Section II

Methods and Conventions of Nutrient Analysis

Supported by AOAC INTERNATIONAL

This session was chaired by Dr Doreen Clark, Managing Director of Analchem Bioassay, Sydney. The keynote address AOAC INTERNATIONAL-Validated Methods for Nutrient Analysis — Method Availability and Method Needs was given by J. DeVries. This was followed by papers on Analysis and Classification of Digestible and Undigestible Carbohydrates by N-G. Asp, Recent Developments in the Determination of Water-soluble Vitamins in Food—Impact on the Use of Food Composition Tables for the Calculation of Vitamin Intakes by P.M. Finglas, Update on the Analysis of Total Lipids, Fatty Acids and Sterols in Foods by A.J. Sinclair and Conventions for the Expression of Analytical Data by D.A.T. Southgate. All of these papers are published on the following pages.

G.R. Beecher, F. Khachik and J.T. Vanderslice did not provide a print version of their presentation Recent Advances in the Analysis of Fat-Soluble Vitamins in Foods. They can be contacted at the Nutrient Composition Laboratory, Beltsville Human Nutrition Research Center, ARS/USDA, Beltsville MD 20705, USA for further information.

AOAC INTERNATIONALValidated Methods for Nutrient Analysis — Method Availability and Method Needs

Jonathan W. DeVries

Medallion Laboratories, General Mills Inc., 9000 Plymouth Ave No., Minneapolis, MN 55427, USA

Adequate analytical methods for nutrients in foods, food ingredients, and food products are the basic first step in determining the nutritional adequacy of a food supply. Whatever the ultimate use of nutrition data, i.e. consumer education via the food label, or databases for nutrient and deficiency disease studies, the assay used to provide the data must determine the analyte of interest adequately. AOAC INTERNATIONAL (formerly the Association of Official Agricultural Chemists then the Association of Official Analytical Chemists) has been systematically validating methods for nutrition analysis for over 100 years. This validation includes a complete peer review system, with study of the method in multiple laboratories and multilevel review of the study results to assure adequacy of a proposed method for its intended purpose. With the passage in the USA of the Nutrition Labeling and Education Act, concern arose regarding the availability and adequacy of validated analytical methods to meet the requirements of the labeling act. A special AOAC task force with members drawn from regulatory agencies, the food industry, academia, and analytical suppliers was formed to address the concerns. In this paper, results of the task force assessment of adequacy of current Official Methods for nutrition analysis are presented. Method matrix combinations where updated methods or method modifications are needed are covered. In addition, a number of special issues addressed by the task force relating to the analysis of fat, moisture, and carbohydrate, reference materials (certified and in-house), and the methods validation process are discussed.

Adequate analytical methods for nutrients in foods, food ingredients, and food products are the basic first step in determining the nutritional adequacy of a food supply. Whether the nutrition data are ultimately used to inform consumers with information on the food label, or to build databases to study correlations between nutrient(s) and deficiency diseases, the assay used to provide the data must determine the analyte of interest adequately. AOAC INTERNATIONAL (formerly the Association of Official Agricultural Chemists and then the Association of Official Analytical Chemists) has been systematically validating methods for nutrition analysis for over 100 years. These validated methods provide competent laboratories with a means of supplying dependable data for nutrition labels of databases regarding the nutrition content of foods and food products.

This paper covers three areas, first, the history of AOAC INTERNATIONAL, and its processes and criteria for validation and acceptance of a method as an AOAC Official Method; second, recent activities of AOAC carried out in response to the recently proposed Nutrition Labeling and Education Act (NLEA)(1) in the US; and third, ideas for a methods validation scheme that might be used to improve the method validation process for foods, providing better comparative data and more rugged methods for laboratory use.

• History and Procedures of AOAC INTERNATIONAL

In 1884, a group of regulatory agricultural chemists formed the Association of Official Agricultural Chemists to adopt uniform methods for the analysis of fertilizers. The collaborative study was adopted as a means of validating methods and evaluating their performance. One of the active participants during the early years of the Association, Dr. Harvey W. Wiley carried out extensive studies on the adulteration of foods and drugs using AOAC methods. This ultimately led to the passage of the US Federal Pure Food and Drug Act of 1906. By 1912, the Association had begun publishing its validated methods as Official Methods in USDA bulletins. By 1920, the volume of validated methods had grown to the point where it warranted its own volume, and the Official Methods of Analysis was established. It has been revised and updated every five years since then (2).

In 1965, the name of the association was changed to Association of Official Analytical Chemists. As the Association's activities grew to include microbiologists and other scientists, the membership base became international in scope. As a result, the name was updated to AOAC INTERNATIONAL in 1991. Membership now includes scientists from many fields worldwide interested in improving analytical methodology and results.

Method validation under AOAC INTERNATIONAL auspices includes a complete peer review system, with study of the method in multiple laboratories and multilevel review of the study results to assure adequacy of a proposed method for its intended purpose. The key to rugged effective validated methods in AOAC INTERNATIONAL lies with the Associate Referee. The Associate Referee is appointed on a recommendation of a methods committee or a General Referee on the basis of the Associate Referee's expertise in an analytical area, i.e. active in methods development work, actively carrying out work assignments or projects relating to the analyte of interest, etc. Quite frequently the Associate Referee develops an analytical method to meet a need, or through knowledge of the literature selects an applicable method for study. After a requisite number of laboratories have been found to carry out a collaborative study, the Associate Referee distributes the methodology and samples. After the collaborative study is complete, the Associate Referee collects the data, develops a study report and submits a recommendation for method adoption to the Association. Assisting the Associate Referee is the General Referee, appointed on the basis of expertise and experience in broad analytical areas, who brings this broad knowledge base to bear on the study and its results.

