thermic effect of glucose in obese subjects studied by

12
Br. J. Nutr. (1976), 35, 281 28 I Thermic effect of glucose in obese subjects studied by direct and indirect calorimetry BY PH. PITTET, PH. CHAPPUIS, K. ACHESON, F. DE TECHTERMANN AND E. JGQUIER Division of Clinical Physiology and Institute of Physiology, University of Lausanne, I 01 I Lausanne, Switzerland (Received 8 August 1975 - Accepted 8 September 1975) I. The thermic effect of a glucose load (50 g) was studied in ten control and eleven obese female subjects, using both direct and indirect calorimetry simultaneously. Experiments were done under conditions of thermal equilibrium (28O and 30 % relative humidity). 2. Thermal balance (heat production measured by indirect calorimetry minus heat losses measured directly) was negative in the control group during the fasting period (heat deficit - 14.2 _+ 5.0 kJ/m2 per h), whereas that of the obese group was in equilibrium (+ 1.4 t4.8 kJ/ma per h). 3. After the glucose load, metabolic rate increased 13.0 f 1.5 and 52 1.3 % in the control and obese groups respectively. 4. In contrast to the metabolic rate, total heat losses were not significantly altered in either group after the glucose load. Total heat losses of the obese group were significantly lower than those of the control group throughout the experimental period. 5. During the experiments the amount of heat stored was increased in both groups. Thermal balance in the control group became positive while that of the obese group remained positive. 6. During the fasting period, the control subjects oxidized more carbohydrates (90.4 mglmin) than lipids (68.8 mglmin), whereas obese subjects oxidized more lipids (103.7 mglmin) than carbohydrates (50.2 mg/min). After the glucose load, the oxidation rate of carbohydrates was increased in both groups to 158-1 mg/min in control subjects and 95-6 mg/ min in obese subjects. 7. The mean skin temperature of the control subjects was significantly higher than that of the obese subjects and remained higher throughout the postprandial period. 8. These results indicate that: (a) during the fasting period, the energy sources utilized and the thermal balance of the two groups were different; (b) the thermic effect of glucose was less in the obese subjects and, therefore, might be a factor contributing to their low energy expenditure. It is widely believed that people become obese because they eat relatively more than non-obese individuals. Thus obesity has for many years been attributed to gluttony.. However, this is frequently not the situation (Johnson, Burke & Mayer, 1956; Stefanik, Heald & Mayer, 1959). Many other criteria are involved in the development of obesity: e.g. social, economic and genetic profiles, muscular activity and biochemical characteristics. Galton & Bray (1967) have reported that the activity of the glycerophosphate cycle is reduced in the adipose tissue of obese subjects, suggesting a more efficient use of energy than in control subjects. Miller & Mumford (1966), and Stirling & Stock (1968) suggest that the development of obesity is likely to be due to a thermogenic defect rather than an aberration in appetite, while Linton, Conley, Kuechenmeister & McClusky (1972) suggest that obese individuals take longer to recognize their feeling of satiety and discomfort with continued eating. Cabanac & Duclaux (1970) have also reported a deficiency in the satiety response to sucrose in obese subjects. https:/www.cambridge.org/core/terms. https://doi.org/10.1079/BJN19760033 Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 20:19:22, subject to the Cambridge Core terms of use, available at

Upload: others

Post on 01-Feb-2022

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Thermic effect of glucose in obese subjects studied by

Br J Nutr (1976) 35 281 28 I

Thermic effect of glucose in obese subjects studied by direct and indirect calorimetry

BY PH PITTET PH CHAPPUIS K ACHESON F DE TECHTERMANN AND E JGQUIER

Division of Clinical Physiology and Institute of Physiology University of Lausanne I 01 I Lausanne Switzerland

(Received 8 August 1975 - Accepted 8 September 1975)

I The thermic effect of a glucose load (50 g) was studied in ten control and eleven obese female subjects using both direct and indirect calorimetry simultaneously Experiments were done under conditions of thermal equilibrium (28O and 30 relative humidity) 2 Thermal balance (heat production measured by indirect calorimetry minus heat losses

measured directly) was negative in the control group during the fasting period (heat deficit - 142 _+ 50 kJm2 per h) whereas that of the obese group was in equilibrium (+ 14 t48 kJma per h)

