables and potatoes thrive well in the Nordic climate, as well as oats, rye and barley. Also, the long coastlines of the Nordic countries provide rich sources of fish [31, 46, 47]. Several prospective epidemiological studies have studied Nordic dietary patterns, including traditional Nordic food with expected 16 health-promoting effects, and concluded that it is associated with important health benefits, lower total mortality [48] and incidence of colorectal cancer [47]. The health effects of the healthy Nordic diet have been studied in a number of interventions studies and results suggest beneficial effects on cardiovascular risk factors and mortality [48-52]. Although definitions of the healthy Nordic diet are not completely congruent it is generally defined by healthy foods that can be produced in the Nordic countries and are traditionally consumed (Table 1). Common features include berries, vegetables (mainly cruciferous vegetables), whole grain products (mainly oats, barley and rye), fish and game and free-range or pasture fed animals [31, 51, 53, 54]. When compared to a conventional Western diet the healthy Nordic diet has generally more dietary fibers, lower salt content and higher quality of dietary fat [51, 53, 55, 56]. 17 Table 1. Different definitions of a Healthy Nordic Diet Food choice based on: Bere and Brug (2009) [31] Ability to be produced locally over large areas in the Nordic countries without usage of external energy, e.g. greenhouses, New Nordic diet – OPUS project [54] SYSDIET study [51] Nutritional value Health Traditional Nordic food items Agricultural traditions Gastronomic potential, palatability Seasonal variations Nordic identity and cultural heritage Diet and nutrition profile based on 2004 Nordic Nutrition Recommendation[57] Habitual use of food items, foods commonly available in local super markets Sustainability, locally grown foods and organic food production Tradition as a food source Possessing better potential for health-enhancing effects than similar foods within the same food group Ability to be eaten as foods, not only in small amounts as dietary supplements (e.g. spices) Key food items NORDIET study [49, 53] Native berries Cabbage Oat/barley/rye Rapeseed oil Native fish and other seafood Wild and pasture-fed land-based animals Diet and nutrition profile based on 2004 Nordic nutrition recommendation[57] and 80% of the diet should be ”Nordic” Fruits, berries, vegetables and root vegetables Potatoes Whole grains in general and specific whole grain cereals based on oat, barley and rye Rapeseed oil, vegetable fat spread Nuts, mainly almonds Fish, especially fatty fish Poultry, lamb, game, small portions of red meat and sausages Low fat dairy Fundamental guidelines: - More calories from plant foods and fewer from meat - More foods from the sea and lakes - More foods from the wild countryside Fruits and vegetables, including berries, cabbages, root vegetables and legumes Potatoes Fresh herbs Plants and mushrooms from the wild countryside Whole grains Nuts Fish and shellfish Seaweed Free-range livestock, including pigs and poultry and game Fruits and vegetables, including berries Whole grain, of which at least 50% rye, barley and oat Rapeseed and/or sunflower and/or soybean oil Fish, fatty and lean White meat, poultry, game Low fat dairy 18 2.2. Whole grain Since the beginning of agriculture, cereals have been a part of the human diet. Cereals were traditionally consumed as whole grain and it is only within the last hundred years that people have consumed refined grain products [58]. The American Association of Cereal Chemists (AACC) define whole grain as follows: “Whole grains shall consist of the intact, ground, cracked or flaked caryopsis, whose principal anatomical components – the starchy endosperm, germ and bran – are present in the same relative proportions as they exist in the intact caryopsis” [59]. This definition has also been adopted by the U.S. Food and Drug Administration [60]. The European HEALTHGRAIN Forum suggested that small losses of components, i.e. less than 2% of the grain and/or 10% of the bran, that occur through processing methods consistent with safety and quality, should be allowed [61]. The main cereals consumed as whole grain are wheat, corn, rice, oats, barley and rye. Millets, sorghum, teff, triticale, canary seed, Job’s tears, fonio, wild rice are also included in the definition of whole grain, and some definitions also include the pseudocereals amaranth, quinoa and buckwheat [60, 61]. Although a working definition exists for whole grains and relevant whole grain ingredients, no universal agreement on the exact definition of whole grain products exists, with questions remain regarding the processing of the grains, range of cereals to be included and proportion of whole grain in the food. Typically whole grain food contains a certain proportion of intact, flaked or broken kernels, coarsely ground kernels or flour of whole grain but different proportions have been used across studies. Many of the studies on whole grain and disease have used the definition of whole grain products by Jacobs et al [62], which proposes that ≥ 25% of product weight must be whole grain, but other definitions have been used as well [60, 63-65]. In 2013, the AACC International Board of Directors approved the Whole Grain Working Group’s characterization of a whole grain product: “A whole grain food must contain 8 grams or more of whole grain per 30 grams of product” [66] and this definition was also proposed at a whole grain expert roundtable discussion in 2014 [13]. The Keyhole Symbol is a public symbol on packages for food products used in the Nordic countries, which meet certain criteria on nutritional value. Thus, food products with the keyhole symbol are healthier than other products in the same food category that do not meet the criteria. To be allowed to use the Keyhole Symbol soft bread should contain ≥25% whole grain, and crisp bread, pasta, breakfast cereals, and porridge should contain at least ≥50% whole grain to fulfill the requirements. For flours, flakes, and kernels the content of whole grain should be 100% [67, 68]. In addition to different definitions of whole grain and whole grain products in previous studies, changes in dietary fiber characteristics during processing and preparation of whole grain products, such as molecular weight distribution and extractability, are considerable. This may impact the physiological effects of whole grain consumption, the chemical and nutritional composition as well as physical structures and functional properties in various ways [69-71]. 19 2.2.1. Whole grain cereals in the Nordic diet Food supply quantity kg/capita/year The most commonly consumed cereals in the Scandinavian countries are wheat, rye and oats, while barley has been used to a lesser extent [31, 72]. Oats, barley and rye do, however, grow better in colder climates than e.g. wheat and are produced in many areas in the Nordic countries [31]. Still, the proportion of wheat as a total of cereal consumption has increased in the last decades and today, around 80% of all grains consumed in the Nordic countries are wheat [31, 72]. Wheat is usually consumed as white flour, while oats, barley and rye are most often consumed as whole grains which make them an important constituent of the healthy Nordic diet. Among the Nordic countries, Sweden is the only country that has data on whole grain intake in their most recent national dietary survey among adults. The data showed an average intake of 42 g/day of whole grain for adults age 18-80 years old [73]. In Denmark, data from the 2011-2013 Danish National Survey of Dietary habits and Physical activity were used to estimate the whole grain intake and the results showed an average intake of 55 g/day [74]. Rye is the second most commonly consumed grain, behind wheat, in the Nordic countries [72, 75]. In Iceland, rye and oatmeal were the main types of whole grain in the diet during the first half of the 20th century and in 1930 the average rye consumption was 152 g/day [76]. However, according to food supply data, rye consumption has decreased dramatically in the last 50 years in Iceland, as well as in other countries (Figure 1) [72, 77]. In 2011, the average rye consumption in Iceland was 8 g/day, the lowest among the Nordic countries while the average was 47 g/day in Denmark, 45 g/day in Finland, 32 g/day in Sweden and 18 g/day in Norway [72]. The change in oat consumption has not been as dramatic. In Iceland the average oat consumption was 13 g/day in 2011 compared to 25 g/day in 1961. In 2011 the average oat consumption was 28 g/day in Denmark, 18 g/day in Finland, 13 g/day in Norway and negligible in Sweden [72]. 20 18 16 14 12 10 8 6 4 2 0 Barley and products Rye and products Oats 1961 1970 1980 1990 2000 2011 Figure 1. Yearly supply of barley, rye and oats for human consumption in the Nordic countries, kg per capita 1961-2011 [72] 20 In line with these results a subgroup analysis of the participants in the HELGA study, including data from Denmark, Sweden and Norway, showed a clear national difference in the sources of whole grain intake [75]. In the populations studied whole grain rye constituted more than 70% of the total whole grain intake in Denmark, more than 50% in Sweden but only 20% in Norway. Whole grain oats made up 6 to 19% of the total whole grain intake in the three countries [75]. Bread is by far the most common rye product, followed by rye flakes, which are frequently used for baking, porridge and breakfast cereals. Oats are not as suitable for making bread due to the lack of gluten and are therefore often prepared as a porridge, flakes or breakfast cereals made from crushed or rolled oats. Oats are however utilized in a variety of baked items as oat flour or oatmeal, and breads can be made from a mixture of oatmeal and wheat flour [78]. Barley has primarily been used in animal feed as well as for production of malt and only to a limited extent for human consumption [31, 72]. Rye, oats and barley are rich sources of vitamins, minerals and bioactive compounds [79-81]. Rye is particularly high in dietary fiber with 20% of dry matter consisting of fiber, compared to 10% in oats and 15% in barley (Table 2) [69, 82]. The main dietary fiber components of rye are arabinoxylan, fructan, β-glucan and cellulose [71, 83] while β-glucans are more abundant in oats and barley (Table 2) [71]. Thus, the physiological effects of dietary fiber differ depending on which cereal has been consumed and thereby the nutritional and potential health effects are also influenced. Table 2. Dietary fiber content and composition of whole grain wheat, whole grain rye, dehulled oats, and naked barley [71] Component Wheat Total dietary fiber, % 13.5 Arabinoxylan, % 5.6 Cellulose, % 2.5 β-Glucan, % 0.8 Fructan, % 1.3 Klason lignin, % 0.8 Data are presented as percentage of dry matter Rye Oats Barley 19.9 8.9 2.9 1.5 4.1 1.1 10.2 2.0 1.3 5.0 0.2 1.4 15.2 5.2 1.9 4.6 1.6 0.7 21 2.2.2. Whole grain intake estimation methods and their limitations Nutritional science relies on accurate dietary information to assess compliance to dietary interventions and to investigate diet-disease associations. Dietary assessment methods include food frequency questionnaires (FFQs), diet histories, 24-hour recalls and weighed or estimated food records [84]. However, most dietary assessment instruments rely on selfreporting which is prone to both systematic and random errors. These errors are due to e.g. the use of food tables, variations in frequency of food consumption or portion size, misreporting or failure to report usual diet, as well as changes in typical habits whilst taking part in a study [6-10]. In addition, dietary assessment methods are often not designed to assess whole grain intake, with e.g. limited number of questions in FFQs which can be used to rank subjects according to their whole grain intake and that may result in an underestimation and inaccurate measures of intake [11]. The increase in the availability of whole grain products during the past decade has made accurate assessment of whole grain intake increasingly challenging. The existing whole grain food composition data is limited and up-to-date nutrient databases are needed to better assess whole grain intake and capture the changing whole grain food supply [13]. The lack of a uniform definition of whole grain products, the increasing breadth of different foods and processing methods this food category covers, and the difficulty that consumers and researchers have in accurately knowing the whole grain content of foods consumed adds to the challenge [11-13]. One way to avoid some of the obstacles associated with traditional dietary assessment instruments is to use dietary biomarkers as an alternative, or as a complement, to self-reports. Biomarkers have the potential to bypass the inherent errors associated with traditional diet assessment tools, be less subject to intra-individual variation, and can allow blind assessment of compliance in dietary intervention trials where a true placebo is not feasible. It also allows for a specific association of certain parts of a complex diet with health variables [14-17]. However, factors such as genetic variability, lifestyle/physiologic factors (e.g. smoking), dietary factors, biological sample and analytical methodology [29, 85] can skew the biomarker measures. As a result, it is imperative to assess the validity, reproducibility of the biomarkers, as well as the ability to detect changes over time and robustness across diverse populations. The use of dietary biomarkers, especially for whole grain intake, will be discussed in the following sections. 2.3. Dietary biomarkers in nutritional science Dietary biomarkers, or nutritional biomarkers, are compounds related to a certain food or nutrient that are measurable in biological tissues and fluids (e.g. plasma, urine and/or adipose tissue), and thus can objectively reflect intake of specific dietary components [84]. There are a number of compounds that have been suggested and used as dietary biomarkers of food and/or nutrients and their purpose may be different depending on the types of study. Firstly, dietary biomarkers can be used for validation of traditional dietary assessment instruments 22 [27, 86-89]. Dietary biomarkers can also be used in combination with, or as an alternative for, self-reports to estimate intake in studies of association between diet and disease risk, or as independent tools to confirm such associations derived using traditional dietary instruments [90-95]. Additionally, in randomized controlled trials, dietary biomarkers can be used to measure compliance or as a tool for secondary analysis, where non-compliant individuals can be excluded on the basis of the biomarker measurements [95-99]. Hence, dietary biomarkers can provide an important link between dietary exposure and disease, especially in situations where traditional dietary assessment methods for different reasons provide poor intake estimates. 2.3.1. Dietary biomarkers of whole grain intake There is currently no generally accepted biomarker of overall whole grain intake. The mammalian lignin enterolactone (ENL) has been evaluated as one such biomarker as whole grain, especially rye, is rich in precursors of ENL and the concentration of ENL in plasma, serum and urine increases with whole grain intake [100-102]. However, there are several other sources of ENL precursors in the diet and the formation of ENL is dependent on the gut flora as well as other non-dietary determinants [25, 102-104]. The ENL concentration may therefore not be a suitable biomarker for whole grain intake. About 10-15 years ago, AR were initially suggested to be a potential biomarker of whole grain wheat and rye intake [20, 105, 106]. A method for the analysis of AR in plasma was developed and initial biomarker evaluation studies were conducted [25, 107-111]. Plasma AR as a biomarker for whole grain wheat and rye intake will be discussed in sections 2.3.2. and 2.3.3. To date, no other biomarker for whole grains, or specific whole grain type, has been specifically identified and validated. Studies on metabolites have discovered potential markers of certain whole grain food intakes using metabolomics, such as metabolites from benzoxazinoids, as a marker for rye intake [112] but further studies need to be developed. 2.3.2. Alkylresorcinols as biomarker of whole grain wheat and rye intake AR belong to a large group of phenolic lipids. Structurally they consist of a 1,3dihydroxybenzene ring (resorcinol), which is alkylated at position 5 by a saturated, oddnumbered hydrocarbon side-chain consisting of 17 to 25 carbon atoms (C13:0 - C27:0) [105]. AR are abundant in the outer layers of rye and wheat, found in low amounts in barley, maize and peas, in trace amounts in refined white flour and absent in most other foods consumed in the human diet [105]. The ratio of different homologues differs between different cereals and can be used to differentiate between AR containing cereals. Barley, for example, has a larger portion of C25:0 compared to other cereals, though overall it has much lower total AR concentrations (Table 3) [105, 113]. 23 Table 3. Alkylresorcinols (AR) content range and AR ratio C17:0/C21:0 reported in commonly consumed cereal grains Cereal Rye (Secale cereal) Common wheat (Triticum aestivum) Durum wheat (Triticum durum) Einkorn wheat (Triticum monococcum) Emmer wheat (Triticum dicoccon) Spelt wheat (Triticum spelta) Barley (Hordeum vulgare) AR range, µg/g dry matter 568-1231 AR C17:0/C21:0 ratio References 0.80-1.30 [28, 79, 114-116] 300-943 0.09-0.24 194-687 0.01-0.02 [28, 115, 117, 118] [105, 117] 545-654 0.04 [105, 117] 531-784 0.05 [105, 117] 490-741 0.09-0.14 [105, 117] 8-210 0.05-0.46 [28, 81, 113] The average daily intake of AR has been estimated, using food supply data or consumption data, and was found to differ within populations and between countries [119]. Food supply data was obtained by converting amounts of milled rye/wheat to per capita amounts and then multiplying this value by the average AR content of that cereal. According to this method, the intake of AR ranged from 12 mg/day in the United Kingdom (UK), where white bread is mostly consumed and 18 mg/day in Sweden, to nearly 40 mg/day in Finland. In the alternative method utilizing food consumption data, individual food records (in Sweden) or household shopping data (in the UK) were used to estimate AR concentrations based on a typical recipe for the food in question. According to this data, the mean intake of AR in Sweden was 23 mg/day and in the UK 12 mg/day. The Swedish data was further analyzed and the intake of AR was found to be higher in males than females and higher in the older age groups [119]. Intakes at the low end of the range may, however, be an underestimate as the small amounts present in white wheat flour were not accounted for. AR are stable during food processing and absorption is around 60% in the human digestive tract [106, 120], though data from pigs and humans suggest that at higher doses the percent absorbed is lower [23, 121]. After absorption, AR are suggested to be transported in lipoproteins (mostly HDL) [25] and may be distributed and stored in some tissues, especially adipose tissue [122]. The absorption half-life of AR is about 6-8 hours [107] and the half-life of AR in plasma is around 5 hours [107]. Thus, strictly defined, plasma AR should be regarded as a short-term biomarker but under stable and frequent intake conditions, it also reflects long-term exposure fairly well [123]. 24 Plasma levels of AR vary widely even at steady state, with mean overnight fasting levels being 38 - 388 nmol/L with a moderate to high rye intake [108-110] and 943 nmol/L with an extremely high (483 g/day) rye intake [24]. Plasma AR hardly ever reach zero, even after avoiding whole grain products for a certain time in trials. This is at least partly due to the storage of dietary AR in adipose tissue and liberation during low or no intake [122]. The total plasma AR concentration and concentrations of five AR homologues (C17:0, C19:0, C21:0, C23:0 and C25:0) was recently evaluated in ten Western European countries using data from the European Prospective Investigation into Cancer and Nutrition (EPIC). A considerable variation was observed across countries, both in plasma concentrations of total AR and concentrations of AR homologues. High concentrations of plasma total AR were found in participants from Scandinavia and Central Europe and lower concentrations in those from the Mediterranean countries. The geometric mean plasma total AR concentrations were between 35 and 41 nmol/L in samples drawn from fasting participants in the Central European and Scandinavian countries and below 23 nmol/L in participants from the Mediterranean countries [124]. Correlations between plasma AR and various measures of their intake (AR intake, whole grain intake, cereal fiber intake) range between 0.25 and 0.58, with better correlations with more detailed dietary intake instruments [18, 19, 26, 109, 125]. Intervention studies focusing on whole grain and bran intake have shown that plasma AR concentration proportionally increases with AR and whole grain intake [18, 22-26]. As a biomarker for whole grain wheat and rye intakes, AR can be used to categorize subjects as low and high consumers of whole grain foods [19, 126] and also as a marker for compliance to a whole grain diet in intervention studies [20, 27]. The reproducibility of plasma AR measurements, evaluating whether a single measurement of a biomarker can reflect average levels over a longer time period, have been reported to be good under intervention conditions (intraclass correlation coefficient ICC = 0.88-0.9), and moderate under free-living conditions (ICC = 0.42-0.48) [19, 22, 127]. These studies further support the proposed use of fasting plasma AR as a biomarker of whole grain intake, especially in populations where whole grain wheat and rye intake is high and relatively consistent. However, further studies are needed to establish cut-off points for determining when a person is complying with a diet or has not been eating a certain amount of whole grain per day. As an alternative to relying on plasma AR, quantification of AR in adipose tissue is possible through adipose tissue biopsy samples [122]. The importance of the pool and factors governing its turnover are, however, still unknown. In addition, the AR metabolites DHPPA and DHBA in plasma and urine have also been suggested to serve as biomarkers for whole grain rye and wheat intake [21, 128-130] although they have been less extensively studied than the intact AR. 25 2.3.3. Alkylresorcinols C17:0/C21:0 homologue ratio as marker of relative rye intake The ratio of C17:0/C21:0 is approximately 0.1 in wheat and 1 in rye, and this ratio has been suggested to be a method for determining if a cereal product contains wheat, rye or a mixture [28]. The C17:0/C21:0 ratio is to some extent reflected in fasting plasma samples and may be used to differentiate sources of cereals consumed [25]. In plasma, the ratio after a rye diet is somewhat lower than 1, usually around 0.6-0.8 after regular whole grain/bran rye consumption [23, 25] and a ratio above 0.15 is usually indicative of some rye intake [12]. In a cross sectional study from Sweden, where rye consumption is appreciable, the C17:0/C21:0 ratio was 0.5 in samples drawn eight hours since last meal [123] and in the EPIC study the ratio was 0.1 in the UK and 0.17 in Italy where the main source was wheat while it was 0.43 in Sweden [124]. The reason for a lower ratio in plasma compared to what is observed in whole grain/bran rye foods is likely due to faster metabolism of the AR homologue C17:0 [107, 131]. However, further insight into absorption and metabolism of the different homologues is needed to clarify to what extent factors other than whole grain rye and wheat affect the C17:0/C21:0 ratio [12]. 2.3.4. Bioactivity of alkylresorcinols A wide variety of bioactivities have been ascribed to AR from in vitro tests. AR have been found to have slight antimutagenic effects and some antioxidant capacity although this is weak compared to known antioxidants such as α-tocopherol [132, 133]. The effective concentrations are high, ranging from 2.5 to 100 mol/L, while the maximum concentration after a single meal containing 190 mg of AR from rye (120 g of rye bran) is 3-4 mol/L [107]. Such high intakes are not likely to be observed at “normal” intakes. Only limited in vivo work has been carried out on possible biological function of AR and oral dosing with pure AR has not been tested in humans [12]. 2.4. Whole grain intake and human health Today, there is a general long-standing view among major governmental, scientific and nonprofit organizations that the Western population would benefit from higher consumption of whole grains as a way to reduce the risk of developing chronic diseases. As a result, whole grain intake, as a part of a healthy diet, has been directly or indirectly promoted in the dietary guidelines in a number of countries as well as by the World Health Organization [33, 134, 135]. Evidence of the health benefits of whole grains have mainly been derived from epidemiological studies which have shown protective associations but not the causal links between whole grain intake and health. Furthermore, most of the prospective cohort studies reported in the literature are from large American cohorts, where the dominant whole grain cereal consumed is wheat. 26 A consistent inverse association between intake of whole grain foods and cardiovascular diseases (CVD) has been reported in these epidemiological studies [1-3]. In a meta-analysis of prospective cohort studies Mellen et al [1] suggested that an average of 2.5 servings/day, compared to 0.2 servings/day, was associated with a 21% lower risk of CVD events in adult populations. This finding was similar to an earlier meta-analysis by Anderson et al [3]. Whole grain and cereal fiber consumption has repeatedly been inversely associated with the risk of type 2 diabetes and insulin sensitivity in observational studies [2, 136-139]. In a systematic review Priebe et al [140] found that 10 out of 11 prospective cohort studies showed that high intakes of whole grains and cereal fiber was associated with a 27-30% and 28-37% decreased risk of developing type 2 diabetes, respectively. Similar results were obtained in the Nurses’ Health Studies I and II that included more than 160 000 women [137]. In a systematic review of 15 observational studies on whole grain consumption and measures of body weight and adiposity, consumption of three servings of whole grain foods per day was associated with 0.63 kg/m2 reduction in body mass index (BMI) and 2.7 cm decrease in waist circumference [141]. Whole grain foods have also been associated with a reduced risk of some cancers and evidence from prospective studies show the associations are strongest for cancer of the gastrointestinal tract [142, 143]. The underlying mechanisms for the protective effect of whole grain foods on development of chronic diseases are thought to be diverse but are not fully understood [144, 145]. Results from epidemiological studies still remain to be confirmed by randomized controlled trials as only few intervention studies on whole grain and disease have been published. Most of these studies have been short-term with a small numbers of subjects as they are difficult to perform. Studies have to rely on compliance over long periods of time if the primary endpoints are the development of chronic diseases, such as CVD and type 2 diabetes. Reports on the effects of various sources of whole grain food on risk markers or intermediate endpoints of CVD and diabetes and results have also been inconsistent [4, 5]. 2.4.1. Whole grain effects on glucose and insulin Potential mechanisms for the beneficial effects of whole grains on the risk of type 2 diabetes are diverse since whole grains are rich in nutrients and phytochemicals. Given that insulin resistance increases the risk of type 2 diabetes [146-148], insulin sensitivity may be one important mechanism through which whole grain consumption confers protection. A whole grain, high fiber diet has been suggested to enhance insulin receptor sensitivity through chronic lowering of the overall dietary glycemic index and related insulin secretion [149-155]. However, findings from intervention studies exploring the effect of whole grain consumption on glucose and insulin concentrations or insulin resistance and sensitivity have been inconsistent [5, 156-160]. Tighe et al [156] found no effects of three whole grain food portions a day on glucose and insulin concentrations or markers of insulin resistance and sensitivity after 12 weeks and Brownlee et al [5] reported the same conclusion in the WHOLEheart Study. Pereira et al [149] did, however, find that whole grain consumption had an effect on fasting plasma insulin concentration and insulin sensitivity and 27 Landberg et al [159] showed positive effects of higher rye consumption for 6 weeks on fasting plasma insulin, using total plasma AR as a measure of compliance. Several intervention studies have also shown diminished levels of postprandial insulin following a rye bread meal [161-163]. The lower postprandial insulin response after rye bread consumption, as compared to wheat bread, may be due to the difference in structure rather than fiber content, as consumption of low fiber rye bread have been shown to have similar effects as high fiber rye bread consumption [164, 165]. Moreover, an improved early insulin response has been shown after rye bread interventions [164, 165]. The inconsistent results in intervention studies might be due to differences in the type, amount, structure, processing and preparation of whole grain products and carbohydrates consumed [71, 166]. Lack of compliance and/or objective measurement of whole grain intake might also be an issue but good adherence has, however, been indicated in most studies using weighed food records. Other suspected mechanisms that might explain the effect of whole grain on insulin sensitivity include slower gastric emptying [167], metabolic effects of certain micronutrients, such as magnesium, selenium and vitamin E [154, 168-172], as well as through short-chain fatty acid production by intestinal bacteria leading to enhanced hepatic glucose oxidation and insulin clearance [173-176]. 2.4.2. Whole grain effects on blood lipids A number of intervention studies have found that whole grains lower serum total and LDLcholesterol [158, 177-179]. In respect to hypocholesterolemic effects oats are probably the most studied cereal [180-183] and it is now well established that minimally processed βglucans in oats and barley reduce blood cholesterol concentration in both normocholesterolemic and hypercholesterolemic subjects [184, 185]. The European Food Safety Authority (EFSA) has approved two health claims on the subject: β-glucans contribute to the maintenance of normal blood cholesterol levels [186] and oat β-glucan has been shown to lower/reduce blood cholesterol. Blood cholesterol lowering may reduce the risk of (coronary) heart disease [187]. Wheat fiber, however, does not appear to affect blood cholesterol [188] and therefore it is often used as a control in experimental studies. The mechanisms that have been proposed to cause the reduction in cholesterol levels revolve around the fiber and phytochemical content of whole grains, particularly water-soluble viscose fiber [144, 189]. Studies have shown that the cholesterol lowering effect of β-glucans is dependent on the molecular weight and subsequent viscosity [70]. Few intervention studies have studied the effects of rye on blood lipids in humans. Rye is especially high in dietary fiber, ranging between 18-22%, and rye contains substantial amounts of extractable beta-glucans (1.5% of dry matter vs. 5.0% and 4.6% in oats and barley, respectively, see Table 2) and arabinoxylans, which could contribute to cholesterol reduction through similar mechanism as β-glucans from oats and barley [82, 83]. Animal studies have shown cholesterol lowering effects of rye consumption [190-192] and that rye could affect blood lipids by increasing the hepatic hydroxymethylglutarate (HMG)-CoA reductase activity, which regulates cholesterol synthesis in the liver [190] or by increasing the total concentration of bile acids in the bile [191, 193]. Bile acid excretion stimulates bile acid 28 synthesis from serum cholesterol, so reducing cholesterol levels in the blood. Studies in humans have, however, been conflicting. One study reported that rye bread consumption decreased total and LDL cholesterol in men with moderately elevated serum cholesterol [177] and another showed lower plasma free-cholesterol, total cholesterol, triglycerides and phospholipids after a high fiber rye diet compared to a low fiber wheat diet [178]. However, Moazzami et al showed an unfavorable shift in serum cholesterol levels after rye bread intake in postmenopausal women [194]. 29 3. Aims The aim of this dissertation was to increase knowledge on the use of plasma AR concentrations as dietary biomarkers for whole grain wheat and rye intake. The specific aims were to evaluate: 1. Whether plasma AR could be used as biomarkers for whole grain intake in the healthy Nordic diet. 2. The association between plasma AR and glucose metabolism in a population with metabolic syndrome. 3. The association between plasma AR and blood lipid concentrations in a population with metabolic syndrome. 4. The feasibility of conducting an intervention with high consumption of whole grain rye in a population of men with habitually low whole grain and fiber intake. 30 4. Methods The thesis is based on two extensive research projects, SYSDIET - Systems biology in controlled dietary interventions and cohort studies and HELGA: Nordic Health — Whole Grain Food, both of which were funded as Nordic Centers of Excellence in Food, Nutrition and Health by NordForsk. Papers I-III are part of the SYSDIET study, a randomized controlled multicenter intervention study performed in six intervention centers (Kuopio and Oulu, Finland, Lund and Uppsala, Sweden, Aarhus, Denmark, and Reykjavik, Iceland). The main aim of the SYSDIET intervention was to clarify whether a Nordic alternative for healthy food pattern in a weightstable condition would have beneficial effects on insulin resistance, glucose tolerance, serum lipids and lipoproteins, and inflammatory markers in people with metabolic syndrome. The main results from the SYSDIET intervention, i.e. the adherence to the diet and the effects on insulin resistance, glucose tolerance, serum lipids and lipoproteins and inflammatory markers, have been published previously [51]. Additionally 7 papers, including 3 which are a part of this thesis, have been published to date from the study [50, 55, 95, 195-197]. Paper IV of the thesis is a part of ARI (A rye intervention), a feasibility study that was conducted as a follow up of HELGA: Nordic Health – Whole grain Food, examining the acceptability and adherence to increased rye consumption in a group of Icelandic men with type 2 diabetes. 4.1. The SYSDIET intervention 4.1.1. Study design The participants were screened for inclusion and exclusion criteria followed by a one month run-in period during which all participants followed their habitual diet. After the run-in period, participants fulfilling the inclusion criteria were randomized into a healthy Nordic diet (ND) group or a control diet group for the next 18 to 24 weeks. In the original protocol, the study length was 24 weeks but after a consultation with the NordForsk Panel and Scientific committee members the study period was shortened to 18 weeks in four centers (Aarhus, Uppsala, Reykjavik and Oulu). The rationale a for shorter intervention period was that an 18 week intervention was considered to give the same information on the main results as that obtained from a 24 week trial, total costs were lowered, and recruitment of study subjects would be easier for a shorter trial. In Lund and Kuopio, where the intervention was started earlier, the original study design, i.e. 24 week intervention, was followed. In Kuopio and Lund the intervention was carried out from October 2009 to June 2010, in Aarhus from January 2010 to September 2010, in Oulu from December 2009 to October 2010, in Reykjavik from March 2010 to October 2010 and in Uppsala from June 2010 to November 2010. 31 The study participants were scheduled to visit the study clinic at 2-4 week intervals with the major visits at week 0, 12 and either 18 or 24 weeks (Figure 2). The staff in all study centers was trained to perform the measurements in a similar way and a common quality management protocol was made familiar to all staff members at each study site. The study participants were advised to keep their weight and physical activity constant during the intervention and not to change their smoking and drinking habits or drug treatment during the study. Figure 2. Study design of the SYSDIET intervention 4.1.2. Study population Participants were recruited through advertising in newspapers, at workplaces and from previous clinical or epidemiological trials. Altogether 309 individuals were contacted and completed the screening visits. From these individuals, 200 started the intervention, 104 in the ND group and 96 in the control group. There were 8 dropouts in the ND group and 26 in the control group during the study period. Thus, 96 individuals in the ND group and 70 in the control group, in total 166 individuals, completed the trial. The inclusion criteria were age 30 - 65 years, BMI 27 - 38 kg/m2 and two other criteria for metabolic syndrome as defined by the International Diabetes Foundation (IDF) [198], except a diagnosis of type 2 diabetes, i.e. fasting plasma glucose <7.0 mmol/l (and a 2-h glucose value <11.1 mmol/l at baseline detected by an oral glucose tolerance test). Antihypertensive and lipid lowering medication was allowed but without changes during the trial. The main exclusion criteria included any chronic disease or condition which could hamper the adherence to the dietary intervention protocol, poor compliance, chronic liver, thyroid and kidney diseases, alcohol abuse (>40 g/d), diabetes, fasting triglycerides >3.0 mmol/l, total cholesterol >6.5 mmol/l and blood pressure >160/100 mmHg. A few study participants with triglycerides between 3 and 4 mmol/l and with BMI between 38 and 40 32 kg/m2 were, however, included due to the fact that they were committed to the trial. Other exclusion criteria were a recent myocardial infarction (<6 months), corticosteroid therapy, psychiatric disorders needing drug treatment, cancer under treatment, celiac disease, allergies to cereals or fish and other serious and extensive food allergies. Exceptional diets were also an exclusion criterion as well as binge eating [199]. Inhaled corticosteroids were permitted. Participants taking fish and vegetable oil supplements or using stanol or sterol esters were asked to discontinue using these supplements at least 4 weeks before the beginning of the intervention. None of the study participants were excluded due to non-compliance with regard to the study protocol. The allowed weight change during the study was aimed to be less than 4 kg. Following the 4 week run-in period randomization (1:1) was performed by matching participants according to gender and medians of age, BMI and fasting plasma glucose at screening resulting in equal amounts of certain strata classes among the groups. 4.1.3. Dietary intervention The study diets were introduced to the participants at week 0 by a clinical nutritionist or dietician. The fourth edition of the Nordic nutrition recommendations formed the basis of the diet in the ND group [57] and the mean nutrient intake in the Nordic countries formed the basis for the control diet. The target nutrient composition and food items recommended in the study diets are summarized in Table 4. The main nutrient differences between the diets were the amount of dietary fiber and salt, and the quality of dietary fat. Both the ND and the control diets were isocaloric based on the evaluation of the habitual diet (calculated from a 4 day food record) during the run in period. Key food products were provided to the study participants in both groups. The individuals in the ND group received whole grain products such as oats and barley, crisp breads, whole grain wheat and rye breads, berry products such as frozen berries and dried berry powder as well as rapeseed oil and vegetable oil based spreads. Furthermore, fish (fatty and lean) was either directly provided or the cost for purchase was reimbursed to the study participants of the ND group. The individuals in the control diet group received low fiber cereal products, e.g. breads with fiber content less than 6 g/100 g and ordinary pasta and dairy fat based spread. The study participants kept dietary records regarding the intake of cereal products, berry products, vegetables and fish with a check-list. These records were checked at each visit to the study center, including the visits when the groceries were delivered to the participants, i.e. at 1-2 week intervals. Any intake of supplements was to be discontinued at least four weeks prior to entering the study. An exception was made for vitamin D supplements in Iceland which the participants could continue to use providing they kept it unchanged. The vitamin D supplementation was not included in the nutrient calculations. 33 Table 4. Advised nutrient composition and suggested food items of for the study diets Nutrients Carbohydrate, E% Sucrose, E% Fiber, g Protein, E% Fat, E% Saturated, E% Monounsaturated + Polyunsaturated, E% Salt, g Healthy Nordic diet 45-52 ≤10, max. 50 g/day ≥35 g/day or 4 g/MJ 18-20 30-35 <10 minimum 2/3 of total fat intake 6 g for women, 7 g for men Control diet 45-47 no restrictions 15-20 18-20 35 15 15 5 ≤10 Advised food items Cereal products ≥25% of total energy as whole grain, of which ≥50% as rye, barley and oat. Whole grain pasta and unpolished rice (≥6 g fiber/100 g) ≥2 meals/week Low salt content (≤1.0%) in breads Cereals with no added sugar or honey Bread (≥6 g fiber/100 g) ≥6 slices/day Vegetables, berries and fruits (potatoes not included) Fruits, vegetables and berries ≥500 g, of which berries ≥150 g/day, fruits ≥175 g/day and vegetables ≥175 g/day ≥25% of energy as refined grain, of which ≥90% as wheat Fruits and vegetables 200250 g/day No bilberries Dietary fats Rapeseed oil Rapeseed and/or sunflower oil and/or soybean oil based margarines with no trans fatty acids and >2/3 of unsaturated fats Unsalted nuts and seeds can be included in the diet Butter or other milk fat based spread (≥50% of total fat as saturated) Dairy products Low fat liquid dairy ≤1% fat Cheese ≤17% fat ≥2 portions/day Sweetened yoghurt and other fruit milk products avoided No limitations Fish ≥3 meals/week, two fatty fish (≥4% of fat) meals and one low fat fish meal ≤1 meal/week Meat Preferably white meat, poultry Low fat choices Game No limitations Avoidance of sugar-sweetened beverages Fruit and berry juices ≤1 glass/day No limitations Beverages 34 4.1.4. Dietary assessment The dietary intake was assessed using a consecutive 4 day (Wednesday to Saturday or Sunday to Wednesday) food record, with either weighed or estimated portion sizes. The baseline food diaries were recorded in the 4 week run-in period of the intervention. Dietary intake during the intervention was recorded at week 2 (data not included in this thesis) 12 and 18/24. Participants received both oral and written instructions on how to fill in the dietary records and were provided with electronic household scales or valid standardized household measure information on how to give information about the amount of food consumed. At each center the nutrient intake from food diaries was calculated, using national nutrient databases (Aivo Finland Ltd, Turku, based on the database of the National Institute of Health and Welfare, Finland, Dietist XP Software Package, version 3.1 [2009] linked to Swedish Food Database 2009, Sweden, Master Dieatist System version 1.235 [2007] based on Danish National Food administration database, Denmark, and The Icelandic Food Composition Database [ISGEM], Iceland). All the databases use EuroFIR definitions for chemicals and total dietary fiber determined by the AOAC methods [200-203]. 4.1.5. Anthropometric measurements Height was measured at week 0 with a calibrated stadiometer with a precision of 0.5 cm. During the measurements participants were in a fully erect position with the shoulder blades, buttocks and heels touching the wall. Participants were asked to stand straight, with heels and knees almost together, shoulders relaxed and arms hanging at sides. Body weight was measured at every visit to the clinic, i.e at screening, week 0, 2, 4, 8, 12, 16 and 18/24. Weight was measured in light clothes on a digital calibrated scale. It was measured to the nearest 0.1 kg with the subject standing motionless on the scale, weight equally distributed on each leg and arms hanging freely. BMI was calculated from the recorded height and weight using the standard formula, dividing the participants´ weight (kg) by the square of the height (m2). The waist circumference was measured at the screening visit, week 0, 12 and 18/24. For the measurement of waist circumference participants stood erect with arms relaxed at the sides and the feet slightly apart from each other with weight equally distributed on each leg. The lowest rib margin and the iliac crest were located and the midway between those two levels was marked. A flexible tape was then placed firmly on the measuring line and ensured that its position was horizontal all around the waist. Participants were asked to breathe normally and a measurement was made at the end of a gentle exhaling. All measures were performed twice using a flexible tape measure and recorded with the accuracy of 0.1 cm. Body composition was determined at week 0, 12 and 18/24 by bioelectrical impedance analysis (Kuopio: STA/ BIA Body Composition Analyzer and Bodygram Software, Akern Bioresearch Srl, Florence, Italy, Reykjavik: STA/BIA Body Composition Analyzer, Akern Bioresearch Srl, Florence, Italy, Lund: Xitron Hydra 4200 instrument, Xitron Technologies, San Diego, USA, Uppsala: Tanita BC-558Segmental Body Composition Monitor, Tokyo, 35 Japan, Aarhus: multi-frequency bioimpedance spectroscopy analyzer, UniQuest-Seac, SFB3, Brisbane, Australia. Body composition was not measured in Oulu). 