1070 Am J C/in Nutr 1995;61:1070-5. Printed in USA. © 1995 American Society for Clinical Nutrition

The effect of raw potato starch on energy expenditure andsubstrate oxidation13

Anna Tagliabue, Anne Raben, Marie Louise Heijnen, Paul Deurenberg, Elisabetta Pasquali,

and Arne Astrup

ABSTRACT Because resistant starch (RS) is not absorbed

as glucose in the small intestine of healthy humans, postpran-

dial thermogenesis should be lower after the intake of RS ascompared with digestible starch. To evaluate this hypothesis,

we measured 5-h postprandial thermogenesis and substrate

oxidation by indirect calorimetry after ingestion of 50 g pre-

gelatinized (0% RS) and 50 g raw potato starch (54% type IIRS) in 15 healthy, normal-weight young males. The subjectsconsumed each starch (mixed in diluted fruit syrup) twice onseparate days and in random order. RS intake was followed bylower thermogenesis (46.5 ± 13.1 compared with 115.4 ± 10.4kJ/5 h; P = 0.008), lower glucose oxidation (P < 0.0005), andgreater fat oxidation (P 0.013) than was pregelatinized

starch consumption. Our results suggest that RS has no ther-

mogenic effect and that its presence does not influence the sizeof the thermic response to digestible starch. Am J Clin Nutr

1995;61:1070-5.

KEY WORDS

sis, substrate oxidation

Introduction

Resistant starch, diet-induced thermogene-

Until recently starch was thought to be completely hydro-

lyzed and absorbed in the small intestine. However, several invivo studies using breath-hydrogen collection (1-4), direct

intubation of the gut (5, 6), or starch measurement in ileostomyeffluents (7-1 1) have shown that starch from many sources

(including cereals, potatoes, bananas, and legumes) is not corn-pletely digested or absorbed in the healthy small bowel. More-

over, in vitro studies have revealed that many processed foodscontain starch resistant to dispersion in boiling water and to

hydrolysis with pancreatic amylase (12). The sum of starch and

starch degradation products not absorbed in the small intestine

of healthy individuals is defined as resistant starch (RS) (13).The incomplete digestion of starch in humans may depend

on several factors, including physical structure, food process-ing, chewing, intestinal transit time, the presence of amylaseinhibitors in the rest of the meal, possible cross-linking and

substitutions in modified starch, and lipid or protein interac-

tions (14, 15).

Among carbohydrate-containing foods, raw potato and un-

ripe banana contain the most RS (type II; 15) because of their

content of B-type starch granules (16). Other types of naturally

occurring RSs in the human diet are represented by physically

inaccessible starch (RS type I) in partially milled grains and

seeds and retrograded starch (RS type III), which is generated

by food processing (eg, bread; cooled, cooked potato; and cornflakes; 15). The RS content of food may affect both gastroin-

testinal function and energy balance. In rats feeding of raw

potato starch resulted in greater luminal acidity, greater luminalbulk, and slower transit time compared with feeding of cooked

starch (17, 18). Starch malabsorption induced by a glucosidase

inhibitor also resulted in a significantly prolonged transit time

in humans (19).

It is also possible that a diet high in RS may be useful in the

treatment of obesity and diabetes because of the reducedamount of available energy present in RS. However, becauseRS is not digested and therefore not absorbed as glucose in thesmall intestine of healthy humans, lower postprandial energy

expenditure can be expected after the intake of RS as comparedwith that of digestible starch. It was observed recently that aparticular type of RS obtained by lintnerization from an amy-lose-rich starch (70% amylose) has no thermogenic effect

either alone or mixed with glucose (20).The present study was performed to obtain information on

the effect of replacing digestible potato starch with resistant

(type II) potato starch on postprandial thermogenesis and sub-strate oxidation in a group of healthy male subjects.

Methods

This research was a joint study within the European FLAIR(Food-Linked Agro-Industrial Research) Concerted Action onResistant Starch (EURESTA) framework. Exactly the same

experiment was conducted at the Departments of Human Nu-

trition of the University of Pavia, Italy; The Royal Veterinary

1 From the Department of Human Nutrition, University of Pavia, Italy;

Research Department of Human Nutrition, Centre for Food Research, The

Royal Veterinary and Agricultural University, Frederiksberg, Denmark;

and the Department of Human Nutrition, Agricultural University of

Wageningen, Netherlands.

