The glycemic index applications in practice

The 2007scientific update of the Food and Agriculture Organization/World Health OrganizationFAO/WHO on carbohydrates in human nutrition endorsed the primary classification,recommended by

GLYCEMIC INDEX

Applications in Practice

Except as permitted under U.S Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com

( http://www.copyright.com/ ) or contact the Copyright Clearance Center, Inc (CCC), 222 Rosewood Drive, Danvers,

MA 01923, 978-750-8400 CCC is a not-for-profit organization that provides licenses and registration for a variety of users For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for

Sports Performance

Lars McNaughton, David Bentley, and S Andy Sparks

Index

By 1981, the United States was in the grip of the low-fat diet craze Based upon tenuousevidence, nutrition authorities began to recommend that everyone consume as little fat aspossible to avoid obesity, diabetes, heart disease, and possibly cancer Instead, the publicwas advised to base their diet on carbohydrates Natural high-fat foods such as nuts,avocado, and whole milk yoghurt acquired a bad reputation, whereas highly processedcarbohydrates inundated the food supply Amazingly, these products—including preparedbreakfast, crackers, baked chips, breads, reduced fat cookies and cakes, and sugarybeverages—were marketed as healthful or at least innocuous, even though they werecomposed primarily of refined grains and concentrated sugar Very soon, this low-fatmessage spread throughout the world

The year 1981 also witnessed the introduction of the glycemic index (GI) by DavidJenkins, Thomas Wolever, and colleagues at the University of Toronto At that time, theconcept of the GI represented a radical departure from conventional thinking, by proposingthat the health effects of carbohydrates differ according to how they affect blood glucose inthe postprandial state On account of the brain’s critical dependence on this metabolic fuelunder most conditions, the concentration of glucose in the blood is ordinarily tightlycontrolled However, most highly processed carbohydrates digest rapidly, raising bloodglucose and insulin levels much more than traditionally consumed carbohydrates such aslegumes, fruits, and minimally processed grains Early investigators in the field recognizedthat a high-GI diet stressed the body’s energy homeostasis mechanisms, with majorimplications not only to diabetes management but also to the prevention of type 2 diabetes,heart disease, obesity, and other modern chronic degenerative conditions Indeed, thescience surrounding the GI helped explain why the conventional low-fat diet loaded withprocessed carbohydrates had actually contributed to many of the diseases it was intended toprevent

Fortunately, the concept of the GI has also spread around the globe and is poised tooutlast and supersede the low-fat diet craze A new scientific study on the topic is nowbeing published at a rate of almost one a day, providing a wealth of new information abouthow diet affects hormones, metabolism, and health All fats are not the same, and neitherare carbohydrates Thus, the GI leads us away from simplistic debates about nutrient

“quantity,” to a critically important focus on food “quality.”

Almost from its inception, the GI elicited controversy, perhaps precisely because itchallenged an entrenched paradigm that implicitly considers all carbohydrates alike Somecritics dismissed the GI, arguing that ostensibly unhealthful foods such as ice cream ratelow on this scale But such arguments miss the point: No one dietary factor can ever define

a healthful diet Others point to the existence of negative studies, neglecting the inherentcomplexity and heterogeneity of nutritional research and the large body of mechanistic,

Boston Children’s Hospital Harvard Medical School Harvard School of Public Health

It is already been 36 years since 1981 when David Jenkins, Thomas Wolever, andcolleagues introduced the concept of glycemic index (GI) to differentiate carbohydratesbased on the rate of blood glucose rise following their consumption Although GI was firstused in the diet therapy of diabetes, since then, research evidence has accumulated tothousands of publications from all over the world with applications for prevention and/ormanagement of metabolic syndrome, cardiovascular disease, obesity, polycystic ovarysyndrome, certain types of cancer, effects on pregnancy outcomes, sports performance, eyehealth, and cognitive functioning

As eloquently put by Professor David S Ludwig in his Foreword to this book, the GIconcept has faced much controversy and criticism arising mainly from misconceptions onits use and application; nevertheless, it has led the way into understanding the importance

of macronutrient quality rather than just quantity on metabolic pathways and diet–disease

relationships

The Glycemic Index: Applications in Practice has gathered in a systematic way all the

up-to-date research in the field of GI It also provides a detailed explanation of how tocorrectly measure a food’s GI, how the GI of food products can be altered, and the use andmisuse of GI labeling around the globe Additionally, it provides practicalrecommendations on how the GI concept can be applied in the dietary management ofcertain disease conditions It is a valuable source of information for healthcareprofessionals of various disciplines, such as nutritionists, dietitians, food scientists, medicaldoctors, sports scientists, psychologists, public health (nutrition) policy makers, andstudents in these fields, as well as an important addition to university libraries for referencepurposes

This book is a result of the combined effort of many experts, including pioneers in thearea of GI research, and I wish to express my sincere gratitude to each one of them formaking it such a valuable addition to the literature I also thank CRC Press and especially

Dr Ira Wolinsky, who invited me to edit this book, Randy Brehm, senior editor of thenutrition program, and Kathryn Everett, production coordinator

I hope you find this book stimulating and useful in your studies and practice

Elena Philippou

University of Nicosia

Dr Elena Philippou is an Assistant Professor in Nutrition and Dietetics at the University

of Nicosia, Cyprus, and a Visiting Lecturer in Nutrition and Dietetics at King’s CollegeLondon, United Kingdom As a registered dietitian, she also holds private consultations ondiet-related issues, including obesity, cardiovascular disease, and diabetes

She obtained a BSc degree in Nutrition and a postgraduate diploma in Dietetics fromKing’s College London, London, England, in 2001 and 2002, respectively She worked as adietitian for the National Health Service in the United Kingdom and in parallel completed apostgraduate certificate in behavioral management of adult obesity awarded by theUniversity of Central Lancashire, Preston, England In 2008, she completed her PhD studies

at Imperial College London, focusing on the role of dietary carbohydrates and specificallydietary GI in weight maintenance and cardiovascular disease prevention Her research hasbeen published in international peer-reviewed scientific journals and presented in scientificconferences

In 2012, Dr Philippou obtained a postgraduate certificate in continuing professionalacademic development program in learning and teaching in higher education awarded bythe University of Hertfordshire, Hatfield, England, and became a member of U.K.’s HigherEducation Academy She lectures on various topics, including public health nutrition,nutritional assessment, and medical nutrition therapy of various diseases

Dr Philippou’s current research interest is in the role of dietary GI manipulation and theMediterranean diet on cognitive function including investigation of the potential underlyingmechanisms

Faculty of Health Sciences

Department of Clinical Epidemiology and BiostatisticsMcMaster University

School of Medicine

Department of Medicine, Faculty of Health SciencesQueen’s University

Kingston, Ontario, Canada

and

Clinical Nutrition and Risk Factor Modification Centre

as in the prevention and management of diseases such as cardiovascular disease (CVD),diabetes and cancer of the large bowel (Mann et al 2007).

