930S Am J Clin Nutr 1995;61(suppl):930S-7S. Printed in USA. © 1995 American Society for Clinical Nutrition

Classification and methodology of food carbohydrates asrelated to nutritional effects1’2

Nils-Georg L Asp

ABSTRACT Dietary guidelines encourage a considerableincrease in carbohydrate intake compared with the present

situation in Western countries. Recent developments regarding

nutritional effects of various digestible and undigestible carbo-

hydrates call for more detailed recommendations. The “carbo-

hydrate by difference” concept emerged 150 y ago because of

the lack of specific analytical techniques and still prevails. The

concept of available compared with unavailable carbohydrates

was introduced in 1929 to obtain a better measure of glu-

cogenic carbohydrates in diabetes. Dietary fiber was first de-

fined as the “skeletal remnants of plant cell walls,” but the

definition was later expanded to include all polysaccharides

and lignin that are not digested in the small intestine. The

gravimetnic method of the Association of Official Analytical

Chemists for total dietary fiber is based on this undigestibilityconcept. However, precipitation of soluble fiber componentswith alcohol, which is used in all current methods, creates an

arbitrary delimitation between oligo- and polysacchanides. The

complex carbohydrates concept is challenged by recent devel-opments regarding nutritional effects of various food carbohy-drates. Am J Clin Nutr 1995;6i(suppl):930S-7S.

KEY WORDS Dietary guidelines, simple sugars, complex

carbohydrates, disaccharides, oligosacchanides, polysacchar-

ides, starch, dietary fiber, inulin, polyols

Introduction

Until recently, recommendations on carbohydrate intakehave been more or less secondary to guidelines regarding

protein and fat intakes. The dietary fiber hypothesis, however,focused attention on the undigestible carbohydrates that are

also in the human diet. Consequently, many dietary guidelinesnow include recommended daily intake of dietary fiber (1).

It has generally been assumed for a long time that sugar andother low-molecular-weight carbohydrates are more rapidly

digested and absorbed in the small intestine than is starch, the

only digestible food polysacchanide. This assumption forms amain part of the concept of “complex carbohydrates” (2). Assummarized below, several food-related factors other than the

molecular size of the carbohydrates are now known to deter-

mine glycemic and hormonal responses after intake of carbo-

hydrate. The glycemic index of foods is important in diabetes

(3) and may be related to blood lipid concentrations, blood

pressure, satiety, and physical performance (4).

Sucrose has been implicated as the arch criminal in dental

caries and there is much evidence that sucrose intake is closely

related to caries. However, dental health has improved in many

countries despite a continuous sucrose intake above the recom-

mended amount. Host resistance factors, including general

nutritional status, obviously modulate the effect of sucrose,

leaving certain groups at risk (5). Furthermore, other ferment-

able carbohydrates, including starch, are capable of lowering

dental plaque pH after a meal and thus are potentially cario-

genic (6, 7).

Historical aspects

The proximate analysis of foods and feed developed in the

middle of the 19th century. Because of the absence of specific

analytical techniques carbohydrates were considered to be the

material remaining after analysis of protein, fat, ash, and mois-

tune (8). Studies of ruminant nutrition at the Weende Expeni-

mental Station in Germany revealed differences in the feed

value of different carbohydrates and this led to the concept and

analysis of crude fiber (9).

In a recent review, Southgate (10) described further devel-

opments regarding carbohydrates. Atwater (1 1) realized that

foods containing significant amounts of crude fiber had lower

digestible energy than predicted by measuring “carbohydrates

by difference.” This led to the introduction, in due time, of the

specific energy conversion factors of Merrill and Watt (12).

The first to report more detailed analytical work on carbohy-

drates in food was Rubner (13) in 1917. He used hydrolysis and

reducing sugar methods and also measured lignin as the acid-

insoluble residue.

A milestone in the understanding of the nutritional impor-tance of food carbohydrates was the differentiation of “avail-

able” and “unavailable” carbohydrates by McCance and Law-

rence (14) in 1929. Their main objective was to define

glucogenic (available) carbohydrates to improve nutritional

counseling to diabetic subjects. These authors also considered

the question, far ahead of their time, of whether short-chain

fatty acids that were produced as the result of fermentation of

unavailable carbohydrates would provide any energy to the

body. In 1935 Widdowson and McCance (15) developed meth-

ods for analyzing reducing sugars, sucrose, and starch in foods

From the Department of Applied Nutrition and Food Chemistry,

Chemical Center, Lund University, Lund, Sweden.2 Reprints not available. Address correspondence to N-G Asp, Chemical

Center, P0 Box 124, S-221 00 LUND, Sweden.

CLASSIFICATION OF FOOD CARBOHYDRATES 931S

as a measure of available carbohydrates, which were subse-

quently introduced in British food tables (16). In the United

States, Williams and Olmstedt (17) developed a method in

1935 that simulated digestion by incubating a food sample with

pepsin and pancreatin. The residue provided a more physio-

logical way to estimate undigestible material than the acid and

alkali treatment used in the crude fiber method. In fact, this

work was the basis for enzymatic gravimetnic methods for

dietary fiber analysis, which were developed later by Hellen-

doom et al (18) in the Netherlands. The introduction of cob-

nimetnic methods that were group-specific (ie, for pentoses,

hexoses, and unonic acids) constitutes the next step in analyt-

ical developments. Such methods form the basis for South-

gate’s analytical scheme including both digestible (available)

and undigestible (unavailable) carbohydrates (19, 20). In the

1960s, Van Soest and Wine (21, 22) provided important con-

tributions toward improvement of the crude fiber method of

animal nutrition, leading to the acid detergent and neutral

detergent fiber methods and subsequently to modifications

suitable for starchy foods (23, 24). In Berlin, Thomas (25)

studied both chemical and physiological aspects of “Ballast-

stoffe” in cereal flours and bran.