2 The experiments performed in Denmark were supported by grants from

the Danish Research and Development programme for Food Technology

1990-1994 and the Danish Medical Research Council, grant 12-9537-3.

3 Address reprint requests to A Tagliabue, Department of Human Nu-

trition, University of Pavia, Via Bassi 21, 27100 Pavia, Italy.

Received February 2, 1994.

Accepted for publication January 5, 1995.

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from

RESISIANI SIARCH-INDUCED ThERMOGENESIS 1071

‘ I ± SEM. RMR, resting metabolic rate.2 Mean of the four measurements per subject.

and Agricultural University of Frederiksberg, Denmark; andthe Agricultural University of Wageningen, Netherlands.

Subjects

In each of the participating centers five subjects were se-lected according to the following criteria: they had normalweight (body mass index < 30 kg/rn2); they were healthy

males, aged 20-31 y; they were non-smokers; they had nofamily history of diabetes or obesity; they had been weight-stable for the past 3 mo; and they had taken no medication for

the past 6 mo. Body composition was assessed by bioelectrical

impedance analysis (21). Females were excluded to avoid

possible periodicity in energy expenditure due to the menstrual

cycle. The characteristics of the volunteers are shown in Table 1.The study was approved by the Municipal Ethical Commit-

tee of Copenhagen and Frederiksberg and by the Committees

on Human Experimentation of the Universities of Pavia andWageningen in observance of the Helsinki II declaration. All

volunteers gave written informed consent before the study.

Starches

Potato starch was chosen as the test substance because theraw type contains a large amount (54% by wt) of RS (type II,

ie, raw starch granules). Pregelatinized potato starch (100%

digestible) was used as the control treatment. Raw and prege-

latinized starches were supplied by the Institut National de laRecherche Agronomique (Nantes, France). Pregelatinized po-

tato starch is obtained by drum-drying rehydrated raw potatostarch and is commercially available (Roquette, Lestrem,

France). Analyses of starch fractions (total starch, rapidly di-gestible, slowly digestible, and RS) were performed at the

Dunn Clinical Nutrition Center, Cambridge, UK, according tothe procedures described by Englyst et al (15) and Englyst and

Cummings (22).

Test meals

The meals containing raw and pregelatinized starch were

identified as meal A and meal B, respectively, and were pre-pared immediately before feeding. Fifty gram of each starch

was mixed into 500 g diluted (1:3) unsweetened fruit syrup by

using a hand mixer at the lowest speed to avoid heating of the

mixture. For meal A this resulted in a drinkable mixture,whereas meal B was a porridge-like gel that was eaten with a

spoon. The composition and energy content of the two testmeals is shown in Table 2. The raw and pregelatinized starchesdiffered in water content (16.6% and 4.9%, respectively) andtherefore meals A and B differed slightly in total starch con-

tent. The amount of digestible starch in meal A was 30% of thatin meal B. The metabolizable energy content of the raw starch

was supposed to derive exclusively from the in vitro digestible

starch, ie, fermentation was not taken into account. This wasdecided because fermentation from raw potato starch has been

demonstrated to occur only 9 h after ingestion (23).

Experimental protocol

The two test meals were given twice to each subject on

separate days in a crossover design, with half of the subjectsstarting with meal A and half with meal B. The test days wereseparated by � 1 wk, but not > 6 wk. Before each test day

subjects were instructed to follow a 3-d standard weight-main-tenance, carbohydrate-rich diet (60% carbohydrate, 12% pro-tein, 28% fat, 3.5 g dietary fiber/MJ ); abstain for 2 d from

strenuous physical activity; and fast for 12 h before energyexpenditure (EE) measurements.

In each center the measurements were carried out at a corn-fortable room temperature, 22-25 #{176}C(which falls within the

thermoneutral zone for clothed individuals; 24, 25), and arelative humidity of 50-60%.