This introductory chapter on carbohydrates and the glycemic index (GI) will provide anoverview of dietary carbohydrates, including their classification, dietary intakerecommendations, roles in the diet and the risks associated with the intake of simple sugars.The concepts of GI and glycemic load (GL) will be introduced and factors affecting the GI

and Stephen 2007) However, not all carbohydrates fit into this scheme, an example beinginulin from plants, which may have between 2 and 200 fructose units and thus crosses theboundary between oligosaccharides and polysaccharides (Roberfroid et al 1993) The 2007scientific update of the Food and Agriculture Organization/World Health Organization(FAO/WHO) on carbohydrates in human nutrition endorsed the primary classification,recommended by the 1997 expert consultation based on chemical form, as explained above,but acknowledging that this classification should also have dimensions of physical effects,functional and/or physiologic effects, and health outcomes (Cummings and Stephen 2007).The explanation of carbohydrate terminology and classification that follows is based on the

2007 scientific update of FAO/WHO on carbohydrates

1.2.1 TOTAL CARBOHYDRATE

“Total carbohydrate” reported in food tables may be derived by using either the “bydifference” approach or the direct measurement of the individual components, which arethen added to give a total (Cummings and Stephen 2007) Determination is done bymeasuring all other components of a food, including moisture, protein, fat, ash, and alcohol,and then subtracting the sum of these from the total weight of the food, thus considering theremainder or “difference” to be the carbohydrate Although the calculation of carbohydrate

by difference for the determination of the nutrient content of foods is used by the U.S.Department of Agriculture (U.S Department of Agriculture 2015), it is limited by the factthat the derived figure includes noncarbohydrate components such as lignin, organic acids,tannins, waxes, and some Maillard products and obviously combines all the analyticalerrors from other analyses (Cummings and Stephen 2007) In addition, knowing only thetotal carbohydrate content of a food without breakdown into the different types does notprovide enough information on the potential health effects Alternatively, direct analysis todetermine the carbohydrate content can be used, and the United Kingdom’s McCance andWiddowson’s composition of foods expresses carbohydrate content in this approach (FoodStandard Agency and Public Health England 2014) The “available carbohydrate” obtained

by the direct method does not include the plant cell wall polysaccharide, fiber, and is notlimited by the errors that occur during analysis of other food components Perhaps moreimportantly, direct analysis of total carbohydrate and its components allows diet-diseaserisks to be explored The 2007 scientific update of FAO/WHO on carbohydratesrecommends that the direct measurement of total carbohydrate should be preferred and thatsimplified methods to do this should be developed (Cummings and Stephen 2007) Here, itshould be noted that the determination of carbohydrate by the above two methods willresult in apparently different carbohydrate content and total energy of certain foods such aspasta (Stephen 2006) Thus, the comparison of carbohydrate intake or carbohydrate content

of foods between countries should be viewed with caution, especially if the method ofdetermination differs

Glucose, galactose, and fructose Sucrose, lactose, maltose, and trehalose Sorbitol, mannitol, lactitol, xylitol, erythritol, isomalt, and maltitol

Oligosaccharides (3–9) Maltooligosaccharides (α-glucans)

Non-α-glucan oligosaccharides

Maltodextrins Raffinose, stachyose, fructo- and galactooligosaccharides, polydextrose, and inulin

Polysaccharides (≥10) Starch (α-glucans)

Nonstarch polysaccharides

Amylose, amylopectin, and modified starches Cellulose, hemicellulose, pectin, arabinoxylans, β-glucan, glucomannans, plant gums and mucilages, and hydrocolloids

in fruit and berries (Holland et al 1992) A nonnatural form of these sugars is the one used

by the food industry as corn syrup and high-fructose corn syrup (HFCS) (Cummings andStephen 2007) In addition to sweetening foods, sugars have a number of functions such asfood preservation and conferring functional characteristics to foods such as viscosity,texture, body, and browning capacity (Institute of Medicine 2006) Sugar alcohols, forexample, sorbitol, may be used to replace sugar and are both found naturally in some fruitsand made commercially (Cummings and Stephen 2007)

The main disaccharides are sucrose, made of glucose and fructose (α-Glc(1 2)β-Fru),and lactose, made of galactose and glucose (β-Gal(1 4)Glc) Sucrose is extracted fromsugar cane or beet and is found widely in fruits, berries, and vegetables, whereas lactose isthe main sugar found in milk In addition, there are other less-abundant disaccharides such

as maltose (α-Glc(1 4)α-Glc), which consists of two glucose molecules and occur insprouted wheat and barley, and trehalose (α-Glc(1 1)α-Glc), which also consists of twoglucose molecules and is found abundantly in yeast and fungi and in small amounts in breadand honey (Cummings and Stephen 2007) Sugars are categorized on food labels by using anumber of different terms, as outlined below; this mainly aims to differentiate their originand thus perceived health impact

1.2.2.1 Total Sugars

Used on labels and accepted by the European Union, Australia, and New Zealand, the term

“all sugars” includes all sugars from whatever source in a food and is defined as “all

monosaccharides and disaccharides other than polyols” (European Union 2011) The

al 1978) The same term was used to describe the carbohydrate components of ahydrolyzed food detected by chromatography or calorimetric methods (Southgate et al.1978) The term has now changed to refer to “monosaccharides and disaccharides added tofoods by the manufacturer, cook, and consumer, plus sugars naturally present in honey,syrups, and fruit juices” (WHO/FAO 2003), and thus, care needs to be taken to avoidconfusion between the two terms (Cummings and Stephen 2007)

1.2.2.3 Added Sugars

The U.S Institute of Medicine defines “added sugars” as “sugars and syrups that are added

to foods during processing or preparation” (Institute of Medicine 2006) Naturallyoccurring sugars, for example, lactose (in milk) or fructose (in fruit), are not included inthis definition Examples of added sugars include white, brown, or raw sugar; syrups such

as corn syrup, HFCS, and malt syrup; liquid fructose; honey; molasses; and dextrose(Institute of Medicine 2006) The Institute of Medicine notes that foods and beverages thatare major sources of added sugars have lower micronutrient densities compared with thosethat contain these sugars naturally However, there is no difference in the chemicalcomposition of the two (Institute of Medicine 2006) (See also Section 1.4.1 that discussesthe possible health risks posed by the consumption of free sugars.)

