Glycaemic and Insulin Index of Four Common German Breads

ORIGINAL ARTICLE Glycaemic and insulin index of four common German breads

J Goletzke1,2, FS Atkinson2,KLEk2, K Bell2, JC Brand-Miller2,3 and AE Buyken1

BACKGROUND/OBJECTIVES: German-style breads are recommended as a lower glycaemic index (GI) alternative, yet little data is available. Our aim was to test the GI and insulin index (II) of four breads and rolls commonly consumed in Germany. SUBJECTS/METHODS: Four German bread products were tested for their GI and II in 12 healthy subjects according to the International Standard Organization guidelines. RESULTS: Only the wholemeal rye bread with visible intact grains and sunflower seeds was identified as low GI (GI = 55). Both the wholemeal spelt wheat (GI = 63) and the rye wheat sourdough bread (GI = 62) were classified as medium GI, whereas soft pretzel was high GI (GI = 80, Po0.05 compared with other products). The II of soft pretzels (II = 102) was also highest and differed significantly from wholemeal rye (II = 70) and rye sourdough bread (II = 72) but not from wholemeal spelt wheat (II = 77). CONCLUSIONS: Contrary to popular belief, these German-style breads are medium to high GI, with the exception of the rye bread with intact grains and seeds. The results highlight the need to test, rather than ‘guestimate’, the GI of local products, and develop a broader range of low-GI breads. European Journal of Clinical Nutrition (2016) 70, 808–811; doi:10.1038/ejcn.2016.9; published online 2 March 2016

INTRODUCTION that this—albeit traditional—bakery product has a high GI, because 20,21 The glycaemic index (GI) is a concept used to categorize carbohydrate- it is less acidic. rich foods according to their blood glucose-raising potential.1 Adiet characterized by a higher GI or a higher glycaemic load, the product of MATERIALS AND METHODS theGIandtherespectivecarbohydratecontent,hasbeenshownto be associated with an increased risk of type 2 diabetes mellitus2–6 and Test foods cardiovascular disease2,7–10 in several meta-analyses based on Four types of breads and rolls, representing typical breads consumed in Germany, were chosen: wholemeal rye bread with intact grains and prospective epidemiological studies. fl To achieve a low dietary GI, it is recommended to focus on sun ower seeds (Kraftkerni; Lieken Urkorn, Dortmund, Germany), organic wholemeal spelt wheat bread (Dinkelvollkornbrot, backdat; local bakery), rye main carbohydrate sources and substitute high-GI foods for 11 wheat sourdough bread (Paderborner; Lieken Urkorn) and soft pretzels alternatives within the same food category with a low(er) GI. (Brezel; Ditsch, Mainz, Germany). The nutrient content is shown in Table 1. Among populations consuming large amounts of breads such as The breads from Germany were freshly purchased, frozen and then placed in – the Nordic European countries,12 15 breads with a low GI are of foam boxes and transported from Germany to Sydney, Australia via express interest. Generally, bread is regarded a medium to high GI food.16 delivery for testing. At the Sydney University Glycaemic Index Research However, it is commonly assumed that breads made from coarsely Service (SUGiRS), the breads were kept frozen until the start of the testing ground flours and whole kernels are often low-GI breads.17,18 procedure. One day before testing, the required portion sizes (see Table 1) were weighed out and stored in the refrigerator to defrost overnight. Each Research has indicated that lowering the pH of a food, the fl bread product was tested once by each participant, whereas the glucose inclusion of intact grains and the use of rye instead of wheat our solution (reference food) was tested on three separate occasions22 in 19 helps to decrease the GI value of breads. This implies that accordance with the International Standard Organization (ISO) methodology. traditional German breads such as rye sourdough bread or The test and reference foods were provided as portions providing 50 g of wholemeal rye breads with intact grains could be lower in GI available carbohydrates. compared with breads consumed in other countries. The aim of the present study was to determine the GI values of GI testing subjects commonly consumed breads and rolls in Germany. Moreover, as The test subjects for the present study were chosen among the regular insulin excursions are closely linked to postprandial glucose participants of the Sydney University Glycaemic index Research Service responses, the insulin index (II) was also determined. Wholemeal (SUGiRS). All 12 subjects (5 Caucasian women and 7 Caucasian men) tested rye bread with intact grains and sunflower seeds and rye wheat each of the four types of breads and rolls. Their mean age was 26.3 (s.d.: 5.7) 2 sourdough bread are two breads representative of traditional years, mean BMI was 22.3 (s.d.: 2.6) kg/m and fasting plasma glucose was German breads. Furthermore, as the proportion of organic 5.22 (s.d.: 0.05) mmol/l. To be eligible for testing, subjects needed to be ‘ ’ healthy and non-smoking individuals between 18 and 65 years of age. bakeries that use more ancient grain varieties is increasing, an In addition, subjects should have no food allergy or intolerance. All organic wholemeal spelt wheat bread was selected to represent procedures involving human subjects were approved by the Human that bread type. Soft pretzels nowadays represent a common Research Ethics Committee of the University of Sydney. Written informed ‘snack-to-go’ available throughout Germany. It is hypothesized consent was obtained from all subjects.