When the General Referee and the Associate Referee agree that a method performs sufficiently well (see below for a discussion of criteria) to be considered as an Official Method, the method is submitted to an AOAC statistician and a safety advisor for review. Upon completion of these reviews, the method is sent to an appropriate Methods Committee for review and recommendation regarding Official Status. Methods committees are constituted of members chosen for their broad expertise in a given analytical area such as Food Nutrition, Food Toxins, or Drug Residues. Recommendation to the Official Methods Board to adopt a method as Official First Action requires agreement of two-thirds of the members of the Methods Committee. If members of the Methods Committee raise significant questions with regard to the method or its performance, the method cannot be recommended for Official Status until those questions have been addressed by the Associate Referee. Upon recommendation from the Methods Committee, the method is considered for Official Status by the Official Methods Board. The Board reviews the actions taken on the method, the review process, and assures consistency between methods and between methods committee reviews. If the method is given “First Action Official Status”, it is published in the Official Methods of Analysis. After “First Action” status for two years, methods which have no unresolved negative comments or issues can be considered for “Final Action Status”, a status achieved through ballot of the entire AOAC INTERNATIONAL membership. There is no difference in the Official Status of Methods, whether “First Action” or “Final Action”. “Final Action” only indicates that a method has withstood some test of time with no substantive questions raised regarding its performance. Any method achieving Official Status through the AOAC process has had both substantial performance testing in multiple laboratories and peer review by scientists who are experts in the analytical area. In addition, it has had intense scrutiny by scientists in related endeavors.

Criteria for validation of a method for Official Status are well established (3). The method must be submitted to participating laboratories written exactly as it is intended to be run. Participating laboratories are expected to run the method exactly as written. For a given collaborative study, participation by no fewer than eight laboratories analyzing a minimum of five sample materials is required for quantitative methods. For qualitative methods, no fewer than 15 laboratories analyzing a minimum of two analyte levels per matrix, five samples per level, and five negative controls are required. Obviously in both cases participation by more laboratories and the inclusion of more samples is encouraged. In extenuating circumstances, for example a particular method being considered has significant regulatory or commercial importance, but can only be carried out in five laboratories anywhere because only they have key instrumentation special consideration is given. Obviously, such circumstances are rare.

After the collaborative study is complete, statistical outliers (laboratories and/or data points) are removed (3). Rejection of data from more than two-ninths of the laboratories (without a valid explanation such as failure to follow the method) is basis for the rejection of the method as being insufficiently rugged to be adequate for intended purpose. Method performance (in statistical terms) will vary depending on analyte, matrix, and/or analyte quantity. Ultimately, therefore, a method must be accepted based on its performance in collaborative study as judged by scientific peer review by experts from government, academia, industry and other organizations. These experts are cognizant of the ultimate use of the methods being validated and judge adequacy for intended purpose. Working together in concert through AOAC INTERNATIONAL, these experts have produced high quality methods for the analytical community to use.

• Methods Needs for Nutrition Labeling

With the recent passage in the USA of the Nutrition Labeling and Education Act, concern arose amongst food consumers, producers, regulators, and laboratories providing nutrition analytical services, regarding the availability and adequacy of validated analytical methods to meet the requirements of the labeling act. The act was passed by the US Congress in November of 1990. The act required the US Food and Drug Administration to promulgate proposed regulations for nutrition labeling of nearly all foods sold in the US. The US Department of Agriculture, although not legally required to do so, initiated activities to adopt labeling regulations essentially equivalent to those of the USFDA. The proposed regulations of November 1991, were open for public comment for a number of months, with final regulations due in November of 1992. The final regulations were actually issued in January 1993, with an effective date for new label implementation of May 8, 1994 (July 8, 1994 for products under USDA jurisdiction).

The NLEA will have a significant impact on industry, consumers, and government agencies. It is estimated that it will cost industry upwards of $1.5 billion for the relabeling required, an estimated $1500 per product for small firms and $900 per product for large firms. Analytical costs will probably range from $750 for the 40 per cent of US foods that need label changes to $1800 for the 60 per cent of foods that had not been previously labeled. Research and development costs for products that will be modified somewhat for marketing advantage under the provisions of the act are hard to estimate, but run anywhere from $20,000 to $400,000 per product. Typically two to five months will be needed to redesign and print new packages. For consumers the cost of relabeling will be passed along in higher product prices. No money has been allocated for “Education”, so it is expected that significant consumer confusion will exist after the label changes occur. Governmental agencies will incur extra costs for interpretation, analysis, and enforcement of the act.

The effective date for NLEA was May 8, 1994, however other aspects of labeling had different effective dates, i.e. juice labeling in May, 1993, health claims in May 1993, and metric weight declarations in February, 1994. The NLEA now mandates nutrition labeling of most products and allows specified uses of nutrient descriptors and health claims related to nutrition.

Label format(s) is(are) rigidly specified under the NLEA (e.g. Figure 1). Mandatory declarations include calories, calories from fat, total fat, saturated fat, cholesterol, sodium, total carbohydrate, dietary fiber, sugars, protein, vitamin A, vitamin C, calcium, and iron. Voluntary declaration is allowed for calories from saturated fat, polyunsaturated fat, monounsaturated fat, stearic acid (USDA products), potassium, soluble fiber, insoluble fiber, sugar alcohols, other carbohydrates, thiamin (vitamin B₁), riboflavin (vitamin B₂), niacin, vitamin D, vitamin E, folate, cyanocobalamin (vitamin B₁₂), phosphorus, iodine, magnesium, zinc, copper, biotin, and pantothenic acid.

Are You Ready for New Food Labels?

NATIONAL FOOD PROCESSORS ASSOCIATION
in cooperation with FDA and FSIS

Figure 1. Example of food label conforming with NLEA requirement.