1.2.2.4 Extrinsic and Intrinsic Sugars

The terms “extrinsic” and “intrinsic” sugars originated from the U.K Department of HealthCommittee report in 1989 in order to “distinguish sugars naturally intergraded into thecellular structure of a food (intrinsic) from those that are free in the food or added to it(extrinsic)” (Department of Health 1989) Examples of intrinsic sugars include whole fruitsand vegetables that contain mainly fructose, glucose, and sucrose, whereas examples ofextrinsic sugars include fruit juice and sugars added to processed foods (Cummings andStephen 2007) The term “nonmilk extrinsic sugar” was also introduced to differentiate thesugar present in milk, lactose, which is nutritionally beneficial despite being extrinsic(Department of Health 1989) In practical terms, analysis of sugars in this way or the use ofthese terms on food labels is problematic, and although the terminology is used in scientificreports, it is not well understood or used by the public (Cummings and Stephen 2007)

The “Carbohydrate Terminology and Classification” paper of the 2007 updated scientificreport of FAO/WHO on carbohydrates notes that apart from the terms “total sugars” and thesubdivision into mono- and disaccharides, the use of most of the other terms, including

“refined sugars,” “natural sugars,” and “discretionary sugar,” is not really justified Inaddition, a uniform terminology is important in order to be able to make direct comparisonsbetween foods and intakes in different populations (Cummings and Stephen 2007)

1.2.3 OLIGOSACCHARIDES

Oligosaccharides are defined as “compounds in which monosaccharide units are joined byglycosidic linkages,” but their DP definition may vary from 2 to 19 monosaccharide units(Cummings and Stephen 2007) As shown in Table 1.1, food oligosaccharides can be eithermaltodextrins, which are used in the food industry as sweeteners and fat substitutes and fortexture modification, or non-α-glucan oligosaccharides, which are found in peas, beans, andlentils The latter group also includes inulin and fructooligosaccharides, which are storagecarbohydrates in artichokes and chicory and are also found in smaller amounts in wheat,rye, asparagus, onion, leek, and garlic The above-mentioned oligosaccharides are also used

by the industry and are referred to as “nondigestible oligosaccharides” because they are notsusceptible to pancreatic or brush border enzyme breakdown Some members of this groupsuch as fructans and galactans are also known for their prebiotic properties, discussed in

1.2.4 STARCH

Starch consists of only glucose molecules and is the storage carbohydrate of plants such ascereals, root vegetables, and legumes It is mainly composed of two polymers: thenonbranched helical chain of glucose linked by α-1,4 glucosidic bonds, called “amylose,”which has a DP of about 103, shown in Figure 1.1, and the highly branched polymercontaining both α-1,4 and α-1,6 bonds, called amylopectin, which has a DP of 104–105,shown in Figure 1.2 Although most starches contain 10%–30% amylose, “waxy” varieties

of starches from maize, rice, barley, and sorghum contain largely amylopectin Differentvarieties of cereals such as rice have different proportions of amylose and amylopectin(Kennedy and Burlingame 2003), which, as discussed in detail under Section 1.7.1, affecttheir GI

FIGURE 1.2 Amylopectin molecule.

Heating starch in water results in the loss of its crystalline structure, which is referred to

as gelatinization, whereas recrystallization or retrogradation results when cooked starch iscooled down, such as that found in cold potato salad (Cummings and Stephen 2007).Gelatinization occurs at higher temperatures in higher-amylose starches, which are alsomore prone to retrograde and form amylose-lipid complexes Thus, these types of starchescan be used to form foods with high-resistant starch (high-RS) content, the definition andproperties of which are explained in Section 1.3.2

Starches can also be modified chemically (modified starch) to change their properties,resulting in qualities such as gel stability; decrease in viscosity; changes in mouth feel,appearance, and texture; and resistance to heat treatment, which are important in the foodindustry (Cummings and Stephen 2007) The applications are so diverse that some of thesemodifications are classed as additives and others as ingredients (Coultate 2009) The twomost important processes to modify starch are substitution and cross-linking Substitutioninvolves esterification of a small proportion (<1%) of glucose units with organic acids orphosphates to produce “stabilized” starches (Coultate 2009) Depending on which groupsare attached, the resulting modified starch may have properties such as freeze stability in

gels, resistance to retrogradation, which is involved in bread staling, and increase inviscosity (Coultate 2009) Cross-linking is a process in which a limited number of linkagesbetween the chains of amylose and amylopectin are introduced (Cummings and Stephen2007) In fact, fewer than one cross-linkage per 1000 glucose units is enough to producesignificant changes in starch properties, which result in strengthening the starch granule(Coultate 2009) The resulting properties include resilience to low pH and extendedcooking, control of viscosity during processing, and also resistance to digestion (Coultate2009; Cummings and Stephen 2007) The use of starch modification to alter GI is discussedextensively in Chapter 13.

1.2.5 NONSTARCH POLYSACCHARIDES

Nonstarch polysaccharides are non-α-glucan polysaccharides, principally found in the plantcell wall and are defined as “macromolecules consisting of a large number ofmonosaccharide residues joined to each other by glycosidic linkages” (IUB-IUPAC JointCommission on Biochemical Nomenclature 1982) Cellulose, a straight-chain β1–4-linkedglucan (DP 103–106), comprises 10%–30% of NSP in foods and gives the plant cell wall itsstructure by its close packing to form microfibrils The hemicelluloses (DP 150–200) arehighly branched chains containing a mixture of hexose (6C) and pentose (5C) sugars,mostly comprising a backbone of xylose sugars with branches of arabinose, mannose,galactose, and glucose An example of NSP is arabinoxylans, found in cereals containinguronic acids—the carboxylated derivatives of glucose and galactose; they are able to formsalts with calcium and zinc, an important determinant of their properties (Cummings andStephen 2007) Another NSP is pectin, a 1–4β-d galacturonic acid polymer, with possibleside chains of other sugars such as rhamnose, galactose, and arabinose, known for its gel-forming properties Other NSPs include plant gums and mucilages (Cummings and Stephen2007) Plant gums, mostly highly branched, complex uronic-acid-containing polymers, aresticky exudates formed at the sites of injuries to plants The most well-known plant gum isgum arabic, used as a thickener (Cummings and Stephen 2007) Plant mucilages, the mostwell known of them being guar gum and carob gums, are mixed with the endosperm ofstorage carbohydrates of seeds and have water-retaining and desiccation-preventingproperties (Cummings and Stephen 2007) These are also used by the food andpharmaceutical industries as thickeners and stabilizers Finally, another NSP category isalgal polysaccharides, such as carageenan, which is used in dairy products and chocolatebecause of its ability to react with milk protein Other examples include agar and alginate,which are the NSPs extracted from seaweed or algae, and have gel-forming properties(Cummings and Stephen 2007)