The term “dietary fiber” was first used by Hipsley (26) in

1953 to describe plant cell walls in the diet, which he thought

were protective against toxemia of pregnancy. In 1972, Trowell

(27) detected differences in the incidence of noninfective dis-

eases in rural Africa and Western countries and defined dietary

fiber as “the skeletal remains of plant cells,” later rephrased to

“remnants of the plant cell wall” (28). In 1976, the definition

was extended to include all undigestible polysacchanides and

lignin (29). Two factors led to this new definition: first, isolated

polysacchanides such as pectin and guar gum were used to

study physiological effects of dietary fiber components, and

second, such isolated polysacchanides, which were often used

as food additives, could not be differentiated analytically from

plant cell wall polysaccharides (30). In fact, knowledge about

effects of dietary fiber on carbohydrate and lipid metabolism is

still based mainly on studies with isolated polysacchanides

(3 1 , 32). These have often been used in amounts far exceeding

those occurring naturally in foods and the effects extrapolated

to foods. To date, no physiological effects of dietary fiber

components per se have been attributed to their presence in the

plant cell wall; rather, the consequences of dietary fiber com-

ponents are attributed to their undigestibility and to their phys-

ical properties such as viscosity (33). The presence of cells

with intact walls composed of fiber polysacchanides is impor-

tant for the enclosure of nutrients such as sugars in fruits (34)

and starch in beans (35, 36), slowing their absorption.

As described by Southgate (37), the 1976 definition of

dietary fiber of Trowell et al (29) is virtually identical with that

for unavailable carbohydrates (14). This definition could in-

dude the recently discovered resistant starch (38), and has been

widely accepted throughout the world by both scientists and

policymakers.

Primary carbohydrates of food

The most important properties of the primary carbohydrates

of food (Table 1) that influence nutritional value include the

following: the extent of absorption in the small intestine, the

TABLE 1

Main food carbohydrates

Monosaccharides

GlucoseFructose

Disaccharides

Sucrose

Lactose

Oligosaccharides

cr-Galactosides

Raffinose, stachyose

Fructans

Fructooligosaccharides

Polysaccharides’

Starch

Amylose

Amylopectin

Modified food starches

Nonstarch polysaccharides

CelluloseHemicellulosePectins

�3-glucans

Fructans

GumsMucilagesAlgal polysaccharides

‘ Degree of polymerization > 10-20.

rate of absorption in the small intestine, the metabolism of

absorbed monomers, and the extent, rate, and nature of prod-

ucts of fermentation in the large intestine. The extent of diges-

tion in the small intestine (digestibility) determines how large

the fraction of total carbohydrates is that will provide glucose

to the organism and the amount of carbohydrate that will pass

to the large bowel for subsequent fermentation. The digestibil-

ity of food carbohydrates is the most important nutritional

property. The rate of absorption in the small intestine primarily

determines the glycemic and hormonal responses after a meal

and is often expressed as the glycemic index, as defined by

Jenkins Ct al (39).

Glucose and fructose, which are present in foods as free

sugars or constituents of sucrose, are metabolized differently;

fructose yields much lower postprandial glycemic and insulin

responses than does glucose. Accordingly, the ratio of glucose

to fructose in the diet is of clinical interest, especially in

diabetes (40). Fermentation of undigestible carbohydrates is

now the subject of much interest and research. There is cvi-

dence, mainly from in vitro studies, that the proportion of the

various short-chain fatty acids formed during fermentation

differs depending on the specific carbohydrate as substrate,

with concomitant differences in physiological effects (41).

Mono- and disaccharides

The free monosaccharides (glucose and fructose) are present

in fruits, berries, and vegetables. Accordingly, intake of thesesugars may be considerable in a vegetarian diet (42). Free

galactose is present in a very small amount, except in fer-

mented milk products, as a result of the preference of micro-

organisms for glucose during fermentation (43). Sucrose is

usually the most abundant disaccharide in both mixed and

vegetarian diets (42). The lactose intake stems from milk

932S ASP

products, with cow milk containing ““50 gIL. The absorption

capacity for mono- and disaccharides is high, except in mdi-

viduals with disacchanidase deficiencies. However, recent stud-

ies have demonstrated a limited absorption capacity for fruc-

tose when it is given alone. Simultaneous intake of glucose

increases the absorption of fructose (44, 45).

Oligosaccha rides

The quantitatively most important oligosacchanides are the

a-galactosides raffinose, stachyose, and verbascose, which

have three, four, and five monomeric units, respectively, and

the fructooligosacchanides (45). The ct-galactoside content is

5-8% on a dry matter basis in beans, lentils, and peas. Fruc-

tooligosacchanides and higher molecular weight fructans con-

stitute up to 60-70% of the dry matter in some plant tissues (eg,

Jerusalem artichokes and onions), and fructooligosacchanides

are found in smaller amounts in many plants, including cereal

grains (46). The a-galactosides are not hydrolyzed by human

intestinal enzymes (47). An increased use of both fructooligo-

saccharides and fructans and other undigestible oligosacchanide

preparations can be expected in foods because of their reduced

energy content, their effects on the intestinal flora, and their

functional properties (48).

Polyols and polydextrose

Small amounts of sugar alcohols (polyols) such as sorbitol

are found in fruits. There is an increasing use of polyols such

as xylitol, sorbitol, mannitol, lactitol, and maltitol as low-

energy and noncaniogenic bulk sweeteners (48). There is a

limited capacity of the small intestine to absorb these sugar

alcohols (49).

