Journal of Food Composition and Analysis 22 (2009) 446452

Contents lists available at ScienceDirect

Journal of Food Composition and Analysis

journal homepage: www.elsevier.com/locate/jfca

Report

New information on carbohydrates in the Brazilian Food Composition Database

E.W. Menezes *, E.B. Giuntini, M.C.T. Dan, F.M. Lajolo *
Departamento de Alimentos e Nutricao Experimental, Faculdade de Ciencias Farmaceuticas, Universidade de Sao Paulo, 05508-900 Sao Paulo, SP, Brazil

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 6 May 2008
Received in revised form 15 January 2009
Accepted 2 February 2009

Foods that contain unavailable carbohydrates may lower the risks for some non-transmissible chronic
diseases because of the potential benets provided by the products of colonic fermentation. On the other
hand, foods that are sources of available carbohydrates may have higher energy value and increase the
post-prandial glycemic response. The biomarker glycemic index and the resulting glycemic load may be
used to classify foods according to their potential to increase blood glucose. Information about glycemic
index and glycemic load may be useful in diet therapy. Currently, food composition tables in Brazil do not
provide data for individually analyzed carbohydrates even though some quality data are available in
scientic publications. The objectives of this work were to produce and compile information about the
concentration of individual carbohydrates in foods and their glycemic responses and to disseminate this
information through the Brazilian Food Composition Database (TBCA-USP). The glycemic index and
glycemic load of foods were evaluated in healthy individuals. Concentrations of available carbohydrates
(soluble sugars and available starch) and unavailable carbohydrates (dietary ber, resistant starch, betaglucans, fructans) were quantied by ofcial methods, and other national data were compiled. TBCA-USP
(http://www.fcf.usp.br/tabela), which is used by professionals and the population in general, now offers
both chemical and biological information for carbohydrates.
2009 Elsevier Inc. All rights reserved.

Keywords:
Available carbohydrates
Unavailable carbohydrates
Carbohydrate analysis
Dietary ber
Fructans
Soluble sugars
Glycemic response
TBCA-USP
Food composition databases

1. Introduction
Carbohydrates are important for human health because they
are a primary source of energy and have benecial effects
resulting mainly from dietary ber and other unavailable
carbohydrates that are resistant to digestion. Colonic microbiota
ferments, partially or totally, the carbohydrates that escape
digestion and absorption in the small intestine and produce,
mainly, short chain fatty acids. The products of fermentation
cause several benecial effects both locally and systemically (Elia
and Cummings, 2007; Gray, 2006). Some unavailable carbohydrates (dietary ber, resistant starch, beta-glucans and fructans)
may lower risks for some non-transmissible chronic diseases
(NTCD), such as diabetes, obesity, cardiovascular diseases and
cancer (Gray, 2006; WHO/FAO, 2003).
One important role of carbohydrates is the glycemic response,
which differs depending on the food source and on extrinsic and
intrinsic factors that affect carbohydrate digestion and absorption.
The biomarker glycemic index (GI) and the resulting glycemic load
(GL) may be used to classify foods according to their potential to
increase blood glucose. The GI distinguishes food carbohydrates

* Corresponding author. Tel.: +55 11 3091 36 24; fax: +55 11 3815 44 10.
E-mail address: wenzelde@usp.br (E.W. Menezes).
0889-1575/$ see front matter 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jfca.2009.02.001

based on their potential to increase the glycemic response after the

ingestion of a xed amount of available carbohydrates (50 g or
25 g) in relation to the ingestion of the same amount of
carbohydrates from a standard food (white bread or glucose)
(Jenkins et al., 1981). The GL reects dietary glycemic impact and is
calculated by multiplying the GI of each food (using glucose as the
standard) by the amount of available carbohydrate in the
consumed food portion (Salmeron et al., 1997).
Studies on GI are important because of potential physiologic
and therapeutic effects of low GI diets for healthy individuals, as
well as for those who are obese, diabetic and/or hyperlipidemic
(Venn and Green, 2007; Brand-Miller, 2003; WHO/FAO, 2003;
Augustin et al., 2002). The global strategy on diet, physical activity
and health, proposed by WHO (2004), aims to control and prevent
NTCDs, and considers, among other things, that consumers have
the right to receive exact, standardized and understandable
information about food product content, so that it is possible for
them to make healthy food choices. The use of GI as a tool to
facilitate food planning through glycemic control has been used
and encouraged in some countries (SUGIRS, 2007; Danone
Vitapole/FAO, 2001; Frost and Dornhorst, 2000).
In Brazil, there is not much information about the different
carbohydrate fractions in foods or about the biological availability
of carbohydrates when consumed. At the same time, there is a high
incidence of NTCDs, which were responsible, in 2001, for 62% of all