1.3 CARBOHYDRATE TERMINOLOGY BASED ON PHYSIOLOGY

The “Carbohydrate Terminology and Classification” paper of the 2007 updated scientific

report of FAO/WHO on carbohydrates explains that the classification of carbohydratesbased on their chemistry does not allow a simple translation into nutritional benefitsbecause their physiologic effects are varied and overlapping (Cummings and Stephen2007) It is thus preferable to classify them based on their physiologic properties, whichfocuses more on the potential health benefits of carbohydrates An example of physiologicgrouping is that based on the effect of carbohydrates on stool weight, where somecarbohydrates such as polyols (except erythritol), some starches, NSP, lactose (in lactose-intolerant populations), and fructose (in large amounts) increase stool weight, whereasothers such as glucose, galactose, sucrose, maltose, trehalose, maltodextrins,oligosaccharides, and most starches have no effect on stool weight (Cummings and Stephen2007) However, it should be noted that the physiology of carbohydrates can vary amongindividuals and populations, with examples including lactose, which is poorly hydrolyzed

by most adults, except Caucasians, and polyols and starch, whose digestion and absorptionare variable (Cummings and Stephen 2007) Table 1.2 shows the preferred terminology ofdietary carbohydrates suggested by the 2007 scientific update of FAO/WHO on dietarycarbohydrates and also lists the less useful terms In the following sections, the physiologic

or botanical terminology will be discussed

1.3.1 PREBIOTICS

The first definition of prebiotics, just more than 20 years ago, was “nondigestible foodingredients that beneficially affect the host by selectively stimulating the growth and/oractivity of one or a limited number of bacteria in the colon, thus improving host health”(Gibson and Roberfroid 1995) This was later refined to include other areas that may benefitfrom selective targeting of particular microorganisms to: “a selectively fermentedingredient that allows specific changes, both in the composition and/or activity in thegastrointestinal microbiota that confers benefits” (Gibson et al 2004) For an ingredient to

be characterized as prebiotic, it has to abide to the following criteria: (1) resist gastricacidity, hydrolysis by mammalian enzymes, and absorption in the upper GI tract; (2) befermented by the intestinal microbiota; and (3) selectively stimulate the growth and/oractivity of intestinal bacteria potentially associated with health and well-being (Gibson andRoberfroid 1995; Gibson et al 2004;) The latter of these criteria is what separatesprebiotics from traditional fibers (Brownawell et al 2012) The benefits of prebiotics aremany and diverse; however, their discussion is outside the scope of this chapter

Prebiotic Resistant starch Dietary fiber a Glycemic

Polysaccharides Starch Nonstarch polysaccharides Total carbohydrate Less useful Sugars

Sugar Free sugars Refined sugars Added sugars Extrinsic and intrinsic sugars

Nondigestible oligosaccharides Soluble and insoluble fiber Available and unavailable carbohydrate Complex carbohydrate

to which starch is broken down (Cummings and Stephen 2007), leading to a suggestedclassification of RS as follows: physically enclosed starch, for example, within intact cellstructures (RS1), in foods such as bulgur wheat, legumes, and pumpernickel bread (wholegrains); (RS2), which is the starch resistant to amylolytic digestion because of its compactunbranched nature such as amylose; retrograded amylose (RS3) formed by the cooling ofgelatinized high-amylose starch; and modified starches (RS4) (Englyst et al 1992; Englystand Cummings 1990)

The fact that the rate and extent of starch digestion varies formed the basis of GI, and the

2007 scientific update of FAO/WHO on carbohydrates reports this as “one of the mostimportant developments in the understanding of carbohydrates in the past 30 years”(Cummings and Stephen 2007)

Trowell defined dietary fiber as “the cellular walls of plants that are resistant tohydrolysis by the enzymes of man” (Trowell 2006) This term, however, does not referprecisely to a chemical component of the diet, and the nondigestibility of plant cell wallsvaries from person to person and is affected by food storage, cooking, chewing, ripeness,and the presence of other foods (Cummings and Stephen 2007) Cummings and Stephen(2007) explain that apart from the plant cell wall, it includes many dietary components such

as lactose in some populations, some polyols, and some RS, and there is no enforceablemethod that can be used to measure this physiologic fraction of the diet Based on this and

on the fact that dietary fiber has been linked to many health benefits, the FAO/WHO in itsscientific update meeting on carbohydrates in human nutrition (July 2006) agreed that thedefinition of dietary fiber should be more clearly linked to health Thus, the followingdefinition was proposed: “dietary fiber consists of intrinsic plant cell wall polysaccharides”(Cummings and Stephen 2007) As described previously, NSPs comprise a mixture of manymolecular forms, of which cellulose is the most widely distributed The plant cell wallpolysaccharides can be determined using the enzymatic-chemical method This method isdesigned to remove all starch, and thus, it measures NSP as the sum of chemically

identified NSP constituent sugars (Englyst et al 1994) The advantage of measuring NSP isthat it is not in itself created or destroyed by normal food preparation or storage techniques,and thus, it is a consistent indicator of plant cell wall material Any added preparations ofNSP will also be measured (Englyst et al 2007) Dietary fiber may also be measured usingthe enzymatic-gravimetric method, which is based on the “indigestibility” approach (AOAC2007; Englyst et al 2007) This aims to measure the sum of indigestible polysaccharidesand lignin, and in practice, it includes RS, the amount of which may be affected by foodprocessing and the addition of RS preparations; noncarbohydrate materials such as lignin;and food-processing artifacts such as Maillard reaction products (Englyst et al 2007; Tuohy

et al 2006)

1.3.2.1 Soluble and Insoluble Fiber

The terms “soluble fiber” and “insoluble fiber” are based on the fractional extraction ofNSP, which can be controlled under laboratory conditions by changing the pH of solutions(Joint FAO/WHO Expert Consultation 1998) In the initial understanding of the properties

of dietary fiber, these terms proved very useful, allowing a simple division of NSPs intothose that were soluble and had effects on glucose and lipid absorption in the smallintestine (Lairon 1994) and those that were insoluble and thus fermented more slowly andincompletely and had more pronounced effects on bowel habit (Cummings 1997) However,their physiologic differences are not always so distinct; for example, much insoluble fiber

is completely fermented and not all soluble fiber has an effect on glucose and lipidabsorption Moreover, their separation is dependent on the conditions of extraction (Asp et

al 1992; Cummings and Stephen 2007) It should be noted that a lot of the earlier work wasdone on isolated gums or extracts of cell walls; however, fiber exists together mostly inintact plant cell walls (Cummings and Stephen 2007)