Subjects were admitted to the research center 30 mm before

resting metabolic rate (RMR) measurements. Soon after their

arrival the participants voided and were weighed. Afterwardthey rested on a comfortable bed and bioelectrical impedance

was measured on the left side of the body.After 30 mm of relaxation, RMR was measured for 45 mm.

After this subjects consumed the test meal within 15 mm, andpostprandial metabolic rate (PMR) was measured for 5 h. Twoand 4 h after the measurements began subjects were allowed to

move around and to void.While under the ventilated hood, participants were permitted

to watch videotapes. A subgroup of the subjects underwent

blood sampling during the measurements. Details of bloodsampling and biochemical results are being published sepa-rately (26).

Energy expenditure

EE was measured by open-circuit indirect calorimetry with aventilated hood. The systems were described in detail else-

where (27-29). EE was calculated by using the Weir formula

(30). RMR and PMR were measured for 45 mm and 5 h,respectively.

Postprandial or diet-induced thermogenesis (DII, Id) over 5

h was obtained for each subject as the sum of the DII calcu-lated for 30-mm intervals as follows:

TABLE 1

Characteristics of the subjects’

Denmark (n = 5) Italy (n = 5) Netherlands (n = 5) All (n = 15)

Age (y) 25.6 ± 1.9 21.2 ± 0.7 23.2 ± 1.0 23.3 ± 0.8

Height (m) 1.80 ± 0.01 1.78 ± 0.02 1.83 ± 0.03 1.80 ± 0.01

Weight (kg) 71.2 ± 3.0 74.0 ± 3.5 71.7 ± 3.0 72.3 ± 1.7

BMI (kg/m2) 22.0 ± 1.1 23.1 ± 0.6 21.4 ± 0.7 22.2 ± 0.5

Fat (% of wt) 18.6 ± 2.4 20.8 ± 1.3 16.2 ± 1.1 18.5 ± 1.0

RMR (kJ/min)2 4.81 ± 0.19 4.98 ± 0.19 4.61 ± 0.27 4.80 ± 0.12

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from

1072 IAGLIABUE El AL

TABLE 2Composition and energy content of test meals A and B (50 g potato

starch + 500 mL diluted fruit syrup)’

Meal A Meal B

Total starch (g) 40.6 46.5

RS (g) 27.0 0.0

RDS (g) 3.0 44.0SDS (g) 10.6 2.6

Fruit sugars from syrup (g) 8.4 8.4

Glucose (g) 2.9 2.9

Fructose (g) 4.4 4.4

Sucrose (g) 1.1 1.1

Gross energy (�J)2 843 946

Metabolizable energy (kJ)3 370 946

1 Meal A, raw potato starch; meal B, pregelatinized potato starch; RS,

resistant starch; RDS, rapidly digestible starch; SDS, slowly digestible

starch.

2 Heat of combustion of starch (17.5 kJ/g) and glucose (15.7 kJ/g).

3 Calculated from the amount of carbohydrate digested and absorbed in

the small intestine and does not take into account intestinal fermentation.

DII (kJ/30 mm) = PMR (kJ/30 mm) - RMR (kJ/30min)

(1)

Total postprandial thermogenesis was also expressed as a

percentage of metabolizable energy as follows:

DIT(%)

= DIT (kJ/5 h)/metabolizable energy of the meal (Id) X 100

Substrate oxidation

Substrate oxidation in the postabsorptive and postprandial

conditions after the two meals was calculated according toFerranini (31):

G (g/min) = 4.55 CO2 - 3.21 02 2.87 UN (3)

G (from glycogen, g/min) = 4.09 CO2 - 2.88 02 - 2.59 UN

(4)

F (g/min) = 1.67 02 - 1.67 CO2 - 1.92 UN

P (g/min) = 6.25 UN

where G is glucose, F is fat, P is protein, CO2 is CO2 produced

(L/min), 02 is 02 consumed (Lfmin), and UN is urinary nitro-gen (g/min).