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

deaths and 39% of all hospitalizations registered in the Brazilian

public health system (Brasil, 2005; Wang et al., 2002; Narayan
et al., 2000). Cardiovascular diseases accounted for 31% of all
known deaths (Brasil, 2005).
The Brazilian Food Composition Database (TBCA-USP), developed and updated by the Brazilian Network of Food Composition
Data Systems (BRASILFOODS), has aimed to provide quality
information about Brazilian food composition since its creation
in 1998 (USP, 1998). BRASILFOODS, housed in the University of Sao
Paulo, Department of Food Science and Experimental Nutrition of
Pharmaceutical Science School (Menezes et al., 2002), is connected
to INFOODS (International Network Food Data Systems) and
LATINFOODS (Latin American Network of Food Composition Data
Systems) and has been cooperating with the Tabla de Composicion
de Alimentos de America Latina (FAO/LATINFOODS, 2000). TBCAUSP currently has a total of 1838 food items, 1200 of which contain
proximate composition of raw, cooked and industrialized foods
and recipes (Giuntini et al., 2006). In this database, the
carbohydrates were estimated by difference (available carbohydrates) to calculate energy, and data on dietary ber was
determined by an enzymatic-gravimetric method (Lee et al.,
1992). However, it has been suggested that the presentation of
carbohydrates by difference should be eliminated from food
composition tables (Greeneld and Southgate, 2003; FAO/WHO,
1998). These estimates are not sufcient for users needs in view of
the fact that the physiological effects of each carbohydrate fraction
may be important in lowering risks for NTCDs (Gray, 2006; FAO/
WHO, 1998). Quantifying each individual carbohydrate and
evaluating foods on a case-by-case basis is the current tendency
of worldwide databases (Greeneld and Southgate, 2003; Li et al.,
2002; Monro and Burlingame, 1996). The goal for TBCA-USP is to
have data for individual carbohydrate fractions for foods that are
sources of carbohydrates. Having data for available and unavailable carbohydrates is also important for the accomplishment of the
global strategy on diet, physical activity and health.
2. Objectives
The objectives of this work were to produce and compile
information about the concentration of the different carbohydrates
in foods and their glycemic effects and to disseminate this
information through the TBCA-USP.
3. Methodology
The data for individual carbohydrates were obtained both by
direct chemical analysis and by compilation. All the information
related to the glycemic responses of Brazilian foods was obtained
from studies using healthy volunteers at the University of Sao
Paulo by BRASILFOODS.

447

The second part of the form contained different spreadsheets to

be lled in according to the analytical methods used. The INFOODS
tagnames for nutrients were adopted to improve data interchange
(Monro and Burlingame, 1996; Klensin, 1992).
The content of carbohydrates in foods can be determined
according to several procedures, which vary according to the type
of carbohydrate and its methodology of analysis (Monro and
Burlingame, 1996). Total carbohydrates can be calculated by the
aggregation of compounds (the sum of available carbohydrate
values analyzed individually by specic methods) or by difference
(100 gthe sum of proximal composition values per 100 g).
Carbohydrates calculated by difference were not useful for this
database; however, this information was documented and stored
for possible consultation. Available carbohydrates can be analyzed
individually, i.e., monosaccharides, disaccharides and polysaccharides obtained by direct analysis. Dietary ber can be determined
according to analytical methods (gravimetric). Monosaccharide
compounds of dietary ber can be obtained by non-gravimetric
methods. Other compounds such as polysaccharides and cell wall
fractions that would also be classied as dietary bers can be
obtained by specic analysis.
The form for compilation of food composition data was
structured to accommodate the main patterns and methods of
analysis for carbohydrates used by national researches and in the
national data publications. Sources of all information (laboratory
or bibliographic reference) are documented in the TBCA-USP.
3.2. Studied foods
The foods selected for the evaluation of glycemic response are
all habitually consumed by the Brazilian population according to a
Brazilian research on family budget, done between 2002 and 2003
(IBGE, 2004). The products were provided by CEAGESP, a company
that centralizes food marketing in the state of Sao Paulo and
receives foods that are produced in the whole country. Depending
on the food characteristics, the most commonly consumed kinds or
preparations were selected (for example, maturation state, raw or
cooked, time of cooking, among others); all this information was
documented when identifying the food. For carbohydrate determination, all samples (raw and cooked ones) were lyophilized and
stored at 20 8C. Foods that needed to be cooked were prepared on
the day of the clinical assay. After this, the samples were
lyophilized, milled to a particle size of <0.250 mm and frozen,
for later completion of chemical analysis (triplicate samples). In
the case of commercial bread, a composite sample was made from
samples of the same kinds of bread produced in three different
weeks, with consideration for the number of industries and their
representativeness in the market.
In the case of data compiled from published sources, the food
was identied from information available in the publication.
3.3. Quantity of food consumed

3.1. Data compilation

To facilitate data compilation and to guarantee data quality and
standardization, the Form for Compilation of Food Composition Data,
developed by BRASILFOODS (Menezes et al., 2005) was updated.
The spreadsheets for identication of foods, proximal composition
and analytical quality control were not modied. However, a
spreadsheet for carbohydrates was expanded to accommodate the
wide range of individual carbohydrates in foods and to include data
on glycemic response.
The rst part of the form included the food identication. The
INFOODS guidelines (Truswell et al., 1991) to describe foods were
adopted with modications made by LATINFOODS (FAO/LATINFOODS, 2004; FAO, 1995; Menezes et al., 2005).

The quantity of food to be consumed by the healthy volunteers

was calculated based on its content of available carbohydrate. This
content is calculated by the sum of the concentration of soluble
sugars (mainly glucose, fructose and sucrose) and available starch
(the difference between total starch and resistant starch) in the
food. Depending on the sample characteristics, the quantity of
available carbohydrate to be ingested was 25 g or 50 g; the same
amount of available carbohydrate was ingested in the standard
food (white bread) (Brouns et al., 2005). Some fruits had to be
consumed dried due to their reduced concentration of available
carbohydrates. In the case of commercial breads, the amount
consumed was a proportion of the bread produced in each of three
industries.