1Department of Nutrition and Food Sciences, Nutritional Epidemiology, University of Bonn, DONALD Study at the Research Institute of Child Nutrition, Dortmund, Germany; 2School of Molecular Bioscience, The University of Sydney, Sydney, NSW, Australia and 3Boden Institute of Obesity, Nutrition and Exercise and School of Molecular Biosciences, The University of Sydney, Sydney, NSW, Australia. Correspondence: Dr AE Buyken, Department of Nutrition and Food Sciences, Nutritional Epidemiology, University of Bonn, DONALD Study at the Research Institute of Child Nutrition, Heinstück 11, 44225 Dortmund, Germany. E-mail: [email protected] Received 2 September 2015; revised 22 December 2015; accepted 13 January 2016; published online 2 March 2016 Glycaemic and insulin index of four German breads J Goletzke et al 809

Table 1. Nutrient content (per 100 g) and glycaemic/insulin index testing portion of four German-type breads 4

Bread kcal CHO Fibre Fat Protein GI/II (g) (g) (g) (g) testing 3 portion (g)

Wholemeal rye bread with intact 248 33.0 8.0 7.5 8.0 151.5 2 grains and sunflower seeds (12%) Wholemeal spelt wheat bread 188 36.7 Not 1.6 6.7 136.2 provided 1 Rye wheat sourdough bread 216 42.0 6.0 1.3 6.0 119.0 Soft pretzel 279 49.9 Not 5.2 8.1 100.2 Change in BG (mmol/l) provided 0 0 153045607590105120 Time (min) GI and II testing procedures -1 The protocol used for the present study was in accordance with the ISO standard for the determination of GI of foods.22 On the day before testing, the subjects were instructed to refrain from unusual physical activity, as well as 200 from alcohol and legume consumption, and to eat a dinner characterized by a high carbohydrate and low fat content. After a 10- to 12-h overnight fast, the subjects arrived on the morning of the test day at the metabolic kitchen. The 150 subjects warmed their hands in hot water and collected two baseline finger- prick blood samples (0.7 ml each) with 5 min between the samples. 100 Thereafter, they consumed the test or reference food with 250 ml of water at a comfortable pace within 12 min. Subsequently, finger-prick blood samples were collected 15, 30, 45, 60, 90 and 120 min from the 50 commencement of eating the test food. Throughout the test session, the subjects remained seated and were not permitted to eat or drink. Blood 0 samples were collected into anticoagulant-coated tubes (Eppendorf tubes, Change in plasma insulin (pmol/L) 0 153045607590105120 grade II; Sigma Chemical Company, St Louis, MO, USA) containing 10 IU Time (min) heparin Na salt, and were centrifuged immediately (10 000 g for 1 min at -50 room temperature). The plasma layer was pipetted into a labelled tube and Figure 1. (a) Blood glucose response curves for the four breads: − stored at 20 °C until analysis. Plasma glucose concentration was analysed in --● - - wholemeal rye bread with intact grains and sunflower seeds … … duplicate using the glucose hexokinase enzymatic essay on a centrifugal (n = 12); .▲ wholemeal spelt wheat bread (n = 12); —◊— rye analyser (model HITACHI 912; Hitachi, Chiyoda, Japan). The mean within- and wheat sourdough bread (n = 11), —■— soft pretzel (n = 11); --■-- between-assay precisions (CV) were both o6%. Plasma insulin concentration glucose reference. Values are the mean change in blood glucose was determined using a solid-phase antibody-coated tube radioimmunoassay levels with their standard errors represented by vertical bars. kit (Coat-A-Count Insulin RIA Kit; Diagnostic Products Corporation, Los (b) Insulin response curves for the four breads: - - ● - - wholemeal … … Angeles, CA, USA) with internal controls. rye bread with intact grains and sunflower seeds (n = 10); .