Statistical analysis

One-way analysis of variance (ANOVA) techniques, includ-ing Duncan’s test, were used to test for differences in age,

weight, height, body mass index, fat percentage, and RMRbetween countries. Results of duplicate respiratory measure-

ments obtained on each subject with each starch were averaged

and the mean value was used in the statistical calculations.Repeated-measures ANOVA was used to test the effect of

time and meal on energy expenditure, respiratory quotient, andsubstrate oxidation. The order in which the meals were offered(period effect) and meal X period interaction were statistically

negligible. Paired Student’s t test was used to compare mean

postprandial thermogenesis and substrate oxidation induced by

the two meals. All results were expressed as mean ± SEM

unless otherwise specified. SEM includes between-countryvariability. SPSS statistical package version 3. 1 (SPSS Inc,

Chicago) was used in the statistical calculations.

Results

Subjects

There were no differences among the participants from the

three countries with regard to age (P = 0.09), weight (P =

0.81), height (P = 0.47), body mass index (P = 0.37), body fat

content (P = 0.20), or RMR (P = 0.51). Age differed slightly(P < 0.05) between Denmark (25.6 ± 1.9 y) and Italy (21.2 ±

0.7 y). Because this difference was mainly due to one subject

(31 y old), this age difference was not taken into consideration

in the subsequent statistical analyses. One subject had a clearlyhigher body fat content that exceeded the normal range of thetotal group, and one subject had a low RMR, which could beattributed to low body weight. Given the generally small dif-

ferences in the data between the participating countries (Table

1), it was decided to treat the data as one entity.

Energy expenditure

Mean RMR of the 15 subjects did not differ before test meals

A and B (4.83 ± 0.13 and 4.78 ± 0.12 kJ/min, respectively,

NS). Time courses of EE after the two test meals are shown in

Figure la. A significant increase was observed after both test

meals (A: P 0.001, B: P = 0.001). This increase was

(2) significantly greater after test meal B than after test meal A(interaction meal X time, P = 0.006). DIT averaged 46.5 ±

18.0 kJ/5 h after raw starch compared with 1 15.4 ± 12.4 kJ/5

h after pregelatinized starch, yielding a difference between the

two meals of 68.8 ± 22.2 kJ/5 h (P = 0.008). DIT expressedas a percentage of metabolizable energy did not differ between

the meals (A: 12.6 ± 4.9%, B: 12.2 ± 1.3%; NS).

Substrate oxidation

Values for the respiratory quotient and substrate oxidation

did not differ before the two test meals (Table 3). A significant

(5) increase in respiratory quotient was observed after both test

meals (A: P = 0.008; B: P < 0.0005). The increase was,

(6) however, significantly greater after test meal B (interactionmeal X time: P = 0.002) (Figure ib). Mean postprandial

substrate oxidation over 5 h is shown in Table 3. A significant

time effect in fat oxidation was observed after both test meals(A: P = 0.05; B: P = 0.005); there was also a significant

difference in the oxidation pattern between the meals (interac-tion meal X time: P < 0.0005) (Figure ic). Total postprandial

fat oxidation averaged 15.0 ± 1.0 g/5 h after meal A and 12.9

± 0.9 g/5 h after meal B (P = 0.013). These values correspond

with 37.7 ± 1.9% and 31.2 ± 1.7% of 5-h EE, respectively(P = 0.001). Net postprandial fat oxidation was significantly

greater after test meal A (1.59 ± 0.74 g/5 h) than after test meal

B (-0.17 ± 1.08 g/5 h) (P = 0.033).

Postprandial carbohydrate oxidation increased significantly

after both test meals (A and B: P = 0.002), but the increase was

significantly greater after pregelatinized starch than after raw

starch (interaction meal X time: P = 0.002) (Figure ld). Total

postprandial carbohydrate oxidation (expressed as glucose

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from

-a- A

-4- B

a)

1073

b)

Energy expenditure

(kJ/min)

Respiratory quotient

5.9 1.00

5.7

5.5 0.95

53

0.905.1

4.9 0.85

4.7

4.5 0.80

0 60 120 180 240 300 0 60 120 180 240 300

Fat oxidation

(g/30 mm) Carbohydrate oxidation

2.2 (g/30 mm)c)