1.3.3 AVAILABLE AND UNAVAILABLE CARBOHYDRATES

McCance and Lawrence in 1929 introduced the terms “available” and “unavailable”carbohydrates in their attempt to prepare food tables for diabetic diets and the realizationthat not all carbohydrates could be “utilized and metabolized,” to the same extend Thisdefinition referred to available carbohydrate as “starch and soluble sugars” and unavailablecarbohydrate as “mainly hemicelluloses and fiber (cellulose)” (McCance and Laurence1929) The concept drew attention to the fact that some carbohydrates are not digested andabsorbed in the small intestine but reach the large bowel, where they are fermented andexcreted in feces (Cummings and Stephen 2007) A later definition of availablecarbohydrate given by an FAO technical workshop was: “that fraction of carbohydrate thatcan be digested by human enzymes, is absorbed and enters into intermediary metabolism”(FAO 2003)

The “Carbohydrate Terminology and Classification” paper of the 2007 updated scientificreport of FAO/WHO on carbohydrates argues that the use of the term “unavailable” for

carbohydrates is misleading because carbohydrates that reach the colon would still provideenergy through fermentation and absorption of short-chain fatty acids (Cummings andStephen 2007) Alternatively, the terms “glycemic,” which means providing carbohydratefor metabolism, and “non-glycemic” were recommended and referred to as more preciseand measurable fractions (Cummings and Stephen 2007) These terms are discussed inmore detail below.

1.3.3.1 Glycemic Carbohydrate

Depending on their gastrointestinal handling, carbohydrates can be referred to as “glycemiccarbohydrates,” which refer to the carbohydrates that are digested and absorbed in the smallintestine and cause a rise in blood glucose (e.g., free sugars, maltodextrins, and starch [even

if it is slowly digested]), and “non-glycemic,” which refer to the carbohydrates (or theircomponents) that are not absorbed in the small intestine and move down to becomefermented in the colon, with the production of short-chain fatty acids, methane, andhydrogen gas Examples of the latter are RS, NSPs, and sugar alcohols (Englyst and Englyst2005) Most unprocessed foods containing carbohydrates include both glycemic andnonglycemic types From the categorization of glycemic carbohydrates stems the term

“glycemic index,” which refers to the extent to which carbohydrate in foods raises the bloodglucose concentration compared with an equivalent amount of reference carbohydrate(glucose or white bread) (Jenkins et al 1981)

1.3.4 COMPLEX CARBOHYDRATES

The term “complex carbohydrates” is mostly used in the United States and was firstintroduced in 1977 in the McGovern report “Dietary Goals for the United States,” in which

it was denoted to include “fruit, vegetables, and whole grains.” Although the idea behind itsuse was to encourage the consumption of healthy foods, it is limited by the fact that fruitsand vegetables are low in starch The “Carbohydrate Terminology and Classification” paper

of the 2007 updated scientific report of FAO/WHO on carbohydrates recommendsdiscussing carbohydrates by using their common chemical names rather than by using thisterm (Cummings and Stephen 2007)

1.3.5 WHOLE GRAIN

Intake of whole grains has been associated with lower risks of CVD, type 2 diabetes, andweight gain (Ferruzzi et al 2014) and is embodied in many dietary recommendations,including the WHO/FAO report on “Diet, Nutrition, and the Prevention of ChronicDiseases” (WHO/FAO 2003) However, the definition of whole grains may vary fromcountry to country In 1999, the definition of a wholegrain ingredient was developed by theWhole Grains Working Group of the American Association of Cereal ChemistsInternational, which stated that whole grains are “intact, ground, cracked or flaked fruit of

the grain whose principal components, the starchy endosperm, germ and bran, are present inthe same relative proportions as they exist in the intact grain” (American Association ofCereal Chemists International 1999) The United States adopted this definition in its WholeGrain Label Guidance (U.S Food and Drug Administration 2006) Another definition ofwhole grains published by the European HEALTHGRAIN Forum is “consisting of theintact, ground, cracked or flaked kernel after the removal of inedible parts such as the hulland husk The principle anatomic components—the endosperm, germ and bran—are to bepresent in the same relative proportions as they exist in the intact kernel Small losses ofcomponents, that is, <2% of the germ or <10% of the bran, which may occur throughprocessing methods consistent with safety and quality, are allowed” (Bjorck et al 2012).The difference between the two definitions is that the latter allows for small losses duringthe initial processing and/or cleaning of the grain (Ferruzzi et al 2014) A wholegrain

“food” had not been defined up until 2012, when a group of experts held a Whole GrainRoundtable in Chicago, Illinois, to discuss this After examining the scientific evidence, theexpert panel recommended that 8 g of whole grain/30 g serving (27 g/100 g), without a fiberrequirement, be considered the minimum content of whole grains that is nutritionallymeaningful and that a food providing at least 8 g of whole grains/30 g serving be defined as

a wholegrain food (Ferruzzi et al 2014) However, Cumming and Stephen rightly pointedout that because the type of grain contributing to whole grains varies from country tocountry, with most of the wholegrain intake being wheat in the United Kingdom and oats inthe United States, the differences in their physical and physiologic properties, including, wecould add, GI, need to be considered when examining the health impacts of wholegrainconsumption (Cummings and Stephen 2007)

1.4 CARBOHYDRATE INTAKE RECOMMENDATIONS AND

DIETARY ROLES

Dietary recommendations across the world point to the essential and central role ofcarbohydrates in the diet The 1998 FAO/WHO Expert Consultation on carbohydrates inhuman nutrition recommends that carbohydrates provide 55%–75% of the total energyintake from a variety of sources and that excessive intake of sugars, which compromisemicronutrient density, should be avoided (Joint FAO/WHO Expert Consultation 1998) Theconsultation endorses that the bulk of carbohydrate-containing foods consumed should berich in NSP and should have low GI (Joint FAO/WHO Expert Consultation 1998) However,the 2007 scientific update of FAO/WHO on carbohydrates identified the need to review therecommended lower limit because of insufficient justification and suggested a possiblerevision of 50% of total energy (Cummings and Stephen 2007)

The dietary recommendations of countries around the globe for carbohydrates aresimilar The U.S Institute of Medicine recommends the acceptable macronutrientdistribution range (AMDR) for carbohydrates to be between 45% and 65% of total energyand suggests a tolerable upper intake level of 25% of total energy from added sugars For

all ages, starting from 1 to more than 70 years, the Dietary Reference Intakes forcarbohydrates are set as: Estimated Average Requirement, 100 g/day, and RecommendedDietary Allowance, 130 g/day (Institute of Medicine 2006) (The Estimated AverageRequirement refers to “the average daily nutrient intake level estimated to meet therequirements of half of the healthy individuals in a group” whereas the RecommendedDietary Allowance is the “average daily dietary intake level sufficient to meet the nutrientrequirements of 97%–98% of the health individuals in a group” [Institute of Medicine2006].) The Dietary Guidelines of Americans 2010 recommend that at least half of allgrains consumed should be whole grains and also recommend to achieve this by replacingrefined grains with whole grains (U.S Department of Agriculture and U.S Department ofHealth and Human Services 2010) Similarly, Canada’s Food Guide recommends that atleast half of the grain products each day should be whole grain and also recommends tochoose grain products that are low in fat, sugar, or salt (Health Canada 2015).