3 After the glucose load metabolic rate increased 130 f 15 and 5 2 13 in the control and obese groups respectively 4 In contrast to the metabolic rate total heat losses were not significantly altered in either

group after the glucose load Total heat losses of the obese group were significantly lower than those of the control group throughout the experimental period

5 During the experiments the amount of heat stored was increased in both groups Thermal balance in the control group became positive while that of the obese group remained positive

6 During the fasting period the control subjects oxidized more carbohydrates (904 mglmin) than lipids (688 mglmin) whereas obese subjects oxidized more lipids (1037 mglmin) than carbohydrates (502 mgmin) After the glucose load the oxidation rate of carbohydrates was increased in both groups to 158-1 mgmin in control subjects and 95-6 mg min in obese subjects

7 The mean skin temperature of the control subjects was significantly higher than that of the obese subjects and remained higher throughout the postprandial period

8 These results indicate that (a) during the fasting period the energy sources utilized and the thermal balance of the two groups were different (b) the thermic effect of glucose was less in the obese subjects and therefore might be a factor contributing to their low energy expenditure

It is widely believed that people become obese because they eat relatively more than non-obese individuals Thus obesity has for many years been attributed to gluttony However this is frequently not the situation (Johnson Burke amp Mayer 1956 Stefanik Heald amp Mayer 1959)

Many other criteria are involved in the development of obesity eg social economic and genetic profiles muscular activity and biochemical characteristics Galton amp Bray (1967) have reported that the activity of the glycerophosphate cycle is reduced in the adipose tissue of obese subjects suggesting a more efficient use of energy than in control subjects Miller amp Mumford (1966) and Stirling amp Stock (1968) suggest that the development of obesity is likely to be due to a thermogenic defect rather than an aberration in appetite while Linton Conley Kuechenmeister amp McClusky (1972) suggest that obese individuals take longer to recognize their feeling of satiety and discomfort with continued eating Cabanac amp Duclaux (1970) have also reported a deficiency in the satiety response to sucrose in obese subjects

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

282 PH PITTET AND OTHERS I976

Table I Main physical characteristics of the two groups of female subjects used in the study (Mean values with their standard errors)

Body-wt Age Height Body surface A

Subjects n (years) (m) bZ) kg ideal wt

Control 10 2425 17 169 5 21 1635004 566f 19 1020 amp 20 Obese I1 289amp32 1645 1-7 190 005 837 amp 44 1593 5 79

Ns not significant Differences between mean values for the two groups were statistically significant (unpaired t test)

X + X I ++ NS NS

++ P lt O O O I

The thermic effect of nutrients represents a loss of energy for the body Since frankly obese subjects can maintain their weight with a lower energy intake than lean subjects it was of interest to study whether the thermic effect of glucose might be lower in obese subjects than in lean controls

We studied oxygen consumption the oxidation rate of lipids carbohydrates and proteins total heat losses and thermal balance in two groups of female subjects after an oral load of 50 g glucose

E X P E R I M E N T A L

Two groups of females who were unfamiliar with the equipment were studied in a direct calorimeter as described previously (Pittet Gygax amp Jkquier 1974)

The first group consisted of eleven obese females with no evidence of endo- crinological disturbances and the second was a group of ten healthy lean female subjects Their main physical characteristics are given in Table I Each subject had fasted for 12 h overnight and was allowed to rest for I h in a constant temperature room at 28 and 30 yo relative humidity before measurements were made The sub- ject wearing a minimum of underclothing was introduced into the calorimeter which was regulated to the same ambient conditions as those of the constant temperature room Initial (fasting) measurements were taken for 40 min or until stable values were obtained The subject was then given an oral dose of 50 g glucose and the measure- ments were continued for a further 150 min The following measurements were made continuously throughout the experiment

Internal temperature This was measured using a tympanic and a sublingual probe Mean skin temperature This was estimated by weighting eight different skin tem-

peratures according to the procedure of Hardy amp Du Bois (1937) Metabolic rate and respiratory quotient (RQ) These were estimated from determina-

tions of ventilation and the concentrations of 0 and carbon dioxide in the expired air using an open-circuit system (Gomez JCquier Chabot Buber amp Felber 1972) Although the RQ can be influenced by changes in ventilation over short periods (ie hypo- or hyperventilation) its integration over 30 min periods corrects for these changes and results in a mean RQ which reflects the true oxidation rate of energy substrates