Labels will list the quantity of a given nutrient, along with a percentage of a daily recommended dietary intake value guideline for the consumer to use for comparison. The percentage of daily value is determined against either a Reference Daily Intake (RDI) value (typically for micronutrients) or against a Daily Reference Value (typically for macronutrients). For example, the daily reference value (based on 2000 calories/day) for fat is 65 g, for saturated fat is 20 g, for cholesterol is 300 mg, and for dietary fiber is 25 g. To encourage consistency in reporting of daily values, reference amounts relating to serving sizes have been published for common food items, e.g. 30 g for ready to eat cereals and cookies, 55 g for cake. Label serving sizes are to be in common household units, e.g. cups, teaspoons etc.

Nutrient claims can be made regarding the food product. However, if fat, saturated fat, cholesterol, or sodium exceed certain levels, this must be disclosed on the package label along with the nutrient claim. Adequate analytical methods are obviously needed to assure compliance both with the spirit of the nutrient claim, as well as to monitor the disclosure level compliance.

For added nutrients (referred to as Class I nutrients), the nutrient must be present at 100 per cent or greater than declared. For naturally occurring nutrients, (Class II), the nutrient must be present at a level at least 80 per cent or greater than declared, but less than or equal to 120 per cent of declared. Examples of nutrients that must be greater than 80 per cent of declared are dietary fiber and potassium. Examples of nutrients that must be less than 120 per cent of declared are fat, saturated fat, and sugar. Analytical variability is taken into account for enforcement, so well-characterized validated methods are necessary for compliance monitoring.

• AOAC Response to Nutrition Labeling Needs

To deal with concerns regarding availability of adequate methods to meet the needs of NLEA, a special task force of the AOAC with members drawn from regulatory agencies, the food industry, academia, and analytical suppliers was formed. The objectives of the task force were to: determine which Official Methods are adequate to meet current nutrition labeling analysis requirements; determine which Official Methods need revisions or modifications to meet current nutrition labeling analysis requirements; determine which nutrient-matrix combinations require the development and validation of Official Methods; propose means by which AOAC INTERNATIONAL can supply needed methods and/or modifications; identify means by which reference materials might be incorporated into AOAC Official Methods and into the validation process for AOAC Official Methods, further assuring the quality and performance of those methods.

The task force began informally at the AOAC INTERNATIONAL Annual Meeting in 1991, and was formally appointed by the board of directors in December of that year. Efforts were initiated immediately to obtain feedback regarding the status of Official Methods used for nutrition labeling. A survey was conducted of laboratories carrying out nutrition analysis and using AOAC methods. An information gathering session was also held in March of 1992. A number of task force meetings were held in the succeeding months to carry out the assigned objectives and fulfill the task force's mission.

Under the proposed nutrition labeling regulations, up to 54 nutrition-related items were either required or could be placed on the label, everything from A (ash) to Z (zinc). To organize the task of evaluating methods for these analytes, the task force divided foods into 20 different matrix groups that were felt at the time to cover the scope of foods and food products. This resulted in 1080 analyte-matrix combinations to be assessed regarding availability of adequate methods. Individual committee members took upon themselves assignments to review AOAC methods on an analyte-matrix basis. After this preliminary review was done, the entire task force, along with aid solicited from others, reviewed the assessments of the individual members. The analyte-matrix grid of adequate methods began to fill in. As the task force progressed, the information being generated was regularly reported in The Referee to keep the AOAC membership informed of progress and to allow feedback. For example the assessment of adequate methods under the proposed regulations was published in the July 1992 issue of The Referee (4).

Initial review of adequate methods under the proposed regulations, indicated that 947 of the 1080 possible matrix-analyte combinations had adequate methods. This meant that 88 per cent of the methods needs were addressed. In some cases, the Official Methods were deemed adequate for the need, but newer technologies can be brought to bear on the analyte-matrix combination to provide better methods at this point in time. An example might be vitamin A. The Carr-Price (5) method provides adequate results for labeling purposes, however most laboratories today would rather use liquid chromatography (e.g. (6)) and avoid handling the corrosive antimony trichloride. Therefore, although the task force accepted the adequacy of the Carr-Price method, it is recommending that validation of liquid chromatography methods be undertaken.

As the list of adequate methods was being generated, a complementary list of methods in need of validation or revision was also developed. This was published in October 1992 (7) to alert members of methods needs.

Special Nutrition Labeling Issues

As the task force evaluated methods for nutrition analysis, a number of issues were raised, in particular, methods for fat, dietary fiber, moisture, carbohydrates, standards and reference materials for Official Methods, and the need for a clear-cut means of determining if a particular Official Method is applicable to all foods. Subcommittees of the task force were formed to address each of these issues.

Fat has traditionally been analyzed by a variety of methods depending upon matrix, analyst carrying out the analysis, and intended use of the resulting data. Typically, the result was dependent upon determination of some solvent-soluble (solvents varied depending on the method) fraction of the food being analyzed. The task force realized that a single concise definition for fat was needed. AOAC INTERNATIONAL does not set definitions for nutrients, but provides validated analytical methods to quantify defined nutrients. Therefore, the subcommittee recommended, and the task force concurred, that the regulatory agencies, the USDA and FDA, adopt a single concise definition for fat. The agencies responded by adopting a definition of fat as the sum of the fatty acids (regardless of source) in the food, expressed as triglycerides (8). This concise definition provides a “gold standard” for evaluating fat analysis methods in the future.

The carbohydrates subcommittee determined that methods for total, soluble, and insoluble dietary fiber are adequate. Sugar methods, in particular the liquid chromatography methods with defatting steps, while adequate, should be further studied to assure validity across a broader matrix base. Complex carbohydrates as a nutrition label item had been included in the labeling proposal, but eliminated from the final regulations due to the lack of a clear definition of the nutrient, and lack of analytical methods to measure it. The subcommittee (and the task force) recommends a concise definition for complex carbohydrates be adopted and has committed AOAC to validating appropriate methodology when a definition is adopted.