6.5

5.0

3.5

2.0

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

RESISIANI STARCH-INDUCED ThERMOGENESIS

0 60 120 180 240 300 0 60 120 180 240 300

Minutes after the meal Minutes after the meal

FIGURE 1. Postprandial energy expenditure (a), respiratory quotient (b), fat oxidation (c), and carbohydrate oxidation (d) after the intake of

pregelatinized (meal B) and raw potato starch (meal A). i ± SEM; n 15. By ANOVA-a, meal effect: P 0.024, interaction meal X time: P = 0.006;

b, meal effect: P < 0.0005, interaction meal X time: P = 0.002; c, meal effect: P 0.013, interaction meal X time: P < 0.0005; d, meal effect: P < 0.0005,

interaction meal X time: P = 0.002.

equivalents) averaged 39.2 ± 2.1 g/5 h after meal A and 49.5± 1.7 g/5 h after meal B (P < 0.0005). These values corre-

spond with 44.0 ± 2.1% and 53.6 ± 2.0% of 5-h EE, respec-

tively (P < 0.0005). Net postprandial carbohydrate oxidation

was significantly greater after test meal B (12.0 ± 2.7 g/5 h)

than after meal A (2.4 ± 2.0 g/5 h, P < 0.0005).

The balance between digestible carbohydrate intake and 5-h

total carbohydrate oxidation was negative after meal A (-15.7

± 2.1 g) and positive after meal B (10.5 ± 1.7 g, P < 0.0005).

Discussion

The thermogenic effect of the meal containing 50 g prege-

latinized potato starch amounted to 1 15 kJ in 5 h, equivalent to

12.2% of ingested energy. Substituting about two-thirds of the

digestible starch with RS (Table 2) reduced the thermogenic

effect from 1 15 to 46 kJ over 5 h, ie, a 60% reduction.

Both meals contained the same small quantity (8.4 g) of free

sugars (Table 2). If it is supposed that the thermic effect of

those sugars represents � 10% of their energy content (32-34),

14 kJ/5 h was derived from the syrup. Therefore, pure starch-

induced thermogenesis would be 1 15 - 14 = 101 kJ for the

pregelatinized starch and 46.5 - 14 = 32.5 U for the raw starch.

Dividing the value of pregelatinized starch-induced thermogen-

esis (101 U) by its digestible starch content (46.5 g), we obtain

a value for starch-induced thermogenesis of 2.2 kJ/g. This

value encompasses the cost of starch digestion, glucose absorp-

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from

0.85 ± 0.01 0.85 ± 0.01 0.84 ± 0.01 0.87 ± o.Oi�44.77 ± 3.99 43.66 ± 5.47 50.08 ± 3.26 43.10 ± 2.97�

49.15 ± 3.24 46.91 ± 2.93 49.15 ± 3.24 46.91 ± 2.93

122.94 ± 9.35 124.78 ± 10.96 130.85 ± 6.98 164.93 ± 5.62�

1074 TAGLIABUE El AL

TABLE 3

Fasting and postprandial values for respiratory quotient and substrate oxidation in 15 subjects’

Fasting Postprandial2

Meal A Meal B Meal A Meal B

Respiratory quotient

Fat (mg/mm)

Protein (mg/mm)

Carbohydrates (mg/mm)

‘ .1 ± SEM. Meal A, raw potato starch; meal B, pregelatinized potato starch.2 Mean of 5 h.

3 Significantly different from postprandial values for meal A, P < 0.05.

tion and storing (obligatory thermogenesis), as well as the

facultative thermogenesis that is at least partly due to insulin-

mediated activation of the sympathetic nervous system (35,

36). The energy cost of storing glucose as glycogen has been

calculated to be 0.2 kcal/g (0.8 kJ/g) by Flatt (37). It is wellestablished, however, that the measured increase in EE is well

in excess of this theoretically calculated value (36, 38). Thie-

baud et al (38) and Acheson et al (36) estimated the thermic

effect of intravenous glucose to be 1 .7 kJ/g (“�‘0.4 kcal/g).

Because hydrolysis of starch yields 1. 1 1 g glucose/g starch, thethermic effect of starch should be 1 .9 kJ/g, which is close to our

observed value of 2.2 kJ/g. The latter includes the additional

cost of digesting and absorbing the nutrient.