448

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

3.4. Evaluation of glycemic response of foods from healthy volunteers

The GI of foods were evaluated from healthy volunteers and the
experimental protocol was submitted and approved by the Ethical
Research Committee of the School of Pharmaceutical Science,
University of Sao Paulo (USP), according to the rules of the
National Committee for Ethical Research of the Brazilian Health
Ministry (CONEP/MS). All subjects signed informed consent to
participate.
The GI and GL produced by the foods were determined
according to the FAO protocol (Brouns et al., 2005; FAO/WHO,
1998), which was validated by the Laboratory of Chemistry,
Biochemistry and Molecular Biology/FCF/USP/BRASILFOODS. The
in vivo study was done with healthy volunteers (616 volunteers/
food), both genders, aged between 21 and 39 years. The volunteers
had body mass indices (BMI) ranging from 18.5 to 24.9 kg/m2, were
classied as eutrophic according to the World Health Organization
(WHO, 1997), and were without previously diagnosed diabetes or
diabetes in the family. The volunteers came to the laboratory once
a week after a 10- to 12-h overnight fast. White bread (standard
food) was tested twice, in the rst two weeks. On subsequent
weeks (until the 5th week), the volunteers ingested a portion of
each test food. Each portion of food contained exactly 25 g or 50 g
of available carbohydrates, depending on the food prole. The
volunteers had 10-15 min to ingest each portion with 250 mL of
water. Blood glucose was determined in each subject fasted (time
zero) and after ingesting a test food. Blood samples were taken at
15, 30, 45, 60, 90 and 120 min after food intake to construct a
glycemic response curve (Brouns et al., 2005; Wolever et al., 1991).
Glucose was measured in capillary whole blood using a blood
glucose meter (Accu-Check Go, Roche Diagnostics1). The GI of each
sample was estimated by the relation between the areas under the
curves produced by the test food and the bread (standard-100%).
The GI is expressed as the mean  SEM. The GL of each food was
calculated according to the following equation: GL = GI (glucose as
standard)  available carbohydrate (g) per portion  1/100 (Ludwig,
2003; Liu et al., 2000). To obtain GI values, with glucose as standard,
from the GI values with white bread as standard, the results of white
bread must be multiplied by 0.7 (SUGIRS, 2007). For each food, the
portions used were the ones recommended by the Brazilian Ministry
of Health (Brasil, 2003).
3.5. Chemical methods
The samples were analyzed in triplicate and the results are
expressed as mean  SD.
3.5.1. Proximate composition
Total protein content was determined by a semi-micro Kjeldahl
method (AOAC, 1995). The conversion factor used was %N  6.25
or specic for each food (Greeneld and Southgate, 2003). Ash
content was determined by incineration in a mufe furnace at
520 8C. The moisture content of the sample was calculated based
on weight loss after the sample was heated in an oven at 105 8C. Fat
was determined by the Soxhlet method (AOAC, 1995).
3.5.2. Soluble sugars
Soluble sugars were extracted three times with 80% ethanol at
80 8C. The supernatants were combined and ethanol was evaporated
under vacuum. The residues were reconstituted with water, ltered
through 0.22 mm membrane lters and analyzed by high performance anion exchange chromatography with pulsed amperometric
detection (HPAE-PAD). The chromatographic analysis was performed on a Dionex DX 500 instrument equipped with a PAD system
(ED 40). The analytical column employed was Carbopac PA1
(250  4 mm, 5 mm particle size). The mobile phase was 18 mM

NaOH and the ow rate was kept constant at 1.0 mL/min. Injections
(25 mL) were made by using an AS 500 autosampler. The sugars
glucose, fructose, galactose, maltose and sucrose from Sigma were
used as standard reference material.
3.5.3. Total starch
Starch from samples was solubilized in 0.5 M NaOH. After
neutralization with 0.5 M acetic acid, an aliquot was precipitated
with 80% ethanol. The precipitated starch was hydrolyzed with
amyloglucosidase (28 U/mL) and the resultant glucose determined
by the glucose-oxidase-peroxidase-ABTS (2,20 -Azino-di-[3-ethylbenzthiazoline] sulfonate) system, as described by Cordenunsi and
Lajolo (1995).
In the case of commercial bread, which contains our improver
(to expand fermentation), it was necessary to use a different
method (acid hydrolysis with anthrone) (McCready et al., 1950). In
this method, starch from samples (0.5 g) was solubilized in 0.5 M
NaOH. After neutralization with 0.5 M acetic acid, an aliquot was
washed with 80% ethanol (2x). Starch was precipitated and the
supernatant was discarded. The dried precipitate was incubated in
water bath at 100 8C with sulfuric acid 0.5 N for 1 h. Aliquots were
incubated with anthrone (0.1%) for 10 min in water bath at 100 8C
for reductor sugars determination.
Starch from Sigma Chemical Co. (St. Louis, MO) was used as
standard reference material and total starch was calculated as
glucose  0.9.
3.5.4. Resistant starch
The resistant starch analysis was done based on the method
described by McCleary and Monaghan (2002) and McCleary et al.
(2002). A sample of boiled beans was used as in-house reference
material. Glucose was quantied on the supernatants with GOD/
POD/ABTS mixture as described for total starch. The available
starch was calculated by the difference between the content of
total starch and resistant starch.
3.5.5. Fructans
The fructans quantication was based on the treatment of the
sample with amyloglucosidase and fructanase according to
Hoebregs (1997) and on the determination of soluble sugars by
high performance liquid chromatography with pulse amperometric detector (HPLCPAD), as described above. Inulin from
Sigma Chemical Co. (St. Louis, MO) was used as standard reference
material.
3.5.6. Dietary ber
The dietary ber (total, soluble and insoluble) was quantied by
the enzymatic-gravimetric method according to Lee et al. (1992).
For foods with a high content of resistant starch, the methodology
for dietary ber was done with modications proposed by
McCleary and Rossiter (2004) so as not to include resistant starch
in the dietary ber fraction; the resistant starch was analyzed
separately. For foods with a high content of fructans, the
methodology of dietary ber was done with modications
proposed by Prosky and Hoebregs (1999) so as not to include
fructans in the dietary ber fraction; the content of fructans was
analyzed separately.
4. Results and discussion
The present paper presents data on some foods to illustrate how
the information for carbohydrates and glycemic response is
structured in TBCA-USP. For each food, food identication and
data references (Table 1), GI and GL produced by the intake of
Brazilian foods (Table 2) and content of different carbohydrates in
those foods (Table 3) are described.