The European Food Safety Authority (EFSA) Scientific Panel’s recommended referenceintake for carbohydrates is 45%–60% of energy, with an adequate fiber intake considered as

25 g/day (EFSA 2010) The U.K.’s recommendations refer to 50% of total food energyconsumed as carbohydrates and this to be broken down to not more than 11% nonmilkextrinsic sugars and 39% intrinsic and milk sugars and starch The individual DietaryReference Intakes (minimum to maximum) for NSPs are 12–24 g/day, with 18 g/day beingthe recommended population average intake With regard to public health messages, theU.K Government recommends consuming “plenty of starchy foods such as rice, bread,pasta, and potatoes (using wholegrain varieties when possible).” Australia’s dietaryguidelines also recommend daily consumption of “grain (cereal) foods, mostly wholegrainand/or high-cereal-fiber varieties, such as breads, cereals, rice, pasta, noodles, polenta,couscous, oats, quinoa, and barley” (National Health and Medical Research Council 2013).The dietary roles of carbohydrates are shown in Table 1.3 Carbohydrate-containingfoods are the staple foods in most countries, because their most important role is to provideenergy (Joint FAO/WHO Expert Consultation 1998), with significant effects on satiety(Blundell et al 1994) In addition, they are an important vehicle for protein intake and asource of vitamin B complex and minerals (calcium, magnesium, potassium, phosphorus,selenium, manganese, zinc, and iron) (Joint FAO/WHO Expert Consultation 1998) Inparticular, wholegrain carbohydrate foods are important sources of fiber, vitamin E,antioxidants, and phytoestrogens, found in their bran and germ components (Anderson et al.2000)

1.4.1 POSSIBLE HEALTH RISKS POSED BY CONSUMPTION OF FREE

SUGARS

Consumption of free sugars, being, as defined by WHO “monosaccharides anddisaccharides added to foods and beverages by the manufacturer, cook or consumer, andsugars naturally present in honey, syrups, fruit juices, and fruit juice concentrates” (WHO

2015), has been implicated in the development of dental caries and overweight and obesity.With regard to dental caries, sugars are reported to be “cariogenic,” which refers to

“foods/drinks containing fermentable carbohydrates that can cause a decrease in salivary

pH to <5.5 and demineralization when in contact with microorganisms in the mouth”(American Dietetic Association 2003), but not all sugars have the same potency The mostcariogenic sugar is sucrose, followed by fructose, glucose, and maltose, whereas lactose,galactose, maltodextrins, and polysaccharides have very little effect and sorbitol and xylitolare noncariogenic and are used in products such as sugar-free chewing gums Thosecarbohydrate-containing foods that are chewy and/or sticky are particularly detrimental toteeth A WHO-commissioned meta-analysis of the effect of sugar on dental caries found astrong positive association between consumption of free sugars and dental caries inchildren, with higher rates of dental caries when consumption of free sugar was more than10% of total energy In three national population studies, per-capita sugar intake below 10kg/person/year (approximately 5% of total energy intake) was associated with lower levels

of type 2 diabetes (Odegaard et al 2010; Palmer et al 2008; Schulze et al 2004).Consumption of 1–2 servings of SSB per day was associated with a 26% (95% confidenceinterval [CI] 12%–41%) greater risk of developing type 2 diabetes compared with just

occasional intake (Malik et al 2010), whereas a meta-analysis of randomized controlledtrials (RCTs) commissioned by WHO found that decreased intake of added sugars

significantly reduced body weight (0.80 kg, 95% CI 0.39–1.21; P < 001), whereas increased sugar intake led to a comparable weight increase (0.75 kg, CI 0.30–1.19; P = 001) (Te et al.

2013) Similarly, in children, it has been shown that a higher intake of SSB is associatedwith a 55% (95% CI 32%–82%) higher risk of overweight or obesity compared with a lowerintake (Te et al 2013) Moreover, considerable epidemiologic evidence also suggests thatincreased intake of added sugars, sucrose, and/or HFCS, or SSB is associated withdyslipidemia, CVD, and metabolic syndrome (Richelsen 2013), whereas the higher theintake of added sugar, the greater the risk (Yang et al 2014)

With regard to the effect of reduction of intake, two large long-term RCTs in childrenand adolescents showed that reduction in the consumption of SSB leads to significantreduction in weight gain and adiposity (de Ruyter et al 2012; Ebbeling et al 2012).Furthermore, in a study lasting 6 months, sugar-sweetened cola was shown to significantlyincrease visceral, liver, and muscle fat; triglycerides; total cholesterol; and systolic bloodpressure in comparison with milk, diet cola, or water, which did not affect body weight ortotal body fat, showing that SSB can mimic many features of the metabolic syndrome(Maersk et al 2012) Even so, Kaiser et al (2013) in an updated meta-analysis of RCTsattempting to reduce SSB consumption concluded that the evidence is equivocal on whetherreducing SSB will reduce the prevalence of obesity To this, Hu (2013) answered thatprospective cohort studies that address dietary determinants of long-term weight gain andchronic diseases are as critical as RCTs in evaluating causality and thus cannot be ignored.Another issue of debate is where the metabolism of HFCS, typically composed of 55%fructose and 45% glucose (for beverages) or 42% fructose and 58% glucose (for bakedgoods) (White et al 2010), is different to that of sucrose, composed of 50% fructose and50% glucose It has been demonstrated that acute responses to HFCS and sucrose areidentical with regard to glucose, insulin, leptin, ghrelin, triglycerides, and appetite(Melanson et al 2007; Soenen and Westerterp-Plantenga 2007) Whether researchcomparing fructose with glucose is relevant to human nutrition has also been questioned,because these sugars are rarely consumed in isolation in the human diet (Rippe andAngelopoulos 2013) The same researchers reported that 10-week consumption of addedsugar (sugar and HFCS at 8%, 18%, or 30% of calories) up to the 90th percentile populationconsumption for fructose did not affect blood pressure and had modest effects on bloodlipids (Lowndes et al 2014), whereas a meta-analysis of controlled feeding studies on theeffect of fructose on blood pressure also reported no effect (Ha et al 2012) However, yetanother study produced conflicting results by finding that consumption of beveragescontaining 10%, 17.5%, or 25% of estimated energy requirements from HFCS resulted indose-dependent increases of circulating lipid and/or lipoprotein risk factors for CVD anduric acid within 2 weeks, prompting the authors to suggest that these results providemechanistic support for the epidemiologic evidence, linking the increasing consumption ofadded sugar with cardiovascular mortality (Stanhope et al 2015)

Based on the available evidence on body weight and dental caries, WHO issued aguideline in 2015 recommending that adults and children reduce their daily intake of freesugars to less than 10% of total energy intake A further reduction to less than 5% orroughly 25 g (6 teaspoons) per day would provide additional health benefits (WHO 2015).Finally, there is no doubt that sugar-sweetened beverages and sucrose provide only emptycalories and they have never been shown to provide any benefit that would support theirintake.