4.1.6. Oral glucose tolerance tests Oral glucose tolerance tests (OGTT) were performed at week 0, 12 and 18/24. Participants were instructed to have an overnight fast before the OGTT and on arrival to the study clinics, fasting venous blood samples for plasma glucose and plasma or serum insulin measurements were collected. The participants were then asked to drink a glucose solution and finish it in less than 3 minutes. The glucose solution contained 75 g of D(+)-Glucose anhydrous Ph.Eur dissolved in 300 ml of water. In some cases 1.6 g of citric acid was also added to the solution. Participants were asked to remain in the testing area during the test and were not allowed to smoke or eat during the examination. Venous blood samples for plasma glucose and plasma or serum insulin measurements were collected 30 and 120 minutes after the glucose load. The samples were kept in ice until centrifugation in the local laboratory. 4.1.7. Biochemical measurements Blood samples were taken at the screening visit and at week 0, 12 and 18/24. Glucose and lipid analyses (triglycerides, total cholesterol and HDL cholesterol) were performed locally and apolipoproteins A1 (Apo A1) and B (Apo B) centrally in the University of Eastern Finland and Kuopio University Hospital according to the standard operational procedures agreed by all centers using automated clinical chemistry analyzers and routine clinical chemistry methods. For these automated analyses, the equipment and system reagents were from Siemens Healthcare Diagnostics (Tarrytown, NY, USA), Thermo Fisher Scientific (Vantaa, Finland), Roche Diagnostics GmbH (Mannheim, Germany) (in two laboratories), Toshiba (Japan), Abbott Laboratories (Irving, TX, USA), Ortho-Clinical Diagnostics and Johnson & Johnson (New Brunswick, NJ, USA). The sample material was either plasma or serum as required by the analytical method in each laboratory. The within-run variations (CV%) for the biochemical measurements were 0.4–4.6%. The between-run CV% was 0.5– 6%. Plasma insulin was measured centrally in Aarhus University Hospital, Denmark using the ELISA method from Dako Denmark A/S, Glostrup, Denmark, with a within-run and betweenrun CV% <8% and <10%, respectively. LDL-C concentrations were calculated using Friedewald’s formula. From data of the oral glucose tolerance test insulinogenic index (IGI), the ratio of the 30 minutes areas under the curve for insulin and glucose (InsAUC0-30/GluAUC0-30) and Stumvoll was calculated for evaluation of early phase insulin secretion (IGI = (Insulin 30 min–Insulin 0 min) / (Glucose 30 min–Glucose 0 min) and Stumvoll = 1283 + 1.829 × insulin 30 min (pmol/l) − 138.7 × glucose 30 min (mmol/L) + 3.772 × insulin 0 min (pmol/l)). Insulinogenic index at 120 minutes (IGI 120) was calculated for late phase insulin secretion (IGI 120 = (insulin 120 min–insulin 0 min) / (glucose 120 min–glucose 0 min)). Matsuda insulin sensitivity index (ISI) and disposition index (DI) were calculated for evaluation of 36 insulin sensitivity (Matsuda ISI=(10000/√([glucose 0 min x insulin 0 min] x [arithmetic mean of glucose 0, 30 and 120 min x arithmetic mean of insulin 0, 30 and 120 min]) [204] and DI=Matsuda ISI x insulinogenic index). 4.1.8. Dietary biomarkers Blood samples used for analysis of plasma AR were collected at week 0, 12 and 18/24 (N = 158). AR in plasma samples were analyzed at the Department of Food Science, BioCenter at the Swedish University of Agricultural Sciences. AR homologues C17:0-C25:0 were measured in 0.2 ml plasma samples according to a previously published gas chromatography– mass spectrometry (GC-MS) method [205]. Quality control samples (N = 4) were included in each batch to ensure appropriate within- and between batch precision (CV <15%). Samples from the same individual were analyzed within the same batch and samples from different centers were allocated randomly between batches. The ratio between the AR homologues C17:0 and C21:0, a marker of relative intake of whole grain rye of overall whole grain rye and wheat intake, was calculated. 4.2. ARI feasibility study 4.2.1. Participants and study design Participants were referred to the study by general practitioners in three health care clinics in the capital area of Iceland. Participants were 23 males, aged 42 to 80 diagnosed with type 2 diabetes or prediabetes by their family doctor. The exclusion criteria included any chronic disease and/or condition, which could hamper the adherence to the dietary intervention protocol. Participants followed their habitual diet for a week before randomization into a rye group (N=14) or a control group (N=9) for 26 weeks. All participants had three major visits, at week 0, 13 and 26 (Figure 3). Additionally, there was a group meeting with a clinical nutritionist in week 2 where participants received general recommendations on a healthy diet as well as a meeting with a physical therapist and a personal trainer during the first weeks. The participants in the rye group came every 2-3 weeks, to pick up the rye products that were provided and both groups were offered to pick up vegetable packages free of charge every 4 weeks. The intervention was carried out in April to November 2013. 37 Figure 3. Outline of the study design and main outcome measurements at each visit 4.2.2. Study diet Participants in the rye group were instructed to consume 100 g of whole grain rye each day for 26 weeks and rye products equivalent to this intake were provided free of charge. The products provided to the rye group and their rye content are listed in Table 5. The participants in the control group were asked to remain their usual diet, but avoid rye products. Weight loss during the study period was allowed. Table 5. Whole grain rye, energy, fiber, added sugar and salt content for the rye products provided for the rye group in the study Whole grain rye, Energy, g kcal Rye flakes (used for porridge) 100 335 Rye bread 1 50 193 Rye bread 2 38 235 Rye bread 3 59 179 Rye bread 4 59 162 Rye crisp bread 94 360 Data are presented as g or kcal per 100 g of product Fiber, Sugar, Salt, g g g 11.7 0 <0.01 8.3 1.5 1.2 7.5 0.8 1.2 8.0 N/A 1.5 10.6 2.9 1.2 19 0.9 <0.01 Participants in the rye group received both oral and written information on the amount of rye in the products provided and other common whole grain rye products, and how to reach 100 g of rye a day in their diet. To simplify, the participants in the rye group were asked to obtain ten points every day, where one point was earned for every 10 g of whole grain rye consumed. A color-printed booklet with names and pictures of the rye products provided in the study, as well as other common whole grain rye products, was handed to the participants. The booklet included information on nutritional value, amount of whole grain rye per 100 g of product and 38 per serving (e.g. slice of bread) and amount of points earned per serving. Altogether the booklet contained information on 32 rye products. Dietary intake was assessed using a consecutive 4-day (Wednesday to Saturday) weighed food record before week 0, at week 13 and 26. Participants received both written and oral instructions on how to fill in the diet records and were provided with electronic household scales. At the 0 week visit a nutritionist reviewed the first food diary with each participant and suggested exchanges for rye products. Rye porridge (from rye flakes) was recommended daily, to facilitate the rye consumption without adding increased energy to the diet. In addition, daily consumption of 1 - 4 slices of rye bread (dependent on the type of bread) was recommended. The nutrient intake from food diaries was calculated using The Icelandic Food Composition Database [ISGEM], Iceland). The database use EuroFIR definitions for chemicals and total dietary fiber determined by the AOAC methods [200]. 4.2.3. Physical activity After randomization, all participants (rye and control groups) were asked to exercise vigorously for at least 45 minutes three days/week, and walk at least 10 000 steps on other days, for 26 weeks. A gym membership and a pedometer were provided to all participants free of charge. In the beginning of the study each participant met with a physical therapist who evaluated the physical condition of the participants. Following that meeting the participants met with a personal trainer who gave them an individualized exercise program and showed them the exercises and the workout equipment. The participants were free to contact the personal trainer during the intervention period to get different (less/more advanced) programs. Thus, every participant got an exercise program designed for their needs, e.g. interests, age, strengths and weaknesses. 4.2.4. Compliance and acceptability AR in plasma were used to assess compliance to the whole grain rye intake with measures of total plasma AR and the AR C17:0/C21:0 ratio [12, 20, 196, 197], in addition to a weighed food diary. The plasma samples were analyzed at the Department of Food Science, BioCenter at the Swedish University of Agricultural Sciences as described in section 4.1.8. To measure the acceptability of the products at week 13 and 26, participants in the rye group were given three statements and asked to check whether they strongly agreed, agreed, neither nor, disagreed or strongly disagreed. The statements were: 1) I have experienced digestive discomfort after I started eating rye products 2) I like food products that contain rye and 3) I would prefer products that contain rye more than other products. 39 4.2.5. Anthropometric and biochemical measurements Height was measured at week 0 with a calibrated stadiometer with a precision of 0.5 cm. Body weight, waist circumference and body composition was measured at weeks 0, 13 and 26 according to the protocol described in section 4.4.2. OGTTs were performed at week 0 and week 26 according to protocol described in section 4.1.6. except for venous blood samples for plasma glucose and plasma or serum insulin measurements were collected 60 and 120 minutes after the glucose load. Measurements on glucose and blood lipids (triglycerides, total cholesterol and HDL cholesterol) were performed in the local laboratory according to standard operational procedures. Automated clinical chemistry analyzers and routine clinical chemistry methods were used to measure glucose, insulin, triglycerides, total cholesterol and HDL cholesterol. LDL-C concentrations were calculated using Friedewald’s formula. 4.3. Statistical analysis Data entering All intervention centers in the SYSDIET study entered their data into Microsoft Office Excel 2007 (Windows) data form prior to pooling the data in a joint database maintained at VTT Technical Research Centre of Finland. Data was exported from the database and imported to SPSS (Statistical Package for the Social Sciences) for Windows, version 20.0 (SPSS Inc., Chicago, Illinois, USA) in which most statistical analysis was done for the SYSDIET study, as well as the ARI feasibility study. Descriptive analyses All analyses for the SYSDIET intervention illustrated in the thesis were performed with N = 158, i.e. the participants who completed the intervention and had valid plasma samples for AR analysis at both week 0 and 18/24, unless otherwise indicated. Power analyses were done during the study design phase in the SYSDIET study. Based on scientific experience and the power calculations, 80-90 participants were expected to provide sufficient statistical power (alpha <0.05, beta >0.8) to detect 10% difference in serum cholesterol levels between groups. For the ARI feasibility study all participants that completed the intervention were included in the analyses (N = 20). Normality of variables was checked by visual inspection. Normal variables were described as means and standard deviations (SD) while skewed variables were described using medians and interquartile range (IQR). Independent sample t-tests or Mann-Whitney U test were used to compare medians between groups and genders. Paired sample t-tests or Wilcoxon´s signed rank tests were used to compare values between baseline and midpoint (week 12 in SYSDIET and week 13 in ARI) or the end of the interventions (week 18/24 in SYSDIET and week 26 in ARI). The answers to the acceptability questions in the ARI feasibility study were combined before statistical analysis combining strongly agree and agree to agree and strongly disagree and disagree to disagree. 40 Determinants of plasma AR To explore the determinants for total plasma AR concentration at 18/24 weeks in the SYSDIET study a linear regression for total plasma AR at 18/24 weeks was performed with group, baseline concentration for total plasma AR, gender, age, BMI, study center and triglycerides at 18/24 weeks as factors in the model. As the distribution of total plasma AR was right skewed, log transformation was performed before this analysis to better approximate normal distribution. One-way ANOVA was used to compare means between study centers and Spearman rank correlation coefficient (rs) was used to calculate correlation between total plasma AR concentration and various nutrients abundant in whole grains. Reproducibility Reproducibility, a measure of intra-person stability in dietary intake or biomarker concentration during the intervention, was assessed by estimating the intra-class correlation coefficient (ICC). ICC is defined as the intra-person variance divided by the total variance (i.e. the sum of intra- and inter-person variance), and it assumes a common mean between replicate measurements. ICC was calculated using freely available SAS macro [206] that was imported to SAS 9.2 (SAS Institute). All ICC estimates were based on log-transformed values. Examining associations between plasma AR and cardiovascular risk factors For the SYSDIET study, secondary analyses were performed with data from the ND and control groups combined. Associations between total plasma AR and AR C17:0/C21:0 ratio at 18/24 weeks and various glucose, insulin and blood lipid outcomes were analyzed using linear regression. Where distributions were skewed log-transformation was performed on the dependent variables for linear regression. For evaluating the association to glucose and insulin outcomes, linear regression for fasting plasma glucose, insulin, Matsuda, DI and IGI was performed with AR C17:0/C21:0 ratio, baseline values for the dependent variable, age, gender, BMI and smoking as factors (model 1). For evaluating the association to blood lipid outcomes linear regression for total, LDL, HDL, non-HDL cholesterol, Apo A1 and B and LDL/HDL cholesterol ratio was performed with AR C17:0/C21:0 ratio, baseline values for the dependent variable, BMI and statin use as factors (model 1). The linear regressions were also performed with total plasma AR concentration instead of C17:0/C21:0 in model 1. As rye often contains more AR compared to wheat, total plasma AR concentration was added as a variable (model 2) in addition to the variables in model 1, to prevent confounding of high AR content. Furthermore, we individually adjusted for saturated fat intake calculated from food records and serum phospholipid fatty acids myrisitc acid, palmitic acid, omega-3 fatty acids and C15:0 for the blood lipid analyses, to prevent confounding of dietary fat intake. In all analyses, model residuals were checked normality visually, with help of QQ-plots. 41 4.4. Approvals Local ethical committees in Denmark, Finland, Iceland and Sweden reviewed and approved the SYSDIET study protocol which followed the Helsinki declaration guidelines. The ARI feasibility study was approved by the National Bioethical Committee in Iceland, Local Ethical Committee at the Landspitali-University Hospital in Iceland, Local Ethical Committee at the Primary Health Care of the Capital Area and by the Icelandic Data Protection Commission. Informed written consent was obtained from all participants in both studies prior to participation. 42 5. Results and discussion The main findings from the four papers are hereby discussed. More detailed descriptions of the results are found in papers I-IV. 5.1. The SYSDIET intervention 5.1.1. Baseline characteristics of the participants In total, 200 participants started the intervention, 104 in the ND group and 96 in the control group. Participants completing the trial were 96 in the ND group (92%) and 70 in the control group (72%). Blood samples for AR analysis at week 0 and week 18/24 were obtained from 158 participants (Figure 4). Figure 4. Flow chart of the SYSDIET study 43 Baseline characteristics of the participants that finished the study and had valid blood sample for AR analyses at week 0 and 18/24 are shown in Table 6. Approximately two-thirds of the participants were classified as obese (BMI ≥30 kg/m2) and one-third as overweight (BMI 25– 29.99 kg/m2). Eighty participants had fasting plasma glucose ≥5.6 mmol/L at week 0. Baseline characteristics between the two groups were not significantly different. Table 6. Baseline characteristics of the participants in the healthy Nordic diet (ND) group and the control group who completed the SYSDIET study and had valid blood samples for alkylresorcinols analyses at week 0 and 18/24 ND group (N = 93) Age, years 54.8 (8.1) Weight, kg 89.7 (13.5) 2 BMI, kg/m 31.6 (3.4) Waist circumference, cm 103 (10) † Fat , % 39 (8) Blood pressure systolic, mm Hg 130 (15) diastolic, mm Hg 82 (11) ‡ Smoking , n (%) 6 (6) ‡ Statin use , n (%) 19 (20) ‡ Metabolic syndrome , n (%) 85 (91) Data are presented as mean (SD) unless otherwise indicated Abbreviations: BMI, body mass index † n = 122, n = 72 in ND group and n = 50 in control group ‡ Data are presented as number (%) Control group (N = 65) 54.2 (8.3) 93.2 (13.8) 31.8 (2.9) 106 (3) 39 (7) 130 (17) 80 (11) 7 (9) 18 (28) 56 (86) 5.1.2. Changes observed during the intervention The main results from the SYSDIET intervention, dietary intake and health effects have been published previously [51] and will only shortly be discussed here. Consistent with the isocaloric study design there was no change in total energy intake during the study and energy intake was not different between the groups. Hence, body weight remained constant during the intervention in both groups. In general, dietary compliance was satisfactory as evidenced from the dietary records [51]. One of the most prominent changes in nutrient composition during the intervention was increased fiber intake in the ND group [51]. The median fiber intake at week 0 was 21 (8) g/day and 2.5 (1.0) g/MJ and during the intervention, the median intake of dietary fiber increased by 71% (P <0.001) in the ND group and decreased by 20% in the control group 44 (P<0.001) (Table 7). By week 12, both groups had reached the predetermined goals for fiber intake of at least 35 g/day (or 4 g/MJ) and 15-20 g/day, in the ND and control groups respectively, and the intake was steady throughout the intervention. Table 7. Total dietary fiber intake in healthy Nordic diet (ND) and control groups during the intervention ND group (N = 80) Week 0 Week 18/24 Control group (N = 60) Week 0 Week 18/24 Dietary fiber intake, g/day 21.1 (8.8) 36.0 (15.1)# 20.0 (7.1) Data are presented as medians (IQR) * Significantly different from ND group at that time point, P <0.05 # Significantly different from baseline in that group, P <0.05 16.0 (6.2)*# The main results from the SYSDIET intervention were that the healthy Nordic diet did not modify insulin sensitivity and glucose tolerance but resulted in significant between-group changes in non-HDL cholesterol, LDL to HDL cholesterol ratio and Apo A1 to Apo B ratio favoring protection from atherosclerosis. Additionally, there was a significant increase in interleukin-1 Ra (IL-1 Ra) levels [51]. Exploration of the participants that had valid plasma samples for AR analyses (N = 158) showed similar results with no significant differences between groups in indices of insulin sensitivity or -secretion. At week 0, median fasting plasma glucose was 5.7 (0.9) mmol/L in both groups and did not change significantly during the intervention period [197]. There were small, but significant, decreases in concentrations of total cholesterol (P = 0.018), LDL (P = 0.006) and non-HDL cholesterol (P = 0.006) and LDL/HDL cholesterol ratio (P = 0.003) within the ND group from week 0 to 18/24 but not within the control group [207]. 45 5.2. Plasma alkylresorcinols as dietary biomarkers in healthy Nordic diet (Paper I) 5.2.1. Changes in plasma alkylresorcinols concentrations during the intervention The change in dietary fiber intake in the present study was reflected in increased plasma AR concentration in the ND group, suggesting that cereal fiber from wheat and rye substantially contributed to the increased fiber intake. There was a 68% increase in total plasma AR concentration in the ND group during the intervention, with median concentrations increasing from 73 (88) nmol/L at week 0 to 106 (108) nmol/L (P = 0.001) at week 18/24, while a slight decrease was seen in the control group (P = 0.021) (Table 8). Even at baseline, the total plasma AR was relatively high compared to the participants in the EPIC study where the geometric mean plasma total AR concentrations were between 35 and 41 nmol/L in participants from the Central European and Scandinavian countries and below 23 nmol/L in participants from the Mediterranean countries [124]. Plasma levels have, however, been found to vary widely, with previous studies reporting mean levels between 38 - 388 nmol/L with a moderate to high rye intake [108-110]. Table 8. Total plasma alkylresorcinols (AR) concentration and AR C17:0/C21:0 ratio for the healthy Nordic diet group (ND) and control group at baseline and after the 18/24 week intervention ND group Week 0 Week 18/24 71 (70) 61 (40)*# C17:0/C21:0 ratio 0.3 (0.2) 0.3 (0.2) 0.3 (0.2) Data are presented as medians (IQR) * Significantly different from ND group at that time point, P <0.05 # Significantly different from baseline, P <0.05 0.2 (0.2)*# Total plasma AR, nmol/L 73 (88) 106 (108)# Control group Week 0 Week 18/24 There was no difference between the groups at baseline in total plasma AR concentration but the difference was significant both at week 12 and 18/24 (P <0.001) (Figure 5). Previous data suggest that total plasma AR meets most of the criteria of a good biomarker for intake of whole grain cereal fiber and whole grain wheat and rye, particularly with frequent consumption and a wide range of intake [12, 18, 19, 21, 23]. Total plasma AR concentration increases proportionally with AR and whole grain intake [18, 22-26] and can be used to categorize subjects as low and high consumers of whole grain foods [19, 126]. Furthermore, it can be used as a marker for compliance to a whole grain diet in intervention studies [20, 27]. 46 Plasma total alkylresorcinols, nmol/L 160 * 140 * 120 ND 100 80 60 Control 40 20 0 Week 0 Week 12 Week 18/24 Figure 5. Total plasma alkylresorcinols concentrations for participants in healthy Nordic diet (ND) and control groups at week 0, 12 and 18/24. Values are medians and bars represent the interquartile range (IQR) * Different from control group at that time, P <0.05 The C17:0/C21 ratio for the ND group and control group at week 0 and week 18/24 are shown in Table 8. The median AR C17:0/C21:0 did not change in the ND group during the intervention and at 18/24 weeks it was 0.3 (0.2). However, the C17:0/C21:0 ratio decreased significantly in the control group and was 0.2 (0.2) at week 18/24. Although the ratio of C17:0/C21:0 is approximately 1.0 in rye the ratio is expected to be somewhat lower in plasma samples and has typically been around 0.6-0.8 in fasting samples after a whole grain rye intervention for 1-8 weeks [23, 25]. It is important to note that the AR C17:0/C21:0 ratio is a marker of relative rye intake, not absolute intake, and the lower ratio in the present study than in previous studies may be indicative of higher whole grain wheat intake as other whole grain products than provided were not allowed in these studies. A ratio above 0.15 is usually indicative of some rye intake [12]. The reason for the lower ratio in plasma compared to what is found in whole grain/bran rye foods is likely due to faster metabolism of the AR homologue C17:0 [107, 131] but further insight into absorption and metabolism of the different homologues is necessary [12]. 47 5.2.2. Determinants of plasma alkylresorcinols concentration Diet group allocation (ND group vs. control group; P < 0.001) and total plasma AR concentration at baseline (P < 0.001) were predictors of total plasma AR concentration at week 18/24 with adjusted R2 = 0.38 (Table 9). Table 9. Regression coefficients for factors affecting plasma total alkylresorcinols (AR) at week 18/24 (log transformed) B† (95% CI) P-value Model 1 log AR 18/24WK Adjusted R Squared = 0.38 Study group‡ 0.234 (0.157, 0.311) <0.001 AR week 0 0.354 (0.212, 0.496) <0.001 Gender 0.026 (-0.055, 0.106) 0.53 Age 0.003 (-0.001, 0.008) 0.17 BMI 0.002 (-0.009, 0.014) 0.70 Triglycerides 18/24 weeks 0.110 (0.042, 0.177) 0.002 Study center -0.019 (-0.041, 0.003) 0.09 Abbreviations: BMI, body mass index † B reflects the change in the log-transformed outcome variable for each 1-unit changes in the independent variable. ‡ Healthy Nordic diet group (ND) compared with control group Serum triglycerides concentration at week 18/24 was also a significant determinant for total plasma AR at 18/24 weeks (P = 0.002). Blood lipids were expected to correlate with total plasma AR concentration as studies have suggested that AR is transported in lipoproteins [19, 25]. The finding that serum triglycerides were a significant predictor of total plasma AR concentration suggests that factors that cause transient increases in plasma triglycerides may also be important for the repeatability of plasma AR measurements [125]. Study center was of borderline significance (P = 0.09) as a determinant for total plasma AR at week 18/24. Significant differences were in baseline values for total plasma AR concentration, as well as the AR C17:0/C21: ratio, among study centers (Table 10). At the two Finnish centers, Oulu and Kuopio, the participants had the highest median (IQR) concentrations, 108 (96) nmol/L and 108 (72) nmol/L, respectively, whereas in Reykjavik the lowest median concentration was observed, 43 (37) nmol/L. The differences in total plasma AR among centers are likely to reflect differences in whole grain wheat and rye intake in the source populations, which also resulted in differences in total dietary fiber intake. Hence, the increase in total plasma AR concentration in Finnish participants during the intervention was 48 not as high as in participants from other centers. Instead, Finnish participants in the control group had a greater decrease in AR concentration. Similarly, for the other centers, total plasma AR concentration increased substantially in the ND group and decreased in the control group, but not as sharply as in the Finnish control group. Table 10. Baseline dietary intake of fiber, plasma total alkylresorcinols (AR) and AR C17:0/C21:0 ratio at different study centers† Dietary intake‡ Plasma concentration Fiber, Fiber, Total AR, AR C17:0/C21:0 Study center g/day g/MJ nmol/L ratio b ab a Oulu (N = 35) 21 (7) 2.7 (0.9) 108 (96) 0.3 (0.2)a Kuopio (N = 32) 25 (11)a 3.1 (1.5)a 108 (72)a 0.4 (0.2)a Reykjavik (N = 15) 18 (8)b 1.8 (0.7)c 43 (37)c 0.2 (0.2)b Lund (N = 37) 19 (8)b 2.2 (0.7)c 59 (36)b 0.2 (0.1)b Uppsala (N = 9) 22 (8)ab 2.5 (0.7)bc 56 (46)bc 0.2 (0.3)b 0.3 (0.2)b Aarhus (N = 30) 22 (12)ab 2.4 (1.1)bc 67 (77)b Data are presented as median (IQR) † Labeled medians in a column without a common letter differ, P <0.05. ‡ N = 151, Oulu N = 35, Kuopio N = 31, Reykjavik N = 13, Lund N = 36, Uppsala N = 8, Aarhus N = 28 Total plasma AR concentration was not significantly different between genders, and gender was not a significant factor in the model (Table 9). Several studies have found higher plasma AR concentration in men compared to women which has been attributed to higher intake of foods containing AR by men and more rapid elimination of AR in women [12, 19, 22, 28]. AR is present in adipose tissue [122, 208] and as women generally have greater fat stores than men, it may result in lower plasma AR. In the present study, the women had a higher body fat percentage than men, suggesting that factors other than variations in body fat may have caused the lack of gender difference. 5.2.3. Six to twelve week reproducibility of plasma alkylresorcinols The reliability of AR plasma measurements at week 12 and 18/24 was moderate to good, with intra-class correlation (ICC) 0.44 – 0.72 (Table 11). The ICC was higher among men than women and higher in ND group than control group. This is similar or higher than seen in previous studies [12, 22, 24, 127]. The ICC has been reported to be 0.42-0.48 over a 2- to 4month period under free living condition [12, 19, 22, 127] and a ICC of 0.88 - 0.90 was reported in a 6 week intervention setting with extremely high daily intake of rye bran/wholegrain rye products [24]. Similarly, higher reliability was observed in the ND group in the 49 present study than in the control group as ND group had higher and more frequent intake of AR. The variation in free-living subjects makes it difficult to classify an individual’s whole grain wheat and rye intake with precision, from a single sample. Nevertheless AR are considered a valid measure for comparing different populations, as mean plasma AR are well correlated with mean AR intake when results from relevant studies are combined [12]. Table 11. Estimated intra-class correlation coefficients (ICC, based on log-transformed values) between week 12 and 18/24 for all participants combined, healthy Nordic diet (ND) group and control groups All Dietary fiber intake Both genders 0.86 (0.82-0.90) Men 0.89 (0.84-0.94) Women 0.84 (0.77-0.89) Total plasma alkylresorcinols Both genders 0.68 (0.59 - 0.76) Men 0.72 (0.58-0.83) Women 0.72 (0.64-0.79) Data are presented as ICC (95% CIs) ND group Control group 0.74 (0.63-0.82) 0.42 (0.24-0.62) 0.80 (0.64-0.90) 0.49 (0.24-0.75) 0.69 (0.54-0.80) 0.36 (0.15-0.65) 0.66 (0.54-0.77) 0.49 (0.32-0.67) 0.69 (0.49-0.84) 0.55 (0.30-0.78) 0.64 (0.49-0.77) 0.44 (0.22-0.69) The half-life of AR is relatively short, about 5 hours, and the absorption half-life is about 6-8 hours [107]. Strictly defined, AR should be regarded as a short-term biomarker but under stable and frequent intake conditions absorption and elimination counteract each other and it may reflect exposure fairly well [123]. In the present study, cereals and bread were consumed on a daily basis thus, plasma AR concentrations should be stable and appropriately reflect intake during the intervention period. It is, however, important to note that the use of plasma AR as a biomarker of whole grain intake is not practical in populations who primarily consume other types of whole grains which do not contain AR, such as oats or brown rice, or when whole grains are not regularly consumed. 50 5.2.4. Correlations between alkylresorcinols and nutrient intake Positive correlations were observed between plasma total AR concentration and total fiber intake at week 0, 13 and 26, as well for vitamin E, folate and magnesium (Table 12), nutrients abundant in whole grains [33]. Under dietary intervention settings, total plasma AR have previously been found to correlate with insoluble fiber [109] as well as dietary fiber [18]. Table 12. Correlations for plasma total alkylresorcinols and intakes of various nutrients abundant in whole grain for both healthy Nordic diet (ND) and control group Baseline Week 12 ND Control All All group group Total fiber intake 0.26* 0.47* 0.19 0.18 Vitamin E 0.26* 0.44* 0.33* 0.18 Folate 0.32* 0.24* -0.01 -0.01 Magnesium 0.35* 0.36* 0.23* -0.01 Values are presented as Spearman´s correlation coefficients (rs). * Significant correlation, P < 0.05 Week 18/24 ND Control All group group 0.57* 0.37* 0.34* 0.49* 0.27* 0.20 0.40* 0.21* 0.08 0.44* 0.39* 0.02 5.3. Plasma alkylresorcinols and cardiovascular risk factors (Papers II-III) The range of total plasma AR at week 18/24 was 14 to 499 nmol/L when the ND and control groups were combined for secondary analyses to investigate the associations between plasma AR at 18/24 weeks and 1) glucose metabolism and 2) blood lipid concentrations. 5.3.1. Glucose and insulin parameters Total plasma AR concentration at week 18/24 was not significantly correlated with concentrations of fasting plasma glucose (P = 0.51), insulin (P = 0.57) or other measures of insulin sensitivity or –secretion, when adjusted for baseline value of the dependent variable, age, gender, BMI and smoking. Combining the data from both weeks 12 and 18/24 did not change the results. Interestingly, the AR ratio C17:0/C21:0, reflecting the relative intake of whole grain rye of whole grain rye and wheat intake, was inversely associated with fasting insulin concentrations when adjusted for the same covariates (model 1, P = 0.016) and also when total plasma AR was added to the regression model to prevent confounding of high AR content (model 2, P = 0.011) (Table 13). Total plasma AR and the AR C17:0/C21:0 ratio were weakly correlated (Spearman´s rho = 0.20, P = 0.012). 51 The AR C17:0/C21:0 ratio was also positively associated with the insulin sensitivity indices Matsuda ISI (P = 0.005 and P=0.004 for model 1 and 2, respectively) and DI (P = 0.022 and P=0.023 for model 1 and 2, respectively) (Table 13). The AR C17:0/C21:0 ratio was, however, not associated with fasting plasma glucose or the indices for early and late insulin secretion [197]. Thus, relative whole grain rye intake, reflected by the AR ratio C17:0/C21:0, seems to be associated with lower fasting plasma insulin and increased insulin sensitivity. Table 13. Regression models for fasting insulin and insulin sensitivity indices at 18/24 weeks (log transformed values) Dependent variable Independent variable B† (95% CI) at 18/24 weeks at 18/24 weeks Fasting insulin Model 1 AR C17:0/C21:0 -0.43 (-0.77 to -0.08) Model 2 AR C17:0/C21:0 -0.45 (-0.79 to -0.10) Matsuda Model 1 AR C17:0/C21:0 0.23 (0.07 to 0.40) Model 2 AR C17:0/C21:0 0.24 (0.08 to 0.40) DI Model 1 AR C17:0/C21:0 0.77 (0.11 to 1.43) Model 2 AR C17:0/C21:0 0.77 (0.11 to 1.43) † B reflects the change in the log-transformed outcome variable for each 1-unit changes in the independent variable. Model 1: adjusted for baseline value for the dependent variable, age, gender, BMI and smoking Model 2: adjusted for baseline value for the dependent variable, total plasma AR, age, gender, BMI and smoking Abbreviations: AR, alkylresorcinols; DI, disposition index; BMI, body mass index There are conflicting findings from previous intervention studies evaluating the effects of whole grain intake on glucose metabolism [159, 160, 209, 210], presumably due to differences in the type, amount, structure and preparation of whole grain products consumed [166] and different methodologies used for glucose and insulin analyses [157]. Poor compliance may also contribute to these varied findings, although good adherence has been indicated in most studies using weighed food records. In a whole grain intervention study by Kristensen et al [211] the effects on body composition were clearer after excluding noncompliant participants based on total plasma AR concentration. Marklund et al [95] also got greater cardiometabolic effects of a healthy Nordic diet among compliant participants using a combination of dietary biomarkers to assess compliance. In both studies food records indicated good compliance. Hence, without an objective measurement of compliance, such as 52 dietary biomarkers, it is difficult to rule out the possibility of low adherence to the intervention diet. Nevertheless, whole grain intake has repeatedly been associated with decreased risk of type 2 diabetes in epidemiological studies [2, 136-140], presumably due to enhanced insulin receptor sensitivity through chronic lowering of the overall dietary glycemic index and related insulin secretion [149-155]. The inverse associations between C17:0/C21:0 and fasting insulin and insulin sensitivity indices observed in the present study are in line with previous findings showing positive effects of higher rye consumption on fasting plasma insulin and to a lesser extent on fasting plasma glucose [159]; further supporting the use of the ratio of the homologues C17:0 and C21:0 as an indicator of whole grain/bran rye intake. Although the AR C17:0/C21:0 ratio reflects only the relative wholegrain/bran rye consumption of total whole grain wheat and rye intake, not total rye intake, the advised whole grain rye intake was high in the present study in relation to other sources of AR and therefore the ratio should reflect the total rye intake reasonably well. Rye was mostly consumed in the form of bread. Several intervention studies have shown diminished levels of postprandial insulin following a rye bread meal [161, 162, 164], possibly explained by structural differences between rye and whole-wheat breads [164, 212]. The lower postprandial insulin response after rye bread consumption as compared to wheat bread may be due to the difference in structure rather than fiber content of whole grain, as low fiber rye bread showed the same results as high fiber rye bread consumption [163, 165]. Moreover, an improved early insulin response has been exhibited after rye bread interventions [163, 165]. In the present study there was, however, no significant association between the AR C17:0/C21:0 ratio and early or late insulin secretion phase as measured by insulin secretion indices. 5.3.2. Blood lipids Total plasma AR concentration was not significantly associated with concentrations of fasting serum total cholesterol (P = 0.95), LDL cholesterol (P = 0.23), HDL cholesterol (P = 0.25), LDL-C/HDL-C ratio (P = 0.13), triglycerides (P = 0.14), Apo A1 (P = 0.09) or Apo B (P = 0.86) after adjusting for confounders. Several animal and human studies have shown cholesterol lowering effects of whole grain rye [177, 178, 190-192] but the lack of associations in these studies may be due to considerable amount of AR intake originating from whole grain wheat. Whole grain wheat intake does not typically affect blood cholesterol and is thus often used as a control in experimental studies [188]. Supporting that hypothesis, the AR C17:0/C21:0 ratio was inversely associated with LDL cholesterol concentration when adjusted for baseline LDL cholesterol concentration, BMI, age and statin use (model 1, P = 0.040) and also when total plasma AR was added to the regression model (model 2, P = 0.046) (Table 14). Similarly, AR C17:0/C21:0 ratio was inversely associated with LDL/HDL cholesterol ratio (P = 0.020 and P = 0.024 for model 1 and model 2, respectively), non-HDL cholesterol (P <0.001 for both model 1 and model 2), triglycerides (P = 0.004 and P=0.002 for model 1 and model 2, respectively) and apolipoprotein B concentrations (P = 0.047 for both model 1 and model 2) and positively associated with HDL cholesterol concentration (P = 0.044 and P = 0.049 for model 1 and 53 model 2, respectively) (Table 14). Including saturated fatty acids calculated from food records to the models did not change the results nor did including the serum phospholipid fatty acids myristic acid, palmitic acid, omega-3 fatty acids and C15:0, except for HDL cholesterol in which the association with C17:0/C21:0 became borderline significant (P = 0.067). Thus, increased relative whole grain rye intake, reflected by the AR ratio C17:0/C21:0, seems to be associated with favorable blood lipid outcomes. Table 14. Regression models for concentrations of serum LDL, HDL, non-HDL cholesterol, triglycerides and apolipoprotein B and the cholesterol LDL-C/HDL-C ratio at 18/24 weeks Dependent variable at Independent variable at B† (95% CI) 18/24 weeks 18/24 weeks LDL cholesterol Model 1 AR C17:0/C21:0 -0.41 (-0.80 to -0.02) Model 2 AR C17:0/C21:0 -0.40 (-0.79 to -0.01) Log HDL cholesterol Model 1 AR C17:0/C21:0 0.11 (0.00 to 0.21) Model 2 AR C17:0/C21:0 0.11 (0.00 to 0.21) Log LDL/HDL cholesterol Model 1 AR C17:0/C21:0 -0.20 (-0.37 to -0.03) Model 2 AR C17:0/C21:0 -0.20 (-0.37 to -0.03) Log non HDL cholesterol Model 1 AR C17:0/C21:0 -0.84 (-1.22 to -0.45) Model 2 AR C17:0/C21:0 -0.81 (-1.19 to -0.44) Log triglycerides Model 1 AR C17:0/C21:0 -0.35 (-0.59 to -0.12) Model 2 AR C17:0/C21:0 -0.37 (-0.60 to -0.13) Log apoliprotein B Model 1 AR C17:0/C21:0 -0.12 (-0.24 to 0.00) Model 2 AR C17:0/C21:0 -0.12 (-0.24 to 0.00) † B reflects the change in the log-transformed outcome variable for each 1-unit changes in the independent variable. Model 1: adjusted for baseline value of the dependent variable, BMI, age and statin use Model 2: adjusted for baseline value of the dependent variable, total plasma AR, BMI, age and statin use Abbreviations: AR, alkylresorcinols; HDL, high-density lipoprotein; LDL, low-density lipoprotein; BMI, body mass index 54 Health effects of rye are far less studied than oats and barley and studies on the effects of rye consumption on blood cholesterol are scarce. The mechanisms that have been proposed to cause the reduction in blood cholesterol levels focus on the fiber and phytochemical content of whole grains, particularly water-soluble viscose fiber [144, 189] and it has been established that the cholesterol lowering effect of β-glucans is dependent on the molecular weight and subsequent viscosity [70]. Rye is particularly high in dietary fiber, ranging between 18-22%, and it contains substantial amounts of extractable β-glucans and arabinoxylans, which may contribute to cholesterol reduction through a similar mechanism as β-glucans from oats and barley [70, 82]. Animal studies have shown that rye can affect blood lipid concentrations by increasing the total concentration of bile acids in the bile [190, 191]. Previous studies have shown that rye bread consumption decreases total and LDL cholesterol concentrations in men with moderately elevated serum cholesterol [177], and lower concentrations of plasma free-cholesterol, total cholesterol, triglycerides and phospholipids have been demonstrated after a high fiber rye diet than a low fiber wheat diet [178]. Additionally, animal studies have shown cholesterol lowering effects of rye consumption [190-192]. However, Moazzami et al showed an unfavorable shift in serum cholesterol levels after rye bread intake in postmenopausal women [194]. Processing of rye and other whole grains, such as fermentation, have been reported to influence the molecular weight of cereal dietary fiber, hence it is possible that the lower viscosity of the fibers could explain the lack of beneficial effects reported in some studies [82]. Nevertheless, these results suggest that higher relative whole grain rye consumption is associated with favorable outcomes in blood lipids and this association needs to be further studied. 5.4. ARI feasibility study (Paper IV) As Icelanders had the lowest intake of dietary fiber and whole grain rye in the SYSDIET intervention [196], the acceptance and compliance to a high whole grain rye intake was explored in a feasibility study (ARI). Altogether, 23 men started the intervention, 14 in the rye group and 9 in the control group. The drop-out rate was low with only three men dropping out of the study, two in the rye group and one in the control group. Two men dropped out during the first weeks for personal reasons not related to the intervention and one had a medical condition and was not able to finish the intervention. Thus, 20 men (87%) completed the study, 12 in the rye group and 8 in the control group. 5.4.1. Baseline The mean age of the participants was 59 (11) years. Baseline characteristics are presented in Table 15. The BMI for the participants at baseline was 24 – 38 kg/m2. There was no significant difference between the groups in baseline characteristics except that participants in the control group took more steps/day than participants in the rye group. 55 Table 15. Baseline characteristics of the participants in the ARI feasibility study Age, years Height, cm Weight, kg Body mass index, kg/m2 Waist circumference, cm Fat, % Data are presented as mean (SD) Rye group (n = 14) 57 (11) 178 (6) 99.4 (11.8) 31.5 (4.4) 114 (10) 32.9 (5.0) Control group (n = 9) 62 (11) 178 (8) 93.5 (13.8) 29.4 (3.0) 109 (8) 34.4 (3.5) 5.4.2. Rye and dietary fiber consumption At week 0 the rye consumption was 11 (19) g/day and as expected, there was a highly significant increase in rye consumption in the rye group from week 0 to week 26, with reported rye intake increasing to 103 (38) g/day at week 13 and 86 (41) g/day at week 26 (P = 0.003). The rye consumption in the control group was unchanged (Figure 6).The mean rye consumption for the rye group in the intervention was 93 (44) g/day, equivalent to an 8-fold increase in daily rye consumption. The mean rye consumption in the control group was 10 (8) g/day. The rye consumption was right skewed with 7 out of 12 participants consuming ≥100 g/day of rye during the intervention, three consuming 60 to 99 g/day and two consuming below 60 g/day. The fiber intake at week 0 was also low, 19 g/day, comparable to the average consumption of Icelandic men, 17.7 g/day, according to recent National Dietary Survey [213]. Reflecting the changes in rye intake, there was a borderline significant increase in fiber intake in the rye group, with mean intake 19 (10) g/day at week 0, 33 (10) g/day at week 13 and 28 (7) g/day at week 26 (P = 0.06). On average, 72% of daily rye consumption in the rye group came from the rye flakes which emphasizes the importance of including rye porridge in future studies. Whole grain rye porridge has also been shown to improve satiety compared to wheat bread breakfast, subsequent decreased hunger and desire to eat up to 8 hours after consumption [214, 215]. Possibly the habitually low bread consumption of Icelandic adults may favor the acceptability of the porridge, as introduction of a bread meal may cause greater changes in the daily diet than changes in breakfast choice. 