▲ . wholemeal spelt wheat bread (n = 12); —◊— rye wheat sourdough GI and II calculations bread (n = 11), —■— soft pretzel (n = 12); --■-- glucose reference. A 2-h plasma glucose response curve was constructed for each subject’stest Values are the mean change in insulin levels with their standard session using the average plasma glucose concentration for each of their errors represented by vertical bars. eight blood samples. A 2-h plasma insulin response curve was also constructed for each of the subjects’ test sessions. To provide one baseline wholemeal rye bread with intact grains and sunflower seeds, value, the two fasting blood samples were averaged. The incremental area wholemeal spelt wheat bread, rye wheat sourdough bread and soft under each 2-h plasma and insulin curve (iAUC) was calculated using the pretzels were 108 ± 13, 128 ± 17, 117 ± 12 and 169 ± 27 mmol/l/min, trapezoidal method.22 The GI of each bread is defined as the iAUC of the respectively. The glucose iAUC of soft pretzels was significantly blood glucose response curve of a 50-g glycaemic carbohydrate portion higher compared with the other three breads (Po0.05). Similarly, the expressed as the percentage of the response to the same amount of GI of soft pretzels was significantly higher compared with the other carbohydrates from the reference food, that is, glucose (GI = 100). Similarly, fi three types of bread (Po0.05), but all four products were lower the II is de ned as the iAUC of the plasma insulin response curve of a 50-g o available carbohydrate portion in relation to the iAUC of the insulin response compared with the glucose solution (P 0.05, Table 2). According to curve of the reference food glucose. Values exceeding the mean by more the ISO-recommended classification, only the wholemeal rye bread than 2 s.d. were considered outliers and excluded from GI and II calculations with intact grains and sunflower seeds fell into the low-GI category, according to the ISO guidelines.22 whereas wholemeal spelt wheat and rye wheat sourdough bread were considered medium GI and soft pretzels high GI (Table 2). Statistical analysis Statistical analyses were performed using the SAS software (version 9.1.3; Postprandial insulinaemic excursions SAS, SAS Institute Inc., Cary, NC, USA). Inter- and intraindividual variations of the three standard glucose tests were assessed by determining the CV, Insulin concentrations increased to the maximum value at 30 min, which was 13.5%. One-way analysis of variance followed by Duncan’s followed by a steady decrease until 120 min for wholemeal multiple range test (Po0.05) was used to determine whether there was a rye bread with intact grains and sunflower seeds and soft pretzels. significant difference among the glucose and insulin iAUC responses, as For wholemeal spelt wheat bread and rye wheat sourdough well as the GI and II values of the four breads/bakery products. bread, a plateau of insulin concentrations was observed at 30 and 45 min, which was followed by a decline until 120 min (Figure 2). RESULTS Insulin iAUC values for the four types of bread (wholemeal rye Postprandial glycaemic excursions bread with intact grains and sunflower seeds, wholemeal spelt For all four types of bread, blood glucose concentrations peaked at wheat bread, rye wheat sourdough bread and soft pretzels) were 30 min and then decreased until 120 min (Figure 1). Glucose iAUC for 6730 ± 1360, 8275 ± 2305, 6360 ± 940 and 9220 ± 1245 pmol/l/min,

Table 2. GI, GI classiﬁcation and II of four German-type breads

GIa IIb

Mean s.e.m. Classiﬁcation Mean s.e.m.