The amount of protein (estimated in terms of amino acids) metabolized was cal-

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic eflect of glucose in obese subjects 283 culated from the urinary nitrogen excretion (Du Bois 1924) The lsquonon-proteinrsquo RQ was then determined and the tables of Lusk (1924) were used to compute the amount of carbohydrate and lipid (as fatty acids) oxidized during the experiment Total heat losses were measured by gradient-layer direct calorimetry (Spinnler JCquier Favre Dolivo amp Vannotti 1973) which enabled the separate measurement of lsquo dry rsquo heat losses (radiation + convection) and lsquo evaporative rsquo heat losses (insensible perspiration and sweating) to be made

Statistical analyses were done using Studentrsquos t test for paired and unpaired results The threshold of significance was P lt 005

R E S U L T S

Metabolic rate and heat losses After the glucose load the metabolic rate measured over a period of 150 min

increased significantly in both groups (Table 2) In the control group this increase was 130 of the initial (fasting) value whereas in the obese group it was only 52 ie about half the thermic effect found in the control group This difference in metabolic rate stimulation was highly significant (Table 3)

During the 150 min after the glucose load neither the total nor the lsquoevaporativersquo heat losses (which represent about 29 of the total heat loss) were changed in either group

However it is important to note that during the fasting period the total heat losses were significantly higher in the control group (Table 3) and that they remained higher throughout the postprandial period

Oxidation rate of the d$ferent substrates The mean oxidation rates of the different substrates are shown in Fig I and Table 4 During the fasting period the control group oxidized mainly carbohydrate (904 mg

min) whereas the principal energy source utilized by the obese group was lipid (103lsquo7 mgmin) The oxidation rates (mglmin) of the other substrates were control group 688 and 348 for lipid and protein respectively obese group 502 and 268 for carbohydrate and protein respectively

After the glucose load the oxidation rate of this substrate (carbohydrate) increased significantly in the control group lipid utilization decreased slightly and protein metabolism remained virtually unchanged Although there were similar over-all changes in substrate metabolism in the obese group the fasting level for the oxidation rate of carbohydrate was approximately half that in the control group for the fasting period and after the glucose load the value for the oxidation rate of carbohydrate was similar to that in the control group for the fasting period

Another point of interest is that the high level of utilization of lipid by the fasting obese subjects remained high after the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