56 40 RG 100 80 60 40 CG 20 0 Fiber consumption, g/day Rye consumption, g/day 120 35 RG 30 25 20 CG 15 10 5 0 Week 0 Week 13 Week 26 Week 0 Week 13 Week 26 Figure 6. Rye consumption and total dietary fiber intake for rye group (RG) and control group (CG) at week 0, 13 and 26 of the intervention. There was no significant difference between groups in energy or macronutrient intake during the intervention (Table 16) indicating that the rye products were not added to the baseline diet but exchanged for other carbohydrate rich food. The carbohydrate intake seemed to increase in the rye group between weeks 13 and 26 and decrease in the control group but the change was not statistically significant nor was the difference between the groups. Table 16. Energy and macronutrient intake for both groups at week 0, 13 and 26 Rye group Control group Week 0 Week 13 Week 26 Week 0 Week 13 Week 26 Energy, kcal 2000 (534) 2175 (629) 1995 (631) 2410 (325) 2249 (655) 2230 (578) Carbohydrates, E% 39 (5) 39 (4) 43 (5) 43 (8) 43 (9) 38 (6) Protein, E% 21 (5) 21 (5) 20 (4) 19 (3) 18 (2) 17 (3) Fat, E% 34 (7) 36 (5) 37 (6) 35 (8) 38 (7) 39 (6) Added sugar, E% 6 (2) 6 (2) 5 (3) 8 (7) 7 (7) 5 (4) Data are presented as means (SD) 5.4.3. Compliance In addition to dietary records, compliance to the study diet was assessed with plasma AR as dietary biomarker. Similar to rye and dietary fiber intake, the total plasma AR concentration and the AR C17:0/C21:0 ratio increased in the rye group but not in the control group indicating good compliance to the whole grain rye intake (Figure 7). As discussed previously plasma AR has been established as a biomarker for whole grain wheat and rye intake in numerous studies [18, 19, 196] and the AR C17:0/C21:0 ratio has been suggested as an indicator of relative rye intake [25, 28, 197]. For the rye group in the present study, the mean 57 AR C17:0/C21:0 ratio went from 0.22 (0.14) at week 0 to 0.34 (0.15) at week 13 and then decreased slightly, to 0.29 (0.10), at week 26. Comparing the results to ND group in the SYSDIET study, the AR C17:0/C21:0 ratio at week 0 was 0.3 (0.2) but did, however, not change during the intervention, probably due to already high ratio in the Finnish participants at baseline [197]. This might be one reason for the lack of health effects in the SYSDIET study. Good compliance has also been reported in similar intervention studies, conducted in Sweden and Finland where overall rye consumption is higher. In all studies rye products were provided to the participants free of charge, as was done in the present study [159, 163, 165, 178]. RG 80,0 60,0 40,0 CG 20,0 0,0 Week 0 Week 13 Week 26 AR C17:0/C21:0 ratio Total plasma AR, nmol/L 100,0 0,4 0,35 0,3 0,25 0,2 0,15 0,1 0,05 0 RG CG Week 0 Week 13 Week 26 Figure 7. Total plasma alkylresorcinols (AR) concentration and AR C17:0/C21:0 ratio for rye group (RG) and control group (CG) at week 0, 13 and 26 of the intervention 5.4.4. Acceptability The men in the rye group were generally pleased with the rye products and few reported digestive discomfort. Four of the twelve participants in the rye group had experienced digestive discomfort in the first 13 weeks of rye consumption and two participants in the latter 13 weeks (Figure 8). The rye intake tended to be higher for those who claimed they had felt discomfort. The mean rye consumption at 13 weeks for the four men that had experienced discomfort in the first weeks was 130 (13) g/day while it was 94 (42) g/day for those had not experienced any discomfort. Similarly, at week 26, the mean consumption for the two participants that had experienced digestive discomfort in the latter 13 weeks was 123 (9) g/day while it was 91 (31) g/day for those that did not experience any discomfort. The difference in rye consumption between those that did experience discomfort and those that did not was, however, not statistically significant. These results indicate, however, that it might be difficult to increase rye consumption over 100 g/day in this population. However, the participants seemed to adapt to the products, i.e. fewer felt discomfort at week 26 than 13, more participants liked the products and more answered that they would prefer rye products. 58 The results of the acceptability in the present study are similar to the study by Pereira et al [149] where 9 of 11 participants preferred a whole grain diet after a crossover controlled feeding trial with two 6-week feeding periods, whole grain (dietary fiber 28 g/day) vs. refined grain (dietary fiber 17.8 g/day) with no difference between groups in recorded indigestion. Landberg et al [24], also using a crossover design with two 6-week feeding periods but with much higher amounts of whole grain rye (dietary fiber 67 g/day) reported greater dropout, with 7 of 24 participants did not finishing the study because of constipation and 17 participants reporting mild constipation. This indicates that some men may not tolerate such high intakes of dietary fiber, at least not without a longer adaption period. Six participants answered positively at week 13 when asked if they liked the rye products and eight at week 26. At week 26 no participant disliked the products. Similarly, at week 26 eight participants stated that they would prefer products that contain rye more than other products and no participants disagreed (Figure 8). a) 10 b) 10 13 w 26 w 6 4 13 w 26 w 26 w 2 agree 6 13 w 4 13 w 26 w 13 w 2 13 w 0 26 w 8 Participants, n Participants, n 8 26 w 0 neither nor disagree agree neither nor disagree Participants, n c) 10 26 w 8 6 13 w 13 w 4 26 w 2 13 w 26 w 0 agree neither nor disagree Figure 8. The rye group´s acceptability of the rye products (n = 12) assessed by three questions at week 13 (13 w) and week 26 (26 w): a) I have experienced digestive discomfort after I started eating rye products b) I like food products that contain rye and c) I would prefer products that contain rye more than other products 59 6. Strengths and limitations The strengths of the SYSDIET study include reliable and comparable methods that were used in six different centers in four Nordic countries. The same study protocol was used in all centers and analyses of plasma AR were done centrally in Uppsala, Sweden. The SYSDIET study was, however, designed to reveal mechanisms by which Nordic foods and diets could be modified to promote health and prevent chronic diseases. Thus, it was not designed specifically to assess the effects of whole grain intake on health outcomes but of the overall Nordic diet. This is a limitation to the present analyses of the SYSDIET study as confounding effects from other food sources cannot be completely excluded and analyses were partly cross-sectional. Information on the intake of food was not available from the study centers but detailed information on whole grain and/or whole grain products, and other food products, would have strengthened the conclusions in the present analyses. Nevertheless, the results suggest that participants were compliant with the whole grain intake recommended (at least whole grain rye and wheat). The strengths of the ARI feasibility study are that the population studied had habitual low intake of whole grain and fiber and the results are of high importance for the design of future studies. Additionally, a rye intervention with patients with type 2 diabetes has not been done before. The number of participants was, however, low and resulted in power issues for analyses of health effects that would have been an interesting addition to the study. In both studies plasma AR were used as biomarkers for whole grain wheat and rye intake instead of (SYSDIET), and in conjunction with (ARI feasibility study), self-reported estimates. The suitability of AR as a biomarker for whole grain wheat and rye intake has been shown in many studies, and at present it appears to be the best option. The use of dietary biomarkers instead of, or preferably in conjunction with, self-reported intake to measure compliance or diet-disease relationships is an approach that should be further studied and utilized in future research. It is, however, important to note that the half-life of plasma AR is short and therefore it may not be suitable at irregular and small intakes and although the reproducibility has been reported to be good in intervention settings it is somewhat lower in free-living subjects. Nevertheless, plasma AR can be used to distinguish between high and very low consumers of whole grain rye and wheat in epidemiological studies and as a compliance marker in dietary interventions, including whole grain wheat and rye intake. AR is a dietary biomarker for whole grain wheat and rye intake, not other whole grains, such as oats, and barley contains only a small amount compared to wheat and rye. Additionally, the AR ratio C17:0/C21:0 only reflects relative intake of whole grain rye, not absolute intake, but where intake is high and stable it reflects absolute intake relatively well. Other markers are therefore warranted to monitor whole grain intake from other cereal sources. To date, no such markers have, however, been identified or validated. 60 7. Conclusion The results of the present study suggest that fasting total plasma AR concentration may be used as a biomarker in a population consuming the healthy Nordic diet, in which whole grain wheat and rye are important contributors. Hence, the results further support the proposed use of fasting plasma AR as a biomarker of whole grain wheat and rye intake, especially in populations where whole grain wheat and rye intake is high and relatively consistent. The results of the present study also support the use of the ratio of the AR homologues C17:0 and C21:0 as an indicator of relative rye intake in intervention studies where a diet rich in whole grain rye is being investigated. Although the ratio reflects only the relative whole grain/bran rye consumption of total whole grain wheat and rye intake the ratio may reflect total rye intake reasonably well where rye intake is high in relation to other sources of AR. In cross-sectional settings the AR C17:0/C21:0 ratio was inversely associated fasting insulin concentration and positively associated with insulin sensitivity indicating that increased proportion of whole grain rye in the diet may be associated with lower fasting insulin concentration and increased insulin sensitivity. Increased proportion of whole grain rye, reflected by AR C17:0/C21:0, was also associated with favorable outcomes in blood lipid concentrations. The baseline dietary fiber intake and total plasma AR differed substantially between study centers with the highest fiber and plasma AR concentration in Finland and the lowest in Iceland. The feasibility study, carried out in Iceland, produced valuable information that can be used to design a randomized controlled trial aimed to assess the health effects of high rye consumption in a population not accustomed to high intakes. The dropout-rate in the feasibility study was low and the good compliance and in general good acceptability of the rye products showed that the intervention was well received by the population studied. 61 8. Future perspectives The present study provides useful knowledge regarding the use of plasma AR as a biomarker for whole grain wheat and rye intake as a part of the healthy Nordic diet and the utilization of plasma AR as a biomarker in nutritional science. Plasma AR is a highly promising biomarker for whole grain wheat and rye intake but still there are many factors that are poorly understood. The validity of plasma AR as a biomarker for various amounts of whole grain intake needs to be determined, especially at low intakes and in non-European populations in order to assess their applicability across different range/types of whole grain intake. The validity of plasma AR against other assessment methods in free-living subjects needs to be evaluated. The potential use of plasma AR measurements in combination with traditional dietary assessment methods for intake estimation should be evaluated. For measure of compliance there is still uncertainty around cut-off points for determining when a person has consumed a certain quantity of whole grain per day, or if they are naturally low absorbers/metabolizers of AR. Deeper understanding of the absorption and metabolism of AR is needed with more strict, high-quality interventions on the effects of rye and rye products on changes in plasma AR and on human physiology. Deeper understanding of the absorption and metabolism of the different homologues is needed for more accurate use of C17:0/C21:0 ratio for determining the source of AR consumed. It is important to keep in mind that AR is only a marker of whole grain wheat and rye intake and an objective marker of overall whole grain intake and/or for other types of cereals is desirable. Studies on metabolites have discovered potential markers of certain whole grain food intakes using metabolomics, an interesting field of research. Although rye is the second most consumed cereal in the Nordic countries after wheat, studies on the health effects of whole grain rye are scarce. The results from the ARI feasibility study showed that increased whole grain rye intake was well accepted by Icelandic men but further studies on how to increase compliance in interventions, and in clinical work, is needed. The knowledge gained from the study should be utilized for future research on the effects of whole grain rye consumption, using plasma AR and AR C17:0/C21:0 ratio as a measure of compliance, on insulin sensitivity and, in particular, blood lipids, which is an area of research that is very limited. 62 9. Acknowledgements The present work was carried out at the Unit for nutrition research, Faculty of Food Science and Nutrition, School of Health Science, University of Iceland and Landspítali – The National University Hospital of Iceland. Funding was provided by Nordforsk NcoE in Food, Nutrition and Health: Project 070014 SYSDIET (Systems biology in controlled dietary interventions and cohort studies) and project HELGA: Nordic Health – Whole Grain Food, the Consortium Agreement for Consortium on Health and Ageing Network of Cohorts in Europe and the United States (CHANCES) (EG Grant agreement No: 242244) and the Icelandic Research Fund for Graduate Students (1206880061). I would never have been able to complete this huge task on my own so in the following paragraphs I would like to express gratitude to all the people who played a part. First of all I would like to thank my hard working supervisors, Inga Þórsdóttir and Ingibjörg Gunnarsdóttir, for giving me this opportunity and for the guidance and support through this process. Furthermore I would like to thank Laufey Steingrímsdóttir, who initiated and coordinated the ARI feasibility study and Þórhallur Ingi Halldórsson for his advice, statistics lessons and reading over my thesis. I want to thank all my co-workers at the Unit for Nutrition Research and the team of dietitians at Landspitali National-University Hospital, you made the work so much more fun. It is a pleasure to have worked with such professional, practical and experienced individuals. I want to thank those that took part in the data collection. Svandís Erna Jónsdóttir who together with me collected the data in the SYSDIET study and I also thank Laufey Hrólfsdóttir and Auður Benediktsdóttir who worked with me to collect data in the ARI feasibility study. My endless gratitude goes also to the participants of the studies as it goes without saying that without their contribution the research could not have been conducted. I would furthermore like to thank my co-authors of the papers, who are all outstanding scientists, for providing their useful contributions and comments. I especially want to thank Rikard Landberg, who has been my “alkylresorcinols mentor“. He has been extremely helpful and willing to share his knowledge and experience and he has served me incredibly well throughout this work. I would like to thank my dear friends and fellow PhD students, Ása Vala, Cindy, Ólöf Helga, Svandís and Tinna. It is hard to describe it in words how much fun I´ve had with you these years and your friendship alone made all the work worth it. Cindy gets special thanks for the support during the last year, reading over the thesis and answering endless English grammar questions and Svandís for mental support over these stressful last couple of months. Finally I want to thank my family. I thank my parents and my parents in law for all their love, support and the faith they have in me and everything I do, kinder and more generous people are hard to find. I am forever grateful to my wonderful husband, Guðmundur, for his endless patience and encouragement throughout this journey and always believing in me. Last, but certainly not least, I would like to thank my two daughters, Þórunn and Hrafnhildur, for making me smile and reminding me every day what truly matters in life. Reykjavik, September 2014 Óla Kallý Magnúsdóttir 63 10. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Mellen, P.B., Walsh, T.F., Herrington, D.M. Whole grain intake and cardiovascular disease: a meta-analysis. Nutr Metab Cardiovasc Dis, 2008. 18(4): p. 283-90. McKeown, N.M., Meigs, J.B., Liu, S., Wilson, P.W. , Jacques, P.F. Whole-grain intake is favorably associated with metabolic risk factors for type 2 diabetes and cardiovascular disease in the Framingham Offspring Study. Am J Clin Nutr, 2002. 76(2): p. 390-8. Anderson, J.W., T.J. Hanna, X. Peng, and R.J. Kryscio, Whole grain foods and heart disease risk. J Am Coll Nutr, 2000. 19(3 Suppl): p. 291S-299S. Jonnalagadda, S.S., Harnack, L., Liu, R.H., McKeown, N., Seal, C., Liu, S., Fahey, G.C. Putting the whole grain puzzle together: health benefits associated with whole grains--summary of American Society for Nutrition 2010 Satellite Symposium. J Nutr, 2011. 141(5): p. 1011s-22s. Brownlee, I.A., Moore, C., Chatfield, M., Richardson, D.P., Ashby, P., Kuznesof, S.A., Jebb, S.A., Seal, C.J. Markers of cardiovascular risk are not changed by increased whole-grain intake: the WHOLEheart study, a randomised, controlled dietary intervention. Br J Nutr, 2010. 104(1): p. 125-34. Fraser, G.E., Butler, T.L., Shavlik, D. Correlations between estimated and true dietary intakes: using two instrumental variables. Ann Epidemiol, 2005. 15(7): p. 509-18. Lissner, L., Troiano, R.P., Midthune, D., Heitmann, B.L., Kipnis, V., Subar, A.F., Potischman, N. OPEN about obesity: recovery biomarkers, dietary reporting errors and BMI. Int J Obes (Lond), 2007. 31(6): p. 956-61. Bingham, S., Luben, R., Welch, A., Tasevska, N., Wareham, N., Khaw, K.T. Epidemiologic assessment of sugars consumption using biomarkers: comparisons of obese and nonobese individuals in the European prospective investigation of cancer Norfolk. Cancer Epidemiol Biomarkers Prev, 2007. 16(8): p. 1651-4. Horner, N.K., Patterson, R.E., Neuhouser, M.L., Lampe, J.W., Beresford, S.A., Prentice, R.L. Participant characteristics associated with errors in self-reported energy intake from the Women's Health Initiative food-frequency questionnaire. Am J Clin Nutr, 2002. 76(4): p. 766-73. Bingham, S.A. Biomarkers in nutritional epidemiology. Public Health Nutr, 2002. 5(6a): p. 821-7. Seal, C.J., Jones, A.R., Whitney, A.D. Whole grains uncovered. Nutrition Bulletin, 2006. 31(2): p. 129-137. Ross, A.B. Present status and perspectives on the use of alkylresorcinols as biomarkers of wholegrain wheat and rye intake. J Nutr Metab, 2012. 2012: p. 462967. Ferruzzi, M.G., Jonnalagadda, S.S., Liu, S., Marquart, L., McKeown, N., Reicks, M., Riccardi, G., Seal, C., Slavin, J., Thielecke, F., van der Kamp, J.-W., Webb, D. Developing a Standard Definition of Whole-Grain Foods for Dietary Recommendations: Summary Report of a Multidisciplinary Expert Roundtable Discussion. Advances in Nutrition: An International Review Journal, 2014. 5(2): p. 164-176. 64 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. Kuhnle, G.G. Nutritional biomarkers for objective dietary assessment. J Sci Food Agric, 2012. 92(6): p. 1145-9. Prentice, R.L., Sugar, E., Wang, C.Y., Neuhouser, M., Patterson, R. Research strategies and the use of nutrient biomarkers in studies of diet and chronic disease. Public Health Nutr, 2002. 5(6A): p. 977-84. Gibson, R.S. Principles of nutritional assessment. 2nd ed. 2005, New York: Oxford University Press. xx, 908 p. Alfthan G. Nordic Biomarker Seminar. Nordic Council of Ministers, Tema Nord. 2005. 554: p. 1-6. Landberg, R., Kamal-Eldin, A., Andersson, A., Vessby, B., Aman, p. Alkylresorcinols as biomarkers of whole-grain wheat and rye intake: plasma concentration and intake estimated from dietary records. Am J Clin Nutr, 2008. 87(4): p. 832-8. Ross, A.B., Bourgeois, A., Macharia, H.N., Kochhar, S., Jebb, S.A., Brownlee, I.A., Seal, C.J. Plasma alkylresorcinols as a biomarker of whole-grain food consumption in a large population: results from the WHOLEheart Intervention Study. Am J Clin Nutr, 2012. 95(1): p. 204-11. Ross, A.B., Kamal-Eldin, A., Aman, P. Dietary alkylresorcinols: absorption, bioactivities, and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr Rev, 2004. 62(3): p. 81-95. Aubertin-Leheudre, M., Koskela, A., Marjamaa, A., Adlercreutz, H. Plasma alkylresorcinols and urinary alkylresorcinol metabolites as biomarkers of cereal fiber intake in Finnish women. Cancer Epidemiol Biomarkers Prev, 2008. 17(9): p. 2244-8. Montonen, J., Landberg, R., Kamal-Eldin, A., Aman, P., Knueppel, S., Boeing, H., Pischon, T. Reliability of fasting plasma alkylresorcinol concentrations measured 4 months apart. Eur J Clin Nutr, 2010. 64(7): p. 698-703. Landberg, R., Aman, P., Friberg, L.E., Vessby, B., Adlercreutz, H., Kamal-Eldin, A. Dose response of whole-grain biomarkers: alkylresorcinols in human plasma and their metabolites in urine in relation to intake. Am J Clin Nutr, 2009. 89(1): p. 290-6. Landberg, R., Kamal-Eldin, A., Andersson, S.O., Johansson, J.E., Zhang, J.X., Hallmans, G., Aman, P. Reproducibility of plasma alkylresorcinols during a 6-week rye intervention study in men with prostate cancer. J Nutr, 2009. 139(5): p. 975-80. Linko-Parvinen, A.M., Landberg, R., Tikkanen, M.J., Adlercreutz, H., Penalvo, J.L. Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. J Nutr, 2007. 137(5): p. 1137-42. Ross, A.B., Pineau, N., Kochhar, S., Bourgeois, A., Beaumont, M., Decarli, B. Validation of a FFQ for estimating whole-grain cereal food intake. Br J Nutr, 2009. 102(11): p. 1547-51. Ocke, M.C., Kaaks, R.J. Biochemical markers as additional measurements in dietary validity studies: application of the method of triads with examples from the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr, 1997. 65(4 Suppl): p. 1240s-1245s. Chen, Y., Ross, A.B., Aman, P., Kamal-Eldin, A. Alkylresorcinols as markers of whole grain wheat and rye in cereal products. J Agric Food Chem, 2004. 52(26): p. 8242-6. 65 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. Hedrick, V.E., Dietrich, A.M., Estabrooks, P.A., Savla, J., Serrano, E., Davy, B.M. Dietary biomarkers: advances, limitations and future directions. Nutr J, 2012. 11: p. 109. World Health Organization (WHO), 2008-2013 Action Plan for the Global Strategy for the Prevention and Control of Noncommunicable Diseases. 2009, World Health Organization (WHO). p. 1-42. Bere, E., Brug, J. Towards health-promoting and environmentally friendly regional diets - a Nordic example. Public Health Nutr, 2009. 12(1): p. 91-6. Popkin, B.M. Global nutrition dynamics: the world is shifting rapidly toward a diet linked with noncommunicable diseases. Am J Clin Nutr, 2006. 84(2): p. 289-98. Nordic Nutrition Recommendations 2012. Integrating nutrition and physical activity. 5th edition. 2014, Copenhagen: Nordic Council of Ministers. 627. Eaton, S.B., Strassman, B.I., Nesse, R.M., Neel, J.V., Ewald, P.W., Williams, G.C. Weder, A.B., Eaton 3rd, S.B., Lindeberg, S., Konner, M.J., Mysterud, I. Cordain, L. Evolutionary health promotion. Prev Med, 2002. 34(2): p. 109-18. Kourlaba, G., Panagiotakos, D.B. Dietary quality indices and human health: a review. Maturitas, 2009. 62(1): p. 1-8. Lopez-Garcia, E., Schulze, M.B., Fung, T.T., Meigs, J.B., Rifai, N., Manson, J.E., Hu, F.B. Major dietary patterns are related to plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr, 2004. 80(4): p. 1029-35. Salas-Salvado, J., Garcia-Arellano, A., Estruch, R., Marquez-Sandoval, F., Corella, D., Fiol, M., Gomez-Gracia, E., Vinoles, E., Aros, F., Herrera, C., Lahoz, C., Lapetra, J., Perona, J.S., Munoz-Aguado, D., Martinez-Gonzalez, M.A., Ros, E. Components of the Mediterranean-type food pattern and serum inflammatory markers among patients at high risk for cardiovascular disease. Eur J Clin Nutr, 2008. 62(5): p. 651-9. Sofi, F., Cesari, F., Abbate, R., Gensini, G.F., Casini, A. Adherence to Mediterranean diet and health status: meta-analysis. BMJ, 2008. 337: p. a1344. Fung, T.T., Rexrode, K.M., Mantzoros, C.S., Manson, J.E., Willett, W.C., Hu, F.B. Mediterranean diet and incidence of and mortality from coronary heart disease and stroke in women. Circulation, 2009. 119(8): p. 1093-100. Willett, W.C., Sacks, F., Trichopoulou, A., Drescher, G., Ferro-Luzzi, A., Helsing, E., Trichopoulos, D. Mediterranean diet pyramid: a cultural model for healthy eating. Am J Clin Nutr, 1995. 61(6 Suppl): p. 1402s-1406s. Papadaki, A. and J.A. Scott, The impact on eating habits of temporary translocation from a Mediterranean to a Northern European environment. Eur J Clin Nutr, 2002. 56(5): p. 455-61. Saxe, H. The New Nordic Diet is an effective tool in environmental protection: it reduces the associated socioeconomic cost of diets. Am J Clin Nutr, 2014. 99(5): p. 1117-25. Roininen, K., Tuorila, H., Zandstra, E.H., de Graaf, C., Vehkalahti, K., Stubenitsky, K., Mela, D.J. Differences in health and taste attitudes and reported behaviour among Finnish, Dutch and British consumers: a cross-national validation of the Health and Taste Attitude Scales (HTAS). Appetite, 2001. 37(1): p. 33-45. 66 44. 45. 46. 47. 48. 49. 50. 51. Akesson, A., Andersen, L.F., Kristjansdottir, A.G., Roos, E., Trolle, E., Voutilainen, E., Wirfalt, E. Health effects associated with foods characteristic of the Nordic diet: a systematic literature review. Food Nutr Res, 2013. 57. Slimani, N., Fahey, M., Welch, A.A., Wirfalt, E., Stripp, C., Bergstrom, E., Linseisen, J., Schulze, M.B., Bamia, C., Chloptsios, Y., Veglia, F., Panico, S., Bueno-deMesquita, H.B., Ocke, M.C., Brustad, M., Lund, E., Gonzalez, C.A., Barcos, A., Berglund, G., Winkvist, A., Mulligan, A., Appleby, P., Overvad, K., Tjonneland, A., Clavel-Chapelon, F., Kesse, E., Ferrari, P., Van Staveren, W.A., Riboli, E. Diversity of dietary patterns observed in the European Prospective Investigation into Cancer and Nutrition (EPIC) project. Public Health Nutr, 2002. 5(6b): p. 1311-28. Roswall, N., Olsen, A., Boll, K., Christensen, J., Halkjær, J., Sørensen, T.I., Dahm, C.C., Overvad, K., Clavel-Chapelon, F., Boutron-Ruault, M.C., Cottet, V., Teucher, B., Kaaks, R., Boeing, H., von Ruesten, A., Trichopoulou, A., Oikonomou, E., Vasilopoulou, E., Pala, V., Sacerdote, C., Mattiello, A., Masala, G., Peeters, P.H., Bueno-de-Mesquita, H.B., Engeset, D., Skeie, G., Åsli, L.A., Amiano, P., Jakszyn, P., Ardanaz, E., Huerta, J.M., Quirós, J.R., Molina-Montes, E., Nilsson, L.M., Johansson, I., Wirfält, E., Drake, I., Mulligan, A.A., Khaw, K.T., Romaguera, D., Vergnaud, A.C., Key, T., Riboli, E., Tjønneland, A. Consumption of predefined ‘Nordic’ dietary items in ten European countries – an investigation in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Public Health Nutrition, 2014. FirstView: p. 1-10. Kyro, C., Skeie, G., Loft, S., Overvad, K., Christensen, J., Tjonneland, A., Olsen, A. Adherence to a healthy Nordic food index is associated with a lower incidence of colorectal cancer in women: the Diet, Cancer and Health cohort study. Br J Nutr, 2013. 109(5): p. 920-7. Olsen, A., Egeberg, R., Halkjaer, J., Christensen, J., Overvad, K., Tjonneland, A. Healthy aspects of the Nordic diet are related to lower total mortality. J Nutr, 2011. 141(4): p. 639-44. Adamsson, V., Reumark, A., Fredriksson, I.B., Hammarstrom, E., Vessby, B., Johansson, G., Riserus, U. Effects of a healthy Nordic diet on cardiovascular risk factors in hypercholesterolaemic subjects: a randomized controlled trial (NORDIET). J Intern Med, 2011. 269(2): p. 150-9. Brader, L., Uusitupa, M., Dragsted, L.O., Hermansen, K. Effects of an isocaloric healthy Nordic diet on ambulatory blood pressure in metabolic syndrome: a randomized SYSDIET sub-study. Eur J Clin Nutr, 2014. 68(1): p. 57-63. Uusitupa, M., Hermansen, K., Savolainen, M.J., Schwab, U., Kolehmainen, M., Brader, L., Mortensen, L.S., Cloetens, L., Johansson-Persson, A., Onning, G., LandinOlsson, M., Herzig, K.H., Hukkanen, J., Rosqvist, F., Iggman, D., Paananen, J., Pulkki, K.J., Siloaho, M., Dragsted, L., Barri, T., Overvad, K., Bach Knudsen, K.E., Hedemann, M.S., Arner, P., Dahlman, I., Borge, G.I., Baardseth, P., Ulven, S.M., Gunnarsdottir, I., Jonsdottir, S., Thorsdottir, I., Oresic, M., Poutanen, K.S., Riserus, U., Akesson B., Effects of an isocaloric healthy Nordic diet on insulin sensitivity, lipid profile and inflammation markers in metabolic syndrome -- a randomized study (SYSDIET). J Intern Med, 2013. 274(1): p. 52-66. 67 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. Poulsen, S.K., Due, A., Jordy, A.B., Kiens, B., Stark, K.D., Stender, S., Holst, C., Astrup, A., Larsen, T.M. Health effect of the New Nordic Diet in adults with increased waist circumference: a 6-mo randomized controlled trial. Am J Clin Nutr, 2014. 99(1): p. 35-45. Adamsson, V., Reumark, A., Cederholm, T., Vessby, B., Riserus, U., Johansson, G. What is a healthy Nordic diet? Foods and nutrients in the NORDIET study. Food Nutr Res, 2012. 56. Mithril, C., Dragsted, L.O., Meyer, C., Blauert, E., Holt, M.K., Astrup, A. Guidelines for the New Nordic Diet. Public Health Nutr, 2012. 15(10): p. 1941-7. Jonsdottir, S.E., Brader, L., Gunnarsdottir, I., Magnusdottir O.K., Schwab, U., Kolehmainen, M., Riserus, U., Herzig, K.H., Cloetens, L., Helgegren, H., JohanssonPersson, A., Hukkanen, J., Poutanen, K., Uusitupa, M., Hermansen, K., Thorsdottir, I. Adherence to the Nordic Nutrition Recommendations in a Nordic population with metabolic syndrome: high salt consumption and low dietary fibre intake (The SYSDIET study). Food Nutr Res, 2013. 57. Mithril, C., Dragsted, L.O., Meyer, C., Tetens, I., Biltoft-Jensen, A., Astrup, A. Dietary composition and nutrient content of the New Nordic Diet. Public Health Nutr, 2013. 16(5): p. 777-85. Nordic Nutrition Recommendations 2004. Intergrating nutrition and physical activity. 2004, Copenhagen: Nordic Council of Ministers. Spiller, G.A. Whole grains, whole wheat abd white flours in history, in Whole-Grain Foods in Health and Disease, L. Marquart, J.L. Slavin, and R.G. Fulcher, Editors. 2002, Eagan Press: St. Paul, Minnesota, USA. p. 1-7. AACC. Definition of Whole Grain. 1999 [cited 2014 1 July]; Available from: http://www.aaccnet.org/initiatives/definitions/Pages/WholeGrain.aspx. FDA. Guidance for industry and FDA staff: Whole grain label statements. Draft Guidance 2006 [cited 2014 1. July]; Available from: http://www.fda.gov/ohrms/dockets/98fr/06d-0066-gdl0001.pdf. Asp, N.G., Poutanen, K., Richardson, D.P., van der Kamp, J.W. HEALTHGRAIN Consortium whole grain definition. 2013 May 2013 [cited 2014 1. July]; Available from: https://www.healthgrain.org/webfm_send/601. Jacobs Jr, D.R., Meyer, K.A., Kushi, L.H., Folsom, A.R. Whole-grain intake may reduce the risk of ischemic heart disease death in postmenopausal women: the Iowa Women's Health Study. Am J Clin Nutr, 1998. 68(2): p. 248-57. Thane, C.W., Jones, A.R., Stephen, A.M., Seal, C.J., Jebb, S.A. Whole-grain intake of British young people aged 4-18 years. Br J Nutr, 2005. 94(5): p. 825-31. Liu, S., Manson, J.E., Stampfer, M.J., Rexrode, K.M., Hu, F.B., Rimm, E.B., Willett, W.C. Whole grain consumption and risk of ischemic stroke in women: A prospective study. Jama, 2000. 284(12): p. 1534-40. McKeown, N.M., Troy, L.M., Jacques, P.F., Hoffmann, U., O'Donnell, C.J., Fox, C.S. Whole- and refined-grain intakes are differentially associated with abdominal visceral and subcutaneous adiposity in healthy adults: the Framingham Heart Study. Am J Clin Nutr, 2010. 92(5): p. 1165-71. 68 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. AACCI. AACCI’s Whole Grains Working Group Unveils New Whole Grain Products Characterization. 2013 [cited 2014 2. July 2014]; Available from: http://www.aaccnet.org/about/newsreleases/Pages/WholeGrainProductCharacterizatio n.aspx. Livsmedelsverket. The Keyhole Symbol. 2007 [cited 2014 August 13]; Available from: http://www.slv.se/en-gb/Group1/Food-labelling/Keyhole-symbol/. Matvælastofnun. Hvaða matvælaflokkar mega bera Skráargatið? [cited 2014 August 13]; Available from: http://www.mast.is/matvaeli/skraargat/matvaelaflokkar/. Andersson, R., Fransson, G., Tietjen, M., Aman, P. Content and molecular-weight distribution of dietary fiber components in whole-grain rye flour and bread. J Agric Food Chem, 2009. 57(5): p. 2004-8. Wolever, T.M., Tosh, S.M., Gibbs, A.L., Brand-Miller, J., Duncan, A.M., Hart, V., Lamarche, B., Thomson, B.A., Duss, R., Wood, P.J. Physicochemical properties of oat beta-glucan influence its ability to reduce serum LDL cholesterol in humans: a randomized clinical trial. Am J Clin Nutr, 2010. 92(4): p. 723-32. Frølich, W., Aman, P., Tetens, I. Whole grain foods and health - a Scandinavian perspective. Food Nutr Res, 2013. 57. FAO Faostat. 2011 [cited 2012 13 Jan]. Available from: http://faostat.fao.org. Amcoff, E., Edberg, A., Barbieri, H.E., Lindroos, A.K., Nälsén, C., Pearson, M., Lemming, E.W. Riksmaten – vuxna 2010–11. Livsmedels- och näringsintag bland vuxna i Sverige. 2012, Livsmedelsverket: Uppsala, Sweden. Mejborn, H., Ygil, K.H., Fagt, S., Trolle, E., Christensen, T. Wholegrain intake of Danes 2011-2012. 2013, DTU Fødevareinstituttet. Kyro, C., Skeie, G., Dragsted, L.O., Christensen, J., Overvad, K., Hallmans, G., Johansson, I., Lund, E., Slimani, N., Johnsen, N.F., Halkjaer, J., Tjonneland, A., Olsen, A. Intake of whole grain in Scandinavia: intake, sources and compliance with new national recommendations. Scand J Public Health, 2012. 40(1): p. 76-84. Jonsson, G. Changes in food consumption in Iceland, 1770–1940. Scandinavian Economic History Review, 1998. 46(1): p. 24-41. Steingrimsdottir, L., Thorgeirsdottir, H., Olafsdottir, A. The Icelandic National Nutrition Survey 2002. 2003, Public Health Institute: Reykjavík. Butt, M., Tahir-Nadeem, M., Khan, M., Shabir, R., Butt, M. Oat: unique among the cereals. European Journal of Nutrition, 2008. 47(2): p. 68-79. Nystrom, L., Lampi, A.M., Andersson, A.A., Kamal-Eldin, A., Gebruers, K., Courtin, C.M., Delcour, J.A., Li, L., Ward, J.L., Fras, A., Boros, D., Rakszegi, M., Bedo, Z., Shewry, P.R., Piironen, V. Phytochemicals and dietary fiber components in rye varieties in the HEALTHGRAIN Diversity Screen. J Agric Food Chem, 2008. 56(21): p. 9758-66. Shewry, P.R., Piironen, V., Lampi, A.M., Nystrom, L., Li, L., Rakszegi, M., Fras, A., Boros, D., Gebruers, K., Courtin, C.M., Delcour, J.A., Andersson, A.A., Dimberg, L., Bedo, Z., Ward, J.L. Phytochemical and fiber components in oat varieties in the HEALTHGRAIN Diversity Screen. J Agric Food Chem, 2008. 56(21): p. 9777-84. Andersson, A.A., Lampi, A.M., Nystrom, L., Piironen, V., Li, L., Ward, J.L., Gebruers, K., Courtin, C.M., Delcour, J.A., Boros, D., Fras, A., Dynkowska, W., 69 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. Rakszegi, M., Bedo, Z., Shewry, P.R., Aman, P. Phytochemical and dietary fiber components in barley varieties in the HEALTHGRAIN Diversity Screen. J Agric Food Chem, 2008. 56(21): p. 9767-76. Rakha, A., Åman, P., Andersson, R. Characterisation of dietary fibre components in rye products. Food Chemistry, 2010. 119(3): p. 859-867. Hansen, H.B., Rasmussen, C.V., Bach Knudsen, K.E., Hansen, Å. Effects of genotype and harvest year on content and composition of dietary fibre in rye (Secale cereale L) grain. Journal of the Science of Food and Agriculture, 2003. 83(1): p. 76-85. Willett, W. Nutritional epidemiology. 2013, Oxford; New York: Oxford University Press. Jenab, M., Slimani, N., Bictash, M., Ferrari, P., Bingham, S.A. Biomarkers in nutritional epidemiology: applications, needs and new horizons. Hum Genet, 2009. 125(5-6): p. 507-25. Bingham, S.A. Urine nitrogen as a biomarker for the validation of dietary protein intake. J Nutr, 2003. 133 Suppl 3: p. 921s-924s. Kaaks, R.J. Biochemical markers as additional measurements in studies of the accuracy of dietary questionnaire measurements: conceptual issues. Am J Clin Nutr, 1997. 65(4 Suppl): p. 1232S-1239S. Livingstone, M.B., Black, A.E. Markers of the validity of reported energy intake. J Nutr, 2003. 133 Suppl 3: p. 895s-920s. Bogers, R.P., Dagnelie, P.C., Westerterp, K.R., Kester, A.D., van Klaveren, J.D., Bast, A., van den Brandt, P.A. Using a correction factor to correct for overreporting in a food-frequency questionnaire does not improve biomarker-assessed validity of estimates for fruit and vegetable consumption. J Nutr, 2003. 133(4): p. 1213-9. Kaaks, R., Riboli, E., Sinha, R. Biochemical markers of dietary intake. IARC Sci Publ, 1997(142): p. 103-26. Bingham, S., Luben, R., Welch, A., Low, Y.L., Khaw, K.T., Wareham, N., Day, N. Associations between dietary methods and biomarkers, and between fruits and vegetables and risk of ischaemic heart disease, in the EPIC Norfolk Cohort Study. Int J Epidemiol, 2008. 37(5): p. 978-87. Grace, P.B., Taylor, J.I., Low, Y.L., Luben, R.N., Mulligan, A.A., Botting, N.P., Dowsett, M., Welch, A.A., Khaw, K.T., Wareham, N.J., Day, N.E., Bingham, S.A. Phytoestrogen concentrations in serum and spot urine as biomarkers for dietary phytoestrogen intake and their relation to breast cancer risk in European prospective investigation of cancer and nutrition-norfolk. Cancer Epidemiol Biomarkers Prev, 2004. 13(5): p. 698-708. Key, T.J., Appleby, P.N., Allen, N.E., Travis, R.C., Roddam, A.W., Jenab, M., Egevad, L., Tjonneland, A., Johnsen, N.F., Overvad, K., Linseisen, J., Rohrmann, S., Boeing, H., Pischon, T., Psaltopoulou, T., Trichopoulou, A., Trichopoulos, D., Palli, D., Vineis, P., Tumino, R., Berrino, F., Kiemeney, L., Bueno-de-Mesquita, H.B., Quiros, J.R., Gonzalez, C.A., Martinez, C., Larranaga, N., Chirlaque, M.D., Ardanaz, E., Stattin, P., Hallmans, G., Khaw, K.T., Bingham, S., Slimani, N., Ferrari, P., Rinaldi, S., Riboli, E. Plasma carotenoids, retinol, and tocopherols and the risk of 70 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. prostate cancer in the European Prospective Investigation into Cancer and Nutrition study. Am J Clin Nutr, 2007. 86(3): p. 672-81. Ward, H., Chapelais, G., Kuhnle, G.G., Luben, R., Khaw, K.T., Bingham, S. Breast cancer risk in relation to urinary and serum biomarkers of phytoestrogen exposure in the European Prospective into Cancer-Norfolk cohort study. Breast Cancer Res, 2008. 10(2): p. R32. Marklund, M., Magnusdottir, O.K., Rosqvist, F., Cloetens, L., Landberg, R., Kolehmainen, M., Brader, L., Hermansen, K., Poutanen, K.S., Herzig, K.H., Hukkanen, J., Savolainen, M.J., Dragsted, L.O., Schwab, U., Paananen, J., Uusitupa, M., Akesson, B., Thorsdottir, I., Riserus, U. A Dietary Biomarker Approach Captures Compliance and Cardiometabolic Effects of a Healthy Nordic Diet in Individuals with Metabolic Syndrome. J Nutr, 2014. Jul 30. [Epub ahead of print] Kristensen, M., Toubro, S., Jensen, M.G., Ross, A.B., Riboldi, G., Petronio, M., Bügel, S., Tetens, I., Astrup, A. Whole grain compared with refined wheat decreases the percentage of body fat following a 12-week, energy-restricted dietary intervention in postmenopausal women. Journal of Nutrition, 2012. 142(4): p. 710-716. Knudsen, M.D., Kyro, C., Olsen, A., Dragsted, L.O., Skeie, G., Lund, E., Aman, P., Nilsson, L.M., Bueno-de-Mesquita, H.B., Tjonneland, A., Landberg, R. Self-reported whole-grain intake and plasma alkylresorcinol concentrations in combination in relation to the incidence of colorectal cancer. Am J Epidemiol, 2014. 179(10): p. 1188-96. Baldrick, F.R., Woodside, J.V., Elborn, J.S., Young, I.S., McKinley, M.C. Biomarkers of fruit and vegetable intake in human intervention studies: a systematic review. Crit Rev Food Sci Nutr, 2011. 51(9): p. 795-815. Bjermo, H., Iggman, D., Kullberg, J., Dahlman, I., Johansson, L., Persson, L., Berglund, J., Pulkki, K., Basu, S., Uusitupa, M., Rudling, M., Arner, P., Cederholm, T., Ahlstrom, H., Riserus, U. Effects of n-6 PUFAs compared with SFAs on liver fat, lipoproteins, and inflammation in abdominal obesity: a randomized controlled trial. Am J Clin Nutr, 2012. 95(5): p. 1003-12. Jacobs, D.R., Jr., Pereira, M.A., Stumpf, K., Pins, J.J., Adlercreutz, H. Whole grain food intake elevates serum enterolactone. Br J Nutr, 2002. 88(2): p. 111-6. Juntunen, K.S., Mazur, W.M., Liukkonen, K.H., Uehara, M., Poutanen, K.S., Adlercreutz, H.C., Mykkanen, H.M. Consumption of wholemeal rye bread increases serum concentrations and urinary excretion of enterolactone compared with consumption of white wheat bread in healthy Finnish men and women. Br J Nutr, 2000. 84(6): p. 839-46. Kilkkinen, A., Stumpf, K., Pietinen, P., Valsta, L.M., Tapanainen, H., Adlercreutz, H. Determinants of serum enterolactone concentration. Am J Clin Nutr, 2001. 73(6): p. 1094-100. Johnsen, N.F., Hausner, H., Olsen, A., Tetens, I., Christensen, J., Knudsen, K.E., Overvad, K., Tjonneland, A. Intake of whole grains and vegetables determines the plasma enterolactone concentration of Danish women. J Nutr, 2004. 134(10): p. 26917. 71 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. Sonestedt, E., Ericson, U., Gullberg, B., Penalvo, J.L., Adlercreutz, H., Wirfalt, E. Variation in fasting and non-fasting serum enterolactone concentrations in women of the Malmo Diet and Cancer cohort. Eur J Clin Nutr, 2008. 62(8): p. 1005-9. Ross, A.B., Shepherd, M.J., Schupphaus, M., Sinclair, V., Alfaro, B., Kamal-Eldin, A., Aman, P. Alkylresorcinols in cereals and cereal products. J Agric Food Chem, 2003. 51(14): p. 4111-8. Ross, A.B., Kamal-Eldin, A., Lundin, E.A., Zhang, J.X., Hallmans, G., Aman, P. Cereal alkylresorcinols are absorbed by humans. J Nutr, 2003. 133(7): p. 2222-4. Landberg, R., Linko, A.M., Kamal-Eldin, A., Vessby, B., Adlercreutz, H., Aman, P. Human plasma kinetics and relative bioavailability of alkylresorcinols after intake of rye bran. J Nutr, 2006. 136(11): p. 2760-5. Linko, A.M., Adlercreutz, H. Whole-grain rye and wheat alkylresorcinols are incorporated into human erythrocyte membranes. Br J Nutr, 2005. 93(1): p. 11-3. Linko, A.M., Juntunen, K.S., Mykkanen, H.M., Adlercreutz, H. Whole-grain rye bread consumption by women correlates with plasma alkylresorcinols and increases their concentration compared with low-fiber wheat bread. J Nutr, 2005. 135(3): p. 580-3. Linko, A.M., Parikka, K., Wahala, K., Adlercreutz, H. Gas chromatographic-mass spectrometric method for the determination of alkylresorcinols in human plasma. Anal Biochem, 2002. 308(2): p. 307-13. Linko, A.M., Ross, A.B., Kamal-Eldin, A., Serena, A., Kjaer, A.K., Jorgensen, H., Penalvo, J.L., Adlercreutz, H., Aman, P., Knudsen, K.E. Kinetics of the appearance of cereal alkylresorcinols in pig plasma. Br J Nutr, 2006. 95(2): p. 282-7. Hanhineva, K., Keski-Rahkonen, P., Lappi, J., Katina, K., Pekkinen, J., Savolainen, O., Timonen, O., Paananen, J., Mykkanen, H., Poutanen, K. The postprandial plasma rye fingerprint includes benzoxazinoid-derived phenylacetamide sulfates. J Nutr, 2014. 144(7): p. 1016-22. Zarnowski, R., Suzuki, Y., Yamaguchi, I., Pietr, S.J. Alkylresorcinols in barley (Hordeum vulgare L. distichon) grains. Z Naturforsch C, 2002. 57(1-2): p. 57-62. Kulawinek, M., Jaromin, A., Kozubek, A., Zarnowski, R. Alkylresorcinols in selected Polish rye and wheat cereals and whole-grain cereal products. J Agric Food Chem, 2008. 56(16): p. 7236-42. Mattila, P., Pihlava, J.M., Hellstrom, J. Contents of phenolic acids, alkyl- and alkenylresorcinols, and avenanthramides in commercial grain products. J Agric Food Chem, 2005. 53(21): p. 8290-5. Ross, A.B., Kamal-Eldin, A., Jung, C., Shepherd, M.J., Åman, P. Gas chromatographic analysis of alkylresorcinols in rye (Secale cereale L) grains. Journal of the Science of Food and Agriculture, 2001. 81(14): p. 1405-1411. Andersson, A.A., Kamal-Eldin, A., Fras, A., Boros, D., Aman, P. Alkylresorcinols in wheat varieties in the HEALTHGRAIN Diversity Screen. J Agric Food Chem, 2008. 56(21): p. 9722-5. Hengtrakul, P., Lorenz, K., Mathias, M. Alkylresorcinols in U.S. and Canadian wheats and flours. Cereal Chemistry, 1990. 67(5): p. 413-417. 72 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. Ross, A.B., Becker, W., Chen, Y., Kamal-Eldin, A., Aman, P. Intake of alkylresorcinols from wheat and rye in the United Kingdom and Sweden. Br J Nutr, 2005. 94(4): p. 496-9. Cheng, H.H., Lai, M.H., Hou, W.C., Huang, C.L. Antioxidant effects of chromium supplementation with type 2 diabetes mellitus and euglycemic subjects. J Agric Food Chem, 2004. 52(5): p. 1385-9. Ross, A.B., Shepherd, M.J., Bach Knudsen, K.E., Glitso, L.V., Bowey, E., Phillips, J., Rowland, I., Guo, Z.X., Massy, D.J., Aman, P., Kamal-Eldin, A. Absorption of dietary alkylresorcinols in ileal-cannulated pigs and rats. Br J Nutr, 2003. 90(4): p. 787-94. Jansson, E., Landberg, R., Kamal-Eldin, A., Wolk, A., Vessby, B., Aman, P. Presence of alkylresorcinols, potential whole grain biomarkers, in human adipose tissue. Br J Nutr, 2010. 104(5): p. 633-6. Landberg, R., Aman, P., Hallmans, G., Johansson, I. Long-term reproducibility of plasma alkylresorcinols as biomarkers of whole-grain wheat and rye intake within Northern Sweden Health and Disease Study Cohort. Eur J Clin Nutr, 2013. 67(3):25963 Kyro, C., Olsen, A., Bueno-de-Mesquita, H.B., Skeie, G., Loft, S., Aman, P., Leenders, M., Dik, V.K., Siersema, P.D., Pischon, T., Christensen, J., Overvad, K., Boutron-Ruault, M.C., Fagherazzi, G., Cottet, V., Kuhn, T., Chang-Claude, J., Boeing, H., Trichopoulou, A., Naska, A., Oikonomidou, D., Masala, G., Pala, V., Tumino, R., Vineis, P., Mattiello, A., Peeters, P.H., Bakken, T., Weiderpass, E., Asli, L.A., Sanchez, S., Jakszyn, P., Sanchez, M.J., Amiano, P., Huerta, J.M., Barricarte, A., Ljuslinder, I., Palmqvist, R., Khaw, K.T., Wareham, N., Key, T.J., Travis, R.C., Slimani, N., Freisling, H., Ferrari, P., Gunter, M.J., Murphy, N., Riboli, E., Tjonneland, A., Landberg, R. Plasma alkylresorcinol concentrations, biomarkers of whole-grain wheat and rye intake, in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Br J Nutr, 2014. 111(10): p. 1881-90. Landberg, R., Kamal-Eldin, A., Aman, P., Christensen, J., Overvad, K., Tjonneland, A., Olsen, A. Determinants of plasma alkylresorcinol concentration in Danish postmenopausal women. Eur J Clin Nutr, 2011. 65(1): p. 94-101. Wild, C.P., Andersson, C., O'Brien, N.M., Wilson, L., Woods, J.A. A critical evaluation of the application of biomarkers in epidemiological studies on diet and health. Br J Nutr, 2001. 86 Suppl 1: p. S37-53. Andersson, A., Marklund, M., Diana, M., Landberg, R. Plasma alkylresorcinol concentrations correlate with whole grain wheat and rye intake and show moderate reproducibility over a 2- to 3-month period in free-living Swedish adults. J Nutr, 2011. 141(9): p. 1712-8. Ross, A.B., Aman, P. Kamal-Eldin, A. Identification of cereal alkylresorcinol metabolites in human urine-potential biomarkers of wholegrain wheat and rye intake. J Chromatogr B Analyt Technol Biomed Life Sci, 2004. 809(1): p. 125-30. Aubertin-Leheudre, M., Koskela, A., Samaletdin, A., Adlercreutz, H. Plasma alkylresorcinol metabolites as potential biomarkers of whole-grain wheat and rye cereal fibre intakes in women. Br J Nutr, 2010. 103(3): p. 339-43. 73 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. Guyman, L.A., Adlercreutz, H., Koskela, A., Li, L., Beresford, S.A., Lampe, J.W. Urinary 3-(3,5-dihydroxyphenyl)-1-propanoic acid, an alkylresorcinol metabolite, is a potential biomarker of whole-grain intake in a U.S. population. J Nutr, 2008. 138(10): p. 1957-62. Marklund, M., Stromberg, E.A., Hooker, A.C., Hammarlund-Udenaes, M., Aman, P., Landberg, R., Kamal-Eldin, A. Chain Length of Dietary Alkylresorcinols Affects Their In Vivo Elimination Kinetics in Rats. J Nutr, 2013. 