Wholemeal rye bread with intact grains and sunﬂower seeds 55c 4 Low 70c 10 Wholemeal spelt wheat bread 63c 4 Medium 77c,d 10 Rye wheat sourdough bread 62c 4 Medium 72c 9 Soft pretzel 80d 6 High 102d 10

Abbreviations: GI, glycaemic index; II, insulin index. aFor both rye wheat sourdough bread and soft pretzel results, one outlier was excluded from the data set and the calculations for these breads are based on 11 participants. bTwo results were outliers for sunﬂower seeds and one result for rye wheat sourdough bread was an outlier that exceeded the mean by more than 2 s.d. and were hence considered outliers. c,dMean values within a column with unlike superscript letters are signiﬁcantly different (Po0.05).

respectively; insulin iAUC did not differ between the four breads undesirable for the increasing number of persons with insulin (P40.05). The II values of wholemeal rye bread with intact grains resistance and type 2 diabetes.28 and sunflower seeds, wholemeal spelt wheat bread and rye wheat To date, few studies have investigated the insulinaemic responses sourdough bread were significantly lower compared with the to common bread and bakery products. With the exception of glucose reference (Po0.05). Soft pretzels had a significantly higher II the soft pretzels, the products produced similar II values despite value compared with wholemeal rye bread with intact grains and differences in the GI. This may be because there is more within- and sunflower seeds and rye wheat sourdough bread (Po0.05) (Table 2). between-person variation in postprandial insulinaemia compared with glycaemia.29 The rising awareness of the GI and its inclusion in dietary advice DISCUSSION has led to efforts to decrease the GI of staple foods, particularly The present study generated new GI values for four German-type cereal products.12,21,27,30,31 Different studies show that focusing on breads, and it contributes to our understanding of how modern foods bread is a promising approach. A study in European outpatients affect postprandial glycaemia and insulinaemia. Contrary to common with type 1 diabetes showed that bread and pasta make the belief, the majority of breads tested were found to be medium or largest contribution to the overall dietary GI.32 Furthermore, high GI, with the exception of wholemeal rye bread with intact grains a small study showed that the simple substitution of a high-GI and seeds. The latter product can be regarded as a typical German- bread for a low-GI bread (while the rest of the diet was identical) style bread that is universally available in the German market but its resulted in improvements in the 24-h glucose profiles.33 GIwasunknown.Ithas,however,beenshownthatwholegrain In the present study, only one of the four ‘traditional’ German breads containing a large proportion of intact kernels elicit lower breads and rolls was found to have a low GI. This type of bread glucose profiles compared with wholemeal breads made from milled is a realistic alternative for the German, as well as the Nordic, flour, because they are digested more slowly.23,24 Different pumper- population in Europe, but not for those in Southern Europe, the nickel breads, which are comparable to wholemeal rye bread with United States or Australia. In these countries, light white sandwich intact grains and sunflower seeds, have been tested to be low-GI breads, rather than dense breads with a chewy texture, are (GI = 53, mean of 7 studies) in patients with type 2 diabetes.16 the norm.14 Indeed, many test subjects in this study did not like The findings for wholemeal spelt wheat and rye wheat the breads and reported difficulties consuming the test portion sourdough bread are in accordance with results from other within the allotted 12 min. Another approach to lower the GI of studies. It has been shown that the glycaemic response of white other commonly consumed breads might be recipe modifications, and wholemeal bread is similar and both types of breads are such as using legume flours.31 In a country like Germany, where considered high GI.16 The fact that the current spelt wheat bread bread resembles the most important source of overall carbohy- consisted of 40% ‘bruised’ spelt grains may have contributed to drate and fibre intake,13 focusing on both strategies might be of the medium GI.23,24 Furthermore, the use of genuine sourdough special relevance. fermentation has been shown to decrease the postprandial Our study has particular strengths. It was conducted according to glycaemic and insulinaemic responses.20,25 During fermentation, ISO guidelines, and the breads were purchased and sent to Sydney organic acids such as lactic acid are generated by bacteria, which directly from Germany. This ensures that the breads tested were lower the overall pH and inhibit gastric emptying.26 Rye products made with traditional German baking methods with no modifica- have been shown to be associated with improved glycaemic tions. The fact that only four types of breads and rolls were tested in profiles and lower acute insulinaemic responses.27 the present project can be regarded as a limitation, as numerous Finally, soft pretzels were tested as high GI. Even though pretzels other varieties exist on the German market. However, in light of GI are rolls rather than breads per se, they were tested because research, the present results underline the importance of further they are a traditional part of the southern German diet and are examining local products. The lack of data on fibre content for two of increasing popularity as a snack food throughout Germany. breads is a further limitation—although it has been shown in several We speculate that dipping the raw pretzels into sodium hydroxide studies that GI values of foods do show low correlations with the solution before baking could be responsible for a higher GI fibre content of the food.34–36 Also, other characteristics of the test compared with white wheat bread (GI value: 75; Atkinson et al.16). foods such as physicochemical properties were not available for This would raise the pH20,21 and encourage faster gastric emptying. interpretation of the results. Alternately, the nature of the flours used for pretzels may be more Taken together, this study revealed that, contrary to popular similar to cake flours with less protein and added gluten and finer belief, common German-type breads are medium or high GI, with particle size. Our results highlight the trend to consuming modern the exception of wholemeal rye bread with intact grains and seeds. starchy foods that, even with a high fibre content, are associated Hence, a broader range of low-GI breads needs to be developed, with high glycaemic and insulinaemic excursions. This may be taking into account local taste preferences and nutritional habits.