A complete listing of moisture methods, along with their characteristics has been published by the moisture subcommittee (9). As with complex carbohydrates, a clearer definition of moisture will be helpful in validating more concise methodology for this analyte.

The subcommittee on reference materials published a listing of commercially available reference materials for the nutrients requiring mandatory labeling in August, 1992 (10). The subcommittee further went on to publish Guidelines for the Preparation of Inhouse Quality Assurance Materials in the May, 1993 issue of The Referee (11). Recognizing that assuring an adequate supply of reference materials was an ongoing task, requiring significant follow-up long after the nutrition labeling task force would be disbanded, the task force supported the formation of the first technical division of AOAC INTERNATIONAL, namely the Technical Division on Reference Materials. This division will continue the efforts initiated through the task force and will expand to reference materials beyond food nutrition. This division already has over 125 members and held its first annual meeting in conjunction with the AOAC INTERNATIONAL Annual Meeting in July, 1993.

• Method Validation Needs

After the final regulations for Nutrition Labeling in the US were issued by the USDA (12) and the USFDA (13), the task force reassessed methods adequacy and needs. The updated listings were published in the March (14) and April (15), 1993 issues of The Referee, respectively. In particular, methods and/or collaborative studies are needed for β-carotene, biotin, sugar alcohols, sugars (verification for certain matrices), cholesterol, copper, cyanocobalamin, defatting of samples for dietary fiber, fat (total, saturated, monounsaturated, and stearic acid), folate, iodine, niacin (microbiological method), pantothenate, protein (eliminating mercury use), pyridoxine, tryptophan (microbiological method), vitamin A, vitamin C (where erythorbate is present), and vitamin E. Some of these nutrients do have adequate methods, however, the methods are in need of modernization and therefore are recommended for further study.

• The Food Triangle as a Systematic Approach to Method Validation

A question that arose during the task force deliberations was: How does one ascertain with reasonable confidence that a method is applicable to all foods without a substantial history of trouble-free application to a wide variety of food samples? Clearly, a defined systematic approach might be helpful to assure method ruggedness across all food types while minimizing the analyst's efforts in assessing the method. The task force Subcommittee on Definition of Foods for Analytical Purposes has proposed an approach that is currently being considered by the Foods committees and the Official Methods Board (16).

The idea of requiring a collaborative study of 40 or more samples can be very discouraging, both for the associate referee organizing the study and for potential participants. There are five macronutrient components of any given food that have a significant impact on the performance of a method, no matter what the analyte being measured. The macronutrients impact analysis of various analytes by causing extraction difficulties or analyte interferences. The five macronutrients are moisture, ash, protein, fat, and carbohydrate. Moisture of nearly all samples can be adjusted if the level affects an assay. Water can be added, or the sample dried. Ash content of a sample usually has little effect on assays, particularly for organic nutrients. Therefore, the remaining three macronutrients, fat, protein, and carbohydrate have the major impact on the effectiveness of an analytical method. In a picture of a triangle with fat, protein, and moisture at the apices, all food samples will fit somewhere on that triangle, assuming the sum of fat, protein, and carbohydrate is normalized to 100 per cent, and these components are expressed as a percentage thereof. For example, a sample with 10 per cent fat, 30 per cent carbohydrate, and 10 per cent protein will have normalized values of 20 per cent fat, 60 per cent carbohydrate, and 20 per cent protein.

The triangle can be split equally in nine subtriangles, with any particular nutrient lying between 0–33 per cent, 33–67 per cent, and 67–100 per cent, respectively. By choosing 18 samples (two from each subtriangle), the analyst would be reasonably certain of covering foods characteristic of most foods. To develop further confidence in a method, samples taken from a subtriangle can be purposefully chosen to represent particular characteristics, i.e. for the 67–100 per cent carbohydrate subsection, a high fiber and a high starch sample might be used. For the 67–100 per cent fat section, a milk or animal fat and a vegetable fat might be chosen. The system could be applied to any nutrient being analyzed by using a Youden pairing technique [closely matched sample pairs as opposed to blind duplicates] (3) for determination of within laboratory variability for the analyte of interest. If difficulty is experienced with getting acceptable results for the method in question for samples from certain subtriangles, this information could be quite helpful for understanding and delineating the cause of the ineffectiveness. A similar approach had earlier been suggested for reference materials for nutrition analysis (17). The concept is extended by Tanner et al. (18).

• Conclusion

The task force has completed its objectives and reported the results of its deliberations on an ongoing basis in The Referee, the official house organ of AOAC. The final report has been published (19). The task force disbanded at the July 1993 Annual Meeting of AOAC INTERNATIONAL.

• References

(1)   Anon. (Nov 27, 1991) Federal Register 56, no 229, 60301–60891

(2)   Official Methods of Analysis, (1920, 1925, 1930, 1935, 1940, 1945, 1950, 1955, 1960, 1965, 1970, 1975, 1980, 1985, 1990) Association of Official Analytical Chemists, Arlington, VA

(3)   Manual for the Development, Study, Review, and Approval Process for AOAC Official Methods (1993) AOAC INTERNATIONAL, Arlington, VA

(4)   Nutrient Labeling Task Force (1992) The Referee 16, 1, 7–12

(5)   Official Methods of Analysis (1990) 15th Ed., secs 1045–1047, AOAC, Arlington, VA

(6)   DeVries, J.W. (1985) in Methods of Vitamin Assay, J. Augustin, B.P. Klein, D. Becker, & P.B. Venugopal, (Eds.), John Wiley and Sons, New York, NY