143(10):1573-8 Kamal-Eldin, A., Pouru, A., Eliasson, C., Åman, P. Alkylresorcinols as antioxidants: hydrogen donation and peroxyl radical-scavenging effects. Journal of the Science of Food and Agriculture, 2001. 81(3): p. 353-356. Parikka, K., Rowland, I.R., Welch, R.W., Wahala, K. In vitro antioxidant activity and antigenotoxicity of 5-n-alkylresorcinols. J Agric Food Chem, 2006. 54(5): p. 1646-50. Dietary Guidelines for Americans 2010. 2010, U.S. Department of Agriculture and U.S. Department of Health and Human Services: Washington D.C. World Health Organization. W.H.O. Global Strategy on Diet, Physical Activity and Health. 2014 [cited 2014 August 19th]; Available from: http://www.who.int/dietphysicalactivity/diet/en/. Ye, E.Q., Chacko, S.A., Chou, E.L., Kugizaki, M., Liu, S. Greater whole-grain intake is associated with lower risk of type 2 diabetes, cardiovascular disease, and weight gain. J Nutr, 2012. 142(7): p. 1304-13. de Munter, J.S., Hu, F.B., Spiegelman, D., Franz, M., van Dam, R.M. Whole grain, bran, and germ intake and risk of type 2 diabetes: a prospective cohort study and systematic review. PLoS Med, 2007. 4(8): p. e261. Montonen, J., Knekt, P., Jarvinen, R., Aromaa, A., Reunanen, A. Whole-grain and fiber intake and the incidence of type 2 diabetes. Am J Clin Nutr, 2003. 77(3): p. 6229. Fung, T.T., Hu, F.B., Pereira, M.A., Liu, S., Stampfer, M.J., Colditz, G.A., Willett, W.C. Whole-grain intake and the risk of type 2 diabetes: a prospective study in men. Am J Clin Nutr, 2002. 76(3): p. 535-40. Priebe, M.G., van Binsbergen, J.J., de Vos, R., Vonk, R.J. Whole grain foods for the prevention of type 2 diabetes mellitus. Cochrane Database Syst Rev, 2008(1): p. Cd006061. Harland, J.I., Garton, L.E. Whole-grain intake as a marker of healthy body weight and adiposity. Public Health Nutr, 2008. 11(6): p. 554-63. Aune, D., Chan, D.S., Lau, R., Vieira, R., Greenwood, D.C., Kampman, E., Norat, T. Dietary fibre, whole grains, and risk of colorectal cancer: systematic review and doseresponse meta-analysis of prospective studies. Bmj, 2011. 343: p. d6617. Hughes, S.A., Shewry, P.R., Li, L., Gibson, G.R., Sanz, M.L., Rastall, R.A. In vitro fermentation by human fecal microflora of wheat arabinoxylans. J Agric Food Chem, 2007. 55(11): p. 4589-95. Fardet, A. New hypotheses for the health-protective mechanisms of whole-grain cereals: what is beyond fibre? Nutr Res Rev, 2010. 23(1): p. 65-134. Seal, C.J. Whole grains and CVD risk. Proc Nutr Soc, 2006. 65(1): p. 24-34. 74 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. Reaven, G.M. Pathophysiology of insulin resistance in human disease. Physiol Rev, 1995. 75(3): p. 473-86. Hsueh, W.A., Law, R.E. Cardiovascular risk continuum: implications of insulin resistance and diabetes. Am J Med, 1998. 105(1a): p. 4s-14s. Ruige, J.B., Assendelft, W.J., Dekker, J.M., Kostense, P.J., Heine, R.J., Bouter, L.M. Insulin and risk of cardiovascular disease: a meta-analysis. Circulation, 1998. 97(10): p. 996-1001. Pereira, M.A., Jacobs Jr, D.R., Pins, J.J., Raatz, S.K., Gross, M.D., Slavin, J.L. Seaquist, E.R. Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr, 2002. 75(5): p. 848-55. Jarvi, A.E., Karlstrom, B.E., Granfeldt, Y.E., Bjorck, I.E., Asp, N.G., Vessby, B.O. Improved glycemic control and lipid profile and normalized fibrinolytic activity on a low-glycemic index diet in type 2 diabetic patients. Diabetes Care, 1999. 22(1): p. 108. Frost, G., Keogh, B., Smith, D., Akinsanya, K., Leeds, A. The effect of low-glycemic carbohydrate on insulin and glucose response in vivo and in vitro in patients with coronary heart disease. Metabolism, 1996. 45(6): p. 669-72. Frost, G., Leeds, A., Trew, G., Margara, R., Dornhorst, A. Insulin sensitivity in women at risk of coronary heart disease and the effect of a low glycemic diet. Metabolism, 1998. 47(10): p. 1245-51. Keenan, J.M., Pins, J.J., Frazel, C., Moran, A., Turnquist, L. Oat ingestion reduces systolic and diastolic blood pressure in patients with mild or borderline hypertension: a pilot trial. J Fam Pract, 2002. 51(4): p. 369. Slavin, J. Why whole grains are protective: biological mechanisms. Proc Nutr Soc, 2003. 62(1): p. 129-34. Slavin, J. Whole grains and human health. Nutr Res Rev, 2004. 17(1): p. 99-110. Tighe, P., Duthie, G., Vaughan, N., Brittenden, J., Simpson, W.G., Duthie, S., Mutch, W., Wahle, K., Horgan, G., Thies, F. Effect of increased consumption of whole-grain foods on blood pressure and other cardiovascular risk markers in healthy middleaged persons: a randomized controlled trial. Am J Clin Nutr, 2010. 92(4): p. 733-40. Giacco, R., Lappi, J., Costabile, G., Kolehmainen, M., Schwab, U., Landberg, R., Uusitupa, M., Poutanen, K., Pacini, G., Rivellese, A.A., Riccardi, G. Mykkanen, H. Effects of rye and whole wheat versus refined cereal foods on metabolic risk factors: A randomised controlled two-centre intervention study. Clin Nutr, 2013. 32(6):941-9 Giacco, R., Clemente, G., Cipriano, D., Luongo, D., Viscovo, D., Patti, L., Di Marino, L., Giacco, A., Naviglio, D., Bianchi, M.A., Ciati, R., Brighenti, F., Rivellese, A.A., Riccardi, G. Effects of the regular consumption of wholemeal wheat foods on cardiovascular risk factors in healthy people. Nutr Metab Cardiovasc Dis, 2010. 20(3): p. 186-94. Landberg, R., Andersson, S.O., Zhang, J.X., Johansson, J.E., Stenman, U.H., Adlercreutz, H., Kamal-Eldin, A., Aman, P., Hallmans, G. Rye whole grain and bran intake compared with refined wheat decreases urinary C-peptide, plasma insulin, and prostate specific antigen in men with prostate cancer. J Nutr, 2010. 140(12): p. 21806. 75 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. Andersson, A., Tengblad, S., Karlstrom, B., Kamal-Eldin, A., Landberg, R., Basu, S., Aman, P., Vessby, B. Whole-grain foods do not affect insulin sensitivity or markers of lipid peroxidation and inflammation in healthy, moderately overweight subjects. J Nutr, 2007. 137(6): p. 1401-7. Heinonen, M.V., Karhunen, L.J., Chabot, E.D., Toppinen, L.K., Juntunen, K.S., Laaksonen, D.E., Siloaho, M., Liukkonen, K.H., Herzig, K.H., Niskanen, L.K., Mykkanen, H.M. Plasma ghrelin levels after two high-carbohydrate meals producing different insulin responses in patients with metabolic syndrome. Regul Pept, 2007. 138(2-3): p. 118-25. Leinonen, K., Liukkonen, K., Poutanen, K., Uusitupa, M., Mykkanen, H. Rye bread decreases postprandial insulin response but does not alter glucose response in healthy Finnish subjects. Eur J Clin Nutr, 1999. 53(4): p. 262-7. Juntunen, K.S., Laaksonen, D.E., Poutanen, K.S., Niskanen, L.K. Mykkanen, H.M. High-fiber rye bread and insulin secretion and sensitivity in healthy postmenopausal women. Am J Clin Nutr, 2003. 77(2): p. 385-91. Juntunen, K.S., Laaksonen, D.E., Autio, K., Niskanen, L.K., Holst, J.J., Savolainen, K.E., Liukkonen, K.H., Poutanen, K.S., Mykkanen, H.M. Structural differences between rye and wheat breads but not total fiber content may explain the lower postprandial insulin response to rye bread. Am J Clin Nutr, 2003. 78(5): p. 957-64. Laaksonen, D.E., Toppinen, L.K., Juntunen, K.S., Autio, K., Liukkonen, K.H., Poutanen, K.S., Niskanen, L., Mykkanen, H.M. Dietary carbohydrate modification enhances insulin secretion in persons with the metabolic syndrome. Am J Clin Nutr, 2005. 82(6): p. 1218-27. Hallfrisch, J., Facn, Behall, K.M. Mechanisms of the effects of grains on insulin and glucose responses. J Am Coll Nutr, 2000. 19(3 Suppl): p. 320S-325S. Yu, K., Ke, M.Y., Li, W.H., Zhang, S.Q., Fang, X.C. The impact of soluble dietary fibre on gastric emptying, postprandial blood glucose and insulin in patients with type 2 diabetes. Asia Pac J Clin Nutr, 2014. 23(2): p. 210-8. Fung, T.T., Manson, J.E., Solomon, C.G., Liu, S., Willett, W.C., Hu, F.B. The association between magnesium intake and fasting insulin concentration in healthy middle-aged women. J Am Coll Nutr, 2003. 22(6): p. 533-8. Gross, L.S., Li, L., Ford, E.S., Liu, S. Increased consumption of refined carbohydrates and the epidemic of type 2 diabetes in the United States: an ecologic assessment. Am J Clin Nutr, 2004. 79(5): p. 774-9. Marshall, J.A., Bessesen, D.H., Hamman, R.F. High saturated fat and low starch and fibre are associated with hyperinsulinaemia in a non-diabetic population: the San Luis Valley Diabetes Study. Diabetologia, 1997. 40(4): p. 430-8. McCarty, M.F. Exploiting complementary therapeutic strategies for the treatment of type II diabetes and prevention of its complications. Med Hypotheses, 1997. 49(2): p. 143-52. Paolisso, G., D'Amore, A., Giugliano, D., Ceriello, A., Varricchio, M., D'Onofrio, F. Pharmacologic doses of vitamin E improve insulin action in healthy subjects and noninsulin-dependent diabetic patients. Am J Clin Nutr, 1993. 57(5): p. 650-6. 76 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. Priebe, M.G., Wang, H., Weening, D., Schepers, M., Preston, T., Vonk, R.J. Factors related to colonic fermentation of nondigestible carbohydrates of a previous evening meal increase tissue glucose uptake and moderate glucose-associated inflammation. Am J Clin Nutr, 2010. 91(1): p. 90-7. Song, Y.J., Sawamura, M., Ikeda, K., Igawa, S., Yamori, Y. Soluble dietary fibre improves insulin sensitivity by increasing muscle GLUT-4 content in stroke-prone spontaneously hypertensive rats. Clin Exp Pharmacol Physiol, 2000. 27(1-2): p. 41-5. Thorburn, A., Muir, J., Proietto, J. Carbohydrate fermentation decreases hepatic glucose output in healthy subjects. Metabolism, 1993. 42(6): p. 780-5. Venter, C.S., Vorster, H.H., Cummings, J.H. Effects of dietary propionate on carbohydrate and lipid metabolism in healthy volunteers. Am J Gastroenterol, 1990. 85(5): p. 549-53. Leinonen, K.S., Poutanen, K.S., Mykkanen, H.M. Rye bread decreases serum total and LDL cholesterol in men with moderately elevated serum cholesterol. J Nutr, 2000. 130(2): p. 164-70. Lundin, E.A., Zhang, J.X., Lairon, D., Tidehag, P., Aman, P., Adlercreutz, H., Hallmans, G. Effects of meal frequency and high-fibre rye-bread diet on glucose and lipid metabolism and ileal excretion of energy and sterols in ileostomy subjects. Eur J Clin Nutr, 2004. 58(10): p. 1410-9. Ross, A.B., Bruce, S.J., Blondel-Lubrano, A., Oguey-Araymon, S., Beaumont, M., Bourgeois, A., Nielsen-Moennoz, C., Vigo, M., Fay, L.B., Kochhar, S., Bibiloni, R., Pittet, A.C., Emady-Azar, S., Grathwohl, D., Rezzi, S. A whole-grain cereal-rich diet increases plasma betaine, and tends to decrease total and LDL-cholesterol compared with a refined-grain diet in healthy subjects. Br J Nutr, 2011. 105(10): p. 1492-502. Glore, S.R., Van Treeck, D., Knehans, A.W., Guild, M. Soluble fiber and serum lipids: a literature review. J Am Diet Assoc, 1994. 94(4): p. 425-36. Ripsin, C.M., Keenan, J.M., Jacobs Jr, D.R., Elmer, P.J., Welch, R.R., Van Horn, L., Liu, K., Turnbull, W.H., Thye, F.W., Kestin, M., Hegsted, M., Davidson, D.M., Davidson, M.H., Dugan, L.D., Demark-Wahnefried, W., Beling, S. Oat products and lipid lowering. A meta-analysis. JAMA, 1992. 267(24): p. 3317-25. Reyna-Villasmil, N., Bermudez-Pirela, V., Mengual-Moreno, E., Arias, N., CanoPonce, C., Leal-Gonzalez, E., Souki, A., Inglett, G.E., Israili, Z.H., HernandezHernandez, R., Valasco, M., Arraiz, N. Oat-derived beta-glucan significantly improves HDLC and diminishes LDLC and non-HDL cholesterol in overweight individuals with mild hypercholesterolemia. Am J Ther, 2007. 14(2): p. 203-12. Wolever, T.M., Gibbs, A.L., Brand-Miller, J., Duncan, A.M., Hart, V., Lamarche, B., Tosh, S.M., Duss, R. Bioactive oat beta-glucan reduces LDL cholesterol in Caucasians and non-Caucasians. Nutr J, 2011. 10: p. 130. Tiwari, U., Cummins, E. Meta-analysis of the effect of beta-glucan intake on blood cholesterol and glucose levels. Nutrition, 2011. 27(10): p. 1008-16. Othman, R.A., Moghadasian, M.H., Jones, P.J. Cholesterol-lowering effects of oat beta-glucan. Nutr Rev, 2011. 69(6): p. 299-309. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on the substantiation of health claims related to beta glucans and maintenance of 77 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. normal blood cholesterol concentrations (ID 754, 755, 757, 801, 1465, 2934) and maintenance or achievement of a normal body weight (ID 820, 823) pursuant to Article 13(1) of Regulation (EC) No 1924/2006. EFSA Journal 2009; 7(9):1254 [18 pp.]. doi:10.2903/j.efsa.2009.1254; Available online: www.efsa.europa.eu/efsajournal EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on the substantiation of a health claim related to oat beta glucan and lowering blood cholesterol and reduced risk of (coronary) heart disease pursuant to Article 14 of Regulation (EC) No 1924/2006. EFSA Journal 2010;8(12):1885 [15 pp.]. doi:10.2903/j.efsa.2010.1885; Available online: www.efsa.europa.eu/efsajournal Jenkins, D.J., Kendall, C.W., Vuksan, V., Augustin, L.S., Mehling, C., Parker, T., Vidgen, E., Lee, B., Faulkner, D., Seyler, H., Josse, R., Leiter, L.A., Connelly, P.W., Fulgoni 3rd, V. Effect of wheat bran on serum lipids: influence of particle size and wheat protein. J Am Coll Nutr, 1999. 18(2): p. 159-65. Slavin, J.L., Jacobs, D., Marquart, L., Wiemer, K. The role of whole grains in disease prevention. J Am Diet Assoc, 2001. 101(7): p. 780-5. Lund, E.K., Salf, K.L., Johnson, I.T. Baked rye products modify cholesterol metabolism and crypt cell proliferation rates in rats. J Nutr, 1993. 123(11): p. 183443. Zhang, J.X., Lundin, E., Reuterving, C.O., Hallmans, G., Stenling, R., Westerlund, E., Aman, P. Effects of rye bran, oat bran and soya-bean fibre on bile composition, gallstone formation, gall-bladder morphology and serum cholesterol in Syrian golden hamsters (Mesocricetus auratus). Br J Nutr, 1994. 71(6): p. 861-70. Andersson, U., Rosen, L., Ostman, E., Strom, K., Wierup, N., Bjorck, I., Holm, C. Metabolic effects of whole grain wheat and whole grain rye in the C57BL/6J mouse. Nutrition, 2010. 26(2): p. 230-9. Lia, A., Hallmans, G., Sandberg, A.S., Sundberg, B., Aman, P., Andersson, H. Oat beta-glucan increases bile acid excretion and a fiber-rich barley fraction increases cholesterol excretion in ileostomy subjects. Am J Clin Nutr, 1995. 62(6): p. 1245-51. Moazzami, A.A., Bondia-Pons, I., Hanhineva, K., Juntunen, K., Antl, N., Poutanen, K., Mykkanen, H. Metabolomics reveals the metabolic shifts following an intervention with rye bread in postmenopausal women--a randomized control trial. Nutr J, 2012. 11: p. 88. Brader, L., Rejnmark, L., Carlberg, C., Schwab, U., Kolehmainen, M., Rosqvist, F., Cloetens, L., Landin-Olsson, M., Gunnarsdottir, I., Poutanen, K.S., Herzig, K.H., Riserus, U., Savolainen, M.J., Thorsdottir, I., Uusitupa, M., Hermansen, K. Effects of a healthy Nordic diet on plasma 25-hydroxyvitamin D concentration in subjects with metabolic syndrome: a randomized, placebo-controlled trial (SYSDIET). Eur J Nutr, 2014. 53(4): p. 1123-34. Magnusdottir, O.K., Landberg, R., Gunnarsdottir, I., Cloetens, L., Akesson, B., Onning, G., Jonsdottir, S.E., Rosqvist, F., Schwab, U., Herzig, K.H., Savolainen, M.J., Brader, L., Hermansen, K., Kolehmainen, M., Poutanen, K., Uusitupa, M., Thorsdottir, I., Riserus, U. Plasma alkylresorcinols reflect important whole-grain components of a healthy Nordic diet. J Nutr, 2013. 143(9): p. 1383-90. 78 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. Magnusdottir, O.K., Landberg, R., Gunnarsdottir, I., Cloetens, L., Akesson, B., Landin-Olsson, M., Rosqvist, F., Iggman, D., Schwab, U., Herzig, K.H., Savolainen, M.J., Brader, L., Hermansen, K., Kolehmainen, M., Poutanen, K., Uusitupa, M., Thorsdottir, I., Riserus, U. Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole-grain rye intake, is associated to insulin sensitivity: a randomized study. Eur J Clin Nutr, 2014. 68(4): p. 453-8. Alberti, K.G., Eckel, R.H., Grundy, S.M., Zimmet, P.Z., Cleeman, J.I., Donato, K.A., Fruchart, J.C., James, W.P., Loria, C.M., Smith Jr, S.C. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation, 2009. 120(16): p. 1640-5. Spitzer, R.L., Yanovski, S., Wadden, T., Wing, R., Marcus, M.D., Stunkard, A., Devlin, M., Mitchell, J., Hasin, D., Horne, R.L. Binge eating disorder: its further validation in a multisite study. Int J Eat Disord, 1993. 13(2): p. 137-53. The Icelandic Food Composition Database (ISGEM). [cited 2013 22.01]; Available from: http://www.matis.is/ISGEM/en/. Danish Food Composition Database - version 7.1. [cited 2012 22.01]; Available from: http://www.foodcomp.dk/v7/fcdb_default.asp. Livsmedelsverket – the National Food Agency. Livsmedelsdatabasen. [cited 2013 22.01]; Available from: http://www.slv.se/en-gb/. Fineli ® - Finnish Food Composition Database. [cited 2013 22.01]; Available from: http://www.fineli.fi/index.php?lang=en. de Mello, V.D., Lindstrom, J., Eriksson, J., Ilanne-Parikka, P., KeinanenKiukaanniemi, S., Sundvall, J., Laakso, M., Tuomilehto, J., Uusitupa, M. Insulin secretion and its determinants in the progression of impaired glucose tolerance to type 2 diabetes in impaired glucose-tolerant individuals: the Finnish Diabetes Prevention Study. Diabetes Care, 2012. 35(2): p. 211-7. Landberg, R., Åman, P., Kamal-Eldin, A. A rapid gas chromatography-mass spectrometry method for quantification of alkylresorcinols in human plasma. Anal Biochem, 2009. 385(1): p. 7-12. Hertzmark, E., Spiegelman, D. The SAS ICC9 Macro. 2010. Magnusdottir, O.K., Landberg, R., Gunnarsdottir, I., Cloetens, L., Åkesson, B., Rosqvist, F., Schwab, U., Herzig, K.H., Hukkanen, J., Savolainen, M.J., Brader, L., Hermansen, K., Kolehmainen, M., Poutanen, K., Uusitupa, M., Riserus, U., Thorsdottir, I. Whole grain rye intake, reflected by a biomarker, is associated with favorable blood lipid outcomes in subjects with the metabolic syndrome – a randomized study. PLOS ONE, 2014. In press Ross, A.B., Chen, Y., Frank, J., Swanson, J.E., Parker, R.S., Kozubek, A., Lundh, T., Vessby, B., Aman, P., Kamal-Eldin, A. Cereal alkylresorcinols elevate gammatocopherol levels in rats and inhibit gamma-tocopherol metabolism in vitro. J Nutr, 2004. 134(3): p. 506-10. 79 209. 210. 211. 212. 213. 214. 215. Ma, J., Ross, A.B., Shea, M.K., Bruce, S.J., Jacques, P.F., Saltzman, E., Lichtenstein, A.H., Booth, S.L., McKeown, N.M. Plasma Alkylresorcinols, Biomarkers of WholeGrain Intake, Are Related to Lower BMI in Older Adults. J Nutr, 2012. 142(10): p. 1859-64. Davy, B.M., Davy, K.P., Ho, R.C., Beske, S.D., Davrath, L.R., Melby, C.L. Highfiber oat cereal compared with wheat cereal consumption favorably alters LDLcholesterol subclass and particle numbers in middle-aged and older men. Am J Clin Nutr, 2002. 76(2): p. 351-8. Kristensen, M., Toubro, S., Jensen, M.G., Ross, A.B., Riboldi, G., Petronio, M., Bugel, S., Tetens, I., Astrup, A. Whole grain compared with refined wheat decreases the percentage of body fat following a 12-week, energy-restricted dietary intervention in postmenopausal women. J Nutr, 2012. 142(4): p. 710-6. Rosen, L.A., Silva, L.O., Andersson, U.K., Holm, C., Ostman, E.M., Bjorck, I.M. Endosperm and whole grain rye breads are characterized by low post-prandial insulin response and a beneficial blood glucose profile. Nutr J, 2009. 8: p. 42. Thorgeirsdottir, H., Valgeirsdottir, H., Gunnarsdottir, I., Gisladottir, E., Gunnarsdottir, B., Thorsdottir, I., Stefansdottir, J., Stengrimsdottir, L. The Icelandic National Nutrition Survey 2011. 2011, Icelandic Directorate of Health, Icelandic Food and Veterinary Authority and Unit for Nutrition Research: Reykjavik. Isaksson, H., Sundberg, B., Aman, P., Fredriksson, H., Olsson, J. Whole grain rye porridge breakfast improves satiety compared to refined wheat bread breakfast. Food Nutr Res, 2008. 52. Isaksson, H., Tillander, I., Andersson, R., Olsson, J., Fredriksson, H., Webb, D.L., Aman, P. Whole grain rye breakfast - sustained satiety during three weeks of regular consumption. Physiol Behav, 2012. 105(3): p. 877-84. 80 Paper I Supplemental Material can be found at: http://jn.nutrition.org/content/suppl/2013/08/11/jn.113.17558 8.DCSupplemental.html The Journal of Nutrition Nutrition and Disease Plasma Alkylresorcinols Reflect Important Whole-Grain Components of a Healthy Nordic Diet1-4 Ola Kally Magnusdottir,5* Rikard Landberg,6 Ingibjorg Gunnarsdottir,5 Lieselotte Cloetens,7 Björn Åkesson,7,8 Gunilla Önning,7 Svandis Erna Jonsdottir,5 Fredrik Rosqvist,9 Ursula Schwab,10,11 Karl-Heinz Herzig,12 Markku J. Savolainen,13 Lea Brader,14 Kjeld Hermansen,14 Marjukka Kolehmainen,10,15 Kaisa Poutanen,10,15 Matti Uusitupa,10,16 Inga Thorsdottir,5 and Ulf Risérus9 Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 5 Unit for Nutrition Research, Landspitali, The National University of Iceland, Faculty of Food Science and Nutrition and School of Health Sciences, University of Iceland, Reykjavı́k, Iceland; 6Department of Food Science, BioCenter, Swedish University of Agricultural Sciences, Uppsala, Sweden; 7Biomedical Nutrition, Pure and Applied Biochemistry, Lund University, Lund, Sweden; 8Department of Clinical Nutrition, Skåne University Hospital, Lund, Sweden; 9Department of Public Health and Caring Sciences, Clinical Nutrition and Metabolism, Uppsala University, Uppsala, Sweden; 10Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland; 11Institute of Clinical Medicine, Internal Medicine, Kuopio University Hospital, Kuopio, Finland; 12Institute of Biomedicine, Department of Physiology and Biocenter of Oulu, Oulu University, Oulu, Finland, and Department of Psychiatry, Kuopio University Hospital, Kuopio, Finland; 13University of Oulu, Institute of Clinical Medicine, Department of Internal Medicine and Biocenter Oulu, Oulu, Finland, and Oulu University Hospital, Clinical Research Center, Oulu, Finland; 14Department of Medicine and Endocrinology MEA, Aarhus University Hospital, Aarhus, Denmark; 15Valtion Teknillien Tutkimuskeshus, Technical Research Center of Finland, Espoo and Kuopio, Finland; and 16Research Unit, Kuopio University Hospital, Kuopio, Finland Abstract Biomarkers of dietary intake can be important tools in nutrition research. Our aim was to assess whether plasma alkylresorcinol (AR) and b-carotene concentrations could be used as dietary biomarkers for whole-grain, fruits and vegetables in a healthy Nordic diet (ND). Participants (n = 166), 30–65 y with a body mass index of 27–40 kg/m2 and two more features of metabolic syndrome (International Diabetes Federation definition, slightly modified), were recruited through six centers in the Nordic countries and randomly assigned to an ND or control diet for 18 or 24 wk, depending on study center. Plasma AR and b-carotene were analyzed and nutrient intake calculated from 4-d food records. Median fiber intake increased in the ND group from 2.5 g/MJ at baseline to 4.1 g/MJ (P < 0.001) at end point (week 18 or 24), and median (IQR) fasting plasma total AR concentration increased from 73 (88) to 106 (108) nmol/L, or 45%, from baseline to end point (P < 0.001). The AR concentration was significantly higher in the ND group (P < 0.001) than in the control group at end point. b-Carotene intake tended to increase in the ND group (P = 0.07), but the plasma b-carotene concentration did not change significantly throughout the study and did not differ between the groups at follow-up. In conclusion, an ND resulted in higher dietary fiber intake and increased plasma total AR concentration compared with the control diet, showing that the total AR concentration might be a valid biomarker for an ND in which whole-grain wheat and rye are important components. No significant difference in plasma b-carotene concentrations was observed between the ND and control groups, suggesting that b-carotene may not be a sensitive enough biomarker of the ND. Introduction Several studies have shown a strong inverse relationship between fruit and vegetable consumption and prevalence of chronic diseases such as hypertension, coronary heart disease, and stroke 1 The study was supported by NordForsk NCoE in Food Nutrition and Health: Project 070014 SYSDIET (Systems Biology in Controlled Dietary Interventions and Cohort Studies). This study was also funded by the Academy of Finland, Finnish Diabetes Research Foundation, Finnish Foundation for Cardiovascular Research, the Sigrid Juselius Foundation, and EVO funding from Kuopio University Hospital (Finland); the Druvan Foundation, ESPEN, Skåne County Council Research and Development Foundation, Swedish Council for Working Life and Social Research, the Heart-Lung Foundation, Diabetesfonden and Foundation Cerealia (Sweden); the Danish Obesity Research Centre (DanORC, www.danorc.dk), the Danish Council for Strategic Research (DairyHealth, J. Nutr. 143: 1383–1390, 2013. (1–4). Moreover, the intake of whole-grain cereals has been inversely associated with type 2 diabetes, coronary heart disease, and other cardiovascular risk factors (5–8). Despite this evidence, no consistent findings have been demonstrated in intervention BioFunCarb) (Denmark), the Agricultural Productivity Fund (Iceland), and the Icelandic Research Fund for graduate students (1206880061). The following companies provided food products for the study participants: Kesko Food and Raisio Group (Finland); Belico Food AB, Bunge Finland, Fazer Bageri Sverige, Lantmännen, Oatly AB, Olle Svensson AB, Procordia Food AB, Pågen AB, Unilever, and Wasabröd AB (Sweden); Lantmännen Food R&D, Jan Import A/S, Ardo A/S, Scandic Food A/S, WASA, Glyngøre Limfjord A/S, Royal Greenland A/S, and Arla Foods (Denmark); the Mother Earth Farm at Vallanes (Iceland); and Unilever Nordic (Sweden, Denmark, Iceland). ã 2013 American Society for Nutrition. Manuscript received February 11, 2013. Initial review completed March 27, 2013. Revision accepted June 03, 2013. First published online July 10, 2013; doi:10.3945/jn.113.175588. 1383 Methods The current analysis is part of the SYSDIET study, a randomized controlled multicenter study performed in six intervention centers (Kuopio 2 Author disclosures: O. K. Magnusdottir, R. Landberg, I. Gunnarsdottir, L. Cloetens, B. Åkesson, G. Önning, S. E. Jonsdottir, F. Rosqvist, U. Schwab, K.-H. Herzig, M. J. Savolainen, L. Brader, K. Hermansen, M. Kolehmainen, K. Poutanen, M. Uusitupa, I. Thorsdottir, and U. Risérus, no conflicts of interest. 3 The study was registered at clinicalTrials.gov as NCT00992641. 4 Supplemental Tables 1-5 are available from the ‘‘Online Supporting Material’’ link in the online posting of the article and from the same link in the online table of contents at http://jn.nutrition.org. 17 Abbreviations used: AR, alkylresorcinols; ICC, intraclass correlation coefficient; ND, Nordic diet; SYSDIET, Systems Biology in Controlled Dietary Interventions and Cohort Studies. *To whom correspondence should be addressed: E-mail: [email protected]. 1384 Magnusdottir et al. and Oulu, Finland; Lund and Uppsala, Sweden; Aarhus, Denmark; and Reykjavik, Iceland). This study examined the health effects of a healthy ND, compared with a habitual diet in the Nordic countries, in a study group with features of metabolic syndrome. Local ethical committees approved the SYSDIET study protocol, which followed the Helsinki declaration guidelines. Informed written consent was obtained from all participants. The study design was reported in detail previously (12). Study participants Study participants were mainly recruited through newspaper advertisements or previously took part in clinical or epidemiological trials in the study centers. The inclusion criteria were 30–65 y of age, BMI of 27–40 kg/m2, and two other criteria for metabolic syndrome as defined by the International Diabetes Foundation (39), although individuals with type 2 diabetes were excluded. Antihypertensive and lipid-lowering medications were allowed but without changes during the trial. The comprehensive inclusion and exclusion criteria have been described previously (12). In total, 201 individuals started the intervention: 104 in the ND group and 97 in the control group. At trial completion, 96 participants were in the ND group (92%) and 70 in the control group (72%). Complete dietary records from both baseline (week 0) and end point (week 18 or 24) were available for 158 participants. Study design The study design and the main clinical and biochemical measurements were described previously (12). Before randomization into the ND or control group, for 1 mo participants followed their habitual diet except oil supplements and food enriched with plant stanol or sterol were discontinued. Participants then followed either the ND or control diet for 18 or 24 wk. For financial and logistic reasons the study period was 18 wk for four centers (Aarhus, Uppsala, Reykjavik, and Oulu) and 24 wk for two centers (Lund and Kuopio) where the intervention was started earlier (12). Staff in all study centers were trained to perform measurements in a similar way, and the quality management protocol was made available to all of them. The study participants were advised to keep physical activity constant and not to change their smoking and drinking habits or drug treatment during the study. Study diets Nordic nutrition recommendations formed the basis of the ND (40), and the mean nutrient intake in the Nordic countries formed the basis for the control diet. Participants in the ND group were, among other things, advised to consume $25% of total energy as whole grain, of which $50% should be rye, barley, and oats, and $6 slices/day of bread with $6 g fiber/100 g (one slice equals 30–40 g of bread). They were asked to consume $500 g of fruit/d ($175 g/d), vegetables ($175 g/d), and berries ($150 g/d). Participants in the control group were asked to consume $25% of energy as refined grain: $90% wheat. They were also asked to keep fruit and vegetable consumption between 200 and 250 g/d. The food items recommended are summarized (Supplemental Table 1). Key products were provided to the participants in both groups. The ND group received whole-grain products such as oats and barley, crisp bread, whole-grain wheat and rye breads, berry products (frozen berries and dried powder), and rapeseed oil– and vegetable oil–based spreads. Furthermore, fish (fatty and lean) was either directly provided or the cost for purchase was reimbursed to ND group participants. Control group participants received low-fiber cereal products, for example, breads with fiber content <6 g/100 g and ordinary pasta and dairy fat–based spread. Participants were instructed by a clinical nutritionist or dietitian during all visits. Participants kept dietary records on their intake of cereal products, berry products, vegetables, and fish. These records were checked at each visit to the study center, including visits when groceries were delivered to participants, which was at 1- to 2-wk intervals. Any intake of supplements was to be discontinued at least 4 wk before entering the study. An exception was made for vitamin D supplements in Iceland, which the participants could continue to use providing they kept dosage unchanged. Vitamin D supplementation was not included in the nutrient calculations. The target nutrient composition of the diets is summarized in Table 1. Dietary assessment Dietary intake was assessed using a consecutive 4-d (Wednesday to Saturday or Sunday to Wednesday) food record, with Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 studies. Recent Danish and Swedish studies suggest that a ‘‘healthy Nordic diet" composed of typical Nordic food items may induce similar health effects as the Mediterranean diet (9–11). In the controlled intervention study Systems Biology in Controlled Dietary Interventions and Cohort Studies (SYSDIET),17 a healthy Nordic, isocaloric diet caused moderate improvements in lipid profile and resulted in lower levels of the inflammation marker IL 1 receptor antagonist compared with a control diet (12). Most dietary assessment instruments rely to some extent on self-reporting, which is prone to systematic and random measurement errors as a result of, for example, under or overreporting and imprecision (13,14). Collecting valid data on dietary fiber and whole-grain intake can be especially challenging because of the broad selection of foods contributing to these intakes (15). One way to avoid some of the obstacles associated with traditional dietary assessment instruments is to use dietary biomarkers as a complement to self-reports (16–20). Alkylresorcinols (ARs) have been proposed as biomarkers for whole-grain and bran intake of wheat and rye (15,21,22) and for cereal fiber intake (23). ARs are 3,5-dihydroxy-alkylbenzenes (phenolic lipids) abundant in the outer layers of rye and wheat, found in low amounts in barley, and absent in highly refined white flour and in most other foods consumed by humans. Controlled whole-grain and bran intervention studies have shown that plasma AR concentrations increase proportionally with AR and wholegrain intake (22,24–28). Thus, AR can be a useful biomarker to categorize participants as low or high consumers of whole-grain foods (15,29) and to monitor their compliance with a wholegrain diet in intervention studies (21,30). Carotenoids have been proposed as biomarkers of fruit and vegetable intake (31,32). They are generally found in plant foods, mainly yellow and orange fruits, orange root vegetables, and green leafy vegetables. Population studies have indicated that b-carotene may be a reliable biomarker for the intake of fruits and vegetables (33), although not all studies agree (32,34). In addition to being a possible biomarker of dietary intake, plasma b-carotene is linked as a risk factor for diseases as previously reviewed (35). Low plasma b-carotene is a risk factor for insulin resistance, type 2 diabetes, and metabolic syndrome (36, 37), and a relationship between dietary b-carotene and metabolic syndrome has also been observed (38). The aim of the present study was to assess whether AR and b-carotene plasma concentrations could be used as dietary biomarkers of a healthy Nordic diet (ND), rich in whole-grain wheat and rye as well as fruits and vegetables, and a control diet in the SYSDIET study and to determine their relationships with reported dietary fiber intake in individuals with markers of metabolic syndrome (12). either weighed or estimated portion sizes. The baseline food diaries were recorded during the 1-mo period before the intervention and then at week 2 (data not included in this paper), week 12, and the end point. The nutrient intake during the intervention has been published elsewhere (12). The main focus of our study was on the intake of dietary fiber, b-carotene, and other nutrients abundant in fruits, vegetables, and whole grains. At each center information from the food diaries was used to calculate nutrient intake, using national nutrient databases. All the databases used EuroFIR definitions for chemicals and total dietary fiber determined by AOAC methods (41–44). TABLE 1 Advised nutrient composition of a healthy Nordic diet and control diet during the intervention1 Nutrient Carbohydrate, E% Sucrose, E% Fiber, g/d Protein, E% Fat, E% Saturated, E% Monounsaturated fat + polyunsaturated fat, E% Salt, g Women Men 1 E%, percent of daily energy intake. Maximum: 50 g/d. 3 Or 4 g/MJ. 2 Nordic diet Control diet 45–52 #102 $353 18–20 30–35 ,10 At least 2/3 of total fat intake 45–47 No restrictions 15–20 18–20 35 15 15 + 5 6 7 #10 #10 Results Data at baseline Baseline characteristics of the participants are shown in Supplemental Table 2. No difference in basic baseline characteristics existed between dropouts and those who completed the study (data not shown). No significant difference in AR or b-carotene concentration in plasma was observed between the ND and control groups at randomization (Table 2).When all participants were analyzed together at baseline, median (IQR) intake of dietary fiber was 21 (8) g/d and 2.5 (1.0) g/MJ, intake of b-carotene was 2.0 (3.1) mg/d, the plasma AR concentration was 72 (73) nmol/L, and the plasma b-carotene concentration was 0.35 (0.32) mmol/L. Changes observed during the intervention Dietary fiber intake. Consistent with the isocaloric study design, no change in total energy intake occurred during the study. During the intervention, the median (IQR) intake of dietary fiber increased by 71%, to 36 (15) g/d, in the ND group (P < 0.001) at end point and decreased by 20%, to 16 (6) g/d, in the control group (P < 0.001). The goal of dietary fiber intake of at least 35 g/d or 4 g/MJ in the ND group was reached at week 12 and was steady throughout the intervention. Plasma AR concentrations. No difference in plasma total AR concentrations was observed between sexes within the groups at any time. However, at end point, median (IQR) plasma total AR concentrations increased to 106 (108) nmol/L in the ND group (P = 0.001) but decreased slightly in the control group, to 61 (40) nmol/L (P = 0.021) (Table 2). The difference in plasma total AR concentrations between the two groups was significant at week 12 and at end point (P < 0.001) (Fig. 1A). Dietary biomarkers of a healthy Nordic diet 1385 Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 Biochemical measurements and body composition Analyses of concentrations of plasma glucose, cholesterol, HDL cholesterol, and TGs were performed in local laboratories as described elsewhere (12). Fasting LDL cholesterol concentrations were calculated by the Friedewald equation. Blood pressure was measured using automatic blood pressure devices. Waist circumference was measured midway between the lowest rib and iliac crest. Body composition was determined by bioelectrical impedance analysis (Kuopio: STA/BIA Body Composition Analyzer and Bodygram Software; Akern Bioresearch. Reykjavik: STA/ BIA Body Composition Analyzer, Akern Bioresearch. Lund: Xitron Hydra 4200, Xitron Technologies. Uppsala: Tanita BC-558 Segmental Body Composition Monitor. Aarhus: UniQuest-SEAC SFB-3; UniQuest Oulu: body composition not measured). Staff at each center were trained to perform the measurements according to the standard operational procedures agreed on by all centers. Blood samples used for analysis of plasma AR and b-carotene were collected at baseline, week 12, and end point. Plasma AR samples were analyzed at the Department of Food Science, Uppsala BioCenter, Swedish University of Agricultural Sciences. AR homologs C17:0–C25:0 were measured in 0.2-mL plasma samples according to a previously published GC-MS method (45). Quality control samples (n = 4) were included in each batch to ensure appropriate within- and between-batch precision (CV < 10–15%). Samples from the same individual were analyzed within the same batch, and samples from different centers were allocated randomly between batches. Plasma b-carotene samples were analyzed at Biomedical Nutrition, Lund University, using a commercial HPLC kit (catalog no. 32000, Chromsystems). Samples from the same individual were included in the same batch. The kit consisted of mobile phase, calibrator, control samples, column, internal standard, precipitation reagent, and extraction buffer. All samples were prepared in a room protected from direct sunlight. The manufacturerÕs manual was followed, and quality controls were included in each batch (CV = 10%). HPLC analysis was performed using a Shimadzu liquid chromatograph system equipped with a vacuum degasser, a solvent delivery module, an autoinjector, a column oven, and a diode-array detector. Data pooling and statistical analysis Data from all intervention centers were entered into Microsoft Office Excel 2007 before they were pooled in a joint centralized database maintained at VTT Technical Research Centre, Helsinki, Finland. Data were imported to SPSS for Windows, version 20.0 (SPSS) for statistical analysis. Level of significance in all analyses was set at P < 0.05. Normality of variables was checked by inspection of histograms and by using the Kolmogorov-Smirnov one-sample test. Values are reported as means 6 SDs if normally distributed or as medians and IQRs if not normally distributed. Independent sample t test or Mann-Whitney U test was used to compare data between sexes and groups. Paired samples t test or WilcoxonÕs signed rank test was used to compare values between baseline and week 12 or the end point. One-way ANOVA was used to compare baseline plasma AR and plasma b-carotene between centers. Spearman rank correlation coefficient (rs) was used to calculate the correlation between both plasma AR and b-carotene concentrations and various nutrients abundant in whole-grain food and vegetables. As the distributions of plasma AR and b-carotene concentrations were right skewed, log transformation was performed before further statistical analysis to better approximate normal distribution. Linear regression for plasma total AR and plasma b-carotene was used with group, baseline concentration, sex, age, BMI, and TGs as factors in the model. Additionally, study center was included in the model for plasma AR. The residuals from the model were examined for normality. Reproducibility, a measure of the intraperson stability in dietary intake or biomarker concentration during the intervention, was assessed by estimating the intraclass correlation coefficient (ICC). ICC is defined as the intraperson variance divided by the total variance (i.e., the sum of intra- and interperson variance), and it assumes a common mean between replicate measurements. We calculated ICC using a freely available SAS macro (46) that was imported to SAS 9.2 (SAS Institute). All ICC estimates are based on log-transformed values. TABLE 2 Dietary intakes of energy, fiber, and b-carotene and plasma biomarker concentrations in a healthy Nordic diet group and control group at baseline (week 0), week 12, and end point (week 18 or 24)1 Week 0 Week 12 Week 18 or 24 Change2 % 8.22 (2.72) 8.31 (2.98) 8.53 (2.05) 8.80 (2.56) 8.19 (2.66) 8.35 (2.60) 0 0 21 (9) 20 (7) 35 (13)3 15 (7)4 36 (15)3,4 16 (6)4 71 220 2.5 (1.2) 2.5 (1.1) 4.1 (1.2)3,4 1.8 (0.7)4 4.1 (1.1)3,4 1.8 (0.8)4 38 227 2.4 (3.0) 1.7 (3.3) 3.1 (4.3)3,4 1.7 (1.6)4 2.9 (3.5)3 1.4 (2.1)4 22 221 73 (87) 70 (70) 91 (88)3 57 (43)4 106 (105)3,4 61 (40)4 45 214 0.36 (0.34) 0.29 (0.34) 0.39 (0.37)4 0.36 (0.27) 0.36 (0.32) 0.33 (0.34) 0 14 1 Values are presented as median (IQR). AR, alkylresorcinols. Proportional changes between week 0 and week 18 or 24. 3 Different from control at that time (P , 0.05). 4 Different from baseline in that group (P = 0.05). 2 Determinants of plasma AR concentrations. Diet group allocation (ND group vs. control group; P < 0.001) and plasma AR concentration at baseline (P < 0.001) were predictors of AR concentration at end point (with adjusted R2 = 0.38). Serum TG concentration at end point was also an important factor in the model, and study center was of borderline significance (P 5 0.09) (Table 3). Considerable differences were found in the plasma AR concentrations among all study centers at baseline, and at the two Finnish centers, Kuopio and Oulu, the participants had the highest median (IQR) concentrations [108 (72) and 108 (96) nmol/L, respectively], whereas in Iceland the lowest concentration was observed [43 (37) nmol/L]. The differences among all centers were explained by differences in whole-grain and/or bran intake from wheat and rye, which resulted in differences in total dietary fiber intake (Supplemental Table 3). Plasma AR concentrations showed moderate to good stability over a period of 6–12 wk, from week 12 to end point [(ICC = 0.44–0.72) Supplemental Table 4]. Positive correlations were noted between plasma total AR and total fiber intake at all time points as well as between plasma total AR and intake of vitamin E, folate, and magnesium (rs = 0.24–0.57) (Supplemental Table 5). Dietary b-carotene and plasma b-carotene concentration. b-Carotene intake increased during the intervention and was higher in the ND than the control group (Table 2). In the ND group median (IQR) intake of b-carotene was 2.4 (3.0) mg/d at baseline, increasing to 3.1 (4.3) mg/d (P = 0.008) at week 12 and borderline significantly to 2.9 (3.5) mg/d (P = 0.07) at end point. In the control group b-carotene intake decreased from 1.7 (3.3) at baseline to 1.7 (1.6) mg/d at week 12 (P = 0.007) and 1.4 (2.1) mg/d at end point (P = 0.019). 1386 Magnusdottir et al. The median (IQR) plasma b-carotene concentration did not change significantly throughout the study either in the ND group or in the control group [0.36 (0.32) mmol/L and 0.33 (0.34) mmol/L, respectively, at endpoint] (Fig. 1B). Dividing b-carotene values by LDL cholesterol, TGs, total cholesterol and TGs, or body fat mass did not change the results. Determinants of plasma b-carotene concentrations. The ND group was not a significant predictor of plasma b-carotene concentration at end point, but baseline b-carotene concentration and BMI were significant predictors (with adjusted R2 = 0.63). The length of the intervention (18 or 24 wk) was of borderline significance (P = 0.054) in the model (Table 3). Plasma b-carotene concentrations showed good stability over the period of 12 wk (ICC = 0.83–0.88) (Supplemental Table 4). A positive correlation was found between plasma b-carotene and fiber intake at baseline (P = 0.005), but it was borderline significant at week 12 (P = 0.054) and end point (P = 0.110) (Supplemental Table 5). Positive correlations were found between plasma b-carotene and b-carotene as calculated from food records at all time-points, with correlation coefficients in the range rs = 0.28–0.39. Additionally, there were significant correlations of plasma b-carotene to vitamin C intake at baseline (P = 0.011) and to the intake of vitamin C and folate at end point (P = 0.005 and P = 0.039, respectively). Discussion One of the most prominent changes in nutrient composition during the SYSDIET intervention was increased fiber intake in the ND group, whereas slightly decreased fiber intake was seen Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 Dietary intake (n = 154) Energy, MJ/d Nordic diet Control diet Fiber intake, g/d Nordic diet Control diet Fiber intake, g/MJ Nordic diet Control diet b-Carotene, mg/d Nordic diet Control diet Plasma biomarker concentrations (n = 149) Total AR, nmol/L Nordic diet Control diet b-Carotene, mmol/L Nordic diet Control diet in the control group. Both groups reached the predetermined goals (35 g/d or 4.0 g/MJ in the ND group and 15–20 g/d in the control group) by week 12. The change in fiber intake was reflected in increased plasma AR concentrations in the ND group, showing that cereal fiber from wheat and rye substantially contributed to the increased fiber intake. Our findings suggest that plasma total AR concentration is a good biomarker for wholegrain wheat and rye, the main contributing foods in the ND. In addition, previous data suggested that AR meets most of the criteria of a good biomarker for intake of whole-grain wheat and rye foods and cereals, particularly at frequent consumption and wide range of intake (15,22,23). Plasma AR is responsive to whole-grain wheat and rye intake, and the long-term reproducibility is moderate (24–26,47). In the present study there was a great difference in AR concentration between the two groups, with an increase in the ND group and a decrease in the control group. Blood lipids were expected to correlate with AR concentration because plasma AR is transported in lipoproteins (15,27). The finding that serum TGs were a significant predictor for AR concentration suggests that factors that result in transient increases in plasma TG concentration may also be important for the repeatability of plasma AR measurements (48). Unlike in previous findings, we found that the plasma AR concentration was not significantly different between sexes at any time, and sex was not a significant covariate in the regression model. In several studies men had higher plasma AR concentration than women (15,24,47,49), which has been attributed to higher intake of foods containing AR by men compared with women and a more rapid elimination of AR in TABLE 3 Regression coefficients for factors affecting plasma total alkylresorcinols and plasma b-carotene at end point (week 18 or 24)1 B (95% CI) P 2 Model 1 log, ARs (n = 154) ND vs. control group Baseline plasma log Sex (F/M) Age (y) BMI (kg/m2) TGs (mmol/L) Study center Model 2 log, b-carotene (n = 166)3 ND vs. control group Baseline plasma log Sex (F/M) Age (y) BMI (kg/m2) TGs (mmol/L) 0.234 (0.157, 0.311) 0.354 (0.212, 0.496) 0.026 (20.055, 0.106) 0.003 (20.001, 0.008) 0.002 (20.009, 0.014) 0.110 (0.042, 0.177) 20.019 (20.041, 0.003) ,0.001 ,0.001 0.53 0.17 0.70 0.002 0.09 20.002 (20.057, 0.053) 0.746 (0.645, 0.847) 0.048 (20.012, 0.109) 0.001 (20.002, 0.005) 20.010 (20.019, 20.002) 20.024 (20.071, 0.023) 0.95 ,0.001 0.12 0.39 0.017 0.32 1 Log-transformed values. B reflects the change in the log-transformed outcome variable for each 1-unit changes in the independent variable. AR, alkylresorcinols; ND, Nordic diet. 2 Adjusted r2 = 0.38. 