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

449

Table 1
Examples of food identication and data source of the database.
Entry number

Complete food identication

Source

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Bread, white, French

Pineapple, pulp, dried, Ananas comosus, Perola
Pineapple, pulp, raw, Ananas comosus, Perola
Pineapple, core, dried, Ananas comosus, Perola
Pineapple, core, raw, Ananas comosus, Perola
Pineapple, pulp and core, raw, Ananas comosus, Perola
Seed of Brazilian pine, native, raw, Araucaria angustifolia
Seed of Brazilian pine, native, boiled without seed coat, 60 min, Araucaria angustifolia
Seed of Brazilian pine, native, boiled with seed coat, 90 min, Araucaria angustifolia
Rice, brown, boiled, 28 min, Orysa sativa L.
Rice, white, boiled, 28 min, Orysa sativa L.
Sweet corn, yellow, boiled, 45 min, Zea mays
Corn meal, cooked, 90 min, Zea mays
Corn meal, cooked, 90 min, stored, 20 8C/30 days, Zea mays
Pasta, spaghetti, boiled, 20 min
Potatoes, Inglesa, boiled with skin, 33 min, Solanum tuberosun
Peas, split, dry, boiled, 30 min, Pisum sativum L.
Beans, Carioca, dry, soaked overnight, boiled with pressure, 36 min, without juice, Phaseolus vulgaris L.
Beans, Carioca, dry, soaked overnight, boiled with pressure, 36 min, stored, 20 8C/30 days, Phaseolus vulgaris L.
Lentils, dry, boiled, 21 min, Lens culinaria
Chickpeas, boiled, 70 min, Cicer arietinum L.
Chickpeas, boiled, 70 min, stored, 20 8C/30 days, Cicer arietinum L.
Bread, white, loaf, Pullman/Plus Vita
Bread, carrot, loaf, low-calorie, Plus Vita
Bread, seven grains, loaf, low-calorie, Plus Vita

6
1
1
1
1
1
2
2
2
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
6
6
6

5
5
5
4,
4,
4,
5
5
4,
4,
5
4,
4,

5
5
5

5
5
5
5

Source: 1 Cordenunsi et al. (in press); 2 Cordenunsi et al. (2004); 3 Rosin et al. (2002); 4 Carreira et al. (2004); 5 Menezes et al. (2004); 6 Clariant Project (2007).
Boiled, cooked in water without pressure. Cooked, heat without water.

4.1. Food identication and information source

Complete information about the food and reference to the
sources of information for each are described in Table 1. Each food
is described individually, providing detailed information on

variety, species, maturation degree, season, time and kind of

cooking and storage, among others. This information is extremely
important because it can affect the nutrient content of the foods.
This detailed information is also relevant to the international
concern about biodiversity (Toledo and Burlingame, 2006).

Table 2
Examples of glycemic index (GI) and glycemic load (GL) of the database.
Entry
numbera

1
2
3
4
5
6
8
9
13
14
15
18
19
21
22
23
24
25
a

Short food
identication

Bread, white, French

Pineapple, pulp, dried
Pineapple, pulp, raw
Pineapple, core, dried
Pineapple, core, raw
Pineapple, pulp and
core, raw
Seed B. pine, boiled
without s.c.
Seed B. pine, boiled
with s.c.
Corn meal, cooked
Corn meal,
cooked, stored
Pasta, spaghetti, boiled
Beans, boiled
Beans, boiled, stored
Chickpeas, boiled
Chickpeas,
boiled, stored
Bread, white, loaf
Bread, carrot, loaf,
low-calorie
Bread, s. grains,
loaf, low-calorie

GI (%)b
(Bread = 100)

Available
carbohydrate
(g per portion)

GL
(per portion)d

GL
classication

50

20

14

90

10
100

2
11

1
7

L
L

Healthy, 6

50

16

47

Healthy, 6

50

16

M
H

52
68

Healthy, 10
Healthy, 10

150
150

21
21

11
14

M
M

62  5
38  4
18  2
34  4
22  2

L
L
L
L
L

43
27
13
24
15

Healthy,
Healthy,
Healthy,
Healthy,
Healthy,

10
10
10
10
10

180
170
170
170
170

45
25
24
38
33

19
7
3
9
5

M
L
L
L
L

97  13
67  13

H
L

68
46

Healthy, 16
Healthy, 16

50
50

21
14

14
7

M
L

60  11

42

Healthy, 16

50

14

GI
classication

GI (%)
(Glucose = 100)c

Volunteers
(kind, n)

100
93  6
93  6
95  7
95  7
94

H
H
H
H
H
H

70
65
65
67
67
66

Healthy, 16
Healthy, 7
Healthy, 7
Hhealthy, 7
Healthy, 7
Healthy, 7

62  2

43

67  5

74  8
96  7

Recommended
portion (g)

The complete description of foods is presented in Table 1.

GI (bread as standard) is expressed as mean %  SEM, except for food 6, which was calculated from the results of foods 3 and 5.
c
GI (glucose as standard) = GI (bread as standard)  0.7.
d
GI = GI (glucose as standard)  available carbohydrate (g) in the recommended portion  1/100. GI classication (bread as standard = 100): Low  75; Medium 75.1
94.9; High  95. GL classication (glucose as standard = 100): Low  10; Medium 10.119.9; High  20. Recommended portion according to the Brazilian legislation
ANVISA (Brasil, 2003). For foods 7, 10, 11, 12, 16, 17, 20, omitted in this table, the glycemic response of healthy volunteers was not evaluated.
b

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

450

Table 3
Examples of content of individual carbohydrates in foods (g/100 g of food as eaten)a of the database.
Entry
numberb