The expected thermogenesis induced by 50 g raw starchcontaining 14 g digestible starch would then be “�30 kJ in 5 h,which is quite comparable with the measured value of 32.5 U/S

h. These theoretical calculations show that RS has no thermo-

genic effect, at least not in the amount used in the present

experiment, and probably does not affect the availability of

digestible carbohydrate in the meal. The lower postprandialthermogenesis induced by raw starch is probably due to lower

obligatory thermogenesis as well as to lower sympathetic stim-ulation after the ingestion of a smaller quantity of digestiblenutrient.

No difference was found when we expressed the thermo-

genic effect as a percentage of ingested energy; in other words,

DIT was proportional to the amount of digestible carbohy-

drates. Similar but smaller reductions in DII have been found

in studies comparing high-fiber with low-fiber meals (39, 40).

These reductions have been explained by differences in oblig-atory thermogenesis and/or in different nutrient bioavailability

in high-fiber meals. The present study differs in at least tworespects from these fiber studies. In fact, we did not use mixed

meals, but rather meals composed of pure carbohydrates; wedid not add RS to a meal, but tested equal amounts of starch

that differed in the percentage of digestible starch. Under these

experimental conditions RS had no effect on the apparent

availability of digestible carbohydrates. Furthermore, we ob-

served greater fat oxidation after ingestion of raw starch than

after pregelatinized starch. We cannot exclude the possibility

that this effect might depend on an extension of the fasting

state, because the energy content of meal A was very small

(367 U). In particular, we noted that meal A produced an early

increase (at 30 mm) in respiratory quotient (Figure ib), fol-

lowed by a subsequent decrease, so that mean postprandial

respiratory quotient was significantly lower after this meal than

after meal B (Table 3). This initial rise was probably attribut-

able to the different viscosity of the two meals. Meal A was

less viscous and could have emptied from the stomach more

rapidly, resulting in faster carbohydrate absorption in the in-testine. After 30 mm, the low energy content of the meal could

have led to a resumption of fat oxidation, as occurs during

periods of fasting. On the other hand, the greater carbohydrate

content of meal B maintained an elevated respiratory quotient

up to 120 mm.One of the possibly useful effects of increasing the amount

of RS in foods is the reduced energy density that would be

achieved. This could be useful in the prevention and treatment

of obesity and diabetes (20). In the present experiment, the

same amount of food (50 g starch + 500 mL diluted syrup)

provided almost the same gross energy intake with or without

RS (Table 2); the 100-kJ difference was due to the higher

moisture content of RS. On the contrary, the difference inmetabolizable energy calculated from the amount of carbohy-

drate absorbed in the small intestine is rather large (576 kJ;

Table 2). However, the metabolizable energy of the meal

containing RS should be increased by the energy contribution

resulting from the absorption of short-chain fatty acids (SC-

FAs) produced by fermentation of undigested starch in thelarge intestine (41). An increase in breath hydrogen, indicativeof fermentation, has been observed ‘9 h after ingestion of raw

potato starch (23).The energy contribution supplied by intestinal fermentation

of RS was estimated at �“8 kJ/g (2 kcal/g; 42). On the basis of

this assumption, the difference in metabolizable energy be-tween the two test meals would be reduced to only 350 kJ.However, the value of 2 kJ/g RS depends on the ratio of SCFAs

(acetate, propionate, and butyrate) resulting from the fermen-tation (42). More studies are needed on the pattern of SCFAs

arising from in vivo fermentation of various RS types, to learn

the exact energy contribution it provides.

At present, our data allow us to conclude that RS (type II)has no thermogenic effect during the 5 h after its intake, and

that it does not seem to influence the size of the thermicresponse to digestible starch. U

We thank Narciso Cazzato, Hellas Cena, Deborah Terracina; the dieti-

tians Claudia Trentani and Caterina Mamini (Department Human Nutri-

tion, Pavia, Italy); Franco Bobbio Pallavicini (Policlinico San Matteo,

Pavia, Italy); Lene Hellemkjaer Jensen, John Lind. Inger-Lise Gr#{248}nfeldt,

and her kitchen staff (Frederiksberg, Denmark) for their help in conducting

the experiments; and Cristina Montomoli (Department of Bio Statistics,

Pavia, Italy) for statistical assistance.