3 Adjusted r2 = 0.63. Dietary biomarkers of a healthy Nordic diet 1387 Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 FIGURE 1 Plasma concentrations for participants in healthy Nordic diet (ND) and control (C) groups at baseline, 12 wk, and 18 or 24 wk (end point). Values are medians, and bars represent the IQR (A) Plasma total alkylresorcinols concentration (ND group, n = 94; control group, n = 65). *Different from C at that time, P , 0.05. (B) Plasma b-carotene concentration (ND group, n = 96; control group, n = 70). women. In general, women have a larger distribution volume because of their higher body fat content, which may result in lower plasma AR concentration. In the present study, women had a higher body fat percentage as well, suggesting that factors other than variations in body fat may have caused the lack of a sex difference in AR concentration. A marked difference existed in baseline values for plasma AR concentration among study centers. Rye consumption is widespread in Finland and more common than in other Nordic countries (50). Hence, the increase in plasma AR concentration in Finnish participants during the intervention was not as sharp as in participants from other centers. Instead, Finnish participants in the control group had a greater decrease in AR concentration. Similarly, for the other centers, AR concentration greatly increased in the ND group and decreased in the control group but not as sharply as in the Finnish participants. The reliability of plasma measurements of AR at week 12 and end point was moderate to good, with the ICC being similar or higher than in previous studies (24,26,51). Higher reliability was observed in the ND group than the control group. Strictly defined, AR should be regarded as a short-term biomarker, but under stable and frequent intake conditions, it also reflects longterm exposure fairly well (52). The half-life of AR is relatively short (5 h), and the absorption half-life is about 6–8 h (53). Absorption and elimination counteract each other when whole grains are consumed regularly, with resulting stable plasma concentrations. Because cereals and bread were consumed on a daily basis in the present study, plasma AR concentrations would be stable and appropriately reflect intake during the intervention period. Note, however, that the use of plasma AR as a biomarker of whole-grain intake is not practical in populations who primarily consume other types of whole grains such as oats or brown rice, because these cereals do not contain AR. No significant difference in plasma b-carotene concentration was found between the ND and the control group. The baseline concentration of plasma b-carotene corroborates with data obtained in nonvegetarians in similar groups (54). The reproducibility 1388 Magnusdottir et al. are important contributors to fiber intake and concomitantly AR intake. Despite a significant correlation between plasma b-carotene and b-carotene intake at baseline, the ND did not cause a significant change, suggesting plasma b-carotene may not be a sensitive enough biomarker of the foods in the ND under the given time frame. Taken together, these findings show that parallel assessment of dietary biomarkers and dietary intakes is useful and may facilitate accurate interpretation of dietary intervention studies. Acknowledgments We thank Ms. M. Debiec for help with the analysis of b-carotene in plasma and Cindy Mari Imai for proofreading the manuscript. M.U., K.P., M.J.S., I.T., B.Å., U.R., and K.H. designed the research. U.S., M.K., L.C., G.Ö., L.B., I.G., O.K.M., S.E.J., and F.R. conducted the research. R.L., B.Å., and L.C. provided essential materials. O.K.M. and R.L. performed statistical analysis, and O.K.M. wrote the paper. All authors have read and approved the final manuscript. Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Buijsse B, Feskens EJ, Kwape L, Kok FJ, Kromhout D. Both alpha- and beta-carotene, but not tocopherols and vitamin C, are inversely related to 15-year cardiovascular mortality in Dutch elderly men. J Nutr. 2008; 138:344–50. Esmaillzadeh A, Kimiagar M, Mehrabi Y, Azadbakht L, Hu FB, Willett WC. Fruit and vegetable intakes, C-reactive protein, and the metabolic syndrome. Am J Clin Nutr. 2006;84:1489–97. Panagiotakos DB, Pitsavos C, Skoumas Y, Stefanadis C. The association between food patterns and the metabolic syndrome using principal components analysis: The ATTICA Study. J Am Diet Assoc. 2007;107:979–87, quiz 97. Boeing H, Bechthold A, Bub A, Ellinger S, Haller D, Kroke A, LeschikBonnet E, Muller MJ, Oberitter H, Schulze M, et al. Critical review: vegetables and fruit in the prevention of chronic diseases. Eur J Nutr. 2012;51:637–63. de Munter JS, Hu FB, Spiegelman D, Franz M, van Dam RM. Whole grain, bran, and germ intake and risk of type 2 diabetes: a prospective cohort study and systematic review. PLoS Med. 2007;4:e261. Jacobs DR, Jr., Gallaher DD. Whole grain intake and cardiovascular disease: a review. Curr Atheroscler Rep. 2004;6:415–23. Sahyoun NR, Jacques PF, Zhang XL, Juan W, McKeown NM. Wholegrain intake is inversely associated with the metabolic syndrome and mortality in older adults. Am J Clin Nutr. 2006;83:124–31. McKeown NM, Meigs JB, Liu S, Wilson PW, Jacques PF. Whole-grain intake is favorably associated with metabolic risk factors for type 2 diabetes and cardiovascular disease in the Framingham Offspring Study. Am J Clin Nutr. 2002;76:390–8. Olsen A, Egeberg R, Halkjaer J, Christensen J, Overvad K, Tjonneland A. Healthy aspects of the Nordic diet are related to lower total mortality. J Nutr. 2011;141:639–44. Adamsson V, Reumark A, Fredriksson IB, Hammarstrom E, Vessby B, Johansson G, Riserus U. Effects of a healthy Nordic diet on cardiovascular risk factors in hypercholesterolaemic subjects: a randomized controlled trial (NORDIET). J Intern Med. 2011;269:150–9. Adamsson V, Reumark A, Cederholm T, Vessby B, Riserus U, Johansson G. What is a healthy Nordic diet? Foods and nutrients in the NORDIET study. Food Nutr Res. 2012;56. Uusitupa M, Hermansen K, Savolainen MJ, Schwab U, Kolehmainen M, Brader L, Mortensen LS, Cloetens L, Johansson-Persson A, Önning G, et al. Effects of an isocaloric healthy Nordic diet on insulin sensitivity, lipid profile and inflammation markers in metabolic syndrome - a randomized study (SYSDIET). J Intern Med. 2013;274: 10.1111/joim.12044. Lissner L, Troiano RP, Midthune D, Heitmann BL, Kipnis V, Subar AF, Potischman N. OPEN about obesity: recovery biomarkers, dietary reporting errors and BMI. Int J Obes (Lond). 2007;31(6):956–61. Bingham S, Luben R, Welch A, Tasevska N, Wareham N, Khaw KT. Epidemiologic assessment of sugars consumption using biomarkers: comparisons of obese and nonobese individuals in the European prospective Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 for plasma b-carotene was high (ICC = 0.83–0.87), in line with the long half-life of b-carotene. No significant change occurred in the groups during the study, and study group was not a significant covariate in the regression model. This finding was not surprising, because the change in b-carotene intake was small. Similar results for the relation between b-carotene intake and plasma concentration were noted elsewhere (33). Various explanations for this lack of response in plasma b-carotene concentration are possible. First, the change in b-carotene intake may be too small to be reflected in a significant change in plasma b-carotene concentration, possibly because of the high intra- and interindividual variations in both the intake and plasma concentration of b-carotene (55). In addition, in intervention studies where significant responses have been observed, increases in b-carotene intake were greater (31,33), and although it would be expected that such an increase would increase concentration, it may not because of variations in the bioavailability of b-carotene from different diets (34). Moreover, in addition to b-carotene, other carotenoids, such as a-carotene, lutein, b-cryptoxanthin, lycopene, and zeaxanthin, have been recommended for use as biomarkers for fruit and vegetable intake (33). Plasma lycopene in particular seems to respond differently than plasma b-carotene, but this difference was not explored in the present study. Second, several factors can affect the bioavailability and metabolism of b-carotene, including sex; smoking; dose; gene polymorphisms; and dietary factors such as fat, protein, and fiber content (56,57). In addition, controlled human studies have shown that the bioavailability of b-carotene from highly processed food is higher than that from raw material (58,59). Total fat and protein intake was similar in the two groups, but dietary fiber intake was much higher in the ND group than in the control group. Controlled intervention studies have shown that the presence of dietary fiber in vegetables and fruits (i.e., dietary pectin) reduces bioavailability of carotenoids (60), which might be a factor affecting the results in the present study. The calculation of correlation coefficients between plasma b-carotene and intake of nutrients abundant in fruits and vegetables gave mixed results depending on group and time point, indicating that fruits and vegetables are not the only source of b-carotene. Furthermore, the half-life of b-carotene is about 20 d, compared to 5 h for AR, which may influence the results. Recently, polymorphisms in genes such as BMOC1 were linked to plasma b-carotene levels (61–63), but this finding was not explored in the present study. Third, the lack of response in plasma b-carotene concentration between the groups may imply that the participants did not increase their intake of fruits and vegetables as advised or that the intake was not properly recorded. The main emphasis in dietary instructions for the ND group was on whole grains, berries, fish, and dietary fat, and many of these foods were directly provided to participants. Increasing the intake of b-carotene–rich fruits and vegetables was less emphasized. Berries were a key component of the diet, but plasma b-carotene is not a good marker for this food group. One of the strengths of the present study is that reliable and comparable methods were used in six centers in four Nordic countries. The same study protocol was used in all centers, and analyses of AR and b-carotene samples were done centrally. A limitation of this study is that the detailed information on the intake of whole-grain products and fruits and vegetables was not available. Such information would have strengthened the evaluation of whether the biomarkers could adequately reflect dietary intakes. In conclusion, AR may be used as a biomarker in a population consuming a ND, in which whole-grain wheat and rye 15. 16. 17. 18. 19. 20. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. Swedish men: report from the Uppsala Longitudinal Study of Adult Men (ULSAM) study. Diabetologia. 2009;52:97–105. Czernichow S, Vergnaud AC, Galan P, Arnaud J, Favier A, Faure H, Huxley R, Hercberg S, Ahluwalia N. Effects of long-term antioxidant supplementation and association of serum antioxidant concentrations with risk of metabolic syndrome in adults. Am J Clin Nutr. 2009;90: 329–35. Sluijs I, Beulens JW, Grobbee DE, van der Schouw YT. Dietary carotenoid intake is associated with lower prevalence of metabolic syndrome in middle-aged and elderly men. J Nutr. 2009;139:987–92. Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, Fruchart JC, James WP, Loria CM, Smith SC Jr. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120: 1640–5. Nordic Council of Ministers. Nordic nutrition recommendations 2004: intergrating nutrition and physical activity. Copenhagen: Nordic Council of Ministers; 2004. Matı́s ohf. Icelandic Food Composition Database (ISGEM) [Database on the Internet]. [cited 2013 Jan 22]. Available from: http://www.matis. is/ISGEM/en/. Danish Food Composition Database, version 7.1 [Database on the Internet]. Kongens Lyngby, Denmark: National Food Institute, Technical University of Denmark. [cited 2012 Jan 22]. Available from: http:// www.foodcomp.dk/v7/fcdb_default.asp. Swedish National Food Agency (Livsmedelsverket) Livsmedelsdatabasen [Database on the Internet]. [cited 2013 Jan 22]. Available from: http://www.slv.se/en-gb/. Fineli: Finnish Food Composition Database [Database on the Internet]. Helsinki, Finland: National Institute for Health and Welfare. [cited 2013 Jan 22]. Available from: http://www.fineli.fi/index.php?lang=en. Landberg R, Åman P, Kamal-Eldin A. A rapid gas chromatography-mass spectrometry method for quantification of alkylresorcinols in human plasma. Anal Biochem. 2009;385:7–12. Hertzmark E, Spiegelman D. SAS ICC9 Macro; 2010 [cited 2012 Jun 26]. Available from: http://www.hsph.harvard.edu/faculty/donnaspiegelman/files/icc9.pdf. Ross AB. Present status and perspectives on the use of alkylresorcinols as biomarkers of wholegrain wheat and rye intake. Journal of nutrition and metabolism. 2012;2012:462967. Landberg R, Kamal-Eldin A, Aman P, Christensen J, Overvad K, Tjonneland A, Olsen A. Determinants of plasma alkylresorcinol concentration in Danish post-menopausal women. Eur J Clin Nutr. 2011; 65:94–101. Chen Y, Ross AB, Aman P, Kamal-Eldin A. Alkylresorcinols as markers of whole grain wheat and rye in cereal products. J Agric Food Chem. 2004;52:8242–6. Rasmussen LB, Andersen LF, Borodulin K, Enghardt Barbieri H, Fagt S, Matthiessen J, Sveinsson T, Thorgeirsdottir H, Trolle E. Nordic monitoring of diet, physical activity and overweight. First collection of data in all Nordic countries 2011. Denmark: Nordic Council of Ministers; 2012. Andersson A, Marklund M, Diana M, Landberg R. Plasma alkylresorcinol concentrations correlate with whole grain wheat and rye intake and show moderate reproducibility over a 2- to 3-month period in freeliving Swedish adults. J Nutr. 2011;141:1712–8. Landberg R, Aman P, Hallmans G, Johansson I. Long-term reproducibility of plasma alkylresorcinols as biomarkers of whole-grain wheat and rye intake within Northern Sweden Health and Disease Study Cohort. Eur J Clin Nutr. 2013; 67(3):259–63. Landberg R, Linko AM, Kamal-Eldin A, Vessby B, Adlercreutz H, Aman P. Human plasma kinetics and relative bioavailability of alkylresorcinols after intake of rye bran. J Nutr. 2006;136:2760–5. Rauma AL, Mykkanen H. Antioxidant status in vegetarians versus omnivores. Nutrition. 2000;16:111–9. Forman MR, Borkowf CB, Cantwell MM, Steck S, Schatzkin A, Albert PS, Lanza E. Components of variation in serum carotenoid concentrations: the Polyp Prevention Trial. Eur J Clin Nutr. 2009;63:763–70. Yeum KJ, Russell RM. Carotenoid bioavailability and bioconversion. Annu Rev Nutr. 2002;22:483–504. Dietary biomarkers of a healthy Nordic diet 1389 Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 21. investigation of cancer Norfolk. Cancer Epidemiol Biomarkers Prev. 2007;16:1651–4. Ross AB, Bourgeois A, Macharia HN, Kochhar S, Jebb SA, Brownlee IA, Seal CJ. Plasma alkylresorcinols as a biomarker of whole-grain food consumption in a large population: results from the WHOLEheart Intervention Study. Am J Clin Nutr. 2012;95:204–11. Kuhnle GG. Nutritional biomarkers for objective dietary assessment. J Sci Food Agric. 2012;92:1145–9. Prentice RL, Sugar E, Wang CY, Neuhouser M, Patterson R. Research strategies and the use of nutrient biomarkers in studies of diet and chronic disease. Public Health Nutr. 2002;5: 6A:977–84. Gibson RS. Principles of nutritional assessment. 2nd ed. New York: Oxford University Press; 2005. Alfthan G. Nordic Biomarker Seminar (Team Nord). Copenhagen, Denmark: Nordic Council of Ministers; 2005. p. 13–7. Gibson RS. Nutritional assessment: a laboratory manual. New York: Oxford University Press; 1993. Ross AB, Kamal-Eldin A, Aman P. Dietary alkylresorcinols: absorption, bioactivities, and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr Rev. 2004;62:81–95. Landberg R, Kamal-Eldin A, Andersson A, Vessby B, Aman P. Alkylresorcinols as biomarkers of whole-grain wheat and rye intake: plasma concentration and intake estimated from dietary records. Am J Clin Nutr. 2008;87:832–8. Aubertin-Leheudre M, Koskela A, Marjamaa A, Adlercreutz H. Plasma alkylresorcinols and urinary alkylresorcinol metabolites as biomarkers of cereal fiber intake in Finnish women. Cancer Epidemiol Biomarkers Prev. 2008;17:2244–8. Montonen J, Landberg R, Kamal-Eldin A, Aman P, Knueppel S, Boeing H, Pischon T. Reliability of fasting plasma alkylresorcinol concentrations measured 4 months apart. Eur J Clin Nutr. 2010;64:698–703. Landberg R, Aman P, Friberg LE, Vessby B, Adlercreutz H, Kamal-Eldin A. Dose response of whole-grain biomarkers: alkylresorcinols in human plasma and their metabolites in urine in relation to intake. Am J Clin Nutr. 2009;89:290–6. Landberg R, Kamal-Eldin A, Andersson SO, Johansson JE, Zhang JX, Hallmans G, Aman P. Reproducibility of plasma alkylresorcinols during a 6-week rye intervention study in men with prostate cancer. J Nutr. 2009;139:975–80. Linko-Parvinen AM, Landberg R, Tikkanen MJ, Adlercreutz H, Penalvo JL. Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. J Nutr. 2007;137:1137–42. Ross AB, Pineau N, Kochhar S, Bourgeois A, Beaumont M, Decarli B. Validation of a FFQ for estimating whole-grain cereal food intake. Br J Nutr. 2009;102:1547–51. Wild CP, Andersson C, O’Brien NM, Wilson L, Woods JA. A critical evaluation of the application of biomarkers in epidemiological studies on diet and health. Br J Nutr. 2001;86: Suppl 1:S37–53. Ocké MC, Kaaks RJ. Biochemical markers as additional measurements in dietary validity studies: application of the method of triads with examples from the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr. 1997; 65(4, Suppl)1240S–5S. Martini MC, Campbell DR, Gross MD, Grandits GA, Potter JD, Slavin JL. Plasma carotenoids as biomarkers of vegetable intake: the University of Minnesota Cancer Prevention Research Unit Feeding Studies. Cancer Epidemiol Biomarkers Prev. 1995;4:491–6. Ye Z, Song H. Antioxidant vitamins intake and the risk of coronary heart disease: meta-analysis of cohort studies. Eur J Cardiovasc Prev Rehabil. 2008;15:26–34. Brevik A, Andersen LF, Karlsen A, Trygg KU, Blomhoff R, Drevon CA. Six carotenoids in plasma used to assess recommended intake of fruits and vegetables in a controlled feeding study. Eur J Clin Nutr. 2004;58: 1166–73. Garcia AL, Mohan R, Koebnick C, Bub A, Heuer T, Strassner C, Groeneveld MJ, Katz N, Elmadfa I, Leitzmann C, et al. Plasma betacarotene is not a suitable biomarker of fruit and vegetable intake in german subjects with a long-term high consumption of fruits and vegetables. Ann Nutr Metab. 2010;56:23–30. Rao AV, Rao LG. Carotenoids and human health. Pharmacol Res. 2007;55:207–16. Ärnlöv J, Zethelius B, Risérus U, Basu S, Berne C, Vessby B, Alfthan G, Helmersson J. Serum and dietary beta-carotene and alpha-tocopherol and incidence of type 2 diabetes mellitus in a community-based study of 57. Wallström P, Wirfält E, Lahmann PH, Gullberg B, Janzon L, Berglund G. Serum concentrations of beta-carotene and alpha-tocopherol are associated with diet, smoking, and general and central adiposity. Am J Clin Nutr. 2001;73:777–85. 58. Rock CL, Lovalvo JL, Emenhiser C, Ruffin MT, Flatt SW, Schwartz SJ. Bioavailability of beta-carotene is lower in raw than in processed carrots and spinach in women. J Nutr. 1998;128:913–6. 59. van Het Hof KH, West CE, Weststrate JA, Hautvast JG. Dietary factors that affect the bioavailability of carotenoids. J Nutr. 2000;130:503–6. 60. Rock CL, Swendseid ME. Plasma beta-carotene response in humans after meals supplemented with dietary pectin. Am J Clin Nutr. 1992;55 (1):96–9. 61. Ferrucci L, Perry JR, Matteini A, Perola M, Tanaka T, Silander K, Rice N, Melzer D, Murray A, Cluett C, et al. Common variation in the betacarotene 15,15’-monooxygenase 1 gene affects circulating levels of carotenoids: a genome-wide association study. Am J Hum Genet. 2009; 84:123–33. 62. Hendrickson SJ, Hazra A, Chen C, Eliassen AH, Kraft P, Rosner BA, Willett WC. b-Carotene 15,15’-monooxygenase 1 single nucleotide polymorphisms in relation to plasma carotenoid and retinol concentrations in women of European descent. Am J Clin Nutr. 2012;96: 1379–89. 63. Borel P. Genetic variations involved in interindividual variability in carotenoid status. Mol Nutr Food Res. 2012;56:228–40. Downloaded from jn.nutrition.org at LANDSPITALI UNIVERSITY HOSPITAL LIBRARY on August 21, 2013 1390 Magnusdottir et al. Paper II European Journal of Clinical Nutrition (2014), 1–6 & 2014 Macmillan Publishers Limited All rights reserved 0954-3007/14 www.nature.com/ejcn ORIGINAL ARTICLE Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole-grain rye intake, is associated to insulin sensitivity: a randomized study OK Magnusdottir1,2, R Landberg3,4, I Gunnarsdottir1,2, L Cloetens5, B Åkesson5,6, M Landin-Olsson7, F Rosqvist8, D Iggman8,9, U Schwab10,11, K-H Herzig12,13, MJ Savolainen14,15, L Brader16, K Hermansen16, M Kolehmainen10,17, K Poutanen10,17, M Uusitupa10,18, I Thorsdottir1,2 and U Risérus8 BACKGROUND/OBJECTIVES: Few studies have used biomarkers of whole-grain intake to study its relation to glucose metabolism. We aimed to investigate the association between plasma alkylresorcinols (AR), a biomarker of whole-grain rye and wheat intake, and glucose metabolism in individuals with metabolic syndrome (MetS). SUBJECTS/METHODS: Participants were 30–65 years of age, with body mass index 27–40 kg/m2 and had MetS without diabetes. Individuals were recruited through six centers in the Nordic countries and randomized to a healthy Nordic diet (ND, n ¼ 96), rich in whole-grain rye and wheat, or a control diet (n ¼ 70), for 18–24 weeks. In addition, associations between total plasma AR concentration and C17:0/C21:0 homolog ratio as an indication of the relative whole-grain rye intake, and glucose metabolism measures from oral glucose tolerance tests were investigated in pooled (ND þ control) regression analyses at 18/24 weeks. RESULTS: ND did not improve glucose metabolism compared with control diet, but the AR C17:0/C21:0 ratio was inversely associated with fasting insulin concentrations (P ¼ 0.002) and positively associated with the insulin sensitivity indices Matsuda ISI (P ¼ 0.026) and disposition index (P ¼ 0.022) in pooled analyses at 18/24 weeks, even after adjustment for confounders. The AR C17:0/C21:0 ratio was not significantly associated with insulin secretion indices. Total plasma AR concentration was not related to fasting plasma glucose or fasting insulin at 18/24 weeks. CONCLUSIONS: The AR C17:0/C21:0 ratio, an indicator of relative whole-grain rye intake, is associated with increased insulin sensitivity in a population with MetS. European Journal of Clinical Nutrition advance online publication, 19 February 2014; doi:10.1038/ejcn.2014.12 Keywords: alkylresorcinols; insulin sensitivity; whole grain; rye INTRODUCTION Whole-grain and cereal fiber consumption has been inversely associated with the risk of type 2 diabetes and insulin sensitivity in observational studies,1–5 but findings have been inconsistent in intervention studies.6–11 A possible reason for the inconclusive results seen in intervention studies may be inadequate dietary compliance. Nutritional science relies on accurate dietary information to assess compliance to dietary interventions and to investigate diet–disease associations. However, most dietary assessment instruments rely to some extent on self-reporting, which is prone to measurement errors.12,13 The use of dietary biomarkers could be one way to improve the validity of dietary assessment in combination with, or as an alternative for, selfreports or as a tool for secondary analysis, where noncompliant individuals can be excluded on the basis of the biomarker measurements. Alkylresorcinols (AR) have been proposed as biomarkers for whole-grain and bran intake of wheat and rye14–16 and for cereal fiber intake in populations with a high intake of whole-grain wheat and rye.17 AR are almost exclusively present in the outer parts of wheat and rye grains and not in refined white flour.18,19 Controlled whole-grain and bran intervention studies show that plasma AR concentration increases proportionally with AR and whole-grain intake16,20–24 and can be used as a biomarker of compliance to a whole-grain diet.15,25 Previously, we have demonstrated that AR are useful biomarkers for whole-grain wheat and rye intake as part of a healthy Nordic diet (ND).26 Total plasma AR have been associated with lower body mass index 1 Unit for Nutrition Research, Landspitali-The National University Hospital of Iceland, Reykjavı́k, Iceland; 2Faculty of Food Science and Nutrition and School of Health Sciences, University of Iceland, Reykjavı́k, Iceland; 3Department of Food Science, BioCenter, Swedish University of Agricultural Sciences, Uppsala, Sweden; 4Division of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden; 5Biomedical Nutrition, Pure and Applied Biochemistry, Lund University, Lund, Sweden; 6Department of Clinical Nutrition, Skåne University Hospital, Lund, Sweden; 7Department of Endocrinology, Skåne University Hospital, Lund, Sweden; 8Department of Public Health and Caring Sciences, Clinical Nutrition and Metabolism, Uppsala University, Uppsala, Sweden; 9Center for Clinical Research Dalarna, Falun, Sweden; 10Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland; 11Institute of Clinical Medicine, Internal Medicine, Kuopio University Hospital, Kuopio, Finland; 12 Department of Physiology and Biocenter of Oulu, Institute of Biomedicine, Oulu University, Oulu, Finland; 13Department of Psychiatry, Kuopio University Hospital, Kuopio, Finland; 14Department of Internal Medicine and Biocenter Oulu, Institute of Clinical Medicine, University of Oulu, Oulu, Finland; 15Clinical Research Center, Oulu University Hospital, Oulu, Finland; 16Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Aarhus, Denmark; 17VTT, Technical Research Centre of Finland, Espoo and Kuopio, Kuopio, Finland and 18Research Unit, Kuopio University Hospital, Kuopio, Finland. Correspondence: OK Magnusdottir, Unit for Nutrition Research, Landspitali-The National University Hospital of Iceland, Reykjavik IS-101, Iceland. E-mail: [email protected] Received 4 October 2013; revised 6 December 2013; accepted 4 January 2014 Alkylresorcinols ratio and insulin sensitivity OK Magnusdottir et al 2 (BMI) in older adults, but no associations between AR and glucose metabolism were found.27 The ratio of the AR homologs C17:0 and C21:0 differs between cereal species and can be used to differentiate between the whole-grain sources in food products.19 The ratio of C17:0/C21:0 is typically 0.1 in common wheat and 1.0 in rye, and the relative intake of whole-grain rye and wheat is to some extent reflected by this ratio in plasma samples.23 In the present study the aim was to investigate the association between plasma AR, as a biomarker for whole-grain wheat and rye intake, and glucose metabolism in a population with features of metabolic syndrome participating in a randomized dietary intervention.28 Furthermore, the associations between the AR C17:0/C21:0 ratio, as a marker of relative whole-grain bran/rye intake, and glucose and insulin parameters were analyzed. the participants, that is at 1–2 weeks intervals, and the subjects also returned 4-day food records at baseline, week 12 and 18/24 weeks. Biochemical measurements Automated clinical chemistry analyzers and routine clinical chemistry methods were used to measure cholesterol, triglycerides and high-density lipoprotein cholesterol. The equipment and system reagents were from Siemens Healthcare Diagnostics (Tarrytown, NY, USA), Thermo Fisher Scientific (Vantaa, Finland), Roche Diagnostics GmbH (Mannheim, Germany) (in two laboratories), Toshiba (Tochigi, Japan), Abbott Laboratories (Irving, TX, USA), Ortho-Clinical Diagnostics and Johnson & Johnson (New Brunswick, NJ, USA).28 Low-density lipoprotein cholesterol concentrations were calculated using Friedewald’s formula (low-density lipoprotein ¼ total cholesterol high-density lipoprotein cholesterol (triglycerides/5)). Two-hour oral glucose tolerance test MATERIALS AND METHODS The current study is part of the SYSDIET study, a randomized controlled multicenter parallel group study with balanced randomization (1:1) performed in six intervention centers (Kuopio and Oulu, Finland; Lund and Uppsala, Sweden; Aarhus, Denmark and Reykjavik, Iceland) examining the health effects of a ND in a population with metabolic syndrome. All local ethical committees approved the SYSDIET study protocol which followed the Helsinki declaration guidelines. Informed written consent was obtained from all participants. The study design, inclusion and exclusion criteria, study diets, the main measurements and analyses performed at each time point and main results have been previously reported in detail.28 A 2-h oral glucose tolerance test was performed at baseline, week 12 and at 18/24 weeks. The subjects were instructed to fast for 12 h before the oral glucose tolerance test. Fasting samples for plasma glucose, plasma insulin and triglycerides were collected at time point 0, before the glucose solution was ingested. The subjects were asked to drink 75 g of D( þ ) -glucose anhydrous Ph.Eur and 1.6 g of citric acid dissolved in 300 ml water in o3 min. Blood samples for glucose and insulin analyses were collected 30 and 120 min after the glucose load. Analyses of plasma glucose were performed in local laboratories using standard operational procedures. Plasma insulin was analyzed at Aarhus University Hospital using the ELISA method from Dako Denmark A/S, Glostrup, Denmark, with a within-run and between-run CVo8% and o10%, respectively. Analysis of AR Study participants Participants in the study were primarily recruited through advertisement in newspapers, but also from previous clinical or epidemiological trials. In brief, the inclusion criteria were age 30–65 years, BMI 27–40 kg/m2 and two diagnostic criteria for metabolic syndrome outlined by the International Diabetes Federation,29 except type 2 diabetes. Study design Before randomization into the ND group and the control group, participants had a 1-month run-in period, during which all the participants followed their habitual diet. Owing to financial and logistic reasons, the study period was 18 weeks at four centers (Aarhus, Uppsala, Reykjavik and Oulu) and 24 weeks at two centers (Lund and Kuopio), where the intervention was initiated earlier.28 The major visits to the study clinics were on week 0 (baseline), week 12 and 18/24 weeks. The randomization was done by staff in each study center as described elsewhere.28 The staff in all study centers was trained to perform measurements in a similar way and the quality management protocol was available for all of the staff members. The allowed weight change during the study was aimed to be o±4 kg. Study diets The study diets were isocaloric. Nordic nutrition recommendations formed the basis of the diet in the ND group and the mean nutrient intake in the Nordic countries formed the basis for the control diet. Participants in the ND group were, among other things, advised to consume X25% of total energy as whole grain, of which X50% was comprised of rye, barley and oat that included X6 slices/day of bread with X6 g fiber/100 g (one slice ¼ 30–40 g of bread). Participants in the control group were asked to consume X25% of energy as refined grain, of which X90% was made from wheat. Key products were provided to the study participants in both groups. The individuals in the ND group received whole-grain products such as oats and barley, crisp breads, whole-grain wheat and rye breads. The individuals in the control group received low-fiber cereal products, for example breads with fiber content o6 g/100 g and ordinary pasta. The participants met a clinical nutritionist or a dietitian at all study visits. The study participants kept dietary records regarding the intake of cereal products. These records were checked at each visit to the study center, including the visits when the groceries were delivered to European Journal of Clinical Nutrition (2014) 1 – 6 Blood samples analyzed for AR were collected at baseline, week 12 and 18/24. Determination of AR in plasma samples were made in collaboration with the HELGA-Centre at Department of Food Science, BioCenter at Swedish University of Agricultural Sciences. A gas chromatography mass spectrometry method was used to quantify AR homologs C17:0–C25:0 in 0.2 ml plasma samples, as described previously.30 Quality control samples (n ¼ 4) were included in each batch to ensure appropriate within- and between-batch precision (CVo15%). Samples from each subject were analyzed within the same batch, and samples from different centers were allocated randomly between batches. Data pooling and statistical analysis All intervention centers entered their data into the Microsoft Office Excel 2007 (Windows) data form before pooling the data in a joint database maintained at VTT Technical Research Centre of Finland. Data was exported from the database and imported to SPSS (Statistical Package for the Social Sciences) for Windows, version 20.0 (SPSS Inc., Chicago, IL, USA) for statistical analysis. Level of significance in all analyses was set at Po0.05. Normality of variables was checked by inspection and by using the Kolmogorov–Smirnov test. Values are described as means and s.d. if normally distributed or as medians and interquartile range if not normally distributed. Independent sample t-tests or Mann–Whitney’s U-test were used to compare data between groups. When comparing values at baseline to week 18/24 values, paired samples t-test or the Wilcoxon signed-rank test was used. From data of the oral glucose tolerance test insulinogenic index (IGI), the ratio of the 30-min areas under the curve for insulin and glucose (InsAUC030/GluAUC0-30) and Stumvoll was calculated for evaluation of early-phase insulin secretion (IGI ¼ (insulin 30 min insulin 0 min)/(glucose 30 min glucose 0 min) and Stumvoll ¼ 1283 þ 1.829 insulin 30 min (pmol/l) 138.7 glucose 30 min (mmol/l) þ 3.772 insulin 0 min (pmol/l)). IGI at 120 min (IGI 120) was calculated for late-phase insulin secretion (IGI 120 ¼ (insulin 120 min insulin 0 min)/(glucose 120 min glucose 0 min)). Matsuda insulin sensitivity index (ISI) and disposition index (DI) were calculated for evaluation of insulin sensitivity (Matsuda ISI ¼ (10 000/ O((glucose 0 min insulin 0 min) (arithmetic mean of glucose at 0, 30 and 120 min arithmetic mean of insulin at 0, 30 and 120 min)))31 and DI ¼ Matsuda ISI IGI). Secondary analyses were performed with data from the ND and control groups combined and associations between total AR and the AR C17:0/C21:0 ratio at 18/24 weeks, and various glucose and insulin & 2014 Macmillan Publishers Limited Alkylresorcinols ratio and insulin sensitivity OK Magnusdottir et al 3 outcomes were analyzed. As the distributions were skewed, logtransformation was performed on dependent variables for linear regression to better approximate normal distribution. Linear regressions for fasting plasma glucose, insulin, Matsuda, DI and IGI was calculated, with AR C17:0/C21:0 ratio, baseline value for the dependent variable, age, gender, BMI and smoking as factors in model 1. As rye often contains more AR compared with wheat, total plasma AR was added as a variable to model 2, in addition to the variables in model 1, to prevent confounding of high AR content. Spearman’s r was calculated to estimate the correlation between AR C17:0/C21:0 ratio and total plasma AR. RESULTS In total, 201 individuals started the intervention; 104 in the ND group and 97 in the control group. Individuals completing the trial were 96 in the ND group (92%) and 70 in the control group (72%). There was no difference in basic baseline characteristics between dropouts and those who completed the study (data not shown). Blood samples for AR analysis at week 0 and week 18/24 were obtained from 158 participants. Consistent with the isocaloric study design there was no change in total energy intake during the study and energy intake was not different between the groups. Hence, body weight remained constant during the intervention in both groups. The mean age of the participants was 54.5±8.2 years, body weight 91.2±13.5 kg, BMI 31.7±3.2 kg/m2 and waist circumference 104.3±9.7 cm. Sixty-three percent of the participants were classified as obese (BMIX30 kg/m2) and 37% as overweight (BMI 25–29.99 kg/m2) at baseline. Ninety-eight individuals (59%) had fasting plasma glucose X5.6 mmol/l at baseline and 161 (97%) had metabolic syndrome. There were no differences in baseline characteristics between the ND and control groups.28 Both total plasma AR concentration and the ratio of C17:0/C21:0 were significantly higher in the ND group compared with the control group at 18/24 weeks (Table 1). Fasting plasma glucose was 5.7±0.9 mmol/l in both groups at baseline and did not change significantly during the intervention period. There were no significant differences between the ND group and control group in concentrations of fasting glucose, 2 h glucose (results described elsewhere28), fasting insulin, 2 h insulin, other calculated glucose metabolism parameters or the changes in these values during the intervention (Table 1). All participants (both ND and control groups) were combined for secondary analyses on the associations between AR and glucose metabolism. Total plasma AR concentration at 18/24 weeks was not significantly correlated with concentrations of fasting plasma glucose (P ¼ 0.51) or insulin (P ¼ 0.57) when adjusted for baseline value of the dependent variable, age, gender, BMI and smoking. Combining the data from both weeks 12 and 18/24 did not change the results. Interestingly, the AR ratio C17:0/C21:0, reflecting the proportion of whole-grain rye to wholegrain rye þ whole-grain wheat, was inversely associated with fasting insulin concentrations when adjusted for baseline value of insulin, age, gender, BMI and smoking (model 1, P ¼ 0.016) and also when total plasma AR was added to the regression model (model 2, P ¼ 0.011) (Table 2). Total plasma AR and the AR C17:0/C21:0 ratio was weakly correlated (r ¼ 0.20, P ¼ 0.012). The AR C17:0/C21:0 ratio was also positively associated with the insulin sensitivity indices Matsuda ISI (P ¼ 0.005 and P ¼ 0.004 for model 1 and 2, respectively) and DI (P ¼ 0.022 and P ¼ 0.023 for model 1 and 2, respectively) (Table 2). The AR C17:0/C21:0 ratio was, however, not associated with fasting plasma glucose (P ¼ 0.12 and P ¼ 0.11 for model 1 and 2, respectively), or the indices for early and late insulin secretion, that is, IGI (P ¼ 0.54 and P ¼ 0.59 for model 1 and 2, respectively), InsAUC0-30/GluAUC0-30 (P ¼ 0.65 and P ¼ 0.54 for model 1 and 2, respectively), Stumvoll or IGI 120 (P ¼ 0.93 and P ¼ 0.99 for model 1 and 2, respectively) and IGI 120 (P ¼ 0.11 and P ¼ 0.12 for model 1 and 2, respectively). DISCUSSION In the present study, consuming a ND rich in whole-grain cereals for either 18 or 24 weeks did not improve glucose metabolism compared with a control diet. When values from both groups were pooled, no association between total plasma AR concentration and glucose metabolism parameters were observed. There are conflicting findings from previous intervention studies evaluating the effects of whole-grain intake on glucose and insulin concentrations and calculated parameters of insulin sensitivity8,11,27,32 presumably owing to differences in the type, amount, structure and preparation of whole-grain products consumed33 and different methodologies used for glucose and insulin analyses.7 Poor compliance could also contribute to these diverse findings, although good adherence has been indicated in most studies using weighed food records. In a whole-grain intervention study by Kristensen et al.9 the effects on body composition were clearer after excluding noncompliant subjects based on AR concentration, although food records indicated good compliance. Hence, without an objective measurement of Table 1. Total plasma alkylresorcinol (AR) concentration, AR C17:0/C21:0 ratio and insulin variables for the healthy Nordic diet group (ND) and control group at baseline and after the 18/24-week intervention Healthy ND Week 0 a,b Total plasma AR , nmol/l C17:0/C21:0 ratioa,b Fasting insulin, pmol/l Insulin 2 h, pmol/l Matsuda ISI DI IGI InsAUC0-30/GluAUC0-30 Stumvoll IGI 120 73 0.3 51 210 5.8 392 68 21.6 733 116 (88) (0.2) (33) (180) (4.0) (432) (58) (13.3) (441) (394) Control Week 18/24 106 0.3 52 251 5.7 411 69 24.0 818 100 c (108) (0.2) (42) (225) (4.9) (427) (64) (15.2) (398) (255) Week 0 Week 18/24 71 (70) 0.3 (0.2) 58 (38) 218±255 4.9 (3.8) 400 (461) 87 (85) 24.0 (18.9) 828 (526) 104 (311) 61 (40)c,d 0.2 (0.2)d 61 (43) 236±237 4.7 (3.6) 377 (501) 83 (89) 26.0 (15.1) 850 (567) 120 (276) Abbreviations: AR, alkylresorcinols; DI, disposition index; IGI, insulinogenic index; InsAUC0-30/GluAUC0-30, the ratio of the 30-min areas under the curve for insulin and glucose; IGI 120, insulinogenic index at 120 min. Data are presented as medians (interquartile range), n ¼ 166, ND group n ¼ 96 and control group n ¼ 70. aData on total plasma AR has been published previously.31 bn ¼ 158, ND group n ¼ 93 and control group n ¼ 65. cSignificantly different from week 0 in that group, Po0.05. dSignificantly different from the ND group at that time point, Po0.05. & 2014 Macmillan Publishers Limited European Journal of Clinical Nutrition (2014) 1 – 6 Alkylresorcinols ratio and insulin sensitivity OK Magnusdottir et al 4 Table 2. Regression models for fasting plasma glucose, insulin, insulin sensitivity indices and indices of insulin secretion (log-transformed values) Dependent variable at 18/24 weeks Fasting glucose Model 1 Model 2 Ba (95% CI) Independent variable at 18/24 weeks AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.049 ( 0.111 to 0.014) 0.051 ( 0.113 to 0.012) 0.000 (0.000 to 0.000) AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.425 ( 0.769 to 0.082) 0.445 ( 0.788 to 0.102) 0.0001 (0.000 to 0.001) AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.234 (0.071 to 0.396) 0.240 (0.077 to 0.404) 0.000 ( 0.001 to 0.000) AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.768 (0.111 to 1.426) 0.773 (0.107 to 1.433) 0.000 ( 0.002 to 0.001) AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.072 ( 0.161 to 0.305) 0.065 ( 0.170 to 0.299) 0.000 (0.000 to 0.001) InsAUC-30/GluAUC-30 Model 1 AR C17:0/C21:0 Model 2 AR C17:0/C21:0 Total AR 0.030 ( 0.162 to 0.101) 0.041 ( 0.172 to 0.090) 0.000 (0.000 to 0.001) Fasting insulin Model 1 Model 2 Matsuda Model 1 Model 2 DI Model 1 Model 2 IGI 0 min Model 1 Model 2 Stumvoll Model 1 Model 2 Log IGI 120 min Model 1 Model 2 AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.007 ( 0.147 to 0.161) 0.001 ( 0.155 to 0.154) 0.000 (0.000 to 0.001) AR C17:0/C21:0 AR C17:0/C21:0 Total AR 0.309 ( 0.683 to 0.066) 0.301 ( 0.680 to 0.079) 0.000 ( 0.001 to 0.001) Abbreviations: AR, alkylresorcinols; CI, confidence interval; DI, disposition index; IGI, insulinogenic index; InsAUC0-30/GluAUC0-30, the ratio of the 30 min areas under the curve for insulin and glucose; IGI 120, insulinogenic index at 120 min. Model 1: adjusted for baseline value of the dependent variable, age, gender, BMI and smoking. Model 2: adjusted for baseline value of the dependent variable, age, gender, BMI and smoking. aB reflects the change in the log-transformed outcome variable for each 1-unit change in the independent variable. compliance, such as biomarkers, it is difficult to rule out the possibility of low adherence to the intervention diet. The most interesting results of the present study are the inverse associations between the AR C17:0/C21:0 ratio and fasting plasma insulin concentration and insulin sensitivity indices, Matsuda ISI and DI, that is, the higher the ratio the better was the insulin sensitivity. Matsuda ISI is the most commonly used index to illustrate peripheral insulin sensitivity, of mostly peripheral muscle tissue insulin sensitivity.34 We also found an association between DI and AR C17:0/C21:0, which suggests that AR may also have a role with improvement of beta-cell function when the respective insulin sensitivity has been taken into account, as DI reflects insulin secretion but takes into account the insulin sensitivity effect.34 European Journal of Clinical Nutrition (2014) 1 – 6 Although the ratio of C17:0/C21:0 is B1.0 in rye, the ratio is somewhat lower in plasma samples, typically around 0.6–0.8 in fasting samples after a whole-grain rye intervention for 1–8 weeks.21,23 In populations where rye is not commonly consumed the ratios are generally lower and a ratio above 0.15 is usually indicative of some rye intake.35 The ratio in free-living Swedish population, where rye consumption is appreciable, is 0.5.36 The reason for lower ratio in plasma compared with what is observed in whole-grain bran/rye foods is likely due to faster metabolism of the AR homolog C17:0,37,38 but further insight into absorption and metabolism of the different homologs is needed to explain to what extent factors other than whole-grain rye and wheat affect the C17:0/C21:0 ratio.35 The inverse association between C17:0/C21:0 and fasting insulin and insulin sensitivity indices observed in the present study is in line with previous findings showing positive effects of higher rye consumption on fasting plasma insulin and to a less extent on fasting plasma glucose8 supporting the use of the ratio of the homologs C17:0 and C21:0 as an indicator of whole-grain/bran rye intake. It is, however, important to note that the AR C17:0/C21:0 ratio reflects only the relative whole-grain/bran rye consumption of total whole-grain wheat and rye intake, not total rye intake. In the present study the advised whole-grain rye intake was high in relation to other sources of AR and therefore the ratio should reflect the total rye intake reasonably well. Rye was mostly consumed in the form of bread. Several intervention studies have shown diminished levels of postprandial insulin following a rye bread meal,39–41 possibly explained by structural differences between rye and whole-wheat breads.41,42 The lower postprandial insulin response after rye bread consumption as compared with wheat bread may be due to the difference in structure rather than fiber content in the whole grain, as low-fiber rye bread showed the same results as high-fiber rye bread consumption.43,44 Moreover, an improved early insulin response has been shown after rye bread interventions.43,44 In the present study, there was, however, no significant association between the AR C17:0/C21:0 ratio and early or later insulin secretion phase as measured by insulin secretion indices. One of the strengths of the present study is that AR is used as a dietary biomarker for whole-grain rye and wheat intake instead of self-reported estimates. Previously, we have demonstrated that AR is a good biomarker for the ND under intervention conditions with moderate-to-good reliability.26 Also, reliable and comparable methods were used in six centers in four Nordic countries and the same study protocol was followed in all centers. However, it is important to note that AR is only a biomarker for whole-grain rye and wheat consumption, and therefore other markers are warranted in order to monitor whole-grain intake from other cereal sources. CONCLUSION The results support the use of the ratio of the homologs C17:0 and C21:0 as an indicator of rye intake in intervention studies where a diet rich in whole-grain rye is being investigated and also suggests that increased proportion of whole-grain rye in the diet may be associated with lower fasting insulin concentrations and increased insulin sensitivity. CONFLICT OF INTEREST The authors declare no conflict of interest. ACKNOWLEDGEMENTS We express sincere gratitude to the participants taking part in the study. We also thank Vanessa de Mello Laaksonen and Maria Lankinen for help with calculations and all staff involved in the intervention. The study was supported by NordForsk NCoE in & 2014 Macmillan Publishers Limited Alkylresorcinols ratio and insulin sensitivity OK Magnusdottir et al 5 Food Nutrition and Health: Project 070014 SYSDIET (Systems biology in controlled dietary interventions and cohort studies). This study was also funded by the Academy of Finland, Finnish Diabetes Research Foundation, Finnish Foundation for Cardiovascular Research, The Sigrid Juselius Foundation and EVO funding from Kuopio University Hospital (Finland); the Druvan Foundation, ESPEN, Skåne County Council Research and Development Foundation, Swedish Research Council for Health, EXODIAB (Excellence of Diabetes Research in Sweden), The Heart-Lung Foundation, Diabetesfonden and Foundation Cerealia (Sweden); The Danish Obesity Research Centre (DanORC, www.danorc.dk), The Danish Council for Strategic Research (DairyHealth, BioFunCarb) (Denmark), The Agricultural Productivity Fund (Iceland) and the Icelandic Research Fund for graduate students (1206880061). REFERENCES 1 Ye EQ, Chacko SA, Chou EL, Kugizaki M, Liu S. Greater whole-grain intake is associated with lower risk of type 2 diabetes, cardiovascular disease, and weight gain. J Nutr 2012; 142: 1304–1313. 