1
3
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Short food
identication

Moisture

Total
starch

Resistant
starch

Available
starch

Glucose

Fructose

Sucrose

Total
soluble
sugars

Total
available
carbohydrate

Soluble
dietary
ber

Insoluble
dietary
ber

Bread, white, French

Pineapple, pulp, raw
Pineapple, core, raw
Pineapple, pulp and
core, raw
Seed of B. pine, raw
Seed B. pine, boiled
without s.c.
Seed B. pine, boiled
with s.c.
Rice, brown, boiled
Rice, white, boiled
Sweet corn, yellow,
boiled
Corn meal, cooked
Corn meal, cooked,
stored
Pasta, spaghetti, boiled
Potatoes, boiled
with skin
Peas, boiled
Beans, boiled
Beans, boiled, stored
Lentils, dry, boiled
Chickpeas, boiled
Chickpeas, boiled,
stored
Bread, white, loaf
Bread, carrot, loaf,
low-calorie
Bread, s. grains, loaf,
low-calorie

30
86
78
85

43.2
0.1

2.32
0.00
0.00

40.8
0.1
0.0
0.1

0.6
1.5
0.7

0.8
1.7
0.9

5.4
9.4
5.8

0.1
6.8
12.6
7.4

40.9
6.9
12.6
7.5

0.0
0.1
0.0

0.8
1.7
0.9

0.8
1.9
0.9

0.1

50
50

36.3
34.5

3.27

31.2

2.3
0.6

0.1
0.0

0.1
0.1

2.4
0.6

2.4
31.9

0.6
0.6

4.3
5.2

4.9
5.7

50

34.5

3.27

31.2

0.6

0.0

0.1

0.6

31.9

0.6

5.2

5.7

75
66
77

20.4
29.9
19.8

0.65
0.83
0.9

19.8
29.1
18.9

0.1
0.0
0.0

0.0
0.0
0.1

0.1
0.0
0.0

0.1
0.0
0.1

19.9
29.1
19.0

0.2
0, 0
0.0

1.2
0.8
2.7

1.5
0.7
2.8

80
80

14.5
15.0

0.82
0.98

13.7
14.1

0.0

0.0

0.09

0.2
0.2

13.8
14.2

0, 0

1.1

1.2

70
85

25.5
12.5

0.67
0.5

24.9
12.0

0.0
0.0

0.0
0.0

0.04
0.13

0.1
0.2

24.8
12.2

0.5
0.5

0.5
0.9

1.1
1.3

66
62
62
68
59
64

20.7
15.9
16.1
19.1
24.4
21.2

1.55
1.78
2.66
1.61
2.39
2.29

19.2
14.1
13.4
17.5
22.0
18.9

0.0
0.0

0.0
0.0

0.19
0.37

4.4
7.2

4.7
9.0

0.0
0.0

0.15
0.36

19.4
14.5
13.8
17.6
22.4
19.3

0.3
1.8

0.0
0.0

0.2
0.4
0.4
0.2
0.4
0.4

0.5
0.5

4.4
5.4

4.9
5.7

36
41

41.2
30.1

2.29
2.75

38.9
27.3

1.3
0.5

1.7
0.6

0.05
0.03

3.0
1.0

41.9
28.3

2.6
10.3

41

31.1

2.29

27.3

0.3

0.3

0.04

0.6

27.9

8.9

Total
dietary
ber

Except for entry number 7expressed in raw base.

The complete description of foods is presented in Table 1. The values are expressed as mean (triplicate assays). This table includes only values with coefcients of
variation lower than 10%. Foods 2 and 4 were omitted because they are expressed in dry weight.
b

Information and data related to biodiversity can be important for

databases that specialize in health, agriculture, commerce or
research (Burlingame, 2004). For the TBCA-USP, analytical quality
information is valuable, even if it is from a specic cultivar.
Composite samples for specic products are prepared and
analyzed to guarantee representativeness in relation to their
consumption by the population. For commercial bread (items 23,
24 and 25 in Table 1), composite samples were prepared based on
the producers and the percentage consumption in the region.
4.2. Glycemic response of foods
High GI foods are those with GI  95% and low GI foods are
those with GI  75%, considering bread as standard with GI = 100%
(Menezes and Lajolo, 2006). The GL was calculated for each food
according to its GI and the amount of available carbohydrate
present per portion of food usually consumed by the population.
Considering glucose as the standard, foods were classied as low
GL (GL  10) or high GL (GL  20). Table 2 shows examples of GI
and GL produced by the intake of some Brazilian foods.
The items 1325, in Table 2, represent starchy foods that are
usually consumed by the Brazilian population and they were
studied in relation to proximal composition, individual carbohydrates and glycemic response, including information about storage
at 20 8C for 30 days. The original publication (Carreira et al., 2004)
showed that the GI and GL of legumes (items 18, 19, 21 and 22 in
Table 2) were signicantly reduced after storage under 20 8C for
30 days (p < 0.05).
Glycemic responses for pineapple (item 6 in Table 2) showed
that this fruit can be considered a high glycemic response food

(GI = 94%). This high GI value resulted from the high concentration
of soluble sugars and low concentrations of soluble dietary ber in
pineapple (Table 3). However, when the pineapple GL was
calculated, this fruit was considered a low GL food (GL = 7). For
pineapple, the GL proved to be the better tool for this food for
dietary planning because it expresses not only quality but also
quantity of the carbohydrates in a usual portion. This is one of the
reasons why international databases of glycemic response (FosterPowell et al., 2002), as well as TBCA-USP, express their results not
only as GI but also as GL.
The seed of Brazilian pine (items 8 and 9 in Table 2), although it
is a seasonal and regional food, represents an important
carbohydrate source of low GI (62 and 67%), just like spaghetti.
At the same time, the GL was also low (7-8 per serving size of 50 g),
while spaghetti (item 15 in Table 1; serving size of 180 g) has a
medium GL (GL = 19). The high content of dietary ber in the pine
seeds (11.4% dry weight) may cause difculties in relation to the
complete swelling and gelatinization of starch and hence, interfere
in the carbohydrate availability (Cordenunsi et al., 2004).
The evaluation of glycemic response involves a large number of
volunteers as well as high costs. For BRASILFOODS to overcome
these critical factors, partnerships have been formed between the
University and private industries. The results of the different kinds
of bread (items 2325 in Table 2) were obtained through
partnerships. It is well known that white bread produces high
post-prandial glycemic response, and that is the reason it is used as
a standard food. Therefore, industries have been developing
different kinds of bread that replace available carbohydrates with
unavailable ones to lower post-prandial glycemic responses. Other
results of glycemic response for industrialized foods have been