Polydextrose, a synthetic glucose polymer that is used as a

low-energy bulking agent, is undigestible because of glycosidic

linkages that are essentially resistant to amylase (50).

Starch

Starch is the most important food carbohydrate quantita-

tively and together with glycogen is the only digestible food

polysaccharide. It occurs as a mixture of virtually unbranched

amylose chains composed of a-i,4-linked glucose residues and

of highly branched amylopectin with a-1,4 and ct-i,6 bonds.

Chemically modified food starches in which ester or ether

groups have been introduced are used as food additives.

The generally accepted concept that starch is completely

although slowly digestible has been challenged recently. Hy-

drolysis of starch is initiated by salivary amylase and is com-

pleted in the small intestine. In 1961 it was demonstrated that

amylase activity in the duodenum is sufficient to hydrolyze

starch in a meal within minutes (51). Consequently, starch in

solution, partially degraded starch, and glucose give the same

glycemic and insulin responses when given in equivalent

amounts (52). However, the glycemic index of starchy foods

varies widely (53) because of food properties affecting the

availability of starch for enzymatic degradation (54). Resistant

starch was first discovered as a starch fraction that associated

with the nonstarch polysacchanides in dietary fiber analyses

unless the sample had first been treated with alkali or dimeth-

ylsulfoxide (DMSO) to solubilize this starch (55, 56). How-

ever, this type of resistant starch, identified as retrograded

amylose (57, 58), is only one of several forms of physiologi-

cally resistant starch that pass through the small intestine (59).

Physically enclosed starch in beans, for example (60), and

ungelatinized starch granules of the B-type (eg, green bananas

and raw potato starch) are other forms of resistant starch, as

well as chemically modified (61) or dry-heated (62) food

starches. The physiological effects of resistant starch are cur-

rently being investigated. Resistant starch is fermented (63)

and has a fecal bulking effect (64, 65), representing properties

that are considered to be typical for nonstarch polysacchanides.

Nonstarch polysaccharides

Nonstarch polysaccharides can be storage polysacchanides

such as fructans (eg, inulin), glucomannans, and galactoman-

nans (eg, guar gum), although structural plant cell wall p0-

lysaccharides such as cellulose, hemicellulose, and pectic sub-

stances are the most abundant. Mucilages, alginates, exudate

gums, and various modified polysacchanides are other constit-

uents of the nonstarch polysacchanides (30). The fecal bulking

effect of dietary fiber is most prominent if the nonstarch

polysaccharides are resistant to fermentation in the large intes-

tine (66, 67). Other physiological effects, such as lowering of

plasma low-density-lipoprotein (LDL) cholesterol and attenu-

ation of blood glucose and insulin responses after a meal, are

attributed mainly to soluble, gel-forming polysacchanides.

However, the LDL-bowening properties of dietary fiber in foods

are modest and the fact that high-fiber foods are generally low

in saturated fat may be more important in clinical practice (31).

Similarly, the content of viscous, soluble fiber, such as f3-glu-

cans in cereals, is less important for the glycemic response than

are structural properties that limit the availability of starch for

enzymic degradation (68, 69). An effect of carrots on reducing

the glycemic response after a mixed meal is seen only if �200

g carrots are added to the meal (70, 71). Thus, excessive

emphasis on the soluble fiber content of foods may be mis-

leading and the physiological properties of the soluble fiber

may be lost if viscosity is diminished through processing (33).

Analytical considerations

Mono-, di-, and oligosaccharides, including polyols

Depending on the food matrix to be analyzed, an alcohol

extraction of the free sugars may be necessary. A final ethanol

concentration of �80% (vol:vol) should be used to avoid

extraction of polysacchanides. Some sugars, especially lactose,

have low solubility and may need lower alcohol concentrations

(eg, 50% vol:vol) during extraction with a final increase to

precipitate polysaccharides (72). Physical methods such as

polarimetry, refractive index, or density are still useful in pure

systems, such as in sugar production control. Methods based on

the reduction of copper salts and colonimetnic methods based

on condensation reactions with anthrone, orcinol, and carbazol

can also be used in well-known systems (72). The enzymatic

procedures based on specific, highly purified enzymes (73)

have been instrumental in providing means of specific and

precise analysis of carbohydrates in mixtures without high-

capital investments. On the other hand, gas-liquid chromatog-

raphy and HPLC are preferable when several different carbo-

hydrates are to be determined simultaneously. HPLC analysis

has been hampered by the relative insensitivity of refractory

CLASSIFICATION OF FOOD CARBOHYDRATES 933S

index detectors. However, this problem has been overcome by

systems using amperometric detection (74).

Starch

Enzymatic hydrolysis and specific glucose assay are the

methods of choice for the measurement of starch. Acid hydro-

lysis is less suitable in mixed foods because of glucose liber-

ation from other glucose-containing polysaccharides such as

f3-glucans. However, the enzymes used have to be carefully

monitored for contaminating activities. A heat-stable amylase

such as Termamyl in a combined gelatinization and hydrolysis

step has turned out to be particularly useful (75).