2 de Munter JS, Hu FB, Spiegelman D, Franz M, van Dam RM. Whole grain, bran, and germ intake and risk of type 2 diabetes: a prospective cohort study and systematic review. PLoS Med 2007; 4: e261. 3 Montonen J, Knekt P, Jarvinen R, Aromaa A, Reunanen A. Whole-grain and fiber intake and the incidence of type 2 diabetes. Am J Clin Nutr 2003; 77: 622–629. 4 Fung TT, Hu FB, Pereira MA, Liu S, Stampfer MJ, Colditz GA et al. Whole-grain intake and the risk of type 2 diabetes: a prospective study in men. Am J Clin Nutr 2002; 76: 535–540. 5 McKeown NM, Meigs JB, Liu S, Wilson PW, Jacques PF. Whole-grain intake is favorably associated with metabolic risk factors for type 2 diabetes and cardiovascular disease in the Framingham Offspring Study. Am J Clin Nutr 2002; 76: 390–398. 6 Tighe P, Duthie G, Vaughan N, Brittenden J, Simpson WG, Duthie S et al. Effect of increased consumption of whole-grain foods on blood pressure and other cardiovascular risk markers in healthy middle-aged persons: a randomized controlled trial. Am J Clin Nutr 2010; 92: 733–740. 7 Giacco R, Lappi J, Costabile G, Kolehmainen M, Schwab U, Landberg R et al. Effects of rye and whole wheat versus refined cereal foods on metabolic risk factors: a randomised controlled two-centre intervention study. Clin Nutr 2013; 32: 941–949. 8 Landberg R, Andersson SO, Zhang JX, Johansson JE, Stenman UH, Adlercreutz H et al. Rye whole grain and bran intake compared with refined wheat decreases urinary C-peptide, plasma insulin, and prostate specific antigen in men with prostate cancer. J Nutr 2010; 140: 2180–2186. 9 Kristensen M, Toubro S, Jensen MG, Ross AB, Riboldi G, Petronio M et al. Whole grain compared with refined wheat decreases the percentage of body fat following a 12-week, energy-restricted dietary intervention in postmenopausal women. J Nutr 2012; 142: 710–716. 10 Brownlee IA, Moore C, Chatfield M, Richardson DP, Ashby P, Kuznesof SA et al. Markers of cardiovascular risk are not changed by increased whole-grain intake: the WHOLEheart study, a randomised, controlled dietary intervention. Br J Nutr 2010; 104: 125–134. 11 Andersson A, Tengblad S, Karlstrom B, Kamal-Eldin A, Landberg R, Basu S et al. Whole-grain foods do not affect insulin sensitivity or markers of lipid peroxidation and inflammation in healthy, moderately overweight subjects. J Nutr 2007; 137: 1401–1407. 12 Lissner L, Troiano RP, D, MidthuneHeitmann BL, Kipnis V, Subar AF et al. OPEN about obesity: recovery biomarkers, dietary reporting errors and BMI. Int J Obes 2007; 31: 956–961. 13 Bingham S, Luben R, Welch A, Tasevska N, Wareham N, Khaw KT. Epidemiologic assessment of sugars consumption using biomarkers: comparisons of obese and nonobese individuals in the European prospective investigation of cancer Norfolk. Cancer Epidemiol Biomarkers Prev 2007; 16: 1651–1654. 14 Ross AB, Bourgeois A, Macharia HN, Kochhar S, Jebb SA, Brownlee IA et al. Plasma alkylresorcinols as a biomarker of whole-grain food consumption in a large population: results from the WHOLEheart Intervention Study. Am J Clin Nutr 2012; 95: 204–211. 15 Ross AB, Kamal-Eldin A, Aman P. Dietary alkylresorcinols: absorption, bioactivities, and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr Rev 2004; 62: 81–95. 16 Landberg R, Kamal-Eldin A, Andersson A, Vessby B, Aman P. Alkylresorcinols as biomarkers of whole-grain wheat and rye intake: plasma concentration and intake estimated from dietary records. Am J Clin Nutr 2008; 87: 832–838. 17 Aubertin-Leheudre M, Koskela A, Marjamaa A, Adlercreutz H. Plasma alkylresorcinols and urinary alkylresorcinol metabolites as biomarkers of cereal fiber intake in Finnish women. Cancer Epidemiol Biomarkers Prev 2008; 17: 2244–2248. 18 Ross AB, Shepherd MJ, Schupphaus M, Sinclair V, Alfaro B, Kamal-Eldin A et al. Alkylresorcinols in cereals and cereal products. J Agric Food Chem 2003; 51: 4111–4118. 19 Chen Y, Ross AB, Aman P, Kamal-Eldin A. Alkylresorcinols as markers of whole grain wheat and rye in cereal products. J Agric Food Chem 2004; 52: 8242–8246. & 2014 Macmillan Publishers Limited 20 Montonen J, Landberg R, Kamal-Eldin A, Aman P, Knueppel S, Boeing H et al. Reliability of fasting plasma alkylresorcinol concentrations measured 4 months apart. Eur J Clin Nutr 2010; 64: 698–703. 21 Landberg R, Aman P, Friberg LE, Vessby B, Adlercreutz H, Kamal-Eldin A. Dose response of whole-grain biomarkers: alkylresorcinols in human plasma and their metabolites in urine in relation to intake. Am J Clin Nutr 2009; 89: 290–296. 22 Landberg R, Kamal-Eldin A, Andersson SO, Johansson JE, Zhang JX, Hallmans G et al. Reproducibility of plasma alkylresorcinols during a 6-week rye intervention study in men with prostate cancer. J Nutr 2009; 139: 975–980. 23 Linko-Parvinen AM, Landberg R, Tikkanen MJ, Adlercreutz H, Penalvo JL. Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. J Nutr 2007; 137: 1137–1142. 24 Ross AB, Pineau N, Kochhar S, Bourgeois A, Beaumont M, Decarli B. Validation of a FFQ for estimating whole-grain cereal food intake. Br J Nutr 2009; 102: 1547–1551. 25 Ocke MC, Kaaks RJ. Biochemical markers as additional measurements in dietary validity studies: application of the method of triads with examples from the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr 1997; 65(4 Suppl): 1240S–1245S. 26 Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, Onning G et al. Plasma alkylresorcinols reflect important whole-grain components of a healthy Nordic diet. J Nutr 2013; 143: 1383–1390. 27 Ma J, Ross AB, Shea MK, Bruce SJ, Jacques PF, Saltzman E et al. Plasma alkylresorcinols, biomarkers of whole-grain intake, are related to lower BMI in older adults. J Nutr 2012; 142: 1859–1864. 28 Uusitupa M, Hermansen K, Savolainen MJ, Schwab U, Kolehmainen M, Brader L et al. Effects of an isocaloric healthy Nordic diet on insulin sensitivity, lipid profile and inflammation markers in metabolic syndrome-a randomized study (SYSDIET). J Internal Med 2013; 274: 52–66. 29 Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009; 120: 1640–1645. 30 Landberg R, Åman P, Kamal-Eldin A. A rapid gas chromatography-mass spectrometry method for quantification of alkylresorcinols in human plasma. Anal Biochem 2009; 385: 7–12. 31 de Mello VD, Lindstrom J, Eriksson J, Ilanne-Parikka P, Keinanen-Kiukaanniemi S, Sundvall J et al. Insulin secretion and its determinants in the progression of impaired glucose tolerance to type 2 diabetes in impaired glucose-tolerant individuals: the Finnish Diabetes Prevention Study. Diabetes Care 2012; 35: 211–217. 32 Davy BM, Davy KP, Ho RC, Beske SD, Davrath LR, Melby CL. High-fiber oat cereal compared with wheat cereal consumption favorably alters LDL-cholesterol subclass and particle numbers in middle-aged and older men. Am J Clin Nutr 2002; 76: 351–358. 33 Hallfrisch J, Facn, Behall KM. Mechanisms of the effects of grains on insulin and glucose responses. J Am Coll Nutr 2000; 19(3 Suppl): 320S–325S. 34 Stancakova A, Kuulasmaa T, Paananen J, Jackson AU, Bonnycastle LL, Collins FS et al. Association of 18 confirmed susceptibility loci for type 2 diabetes with indices of insulin release, proinsulin conversion, and insulin sensitivity in 5,327 nondiabetic Finnish men. Diabetes 2009; 58: 2129–2136. 35 Ross AB. Present status and perspectives on the use of alkylresorcinols as biomarkers of wholegrain wheat and rye intake. J Nutr Metab 2012; 2012: 462967. 36 Landberg R, Aman P, Hallmans G, Johansson I. Long-term reproducibility of plasma alkylresorcinols as biomarkers of whole-grain wheat and rye intake within Northern Sweden Health and Disease Study Cohort. Eur J Clin Nutr 2013; 67: 259–263. 37 Landberg R, Linko AM, Kamal-Eldin A, Vessby B, Adlercreutz H, Aman P. Human plasma kinetics and relative bioavailability of alkylresorcinols after intake of rye bran. J Nutr 2006; 136: 2760–2765. 38 Marklund M, Stromberg EA, Hooker AC, Hammarlund-Udenaes M, Aman P, Landberg R et al. Chain length of dietary alkylresorcinols affects their in vivo elimination kinetics in rats. J Nutr 2013; 143: 1573–1578. 39 Heinonen MV, Karhunen LJ, Chabot ED, Toppinen LK, Juntunen KS, Laaksonen DE et al. Plasma ghrelin levels after two high-carbohydrate meals producing different insulin responses in patients with metabolic syndrome. Regul Pept 2007; 138: 118–125. 40 Leinonen K, Liukkonen K, Poutanen K, Uusitupa M, Mykkanen H. Rye bread decreases postprandial insulin response but does not alter glucose response in healthy Finnish subjects. Eur J Clin Nutr 1999; 53: 262–267. European Journal of Clinical Nutrition (2014) 1 – 6 Alkylresorcinols ratio and insulin sensitivity OK Magnusdottir et al 6 41 Juntunen KS, Laaksonen DE, Autio K, Niskanen LK, Holst JJ, Savolainen KE et al. Structural differences between rye and wheat breads but not total fiber content may explain the lower postprandial insulin response to rye bread. Am J Clin Nutr 2003; 78: 957–964. 42 Rosen LA, Silva LO, Andersson UK, Holm C, Ostman EM, Bjorck IM. Endosperm and whole grain rye breads are characterized by low post-prandial insulin response and a beneficial blood glucose profile. Nutr J 2009; 8: 42. European Journal of Clinical Nutrition (2014) 1 – 6 43 Juntunen KS, Laaksonen DE, Poutanen KS, Niskanen LK, Mykkanen HM. High-fiber rye bread and insulin secretion and sensitivity in healthy postmenopausal women. Am J Clin Nutr 2003; 77: 385–391. 44 Laaksonen DE, Toppinen LK, Juntunen KS, Autio K, Liukkonen KH, Poutanen KS et al. Dietary carbohydrate modification enhances insulin secretion in persons with the metabolic syndrome. Am J Clin Nutr 2005; 82: 1218–1227. & 2014 Macmillan Publishers Limited Paper III Title: Whole grain rye intake, reflected by a biomarker, is associated with favorable blood lipid outcomes in subjects with the metabolic syndrome – a randomized study Authors: Ola Kally Magnusdottir1, Rikard Landberg2,3, Ingibjorg Gunnarsdottir1, Lieselotte Cloetens4, Björn Åkesson4,5, Fredrik Rosqvist6, Ursula Schwab7,8, Karl-Heinz Herzig9, Janne Hukkanen10, Markku J Savolainen10, Lea Brader11, Kjeld Hermansen11, Marjukka Kolehmainen7,12, Kaisa Poutanen7,12, Matti Uusitupa7,13, Ulf Risérus6, Inga Thorsdottir1 1 Unit for Nutrition Research, Landspitali -The National University Hospital of Iceland and Faculty of Food Science and Nutrition, School of Health Sciences, University of Iceland, Reykjavík, Iceland 2 Department of Food Science, BioCenter, Swedish University of Agricultural Sciences, Uppsala, Sweden 3 Division of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden 4 Biomedical Nutrition, Pure and Applied Biochemistry, Lund University, Lund, Sweden 5 Department of Clinical Nutrition, Skåne University Hospital, Lund, Sweden 6 Department of Public Health and Caring Sciences, Clinical Nutrition and Metabolism, Uppsala University, Uppsala, Sweden 7 Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland 8 Institute of Clinical Medicine, Internal Medicine, Kuopio University Hospital, Kuopio, Finland 9 Institute of Biomedicine, Department of Physiology and Biocenter, Oulu University, Oulu, Finland and Department of Psychiatry, Kuopio University Hospital, Kuopio, Finland 10 Department of Internal Medicine and Biocenter, Institute of Clinical Medicine, University of Oulu, Oulu, Finland and Clinical Research Center, Oulu University Hospital, Oulu, Finland. 11 Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Aarhus, Denmark 12 VTT Technical Research Centre of Finland, Espoo and Kuopio, Finland 13 Research Unit, Kuopio University Hospital, Kuopio, Finland Corresponding author: Ola Kally Magnusdottir E-mail: [email protected] Address: Eiriksgata 29, IS-101 Reykjavik, Iceland Telephone: +354 543 8421 Fax: +354 543 4824 Abstract Background and aim Few studies have explored the possible plasma cholesterol lowering effects of rye consumption. The aim of this secondary analysis in the SYSDIET study was to investigate the association between plasma alkylresorcinols (AR), a biomarker for whole grain wheat and rye intake, and blood lipid concentrations in a population with metabolic syndrome. Furthermore, we analyzed the associations between the AR C17:0/C21:0 ratio, a suggested marker of the relative intake of whole grain/bran rye, and blood lipid concentrations. Methods Participants were 30-65 years of age, with body mass index (BMI) 27-40 kg/m2 and had metabolic syndrome. Individuals were recruited through six centers in the Nordic countries and randomized either to a healthy Nordic diet (ND, n=93), rich in whole grain rye and wheat, as well as berries, fruits and vegetables, rapeseed oil, three fish meals per week and low-fat dairy products, or a control diet (n=65) for 18/24 weeks. Associations between total plasma AR concentration and C17:0/C21:0 homologue ratio and blood lipids were investigated in pooled (ND + control group) regression analyses at 18/24 weeks adjusted for baseline value for the dependent variable, age, BMI and statin use. Results When adjusted for confounders, total plasma AR at 18/24 weeks was not significantly associated with blood lipids but the AR ratio C17:0/C21:0 was inversely associated with LDL cholesterol concentrations (B (95% CI): -0.41 (-0.80 to -0.02)), log LDL/HDL cholesterol ratio (-0.20 (-0.37 to -0.03)), log non-HDL cholesterol (-0.20 (-0.37 to -0.03)), log apolipoprotein B (-0.12 (-0.24 to 0.00)) and log triglyceride concentrations (-0.35 (-0.59 to -0.12)). Discussion Increased proportion of whole grain rye, reflected by a biomarker, in the diet is associated with favorable blood lipid outcomes, a relationship that should be further investigated. ClinicalTrials.gov identifier: NCT00992641 (http://clinicaltrials.gov/ct2/show/NCT00992641?term=NCT00992641&rank=1) 1 Introduction 2 Epidemiological studies have shown an inverse relationship between diets rich in whole grain 3 and the risk of many diet-related diseases, including cardiovascular diseases [1,2], metabolic 4 syndrome [3] and type 2 diabetes [4]. A number of intervention studies have shown that 5 intake of whole grains lowers serum total and LDL cholesterol concentrations [5-8], blood 6 pressure [9,10] and improves insulin sensitivity [11-13]. Animal studies have shown 7 cholesterol lowering effects of rye consumption [14-16] but few intervention studies have 8 studied the effects of rye on blood lipids in humans, despite the fact that whole grain rye 9 products contain substantial amounts of extractable arabinoxylans and β-glucans [17] that 10 may contribute to cholesterol reduction. One study has shown that rye bread consumption 11 decreased total and LDL cholesterol concentrations in men with moderately elevated serum 12 cholesterol [5] but another study showed an unfavorable shift in serum cholesterol 13 concentrations after rye bread intake in postmenopausal women [18]. 14 In nutritional science it is essential to have accurate dietary information to assess 15 compliance to dietary interventions and to investigate diet-disease associations. Collecting 16 valid data on whole grain intake can, however, be challenging due to the broad selection of 17 foods contributing to these intakes [19]. The use of alkylresorcinols (AR) as dietary 18 biomarkers for whole grain wheat and rye intake could improve the validity of dietary 19 assessment of whole grain in combination with, or as an alternative for, self-reports [19-22]. It 20 may also be used as a tool for secondary analysis, where non-compliant individuals can be 21 excluded on the basis of the biomarker measurements. AR are 3,5-dihydroxy-phenolic lipids 22 with an odd-numbered alkyl chain generally ranging from C17 to C25. Among commonly 23 consumed foods, AR are almost exclusively present in the outer parts of wheat and rye grains, 24 and just at trace amounts in refined white flour [23,24]. The ratio of the AR homologues 25 C17:0 and C21:0 differs between cereals but is consistent within cereal species 26 (approximately 0.1 in wheat and 1 in rye ) [24] and may be used to differentiate sources of 27 cereals consumed via plasma measurements [25]. We have previously demonstrated that AR 28 are useful biomarkers for whole grain wheat and rye intake as part of a healthy Nordic diet 29 [26]. Furthermore, significant between-group changes in non-HDL cholesterol concentrations 30 and LDL to HDL cholesterol ratio was found in the randomized dietary intervention with a 31 healthy Nordic diet (SYSDIET) [27]. 32 The present study is based on a secondary analysis of the SYSDIET intervention trial 33 and the aim was to investigate the association between plasma AR, a biomarker for whole 34 grain wheat and rye intake, and blood lipid concentrations in a population with metabolic 35 syndrome (MetS) participating in a randomized dietary intervention. Furthermore, the 36 associations between the AR C17:0/C21:0 ratio, a suggested marker of the relative whole 37 grain/bran rye intake, and blood lipids were analyzed. 38 39 Methods 40 The current study was a part of the SYSDIET study, a randomized controlled multicenter 41 parallel group study with balanced randomization (1:1) performed in six intervention centers 42 (Kuopio and Oulu, Finland; Lund and Uppsala, Sweden; Aarhus, Denmark and Reykjavik, 43 Iceland) examining the health effects of a healthy Nordic diet (ND) in a population with 44 MetS. The protocol for this trial and supporting CONSORT checklist are available as 45 supporting information; see Checklist S1 and Protocol S1. 46 All local ethical committees (Research Ethics Committee of the Hospital District of 47 Northern Savo and Northern Ostrobothnia Hospital District,Oulu, Finland, Regional Ethical 48 Review Board, Lund and The Regional Ethical Review Board in Uppsala, Sweden, Ethical 49 Committee of Aarhus County, Denmark and The Icelandic Bioethics Committee, Iceland) 50 approved the SYSDIET study protocol which followed the Helsinki declaration guidelines 51 and informed written consent was obtained from all participants. The trial was registered at 52 ClinicalTrials.gov (ClinicalTrials identifier: NCT00992641, 53 http://clinicaltrials.gov/ct2/show/NCT00992641?term=NCT00992641&rank=1) in October 54 2009, once all local ethical committees had given their approval. The authors confirm that all 55 ongoing and related trials for this intervention are registered. 56 The study design, inclusion and exclusion criteria, study diets, the main measurements 57 and analyses performed at each time point and main results have previously been reported in 58 detail [27]. The present study is a secondary analysis where the association between AR and 59 blood lipids at the end of the intervention was examined. 60 61 Study participants 62 Participants in the study were primarily recruited through advertisements in newspapers, but 63 also from previous clinical or epidemiological trials. In brief, the inclusion criteria were age 64 30-65 years, body mass index (BMI) 27 - 40 kg/m2, and two diagnostic criteria for metabolic 65 syndrome as outlined by the International Diabetes Federation [28], except type 2 diabetes. 66 67 Study design 68 Participants were randomly assigned to either a healthy Nordic diet or an average Nordic 69 (control) diet for 18 or 24 weeks. Due to financial and logistic reasons the study period was 70 18 weeks at 4 centers (Aarhus, Uppsala, Reykjavik and Oulu) and 24 weeks at 2 centers 71 (Lund and Kuopio) where the intervention was initiated earlier [27]. The major visits to the 72 study clinics were on week 0 (baseline), week 12 and week 18/24. In Kuopio and Lund, the 73 intervention was carried out from October 2009 to June 2010, in Aarhus, from January 2010 74 to September 2010, in Oulu, from December 2009 to October 2010, in Reykjavik, from 75 March 2010 to October 2010 and in Uppsala, from June 2010 to November 2010. 76 The randomization was done by staff in each study center as described elsewhere [26]. 77 The staff in all study centers was trained to perform measurements in a similar way and the 78 quality management protocol was available for all staff members. The allowed weight change 79 during the study was aimed to be less than ±4 kg due to the isocaloric design applied. 80 81 Study diets 82 The study diets were planned to be isocaloric. Nordic nutrition recommendations formed the 83 basis of the diet in the ND group [29] and the mean nutrient intake in the Nordic countries 84 formed the basis for the control diet. Participants in the ND group were advised to consume 85 ≥25% of total energy as whole grain, of which ≥50% was comprised of rye, barley and oat 86 which included ≥6 slices/day of bread with ≥6 g fiber/100 g (one slice equals 30-40 g of 87 bread). Participants in the control group were asked to consume ≥25% of energy as refined 88 grain, of which ≥90% made from wheat. Furthermore, the participants in the ND group were 89 encouraged to consume ≥500 g of fruits, vegetables and berries daily, use rapeseed and/or 90 sunflower and/or soybean oil based margarines, choose low fat milk and dairy products, 91 consume three fish meals a week, choose low fat meat, and to avoid sugar-sweetened 92 beverages while the control group was advised to keep their habitual diet. Key food products 93 were provided to the study participants in both groups. The individuals in the ND group 94 received whole grain products such as oatmeal and barley, crisp breads, whole grain wheat 95 and rye breads. The individuals in the control group received low fiber cereal products, e.g. 96 breads with fiber content less than 6 g/100 g and ordinary pasta. The study diets have 97 previously been described in detail [27]. 98 The participants were instructed by a clinical nutritionist or a dietitian at all study 99 visits. The study participants kept dietary records regarding the intake of cereal products. 100 These records were checked at each visit to the study center, including the visits when the 101 groceries were delivered to the participants, i.e. at 1-2 week intervals and the subjects also 102 returned 4-day weighed food records at baseline, week 2, 12 and 18/24 [27]. 103 104 Biochemical measurements 105 Blood samples were collected at week 0, 12 and 18/24. Automated clinical chemistry 106 analyzers and routine clinical chemistry methods were used to measure concentrations of total 107 and HDL cholesterol, triglycerides and apoplipoproteins A1 and B. The equipment and 108 system reagents and CV%´s have been described previously [27]. LDL cholesterol 109 concentrations were calculated using Friedewald's formula. 110 Determination of AR in plasma samples were made in collaboration with the HELGA- 111 Centre at Department of Food Science, BioCenter at Swedish University of Agricultural 112 Sciences. A gas chromatography mass spectrometry (GC-MS) method was used to quantify 113 AR homologues C17:0-C25:0 in 0.2 ml plasma samples as described previously [30]. Quality 114 control samples (n=4) was included in each batch to ensure appropriate within- and between 115 batch precision (CV <15%). Samples from each subject were analyzed within the same batch 116 and samples from different centers were allocated randomly between batches. 117 118 Data pooling and statistical analysis 119 All intervention centers entered their data into Microsoft Office Excel 2007 (Windows) data 120 form prior to pooling the data in a joint database maintained at VTT Technical Research 121 Centre of Finland. Data was exported from the database and imported to SPSS (Statistical 122 Package for the Social Sciences) for Windows, version 20.0 (SPSS Inc., Chicago, Illinois, 123 USA) for statistical analysis. Level of significance in all analyses was set at P<0.05. 124 Normality of variables was checked by inspection and by using the Kolmogorov– 125 Smirnov test. Values are described as means and standard deviations (SD) if normally 126 distributed or as medians and interquartile range (IQR) if not normally distributed. 127 The analyses were performed with data from the ND and control groups combined and 128 associations between total AR and the AR C17:0/C21:0 ratio at 18/24 weeks and various 129 blood lipid outcomes were analyzed. Where distributions were skewed (every variable except 130 total and LDL cholesterol concentrations), log-transformation was performed on dependent 131 variables for linear regression to better approximate normal distribution. The distribution of 132 AR C17:0/C21:0 ratio across total plasma AR was checked by inspection. 133 Linear regressions for fasting concentrations of total, LDL, HDL, non-HDL 134 cholesterol, apolipoproteins A1 and B, triglycerides and LDL/HDL cholesterol ratio with total 135 plasma AR or AR C17:0/C21:0 ratio (model 1), baseline value for the dependent variable, 136 age, BMI and statin use, as factors were performed. As rye often contains more AR than 137 wheat, total plasma AR was added as a variable, in addition to the variables in model 1 when 138 the associations between AR C17:0/C21:0 ratio was examined, to prevent confounding of 139 high AR content (model 2). To account for multiple testing, we used the Bonferroni 140 correction approach, using a significance level of alpha divided by the number of independent 141 outcomes, that is 0.05/8 = 0.006. Furthermore, we individually adjusted for saturated fat 142 intake calculated from food records and serum phospholipid fatty acids myristic acid, palmitic 143 acid, omega-3 fatty acids and C15:0, to prevent confounding of dietary fat intake. 144 145 Results 146 Altogether 166 participants finished the study and of those 158 subjects had blood samples for 147 AR and blood lipid analyses from all time points and that group was used for analyses in the 148 present study (Figure 1). There was no difference between dropouts and those who completed 149 the study in baseline BMI (P = 0.797), waist circumference (P=0.228), age (P = 0.959) or 150 triglyceride (P = 0.055), total cholesterol (P=0.743), HDL cholesterol (P = 0.216) or LDL 151 cholesterol (P =0.686) concentrations. 152 There was no change in total energy intake during the study and energy intake was not 153 different between the groups, hence, body weight remained constant during the intervention in 154 both groups. The mean age of the participants was 54.5 (8.2) years, body weight 91.2 (13.5) 155 kg, BMI 31.7 (3.2) kg/m2 and waist circumference 104.3 (9.7) cm as described previously 156 [31]. Baseline lipid concentrations and the number of statin users in by intervention group, are 157 shown in Table 1. The prevalence of MetS was 90%. No differences were observed between 158 the groups at baseline in clinical characteristics. 159 There were small, but significant, decrease in concentrations of total cholesterol (5.1 160 (1.2) mmol/L at 18/24 weeks, P=0.018), LDL (3.2 (1.1) mmol/L at 18/24 weeks, P=0.006) 161 and non-HDL cholesterol (3.8 (1.3) mmol/L at 18/24 weeks, P=0.006) and LDL/HDL 162 cholesterol ratio (2.3 (1.3) at 18/24 weeks, P=0.003) within the ND group from week 0 to 163 18/24 but not within the control group. Results on the effects of the SYSDIET intervention, 164 with ND, on blood lipid concentrations and other health outcomes have been reported 165 previously [27]. Both total plasma AR concentration and the ratio of C17:0/C21:0 were 166 significantly higher in the ND group than in the control group at 18/24 week (Table 2) and the 167 range of total plasma AR was 14 to 499 nmol/L. There was no skew towards higher AR 168 C17:0/C21:0 ratios for higher total plasma AR concentrations. 169 All participants (both ND and control groups) were combined for secondary analyses 170 on the associations between AR and blood lipids concentrations at 18/24 weeks. Total plasma 171 AR concentration at 18/24 weeks was not significantly correlated with concentrations of 172 fasting serum total cholesterol (P=0.95), LDL cholesterol (P=0.23), HDL cholesterol 173 (P=0.25), LDL-C/HDL-C ratio (P=0.13), triglycerides (P=0.14), apolipoprotein A1 (P=0.09) 174 or apolipoprotein B (P=0.86) after adjusting for confounders. However, the AR C17:0/C21:0 175 ratio, which reflects the proportion of whole grain rye of whole grain rye and whole grain 176 wheat combined, was inversely associated with LDL cholesterol concentration when adjusted 177 for baseline LDL cholesterol concentration, BMI, age and statin use (model 1, P=0.040) and 178 also when total plasma AR was added to the regression model (model 2, P=0.046) (Table 3). 179 Similarly, AR C17:0/C21:0 ratio was inversely associated with LDL/HDL cholesterol 180 ratio (P=0.020 and P=0.024 for model 1 and model 2, respectively), non-HDL cholesterol 181 (P<0.001 for both model 1 and model 2), triglyceride (P=0.004 and P=0.002 for model 1 and 182 model 2, respectively) and apolipoprotein B concentrations (P=0.047 for both model 1 and 183 model 2) and positively associated with HDL cholesterol concentration (P=0.044 and 184 P=0.049 for model 1 and model 2, respectively). Including saturated fatty acids calculated 185 from food records to the models did not change the results nor did including the serum 186 phospholipid fatty acids myristic acid, palmitic acid, omega-3 fatty acids and C15:0, except 187 for HDL cholesterol in which the association with C17:0/C21:0 became borderline significant 188 (P=0.067). However, using the Bonferroni correction approach to account for multiple testing 189 the associations for LDL cholesterol, HDL cholesterol, LDL/HDL cholesterol ratio and 190 apolipoprotein B concentrations would not be considered significant while associations for 191 non-HDL cholesterol and triglyceride concentration are robust for this correction (P<0.006). 192 193 Discussion 194 Favorable changes in serum LDL cholesterol concentration and LDL to HDL cholesterol ratio 195 were observed in participants consuming a healthy Nordic diet, established with a healthy 196 dietary pattern approach, and thus also rich in whole grain cereals, for either 18 or 24 weeks 197 [27]. In the intervention the participants in the ND group were advised to consume ≥25% of 198 total energy as whole grain, of which ≥50% was comprised of rye, barley and oat, and to 199 make favorable changes in dietary fat intake, and therefore a change in serum cholesterol 200 concentration in the ND group was expected. Some studies have found that whole grain diets 201 lower serum total and LDL cholesterol concentrations [5-8]. The cholesterol lowering effects 202 of β-glucans in oats and barley have been extensively studied [32,33] and the European Food 203 Safety Authority (EFSA) has approved two health claims on the subject: β-glucans contribute 204 to the maintenance of normal blood cholesterol levels [34] and oat β-glucan has been shown 205 to lower/reduce blood cholesterol. Blood cholesterol lowering by increased oats intake may 206 reduce the risk of (coronary) heart disease [35]. 207 In the present study, both ND and control groups were pooled for secondary analysis 208 on the association between plasma AR and blood lipids at 18/24 weeks. Plasma AR has been 209 established as a biomarker for whole grain wheat and rye intake [19,21,26] and has previously 210 been studied regarding associations with BMI and metabolic risk factors [36,37]. We 211 observed no association between total plasma AR concentration and serum lipid 212 concentrations, possibly due to a considerable amount of AR intake originating from whole 213 grain wheat. Whole grain wheat intake does typically not affect blood cholesterol and is thus 214 often used as a control in experimental studies [38]. Interestingly, the AR C17:0/C21:0 ratio, 215 an indicator of relative rye intake, was associated with favorable changes in concentrations of 216 LDL-, HDL-, and non HDL cholesterol, LDL/HDL cholesterol ratio and triglycerides, even 217 after adjustment for fatty acids composition of serum phospholipids, with the exception of 218 HDL cholesterol where the association became borderline significant. The ratio of the AR 219 homologues C17:0 and C21:0 differs between cereal species and can be used to differentiate 220 between the whole grain sources in food products [24]. The ratio of AR C17:0/C21:0 is 221 approximately 0.1 in wheat and 1 in rye and the ratio is to some extent reflected in fasting 222 plasma samples, although it is generally lower, typically around 0.6-0.8 in fasting samples 223 after a whole grain rye intervention for 1-8 weeks, and can be used to differentiate which AR 224 containing cereals have been consumed [25,39,40]. 225 Health effects of rye are far less studied than oats and barley and studies on the effects 226 of rye consumption on cholesterol are scarce. The mechanisms that have been proposed to 227 cause the reduction in plasma cholesterol levels focus on the fiber and phytochemical content 228 of whole grains, particularly water-soluble viscose fiber [41,42], and it has been established 229 that the cholesterol lowering effect of β-glucans is dependent on the molecular weight and 230 subsequent viscosity [43]. Rye is particularly high in dietary fiber, ranging between 18-22%, 231 and it contains substantial amounts of extractable β-glucans and arabinoxylans, which could 232 contribute to cholesterol reduction through a similar mechanism as β-glucans from oats and 233 barley [17,44]. Animal studies have shown that rye could affect blood lipid concentrations by 234 increasing the total concentration of bile acids in the bile [14-15]. 235 Previous studies have shown that rye bread consumption decreases total and LDL 236 cholesterol concentrations in men with moderately elevated serum cholesterol [5], and lower 237 concentrations of plasma free-cholesterol, total cholesterol, triglycerides and phospholipids 238 have been demonstrated after a high fiber rye diet than a low fiber wheat diet [6]. 239 Additionally, animal studies have shown cholesterol lowering effects of rye consumption [14- 240 16]. However, Moazzami et al showed an unfavorable shift in serum cholesterol levels after 241 rye bread intake in postmenopausal women [18]. Processing of rye and other whole grains, 242 such as fermentation, have been reported to influence the molecular weight of cereal dietary 243 fiber, hence it is possible that lower viscosity of the fibers could explain the lack of beneficial 244 effects reported in some studies [17]. Nevertheless, the results of the present study suggest 245 that whole grain rye consumption is associated with favorable outcomes in blood lipids and 246 this association should be further studied. 247 One of the strengths of the present study is that we used AR as a dietary biomarker for 248 whole grain rye and wheat intake instead of self-reported estimates. As collecting valid data 249 on whole grain intake can be challenging, the use of biomarkers to assess intake and/or 250 compliance to dietary interventions is a way to avoid some of the obstacles associated with 251 self-reporting [19]. Previously, we have demonstrated that AR is a good biomarker for the ND 252 under intervention conditions with moderate to good reliability [26] and our data supports the 253 use of the AR C17:0/C21:0 ratio as an indicator for whole grain rye intake in intervention 254 studies where a diet rich in whole grain rye is being investigated [31]. Although the AR 255 C17:0/C21:0 ratio reflects only the relative whole grain/bran rye consumption of total whole 256 grain wheat and rye intake, and not the total rye intake, the advised whole grain rye intake in 257 the present study was high in relation to other sources of AR and therefore the ratio should 258 reflect the total rye intake reasonably well. However, the present study does not offer any 259 mechanistic explanations behind the possible cholesterol-lowering by whole grain rye, and 260 more needs to be done to explain the association. Nevertheless, the results do motivate further 261 research on this topic. Some confounding effects from oats intake are possible, as participants 262 consuming more rye might also consume more oats, and confounding effects from other food 263 sources cannot be excluded. However, this is probably not explaining the main results. 264 Previous studies suggest that plasma AR may be transported in VLDL and HDL [45], which 265 could affect the results of the present study, as well as numerous dietary and non-dietary 266 factors potentially influencing biomarker concentrations [19]. Nevertheless, since the AR 267 C17:0/C21:0 ratio, but not total plasma AR, was associated with cholesterol concentrations, 268 and the associations did not change when adjusted for total plasma AR, we believe our 269 findings indicate that the associations we found are related to whole grain rye intake and not 270 by the fact that AR may be transported in certain lipoproteins. 271 272 Conclusion 273 Our findings suggest that a greater proportion of whole grain rye in a Nordic diet is associated 274 with favorable outcomes in blood lipid concentrations, an association that warrants further 275 investigation. 276 References 1. McKeown NM, Meigs JB, Liu S, Wilson PW, Jacques PF (2002) Whole-grain intake is favorably associated with metabolic risk factors for type 2 diabetes and cardiovascular disease in the Framingham Offspring Study. Am J Clin Nutr 76: 390-398. 2. Mellen PB, Walsh TF, Herrington DM (2008) Whole grain intake and cardiovascular disease: a meta-analysis. Nutr Metab Cardiovasc Dis 18: 283-290. 3. Sahyoun NR, Jacques PF, Zhang XL, Juan W, McKeown NM (2006) Whole-grain intake is inversely associated with the metabolic syndrome and mortality in older adults. Am J Clin Nutr 83: 124-131. 4. de Munter JS, Hu FB, Spiegelman D, Franz M, van Dam RM (2007) Whole grain, bran, and germ intake and risk of type 2 diabetes: a prospective cohort study and systematic review. PLoS Med 4: e261. 5. Leinonen KS, Poutanen KS, Mykkanen HM (2000) Rye bread decreases serum total and LDL cholesterol in men with moderately elevated serum cholesterol. J Nutr 130: 164170. 6. Lundin EA, Zhang JX, Lairon D, Tidehag P, Aman P, et al. (2004) Effects of meal frequency and high-fibre rye-bread diet on glucose and lipid metabolism and ileal excretion of energy and sterols in ileostomy subjects. Eur J Clin Nutr 58: 1410-1419. 7. Giacco R, Clemente G, Cipriano D, Luongo D, Viscovo D, et al. (2010) Effects of the regular consumption of wholemeal wheat foods on cardiovascular risk factors in healthy people. Nutr Metab Cardiovasc Dis 20: 186-194. 8. Ross AB, Bruce SJ, Blondel-Lubrano A, Oguey-Araymon S, Beaumont M, et al. (2011) A whole-grain cereal-rich diet increases plasma betaine, and tends to decrease total and LDL-cholesterol compared with a refined-grain diet in healthy subjects. Br J Nutr 105: 1492-1502. 9. Behall KM, Scholfield DJ, Hallfrisch J (2006) Whole-grain diets reduce blood pressure in mildly hypercholesterolemic men and women. J Am Diet Assoc 106: 1445-1449. 10. Tighe P, Duthie G, Vaughan N, Brittenden J, Simpson WG, et al. (2010) Effect of increased consumption of whole-grain foods on blood pressure and other cardiovascular risk markers in healthy middle-aged persons: a randomized controlled trial. Am J Clin Nutr 92: 733-740. 11. Leinonen K, Liukkonen K, Poutanen K, Uusitupa M, Mykkanen H (1999) Rye bread decreases postprandial insulin response but does not alter glucose response in healthy Finnish subjects. Eur J Clin Nutr 53: 262-267. 12. Juntunen KS, Laaksonen DE, Poutanen KS, Niskanen LK, Mykkanen HM (2003) Highfiber rye bread and insulin secretion and sensitivity in healthy postmenopausal women. Am J Clin Nutr 77: 385-391. 13. Pereira MA, Jacobs DR, Jr., Pins JJ, Raatz SK, Gross MD, et al. (2002) Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr 75: 848-855. 14. Lund EK, Salf KL, Johnson IT (1993) Baked rye products modify cholesterol metabolism and crypt cell proliferation rates in rats. J Nutr 123: 1834-1843. 15. Zhang JX, Lundin E, Reuterving CO, Hallmans G, Stenling R, et al. (1994) Effects of rye bran, oat bran and soya-bean fibre on bile composition, gallstone formation, gallbladder morphology and serum cholesterol in Syrian golden hamsters (Mesocricetus auratus). Br J Nutr 71: 861-870. 16. Andersson U, Rosen L, Ostman E, Strom K, Wierup N, et al. (2010) Metabolic effects of whole grain wheat and whole grain rye in the C57BL/6J mouse. Nutrition 26: 230239. 17. Rakha A, Åman P, Andersson R (2010) Characterisation of dietary fibre components in rye products. Food Chemistry 119: 859-867. 18. Moazzami AA, Bondia-Pons I, Hanhineva K, Juntunen K, Antl N, et al. (2012) Metabolomics reveals the metabolic shifts following an intervention with rye bread in postmenopausal women--a randomized control trial. Nutr J 11: 88. 19. Ross AB, Bourgeois A, Macharia HN, Kochhar S, Jebb SA, et al. (2012) Plasma alkylresorcinols as a biomarker of whole-grain food consumption in a large population: results from the WHOLEheart Intervention Study. Am J Clin Nutr 95: 204-211. 20. Ross AB, Kamal-Eldin A, Aman P (2004) Dietary alkylresorcinols: absorption, bioactivities, and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr Rev 62: 81-95. 21. Landberg R, Kamal-Eldin A, Andersson A, Vessby B, Aman P (2008) Alkylresorcinols as biomarkers of whole-grain wheat and rye intake: plasma concentration and intake estimated from dietary records. Am J Clin Nutr 87: 832-838. 22. Montonen J, Landberg R, Kamal-Eldin A, Aman P, Knueppel S, et al. (2010) Reliability of fasting plasma alkylresorcinol concentrations measured 4 months apart. Eur J Clin Nutr 64: 698-703. 23. Ross AB, Shepherd MJ, Schupphaus M, Sinclair V, Alfaro B, et al. (2003) Alkylresorcinols in cereals and cereal products. J Agric Food Chem 51: 4111-4118. 24. Chen Y, Ross AB, Aman P, Kamal-Eldin A (2004) Alkylresorcinols as markers of whole grain wheat and rye in cereal products. J Agric Food Chem 52: 8242-8246. 25. Linko-Parvinen AM, Landberg R, Tikkanen MJ, Adlercreutz H, Penalvo JL (2007) Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. J Nutr 137: 1137-1142. 26. Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, et al. (2013) Plasma Alkylresorcinols Reflect Important Whole-Grain Components of a Healthy Nordic Diet. J Nutr. 27. Uusitupa M, Hermansen K, Savolainen MJ, Schwab U, Kolehmainen M, et al. (2013) Effects of an isocaloric healthy Nordic diet on insulin sensitivity, lipid profile and inflammation markers in metabolic syndrome -- a randomized study (SYSDIET). J Intern Med 274: 52-66. 28. Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, et al. (2009) Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 120: 1640-1645. 29. (2004) Nordic Nutrition Recommendations 2004 Intergrating nutrition and physical activity. Copenhagen: Nordic Council of Ministers. 30. Landberg R, Åman P, Kamal-Eldin A (2009) A rapid gas chromatography-mass spectrometry method for quantification of alkylresorcinols in human plasma. Anal Biochem 385: 7-12. 31. Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, et al. (2014) Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole-grain rye intake, is associated to insulin sensitivity: a randomized study. Eur J Clin Nutr 68: 453-458. 32. Tiwari U, Cummins E (2011) Meta-analysis of the effect of beta-glucan intake on blood cholesterol and glucose levels. Nutrition 27: 1008-1016. 