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

obtained and are being compiled, and evaluations of other foods

have been approved by the Ethical Research Committee of the
School of Pharmaceutical Science, University of Sao Paulo. Therefore, BRASILFOODS plans to expand data on the glycemic response
of industrialized foods.
The incidence of NTCDs has been growing year by year; the
forecast is that there will be an increase of 170% in the prevalence
of diabetes between 1995 and 2025 (King et al., 1998; Narayan
et al., 2000). Therefore, relevant information on foods that may
contribute to lowering the risk of NTCDs must be considered.
Hopefully, publishing the information on glycemic response and
individual carbohydrate of Brazilian foods will help improve this
negative prognosis of the population. TBCA-USP currently has data
of glycemic response for 41 foods.
4.3. Carbohydrate content
The form for compilation of food composition data considers
several methods of carbohydrates analysis; however, only a few of
them are effectively used by national researchers. The results
shown in Table 3 were obtained, mainly, from dietary ber
analysis, done by the enzymatic-gravimetric method with
separation of soluble and insoluble fractions.
Although the concentration of sucrose in foods is one of the
main causes for a high post-prandial glycemic response, data on
the mono- and disaccharide content of common foods are scarce.
This information should be included in Food Composition
Databases to improve food planning and help control glycemic
response (Miller et al., 1995; Jenkins et al., 2002).
Among the foods shown in Table 3, the pineapple (item 6) is
the only one that presents a high concentration of total soluble
sugars (7.4% in the pulp and core); however, the pineapple
presents a reduced concentration of total available carbohydrates, in comparison to other foods, which have high
concentrations of available starch, varying from 13.8% (boiled
and stored beans, item 19) to 41.0% (French white bread, item 1).
The different kinds of breads (items 2325) present concentrations that vary from 27.9% to 41.9% of available carbohydrates,
from 2.3% to 2.8% of resistant starch, and from 2.6% to 10.3% of
total dietary ber.
The seed of Brazilian pine (item 9 in Table 3) presents the
highest percentage of resistant starch (3.3%), but this is a typical
food from cold regions in Brazil and is consumed seasonally. Boiled
Carioca beans (item 18), which is a food habitually consumed in
Brazil, presents 1.8% of resistant starch, and when boiled and
stored (item 19), there was a signicant increase in the
concentration of resistant starch (2.7%). The same did not happen
for chickpeas, although both foods produced GIs signicantly more
reduced after storage under 20 8C for 30 days (Carreira et al.,
2004). Other researchers who studied legumes and other kinds of
foods also noticed resistant starch formation when the food was
stored under low temperatures (Tovar et al., 2002; Bravo et al.,
1998). Storage under low temperature resulted in lower glycemic
responses for the legumes. Fructans were analyzed only in the
pineapple pulp (item 3) and pineapple core (item 5) (Table 3). They
presented 0.2 g/100 g of food as eaten.
TBCA-USP currently has available information on individually
analyzed carbohydrates for 112 foods.
5. Final considerations
TBCA-USP offers chemical and biological information about
carbohydrates in Brazilian foods. The data on the individual
carbohydrates of 112 foods and the glycemic responses for 41
foods are important tools for accomplishing the aims of the global
strategy for a healthy diet.