(7)   Nutrient Labeling Task Force (1992) The Referee 16, 5–10

(8)   Anon. (1993) Federal Register 58, 2086–2093

(9)   Anon. (1993) The Referee 17, 6–9

(10) Anon. (1992) The Referee 16, 4–5

(11) Anon. (1993) The Referee 17, 6–8

(12) Anon. (1993) Federal Register 58, 631–2063

(13) Anon. (1993) Federal Register 58, 2065–2964

(14) Nutrient Labeling Task Force (1993) The Referee 17, 6–10

(15) Nutrient Labeling Task Force (1993) The Referee 17, 6–8

(16) Anon. (1993) The Referee 17, 1, 6–7

(17) Southgate, D.A.T. (1987) Fres. J. Anal. Chem. 326, 660–664

(18) Tanner, J.T., Wolf, W.R., & Horwitz, W. (1995) in Quality and Accessibility Food-Related Data, H. Greenfield (Ed.), AOAC INTERNATIONAL, Arlington, VA, pp. 99–104

(19) Methods of Analysis for Nutrition Labeling (1993), D.M. Sullivan, & D.E. Carpenter, (Eds.), AOAC INTERNATIONAL, Arlington, VA, pp. 33–83

Analysis and Classification of Digestible and Undigestible Carbohydrates

Nils-Georg Asp

Department of Applied Nutrition and Food Chemistry, Lund University, Chemical Center, PO Box 124, S-221 00 Lund, Sweden

The current interest in the nutritional properties of various food carbohydrates has increased the demand for compositional data. The small-intestinal digestibility is the most important nutritional property. The digestible carbohydrates provide glucose to body tissues, whereas the undigestible carbohydrates are partially fermented and provide fermentation substrate and bulk in the colon. Mono-, di- and oligosaccharides, as well as polyols, can be determined with specific enzymatic methods, but gas-liquid chromatography (GLC) and especially liquid chromatography (LC) methods are preferable when a range of sugars is to be analyzed. Dietary fiber determination should aim to differentiate between digestible (“available”) and undigestible (“unavailable”) carbohydrates. Gravimetric and component analysis methods are complementary for different purposes. Resistant starch, i.e. undigestible starch, as well as lignin should be included in the dietary fiber. Starch is preferably analyzed with specific enzymatic methods, that should have the same cut-off as for starch removal in dietary fiber analysis.

Dietary guidelines in Western countries recommend that the carbohydrate intake be increased to at least 55–60 per cent of the energy (1). In diets consumed in other parts of the world carbohydrates may contribute more than 70 per cent of energy. Originally, the carbohydrate recommendations came as a consequence of the fat and protein recommendations. In recent years, however, the nutritional importance of the carbohydrates as such has been more and more emphasized, and new developments call for a more nutritional classification of the different food carbohydrates as a basis for more specific recommendations about intake.

Labeling of foods regarding carbohydrate content is a separate, but closely related issue. Most carbohydrate content figures on food labels are still being calculated “by difference”, i.e. the material remaining after moisture, ash, fat and protein determinations. In view of the quite variable nutritional effects of different carbohydrates, this is unsatisfactory.

A thorough characterization of the various digestible and undigestible carbohydrate fractions is required whenever investigating the physiological properties of a carbohydrate-containing food or diet. Compositional data on the food carbohydrates are also essential in epidemiological research.

Table I shows the food carbohydrates that are quantitatively most important. Starch generally occurs in the largest amount in diets, followed by sucrose and — when milk products are consumed — lactose. Glucose, fructose and sucrose are present naturally in fruits, berries and vegetables, and also may be added as refined sugars. Polyols and fructans such as inulin are increasingly used as lowcalorie bulking agents, as is polydextrose. Since many dietary guidelines recommend a limited use of refined sugars (sucrose, fructose, corn syrups, high fructose corn syrup etc.), the contribution of such “extrinsic” sugars is of special interest. However, it is not possible analytically to distinguish between “extrinsic” sugars and the “intrinsic” sugars present naturally.

Table I. Main food carbohydrates

• Nutritional Properties of Food Carbohydrates

Small Intestinal Digestibility

Carbohydrates that are digested and absorbed in the small intestine provide glucose, fructose and galactose to body tissues. Undigestible carbohydrates, on the other hand, are delivered to the large intestine and fermented to various extents. The main products of this anaerobic fermentation are acetate, propionate and butyrate. Acetate and propionate are absorbed and metabolized in peripheral tissues and the liver, respectively, and their possible effects on carbohydrate and lipid metabolism are currently investigated. Butyrate is an important source of energy for the epithelial cells of the large intestine itself, and may be important in protecting against colonic cancer (for review, see e.g. 2). Various fermentable carbohydrates give different proportions of these fermentation products (3).

As early as 1929, McCance and Lawrence emphasized small-intestinal digestibility by introducing the term “available” carbohydrates, based on determination of starch and digestible sugars (4). Correspondingly the term “unavailable carbohydrates” was used for cellulose, non-cellulose polysaccharides and lignin (5). Within the European Community legislators have defined “carbohydrates” as digestible (“metabolizable”) carbohydrates and including polyols (6).

Dietary fiber was first defined as the remnants of plant cell-walls not digested in the small intestine (7). With this definition it constitutes the non-starch polysaccharides of the plant cell-walls, but also undigestible protein, inorganic material, tannins, cutin etc. The redefinition by Trowell et al. (8) restricted the definition to polysaccharides and lignin, but enlarged it to include all undigestible polysaccharides. There were two reasons for this: First, purified polysaccharides such as pectins and gums were frequently used to study the physiological effects of dietary fiber constituents, and second, cellwall polysaccharides could not easily be differentiated analytically from undigestible polysaccharides from other sources (9).