Dietary fiber

There has been extensive controversy regarding the analysis

of dietary fiber, partly because the needs for research have not

been clearly separated from those of food labeling and legis-

lation. Dietary fiber can be analyzed according to two main yet

different principles (30). In the gravimetnic methods the non-

fiber components are removed and the residue is weighed. This

residue can be analyzed for monomeric composition or starch

residues and also for protein and ash. The crude fiber and

detergent fiber methods are both gravimetnic methods. Enzy-

matic gravimetnic methods as developed by Asp et a! (76) and

as approved by the Association of Official Analytical Chemists

(AOAC) (77, 78) use alcohol precipitation to recover soluble

fiber components. Such methods are useful to measure total

dietary fiber or soluble and insoluble components separately,

with appropriate correction for protein and ash in the fiberresidue. The component analysis methods use more or less

specific determination of monomeric constituents, with subse-

quent summing up for a total fiber determination. As in the

gravimetnic methods, soluble and insoluble components can be

determined separately. However, the solubility of polysaccha-

rides is method-dependent and is determined by temperature,

time, and pH (79). The Southgate procedure (19) uses colon-

metric methods to determine hexoses, pentoses, and uronic

acids. The methods of Theander et al (80) and Englyst et al (81)

use gas-liquid chromatography for neutral sugar components

and a colonimetric assay for uronic acids. HPLC determination

is gaining in popularity. A colonimetric measurement of reduc-

ing sugars has been introduced as an alternative to the gas-

liquid determination method developed by Englyst et al (81).Advantages and disadvantages with the two principally dif-

ferent methods of analyzing dietary fiber are summarized in

Table 2. Enzymatic-gravimetric methods are simple and no-bust, with no requirement for advanced equipment. There is a

risk of overestimating the fiber content if other components

remain in the residue. However, the residue can be analyzed for

any such contaminating components. Colorimetnic methods

can be inflated by nonspecific reactions and can give various

response factors for different monomers. Specific gas-liquid

chromatography and HPLC measurement, on the other hand,

require complete and quantitative recovery of monomers after

hydrolysis of the polysacchanides. Incomplete hydrolysis or

losses due to decomposition of monomers will lead to under-

estimation (30).

The current component analysis methods use acid hydroly-

sis. As in amino acid analysis, conditions for hydrolysis have to

be chosen to obtain an optimal compromise between hydrolysis

TABLE 2

Advantages and disadvantages of different kinds of methods for dietary

fiber analysi5’

Component analysisEnzymatic-

gravimetric GLC, HPLC Colorimetry

Equipment Simple Advanced Simple

Information on composition No Yes Yes and no

Risk of overestimation Yes2 No (Yes)

Risk of underestimation No Yes3 Yes3

1 For review, see reference 30. GLC, gas-liquid chromatography; HPLC,

high-pressure liquid chromatography.2 Residue can be analyzed.

3 If hydrolysis is incomplete.

yield and monomer degradation. Corrections are used for hy-

drolysis losses of the different components. Quantitative hy-

drolysis is particularly difficult to obtain with acidic polysac-

charides because of the high acid stability of glycosyl uronic

acid linkages. This fact and the more rapid degradation of

monomeric uronic acids in an acidic condition are reasons whycobonimetnic methods are still preferred for uronic acid deter-

minations (30).Enzymatic, gravimetnic dietary fiber determination has been

tested in several collaborative studies carried out within the

AOAC. Recently this organization also tested the gas-liquid

chromatography component analysis method of Theander et al(80). The gas-liquid chromatography and colorimetnic varieties

of the Englyst method have been tested in studies carried out bythe Ministry of Agriculture, Food and Fisheries in Great Bnit-

am. A comparison of studies indicates improved performanceover time with typical mean R95 values of 2-3 for both the

gravimetnic methods approved by the AOAC and for the En-

glyst method. The best performance thus far has been in a

Swiss study using the enzymatic gravimetric AOAC method

(R95 1.01.1) (82). An R95 of 2 at 100 g dietary fiber/kg meansthat 19 of 20 single estimates run in different laboratories arelikely to fall within the interval of 90-1 10 gfkg.

There are few formal collaborative studies covering morethan one method. In a recent study coordinated by the Euro-

pean Community Bureau of Reference, dietary fiber values

with the AOAC method could be certified for three differentmaterials. Indicative values only could be given for the Englyst

gas-liquid chromatography and colonimetnic methods, butmean values obtained with these methods were similar to thoseobtained with the AOAC method (83). For most foods, esti-mates of total dietary fiber with the enzymatic gravimetnic

method would not be significantly different from estimates ofnonstarch polysacchanides derived from the Englyst method.

This means that the CIs for the different methods overlap (30).

Note also that two collaborative studies have shown consis-

tently higher values with the colonimetric Englyst method thanwith the original gas-liquid chromatography variety (30). Onlyfor foods with particularly high amounts of resistant starch ofthe retrograded amylose type, or lignin, would Englyst-derivedvalues be expected to be significantly lower than estimates withmethods including these components.

Resistant starch

Methods for analysis of resistant starch are listed in Table 3.Originally, resistant starch was determined as the difference in

934S ASP

TABLE 3

Methods for analysis of resistant starch’

Year Principle

1982 Difference between NSP glucan with and without DMSO or

KOH solubilization (55)1984 Starch analysis in dietary fiber residue (56)

1986 Extensive ce-amylase digestion, analysis of remaining starch,

and no gelatinization step (84)

1992 Standardized ball milling and amylase digestion, and

analysis of various resistant starch fractions (85)

1992 Chewing to disintegrate foods, pepsin and pancreatin

digestion, analysis of remaining starch (86)

‘NSP, nonstarch polysaccharide; DMSO, dimethylsulfoxide.

nonstarch-polysaccharide-glucan without and with DMSO sol-

ubilization (55), or by analyzing a gravimetnic fiber residue for

starch associated with the fiber (56, 87). This starch can be

differentiated into “residual starch” or “resistant starch” (87).

The resistant starch determined by these methods consists of

mainly retrograded amylose (57, 58). Benny developed a

method in which the sample was subjected to extensive a-amy-

lase digestion without any heating step to gelatinize the starch.