D b

$

80TO

I -

s I 0

EI +

s0 + 912+ s0 + 601+

1E

-

LL + 1s

-

661 +

n

5 5 k

20 + 601 +

LPl+

10 -

10 -

8+ 0E + E

II+

9E - SEI + s1

-

6L +

96 0061

65 8002

61 9881

8E S661

11 L661

2s

E1

02

80 0161

01

zsoz

12

ES61

E6 S9Ez

80 E281

SE 9161

6E Z

LI-2

LI SS91

98 98

61 96L1

E6 0661

22

1161

0E L2

22

61 9012

60 9881

81 0O

LI

as uaam

95 8681

9L L661

01

6261 09

9861 1

2

oE61

01

0161

21

L161

06 9LEz

SE 6611

61 2681

61

16S1 E6

66Sr L0

EgL1

2E

2681

1E 8602

60

6SSI

86 0

80

2

66 8602

81 E881

E6 L991

8E L

Oamp

Z

9E 1802

06 2061

LS 8661

61 0161

oE 6602

61 1861

8L 8S

OZ

50 s681

PE 6S

oz 2

1

EE61 9E

28

22

L0 sE

81 66

S881 L1

6881 96

06

12

61 6891

62 S281

6S 6261

81 zS91

62 6602

61 6E81

11 8981

66 6812

BS uaam

v

6

L6 E161

65 8902

21

0L81

66 SEoz

60 6961

L2 6

21

2

60 1661

60 Looz

62 L802

06 9zPz

L0 s181

L1 8661

$I Z

ampI

66 E

822 61

6891 E6

6961

6S 2661

0s

oSL1 Sz

6622 61

0LOZ

01 0681

61 0681

6E 6881

29 2902

6S

SSoz

LI 0E

61 1L

2002

26

9202

21

L061

60 0181

81 E

061

0s

6862

L0 5181

8E 0002

91 888

19 S

ozz 11

LS91

62

S6L1

09 2661

Sz

0661 E6

1ZLI

22

0912

89 0

20

2

06 9E

lZ

zS E061

8S 0S61

E1 ISg1

L2 0581

LI 0E61

E9 6E

oz 81

E061 5

21

8E

02 01

00

02

22

L

SZ

Z

SE 2S61

9I 888

11 LS91

62 IZLI

11 StL

1 62

06L1

22

2691

11 E

10

2

L0 s181

69 8612

2E

0861

1s 1922

as ueapq

I

ES 0261

0s

8LLI

E1

9L

gI 6E

0P91 5

2

8E6I

s2

6281

SI 1L61

66 S061

60 6Soz

zE

EL

61 L0

0281 61

8ZLI

91 888

11

LSg1

L6 6291

11

S6LI

62 C

6LI

6E E681

60 6861

8s 6661

1s

1922 62

1S61

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 2 c

ontin

ued

Subj

ect

Bs

M

F

M

DA

L

S

B

SA

M

H

M

GJ

M

S

T

L

B G

C

P M

Mea

n

Cal

orim

etry

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Perio

d be

fore

gl

ucos

e lo

ad

(fas

ting)

Peri

od a

fter

glu

cose

load

I

__h_7

Mea

n

150

148

214

180

181

133

I75

181

184

I77

166

140

162

163

I59

171

167

166

147

I59

178

I 62

212

20

0

SE I I

07

86

6-5

32

29

43

29

14

68

18

I 8

4

0 50

I 4

32

25

29

I 4

71

45

22

22

11

2 3

7- ----

4 5

M

ean

SE

Mea

n

(b)

Obe

se s

ubje

cts

148

158

184

I77

142

207

I74

I75

185

I 86

16

6

148

I59

161

I 63

163

161

I 60

I4

3

I 60

176

163

210

20

0

I 8

1 4

5

4 2

2

11

25

40

3

2 4

7 5

4 4

3 1

8

I 4

104

2

9 2

5

43

22

32

I 8

6

8 4

3

11

11

SE

J 4

0

4

25

I 4

43

25

1 4

5

4

25

65

I I

65

11

11

32

29

36

07

I 8

I 4

29

1 4

6

1 3

7

Mea

n

148

I55

181

171

I44

20

I

20 I

176

176

I74

175

162

144

I 60

163

162

167

164

156

142

I97

I 60

I73

162

SE

07

32

68

32

32

43

I 4

47

3 2

1 4

1 4

0

7 3

2 I 4

1

4 2

9 2

2

1 4

61

3 7

36

22

22

11

Mea

n (

initi

al

rate

)+

+ 2

5 +

64

+ 11

8 -

35

+ 3

9 -

68

+ 81

+ 101

f11

4 - 5

5

+ 5

6 - 2

6

+ 4

0

+ 4

2 +

62

- 1

7 +

56

- 7

6 +

38

0

- 22

- 2

2

+ 5

2k

I3

-

05k

17

Fo

r de

tails

of

expe

rim

enta

l pro

cedu

res

see

p 282

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing

150

min

aft

er g

luco

se lo

ad a

s a

perc

enta

ge of

the

fast

ing

met

abol

ic ra

te (b

efor

e gl

ucos

e lo

ad)

w

cn Y

P F N

h

a N 3

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 2: Thermic effect of glucose in obese subjects studied by

282 PH PITTET AND OTHERS I976

Table I Main physical characteristics of the two groups of female subjects used in the study (Mean values with their standard errors)

Body-wt Age Height Body surface A

Subjects n (years) (m) bZ) kg ideal wt

Control 10 2425 17 169 5 21 1635004 566f 19 1020 amp 20 Obese I1 289amp32 1645 1-7 190 005 837 amp 44 1593 5 79

Ns not significant Differences between mean values for the two groups were statistically significant (unpaired t test)

X + X I ++ NS NS

++ P lt O O O I

The thermic effect of nutrients represents a loss of energy for the body Since frankly obese subjects can maintain their weight with a lower energy intake than lean subjects it was of interest to study whether the thermic effect of glucose might be lower in obese subjects than in lean controls

We studied oxygen consumption the oxidation rate of lipids carbohydrates and proteins total heat losses and thermal balance in two groups of female subjects after an oral load of 50 g glucose

E X P E R I M E N T A L

Two groups of females who were unfamiliar with the equipment were studied in a direct calorimeter as described previously (Pittet Gygax amp Jkquier 1974)