Dietary fiber includes a large number of polysaccharides with quite different properties, both from the chemical and physiological points of view. Insoluble, lignified types of dietary fiber have the most prominent fecal bulking effect due to their resistance to fermentation, whereas soluble, gel-forming polysaccharides are most efficient in lowering blood cholesterol and blood glucose after a meal (for review, see e.g. 10).

In the large intestine, the dietary fiber polysaccharides, oligosaccharides and resistant starches are fermented to various extent with production of acetate, propionate and butyrate in various proportions.

Rate of Carbohydrate Digestion and Absorption

In diabetes, patients have long been advised to choose carbohydrates that are slowly digested and absorbed, giving a limited and sustained blood glucose elevation with minimum insulin requirement. This has been demonstrated to improve the metabolic control in maturity onset diabetes (NIDDM) (11). Generally it has been believed that starch is slowly digested and absorbed due to its high molecular weight (“complex carbohydrate”). Sucrose and other low-molecular weight carbohydrates (“simple sugars”), on the other hand, have been regarded as rapidly absorbed. It is remarkable that this view has been so prevalent in spite of the lack of scientific evidence. On the contrary, data accumulated in the 70s and 80s showing that the height and shape of the blood glucose curve could be quite different after the intake of different foods, and that these differences were unrelated to the molecular size of the carbohydrates. Among the low molecular weight carbohydrates, fructose gives a very low glycemic response, and sucrose is intermediate between glucose and fructose (12). Starchy foods are found across the whole range of “slow” to “rapid” (for review see e.g. 13).

A number of other carbohydrate and food properties determining the glycemic response have now been identified and include gel-forming types of dietary fiber, degree of gelatinization and other properties of the starch, cellular structure and gross structure (14). There is increasing evidence that “slow” properties of food carbohydrates may also be beneficial in relation to blood lipid levels, satiety, physical performance and dental caries (12). The cariogenic properties of foods have mainly been related to added sucrose, but there is evidence that other fermentable carbohydrates are important as well. Even starch can lower dental plaque pH, and this property is related to the availability of starch for enzymatic degradation in the mouth (15).

The term “complex carbohydrates” was used in 1977 by the U.S. Senate Committee on Nutrition and Human Needs (McGovern Report) without any exact definition, but meaning in practice digestible polymeric carbohydrate, i.e. starch. From what has been said above, it is obvious that starch has no nutritional advantage per se, and therefore, the term “complex carbohydrates” is questionable. In Britain, it was reintroduced to mean starch and non-starch polysaccharides (16). Although the grouping together of starch and non-starch polysaccharides may be relevant from the chemistry point of view, its usefulness for nutritional classification is questionable. Recommendations regarding complex carbohydrate intake (1) are very complicated to interpret in terms of foods even for experts.

• Overview of Analytical Methods

Mono-, Di- and Oligosaccharides including Polyols

Depending on the food matrix to be analyzed, an extraction of the free sugars may be necessary. Aqueous ethanol is preferable due to the toxicity of methanol, that was frequently used earlier. A final ethanol concentration of at least 80 per cent (v/v) should be used to avoid extraction of polysaccharides. However, some polysaccharides such as inulin and pectic substances may have considerable solubility also at this alcohol concentration. Some sugars, especially lactose, have a slow rate of dissolution and limited solubility, and may need lower alcohol concentration (e.g. 50 per cent v/v) at extraction with a final increase to precipitate polysaccharides (17).

The International Union of Pure and Applied Chemistry (IUPAC) defines oligosaccharides as having less than ten monomeric residues. In practice, however, oligosaccharides are defined as carbohydrates soluble or extractable in aqueous ethanol. Precipitation in 78–80 per cent ethanol (or dialysis/ultrafiltration) is generally used to separate oligosaccharides and starch degradation products from polysaccharides in dietary fiber analysis (9).

The α-galactosides in leguminous seeds and fructans in onions, artichokes etc. are the quantitatively most important groups of naturally occurring undigestible oligosaccharides. Inulin is a nonstarch polysaccharide, but it is also not determined as dietary fiber with any of the current methods in spite of a degree of polymerization (DP) of 30 or more. Arabans in sugar beet fiber are another example of a dietary fiber polysaccharide that is extremely soluble in alcohol due to its extensive branching (18). Polydextrose also falls into this category.

Physical methods, such as polarimetry, refractive index, or density are still useful in pure systems, e.g. in sugar production control. Methods based on the reduction of copper salts, and colorimetric methods based on condensation reactions with anthrone, orcinol and carbazol, can also be used in well known systems (17).

Table II. Resistant starch definition and determination

The enzymatic procedures based on specific, highly purified enzymes have been instrumental in providing means of specific and precise analysis of carbohydrates in mixtures without high capital investments. GLC and LC procedures, on the other hand, are preferable when a number of different carbohydrates are to be determined simultaneously. LC analysis has for long been hampered by the relative insensitivity of refractory index detectors. However, this has been overcome by systems using amperometric detection (17).

Polysaccharides

Starch. Starch is the predominant dietary carbohydrate, and the only polysaccharide that is digestible in the human small intestine. Enzymatic hydrolysis and specific glucose assay is the method of choice today because of glucose liberation also from for instance α-glucans at acid hydrolysis. However, the enzymes used have to be checked for contaminating activities (19).

A heat stable amylase (Termamyl) in a combined gelatinization and hydrolysis step, has turned out to be particularly useful (e.g. 20).

Resistant Starch. Resistant starch was first defined as a starch fraction resisting amylase hydrolysis unless it was first solubilized in KOH or DMSO (21). It is now generally defined as the sum of starch and products of starch degradation not absorbed in the small intestine (22). It is then an undigestible polysaccharide, that should be included in the dietary fiber. There are different forms of resistant starch: Physically enclosed starch, raw α-type starch granules, retrograded amylose and chemically or physically modified food starches (23).