In this way, resistant starch in the form of intact starch granuleswould also be analyzed. Two methods have been published that

are proposed to measure all of the physiologically resistant

starch (85, 86). A critical step is then to disintegrate the food in

a similar manner as when eaten. One of these procedures uses

chewing as a realistic disintegration step (86), whereas the

other method uses a standardized ball-milling step (85). Both

methods correlate with in vivo resistant starch as measured in

subjects with ileostomies.

Dietary fiber and unavailable carbohydrates

The definition of dietary fiber as undigestible polysacchar-

ides and lignin is similar to the definition of “unavailable

carbohydrates,” which is based on the fundamentally different

nutritional importance of various carbohydrates as determined

by their digestibility in the small intestine. Discussions of the

definition and analysis of dietary fiber have focused on whether

resistant starch and lignin are included (88, 38). Equally im-

portant, however, is awareness of components that are not

precipitated in the 70-80% (vol:vol) ethanol used in all the

methods to separate water-soluble fiber components. Inulin is

an undigestible polysacchanide that is not precipitated and

therefore not recovered in any of the methods. Polydextrose is

another undigestible oligopolysacchanide that also is not deter-

mined as dietary fiber. Physiologically, undigestible oligosac-

chanides have much in common with nonstarch polysacchar-

ides; they are fermented and influence the bacterial flora of the

colon (48, 89). Specific enzymatic or HPLC methods must be

used to determine the undigestible oligo- and polysaccharides

not recovered in the current dietary fiber analyses.

Classification of food carbohydrates for nutritionlabeling

Ever since the introduction of the carbohydrate by difference

concept, it has been customary in most countries to label both

digestible carbohydrates and dietary fiber as carbohydrates.

Chemically this is obviously correct and this convention has

been kept in the recent implementation of the US Nutrition

Labeling and Education Act (90). The Commission of the

European Communities, on the other hand, has defined carbo-

hydrates as metabolizable carbohydrates including polyols

(91). “Metabolizable” obviously means that carbohydrates

must be digestible in the small intestine, and this definition,

therefore, is related to the available carbohydrates concept (14).

Dietary fiber remains to be defined by the Commission.

The unavailable carbohydrates concept preceded the dietary

fiber concept, which was also based on undigestibility in the

small intestine. The plant cell wall origin of dietary fiber was

emphasized in the epidemiological association between dietary

fiber intake and disease. Physiological effects of dietary fiber,

however, have been related to undigestibility and physical

effects of dietary fiber constituents in the intestinal tract but

never to their plant cell wall origin. The importance of cell

walls in legumes and fruits for the glycemic response of these

foods is recognized, but disruption of these structures without

removing the fiber will abolish these effects. Methods pretend-

ing to give an index of plant cell walls cannot reveal such

structural properties (81).

A classification of food carbohydrates for nutritional label-

ing should be based on small intestinal digestibility. The en-

zymic gravimetnic AOAC methods for total fiber (77, 78) or

the component analysis method of Theander et al (80), which

include one important form of resistant starch and lignin, give

a closer measurement of unavailable carbohydrates than do the

Englyst methods (81), which do not include resistant starch and

lignin. However, all the methods including resistant starch need

further development regarding the distinction between digest-

ible and resistant starch. Furthermore, separate estimates of

alcohol-soluble but undigestible components, such as inulin,

polydextrose, and oligosacchanides have to be performed when

such components are present in significant amounts. The rela-

tion between nonstarch polysacchanides, total dietary fiber ac-

cording to the AOAC method, and unavailable carbohydrates is

illustrated in Figure 1.When a dietary fiber definition and an estimate as close as

possible to unavailable carbohydrates are used, available or

FIGURE 1. Relationship between nonstarch polysaccharides (NSP),

total dietary fiber, and unavailable carbohydrates. AOAC, Association of

Official Analytical Chemists.

CLASSIFICATION OF FOOD CARBOHYDRATES 935S

metabolizable carbohydrates can be calculated by difference.

Accordingly, an estimate of dietary fiber can be obtained by

difference if mono- and disaccharides and digestible starch are

determined as originally advised for unavailable carbohydrates

(15).The labeling of soluble fiber according to the AOAC method

is mandatory in the United States if claims regarding choles-

terol-lowening effects of the fiber are to be made. This require-

ment is reasonable because the cholesterol-reducing effects are

specifically related to gel-forming, soluble types of fiber. Note,

however, that the solubility of dietary fiber polysaccharides is

highly method-dependent because of different extraction con-

ditions (79). The most relevant conditions, from a physiologi-

cal perspective, remain to be defined. The monomeric compo-

sition of dietary fiber is of little guidance to predict

physiological effects, which are more related to viscosity,

fermentability, and structural properties of foods. However,

standardized methods for measuring such properties are still

lacking, although in vitro systems for prediction of glycemic

response seem promising (92). Such methods may prove useful

for documentation of food properties as a basis for specific

dietary advice or claims, although they may not be suitable to

produce specific data appropriate for food labels. U

References

1 . World Health Organization. Diet, nutrition, and the prevention of

chronic diseases. World Health Organization. World Health Organ

Tech Rep Ser 1990;797.

2. US Senate Select Committee on Nutrition and Human Needs. Dietary

goals for the United States. 2nd ed. Washington, DC: US Government

Printing Office, 1977.

3. Brand Miller JC. Importance of glycemic index in diabetes. Am J Clin

Nutr 1994;59(suppl):747S-53S.

4. Truswell AS. Glycaemic index of foods. Eur J Clin Nutr 1992;46:

591-101.

5. Navia J. Carbohydrates and dental health. Am J Clin Nutr 1994;

59(suppl):719S-28S.