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from

RESISIANI STARCH-INDUCED ThERMOGENESIS 1075

References

1. Calloway D. Respiratory hydrogen and methane as affected by con-

sumption of gas forming foods. Gastroenterology 1966;51:383-9.

2. Calloway DH, Murphy EL, Bauer D. Determination of lactose intol-

erance by breath analysis. Am I Dig Dis 1969;14:811-5.

3. Levitt MD, Donaldson RM. Use of respiratory hydrogen (H2) excre-

tion to detect carbohydrate malabsorption. I Lab Clin Med 1970;75:

937-45.

4. Anderson IH, Levine AS, Levitt MD. Incomplete absorption of the

carbohydrate in all purpose wheat flour. N EngI I Med 1981;304:

891-2.

5. Stephen AM, Haddad AC, Phillips SF. Passage of carbohydrate into

the the colon: direct measurements in humans. Gastroenterology 1983;

85:589-95.

6. Flouri#{233}B, Leblond A, Florent CH, Rautureau M, Bisalli A, Rambaud

I-C. Starch malabsorption and breath gas excretion in healthy humans

consuming low- and high-starch diets. Gastroenterology 1988;95:356-

63.

7. Sandberg A-S. Andersson H, Hallgren B, Hasselblad K, Isaksson B.

Experimental model for in vivo determination of dietary fibre and its

effect on the absorption of nutrients in the small intestine. Br I Nutr

1981;45:283-94.

8. Chapman RW, Sillery 1K, Graham MM, Saunders DR. Absorption of

starch by healthy ileostomates: effect of transit time and of carbohy-

drates load. Am I Clin Nutr 1985;41:1244-8.

9. Englyst HN, Cummings JH. Digestion of the polysaccharides of some

cereal foods in the human small intestine. Am I Clin Nutr 1985;42:

778-87.

10. Englyst HN, Cummings IH. Digestion of the carbohydrates of banana

(Musa paradisiaca sapiena’um) in the human small intestine. Am I Clin

Nutr 1986;44:42-50.

1 1. Englyst HN, Cummings JH. Digestion of polysaccharides of potato in

the small intestine of man. Am I Clin Nutr 1987;45:423-31.

12. Englyst HN, Anderson V, Cummings IH. Starch and non-starch po-

lysaccharides in some cereal foods. I Sci Food Agric 1983;34:1434-

40.

13. Asp N-G. Resistant starch. Eur I Clin Nutr 1992;46 (suppl 2):S1.

14. Cummings IH, Englyst HN. Measurement ofstarch fermentation in the

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

15. Englyst HN, Kingman SM, Cummings IH. Classification and mea-

surement of nutritionally important starch fractions. Eur I Clin Nutr

1992;46(suppl 2):S33-S50.

16. Gallant DI, Bouchet B, Bul#{233}onA, Perez S. Physical characteristics of

starch granules and susceptibility to enzymatic degradation. Eur I Clin

Nutr 1992;46 (suppl 2):S3-S16.

I 7. Calvert Ri, Otsuka M, Satchithanandam S. Consumption of raw potato

starch alters intestinal function and colonic cell proliferation in the rat.

I Nutr 1989;119:1610-6.

18. Lajvardi A, Mazarin GI, Gillespie MB, Satchithanandam 5, Calvert

Ri. Starches of varied digestibilities differentially modify intestinal

function in rats. I Nutr 1993;123:2059-66.

19. Scheppach W, Fabian C, Sachs M, Kasper H. The effect of starch

malabsorption on fecal short-chain fatty acid excretion in man. Scand

I Gastroenterol 1988;23:755-9.

20. Ranganathan 5, Champ M, Pechard C, et al. Comparative study of the

acute effects of resistant starch and dietary fibers on metabolic indexes

in men. Am I Clin Nutr 1994;59:879-83.