1.5 INSULIN

Central to the health effects of carbohydrates is insulin, a hormone produced by the β (beta)cells of the pancreas, which was isolated by Frederic Banting and Charles Best in 1921 inCanada and which is required for the proper use of glucose by the body Insulin appears toactivate a process that helps glucose molecules enter the cells of striated muscle andadipose tissue In addition, it stimulates the production of glycogen by the liver Insulinalso promotes protein synthesis and helps the body store fat by preventing its breakdownfor energy

1.5.1 INSULIN AND THE METABOLIC SYNDROME

Metabolic syndrome is a cluster of disorders linked with obesity and hyperinsulinemia and

is associated with a markedly increased risk of type 2 diabetes and CVD (Soderberg et al.2005) The metabolic syndrome is characterized by impaired insulin sensitivity (insulinresistance), hyperglycemia, dyslipidemia, and hypertension Insulin resistance is the mostapproved and unifying hypothesis to explain the pathophysiology of the metabolicsyndrome (Eckel et al 2005), and it is strongly associated with obesity, especially itscentral or visceral component A number of mechanisms have been suggested regarding thedevelopment of insulin resistance and the metabolic syndrome, which may be caused byvisceral obesity (Frayn 2000) First, as illustrated in Figure 1.3, visceral adipose tissuesecretes cytokines and hormones that drain into the portal vein and may alter hepaticmetabolism Second, the visceral adipose depot releases nonesterified fatty acids morerapidly that subcutaneous adipose tissue, consequently increasing fatty acid flux to theliver The increased hepatic uptake of nonesterified fatty acids could increase hepaticglucose production while decreasing glucose oxidation, thus resulting in glucoseintolerance (Belfiore and Iannello 1998; Ferrannini et al 1983); increase hepatic very-low-density lipoprotein-triglyceride secretion, which causes hypertriglyceridemia (Frayn 2000);and decrease hepatic insulin removal, thus leading to hyperinsulinemia (Wiesenthal et al.1999) Thus, increased accumulation of visceral fat may be the main factor that leads to thedevelopment of the metabolic syndrome It is apparent that healthy subjects show markedvariability in the location and size of fat depots, and this may contribute to differences indisease risk

FIGURE 1.3 The metabolic syndrome (From Davy, B.M and Melby, C.L 2003 J Am.

Diet Assoc., 103, 1, 86–96 With permission.)

The dietary influences on metabolic syndrome are many and complex, with potentiallysynergistic effects both in protective and detrimental dietary patterns A recently conductedreview identified three dietary patterns that have potentially beneficial effects on theprevalence of the metabolic syndrome: the Mediterranean diet, the Dietary Approaches toStop Hypertension (DASH) diet, and the Nordic diet On the contrary, the Western dietarypattern characterized by high intakes of total and saturated fats and simple and added sugarshas been associated with higher risk of the metabolic syndrome Although it is outside thescope of this book to discuss them in detail, these dietary patterns include increasedconsumption of fruits, vegetables, whole grains, and (low-fat) dairy, and their relatively low

GI, among others (such as calcium, vitamin D, and omega-3 fatty acids), has beensuggested as one of the likely mechanisms in which they may exert their protective effects(Calton et al 2014)

1.6 GLYCEMIC INDEX: A HISTORY

The GI was conceived in 1981 by David Jenkins and his colleagues at the University ofToronto, Canada, as a tool for classifying carbohydrates according to their effect on bloodglucose concentrations (Jenkins et al 1981) At that time, it was thought that all simplesugars caused a more rapid rise in blood sugar than complex carbohydrates, but somestudies were beginning to emerge that challenged this conventional wisdom about sugars.The GI was developed to predict postprandial (after a meal) increases in blood glucose

In 1997, a committee of experts was brought together by FAO and WHO to review theavailable research evidence regarding the importance of carbohydrate in human nutritionand health (FAO 1998) The committee authorized the use of the GI method for classifyingcarbohydrate-rich foods and recommended that the GI values of foods be used incombination with information about food composition to guide food choices Tables on themeasured GIs of various carbohydrate-rich foods have been published, aiming to bringtogether all the published data on the GI values of individual foods for the convenience ofusers (Atkinson et al 2008; Foster-Powell et al 2002)

1.6.1 DEFINITION OF THE GLYCEMIC INDEX

GI is defined as “the incremental area under the blood glucose response curve of a 50 gcarbohydrate portion of a test food, expressed as a percentage of the response to the sameamount of carbohydrate from a standard food taken by the same subject (either white bread

or glucose)” (Jenkins et al 1981) In effect, GI ranks carbohydrate-containing foods based

on how quickly they elevate blood sugar concentration By comparing the area under theblood glucose response curve of the test food with that of the standard food, which is given

a relative value of 100, foods receive a numeric value and are then generally classified ashaving a low, moderate, or high GI (Jenkins et al 1984) Foods containing carbohydratesthat are quickly digested have the highest GI, because the blood sugar response is fast andhigh Slowly digested carbohydrates have a low GI, because they release glucose graduallyinto the bloodstream (Brand-Miller et al 2002; 2003a) In general, most refinedcarbohydrate-rich foods have a high GI, whereas nonstarchy vegetables, fruits, and legumestend to have a low GI (Ludwig 2002) However, there are many factors that determine the

GI of a food, and these are discussed in Section 1.7

The GI of a food is measured by comparing the increase in blood glucose concentrationafter eating 50 g of available carbohydrate (i.e., the total carbohydrate content of foodscompletely hydrolyzed and absorbed in the small intestine and used in metabolism, i.e.,total carbohydrate minus dietary fiber) from a test food with the same quantity of availablecarbohydrate from a standard food, which is either pure glucose or white bread The averagechange in blood sugar concentration over the next 2 h, compared with the change in bloodsugar concentration after consuming the standard food is the GI value of that particularfood The blood sugar response of the standard food, usually glucose or white bread, isgiven a value of 100, and all other foods are compared with this value A detailedexplanation of how GI is calculated, including example calculations, can be found in

Chapter 2

The GI is defined as: (Joint FAO/WHO Expert Consultation 1998)

GI= (iAUC for test food containing 50g available carbohydrate) / (iAUC 50g standard

food) × 100

where iAUC = incremental area under the curve (i.e., area above fasting concentrations).There is a direct correlation between a food’s glycemic response and insulinemicresponse, also referred to as insulin index, calculated in the same way as the GI by using the2-h postprandial insulin iAUC rather than the glucose iAUC (Bornet et al 1987).