6. Lingstrom P, Hoim J, Birkhed D, Bjdrck I. Effects of variously

processed starch on pH of human dental plaque. Scand I Dent Res

1989;97:392-400.

7. Lingstrom P, Birkhed D, Granfeldt Y, BjOrck I. pH measurements of

human dental plaque after consumption of starchy foods using the

microtouch and the sampling method. Caries Res 1993;27:394-401.

8. Henneberg W, Stohmann F. Contribution on the rational feeding of

ruminants. (Beitrage zur BegrUndung einer rationellen FUtterung der

Wiederkhuer, I.) Braunschweig, 1860 (in German).

9. Henneberg W, Stohmann F. On the maintenance feeding of one-year-

old cattle. (Ueber das Erhaltungsfutter volljahrigen Rindviehs.)

J Landw 1859;3:485-551 (in German).

10. Southgate DAT. The dietary fibre hypothesis: a historical perspective.

In: Schweizer IF, Edwards CA, eds. Dietary fibre-a component of

food. Nutritional function in health and disease. London: Springer-

Verlag, 1992:3-20.

1 1. Atwater WO. Discussion of the terms digestibility, availability and

fuel value. Twelfth annual report of the Storrs Agricultural Experi-

mental Station. Storrs Agricultural Experimental Station, 1900:69.

12. Merrill AL, Watt BK. Energy value of foods basis and derivation.

Agriculture handbook no. 74. Washington, DC: US Department of

Agriculture, 1955.

13. Rubner M. Investigations on whole grain bread. (Untersuchungen #{252}ber

Vollkornbrot.) Arch Mat Physiol 1917;74:245-372 (in German).

14. McCance RA, Lawrence RD. The carbohydrate content of foods.

London: Her Majesty’s Stationery Office, 1929. Medical Research

Council special report series no. 135.

15. Widdowson EM, McCance RA. The available carbohydrates of fruits.

Determination of glucose, fructose, sucrose and starch. Biochem J

1935;29:151-6.

16. McCance RA, Widdowson EM. The chemical composition of foods.

London: Her Majesty’s Stationery Office, 1940. (Special report series

of the Medical Research Council, no. 235.)

17. Williams RD, Olmstedt W. A biochemical method for determining

indigestible residue (crude fibre) in faeces: lignin, cellulose, and non-

water soluble hemicelluloses. J Biol Chem 1935;108:653-66.

18. Hellendoorn EW, Noordhoff MG, Slagman J. Enzymatic determina-

tion of the indigestible residue (dietary fibre) content of human food.

J Sci Food Agric 1975;26:1461-8.19. Southgate DAT. Use of the Southgate method for unavailable carbo-

hydrates in the measurement of dietary fiber. In: James WPT, Thean-

der 0, eds. The analysis of dietary fiber in foods. New York: Marcel

Dekker, 1981:1-19.

20. Holland B, Welch AA, Unwin ID, Buss DH, Paul AA, Southgate

DAT. McCance and Widdowsons’ the composition of foods. Cam-

bridge, UK: The Royal Society of Chemistry, 1991.

21. Van Soest PJ. Use of detergents in the analysis of fibrous feeds. I.

Preparation of fiber residues of low nitrogen content. J Assoc Off

Agric Chemists 1963;46:825-9.

22. Van Soest PJ, Wine RH. Use of detergent in the analysis of fibrous

feeds. IV. Determination of plant cell wall constituents. J Assoc Off

Anal Chem 1967;50:50-5.

23. Robertson JB, Van Soest PJ. The detergent system of analysis and its

application to human foods. In: James WPT, Theander 0, eds. The

analysis of dietary fiber in foods. New York: Marcel Dekker, 1981:

123-58.24. Jeraci JL, Lewis BA, Van Soest Pi, Robertson JB. Urea enzymatic

dialysis procedure for determination of total dietary fiber. J Assoc Off

Anal Chem 1989;72:677-81.

25. Thomas B. Contribution to the nomenclature and analysis of plant

cell-wall substances. (Beitrage zur Nomenklatur und Analytik

pflanzlicher Zellwandsubstanzen.) Gertreide Mehi Brot 1972;26:

158-64,168-9 in German.

26. Hipsley EH. “Dietary fibre” and pregnancy toxaemia. Br Med I

1953;2:420-2.

27. Trowell H. lschemic heart disease and dietary fiber. Am I Clin Nutr

1972;25:926-32.

28. Trowell H. Refined carbohydrate foods and fibre. In: Burkitt DP,

Trowell H, eds. Refined carbohydrate foods and disease. London:

Academic Press, 1975:23-41.

29. Trowell HC, Southgate DAT, Wolever TMS, Leeds AR, Gassull MA,

Jenkins DJA. Dietary fibre redefined. Lancet 1976;1:967.

30. Asp N-G, Schweizer iT, Southgate DAT, Theander 0. Dietary fibre

analysis. In: Schweizer TF, Edwards CA, eds. Dietary fibre-a com-

ponent of food. Nutritional function in health and disease. London:

Springer-Verlag, 1992:57-102.

31. Truswell AS, Beynen A. Dietary fibre and plasma lipids; potential for

prevention and treatment of hyperlipidaemia. In: Schweizer TF,

Edwards CA, eds. Dietary fibre-a component of food. Nutritional

function in health and disease. London: Springer-Verlag, 1992:

295-332.

32. Wolever TMS. Dietary management of diabetes. In: Knitchevsky D,

Bonfield C, Anderson JW, eds. Dietary fiber, chemistry, physiology

and health effects. New York: Plenum Press, 1990:287-99.