European Journal of Clinical Nutrition (2016) 808 – 811 © 2016 Macmillan Publishers Limited Glycaemic and insulin index of four German breads J Goletzke et al 811 The present results highlight the need to test the GI of local 14 Roswall N, Olsen A, Boll K, Christensen J, Halkjaer J, Sorensen TI et al. Con- products under standardized conditions. sumption of predefined 'Nordic' dietary items in ten European countries—an investigation in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Public Health Nutr 2014; 17: 2650–2659. CONFLICT OF INTEREST 15 Wirfalt E, McTaggart A, Pala V, Gullberg B, Frasca G, Panico S et al. Food sources of 5 – JBM is a coauthor of The Low GI Handbook book and other books (DeCapo, New carbohydrates in a European cohort of adults. Public Health Nutr 2002; :1197 1215. York), President of the Glycaemic Index Foundation and Director of a non-profitGI- 16 Atkinson FS, Foster-Powell K, Brand-Miller JC. International tables of glycaemic based food endorsement programme in Australia and manages the University of index and glycaemic load values. Diabetes Care 2008; 31: 2281–2283. Sydney GI testing service. The remaining authors declare no conflict of interest. 17 Granfeldt Y, Hagander B, Bjorck I. Metabolic responses to starch in oat and wheat products. On the importance of food structure, incomplete gelatinization or presence of viscous dietary fibre. Eur J Clin Nutr 1995; 49:189–199. ACKNOWLEDGEMENTS 18 Liljeberg H, Granfeldt Y, Bjorck I. Metabolic responses to starch in bread con- fl 46 – We thank Lieken Urkorn for providing two of the breads and sponsoring their taining intact kernels versus milled our. Eur J Clin Nutr 1992; :561 575. fi transport to Sydney, Australia. Furthermore, we thank the SUGiRS staff, 19 Scazzina F, Siebenhandl-Ehn S, Pellegrini N. The effect of dietary bre on reducing 109 – particularly Karola Landrey, Courtney J Wright and Alisha Li, for helping with the glycaemic index of bread. Br J Nutr 2013; : 1163 1174. the organization of the GI/II testing in the metabolic kitchen, with the processing 20 De Angelis M, Rizzello CG, Alfonsi G, Arnault P, Cappelle S, Di Cagno R et al. Use of of the blood samples and with the conduction of the glucose and insulin sourdough lactobacilli and oat fibre to decrease the glycaemic index of white analyses. wheat bread. Br J Nutr 2007; 98: 1196–1205. 21 Novotni D, Curic D, Bituh M, Colic Baric I, Skevin D, Cukelj N. Glycaemic index and phenolics of partially-baked frozen bread with sourdough. Int J Food Sci Nutr AUTHOR CONTRIBUTIONS 2011; 62:26–33. — AEB, JBM and JG conceived the project; AEB and JG selected the breads; FA and KLE 22 International Standards Organisation. Food Products Determination of the Gly- fi supervised the testing procedures and data interpretation; KB assisted in the II result caemic Index (GI) and Recommendation for Food Classi cation, ISO 26642-2010. interpretation; JG conducted the GI/II testing, undertook statistical analyses, International Organisation for Standardisation: Geneva, Switzerland, 2010. interpreted the findings and wrote the manuscript; and all authors read and 23 Fardet A. New hypotheses for the health-protective mechanisms of whole-grain approved the final manuscript. cereals: what is beyond fibre? Nutr Res Rev 2010; 23:65–134. 