The first group consisted of eleven obese females with no evidence of endo- crinological disturbances and the second was a group of ten healthy lean female subjects Their main physical characteristics are given in Table I Each subject had fasted for 12 h overnight and was allowed to rest for I h in a constant temperature room at 28 and 30 yo relative humidity before measurements were made The sub- ject wearing a minimum of underclothing was introduced into the calorimeter which was regulated to the same ambient conditions as those of the constant temperature room Initial (fasting) measurements were taken for 40 min or until stable values were obtained The subject was then given an oral dose of 50 g glucose and the measure- ments were continued for a further 150 min The following measurements were made continuously throughout the experiment

Internal temperature This was measured using a tympanic and a sublingual probe Mean skin temperature This was estimated by weighting eight different skin tem-

peratures according to the procedure of Hardy amp Du Bois (1937) Metabolic rate and respiratory quotient (RQ) These were estimated from determina-

tions of ventilation and the concentrations of 0 and carbon dioxide in the expired air using an open-circuit system (Gomez JCquier Chabot Buber amp Felber 1972) Although the RQ can be influenced by changes in ventilation over short periods (ie hypo- or hyperventilation) its integration over 30 min periods corrects for these changes and results in a mean RQ which reflects the true oxidation rate of energy substrates

The amount of protein (estimated in terms of amino acids) metabolized was cal-

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic eflect of glucose in obese subjects 283 culated from the urinary nitrogen excretion (Du Bois 1924) The lsquonon-proteinrsquo RQ was then determined and the tables of Lusk (1924) were used to compute the amount of carbohydrate and lipid (as fatty acids) oxidized during the experiment Total heat losses were measured by gradient-layer direct calorimetry (Spinnler JCquier Favre Dolivo amp Vannotti 1973) which enabled the separate measurement of lsquo dry rsquo heat losses (radiation + convection) and lsquo evaporative rsquo heat losses (insensible perspiration and sweating) to be made

Statistical analyses were done using Studentrsquos t test for paired and unpaired results The threshold of significance was P lt 005

R E S U L T S

Metabolic rate and heat losses After the glucose load the metabolic rate measured over a period of 150 min

increased significantly in both groups (Table 2) In the control group this increase was 130 of the initial (fasting) value whereas in the obese group it was only 52 ie about half the thermic effect found in the control group This difference in metabolic rate stimulation was highly significant (Table 3)

During the 150 min after the glucose load neither the total nor the lsquoevaporativersquo heat losses (which represent about 29 of the total heat loss) were changed in either group

However it is important to note that during the fasting period the total heat losses were significantly higher in the control group (Table 3) and that they remained higher throughout the postprandial period

Oxidation rate of the d$ferent substrates The mean oxidation rates of the different substrates are shown in Fig I and Table 4 During the fasting period the control group oxidized mainly carbohydrate (904 mg

min) whereas the principal energy source utilized by the obese group was lipid (103lsquo7 mgmin) The oxidation rates (mglmin) of the other substrates were control group 688 and 348 for lipid and protein respectively obese group 502 and 268 for carbohydrate and protein respectively

After the glucose load the oxidation rate of this substrate (carbohydrate) increased significantly in the control group lipid utilization decreased slightly and protein metabolism remained virtually unchanged Although there were similar over-all changes in substrate metabolism in the obese group the fasting level for the oxidation rate of carbohydrate was approximately half that in the control group for the fasting period and after the glucose load the value for the oxidation rate of carbohydrate was similar to that in the control group for the fasting period

Another point of interest is that the high level of utilization of lipid by the fasting obese subjects remained high after the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