33. Othman RA, Moghadasian MH, Jones PJ (2011) Cholesterol-lowering effects of oat betaglucan. Nutr Rev 69: 299-309. 34. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on the substantiation of health claims related to beta glucans and maintenance of normal blood cholesterol concentrations (ID 754, 755, 757, 801, 1465, 2934) and maintenance or achievement of a normal body weight (ID 820, 823) pursuant to Article 13(1) of Regulation (EC) No 1924/2006. EFSA Journal 2009; 7(9):1254 [18 pp.]. 10.2903/j.efsa.2009.1254 Available online: www.efsa.europa.eu/efsajournal. Accessed February 20 2014 35. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on the substantiation of a health claim related to oat beta glucan and lowering blood cholesterol and reduced risk of (coronary) heart disease pursuant to Article 14 of Regulation (EC) No 1924/2006. EFSA Journal 2010;8(12):1885 [15 pp.]. 10.2903/j.efsa.2010.1885 Available online:www.efsa.europa.eu/efsajournal. Accessed February 20 2014 36. Ma J, Ross AB, Shea MK, Bruce SJ, Jacques PF, et al. (2012) Plasma Alkylresorcinols, Biomarkers of Whole-Grain Intake, Are Related to Lower BMI in Older Adults. J Nutr 142: 1859-1864. 37. Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, et al. (2014) Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole-grain rye intake, is associated to insulin sensitivity: a randomized study. Eur J Clin Nutr. 38. Jenkins DJ, Kendall CW, Vuksan V, Augustin LS, Mehling C, et al. (1999) Effect of wheat bran on serum lipids: influence of particle size and wheat protein. J Am Coll Nutr 18: 159-165. 39. Landberg R, Aman P, Friberg LE, Vessby B, Adlercreutz H, et al. (2009) Dose response of whole-grain biomarkers: alkylresorcinols in human plasma and their metabolites in urine in relation to intake. Am J Clin Nutr 89: 290-296. 40. Ross AB (2012) Present status and perspectives on the use of alkylresorcinols as biomarkers of wholegrain wheat and rye intake. J Nutr Metab 2012: 462967. 41. Slavin JL, Jacobs D, Marquart L, Wiemer K (2001) The role of whole grains in disease prevention. J Am Diet Assoc 101: 780-785. 42. Fardet A (2010) New hypotheses for the health-protective mechanisms of whole-grain cereals: what is beyond fibre? Nutr Res Rev 23: 65-134. 43. Wolever TM, Tosh SM, Gibbs AL, Brand-Miller J, Duncan AM, et al. (2010) Physicochemical properties of oat beta-glucan influence its ability to reduce serum LDL cholesterol in humans: a randomized clinical trial. Am J Clin Nutr 92: 723-732. 44. Hansen HB, Rasmussen CV, Bach Knudsen KE, Hansen Å (2003) Effects of genotype and harvest year on content and composition of dietary fibre in rye (Secale cereale L) grain. Journal of the Science of Food and Agriculture 83: 76-85. 45. Linko-Parvinen AM, Landberg R, Tikkarren MJ, Adlercreutz H, Penalvo JL (2007) Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. Journal of Nutrition 137: 1137-1142. Funding The study was supported by NordForsk NCoE in Food Nutrition and Health: Project 070014 SYSDIET (Systems biology in controlled dietary interventions and cohort studies (http://www.nordforsk.org/en?set_language=en). This study was also funded by the Academy of Finland, Finnish Diabetes Research Foundation, Finnish Foundation for Cardiovascular Research, The Sigrid Juselius Foundation, and EVO funding from Kuopio University Hospital (Finland: US, KHH, JH, MJS, MK, KP, MU); the Druvan Foundation, ESPEN, Skåne County Council Research and Development Foundation, Swedish Research Council for Health, Working Life and Welfare, EXODIAB (Excellence of Diabetes Research in Sweden), The Heart-Lung Foundation, Diabetesfonden and Foundation Cerealia (Sweden: RL, LC, BÅ, FR, UR); The Danish Obesity Research Centre (DanORC, www.danorc.dk), The Danish Council for Strategic Research (DairyHealth, BioFunCarb) (Denmark: LB, KH), The Agricultural Productivity Fund (Iceland: IG, OKM, IT) and the Icelandic Research Fund for graduate students (1206880061)http://rannis.is/ (OKM). The following companies provided food products for the study participants: Kesko Food Ltd, Raisio Group, Bunge Finland Ltd (Finland); Belico Food AB, Fazer Bageri Sverige, Lantmännen, Oatly AB, Olle Svensson AB, Procordia Food AB, Pågen AB, Unilever, Wasabröd AB (Sweden); Lantmännen Food R&D, Jan Import A/S, Ardo A/S, Scandic Food A/S, WASA, Glyngøre Limfjord A/S, Royal Greenland A/S, Arla Foods (Denmark); The Mother Earth Farm at Vallanes (Iceland); and Unilever Nordic (Sweden, Denmark, Iceland). Acknowledgement Authors express sincere gratitude to the participants taking part in the study. We also thank all staff involved in the intervention. Figure 1. Flow chart of the study Table 1. Baseline blood lipid concentrations and statin use of the participants. Values are presented as medians (IQR). ND group Control group (n=93) (n=65) Total cholesterol, mmol/L 5.3 (1.2) 5.1 (1.5) HDL cholesterol, mmol/L 1.4 (0.5) 1.3 (0.6) LDL cholesterol, mmol/L 3.2 (1.1) 3.1 (1.5) Non-HDL cholesterol, mmol/L 3.9 (1.3) 3.7 (1.6) LDL-C/HDL-C 2.5 (1.4) 2.4 (1.8) Triglycerides 0 min, mmol/L 1.3 (0.8) 1.4 (0.8) Apolipoprotein A1, g/L 1.4 (0.3) 1.4 (0.3) Apolipoprotein B, g/L 1.0 (0.4) 1.0 (0.4) 19 18 Statin use, n Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein Table 2. Dietary fiber intake, total plasma alkylresorcinols (AR) concentration and AR C17:0/C21:0 ratio for the healthy Nordic diet group (ND) and control group at baseline and after the 18/24 week intervention. Values are medians (IQR). ND group (n = 93) Dietary fiber intake1,2, g/day Control group (n = 65) Week 0 Week 18/24 Week 0 Week 18/24 21.1 (8.8) 36.0 (15.1)# 20.0 (7.1) 16.0 (6.2)*# 73 (88) 106 (108)# 71 (70) 61 (40)*# 0.3 ± 0.2 0.3 ± 0.2 0.3 ± 0.2 0.2 ± 0.2*# Total plasma AR1, nmol/L C17:0/C21:0 ratio1 *significantly different from ND group at that time point # significantly different from baseline 1 Data published previously [31] 2 n = 140, ND group n = 80 and control group n = 40 Abbreviations: ND, healthy Nordic diet; AR, alkylresorcinols Table 3. Regression models for concentrations of serum total, HDL, LDL, non HDL cholesterol and triglycerides and the cholesterol LDL-C/HDL-C ratio Dependent variable at Independent variable at 18/24 weeks 18/24 weeks B1 (95% CI) Total cholesterol Model 1 AR C17:0/C21:0 -0.33 (-0.78 to 0.12) Model 2 AR C17:0/C21:0 -0.33 (-0.78 to 0.12) Model 1 AR C17:0/C21:0 -0.41 (-0.80 to -0.02) Model 2 AR C17:0/C21:0 -0.40 (-0.79 to -0.01) Model 1 AR C17:0/C21:0 0.11 (0.00 to 0.21) Model 2 AR C17:0/C21:0 0.11 (0.00 to 0.21) LDL cholesterol Log HDL cholesterol Log LDL-C/HDL-C cholesterol Model 1 AR C17:0/C21:0 -0.20 (-0.37 to -0.03) Model 2 AR C17:0/C21:0 -0.20 (-0.37 to -0.03) Model 1 AR C17:0/C21:0 -0.84 (-1.22 to -0.45) Model 2 AR C17:0/C21:0 -0.81 (-1.19 to -0.44) Log non HDL cholesterol Log triglycerides Model 1 AR C17:0/C21:0 -0.35 (-0.59 to -0.12) Model 2 AR C17:0/C21:0 -0.37 (-0.60 to -0.13) Model 1 AR C17:0/C21:0 0.05 (-0.03 to 0.13) Model 2 AR C17:0/C21:0 0.05 (-0.03 to 0.13) Model 1 AR C17:0/C21:0 -0.12 (-0.24 to 0.00) Model 2 AR C17:0/C21:0 -0.12 (-0.24 to 0.00) Log apolipoprotein A1 Log apoliprotein B 1 B reflects the change in the log-transformed outcome variable for each 1-unit changes in the independent variable. Model 1: adjusted for baseline value of the dependent variable, BMI, age and statin use Model 2: adjusted for baseline value of the dependent variable, total AR, BMI, age and statin use Abbreviations: ND, healthy Nordic diet; AR, alkylresorcinols; HDL, high-density lipoprotein; LDL, low-density lipoprotein; BMI, body mass index Paper IV Title: Increasing rye consumption among men with type 2 diabetes: a feasibility study Keywords: alkylresorcinols, rye consumption, feasibility, acceptability, whole grain, compliance Authors: Ola Kally Magnusdottir1,Rikard Landberg2,3, Ingibjorg Gunnarsdottir1, Kristjan Thor Magnusson4, Göran Hallmans5, Rikke Dalgaard Hansen6, Cecilie Kyrø6, Anja Olsen6, Anne Tjønneland6, Inga Thorsdottir1, Laufey Steingrimsdottir1 1 Unit for Nutrition Research, Landspitali -The National University Hospital of Iceland, Faculty of Food Science and Nutrition and School of Health Sciences, University of Iceland, Reykjavík, Iceland, 2 Department of Food Science, BioCenter, Swedish University of Agricultural Sciences, Uppsala, Sweden 3 Division of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden 4 Center for Sport and Health Sciences, School of Education, University of Iceland, Reykjavík, Iceland 5 Department of Public Health and Clinical Medicine, Nutritional Research, Umeå University, Umeå, Sweden 6 Danish Cancer Society Research Center, Strandboulevarden 49, DK-2100 Copenhagen, Denmark Corresponding author: Ola Kally Magnusdottir E-mail: [email protected] Address: Eiriksgata 29, IS-101 Reykjavik, Iceland Telephone: +354 543 8422 Fax: +354 543 4824 Conflict of interests The authors declare no conflict of interest. Funding The study was funded by Nordic Center of Excellence, HELGA: Nordic Health – Whole grain Food, the Consortium Agreement for Consortium on Health and Ageing Network of Cohorts in Europe and the United Std States (CHANCES) - EG Grant agreement No: 242244 – WP4 and The Icelandic Research Fund for graduate students (1206880061). Abstract Background and aim: Intervention studies have indicated beneficial effects of whole grain rye consumption but studies on acceptability and compliance to such interventions in populations not accustomed to high intake is needed. The aim was to determine the feasibility of conducting an intervention with high consumption of whole grain rye and vigorous physical activity (PA) for 6 months in a population of men with type 2 diabetes (DM2) and habitually low whole grain intake. Methods: Participants (males, 42-80 years with DM2) were randomized into either a rye group (n=14) or a control group (n=9) for 26 weeks. Participants in the rye group were provided with, and asked to consume, 100 g/day of whole grain rye. Acceptability and compliance was assessed with questionnaires, 4-day food record and plasma alkylresorcinols (AR). Participants in both groups were instructed to exercise vigorously three days/week and walk ≥10000 steps on other days. Results: In total, 87% of the participants finished the study. Rye intake and total plasma AR increased significantly in the rye group (P=0.003 and P=0.004, respectively) and the mean rye consumption during the intervention was 93±44 g/day. In the control group, the mean rye intake was 10±8 g/day during the intervention. Four participants in the rye group experienced digestive discomfort in weeks 0-13 weeks and two in weeks 13-26. PA levels did not change significantly during the intervention for the rye group or the control group and no participant reached 10 000 steps/day. Discussion: The dietary intervention was well received by the intervention group as seen in low dropout, good compliance and acceptability. Poor compliance to prescribed PA in the intervention emphasizes the need for improved methods to encourage PA in diabetic populations. 1 Introduction 2 The most important grains consumed in the Western world are wheat, corn, rice, oats, barley and 3 rye. Rye is a cereal of specific interest as it is particularly high in dietary fiber [1], and 4 traditionally consumed as whole grain. Rye is mostly consumed in the Northern and Eastern 5 European countries [2]. In Iceland, rye and oatmeal were the main types of whole grain in the 6 diet during the first half of the 20th century and in 1930 the average rye consumption was 152 7 g/day according to import figures, but cereals have not been grown in Iceland for human 8 consumption to any extent [3]. Rye consumption has, however, decreased dramatically during 9 the last 50 years in Iceland, as well as in many other countries [2,4]. According to food supply 10 data the average rye consumption in Iceland was 31 g/day in 1961 but 8 g/day in 2011. The 11 average rye consumption in 2011 was higher in the other Nordic countries, 47 g/day in Denmark, 12 45 g/day in Finland, 32 g/day in Sweden and 18 g/day in Norway [2]. Furthermore, Iceland also 13 has the lowest consumption of total dietary fiber among all the Nordic countries [5-8]. 14 Several intervention studies have indicated beneficial effects of rye consumption on 15 metabolic processes [9-13]. Studies have shown positive effects of high rye consumption on 16 fasting plasma insulin and, to a less extent, on fasting plasma glucose [9,10] and an improved 17 early insulin response has been shown after rye bread interventions [11,13]. These studies have, 18 however, been conducted in countries accustomed to a relatively high rye intake (i.e. Finland and 19 Sweden) and the effects of rye consumption were investigated over a short period of time, 20 generally 6-12 weeks. To date, no intervention study on the health effects of whole grain rye 21 intake has been conducted on patients with type 2 diabetes. Before implementing an intervention 22 with high consumption of whole grain rye in a population unaccustomed to high intakes, it is 23 important to explore to what extent it is possible to increase the consumption and the 24 acceptability of whole grain rye products. 25 Physical activity (PA) also plays a crucial role in the management of type 2 diabetes with 26 extensive research reporting improvement in glucose control with increased PA [14-17]. 27 Nevertheless, the majority of people with type 2 diabetes are inactive [15,18,19] and it is 28 therefore of great importance to encourage increased PA in this population. 29 The aim of the present study was to determine the feasibility of conducting an 30 intervention with high consumption of whole grain rye and vigorous PA for 6 months in a 31 population of men with habitually low whole grain and fiber intake. The present feasibility study 32 investigated the dropout, acceptability and adherence to the intervention in a population 33 comprised of 23 men diagnosed with type 2 diabetes. 34 35 Methods 36 The present study is a feasibility study conducted as a follow up of Nordforsk Nordic Center of 37 Excellence Program, HELGA: Nordic Health – Whole grain Food. Informed consent was 38 obtained from all participants and the study received approval from the Icelandic National 39 Bioethics Committee and the Data Protection Commission. 40 41 Participants and study design 42 Participants were referred to the study by general practitioners in three health care clinics in the 43 capital area of Iceland. Participants were 23 males, aged 42 – 80 years and diagnosed with type 2 44 diabetes or pre-diabetes by their family doctor. The exclusion criteria included any chronic 45 disease and/or condition, which could hamper the adherence to the dietary intervention protocol. 46 The duration of the intervention was 26 weeks and participants followed their habitual 47 diet for a week before randomization into either the rye group (n = 14) or the control group (n = 48 9). All participants had three major visits, at week 0, weeks 13 and 26 (Figure 1). Additionally 49 there was a group meeting for all participants with a clinical nutritionist in week 2 where 50 participants received general recommendations on a healthy diet, as well as a meeting with a 51 physical therapist and a personal trainer during the first weeks. The participants in the rye group 52 came to the clinic every 2-3 weeks, to pick up the rye products that were provided and both 53 groups were offered to pick up vegetable packages free of charge every 4 weeks. The 54 intervention was carried out from April to November 2013. 55 56 Study diet 57 Participants in the rye group were instructed to consume 100 g of whole grain rye each day for 58 26 weeks and rye products equivalent to this intake were provided free of charge. The products 59 provided to the rye group were rye flakes to use for rye porridge, rye crisp bread and rye bread. 60 The rye content of the products provided is listed in Table 1. The participants in the control 61 group were asked to maintain their usual diet, but avoid rye products, and were offered whole 62 wheat breads free of charge 63 Participants in the rye group received both oral and written information on the amount of 64 rye in the products provided, as well as other common whole grain rye products, and how to 65 reach 100 g rye a day. To simplify, the participants were asked to obtain ten points every day, 66 where one point was earned for every 10 g of whole grain rye consumed. A color-printed booklet 67 with names and pictures of the rye products provided in the study, as well as other common 68 whole grain rye products available in local supermarkets, was also given to the participants in the 69 rye group. The booklet included information on the total nutritional value, the amount of whole 70 grain rye per 100 g of product and per serving (e.g. slice of bread) and number of points earned 71 per serving. Altogether the booklet contained information on 32 rye products. 72 Dietary intake was assessed using a consecutive 4-day (Wednesday to Saturday) weighed 73 food record collected prior to week 0, at week 13 and 26. Participants received both written and 74 oral instructions on how to fill in the diet records and were provided with electronic household 75 scales. At the week 0 visit, a nutritionist reviewed the first food diary with each participant and 76 suggested exchanges for rye products. Rye porridge (from rye flakes) was recommended daily, 77 to facilitate the rye consumption without adding increased energy to the diet. In addition, daily 78 consumption of 1 - 4 slices of rye bread (dependent on the type of bread) was recommended. 79 The nutrient intake from food diaries was calculated using The Icelandic Food 80 Composition Database [ISGEM], Iceland). The database uses EuroFIR definitions for chemicals 81 and total dietary fiber determined by the AOAC methods [20]. Whole grain rye intake was 82 calculated manually. 83 84 Physical activity 85 After randomization, all participants in both groups were asked to exercise vigorously for at least 86 45 minutes three days/week, and walk at least 10 000 steps on other days, for 26 weeks. A gym 87 membership and a pedometer were provided to all participants free of charge. At the beginning 88 of the study, each participant met with a physical therapist who evaluated the physical condition 89 of the participants. Following that meeting the participants met with a personal trainer who gave 90 them individualized exercise programs and showed them the exercises and the workout 91 equipment. The participants were free to contact the personal trainer during the intervention 92 period to get a different (less/more advanced) program if necessary. Thus, every participant 93 received an exercise program designed for their needs, e.g. interests, age, strengths and 94 weaknesses. 95 Accelerometers were used (Actigraph™ GT1M monitors) to assess both volume and 96 intensity of PA during waking hours for four consecutive days (Wednesday to Saturday) at three 97 time-points during the intervention, before week 0, week 13 and week 26. 98 99 Compliance and acceptability 100 Alkylresorcinols (AR) in plasma were used to assess compliance to the whole grain rye diet with 101 measures of total plasma AR and the AR C17:0/C21:0 ratio [10,21-23], in addition to the 102 weighed food diary. The plasma samples were analyzed at the Department of Food Science, 103 BioCenter at the Swedish University of Agricultural Sciences. AR homologues C17:0-C25:0 104 were measured in 0.2 ml plasma samples according to a previously published GC-MS method 105 [24]. Quality control samples (n = 4) were included in each batch to ensure appropriate within- 106 and between batch precision (CV <10-15%). Samples from the same individual were analyzed 107 within the same batch. 108 To measure the acceptability of the products participants in the rye group were, at week 109 13 and 26, given three statements and asked to check whether they strongly agreed, agreed, 110 neither agreed nor disagreed, disagreed or strongly disagreed. The statements were: 1) I have 111 experienced digestive discomfort after I started eating rye products 2) I like food products that 112 contain rye and 3) I would prefer products that contain rye more than other products. 113 114 115 Anthropometric measurements 116 Height was measured at week 0 with a calibrated stadiometer with a precision of 0.5 cm. Body 117 weight was measured at weeks 0, 13 and 26. Weight was measured in light clothing on a digital 118 calibrated scale to the nearest 0.1 kg. Body mass index (BMI) was calculated from the recorded 119 height and weight using the standard formula, dividing the participants´ weight (kg) by the 120 square height (m2). The waist circumference was measured at week 0, week 13 and 26, midway 121 between the lowest rib margin and the iliac crest. All measures were performed twice using a 122 flexible tape measure and recorded with the accuracy of 0.1 cm. Body composition was 123 determined at weeks 0, 13 and 26 by bioelectrical impedance analysis (STA/BIA Body 124 Composition Analyzer, Akern Bioresearch Srl, Florence, Italy). 125 126 Biochemical measurements 127 Measurements of fasting glucose, insulin and blood lipids (triglycerides, total cholesterol and 128 HDL cholesterol) were performed in Landspitali – University Hospital according to standard 129 operational procedures. Automated clinical chemistry analyzers and routine clinical chemistry 130 methods were used to measure glucose, insulin, triglycerides, total cholesterol and HDL 131 cholesterol. LDL-C concentrations were calculated using Friedewald’s formula. 132 133 Statistical analysis 134 All data analyses were done in SPSS (Statistical Package for the Social Sciences) for Windows, 135 version 20.0 (SPSS Inc., Chicago, Illinois, USA) for statistical analysis. Level of significance in 136 all analyses was set at P<0.05. 137 Normality of variables was checked by inspection and by using the Kolmogorov– 138 Smirnov test. Values are described as means and standard deviations (SD) if normally distributed 139 or as medians and interquartile range (IQR) if not normally distributed. Independent sample t- 140 tests or Mann-Whitney’s U test were used to compare data between groups. When comparing 141 values at week 0 to week 26, paired samples t-test or Wilcoxon signed ranks test was used. 142 The answers to the acceptability questions were combined before statistical analysis 143 combining strongly agree and agree to agree and strongly disagree and disagree to disagree. 144 145 Results 146 Altogether, 23 men started the intervention, 14 in the rye group and 9 in the control group. Three 147 dropped out of the study, two in the rye group and one in the control group. Two dropped out 148 during the first weeks for personal reasons not related to the intervention and one had a medical 149 condition and was not able to finish the intervention. Thus, 20 men completed the study, 12 in 150 the rye group and 8 in the control group. The participants were 42 to 80 years old, with a mean ± 151 SD age of 59 ± 11 years. Baseline characteristics are presented in Table 2. The BMI for the 152 participants at baseline was 24 – 38 kg/m2. There was no significant difference between the 153 groups in baseline characteristics except for participants in the control group had more steps/day 154 than participants in the rye group. 155 156 Dietary intake 157 There was a significant increase in rye consumption in the rye group from week 0 to week 26, 158 with reported rye intake increasing from 11 ± 19 g/day at week 0, to 103 ± 38 g/day at week 13 159 and 86 ± 41 g/day at week 26 (P = 0.003). Rye consumption in the control group did not change. 160 The mean rye consumption during the intervention was 93 ± 44 g/day in the rye group and 10 ± 161 8 g/day for the control group. The rye consumption was right skewed with 7 out of 12 162 participants consuming ≥100 g/day of rye during the intervention, three consuming between 60 163 and 99 g/day and two consuming less than 60 g/day. On average, 72% of the daily rye 164 consumption in the rye group came from the rye porridge. 165 Reflecting the changes in rye intake, there was a borderline significant increase in fiber 166 intake in the rye group, with mean intake 19 ± 10 g/day at week 0, 33 ± 10 g/day at week 13 and 167 28 ± 7 g/day at week 26 (P = 0.06). The fiber intake in the control group was unchanged. 168 Similarly, the total plasma AR concentration and the AR C17:0/C21:0 ratio increased in the rye 169 group, although borderline significant for AR C17:0/C21:0 (P = 0.004 and 0.060, respectively), 170 and not in the control group (Figure 2). There was no significant difference between groups in 171 energy or macronutrient intake during the intervention (Table 3). 172 Four of the twelve participants in the rye group experienced digestive discomfort in the 173 first 13 weeks of rye consumption and two participants in the latter 13 weeks (Figure 3 a)). The 174 mean rye consumption at 13 weeks for the four men that experienced discomfort in the first 175 weeks was 130 ± 13 g/day while it was 94 ± 42 g/day for those who did not experienced any 176 discomfort. Similarly, at week 26, the mean consumption for the two participants that 177 experienced digestive discomfort in the latter 13 weeks was 123 ± 9 g/day while it was 91 ± 31 178 g/day for those that did not experience any discomfort. The difference in rye consumption 179 between those that did experience discomfort and those that did not was, however, not 180 statistically significant. 181 182 Six participants answered positively when asked if they liked the rye products at week 13 and eight at week 26. At week 26 no participants disliked the products Figure 3 b). Similarly, at 183 week 26 eight participants stated that they would prefer products that contain rye more than other 184 products and no participants disagreed (Figure 3 c)). 185 186 Physical activity 187 PA did not change significantly during the intervention neither for the rye group nor the control 188 group (Table 4). For the rye group, the mean steps/day were 3195 ± 988 at week 0, 3581 ± 1298 189 at week 13 and 3817 ± 1401 at week 26. The mean steps/day for the control group were 4347 ± 190 2514 at week 0, 4477 ± 2138 at week 13 and 3970 ± 1076 at week 26. Participants in the control 191 group had significantly more steps/day at baseline than participants in the rye group but there 192 was no significant difference between the groups during the intervention. No participants 193 reached 10 000 steps a day. 194 195 Discussion 196 Findings from this feasibility study suggest that an intervention consisting of up to 100 g whole 197 grain rye a day is well accepted in free living men with habitual low rye and fiber consumption, 198 as measured by drop-out, compliance and acceptability of the products. At week 0 the rye 199 consumption was low, or 11 g/day. In line with the aims of the study, there was a highly 200 significant increase in rye consumption in the rye group from week 0 to week 26, while the rye 201 consumption in the control group was unchanged, indicating a good compliance to the dietary 202 intervention. A total of 20 participants finished the study, or 87% of those who started the 203 intervention, with the rye group increasing their rye intake eight fold on average. Good 204 compliance has also been reported in previous intervention studies, conducted in Sweden and 205 Finland where general rye consumption is greater. In all studies, including the present study, rye 206 products were provided to the participants free of charge [9,11-13]. Most of the rye the 207 participants consumed in the present study came from the rye flakes during breakfast 208 emphasizing the importance of including rye porridge in future studies. Compared to a wheat 209 bread breakfast, whole grain rye porridge has also been shown to improve satiety and decrease 210 subsequent hunger and desire to eat up to 8 hours after consumption [25,26]. Possibly the 211 habitually low bread consumption of Icelandic adults may favor the acceptability of the porridge, 212 as introduction of a bread meal may cause greater changes in the daily diet than changes in 213 breakfast choice. 214 There was no difference between the groups in energy or macronutrient intake during the 215 intervention indicating that the rye products were not added to the baseline diet but were 216 exchanged for other carbohydrate rich food. At week 0 the fiber intake was 19 g/day, comparable 217 to the 17.7 g/day reported for Icelandic men in a recent Icelandic National Dietary Survey [27]. 218 Similar to the increase in whole grain rye consumption, there was an increase in total fiber intake 219 as well as total plasma AR concentration and the AR C17:0/C21:0 ratio. Plasma AR has been 220 established as a biomarker for whole grain wheat and rye intake in numerous studies [23,28,29] 221 and the AR C17:0/C21:0 ratio has been suggested to be an indicator of relative rye intake 222 [10,30,31]. In the present study, the mean AR C17:0/C21:0 ratio went from 0.22 at week 0 to 223 0.34 at week 13 in the rye group and then decreased slightly, to 0.29, at week 26. Although the 224 AR C17:0/C21:0 ratio is approximately 1.0 in whole grain rye and 0.1 in whole grain wheat, in 225 fasting plasma samples the ratio is somewhat lower, around 0.6-0.8 in fasting samples after a 226 whole grain rye intervention for 1-8 weeks [31,32]. It is important to note that the AR 227 C17:0/C21:0 ratio is a marker of relative rye intake, not absolute intake, and the lower ratio in 228 the present study than in previous studies may be indicative of higher whole grain wheat intake 229 as other whole grain products than provided were not allowed in these studies. The ratio in a 230 Swedish population, where rye consumption is appreciable, was 0.5 in samples drawn 8 hours 231 since last meal [33]. The reason for lower ratio in plasma compared to what is observed in whole 232 grain rye foods is likely due to faster metabolism of the AR homologue C17:0 [34,35] and as the 233 absorption half-life of AR is about 6-8 hours and the half-life of AR in plasma is around 5 hours 234 [34], the time since last meal may influence the ratio. In a multicenter intervention with a healthy 235 Nordic diet, rich in whole grain oats, rye and barley, the ratio was comparable to the results of 236 the present study [10]. Ratio above 0.15 is usually indicating some rye intake [21]. 237 The men in the rye group were generally pleased with the rye products and few felt 238 digestive discomfort. The rye intake tended to be higher for those who expressed discomfort 239 indicating that it might be difficult to increase rye consumption beyond 100 g/day in this 240 population. However, the participants seemed to adapt to the products, i.e. fewer felt discomfort 241 at week 26 than 13, more participants liked the products and more answered that they would 242 prefer rye products. The results of the acceptability in the present study is similar to the study by 243 Pereira et al [36] where 9 of 11 participants preferred whole grain diet after a crossover 244 controlled feeding trial with two 6-week feeding periods, whole grain (dietary fiber 28 g/day) vs. 245 refined grain (dietary fiber 17.8 g/day), with no difference between groups in recorded 246 indigestion. Landberg et al [9], also using crossover design with two 6 weeks feeding periods 247 but much higher amounts of whole grain rye (dietary fiber 67 g/day) reported greater drop-out, 248 with 7 of 24 participants not finishing the study because of constipation and 17 participants 249 reporting mild constipation. This indicates that some men may not tolerate such high intake of 250 dietary fiber, at least not without a longer adaption period. 251 While compliance to the dietary intervention was good, the participants did not increase 252 their PA during the intervention and no participant reached the recommended 10 000 steps/day. 253 Other measures for PA showed similar results. The main emphasis in the present study was on 254 the rye consumption and the lack of compliance to the PA advice may be due to the complexity 255 of achieving and maintaining simultaneous changes in both diet and PA. Previous studies have 256 reported that managing changes in multiple behaviors at the same time can be problematic 257 [15,16,18,37,38]. Studies have shown that the most effective PA interventions in diabetic 258 populations are those associated with a greater frequency of contact between the participants and 259 the counselor [39,40]. Thus, better support and monitoring may have been needed to improve the 260 PA in the present study. Pedometers are increasingly being used for self-monitoring and have 261 been shown to increase the number of daily steps taken by people with type 2 diabetes during a 262 16 week intervention although additional strategies are needed for long-term change. [41]. 263 The majority of early studies investigating the effects of PA on type 2 diabetes have 264 incorporated structured exercise programs [42,43] but in recent years interventions aimed at 265 increasing PA using behavior change techniques have been increasingly implemented and 266 recommended [44,45]. The question of the best way to promote PA in this population is, 267 however, still unanswered. Even though the benefits of PA for type 2 diabetes are clear, the 268 majority of people with type 2 diabetes report they are inactive [15,18,19] and attempts to 269 become more active are often met with failure [46-48]. Diabetic patients have been found to be 270 even less ready to engage in PA and other self-care behaviors than patients with other chronic 271 conditions [49] and people with type 2 diabetes perceive the lowest belief in the effectiveness of 272 PA in diabetes management [46]. Thus, increasing PA in this population is challenging and this 273 issue needs to be addressed in the design of future studies. 274 The lack of compliance to the PA advice is a clear limitation to the present study, but at 275 the same time points to the urgent need to find effective means to encourage PA in diabetic 276 populations. Nevertheless, PA was not significantly different between groups during the 277 intervention which makes them comparable for evaluating the health effects of whole grain rye 278 consumption. The strengths of the present study are that the population studied had habitually 279 low intake of whole grain and fiber and the results are important for the design of future studies. 280 Additionally, a rye intervention in patients with type 2 diabetes has not been done before. Self- 281 reported dietary intake was used in conjunction with a dietary biomarker that gives an objective 282 measurement of compliance to rye intake. Previous studies have shown that total plasma AR can 283 be used as a dietary biomarker for whole grain wheat and rye intake [22,23,28,29] and the AR 284 C17:0/C21:0 ratio as an indicator of rye intake [10]. The use of dietary biomarker instead of, or 285 preferably in conjunction with, self-reported intake to measure compliance or investigate a diet- 286 disease relationship is an approach that should be further studied and utilized in future research. 287 288 Conclusion 289 The present feasibility study, carried out in an Icelandic population with habitual low whole 290 grain rye intake, produced valuable results that can be used to design a randomized trial aimed to 291 assess the health effects of high rye consumption. The low dropout-rate, good compliance and in 292 general good acceptability of the rye products showed that the intervention was well received by 293 the intervention group. However, poor compliance to prescribed PA in the intervention 294 emphasizes the need for improved methods to encourage PA in diabetic populations. 295 296 Acknowledgements 297 Authors express sincere gratitude to the participants taking part in the study. We thank Laufey 298 Hrólfsdóttir, Auður Benediktsdóttir and Tinna Eysteinsdóttir for help in planning and conducting 299 the intervention. The following companies provided food products for the study participants: 300 Heilsa Ltd, Myllan, Bernhöftsbakarí, Eggert Kristjánsson Ltd., Sölufélag garðyrkjumanna, 301 Íslenskt ameríska (Ísam), K. Karlsson and Sporthúsið. References 1. Rakha A, Åman P, Andersson R (2010) Characterisation of dietary fibre components in rye products. Food Chemistry 119: 859-867. 2. FAO Faostat (2011) http://faostat.fao.org. 3. Jonsson G (1998) Changes in food consumption in Iceland, 1770–1940. Scandinavian Economic History Review 46: 24-41. 4. Steingrimsdottir L, Thorgeirsdottir H, Olafsdottir A (2003) The Icelandic National Nutrition Survey 2002. Reykjavík: Public Health Institute. 5. Knudsen VK, Fagt S, Trolle E, Matthiessen J, Groth MV, et al. (2012) Evaluation of dietary intake in Danish adults by means of an index based on food-based dietary guidelines. Food Nutr Res 56. 6. Johansson L, Solvoll K (1999) Norkost 1997. Landsomfattende kostholdsundersøkelse blant menn og kvinner i alderen 16-79 år. Rapport 2. Statens råd for ernæring og fysisk aktivitet Oslo. 7. Amcoff E, Edberg A, Barbieri HE, Lindroos AK, Nälsén C, et al. (2012) Riksmaten – vuxna 2010–11. Livsmedels- och näringsintag bland vuxna i Sverige. Uppsala, Sweden: Livsmedelsverket. 8. Pietinen P, Paturi M, Reinivuo H, Tapanainen H, Valsta LM (2010) FINDIET 2007 Survey: energy and nutrient intakes. Public Health Nutr 13: 920-924. 9. Landberg R, Andersson SO, Zhang JX, Johansson JE, Stenman UH, et al. (2010) Rye whole grain and bran intake compared with refined wheat decreases urinary C-peptide, plasma insulin, and prostate specific antigen in men with prostate cancer. J Nutr 140: 2180-2186. 10. Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, et al. (2014) Plasma alkylresorcinols C17:0/C21:0 ratio, a biomarker of relative whole-grain rye intake, is associated to insulin sensitivity: a randomized study. Eur J Clin Nutr 68: 453-458. 11. Laaksonen DE, Toppinen LK, Juntunen KS, Autio K, Liukkonen KH, et al. (2005) Dietary carbohydrate modification enhances insulin secretion in persons with the metabolic syndrome. Am J Clin Nutr 82: 1218-1227. 12. Lundin EA, Zhang JX, Lairon D, Tidehag P, Aman P, et al. (2004) Effects of meal frequency and high-fibre rye-bread diet on glucose and lipid metabolism and ileal excretion of energy and sterols in ileostomy subjects. Eur J Clin Nutr 58: 1410-1419. 13. Juntunen KS, Laaksonen DE, Poutanen KS, Niskanen LK, Mykkanen HM (2003) High-fiber rye bread and insulin secretion and sensitivity in healthy postmenopausal women. Am J Clin Nutr 77: 385-391. 14. Boule NG, Haddad E, Kenny GP, Wells GA, Sigal RJ (2001) Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. Jama 286: 1218-1227. 15. Zanuso S, Jimenez A, Pugliese G, Corigliano G, Balducci S (2010) Exercise for the management of type 2 diabetes: a review of the evidence. Acta Diabetol 47: 15-22. 16. Thomas DE, Elliott EJ, Naughton GA (2006) Exercise for type 2 diabetes mellitus. Cochrane Database Syst Rev: Cd002968. 17. Umpierre D, Ribeiro PA, Kramer CK, Leitao CB, Zucatti AT, et al. (2011) Physical activity advice only or structured exercise training and association with HbA1c levels in type 2 diabetes: a systematic review and meta-analysis. Jama 305: 1790-1799. 18. Taylor R (2008) Pathogenesis of type 2 diabetes: tracing the reverse route from cure to cause. Diabetologia 51: 1781-1789. 19. Michie S, Prestwich A (2010) Are interventions theory-based? Development of a theory coding scheme. Health Psychol 29: 1-8. 20. The Icelandic Food Composition Database (ISGEM). Matís ohf. 21. Ross AB (2012) Present status and perspectives on the use of alkylresorcinols as biomarkers of wholegrain wheat and rye intake. J Nutr Metab 2012: 462967. 22. Ross AB, Kamal-Eldin A, Aman P (2004) Dietary alkylresorcinols: absorption, bioactivities, and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr Rev 62: 81-95. 23. Magnusdottir OK, Landberg R, Gunnarsdottir I, Cloetens L, Akesson B, et al. (2013) Plasma alkylresorcinols reflect important whole-grain components of a healthy Nordic diet. J Nutr 143: 1383-1390. 24. Landberg R, Åman P, Kamal-Eldin A (2009) A rapid gas chromatography-mass spectrometry method for quantification of alkylresorcinols in human plasma. Anal Biochem 385: 7-12. 25. Isaksson H, Sundberg B, Aman P, Fredriksson H, Olsson J (2008) Whole grain rye porridge breakfast improves satiety compared to refined wheat bread breakfast. Food Nutr Res 52. 26. Isaksson H, Tillander I, Andersson R, Olsson J, Fredriksson H, et al. (2012) Whole grain rye breakfast - sustained satiety during three weeks of regular consumption. Physiol Behav 105: 877-884. 27. Thorgeirsdottir H, Valgeirsdottir H, Gunnarsdottir I, Gisladottir E, Gunnarsdottir B, et al. (2011) The Icelandic National Nutrition Survey 2011. Reykjavik: Icelandic Directorate of Health, Icelandic Food and Veterinary Authority and Unit for Nutrition Research. 28. Ross AB, Bourgeois A, Macharia HN, Kochhar S, Jebb SA, et al. (2012) Plasma alkylresorcinols as a biomarker of whole-grain food consumption in a large population: results from the WHOLEheart Intervention Study. Am J Clin Nutr 95: 204-211. 29. Landberg R, Kamal-Eldin A, Andersson A, Vessby B, Aman P (2008) Alkylresorcinols as biomarkers of whole-grain wheat and rye intake: plasma concentration and intake estimated from dietary records. Am J Clin Nutr 87: 832-838. 30. Chen Y, Ross AB, Aman P, Kamal-Eldin A (2004) Alkylresorcinols as markers of whole grain wheat and rye in cereal products. J Agric Food Chem 52: 8242-8246. 31. Linko-Parvinen AM, Landberg R, Tikkanen MJ, Adlercreutz H, Penalvo JL (2007) Alkylresorcinols from whole-grain wheat and rye are transported in human plasma lipoproteins. J Nutr 137: 1137-1142. 32. Landberg R, Aman P, Friberg LE, Vessby B, Adlercreutz H, et al. (2009) Dose response of whole-grain biomarkers: alkylresorcinols in human plasma and their metabolites in urine in relation to intake. Am J Clin Nutr 89: 290-296. 33. Landberg R, Aman P, Hallmans G, Johansson I (2013) Long-term reproducibility of plasma alkylresorcinols as biomarkers of whole-grain wheat and rye intake within Northern Sweden Health and Disease Study Cohort. Eur J Clin Nutr. 34. Landberg R, Linko AM, Kamal-Eldin A, Vessby B, Adlercreutz H, et al. (2006) Human plasma kinetics and relative bioavailability of alkylresorcinols after intake of rye bran. J Nutr 136: 2760-2765. 35. Marklund M, Stromberg EA, Hooker AC, Hammarlund-Udenaes M, Aman P, et al. (2013) Chain Length of Dietary Alkylresorcinols Affects Their In Vivo Elimination Kinetics in Rats. J Nutr. 36. Pereira MA, Jacobs DR, Jr., Pins JJ, Raatz SK, Gross MD, et al. (2002) Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr 75: 848-855. 37. Snowling NJ, Hopkins WG (2006) Effects of different modes of exercise training on glucose control and risk factors for complications in type 2 diabetic patients: a meta-analysis. Diabetes Care 29: 2518-2527. 38. Kavookjian J, Berger BA, Grimley DM, Villaume WA, Anderson HM, et al. (2005) Patient decision making: strategies for diabetes diet adherence intervention. Res Social Adm Pharm 1: 389-407. 39. Greaves CJ, Sheppard KE, Abraham C, Hardeman W, Roden M, et al. (2011) Systematic review of reviews of intervention components associated with increased effectiveness in dietary and physical activity interventions. BMC Public Health 11: 119. 40. Keyserling TC, Samuel-Hodge CD, Ammerman AS, Ainsworth BE, Henriquez-Roldan CF, et al. (2002) A randomized trial of an intervention to improve self-care behaviors of African-American women with type 2 diabetes: impact on physical activity. Diabetes Care 25: 1576-1583. 41. Tudor-Locke C, Bell RC, Myers AM, Harris SB, Ecclestone NA, et al. (2004) Controlled outcome evaluation of the First Step Program: a daily physical activity intervention for individuals with type II diabetes. Int J Obes Relat Metab Disord 28: 113-119. 42. Sevick MA, Dunn AL, Morrow MS, Marcus BH, Chen GJ, et al. (2000) Cost-effectiveness of lifestyle and structured exercise interventions in sedentary adults: results of project ACTIVE. Am J Prev Med 19: 1-8. 43. Hanefeld M, Fischer S, Schmechel H, Rothe G, Schulze J, et al. (1991) Diabetes Intervention Study. Multi-intervention trial in newly diagnosed NIDDM. Diabetes Care 14: 308-317. 44. Avery L, Flynn D, van Wersch A, Sniehotta FF, Trenell MI (2012) Changing physical activity behavior in type 2 diabetes: a systematic review and meta-analysis of behavioral interventions. Diabetes Care 35: 2681-2689. 45. Kirk AF, Barnett J, Mutrie N (2007) Physical activity consultation for people with Type 2 diabetes: evidence and guidelines. Diabet Med 24: 809-816. 46. Wilson W, Ary DV, Biglan A, Glasgow RE, Toobert DJ, et al. (1986) Psychosocial predictors of self-care behaviors (compliance) and glycemic control in non-insulindependent diabetes mellitus. Diabetes Care 9: 614-622. 47. Hays LM, Clark DO (1999) Correlates of physical activity in a sample of older adults with type 2 diabetes. Diabetes Care 22: 706-712. 48. Krug LM, Haire-Joshu D, Heady SA (1991) Exercise habits and exercise relapse in persons with non-insulin-dependent diabetes mellitus. Diabetes Educ 17: 185-188. 