Originally, resistant starch was measured as the starch remaining associated with the dietary fiber if solubilizing agents were not used (21, 24). This type of resistant starch has been identified as mainly retrograded amylose (25, 26). Methods capable of measuring also other forms of resistant starch (27, 28) are currently evaluated against in vivo measurements of starch absorption within the European FLAIR Concerted Action program (EURESTA).

Chemically modified food starches (29) and dry heated starches (30) are degraded by amylases to fragments that are soluble in alcohol. These fragments are neither determined as starch with enzymatic methods, nor as dietary fiber (30). Methods for resistant starch analysis are summarized in Table II.

Dietary Fiber. Dietary fiber is analyzed according to two different principles (for review, see 9). In the gravimetric methods the non-fiber components are removed and a residue weighed. The residue can be analyzed for e.g. protein and ash, and corrections used accordingly. The crude fiber and detergent fiber methods belong to this category. Enzymatic gravimetric methods such as those approved by the AOAC (31, 32) use alcohol precipitation to recover soluble fiber components and can be used to measure total dietary fiber (TDF) or soluble and insoluble components separately. Correction for protein and ash in the fiber residue is needed.

Table III. Advantages and disadvantages of various method for dietary fiber analysis (9).

^a but residue can be analyzed
^b if hydrolysis is complete

The component analysis methods use more or less specific determination of monomeric constituents, that are then summed to yield a total fiber value. As in gravimetric methods, soluble and insoluble components can be determined separately. It should be noted that the solubility of polysaccharides is method dependent and determined by the temperature, time and pH conditions used.

The Southgate procedure (33) employs colorimetric methods to determine hexoses, pentoses and uronic acids. The methods of Theander et al. (34) and Englyst et al. (35) use GLC for neutral sugar components and a colorimetric assay for uronic acids. LC determination is gaining in popularity. A colorimetric measurement of reducing sugars has been introduced as an alternative to the GLC determination by Englyst et al. (35).

Advantages and disadvantages of the two different ways of analyzing dietary fiber are summarized in Table III. Enzymatic gravimetric methods are simple and robust with no requirement of advanced equipment. There is a risk of overestimating the fiber content if other components remain in the residue. However, this can be analyzed for any such contaminating components. Colorimetric methods can also be inflated by unspecific reactions. Specific, GLC or LC measurements on the other hand, require complete hydrolysis and quantitative recovery of monomers after hydrolysis of the polysaccharides. Incomplete hydrolysis or losses due to decomposition of monomers will lead to underestimation (for review, see 9).

The current component analysis methods employ acid hydrolysis and corrections for hydrolysis losses of the different components. As in amino acid analysis, conditions for the hydrolysis have to be chosen to obtain an optimal compromise between hydrolysis yield and monomer degradation. Quantitative hydrolysis yield is particularly difficult to obtain with acidic polysaccharides due to the high stability of glycosyl uronic acid linkages towards acid hydrolysis. This fact and the more rapid degradation of monomeric uronic acids at acidic condition are reasons why colorimetric methods are preferred for uronic acid determination (9).

Collaborative Studies of Dietary Fiber Analysis. A number of collaborative studies of enzymatic, gravimetric dietary fiber determination has been carried out within the AOAC. The component analysis method of Englyst and co-workers has been tested in studies carried out by the Ministry of Agriculture, Food and Fisheries in the UK (MAFF). The studies reported prior to 1990 were reviewed and compared (9). They show gradually improved performance with typical mean reproducibility (R95) values of 2–3 for both the gravimetric methods approved by the AOAC and for the Englyst method. An AOAC study with the method of Theander et al. is about to be finished. The best performance reported so far is a Swiss study with the enzymatic gravimetric AOAC method (R95=1.0–1.1) (36). An R95 value of 2.0 at a dietary fiber content of 10 g/100 g means that 19 out of 20 single determinations coming from various laboratories would fall in the range 9–11 g/100 g.

There are few formal collaborative studies covering more than one method. Usually, studies have included just one or a few laboratories running a different methods, which makes strict intermethod comparison difficult. In a recent study coordinated by the European Community Bureau of Reference (BCR) dietary fiber values with the AOAC method could be certified for three different materials. Indicative values only could be given for the Englyst GLC and colorimetric methods, but means with these methods were similar to those obtained with the AOAC method (37).

For most foods, estimates of total dietary fiber with the enzymatic gravimetric method of the AOAC or according to Theander et al. (both including the retrograded amylose type of resistant starch and lignin) would not be significantly different from estimates of nonstarch polysaccharides with the Englyst methods. This means that the confidence intervals for the different methods overlap (9). It should be noted also that two collaborative studies have shown consistently higher values with the colorimetric Englyst method than with the original GLC variety (9). Only in foods with particularly high levels of resistant starch of the retrograded amylose type, or lignin, would Englyst values be expected to be significantly lower than estimates with methods including these components.

Delimitation problems in definition and analysis of dietary fiber have been much focused on the inclusion or not of resistant starch and lignin. Equally important, however, is the delimitation towards components that are not precipitated in the 70–80 per cent (v/v) ethanol used in all the methods to separate water-soluble fiber components. Inulin is an undigestible polysaccharide that is not precipitated and therefore not recovered in any of the methods. Polydextrose is another undigestible oligo-polysaccharide also not determined as dietary fiber. As discussed above, these components should be grouped together with the dietary fiber rather than with digestible carbohydrates. The same is true for the undigestible oligosaccharides. Specific enzymatic or HPLC methods then have to be employed for these components.