Carbohydrate-containing foods can be ranked according to their glucose response as low,medium, or high GI; however, the cutoff values are arbitrary The most widely acceptableclassification is shown in Table 1.4 (Brand-Miller et al 2003a)

1.6.2 GLYCEMIC LOAD

The GL is a measure of the overall glycemic impact of the food and is the product of thefood’s GI and the amount of carbohydrate it provides It incorporates both the quantity andquality of the dietary carbohydrate consumed, as opposed to GI, which measures only thequality of carbohydrate intake (Wolever 2003) Each unit of GL is equal to the glycemiceffect produced after the ingestion of 1 g of glucose (used as a reference food), and thehigher the GL, the greater the expected elevation in blood glucose and insulin

The GL is defined as: (Salmeron et al 1997)

GL = (GI × amount of carbohydrate)/100

The question of whether the GI or the carbohydrate content of a food is the greatestdeterminant of GL was addressed by Brand-Miller and colleagues in 2003, and it was shownthat the carbohydrate content of foods alone explained 68% of the variance in GL values,whereas the GI value alone explained 49% of the variance, concluding that carbohydratecontent (rather than GI) is the greatest determinant of GL (Brand-Miller et al 2003b).Foods with a GL ≤ 10 are classified as low GL and those with a GL ≥ 20 are classified ashigh GL (Brand-Miller et al 2003b), but this classification is mainly used for researchpurposes

type of carbohydrate, the nature of the starch (amylose or amylopectin), cooking and foodprocessing (degree of starch gelatinization, particle size, cellular form, etc.), the food form,and other food components (fat, protein, and natural or added substances that reducedigestion, such as viscous fiber and acidity) Furthermore, the GI of a food is determined by

a complex interaction of factors, including but not limited to the physical and chemicalproperties of foods, the rate of carbohydrate digestion, the rate of gastric emptying, thepresence of other nutrients or antinutrients, and the insulin response elicited by the food.Therefore, the GI reflects the combined effect of all the properties of a food or meal thatinfluence the rate of influx and removal of glucose from the circulation (Englyst and

Englyst 2005; Wolever 2006) Hence, a GI value is an in vivo measurement and does not always correlate with in vitro measurements of glycemic response, because the latter cannot

fully account for all the factors that determine a GI value (Brand-Miller and Holt 2004) Itshould also be noted that not all the food factors or mechanisms that determine theglycemic response to a food or meal are equally beneficial to health A brief overview ofthe factors that affect a food’s GI is provided below, but further details can be found in

Chapter 13

1.7.1 STARCH TYPE

Different kinds of a specific food, for example, rice, have different GI values, because thetype of starch determines the rate of digestion As explained in Section 1.2.4, there are twotypes of starch: amylose and amylopectin; amylose is a linear molecule and amylopectin ishighly branched In most starchy foods, 70%–80% of the total starch consists ofamylopectin, with the remaining being amylose (Cummings and Englyst 1995) As a result

of its linear structure, amylose has more extensive hydrogen bonding and is easier toretrograde than amylopectin Thus, amylose is more resistant to hydrolytic enzymes, withresultant lower glycemic and insulinemic responses than amylopectin There are alsovarious genotypes of cereals such as barley, corn, and rice with different amyloseto-amylopectin ratios Individuals consuming high amylose-containing rice (Juliano andGoddard 1986) and corn (Behall et al 1988; 1989; Granfeldt et al 1995) were reported tohave significantly lower postprandial serum glucose and insulin responses However, ricevarieties with similar high-amylose contents can also differ in physicochemical properties,and this, in turn, may influence starch digestibility Examples of rice with different GIs arebasmati rice, a high-amylose rice, which has a GI of 58, and instant rice, which has a higherproportion of amylopectin and a GI of 87 (Foster-Powell et al 2002)

1.7.2 PROCESSING

The method of processing a single food can greatly change its GI Grinding, rolling, ormilling starchy foods reduces particle size and makes it easier for water to be absorbed anddigestive enzymes to attack the food Processing can also remove the fibrous outer coat of

the grain that slows down the access of digestive enzymes to the starch inside and at thesame time increases the GI value (Asp 1987) Chemically modifying a food also affects its

GI For instance, 1%–2% acetylated potato starch results in a decrease in GI (Raben et al.1997), as does the addition of β-cyclodextrin to stabilize the carbohydrate (Raben et al.1997)

1.7.3 PREPARATION

The applications of heat and moisture and cooking time, all have a significant effect on GI(Vaaler et al 1984) During cooking, water and heat expand the starch granules to varyingdegrees Foods containing starch that has gelatinized to bursting point, such as boiled orbaked potatoes, are more easily digested and therefore have higher GI values comparedwith foods containing starch granules that are less gelatinized, such as oatmeal or brownrice For illustration, the GI of a baked potato is 85, whereas that of brown rice is 50(Englyst and Cummings 1987; Foster-Powell et al 2002)

1.7.4 PROTEIN, FAT, AND CARBOHYDRATE

Numerous studies have shown that protein and fat decrease the blood glucose response andenhance insulin secretion when added to a carbohydrate meal (Estrich et al 1967; Gulliford

et al 1989) Eating protein-rich food in the same meal stimulates insulin secretion and thusreduces the blood glucose response of that meal (Nuttall et al 1984) Protein foods delaystomach emptying, which in effect delays digestion of starches In addition, foods or mealswith a higher fat content will have a lower GI than those with a lower fat content because ofthe property of fat to delay gastric emptying (Thompson et al 1982) However, foods with ahigh fat content have also been shown to enhance postprandial insulin secretion because ofthe large increase in gastric inhibitory polypeptide concentration (Collier and Odea 1983).Gastric inhibitory polypeptide potentiates glucose-induced insulin secretion (Sarson et al.1984) To demonstrate the effect of protein and fat, the GI of potato chips is 57, that ofFrench fries is 75, and that of baked potato is 85 (Foster-Powell et al 2002)

1.7.5 FIBER

Viscous, soluble fiber thickens the mixture of food in the digestive tract, which slows downenzymes from digesting the starch (Jenkins et al 1986) This results in a lower blood sugarresponse and a lower GI Foods containing viscous fiber such as barley (Wolever et al.1988) and legumes (Jenkins et al 1983) have a low GI However, it is not possible todifferentiate the effects of the type of fiber from those of the food form, particle size, starchtype, antinutrients, and the starch-protein interaction

1.7.6 SUGAR