33. Jenkins DJA, Wolever TMS, Leeds AR, et al. Dietary fibres, fibre

analogues, and glucose tolerance: importance of viscosity. Br Med J

1978;1:1392-4.

34. Haber GB, Heaton KW, Murphy D, Burroughs LF. Depletion and

disruption of dietary fibre. Effects on satiety, plasma-glucose, and

serum insulin. Lancet 1977;2:679-82.

35. Wursch P, Del Vedovo 5, Koellreuter B. Cell structure and starch

936S ASP

nature as key determinants of the digestion rate of starch in legumes.

Am I Clin Nutr 1986;43:25-9.36. Tovar J, Granfeldt Y, Bjdrck I. Effect of processing on blood glucose

and insulin responses to legumes. I Agric Food Chem 1992;40:1846-

51.

37. Southgate DAT. Determination of food carbohydrates. Barking, Essex,

UK: Elsevier Science Publishers Ltd, 1991.

38. Asp N-G, BjOrck I. Resistant starch. Trends Food Sci Technol 1992;

3:111-4.

39. Jenkins DJA, Wolever TMS, Taylor RH, et al. Glycemic index of

foods: a physiological basis for carbohydrate exchange. Am J Clin

Nutr 1981;34:362-6.

40. Uusitupa MIJ. Fructose in the diabetic diet. Am J Clin Nutr 1994;

59(suppl):753S-7S.

41. Edwards CA, Rowland IR. Bacterial fermentation in the colon and its

measurement. In: Schweizer TF, Edwards CA, eds. Dietary fibre-a

component of food. Nutritional function in health and disease. London:

Springer-Verlag, 1992:119-50.42. Abdulla M, Aly K-O, Andersson I, et al. Nutrient intake and health

status of lactovegetanians. Am I Clin Nutr 1983;40:325-38.

43. AIm L. Effect of fermentation on lactose, glucose, and galactose

content in milk and suitability of fermented milk products for lactose

intolerant individuals. I Dairy Sci 1982;65:346-52.

44. Rumessen JJ, Gudmand-Hoyer E. Absorption capacity of fructose in

healthy adults. Comparison with sucrose and its constituent monosac-

charides. Gut 1986;27:1161-8.

45. Rumessen JJ. Review: fructose and related food carbohydrates.

Sources, intake, absorption and clinical implications. Scand I Gastro-

enterol 1992;27:819-28.46. Nilsson U, Nilsson B, Dahlqvist A. Cereal fructosans. Part 2. Charac-

terization and structure of wheat fructosans. Food Chem 1986;22:95-

106.47. Nilsson U, Oste R, Jagerstad M, Birkhed D. Cereal fructans: in vitro

and in vivo studies on availability in rats and humans. J Nutr 1987;

118:1325-30.

48. Bornet FRJ. Undigestible sugars in food products. Am I Clin Nutr

1994;59(suppl):763S-9S.

49. Wflrsch P, Koelireutter B, Schweizer IF. Hydrogen excretion after

ingestion of five different sugar alcohols and lactulose. Eur I Clin Nutr

1989;43:819-25.

50. Oku I, Fujii Y, Okamatsu H. Polydextrose as dietary fiber: hydrolysisby digestive enzymes and its effect on gastrointestinal transit time in

rats. J Clin Biochem Nutr 1991;11:31-40.

51. Dahlqvist A, Borgstram B. Digestion and absorption of disaccharides

in man. Biochem J 1961;81:411-8.

52. Wahlqvist ML, Wilmshurst EG, Murton CR, Richardson EN. Theeffect of chain length on glucose absorption and the related metabolic

response. Am I Clin Nutr 1978;31:1998-2001.

53. WUrsch P. Starch in human nutrition. World Rev Nutr Diet 1989;60:

199-256.

54. Bj#{246}rckI, Granfeldt Y, Liljeberg H, Tovar J, Asp N-G. Food properties

affecting the digestion and absorption of carbohydrates. Am J Clin

Nutr 1994;59(suppl):699S-705S.

55. Englyst HN, Wiggins HS, Cummings Ill. Determination of the non-

starch polysaccharides in plant foods by gas-liquid chromatography of

constituent sugars as alditol acetates. Analyst 1982;107:307-18.

56. Johansson C-G, SiljestrOm M, Asp N-G. Dietary fibre in bread and

corresponding flours-formation of resistant starch during baking. Z

Lebensm Unters Forsch 1984;179:24-8.

57. Siljestrdm M, Eliasson A-C, Bjdrck I. Characterization of resistant

starch from autoclaved wheat starch. St#{228}rke1989;41:147-51.

58. Russel PL, Berry CS, Greenwell P. Characterization of resistant starch

from wheat and maize. J Cereal Sci 1989;9:1-15.

59. Englyst HN, Cummings JH. Non-starch polysacchanides (dietary fiber)

and resistant starch. In: Furda I, Brine CJ, eds. New developments in

dietary fiber. New York: Plenum Press, 1990:205-25.

60. Schweizer IF, Andersson H, Langkilde AM, Reiman 5, Torsdottir I.

Nutrients excreted in ileostomy effluents after consumption of mixed

diets with beans or potatoes. II: starch, dietary fibre and sugars. Eur I

Clin Nutr 1990;44:567-75.

61. Bjdrck I, Gunnarsson A, OstergArd K. A study of native and chemi-

cally modified potato starch. Part II. Digestibility in the rat intestinal

tract. Sthrke 1989;41:128-34.

62. Siljestr#{246}m M, Bj#{246}rckI, Westerlund E. lransglycosylation reactions

following heat treatment of starch-effects on enzymic digestibility.

Sthrke 1989;41:95-100.