24 Jenkins DJ, Wesson V, Wolever TM, Jenkins AL, Kalmusky J, Guidici S et al. Wholemeal versus wholegrain breads: proportion of whole or cracked grain and REFERENCES the glycaemic response. BMJ 1988; 297: 958–960. 1 Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM et al. 25 Najjar AM, Parsons PM, Duncan AM, Robinson LE, Yada RY, Graham TE. The acute Glycaemic index of foods: a physiological basis for carbohydrate exchange. Am J impact of ingestion of breads of varying composition on blood glucose, insulin fi 101 – Clin Nutr 1981; 34:362–366. and incretins following rst and second meals. Br J Nutr 2009; :391 398. 2 Barclay AW, Petocz P, McMillan-Price J, Flood VM, Prvan T, Mitchell P et al. 26 Liljeberg HG, Lonner CH, Bjorck IM. Sourdough fermentation or addition of Glycaemic index, glycaemic load, and chronic disease risk—a meta-analysis of organic acids or corresponding salts to bread improves nutritional properties of observational studies. Am J Clin Nutr 2008; 87:627–637. starch in healthy humans. JNutr1995; 125: 1503–1511. 3 Dong JY, Zhang L, Zhang YH, Qin LQ. Dietary glycaemic index and glycaemic load 27 Rosen LA, Silva LO, Andersson UK, Holm C, Ostman EM, Bjorck IM. Endosperm and in relation to the risk of type 2 diabetes: a meta-analysis of prospective cohort whole grain rye breads are characterized by low post-prandial insulin response studies. Br J Nutr 2011; 106: 1649–1654. and a beneficial blood glucose profile. Nutr J 2009; 8:42. 4 Greenwood DC, Threapleton DE, Evans CE, Cleghorn CL, Nykjaer C, Woodhead C 28 International Diabetes Federation. IDF Diabetes Atlas6th edn (International Dia- et al. Glycaemic index, glycaemic load, carbohydrates, and type 2 diabetes: sys- betes Federation: Brussels, Belgium, 2013. tematic review and dose–response meta-analysis of prospective studies. Diabetes 29 Moazzami AA, Shrestha A, Morrison DA, Poutanen K, Mykkanen H. Metabolomics Care 2013; 36: 4166–4171. reveals differences in postprandial responses to breads and fasting metabolic char- 5 Liu S, Chou EL. Dietary glycaemic load and type 2 diabetes: modeling the glucose- acteristics associated with postprandial insulin demand in postmenopausal women. J raising potential of carbohydrates for prevention. Am J Clin Nutr 2010; 92:675–677. Nutr 2014; 144: 807–814. – 6 Livesey G, Taylor R, Livesey H, Liu S. Is there a dose response relation of dietary 30 Ferrer-Mairal A, Penalva-Lapuente C, Iglesia I, Urtasun L, De Miguel-Etayo P, Remon S glycaemic load to risk of type 2 diabetes? Meta-analysis of prospective cohort et al. In vitro and in vivo assessment of the glycaemic index of bakery products: 97 – studies. Am J Clin Nutr 2013; : 584 596. influence of the reformulation of ingredients. Eur J Nutr 2012; 51:947–954. 