451

Acknowledgements
The authors acknowledge the nancial support from CNPq, USP
and the participation by BRASILFOODS. We also acknowledge the
XI.18 and 106PI0297 CYTED/CNPq projects of international
cooperation for allowing us to use of data in the database.
References
Association of Ofcial Analytical Chemists, AOAC, 1995. Ofcial Methods of
Analysis, 16th ed., vol. 2. AOAC, Arlington, VA, USA.
Augustin, L.S., Franceschi, S., Jenkins, D.J.A., Kendall, C.W.C., La Vecchia, C., 2002.
Glycemic index in chronic disease: a review. European Journal of Clinical
Nutrition 56, 10491071.
Brand-Miller, J.C., 2003. Glycemic load and chronic disease. Nutrition Reviews 61
(5), 4955.
Brasil. Agencia Nacional de Vigilancia Sanitaria, 2003. Regulamento tecnico para
rotulagem nutricional obrigatoria de alimentos e bebidas embaladas. RDC 360,
23 de dezembro de 2003. Retrivied 2008-01-16 from http://www.anvisa.gov.br/
legis/.
Brasil. Ministerio da Saude, 2005. A vigilancia, o controle e a prevencao das
nico de
doencas cronicas nao-transmissveis: DCNT no contexto do Sistema U
Saude Brasileiro/Brasil. Ministerio da Saude. Braslia: Organizacao Pan-Americana da Saude, 80 p. Retrieved 2008-01-16 from http://bvsms.saude.gov.br/
bvs/publicacoes/ DCNT.pdf.
Bravo, L., Siddhuraju, P., Saura-Calixto, F., 1998. Effect of various processing methods on the in vitro starch digestibility and resistant starch content of Indian
pulses. Journal of Agriculture and Food Chemistry 46, 46674674.
Brouns, F., Bjorck, I., Frayn, K.N., Gibbs, A.L., Lang, V., Slama, G., Wolever, T.M.S.,
2005. Glycaemic index methodology. Nutrition Research Reviews 18, 145171.
Burlingame, B., 2004. Fostering quality data in Food Composition Databases: visions
for the future. Journal of Food Composition and Analysis 17, 251258.
Carreira, M.C., Lajolo, F.M., Menezes, E.W., 2004. Glycemic index: effect of food
storage under low temperature. Brazilian Archives of Biology and Technology
47 (4), 569574.
Clariant Project Glycemic response of different kinds of bread, 2007. Coordinator:
Elizabete W. Menezes, Financial support: Clariant S.A., Sao Paulo. Products:
Bimbo do Brasil.
Cordenunsi, B.R., Lajolo, F.M., 1995. Starch breakdown during banana ripening
sucrose synthase and sucrose-phosphate synthase. Journal of Agriculture and
Food Chemistry 43, 347351.
Cordenunsi, B.R., Menezes, E.W., Genovese, M.I., Colli, C., Souza, A.G., Lajolo, F.M.,
2004. Chemical composition and glycemic index of Brazilian pine (Araucaria
angustifolia) seeds. Journal of Agriculture and Food Chemistry 52 (11), 3412
3416.
Cordenunsi, B.R., Saura-Calixto, F., Diaz-Rubio, M.E., Zuleta, A., Tine, M.A., Buckeridge, M.S., Silva, J.B., Carpio, C., Giuntini, E.B., Menezes, E.W., Lajolo, F.M.
Carbohydrate composition of ripe pineapple (cv. Perola) and the glycemic
response in humans. Ci. Tecnol. Aliment, in press.
Danone Vitapole/FAO, 2001. Glycemic index and health: the quality of the evidence.
In: John Libbey Eurotext, Nutrition and Health Collection, France, 48 p.
Elia, M., Cummings, J.H., 2007. Physiological aspects of energy metabolism and
gastrointestinal effects of carbohydrates. European Journal of Clinical Nutrition
61 (Suppl. 1), S40S74.
Food and Agriculture Organization (FAO), 1995. INFORME del taller CTPD sobre
production y manejo de dados de composicion quimica de alimentos en
nutricion de America Latina. FAO/Ocina Regional para la America Latina y
el Caribe (NUT-60), Santiago.
Food and Agriculture Organization/Red Latinoamericana de Datos de
Composicion de Alimentos (FAO/LATINFOODS), 2000. Tabla de Composicion
de Alimentos de America Latina. Retrieved 2008-01-16 from http://
www.rlc.fao.org/bases/alimento.
Food and Agriculture Organization/Red Latinoamericana de Datos de Composicion
de Alimentos (FAO/LATINFOODS), 2004. Segunda Conferencia Eletronica Avaliacao da qualidade dos dados para bases de dados e tabelas de composicao
qumica de alimentos. 11 a 29 de novembro de 2004. Retrieved 2008-01-16
from http://www.rlc.fao.org/foro/latinfoods/.
Food and Agriculture Organization/World Health Organization (FAO/WHO), 1998.
Carbohydrates in human nutrition: Report of a Joint FAO/WHO Expert Consulation, April 1418, 1997, Food and Nutrition Paper, 66, FAO, Rome, 140 p.
Foster-Powell, K., Holt, S.H.A., Brand-Miller, J.C., 2002. International table of glycemic index and glycemic load values. American Journal of Clinical Nutrition 76,
556.
Frost, G., Dornhorst, A., 2000. The relevance of the glycaemic index to our understanding of dietary carbohydrates. Diabetic Medicine 17, 336345.
Giuntini, E.B., Lajolo, F.M., Menezes, E.W., 2006. Tabela Brasileira de Composicao de
Alimentos TBCA-USP: Versoes 3 e 4. Archivos Latinoamericanos de Nutricion 56
(4), 183203.
Gray, J., 2006. Dietary FiberDenition, Analysis, Physiology & Health. ILSI Europe,
Belgium.
Greeneld, H., Southgate, D.A.T., 2003. Food Composition Data: Production, Management and Use, 2nd ed. Roma: Food and Agriculture Organization of United
Nations (FAO), Roma, 288 p.

452

E.W. Menezes et al. / Journal of Food Composition and Analysis 22 (2009) 446452

Hoebregs, H., 1997. Fructans in food and food products, ion-exchange chromatographic method: collaborative study. Journal of AOAC International 80, 1029
1037.
Instituto Brasileiro de Geograa e Estatstica (IBGE), 2004. Pesquisa de Orcamentos
Familiares 20022003. Analise da disponibilidade domiciliar de alimentos e do
estado nutricional no Brasil. Ministerio do Planejamento, Orcamento e Gestao.
Rio de Janeiro.
Jenkins, D.J., Kendall, C.W., Augustin, L.S., Franceschi, S., Hamidi, M., Marchie, A.,
Jenkins, A.L., Axelsen, M., 2002. Glycemic index: overview of implications
in health and disease. American Journal of Clinical Nutrition 76, 266S
273S.
Jenkins, D.J.A., Wolever, T.M.S., Taylor, R.H., Barker, H., Fielder, H., Baldwin, J.M.,
Bowling, A.C., Newman, H.C., Jenkins, A.L., Goff, D.V., 1981. Glycemic index of
foods: a physiological basis for carbohydrates exchange. American Journal of
Clinical Nutrition 34, 362366.
King, H., Aubert, R.E., Herman, W.H., 1998. Global burden of diabetes. Diabetes Care
21, 14141431.
Klensin, J.C., 1992. INFOODS: Food Composition Data Interchange. United Nations:
University Press, Tokyo, Japan. Retrivied 2008-01-16 from http://www.unu.edu/
unupress/unupbooks/80734e/80734E00.htm.
Lee, S.C., Prosky, L., Devries, J.W., 1992. Determination of total, soluble and insoluble
dietary ber in foods. Enzymatic-gravimetric method, Mes-TRIS Buffer: collaborative study. Journal of AOAC International 75, 395416.
Li, B.W., Andrews, K.W., Pehrsson, P.R., 2002. Individual sugars, soluble, and
insoluble dietary ber contents of 70 high consumption foods. Journal of Food
Composition and Analysis 15, 715723.
Liu, S., Willet, W.C., Stampfer, M.J., Hu, F.B., Franz, M., Sampson, L., Hennekens, C.H.,
Manson, J.E., 2000. A prospective study of dietary glycemic load, carbohydraste
intake, and risk of coronary heart disease in US woman. American Journal of
Clinical Nutrition 71, 14551461.
Ludwig, D.S., 2003. Glycemic load comes of age. Journal of Nutrition 133, 2695
2696.
McCleary, B.V., McNally, M., Rossiter, P., 2002. Measurement of resistant starch by
enzymatic digestion in starch selected plant materials: collaborative study.
Journal of AOAC International 85 (5), 11031111.
McCleary, B.V., Monaghan, D.A., 2002. Measurement of resistant starch. Journal of
AOAC International 85 (3), 665675.
McCleary, B.V., Rossiter, P., 2004. Measurement of novel dietary bres. Journal of
AOAC International 87, 707711.
McCready, R.M., Guggolz, J., Silveira, V., Owens, H.S., 1950. Determination of starch
and amylase in vegetables. Application to peas. Analytical Chemistry 22, 1.156
1.158.
Menezes, E.W., Goncalves, F.A.R., Giuntini, E.B., Lajolo, F.M., 2002. Brazilian Food
Composition Database: Internet dissemination and others recent developments. Journal of Food Composition and Analysis 15 (4), 453464.
Menezes, E.W., Lajolo, F.M., 2006. Marcadores in vivo e in vitro de carboidratos. In:
Lajolo, F.M., Menezes, E.W. (Eds.) Carbohidratos en alimentos regionales iberoamericanos. Proyecto CYTED/CNPq XI.18. Composicion, Estructura, Propiedades Biologicas de Carbohidratos y su Utilizacion en Alimentos, EDUSP, Sao
Paulo, pp.309334.
Menezes, E.W., Melo, A.T., Lima, G., Lajolo, F.M., 2004. Carbohydrate measurement
and energy value of foods. Journal of Food Composition and Analysis 17 (34),
331338.