63. Cummings JH, Englyst HN. Measurement of starch fermentation in the

human large intestine. Can J Physiol Pharmacol 1991;69:121-9.

64. Shetty PS, Kurpad A. Increasing starch intake in the human increases

fecal bulking. Am J Clin Nutr 1986;43:567-75.

65. Cummings JH, Beatty SM, Kingman SA, Bingham 5, Englyst HN.

Faecal bulking effect of resistant starch. Gastroenterology 1992;102:

A548(abstr).

66. Nyman M, Asp N-G, Cummings JH, Wiggins H. Fermentation of

dietary fibre in the intestinal tract: comparison between man and rat.

Br J Nutr 1986;55:487-96.

67. Asp N-G, Bj#{246}rckI, Nyman M. Physiological effects of cereal dietary

fibre. Carbohydr Polym 1993;21:183-7.

68. Liljeberg H, Granfeldt Y, Bjdrck I. Metabolic response to starch in

bread containing intact kernels versus milled flour. Eur J Clin Nutr

1992;46:561-75.

69. Granfeldt Y, Liljeberg H, Drews A, Newman R, Bjdrck I. Glucose andinsulin responses to barley products: influence of food structure and

amylose-amylopectin ratio. Am J Clin Nutr 1994;59: 1075-82.

70. Gustafsson K, Asp N-G, Hagander B, Nyman M. Effects of different

vegetables in mixed meals on glucose homeostasis and satiety. Eur J

Clin Nutr 1993;47:192-200.

71. Gustafsson K, Asp N-G, Hagander B, Nyman M. Dose-response

effects of boiled carrots and effect of carrots in lactic acid in mixed

meals on glycaemic response and satiety. Eur I Clin Nutr 1994;48:

386-96.

72. Greenfield H, Southgate DAT. Food composition data. Production,

management and use. London: Elsevier Applied Science, 1992:94-

104.

73. Anonymous. Methods of enzymatic food analysis. Mannheim, Ger-

many: Boehringer Mannheim GmbH, 1984.

74. Lee YC. High-performance anion-exchange chromatography for car-

bohydrate analysis. Anal Biochem 1990;189:151-62.

75. Holm I, Bjdrck I, Drews A, Asp N-G. A rapid method for the analysis

of starch. Sthrke 1986;38:224-6.

76. Asp N-G, Johansson C-G, Hailmer H, SiljestrOm M. Rapid enzymatic

assay of soluble and insoluble dietary fiber. J Agnic Food Chem

1983;31:476-82.77. Prosky L, Asp N-G, Schweizer IF, DeVries JW, Furda I. Determina-

tion of insoluble, soluble, and total dietary fiber in foods and food

products: inter-laboratory study. J Assoc Off Anal Chem 1988;71:

1017-23.

78. Lee 5, Prosky L, DeVries I. Determination of total, soluble and

insoluble dietary fiber in foods: collaborative study. I Assoc Off Anal

Chem 1992;75:395-416.

79. Monro IA. Dietary fiber pectic substances: source of discrepancy

between methods of fiber analysis. I Food Comp Anal 1991;4:88-99.

80. Theander 0, Aman p, Westerlund E, Graham H. The Uppsala method

for rapid analysis of total dietary fiber. In: Furda I, Brine Ci, eds. New

developments in dietary fiber. New York: Plenum Press, 1990:273-81.

81. Englyst HN, Cummings JH, Wood R. Determination of dietary fibre in

cereals and cereal products-collaborative trials. III. Study of further

simplified procedures. I Assoc PubI Analysts 1987;25:73-110.

82. Schweizer TF, Walter E, Venetz P. Collaborative study for the enzy-

matic, gravimetric determination of total dietary fibre in foods. Mitt

Geb Lebensmittelunters Hyg 1988;79:57-68.

83. Hollman PCH, Boenke A, Wagstaffe P1. The certification of major

components and major elements in five food reference materials.

Fresenius J Anal Chem 1993;345:174-9.

CLASSIFICATION OF FOOD CARBOHYDRATES 937S

84. Berry CS. Resistant starch: formation and measurement of starch that

survives exhaustive digestion with amylolytic enzymes during the

determination of dietary fibre. J Cereal Sci 1986;4:301-14.

85. Englyst HN, Kingman SM, Cummings JH. Classification and mea-

surement of nutritionally important starch fractions. Eur J Clin Nutr

1992;46:S33-50.

86. Muir JG, O’Dea K. Measurement of resistant starch: factors affecting

the amount of starch escaping digestion in vitro. Am I Clin Nutr

1992;56:123-7.

87. Bj#{246}rckI, Nyman M, Pedersen B, SiljestrOm M, Asp N-G, Eggum BO.

On the digestibility of starch in wheat bread-studies in vitro and in

vivo. I Cereal Sci 1986;4:1-11.

88. Cummings JH, Englyst HN. What is dietary fibre? Trends Food Sci

Technol 1991;1:99-103.

89. Berggren A, Bj#{246}rckI, Nyman M, Eggum BO. Short-chain fatty acid

content and pH in caecum of rats given various sources of carbohy-

drates. I Sci Food Agnic 1993;63:397-406.

90. Anonymous. Rules and regulations. Fed Regist 1993;58:2095-6.

91. Anonymous. Council directive of 24 September 1990 on nutrition

labelling for foodstuffs (90/496/EEC). Off I Eur Communities 1990;

L276:40-4.92. Granfeldt Y, Bj#{246}rckI, Drews A, Tovar I. An in vitro procedure based

on chewing to predict metabolic response to starch in cereal and

legume products. Eur I Clin Nutr 1992;46:649-60.