7 Fan J, Song Y, Wang Y, Hui R, Zhang W. Dietary glycaemic index, glycaemic load, 31 Radhika G, Sumathi C, Ganesan A, Sudha V, Jeya Kumar Henry C, Mohan V. and risk of coronary heart disease, stroke, and stroke mortality: a systematic Glycaemic index of Indian flatbreads (rotis) prepared using whole wheat flour and review with meta-analysis. PLoS One 2012; 7: e52182. 'atta mix'-added whole wheat flour. Br J Nutr 2010; 103: 1642–1647. 8 Ma XY, Liu JP, Song ZY. Glycaemic load, glycaemic index and risk of cardiovascular 32 Buyken AE, Toeller M, Heitkamp G, Karamanos B, Rottiers R, Muggeo M et al. diseases: meta-analyses of prospective studies. Atherosclerosis 2012; 223:491–496. Glycaemic index in the diet of European outpatients with type 1 diabetes: rela- 9 Mente A, de Koning L, Shannon HS, Anand SS. A systematic review of the evi- tions to glycated hemoglobin and serum lipids. Am J Clin Nutr 2001; 73:574–581. dence supporting a causal link between dietary factors and coronary heart disease. Arch Intern Med 2009; 169:659–669. 33 Henry CJ, Lightowler HJ, Tydeman EA, Skeath R. Use of low-glycaemic index bread 10 Mirrahimi A, de Souza RJ, Chiavaroli L, Sievenpiper JL, Beyene J, Hanley AJ et al. Asso- to reduce 24-h blood glucose: implications for dietary advice to non-diabetic and 57 – ciations of glycaemic index and load with coronary heart disease events: a systematic diabetic subjects. Int J Food Sci Nutr 2006; :273 278. review and meta-analysis of prospective cohorts. JAmHeartAssoc2012; 1: e000752. 34 Salmeron J, Ascherio A, Rimm EB, Colditz GA, Spiegelman D, Jenkins DJ et al. Dietary fi 20 – 11 Brand-Miller J, McMillan-Price J, Steinbeck K, Caterson I. Carbohydrates—the ber, glycaemic load, and risk of NIDDM in men. Diabetes Care 1997; :545 550. good, the bad and the whole grain. Asia Pac J Clin Nutr 2008; 17 (Suppl 1): 16–19. 35 Stevens J, Ahn K, Juhaeri, Houston D, Steffan L, Couper D. Dietary fiber intake and 12 Burton PM, Monro JA, Alvarez L, Gallagher E. Glycaemic impact and health: new glycaemic index and incidence of diabetes in African-American and white adults: horizons in white bread formulations. Crit Rev Food Sci Nutr 2011; 51: 965–982. the ARIC study. Diabetes Care 2002; 25: 1715–1721. 13 Max Rubner-Institut. Nationale Verzehrsstudie II. Ergebnissbericht Teil 1. Available at: 36 van Bakel MM, Kaaks R, Feskens EJ, Rohrmann S, Welch AA, Pala V et al. Dietary www.was-esse-ich.de/uploads/media/NVSII_Abschlussbericht_Teil_1.pdf (last accessed glycaemic index and glycaemic load in the European Prospective Investigation 15 January 2014). into Cancer and Nutrition. Eur J Clin Nutr 2009; 63 (Suppl 4): S188–S205.