Difference methods are unspecific and accumulate analytical errors from the fat, protein, ash and moisture determinations. When keeping these limitations in mind, however, these methods are capable of giving a reasonable estimate of “total available carbohydrates” in many foods if the dietary fiber is measured, e.g. with the enzymatic gravimetric AOAC method. Difference calculations are also useful in the laboratory to check the standardization of methods for proximate analysis.

• References

(1)   World Health Organization (1990) Diet, Nutrition, and the Prevention of Chronic Disease. Technical Report Series, 797, WHO, Copenhagen

(2)   Rémésy, C., Demigné C., & Morand, C. (1992) in Dietary Fibre— A Component of Food, T.F. Schweizer & C.A. Edwards (Eds.), Springer-Verlag, London, pp. 137– 150

(3)   Edwards, C.A., & Rowland, I. (1992) in Dietary Fibre—A Component of Food, T.F. Schweizer & C.A. Edwards (Eds.), Springer-Verlag, London, pp. 119–136

(4)   McCance, R.A., & Lawrence, R.D. (1929) Medical Research Council Special Report Series No. 135, London, HMSO

(5)   McCance, R.A., & Widdowson, E.M. (1940) Medical Research Council Special Report Series No. 235, London, HMSO

(6)   Official Journal of the European Community (1990) Directive NOL 276/40

(7)   Trowell, H.C. (1972) Am. J. Clin. Nutr. 25, 926–932

(8)   Trowell, H.C., Southgate, D.A.T., Wolever, T.M.S., Leeds, A.R., Gassull, M.A., & Jenkins, D.J.A. (1976) Lancet, i, 967

(9)   Asp, N.-G., Schweizer, T.F., Southgate, D.A.T., & Theander, O. (1992) in Dietary Fibre- A Component of Food, T.F. Schweizer & C.A. Edwards (Eds.), Springer-Verlag, London, pp. 57–101

(10) Asp, N.-G., Björck, I., & Nyman, M. (1993) Carbohydrate Polymers 21, 183–187

(11) Brand Miller, J.C. (1994) Am. J. Clin. Nutr. 59 (Suppl), 747S–752S

(12) Truswell, A.S. (1992) Eur. J. Clin. Nutr. 46 (Suppl 2), S91–S101

(13) Würsch, P. (1989) World Rev. Nutr. Diet. 60, 199–256

(14) Björck, I., Granfeldt, Y., Liljeberg, H., Tovar, J., & Asp, N.-G. (1994) Am. J. Clin. Nutr. 59 (Suppl), 699S– 705S

(15) Lingström, P., Holm, J., Birkhed, D., & Björck, I. (1989) Scand. J. Dent. Res. 97, 392–400

(16) The British Nutrition Foundation. (1990) Complex Carbohydrates in Foods, Chapman and Hall, London, pp. 1–164

(17) Greenfield, H., & Southgate, D.A.T. (1992) Food Composition Data Production, Management and Use, Elsevier Applied Science, London, pp. 94–104

(18) Asp, N.-G. (1990) in New Developments in Dietary Fiber, I. Furda & C.J. Brine (Eds.), Plenum Press, New York, pp. 227–236

(19) Åman, P., & Graham, H. (1987) J. Agric. Food Chem. 35, 704–709

(20) Holm, J., Björck, I., Drews, A., & Asp, N.-G. (1986) Starch/Stärke, 38, 224–226

(21) Englyst, H.N., Wiggins, H.S., & Cummings, J.H. (1982) Analyst, 107, 307–318

(22) Asp, N.-G. (1992) Eur. J. Clin. Nutr. 46 (Suppl), S1

(23) Asp, N.-G., & Björck, I. (1992) Trends Food Sci. Technol. 3, 111– 114

(24) Johansson, C.-G., Siljeström, M., & Asp, N.-G. (1984) Lebensm. Unters Forsch. 179, 24–28

(25) Siljeström, M., Eliasson, A.-C., & Björck, I. (1989) Starch/Stärke, 41, 147–151

(26) Russel, P.L., Berry, C.S., & Greenwell, P. (1989) J. Cereal. Sci. 9, 1–15

(27) Berry, C.S. (1986) J. Cereal Sci. 4, 301–314

(28) Englyst, H.N., & Cummings, J.H. (1990) in New Development in Dietary Fiber, Furda, I., Brine, J. (Eds.), Plenum Press, New York, NY, pp. 205–225

(29) Björck, I., Gunnarsson, A., & Östergård, K. (1989) Starch/Stärke 41, 128–134

(30) Siljeström, M., Björck, I., & Westerlund, E. (1989) Starch/Stärke 41, 95–100

(31) Prosky, L., Asp, N.-G., Schweizer, T.F., DeVries, J.W., & Furda, I. (1988) J. Assoc. Off. Anal. Chem. 71, 1017–1023

(32) Lee, S., Prosky, L., & DeVries, J. (1992) J. Assoc. Off. Anal. Chem. 75, 395–416

(33) Southgate, D.A.T. (1969) J. Sci. Food Agric. 20, 331–335

(34) Theander, O., Åman, P., Westerlund, E., & Graham, H. (1990) in New Developments in Dietary Fiber. I. Furda & J. Brine (Eds.), Plenum Press, New York, NY, pp. 273–281

(35) Englyst, H.N., Cummings, J.H., & Wood, R. (1987) J. Assoc. Publ. Analysts 25, 73–110

(36) Schweizer, T.F., Walter, E., & Venetz, P. (1988) Mitteilungen aus dem Gebeit der Lebensmitteluntersuchung und Hygiene 79, 57–68

(37) Hollman, P.C.H., Boenke, A., & Wagstaffe, P.J. (1993) Fres J. Anal. Chem. 345, 174–179

(38) Englyst, H.N., Kingman, S.M., & Cummings, J.H. (1992) Eur. J. Clin. Nutr. 46, S33–S50