49. Boyle RG, O'Connor PJ, Pronk NP, Tan A (1998) Stages of change for physical activity, diet, and smoking among HMO members with chronic conditions. Am J Health Promot 12: 170-175. Figures Figure 1. Outline of the study design and main outcome measurements at each visit Rye consumption, g/day RG 100 80 60 40 CG 20 Fiber consumption, g/day b) a) 120 0 Week 13 RG 30 25 20 CG 15 10 5 Week 26 Week 0 d) 100,0 RG 80,0 60,0 40,0 CG 20,0 0,0 Week 0 Week 13 AR C17:0/C21:0 ratio Total plasma AR, nmol/L 35 0 Week 0 c) 40 Week 13 0,4 0,35 0,3 0,25 0,2 0,15 0,1 0,05 0 Week 26 Week 26 RG CG Week 0 Week 13 Week 26 Figure 2. Changes in a) rye consumption, b) total dietary fiber intake, c) total plasma alkylresorcinols (AR) concentration and d) AR C17:0/C21:0 ratio for the rye group (RG) and the control group (CG) from week 0 to 26 of the intervention. a) 10 b) 10 13 w 26 w 6 4 13 w 26 w 26 w 2 agree 6 13 w 4 13 w 26 w 13 w 2 13 w 0 26 w 8 Participants, n Participants, n 8 26 w 0 neither nor disagree agree neither nor disagree Participants, n c) 10 26 w 8 6 13 w 13 w 4 26 w 2 13 w 26 w 0 agree neither nor disagree Figure 3. The rye group´s acceptability of the rye products (n = 12) assessed by three questions at week 13 (13 w)and week 26 (26 w): a) I have experienced digestive after I started eating rye products b) I like food products that contain rye and c) I would prefer products that contain rye more than other products. Tables Table 1. Whole grain rye, energy, fiber, added sugar and salt content for the rye products provided for the rye group in the study Whole grain rye, Energy, g kcal g g g Rye flakes (porridge) 100 335 11.7 0 <0.01 Rye bread 1 50 193 8.3 1.5 1.2 Rye bread 2 38 235 7.5 0.8 1.2 Rye bread 3 59 179 8.0 N/A 1.5 Rye bread 4 59 162 10.6 2.9 1.2 Rye crisp bread 94 360 19 0.9 <0.01 Data are presented as g or kcal per 100 g of product Fiber, Sugar, Salt, Table 2. Baseline characteristics of the participants Rye group Control group (n = 14) (n = 9) Age, years 57 ± 11 62 ± 11 Height, cm 178 ± 6 178 ± 8 Weight, kg 99.4 ± 11.8 93.5 ± 13.8 2 Body mass index, kg/m 31.5 ± 4.4 29.4 ± 3.0 Waist circumference, cm 114 ± 10 109 ± 8 Fat, % 32.9 ± 5.0 34.4 ± 3.5 Fat free mass, kg 65.7 ± 6.1 61.3 ± 9.3 136 ± 12 Systolic blood pressure, mm Hg 136 ± 11 79 ± 8 Diastolic blood pressure, mm Hg 74 ± 5 1.5 ± 0.7 Triglycerides, mmol/L 1.6 ± 0.9 4.4 ± 1.2 Total cholesterol, mmol/L 4.5 ± 0.8 1.2 ± 0.3 HDL cholesterol, mmol/L 1.2 ± 0.2 2.5 ± 1.1 LDL cholesterola), mmol/L 2.7 ± 0.7 Glucose 0 min, mmol/L 7.0 ± 1.4 6.8 ± 1.2 Insulin 0 min, mU/L 18.2 ± 14.0 12.2 ± 5.0 HOMA-IR 5.4 ± 3.3 3.8 ± 2.0 Marital statusb) 13 (93) 8 (89) Married/in a relationship, n (%) 1 (7) 1 (11) Divorced/widower, n (%) b) Education Elementary school, n (%) 2 (14) 1 (11) 7 (50) 3 (33) College, n (%) 4 (29) 4 (44) University, n (%) 1 (7) 1 (11) Other, n (%) b) Smoking status 7 (50) 2 (22) Never, n (%) 6 (43) 6 (67) Former, n (%) 1 (7) 1 (11) Current, n (%) b) Medication 9 (64) 6 (67) Blood pressure, n (%) 8 (57) 6 (67) Lipid lowering, n (%) 7 (50) 3 (33) Diabetes, n (%) c) Physical activity Steps/day 3195 ± 988 4477 ± 2138* Counts per minute (CPM) 347 ± 80 370 ± 138 Data are presented as mean ± standard deviation unless otherwise indicated * Significantly different from rye group a) n = 22, n = 14 in rye group and n = 8 in control group b) Data are presented as number (percentage) c) n = 21, n = 13 in rye group and n = 8 in control group Table 3. Energy and macronutrient intake for rye and control groups at week 0, 13 and 26. Rye group Week 0 Energy, kcal Week 13 Control group Week 26 Week 0 Week 13 Week 26 2000 ± 534 2175 ± 629 1995 ± 631 2410 ± 325 2249 ± 655 2230 ± 578 Carbohydrates, E% 39 ± 5 39 ± 4 43 ± 5 43 ± 8 43 ± 9 38 ± 6 Dietary fiber, g/day 19 ± 10 33 ± 10 28 ± 7 18 ± 8 21 ± 9 15 ± 7 Added sugar, E% 6±2 6±2 5±3 8±7 7±7 5±4 Protein, E% 21 ± 5 21 ± 5 20 ± 4 19 ± 3 18 ± 2 17 ± 3 Fat, E% 34 ± 7 36 ± 5 37 ± 6 35 ± 8 38 ± 7 39 ± 6 Data are presented as mean ± standard deviation Table 4. Physical activity for the rye and the control groups at the beginning and the end of the intervention. Rye group (n = 12) Control group (n = 8) Week 0 Week 26 Week 0 Week 26 3195 ± 988 3817 ± 1401 4347 ± 2514* 3970 ± 1076 347 ± 80 475 ± 137 370 ± 138 194 ± 90 Sedentary, % 71 ± 5 64 ± 4 69 ± 8 60 ± 7 Light activity, % 24 ± 5 29 ± 2 26 ± 5 33 ± 7 Moderate activity, % 4±2 7±3 4±3 6±2 Vigorous activity, % 0 0 0 1±0 Steps, steps/day Counts per minute Data are presented as mean ± standard deviation * Significantly different from rye group Appendix Instructions for the study groups and questionnaires i ii Instructions for the healthy Nordic diet group in the SYSDIET study - ICELANDIC LEIÐBEININGAR UM MATARÆÐI Tilraunahópur Helstu markmið næstu 6 mánaða (24 vikna) eru að: Auka neyslu grófra kornvara og heilkorns Auka neyslu ávaxta, grænmetis og berja Neyta 3 fiskmáltíða á viku: 2 máltíðir af feitum fisk og 1 máltíð af mögrum fisk Auka neyslu jurtaolíu og viðbits með jurtaolíum iii GRÓFAR KORNVÖRUR Brauð Skiptið út brauðum sem þið eruð vön að borða fyrir eftirtalin gróf brauð (≥6g trefjar/100g) Dæmi: Maltbrauð Normalbrauð Fitty brauð Danskt rúgbrauð (t.d. frá Myllunni) Flatkökur Gróf súrdeigsbrauð úr rúgi Ráðskonubrauð (frá Myllunni) Seydd rúgbrauð (hámark 2 sneiðar á dag) Borðið ≥ 6 sneiðar af grófu brauði daglega (miðað við að hver sneið sé 30-35g). Skipta má 2 brauðsneiðum út fyrir 3 dl (~1 diskur) af hafragraut Vinsamlegast merkjið við daglega neyslu grófra brauða í sérstökum spurningarlista Pasta og hrísgrjón Neytið a.m.k. 2-3 í viku grófs pasta og/eða hýðishrísgrjóna (≥6g trefjar/100g). Vinsamlegast merkjið við vikulega neyslu pasta og hrísgrjóna í sérstökum spurningarlista Almennt: Veljið kornvörur úr rúgi, byggi og höfrum umfram aðrar vörur. Veljið vörur sem innihalda lítið salt. Veljið ósætt morgunkorn, án viðbætts sykurs eða hunangs. Hafragrautur er hollur morgunmatur! iv ÁVEXTIR, GRÆNMETI OG BER Borðið a.m.k. 500 g af ávöxtum, grænmeti og berjum á dag. Skiptið neyslu milli berja, grænmetis og ávaxta á eftirfarandi hátt: Ber: Borðið a.m.k. 150-200g af berjum daglega (t.d. 1,5 dl af aðalbláberjum + 3 tsk af berjadufti). Borðið aðalbláber helst daglega. Borðið einnig jarðarber umfram aðrar tegundir. Dæmi um aðrar tegundir berja (fersk eða frosin) sem hægt er að neyta eru bláber, rifsber, krækiber, hindber, kirsuber, brómber, garðaber, ylliber og týtuber. 15-20 g (u.þ.b. 4,5 – 6 tsk) af berjadufti jafngilda því að borða 150 – 200 g af ferskum eða frosnum berjum. Hægt er að nota berjaduftið út á t.d. mjólkurvörur og grauta Ávextir: Borðið a.m.k. 175g af ávöxtum daglega (dæmi: 1 epli og 1 mandarína). Endilega borðið fjölbreytt úrval ávaxta t.d. epli, perur, appelsínur, mandarínur, banana o.s.frv. Grænmeti: Borðið a.m.k. 175g af grænmeti daglega (dæmi: 1 tómatur, 5 cm gúrkubiti og 1 dl blómkál). Endilega borðið fjölbreytt úrval grænmetis t.d. tómata, gúrku, salat, gulrætur, lauk, sellerí, graslauk, blaðlauk, spergilkál, hvítkál, blómkál, rófur, dill, steinselju o.s.frv. Athugið að kartöflur teljast ekki sem grænmeti! Vinsamlegast merkjið við daglega neyslu berja, ávaxta og grænmetis í sérstökum spurningarlista FISKUR Neytið a.m.k. 3 fiskimáltíða á viku Þar af tvisvar feitan fisk (≥ 4-5% fita), t.d. lax, regnbogasilung, eldissilung, hlýra, rauðmaga, síld, makríl, túnfisksteikur, stórlúðu. Þar af einu sinni magran fisk (< 4% fita), t.d. ýsu, þorsk, skötusel, steinbít, rauðsprettu, kola, saltfisk (útvatnaður), humar, hörpudisk, keilu, kolmunna, rækjur, ufsa, vatnableikju. v KJÖT Veljið magurt kjöt Fuglakjöt Lambakjöt (veljið magra hluta lambsins með ≤10% fitu) Villibráð Svínakjöt (veljð magra hluta með ≤10% fita) Nautakjöt (veljið magra hluta með ≤10% fita) MJÓLK OG MJÓLKURVÖRUR Vinsamlegast notið magrar mjólkurvörur. Borðið a.m.k. 2 skammta af mjólk/mjólkurvörum daglega (1 skammtur = 2,5 dl af drykkjarmjólk, 25g af osti, lítil dós af skyri eða jógúrt) Mjólk, súrmjólk, jógúrt/jógúrtdrykkir, skyr/skyrdrykkir sem er neytt á að vera með ≤1,4g af fitu í 100g Dæmi: o Undanrenna o Fjörmjólk o Skyr o Skyrdrykkir o Mysa o Létt ab mjólk Ostur þarf að innihalda ≤17g af fitu í hverjum 100g en einnig má nota jurtaosta. Dæmi: o Gouda (11% eða 17% fita) o Fjörostur (9% fita) o Létt smurostar (6% fita) o Mysingur (8% fita) o Kotasæla (4,5% fita) vi FITA Repjuolía (rapeseed oil) á að vera notuð í allri matreiðslu. Nota á eingöngu Becel sem viðbit á brauð. DRYKKIR Vatn er besti svaladrykkurinn, neyta má kolsýrðs vatns að vild! Ekki á að neyta sykraðra gosdrykkja. Takmarkið neyslu gosdrykkja með sætuefnum við að hámarki 1-2 glös daglega. Takmarkið neyslu ávaxta og berjasafa við að hámarki 1,5 dl daglega. HREYFING Þar sem líkamsþyngd þín á að haldast stöðug út tímabilið ættirðu ekki að breyta neinu varðandi venjubundna hreyfingu þína. BÆTIEFNI Takið inn eina töflu af D-vítamíni daglega. vii Instructions for the healthy Nordic diet group in the SYSDIET study - ENGLISH DIET INSTRUCTIONS Experimental diet Main goals during the following six months are: to increase the use of whole grain cereal products to increase the use fruits, vegetables and berries to consume 3 fish meals per week (2 fatty fish and 1 low fat fish) to increase the use of vegetable oil and vegetable oil based spreads WHOLE GRAIN CEREAL PRODUCTS Breads Replace all the breads in your habitual diet by test breads. Test breads will be used _________ slices per day as follows: Whole grain rye bread Endosperm rye bread xx xx xx _______ slices per day _______ slices per day _______ slices per day _______ slices per day _______ slices per day If you wish, you may substitute 2 slices of bread for 3,0 dl of whole grain porridge (e.g. made of oat flakes). Please mark the use of test breads daily for the separate questionnaire. Pasta and rice Consume whole grain pasta and unpolished rice (≥ 6 g fiber/100 g) in 2-3 meals or more per week. Please mark the use of pasta and rice weekly for the separate questionnaire. viii FRUITS, VEGETABLES AND BERRIES o Consume 500 g or more of fruits, berries and vegetables per day. o Construct the amount as follows: Berries (≥ 150-200 g/day = e.g. 1,5 dl of frozen bilberries + 3 tbsp of berry powder) o variety of bilberries (e.g. whole berries, puree, extract), strawberries and one of the following: blueberries, black and red currants, raspberries, cherries, blackberries, gooseberries, elderberries, cloudberries, cowberries (wild) Fruits (≥ 175 g/day = e.g. 1 apple + 1 mandarin) o e.g. apples, pears, oranges, mandarins, bananas, quince Vegetables (≥ 175 g/day = e.g. 1 tomato + 5 cm of cucumber + 1 dl of cauliflower) o e.g. tomato, cucumber, lettuce, carrot, leak, onion, rhubarb, sweet beet, turnip, parsnip, parsley, dill, radish, beetroot, chive, brassica [broccoli, Brussels sprout, cabbage (red and white), cale, cauliflower, swede], celery root o Please note that potatoes are not regarded as vegetable. Please mark the use of fruits, vegetables and berries daily for the separate questionnaire. FISH Consume 3 fish meals per week (2 fatty fish meals + 1 low fat fish meal). Fatty fish: e.g. salmon, rainbow trout, Baltic herring, vendace, trout, herring, mackerel, bream, mackerel, sardine, brisling Low fat fish: e.g. whitefish, pike, cod, saithe, perch, pikeperch, haddock, halibut, plaice MEAT Prefer low fat choices. e.g. poultry, lamb (lean products, ≤ 10 % of fat), game, pig (lean products, ≤ 10 %), beef (lean products, ≤ 10 %) ix FATS Rapeseed oil (brands) should be used in cooking and rapeseed and/or sunflower oil and/or soyabean oil based margarines (brands) as spread. MILK PRODUCTS Please use low fat milk products minimum 2 portions daily. (1 portion =2.5 dl of liquid milk products or 2 slices of cheese) milk, sour milk, yogurts, skyrs etc. (≤ 1 % of fat) cheese (≤ 17 % of fat) or vegetable oil based cheese DRINKS Sugar-containing soft drinks should be avoided. Limit the use of fruit and berry juices for max. 1 glass (1.5 dl)/day OTHER ASPECTS DURING THE STUDY Please keep your exercise habits as before, since the aim is that your body weight remains stable during the study. x Instructions for the control group in the SYSDIET study - ICELANDIC LEIÐBEININGAR UM MATARÆÐI Viðmiðunarhópur Helstu markmið næstu 6 mánaða (24 vikna) eru að: Að nota kornafurðir með lágu trefjainnihaldi Að nota viðbit sem inniheldur mjólkurfitu (Smjör eða Smjörva) Borða í mesta lagi 200-250g af berjum, ávöxtum og grænmeti daglega Neyta 1 fiskmáltíðar á viku xi KORNVÖRUR Brauð Skiptið út brauðum sem þið eruð vön að borða fyrir brauð sem innihalda minna en 5-6g af trefjum í 100g. Dæmi: Samlokubrauð, fín (t.d. Bónusbrauð, Heimilisbrauð o.s.frv.) Franskbrauð Snittubrauð Sigtibrauð Skonsur o.s.frv. Vinsamlegast merkjið við daglega neyslu brauða í sérstökum spurningarlista Borðið brauðsneiðar daglega (miðað við að hver sneið sé 30-35g) og helst brauð úr fínmöluðu hveiti. Skipta má 2 brauðsneiðum út fyrir 3 dl (~1 diskur) af graut með lágt trefjainnihald (t.d. hrísmjölsgraut). Pasta og hrísgrjón Notið hveiti pasta (fínt) og hvít hrísgrjón (sem innihalda minna en 5-6 g af trefjum í hverjum 100g). Almennt: Veljið kornvörur úr fínunnu hveiti umfram aðrar vörur. Ekki neyta kornvara/brauðs sem innihalda heil korn Ekki neyta súrdeigsbrauðs sem er búið til úr rúgi xii ÁVEXTIR, GRÆNMETI OG BER Dagleg neysla berja, ávaxta og grænmetis ætti ekki að fara yfir 200-250g. Borðið í mesta lagi 50g (1 dl) af berjum daglega og 150-200g af ávöxtum/grænmeti. Vinsamlegast takið eftir að ekki á að neyta aðalbláberja á tímabilinu! Athugið að kartöflur teljast ekki sem grænmeti! FISKUR Fiskneysla á að takmarkast við í mesta lagi 1 máltíð á viku. KJÖT Neyta má alls kjöts og kjötvara að vild á tímabilinu. MJÓLK OG MJÓLKURVÖRUR Neyta má mjólkurvara að vild. FITA Notið viðbit með mjólkurfitum (t.d. íslenskt Smjör eða Smjörva) við alla matreiðslu og matarundirbúning. DRYKKIR Neyta má allra drykkja að vild, þ.m.t. sykraðra gosdrykkja HREYFING Þar sem líkamsþyngd þín á að haldast stöðug út tímabilið ættirðu ekki að breyta neinu varðandi venjubundna hreyfingu þína. BÆTIEFNI Takið inn eina töflu af D-vítamíni daglega. xiii Instructions for the control group in the SYSDIET study - ENGLISH Instructions for the control group in the SYSDIET study - ENGLISH DIET INSTRUCTIONS Control diet Main goals during the following six months are: to use low fiber cereal products to use milk fat based spreads, butter is also allowed to consume moderate amount of fruits, vegetables and berries to limit the use of fish CEREAL PRODUCTS Breads Replace all the breads in your habitual diet by test breads. Test breads will be used _________ slices per day as follows: Refined wheat bread Wheat oat bread xx xx xx _______ slices per day _______ slices per day _______ slices per day _______ slices per day _______ slices per day If you wish, you may substitute 2 slices of bread for 3,0 dl of porridge that is low in fiber (e.g. rice or semolina). Please mark the use of test breads daily for the separate questionnaire. Pasta and rice Prefer pasta and rice products that are low in fiber (≤ 5-6 g fiber/100 g) xiv FRUITS, VEGETABLES AND BERRIES Consume maximum 50 g (1dl) of berries per day and maximum 150-200 g of fruits and vegetables. Please note that bilberries should NOT be used during the study. FISH Consume maximum 1 fish meal per week. FATS Use milk fat based spreads (brand) or butter on top of breads and in cooking and baking. OTHER ASPECTS DURING THE STUDY Please keep your exercise habits as before, since the aim is that your body weight remains stable during the study. xv Background questionnaire in the SYSDIET study - ICELANDIC Upplýsingar um bakgrunn þátttakenda Númer þátttakanda________________ Svara skal upplýsingum um eigin bakgrunn á þetta eyðublað. Allar upplýsingar eru meðhöndlaðar sem algjört trúnaðarmál. Vinsamlegast svaraðu eftirfarandi spurningum með því að draga hring utan um það númer sem við á. Ef þörf er á skaltu útskýra svarið nánar á tilheyrandi línum eða aftan á blaðið. I) BAKGRUNNUR 1. Kyn 1) Karlkyn 2) Kvenkyn 2. Hver er hjúskaparstaða þín? 1) Gift/ur 2) Í sambúð 3) Einhleyp/ur 4) Fráskilin/n 5) Ekkja/ekkill 3. Hver er menntun þín? 1) Grunnskóli – hef lokið grunnskólaprófi 2) Verkmenntaskóli/iðnskóli – hef lokið verklegu prófi 3) Framhaldsskóli – hef lokið stúdentsprófi 4) Háskóli – hef lokið grunnnámi 5) Háskóli – hef lokið framhaldsnámi 6) Önnur menntun 4. Við hvað starfarðu (þ.e. aðal starfsvettvangur)? 1) Bóndi eða garðyrkjubóndi 2) Byggingavinna eða verksmiðjustarf 3) Skrifstofustarf, þjónustustarf eða önnur kyrrsetustörf 4) Í námi 5) Heimavinnandi 6) Öryrki eða ellilífeyrisþegi 7) Atvinnulaus xvi II) HEILSUFAR 5. Ertu almennt heilsuhraust/ur eða hvernig meturðu núverandi heilsu þína? 1) Góð heilsa 2) Í meðallagi 3) Slæm heilsa 6. Hefur læknir greint þig með eftirtalda sjúkdóma? 1. Kransæðastífla (hjartaáfall) 2. Kransæðasjúkdómar 3. Hjartabilun 4. Aðrir hjartakvillar 5. Hjartakveisa 6. Heilablóðfall 7. Helti (sem kemur og fer vegna hjartasjúkdóms) 8. Háþrýstingur 9. Sykursýki 10. Krabbamein 11. Nýrnabilun 12. Lifrarsjúkdómar 13. Tíðar þvagfærasýkingar 14. Gallsteinar 15. Liðagigt 16. Tíðir smitsjúkdóma 17. Astmi 18. Ofvirkur skjaldkirtill 19. Vanvirkur skjaldkirtill 20. Ristilbólgusjúkdómar 21. Glúteinóþol 22. Mjólkursykursóþol 23. Hægðatregða meðhöndluð með lyfjum 24. Annar langvinnur sjúkdómur 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) 1) Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já Já 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) 2) Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei Nei 7. Tekurðu lyf að staðaldri? 1) Já 2) Nei Ef já, vinsamlegast tilgreindu öll þau lyf sem þú tekur reglulega, þ.e. heiti lyfsins, tilgang notkunar og skammtastærð: _____________________________________________________________________ _____________________________________________________________________ xvii 8. Notar þú reglulega einhver fæðubótarefni, náttúruefni eða náttúrulyf án lyfseðils (s.s. vítamín, steinefni og/eða lýsi) eða fæðuafurðir sem innihalda stanol eða stanolestera? 1) Já – vinsamlegast tilgreindu (tegund og skammtastærð) _____________________________________________________________________ _____________________________________________________________________ 2) Nei III) LÍFSSTÍLL 9. Ertu á einhverju sérstöku mataræði (s.s. mjólkursykurfríu fæði, mjólkursykurskertu fæði, glúteinfríu fæði, megrun, fæðuofnæmi)? 1) Já – vinsamlegast tilgreindu __________________________________________ 2) Nei 10. Reykir þú? 1) Já, reglulega 2) Já, stöku sinnum 3) Ég er hætt/ur að reykja – hvenær?______________________________________ 4) Ég hef aldrei reykt 11. Ef þú reykir daglega, hversu mikið reykirðu? ____________ sígarettur/dag ____________ píputóbak/dag ____________ vindla/dag 12. Hversu oft drekkurðu bjór, pilsner eða annað dauft áfengi? 0) Daglega 1) 3-5 sinnum í viku 2) 1-2 sinnum í viku 3) Nokkrum sinnum í mánuði 4) Einu sinni í mánuði eða sjaldnar 5) Ég drekk ekki dauft áfengi 13. Hversu mikið drekkurðu í einu af daufu áfengi (skrifaðu inn fjölda)? __________ 0,33 l flöskur/dósir __________ 0,5 l flöskur/dósir xviii 14. Hversu oft drekkurðu rauðvín, hvítvín eða annað vín? 0) Daglega 1) 3-5 sinnum í viku 2) 1-2 sinnum í viku 3) Nokkrum sinnum í mánuði 4) Einu sinni í mánuði eða sjaldnar 5) Ég drekk ekki vín 15. Hversu mikið drekkurðu í einu af víni (skrifaðu inn fjölda)? __________ glös __________ flöskur 16. Hversu oft drekkurðu sterkt vín? 0) Daglega 1) 3-5 sinnum í viku 2) 1-2 sinnum í viku 3) Nokkrum sinnum í mánuði 4) Einu sinni í mánuði eða sjaldnar 5) Ég drekk ekki sterkt vín 17. Hversu mikið drekkurðu í einu af sterku víni (skrifaðu inn fjölda)? __________ skammta/skot (4 cl) __________ flöskur (0,5 l) IV) HREYFING 18. Hversu líkamlega erfið er vinnan þín? Ef þú ert atvinnulaus, vinsamlegast merktu við fyrsta valmöguleikann. 1) Aðallega skrifstofuvinna, ekki mikil ganga 2) Geng talsvert mikið en lyfti hvorki né ber þunga hluti 3) Geng mjög mikið og lyfti þungum hlutum eða geng mikið upp og niður stiga 4) Mjög líkamlega erfið vinna, lyfti eða ber mikið af þungum hlutum, moka o.þ.h. 19. Hvernig ferðastu á milli staða? 1) Með bíl eða strætó 2) Geng ____________ mín/dag 3) Hjóla_____________ mín/dag xix 20. Stundarðu einhverja hreyfingu í frístundum þínum? a) Garðyrkja, veiðar, berjatínsla o.þ.h. 1) 4 sinnum í viku eða oftar 2) 2-3 sinnum í viku 3) Einu sinni í viku 4) 2-3 sinnum í mánuði 5) Sjaldnar b) Líkamleg áreynsla 30 mín í senn þannig að þú verðir móð/ur (ganga, hjóla, skokka, skíða o.þ.h.): 1) 4 sinnum í viku eða oftar 2) 2-3 sinnum í viku 3) Einu sinni í viku 4) 2-3 sinnum í mánuði 5) Sjaldnar xx Background questionnaire in the SYSDIET study - ENGLISH Backgound information form CONFIDENTIAL Number of the participant:________________________ The background information for the study is collected with this form. All information will be handled with absolute confidentiality. Kindly answer the following questions by circling the number of the suitable option. If needed, specify the answer on the line or back of the paper. I BACKGROUND 1. Gender 1 male 2 female 2. Marital status 1 Married 2 Domestic partnership 3 Unmarried 4 Judicial separation or divorced 5 Widower/Widow 3. What is your education? 1 Elementary or comprehensive school 2 Middle school 3 Vocational school/secondary modern school 4 Upper secondary school or senior high school 5 College-level training 6 Polytechnic or vocational high school 7 University 6 Other education, please specify______________________________________ 4. What is your profession – what kind of work you do most of your time? 1 Farming, stock raising, forest work 2 Factory, mine or construction work 3 Office work, service trade, mental work 4 Studying 5 House wife or husband 6 Pensioner or senior citizen 7 Unemployed xxi II HEALTH STATUS 5. How do you see your health and well being at the moment? 1. good 2. average 3. poor 6. Has your physician diagnosed some of these diseases ? 1. Coronary thrombosis (myocardial infarction) 1. Yes 2. No 2. Coronary artery disease 1. Yes 2. No 3. Congestive heart failure/cardiac insufficiency 1. Yes 2. No 4. Other cardiac problem 1. Yes 2. No 5. Angina pectoris 1. Yes 2. No 6. Ischaemic attack 1. Yes 2. No 7. Intermittent claudication 1. Yes 2. No 8. Hypertension 1. Yes 2. No 9. Diabetes 1. Yes 2. No 10. Cancer 1. Yes 2. No 11. Renal failure 1. Yes 2. No 12. Hepatic disease 1. Yes 2. No 13. Frequent urinary tract infection/cytitis 1. Yes 2. No 14. Cholelithiasis 1. Yes 2. No 15. Rheumatoid arthritis 1. Yes 2. No 16. Frequent infectious diseases 1. Yes 2. No 17. Asthma bronchiale 1. Yes 2. No 18. Hyperthyroidism 1. Yes 2. No 19. Hypothyroidism 1. Yes 2. No 20. Inflammatory bowel disease 1. Yes 2. No 21. Celiac disease 1. Yes 2. No 22. Lactose intolerance 1. Yes 2. No 23. Constipation with medication 1. Yes 2. No 24. Other chronic disease 1. Yes 2. No Please specify?_____________________________________ 7. Do you have regular medication? 1 Yes 2 No If so, please write here all the medication you use regularly: name, purpose of use and dosage : ______________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ xxii 8. Do you use regularly any nutritional supplements or natural products without description (vitamin and/or mineral supplements, fish or vegetable oil supplements) or stanol- or sterol-ester-containing food products ? 1 Yes, please specify (brand name, dosage/day)? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ 2 No III LIFESTYLE HABITS 9. Do you have a special diet (such as lactose free, low-lactose, gluten free, weight loosing, food allergy)? 1 Yes, please speficy___________________________________________________ 2 No 10. Do you smoke? 1 Yes, regularly 2 Yes, occationally 3 I have ceased smoking, when ?__________________ 4 I have never smoked 11. If you smoke daily, how many portions/ day? ____________ cigarettes/day ____________ pipe tobacco/day ____________ cigars/day 12. How often do you drink beer, cider, or other mild alcoholic beverages? 0 Daily 1 3-5 times/week 2 1-2 times/week 3 Twice/couple of times /month 4 Once a month or more seldom 5 I do not drink mild alcoholic beverages 13. How much do you drink mild alcoholic beverages at once ? __________ 0,33 l bottles/ cans (strike out unnecessary option) __________ 0,5 l bottles/ cans 14. How often do you drink wine? 0 Daily 1 3-5 times/week 2 1-2 times/week 3 Twice/couple of times /month 4 Once a month or more seldom 5 I do not drink wine xxiii 15. How much do you drink wine at once?__________ glasses / bottles (strike out unnecessary option) 16. How often do you drink alcoholic spirits? 0 Daily 1 3-5 times/week 2 1-2 times/week 3 Twice/couple of times /month 4 Once a month or more seldom 5 I do not drink alcoholic spirits 17. How much do you drink alcoholic spirits at once? __________ serving portions (4 cl) / 0,5 l bottles (strike out unnecessary option) IV PHYSICAL EXERCISE 18. How physically demanding is your work? If you are unemployed, please choose the first option. 1 Mainly office work, not much walking 2 Walking quite a lot, but no raising or carrying of heavy objects 3 Very much walking and raising of objects, or climbing of stairs or uphills 4 Heavy physical work, including a lot of raising or carrying of heavy objects, digging, shoveling, or deforesting 19. How do you commute? 1 By car or public transportation 2 Walking ____________ min/day 3 Cycling_____________ min/day 20. Do you do leisure time physical exercise? Gardening, timbering, picking berries, fishing etc.: 1 4 times a week or more 2 2-3 times/week 3 Once a week 4 2-3 times/month 5 More seldom Fitness training ½ hour at once so that you become breathless (walking, cycling, skiing, jogging etc.): 1 4 times a week or more 2 2-3 times/week 3 Once a week 4 2-3 times/months 5 More seldom xxiv Instructions for rye group in the ARI feasibility study - ICELANDIC Rúghópur Helstu markmið næstu 6 mánaða eru: Að borða að minnsta kosti 100 g af heilkorna rúgi daglega Að öðru leyti borða samkvæmt ráðleggingum um mataræði sykursjúkra Að hreyfa þig rösklega þrisvar í viku, í að minnsta kosti 45 mínútur í senn, og að ganga að minnsta kosti 10.000 skref aðra daga Matur Fundur með næringarráðgjafa í sal 9 í Sporthúsinu fimmtudaginn 18.apríl kl.20:00. Makar eru velkomnir á fundinn! Þú getur sótt brauð og aðrar rúgvörur þér að kostnaðarlausu, annan hvern miðvikudag frá kl 11 til 13 á rannsóknarstöð RÍN í Kópavogi (við munum senda ykkur sms daginn áður til að minna á þennan tíma). Einu sinni í mánuði verður líka grænmeti í boði. Þú færð matardagbók til að halda utan um rúgneysluna. Við mælum með því að þið aukið rúgneysluna rólega fyrstu dagana en séuð komnir í fullan skammt (100 g eða 10 punkta) á annarri viku. Mikilvægt er að drekka vel samhliða aukinni trefjaneyslu. Hreyfing Nafn þitt hefur nú verið skráð í Sporthúsinu og þú getur gefið þig fram í afgreiðslu til að opna aðganginn. Vilji maki þinn kaupa kort á afslætti í Sporthúsinu (4400 kr. á mánuði árskort) skaltu tala við Ríkharð Óskar Guðnason, hann er með skrifstofu á móti afgreiðslunni í Sporthúsinu. Þú getur nú farið á fund með sjúkraþjálfara í Sporthúsinu, þú átt tíma ____________________ Þá átt síðan tíma hjá honum Vali einkaþjálfara í Sporthúsinu ____________________ Við mælum með því að byrja rólega og auka svo smátt og smátt hreyfinguna. Sjúkraþjálfari og hann Valur upp í Sporthúsi munu leiðbeina þér varðandi það hvaða álag hentar þér. Þátttakendur munu fá skrefamæla til að fylgjast með hreyfingu sinni. Næstu mælingar Næstu skráningar í matardagbók og hreyfidagbók fara fram _______________ Mælingar í Kópavogi: ___________________ Ef einhverjar spurningar vakna ekki hika við að hafa samband í tölvupósti ([email protected] eða ) eða í síma 543-8420 eða 695-988. xxv Instructions for the control group in the ARI feasibility study - ICELANDIC Hreyfihópur Helstu markmið næstu 6 mánaða eru: Takmarka neyslu matvara sem innihalda rúg Að öðru leyti borða samkvæmt ráðleggingum um mataræði sykursjúkra? Að hreyfa þig rösklega þrisvar í viku, í að minnsta kosti 45 mínútur í senn, og að ganga að minnsta kosti 10.000 skref aðra daga Matur Fundur með næringarráðgjafa í sal 9 í Sporthúsinu fimmtudaginn 18.apríl kl.20:00. Makar eru velkomnir á fundinn! Þú getur sótt heilhveitibrauð þér að kostnaðarlausu annan hvern miðvikudag frá kl 11 til 13 á rannsóknarstöð RÍN í Kópavogi (við munum senda ykkur sms daginn áður til að minna á þennan tíma). Einu sinni í mánuði verður líka grænmeti í boði. Hreyfing Nafn þitt hefur nú verið skráð í Sporthúsinu og þú getur gefið þig fram í afgreiðslu til að opna aðganginn. Vilji maki þinn kaupa kort á afslætti í Sporthúsinu (4400 kr. á mánuði árskort) skaltu tala við Ríkharð Óskar Guðnason, hann er með skrifstofu á móti afgreiðslunni í Sporthúsinu. Þú getur nú farið á fund með sjúkraþjálfara í Sporthúsinu, þú átt tíma ____________________ Þá átt síðan tíma hjá honum Vali einkaþjálfara í Sporthúsinu ____________________ Við mælum með því að byrja rólega og auka svo smátt og smátt hreyfinguna. Sjúkraþjálfari og hann Valur upp í Sporthúsi munu leiðbeina þér varðandi það hvaða álag hentar þér. Þátttakendur munu fá skrefamæla til að fylgjast með hreyfingu sinni. Næstu mælingar Næstu skráningar í matardagbók og hreyfidagbók fara fram _______________ Mælingar í Kópavogi: ___________________ Ef einhverjar spurningar vakna ekki hika við að hafa samband í tölvupósti ([email protected]) eða í síma 543-8420 eða 695-988. xxvi Booklet for rye products in the ARI feasibility study - ICELANDIC Númer þátttakanda: _____________ xxvii Af hverju að velja helkorna rúg? Þegar talað er um heilkorn er átt við þegar allir hlutar kornsins eru notaðir, hvort sem kornið er notað í heilu lagi eða malað, og á það við um bygg, hafra, hveiti, rúgur, hirsi, hrís og maís. Venjulegt hvítt hveiti er til dæmis ekki heilkornaafurð þar sem búið er að fjarlægja „skelina/hýðið“ og kímið og þar með töluvert af hollustuefnum kornsins. Heilkorn eru einstaklega holl þar sem þau innihalda alla hluta kornsins og því öll hollustuefni þess (trefjar, prótein, vítamín og steinefni). Skel: trefjaríkasti hlutinn, steinefni, B-vítamín Frækjarni: aðallega kolvetni og sterkja Kím: mikið af E- og B-vítamínum, próteinum og öðrum hollefnum Heilkorn eru mjög mettandi, sem þýðir að við borðum minna og eigum auðveldara með að stjórna þyngdinni. Einnig eru trefjarnar sérstaklega góðar fyrir meltinguna. Hollustuefnin í heilkorni geta einnig minnkað hættuna á að fá sykursýki, hjarta- og æðasjúkdóma og ákveðnar tegundir krabbameins. Þó er ekki fullvitað hvaða efni það eru í heilkorninu sem hafa þessi áhrif, eða jafnvel hvort það sé samspil milli ákveðinni efna. Því er mikilvægt að neyta heilkorns, en ekki aðeins ákveðinnna hluta kornsins. Sænskir samstarfsaðilar okkar hafa sýnt fram á að mataræði sem inniheldur mikinn rúg og trefjar lækkar insúlínstyrk í blóði og blóðsykur helst stöðugri (hefur jákvæð áhrif á blóðsykurstjórnun). xxviii Hvernig virkar stigakerfið? Þátttakendur í íhlutunarhópi eru beðnir um að borða a.m.k. 100 g af rúgi á dag. Til að auðvelda ykkur að fylgjast með og skrá niður rúgneysluna verður farið eftir stigakerfi, þar sem hver rúgvara gefur ákveðið mörg stig Stigakerfið er hugsað þannig að um 10g af rúgi gefi 1 stig. Í hverri viku eruð þið beðnir um að neyta rúgvara sem gefa samtals um 70 stig, eða um 10 stigum að meðaltali á dag. Hægt er að hliðra neyslunni aðeins til, t.d. borða sem samsvarar færri stigum einn daginn, en þá verður að reyna bæta það upp með því að fá fleiri stig einhvern annan dag. En heildarfjöldi stiga ætti ekki að fara undir 70 á viku. Meðfylgjandi er listi af vörum sem innihalda heilkorna rúg og gefið upp hversu mörg stig einn skammtur af vörunni gefur. Ef verið er að neyta annarra rúgvara en þeirra sem eru á listanum, vinsamlegast hafið samband við rannsóknaraðila hægt sé að reikna út hversu mörg stig hver skammtur gefur ykkur. Við mælum með því að þið aukið rúgneysluna jafnt og þétt fyrstu tvær vikur íhlutunarinnar, byrjið rólega en séuð komnir í fullan skammt (a.m.k. 100 g, eða 10 punkta á dag) í annarri viku. Gott er t.d. að byrja á tveim rúgbrauðsneiðum á dag og bæta svo grautnum inn í vikunni þar á eftir. Þannig ætti meltingarkerfið að ná að aðlagst breyttri neyslu og auknum trefjum. En alltaf er þó einhver möguleiki á því að þið finnið fyrir óróa í maga og auknum vindgangi til að byrja með, það er alveg eðlilegt. Ef spurningar vakna, endilega hafið samband (Óla Kallý Magnúsdóttir, sími 543 8422 ([email protected]) eða Laufey Hrólfsdóttir, sími 695 9898 ([email protected])) xxix Heill rúgur Heilan rúg má leggja í bleyti eða sjóða og nota meðal annars í bakstur og sem meðlæti (svipað og bygg). Stig 10g af ósoðnum rúgi gefa 1 stig Rúgflögur Rúgflögur minna einna helst á grófvalsaða hafra, en eru úr heilum rúgkjörnum, þær eru þó heldur þéttari í sér. Rúgflögur má borða eins og haframjöl, með mjólk eða súrmjólk, einnig er gott að bæta við kanil, rúsínum, eplabitum, múslí eða öðru þess háttar. Einnig má sjóða flögurnar í graut (sjá skjal sem heitir uppskriftir). Rúgflögurnar eru 100% heill rúgur og við mælum með því að borða rúgflögur sem flesta daga, en það auðveldar að ná tilskildum stigafjölda yfir daginn. Fyrirtækið Heilsa ehf. styrkir rannsóknina með því að sjá þátttakendum fyrir rúgflögum allt rannsóknartímabilið. Stig 10g ósoðnum rúgflögum gefa 1 stig 60g af ósoðnum rúgflögum (u.þ.b. 1½ dL) gefa 6 stig 100g af ósoðnum rúgflögum (u.þ.b. 2 ½ dL) gefa 10 stig xxx Rug Fras (Rúgkoddar) Rug Fras er tegund af morgunkorni, sem svipar til Havrefras (Hafrakoddar). Í Rugfras eru um 62% heilkorna rúgur. Rug Fras má t.d. borða sem hluta af morgunmat eða sem millimál. Stig 16g af Rugfras gefa 1 stig 50g af Rugfras (u.þ.b. 4 dL) gefa 3 stig Rúgflakes (unnið morgunkorn) Svipaðar kornflögum, nema innihalda heilkorna rúg ásamt öðrum korntegundum. Henta vel sem hluti af morgunmat eða jafnvel sem millimáltíð. Stig 20g af rugflakes (u.þ.b. 1 dL) gefa 1 stig 40g af rugflakes (u.þ.b. 2 dL) gefa 2 stig Hrökkbrauð Finn crisp, þetta klassíska þunna, inniheldur 94% heilkorna rúg. Finn crisp, kringlótt, inniheldur 72% heilkorna rúg xxxi Wasa, hlutfall rúgs í mismunandi tegundum, ~90% Burger, inniheldur 95% heilmalaðan rúg. Allar tegundir Burger hrökkbrauðs í lagi, innihalda ~90% heilmalaðan rúg eða meira) Ryvita dark rye, 95% heilmalaður rúgur Hver hrökkbrauðssneið gefur 1 stig, nema Finn crisp þunnt – þá eru tvær sneiðar sem gefa 1 stig. xxxii Rúgbrauð – Brauð sem innihalda heilkorna rúg Brauð sem innihalda rúg ættu að koma við sögu á hverju degi. Munið að vigta brauðsneiðarnar, alla vega í byrjun til að geta skráð réttan stigafjölda, ekki er verra að skera brauðið nokkuð þykkt. Búið er að reikna út stig fyrir nokkrar algengar tegundir af brauði eins og sjá má í listanum hér að neðan. Myllan styrkir rannsóknina með brauðgjöfum og þátttakendur geta nálgast brauð frá Myllunni hjá rannsóknaraðilum. Að sjálfsögðu má þó neyta annarra brauða en þeirra sem verða sköffuð, en endilega hafið þá samband við rannsóknaraðila til að hægt sé að reikna út hversu marga punkta ein sneið er að gefa ykkur. Mikið er til af góðum innfluttum rúgbrauðum og auk þess er hægt að fá óseytt rúgbrauð í flestum handverksbakaríum. Mikill munur getur verið á rúginnihaldi brauða og þyngd brauðsneiða, því getur verið mikill munur á því hversu mörg stig fást fyrir eina brauðsneið. Einnig verður að hafa í huga að þó seytt rúgbrauð, þrumari og jöklabrauð innihaldi 40-45% rúg, þá innihalda þessi brauð einnig hátt hlutfall af viðbættum sykri, eða kringum 20% (20g/100g) og eru því ekki hentug sem aðaluppspretta rúgs. Maltbrauð, inniheldur um 50% heilmalaðan rúg. Ein 40 g sneið gefur 2 stig. Inniheldur auk þess 1,5g sykur. 8,3-8,8 g trefjar og 472mg natríum í 100g (jafngildir 1,2g af matarsalt). Danskt rúgbrauð, inniheldur 38% heilan/heilmalaðan rúg. Ein 40 g sneið gefur 1,5 stig. Inniheldur auk þess 0,8g sykur, 7,5 g trefjar og 480mg natríum í 100g (jafngildir 1,2g af matarsalti). xxxiii Ráðskonubrauð, inniheldur rúmlega 30% heilmalaðan rúg. Ein 40 g sneið gefur 1 stig. Inniheldur auk þess 1,5g sykur, 6,2 g trefjar og 498mg natríum í 100g (jafngildir 1,2g af matarsalti). Lífskorn, inniheldur 25-27% heilmalaðan rúg. Ein 35 g sneið gefur 1 stig. Inniheldur auk þess 0g sykur, 7-9g trefjar og 490mg natríum í 100g (jafngildir 1,2g af matarsalti). Sólkjarnarúgbrauð, inniheldur um 24% heilan/heilmalaðan rúg. Ein 40 g sneið gefur 1 stig. Inniheldur auk þess 1,6g sykur, 6.9 g trefjar og 492mg natríum í 100g (jafngildir 1,2g af matarsalti). Jöklabrauð, inniheldur um 45% heilmalaðan rúg. Ein 50 g sneið gefur 2,5 stig. Inniheldur auk þess 21,5g sykur, 7,6 g trefjar og 565mg natríum í 100g (jafngildir 1,4g af matarsalti). xxxiv Þrumari, inniheldur rúmlega 40% heilmalaðan rúg. Ein 50 g sneið gefur 2 stig. Inniheldur auk þess 19,4g sykur, 7,3 g trefjar og 343mg natríum í 100g (jafngildir 0,9g af matarsalti). Seytt rúgbrauð, inniheldur rúmlega 40% heilmalaðan rúg. Ein 40 g sneið gefur 1,5 stig (forskorið), ein 50 g sneið gefur 2 stig. Inniheldur auk þess 19,4g sykur, 7.3 g trefjar og 343 mg natríum í 100 g (jafngildir 0,9g af matarsalti). Delba Sunflower seed bread, inniheldur 56% heilmalaðan rúg. Tvær litlar sneiðar (62.5g) gefa 3,5 stig. Inniheldur auk þess 8,1 g af trefjum og 600 mg natríum í 100 g (jafngildir 1.5g af matarsalti) xxxv Delba RYE BREAD, inniheldur 59% heilmalaðan rúg. Tvær litlar sneiðar (62.5g) gefa 3,5 stig. Inniheldur auk þess 8g af trefjum og 600 mg natríum í 100 g (jafngildir 1.5g af matarsalti) Delba 5 Grain Bread, inniheldur 48% heilmalaðan rúg. Tvær litlar sneiðar (62.5g) gefa 3 stig. Mestemacher Wholemeal Rye Bread, inniheldur 59% heilmalaðan rúg. Ein 71g sneið gefur 4 stig. Inniheldur auk þess 2.9g sykur, 10.6 g trefjar og 500 mg natríum í 100 g (jafngildir 1.2g af matarsalti). Brauð frá Bernhöftsbakarí, inniheldur um 44% heilmalaðan rúg. Tvær sneiðar (80g) gefa 3,5 stig. xxxvi Biona Organic Rye bread, inniheldur 62% heilmalaðan rúg. Ein 71g sneið gefur 4,5 stig. Inniheldur auk þess 10.1 g trefjar og 1.13 g af salti í 100g. Biona Organic Rye bread. Sunflower seed, inniheldur 57% heilmalaðan rúg. Ein 71 sneið gefur 4 stig. Inniheldur auk þess 9.7 g trefjar og 0.95 g af salti í 100g. Biona Organic Rye bread. Amaranth / Quinoa. Inniheldur 54% heilmalaðan rúg. Ein 71 sneið gefur 4 stig. Inniheldur auk þess 10.1 g trefjar og 0.98 g salt í 100 g. Finn Crisp Rye snack. Inniheldur 81% heilmalaðan rúg. 30 g gefa 2,5 stig. Inniheldur auk þess 18 g trefjar og 700 mg natríum í 100 g (jafngildir 1,7 g af matarsalti) xxxvii Dæmi um hvernig skrá má í rúgdagbókina: Dagur:____Mánudagurinn 4.mars 2013___________________ Vara Hlutfall rúgs pr 100g af vöru Stig á einingu Hversu mikið borðað Hversu mörg stig Rúgflögur 100% 1 stig á 10g, t.d. 1dL=40g sem gera 4 stig 1,5dL (í morgungraut) 6 Heill rúgur 100% 1 stig á 10g af ósoðnum rúg. Ef soðinn þá 1 stig á 25g Rugfras 62% 1 stig á 1 dL Hrökkbrauð (Finn Crisp þunnt) 94% 1 sneið 0,5 punktar 2 stk með kvöld kaffinu 1 Hrökkbrauð (Finn Crisp kringlótt) 72% 1 stig á sneið Hrökkbrauð Wasa/Burger ~90% 1 stig á sneið 50% 2 stig á sneið 2 sneiðar í hádegismat 4 Hrökkbrauð annað_________ Brauð: maltbrauð Brauð _____________ Annað _____________ Annað _____________ Annað _____________ Annað _____________ Annað _____________ Alls: 11 stig xxxviii Rúgdagbók Dagur:_____________________________ Vara Hlutfall rúgs pr 100g af vöru Stig á einingu Rúgflögur 100% 1 stig á 10g, t.d. 1dL=40g sem gera 4 stig Heill rúgur 100% 1 stig á 10g af ósoðnum rúg. Ef soðinn þá 1 stig á 25g Rugfras 62% 1 stig á 1 dL Hrökkbrauð (Finn Crisp þunnt) 94% 1 sneið 0,5 stig Hrökkbrauð (Finn Crisp kringlótt) 72% 1 stig á sneið Hrökkbrauð Wasa/Burger ~90% 1 stig á sneið Hversu mikið borðað Hversu mörg stig Hrökkbrauð annað_________ Brauð _____________ Brauð _____________ Annað _____________ Annað _____________ Annað _____________ Annað _____________ Annað _____________ xxxix Food diary and instructions in the ARI feasibility study - ICELANDIC Númer þátttakanda: _____________ Matardagbók Rannsóknastofa í næringarfræði xl LEIÐBEININGAR FYRIR FÆÐUSKRÁNINGU ALMENNAR UPPLÝSINGAR Skráðu allt sem þú borðar og drekkur í 4 daga samfleytt, frá miðvikudegi til laugardags. Skrifaðu í matardagbókina um leið og þú borðar svo ekkert gleymist. Það á að vigta allan mat og drykk. Drykki má þó skrá í millilítrum (ml) eða desilítrum (dl) ef fólk vill það frekar. Reyndu að borða eins og venjulega og ekki láta fæðuskráninguna hafa áhrif á hvað þú borðar. Byrjaðu alltaf að skrá fyrstu máltíð (t.d. morgunmatinn) daginn sem skráningin hefst. FÆÐUSKRÁNING Byrjið á nýrri blaðsíðu fyrir hvern dag og skrifið dagsetninguna efst á hvert blað (sjá dæmi á næstu bls.) Skrifið tímasetningu máltíða við hverja máltíð (fyrsti dálkurinn). Gott er að hafa eitt línubil á milli máltíða hvers dags. Skrifið bara eina fæðutegund í hverja línu (t.d. “þriggjakornabrauð” í eina línu og “Smyrja” í næstu línu). Lýsið mat, drykk og magni eins nákvæmlega og hægt er: o Gefið upp tegund af matvælum (t.d. hvort þið notið undanrennu eða léttmjólk) og jafnvel framleiðanda eða vörumerki matvælanna. Gefið upp fisktegund (skrifið t.d. ”soðin ýsa”) og kjöttegund (t.d. ”ungnautahakk 8-12% fita”) í stað þess að skrifa einungis ”fiskur” eða ”kjöt”. o Lýsið matreiðsluaðferðum, skrifið t.d. hvort kjötið var steikt, soðið eða grillað, hvort grænmetið var hrátt, soðið eða niðursoðið og hvort ávextirnir voru borðaðir ferskir eða niðursoðnir o.s.frv. o Ef um er að ræða heimatilbúna blandaða rétti (t.d. pottrétti þar sem ekki er hægt að vigta hvert matvæli beint á diskinn), þá er best ef grófleg uppskrift fylgir með (a.m.k. hvaða hráefni voru notuð). T.d. ”Pottréttur: 2 msk repjuolía, 300 g nautakjöt (10% fita), 1 laukur, tómatar í dós (stærð dósar), hvítlaukur og krydd.” Og fylla svo inn hversu mikið var borðað af umræddum rétti. Ef maturinn er ekki kláraður af diskinum, vigtið þá afganga og skráið hversu mikið var afgangs. Ekki gleyma að skrá alla drykki (einnig vatn), millibita og sælgæti. Ef lýsi eða önnur bætiefni/fæðubótarefni/náttúruefni eru tekin inn skráið þau þá líka. xli DÆMI UM SKRÁNINGU MÁLTÍÐA Dagssetning: Mánudagur 14.janúar Kl. 7:30 10:15 Matur / drykkur (heiti, lýsing, vöruteg) Skammtað magn (g) Magn afgangs (g) Magn sem er neytt (g) Hafragrautur 294 294 Undanrenna 120 Þriggjakornabrauð (muna að skrifa nafn á bakaríi eða framleiðanda) Smjör 37 37 5 5 Ostur, Gouda 17% 22 22 Epli 137 Kaffi 220 43 31 77 106 220 O.s.fr v. xlii LEIÐBEININGAR FYRIR VIGTUN MATVÆLA OG DRYKKJA NÚLLSTILLING: Til að fá þyngdina á hverri matartegund fyrir sig þarf að núllstilla vogina á milli tegunda með því að ýta á ”reset” eða ”0” (mismunandi eftir vogum). Dæmi 1: Vigtun á brauði með viðbiti og osti Best er að setja disk á vogina og núllstilla, setja síðan brauðsneiðina á diskinn og skrá þyngdina á brauðinu. Vogin er núllstillt aftur með diskinn og brauðið á voginni, smyrjið brauðsneiðina og þá sést þyngd viðbitsins á voginni. Núllstillið svo aftur áður en osturinn er settur á brauðið til að geta skráð þyngd ostsins. Það getur verið erfitt að vigta eitthvað sem er minna en 3 grömm á þennan hátt en í þeim tilvikum er hægt að skrifa ”< 3 g”. Dæmi 2: Vigtun á kvöldmat (t.d. fiskur með kartöflum og grænmeti). Diskurinn er settur á vogina og vogin núllstillt. Vigtið síðan hverja fæðutegund fyrir sig og munið að núllstilla á milli. Setjið fiskinn á diskinn og skráið þyngd fisksins. Stillið aftur á núll með diskinn og fiskinn enn á voginni. Kartöflurnar eru næst settar á diskinn og þyngdin skráð. Stillt aftur á núll eins og áður. Setjið hverja grænmetistegund fyrir sig á vogina, vigtið og núllstillið á milli. Ef forblandað salat er notað tilgreinið þá helstu hráefni í salatinu t.d. salatblöð, gúrka, tómatar. Ef dressing er notuð núllstillið þá vogina áður en dressingunni er hellt yfir salatið. ATH! Ef ekki er klárað af disknum vigtið þá afganga eftir máltíðina og skráið það sem er afgangs (t.d. bein í kjöti eða eplakjarni). Alla óæta hluta þarf að vigta eftir máltíðina og skrá sem afgang. Þegar fæðuskráningu er lokið skilið þá matardagbókinni og voginni í næstu heimsókn. Ef þið hafið einhverjar spurningar vinsamlega hafið samband við: Auði Benediktsóttur í síma 543 8424 (tölvupóstfang [email protected]) eða Laufeyju Hrólfsdóttur í síma 543 8420 (tölvupóstfang [email protected]) eða Ólu Kallý Magnúsdóttur í síma 543 1452 (tölvupóstfang [email protected]) xliii DAGUR 1 (___________________) Dagssetning: Kl. Matur / drykkur (heiti, lýsing, vöruteg) Skammtað magn (g) Magn afgangs (g) Magn sem er neytt (g) xliv Background questionnaire in the ARI feasibility study Númer þátttakanda: ___________ SPURNINGAR UM LÍFSSTÍL Tóbaksnotkun Já Nei Hefur þú einhvern tímann reykt sígarettur, vindla, vindlinga eða pípu daglega yfir eitt ár eða lengur? (Ef nei, farið að spurningum um áfengisneyslu) □ □ Reykir þú núna? (Ef nei, farið á spurningu um fyrri reykingar) □ □ Núverandi reykingar Hversu margar sígarettur/vindla/vindlinga/pípur reykir þú að jafnaði á dag? Sígarettur stk Vindlar stk Vindlingar stk Pípur stk Fyrri reykingar ára Á hvaða aldri byrjaðir þú að reykja? Í hversu mörg ár reyktir þú/hefur þú reykt? ár Hvenær hættir þú að reykja (ef á við)? ára Tímabil þar sem ekki var reykt inn á milli? Ef já, á hvaða aldri reyktir þú ekki? frá til ára frá til ára frá til ára Hversu margar sígarettur/vindla/vindlinga/pípur reyktir þú jafnaði á dag? Sígarettur stk Vindlar stk Vindlingar stk Pípur stk xlv Áfengisnotkun Hversu gamall varst þú þegar þú byrjaðir að neyta áfengra drykkja? (vinsamlegast merkið aðeins við einn valmöguleika) Aldrei drukkið Undir 14 ára 14-16 ára 17-18 ára 19-20 ára 21-25 ára Eldri en 25 ára □ □ □ □ □ □ □ Núverandi áfengisneysla Hversu oft neytirðu áfengra drykkja? (vinsamlegast merkið aðeins við einn valmöguleika) □ □ □ □ □ □ □ Ég drekk ekki áfengi Minna en einu sinni í mánuði 1-3 sinnum í mánuði 1-2 sinnum í viku 3-4 sinnum í viku 5-6 sinnum í viku Daglega Þegar þú neytir áfengis, hversu mikið drekkur þú að jafnaði í hvert skipti? (Einn einfaldur drykkur jafngildir t.d. einu léttvínsglasi, einum litlum bjór eða einföldum sterkum drykk). □ □ □ □ □ Einn drykk 2-3 drykki 4-5 drykki 6-7 drykki 8 drykki eða meira Fyrri áfengisneysla Hversu oft neyttir þú áfengra drykkja á eftirfarandi aldri? (Merkið við eins og við á, vinsamlegast merkið aðeins við einn valmöguleika fyrir hvern aldur) Tvítugur Þrítugur Fertugur Fimmtugur Sextugur Drakk ekki áfengi Minna en einu sinni í mánuði 1-3 sinnum í mánuði 1-2 sinnum í viku 3-4 sinnum í viku 5-6 sinnum í viku Daglega xlvi Þegar þú neyttir áfengis, hversu mikið drakkst þú að jafnaði í hvert skipti? □ □ □ □ □ Einn drykk (einfaldur drykkur t.d.eitt vínglas, einn lítill bjór, eitt staup) 2-3 drykki 4-5 drykki 6-7 drykki 8 drykki eða meira Heilsufarsupplýsingar Ertu almennt heilsuhraustur? Ef nei, hvaða heilsubrestir?______________________________________ Já Nei □ □ Já Nei □ □ Já Nei □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ □ Hvenær varst þú greindur með sykursýki eða forsykursýki?______________________________________ Hefur einhver í fjölskyldu þinni verið greindur með sykursýki af gerð 2? Ef já, hver/hverjir?______________________________________ Tekur þú núna einhver lyf(þ.m.t. náttúrulyf)? Ef já, merkið við það sem við á Hormónalyf (nafn lyfs____________________________________) Bólgueyðandi lyf (nafn lyfs________________________________) Blóðþrýstingslækkandi lyf (nafn lyfs__________________________) Blóðfitulækkandi lyf (nafn lyfs______________________________) Blóðþynningarlyf (nafn lyfs________________________________) Ofnæmislyf (nafn lyfs____________________________________) Lyf við sykursýki (nafn lyfs________________________________) Sýklalyf (nafn lyfs_______________________________________) Verkjalyf (nafn lyfs______________________________________) Annað: ______________________________________________ xlvii Menntun og staða Hver er menntun þín? □ □ □ □ □ □ Grunnskólamenntun Framhaldsskólamenntun Iðnmenntun Háskólamenntun (Ba, BSc, BEd) Æðri háskólagráða (MSc, PhD) Annað: ______________________ Hver er núverandi staða þín? □ □ □ □ □ □ □ Nemi Launþegi Sjálfstætt starfandi Atvinnulaus Öryrki Ellilífeyrisþegi Annað: ______________________ Hver er hjúskaparstaða þín? □ □ □ □ □ □ Einhleypur Í sambandi Í sambúð/giftur Fráskilinn Ekkjumaður Annað: ______________________ Hverjir búa á heimilinu (auk þín)? Maki Börn yngri en 18 Börn eldri en 18 Aðrir: __________________________ Já Nei □ □ □ □ □ □ □ □ xlviii Rúgneysla Eftirfarandi spurningar eiga við upplifun þína af rúgneyslu síðastliðna þrjá mánuði Mjög sammála Sammála Hvorki sammála né ósammála ósammála Mjög ósammála Ég hef fundið fyrir óþægindum í meltingarfærum □ □ □ □ □ Mér finnst gott að borða vörur sem innihalda rúg □ □ □ □ □ Ég myndi kjósa vörur sem innihalda rúg fram yfir vörur sem innihalda ekki rúg □ □ □ □ □ Hvernig hefur upplifun þín verið af rúgneyslu síðastliðna þrjá mánuði? Hvaða rúgvörur myndir þú helst kjósa, (þ.e. rúgbrauð, rúggraut, rúghrökkbrauð, morgunkorn með rúgi) og hvaða vörur eru sístar að þínu mati. xlix