Menezes, E.W., Ratto, A.T., Giuntini, E.B., Lajolo, F.M., 2005. Composicao de alimentos: compilacao e uniformizacao de estruturas para intercambio de dados.
Brazilian Journal of Food Technology 8 (1), 2533.
Miller, J.B., Pang, F., Broomhead, L., 1995. The glycemic index of foods containing
sugarscomparison of foods with naturally occurring vs added sugars. British
Journal of Nutrition 73, 613623.
Monro, J., Burlingame, B., 1996. Carbohydrates and related food compounds:
INFOODS tagnames, meanings, and uses. Journal of Food Composition and
Analysis 9, 100118.
Narayan, K.M.U., Gregg, E.W., Fagot-Campagna, A., Engelgan, M.M., Vinicor, F., 2000.
Diabetesa common, growing, serious, costly and potentially preventable
public health problem. Diabetes Research Clinical Practice 50, S7784.
Prosky, L., Hoebregs, H., 1999. Methods to determine food inulin and oligofructose.
Journal Nutrition 129, S1418SS1423S.
Rosin, P.M., Lajolo, F.M., Menezes, E.W., 2002. Measurement and characterization of
dietary starches. Journal of Food Composition and Analysis 15 (4), 367377.
Salmeron, J., Manson, J.E., Stampfer, M.F., Colditz, G.A., Wing, A.L., Willett, W.C.,
1997. Dieatry ber, glycemic load, and risk of non-insulin-dependent diabetes
mellitus in women. Journal of the American Medical Association 277, 472477.
Sydney University Glycemic Index Research Service (SUGIRS), 2007. Glycemic
index. Retrieved 2008-01-16 from http://www.glycemicindex.com/.
Toledo, A., Burlingame, B., 2006. Biodiversity and nutrition: a common path toward
food security and sustainable development. Journal of Food Composition and
Analysis 19, 477483.
Tovar, J., Melito, C., Herrera, E., Rascon, A., Perez, E., 2002. Resistant starch formation
does not parallel syneresis tendency in different starch gels. Food Chemistry 76,
455459.
Truswell, A.S., Bateson, D.J., Madaglio, K.C., Pennington, J.A.T., Rand, W.M., Klensin,
J.C., 1991. INFOODS Guidelines for describing foods: a systematic approach to
describing foods to facilitate international exchange of food composition data.
Journal of Food Composition and Analysis 4, 1838.
Universidade de Sao Paulo (USP), 1998. Faculdade de Ciencias Farmaceuticas.
Departamento de Alimentos e Nutricao Experimental. Tabela Brasileira de
Composicao de Alimentos-USP. Version 4.1. Retrieved 2008-1-16 from
http://www.fcf.usp.br/tabela.
Venn, B.J., Green, T.J., 2007. Glycemic index and glycemic load: measurement issues
and their effect on diet-disease relationships. European Journal of Clinical
Nutrition 61 (Suppl. 1), S122S131.
Wang, Y., Monteiro, C.A., Popkin, B.M., 2002. Trends of obesity and underweight in
older children and adolescents in the United States, Brazil, China, and Russia.
American Journal of Clinical Nutrition 75 (6), 971977.
Wolever, T.M., Jenkins, D.J.A., Jenkins, A.L., Josse, R.G., 1991. The glycemic index:
methodology and clinical implications. American Journal of Clinical Nutrition
54, 846854.
World Health Organization (WHO), 1997. Lepidemie dobesite expose des millions
de personnes a dautres maladies. Communique OMS/46.
World Health Organization (WHO), 2004. Global strategy on diet, physical activity
and health. Fifty-Seventh World Health Assembly. WHA57.17. Retrieved 200603-25 from http: www.who.int/gb/ebwha/pdf_les/WHA57/A57_R17-en.pdf.
20 p.
World Health Organization/Food and Agriculture Organization (WHO/FAO), 2003.
Diet, nutrition and prevention of chronic diseases. WHO Technical Report
Series, 916. Geneve, 149 p.