21. Deurenberg P, van der Kooy K, Leenen R, Westrate IA, Seidell IC.

Sex- and age specific prediction formulas for estimating body com-

position from bioelectrical impedance: a cross validation study. Int I

Obes 1991;15:17-25.

22. Englyst HN, Cummings JH. An improved method for the measurement

of dietary fiber as non-starch polysaccarides in plant foods. I Assoc

Off Anal Chem 1988;71:808-14.

23. Olesen M, Rumessen II, Gudmand-Hoyer. The hydrogen breath test in

resistant starch research. Eur I Clin Nutr l992;46(suppl 2):S133-4.

24. Houdas I, Ring EFI. Human body temperature (its measurements and

regulation). New York: Plenum Press, 1982:110.

25. Edholm OG, Weiner IS. Thermal physiology. In: Edholm 0G. Weiner

IS, eds. The principles and practice of human physiology. Vol 286.

London: Academic Press 1981:168.

26. Raben A, Tagliabue A, Christensen NJ, Madsen I, Holst II, Astrup A.

Resistant starch: the effect on postprandial glycemia, hormonal re-

sponse, and satiety. Am I Clin Nutr 1994;60:544-51.

27. Takala I, Merilainen P. Handbook of gas exchange and indirect cal-

orimetry. Helsinki: Datex/Instrumentarium Corp, 1989.

28. Fellows 1W, MacDonald I. An automated method for the measurement

of oxygen consumption and carbon dioxide excretion in man. Clin

Phys Physiol Meas 1985;6:349-55.

29. Leenen R, van der Kooy K, Deurenberg P, et al. Visceral fat accumu-

lation in obese subjects: relation to energy expenditure and response to

weight loss. Am I Physiol 1992;263:E913-9.

30. Weir lB de V. New methods for calculating metabolic rate with special

reference to protein metabolism. I Physiol 1949;109:l-9.

31. Ferranini E. The theoretical bases of indirect calorimetry: a review.

Metabolism 1988;37:287-301.

32. Tappy L, Randin I-P. Felber I-P, et al. Comparison of thermogenic

effect of fructose and glucose in normal humans. Am I Physiol

1986;250:E718-24.

33. Schutz Y, Golay A, Felber I-P. Jequier E. Decreased glucose-induced

thermogenesis after weight loss in obese subjects: a predisposing

factor for the relapsing of obesity? Am I Clin Nutr 1984;39:38()-7.

34. Sharief NN, MacDonald I. Differences in dietary-induced thermogen-

esis with various carbohydrates in normal and overweight men. Am I

Clin Nutr 1982;35:267-72.

35. Acheson K, Iequier E, Wahren I. Influence of f3-adrenergic blockade

on glucose-induced thermogenesis in man. I Clin Invest l983;72:

981-6.

36. Acheson KI, Ravussin E, Wahren I, Jequier E. Thermic effect of

glucose in man. Obligatory and facultative thermogenesis. I Clin

Invest 1984;74:1572-80.

37. Flatt JP. The biochemistry of energy expenditure. In: Bray GA, ed.

Obesity research II. London: Newman Publishing Ltd, 1978:211-28.

38. Thiebaud D, Schutz Y, Acheson K, et al. Energy cost of glucose

storage in human subjects during glucose-insulin infusions. Am I

Physiol 1983;244:E216-21.

39. Scalfi L, Coltorti A, D’Arrigo E, et al. Effect of dietary fibre on

postprandial thermogenesis. Int I Obes 1987;1 l(suppl l):95-9.

40. Raben A, Christensen NJ, Madsen I, Hoist II, Astrup A. Decreased

postprandial thermogenesis and fat oxidation, but increased fullness

after a high-fiber meal compared with a low-fiber meal. Am I Clin

Nutr 1994;59:1986-94.

41. Cummings JH, Pomare EW, Branch WI, Naylor CPE, Macfarlane GT.

Short chain fatty acids in human large intestine, portal, hepatic and

venous blood. Gut 1987;28: 1221-7.

42. Mathers IC. Energy value of resistant starch. Eur J Clin Nutr 1992;

46(suppl 2):S129-30.

by guest on October 11, 2015

ajcn.nutrition.orgD

ownloaded from