D b

$

80TO

I -

s I 0

EI +

s0 + 912+ s0 + 601+

1E

-

LL + 1s

-

661 +

n

5 5 k

20 + 601 +

LPl+

10 -

10 -

8+ 0E + E

II+

9E - SEI + s1

-

6L +

96 0061

65 8002

61 9881

8E S661

11 L661

2s

E1

02

80 0161

01

zsoz

12

ES61

E6 S9Ez

80 E281

SE 9161

6E Z

LI-2

LI SS91

98 98

61 96L1

E6 0661

22

1161

0E L2

22

61 9012

60 9881

81 0O

LI

as uaam

95 8681

9L L661

01

6261 09

9861 1

2

oE61

01

0161

21

L161

06 9LEz

SE 6611

61 2681

61

16S1 E6

66Sr L0

EgL1

2E

2681

1E 8602

60

6SSI

86 0

80

2

66 8602

81 E881

E6 L991

8E L

Oamp

Z

9E 1802

06 2061

LS 8661

61 0161

oE 6602

61 1861

8L 8S

OZ

50 s681

PE 6S

oz 2

1

EE61 9E

28

22

L0 sE

81 66

S881 L1

6881 96

06

12

61 6891

62 S281

6S 6261

81 zS91

62 6602

61 6E81

11 8981

66 6812

BS uaam

v

6

L6 E161

65 8902

21

0L81

66 SEoz

60 6961

L2 6

21

2

60 1661

60 Looz

62 L802

06 9zPz

L0 s181

L1 8661

$I Z

ampI

66 E

822 61

6891 E6

6961

6S 2661

0s

oSL1 Sz

6622 61

0LOZ

01 0681

61 0681

6E 6881

29 2902

6S

SSoz

LI 0E

61 1L

2002

26

9202

21

L061

60 0181

81 E

061

0s

6862

L0 5181

8E 0002

91 888

19 S

ozz 11

LS91

62

S6L1

09 2661

Sz

0661 E6

1ZLI

22

0912

89 0

20

2

06 9E

lZ

zS E061

8S 0S61

E1 ISg1

L2 0581

LI 0E61

E9 6E

oz 81

E061 5

21

8E

02 01

00

02

22

L

SZ

Z

SE 2S61

9I 888

11 LS91

62 IZLI

11 StL

1 62

06L1

22

2691

11 E

10

2

L0 s181

69 8612

2E

0861

1s 1922

as ueapq

I

ES 0261

0s

8LLI

E1

9L

gI 6E

0P91 5

2

8E6I

s2

6281

SI 1L61

66 S061

60 6Soz

zE

EL

61 L0

0281 61

8ZLI

91 888

11

LSg1

L6 6291

11

S6LI

62 C

6LI

6E E681

60 6861

8s 6661

1s

1922 62

1S61

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 2 c

ontin

ued

Subj

ect

Bs

M

F

M

DA

L

S

B

SA

M

H

M

GJ

M

S

T

L

B G

C

P M

Mea

n

Cal

orim

etry

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Perio

d be

fore

gl

ucos

e lo

ad

(fas

ting)

Peri

od a

fter

glu

cose

load

I

__h_7

Mea

n

150

148

214

180

181

133

I75

181

184

I77

166

140

162

163

I59

171

167

166

147

I59

178

I 62

212

20

0

SE I I

07

86

6-5

32

29

43

29

14

68

18

I 8

4

0 50

I 4

32

25

29

I 4

71

45

22

22

11

2 3

7- ----

4 5

M

ean

SE

Mea

n

(b)

Obe

se s

ubje

cts

148

158

184

I77

142

207

I74

I75

185

I 86

16

6

148

I59

161

I 63

163

161

I 60

I4

3

I 60

176

163

210

20

0

I 8

1 4

5

4 2

2

11

25

40

3

2 4

7 5

4 4

3 1

8

I 4

104

2

9 2

5

43

22

32

I 8

6

8 4

3

11

11

SE

J 4

0

4

25

I 4

43

25

1 4

5

4

25

65

I I

65

11

11

32

29

36

07

I 8

I 4

29

1 4

6

1 3

7

Mea

n

148

I55

181

171

I44

20

I

20 I

176

176

I74

175

162

144

I 60

163

162

167

164

156

142

I97

I 60

I73

162

SE

07

32

68

32

32

43

I 4

47

3 2

1 4

1 4

0

7 3

2 I 4

1

4 2

9 2

2

1 4

61

3 7

36

22

22

11

Mea

n (

initi

al

rate

)+

+ 2

5 +

64

+ 11

8 -

35

+ 3

9 -

68

+ 81

+ 101

f11

4 - 5

5

+ 5

6 - 2

6

+ 4

0

+ 4

2 +

62

- 1

7 +

56

- 7

6 +

38

0

- 22

- 2

2

+ 5

2k

I3

-

05k

17

Fo

r de

tails

of

expe

rim

enta

l pro

cedu

res

see

p 282

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing

150

min

aft

er g

luco

se lo

ad a

s a

perc

enta

ge of

the

fast

ing

met

abol

ic ra

te (b

efor

e gl

ucos

e lo

ad)

w

cn Y

P F N

h

a N 3

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 3: Thermic effect of glucose in obese subjects studied by

VOl 35 Thermic eflect of glucose in obese subjects 283 culated from the urinary nitrogen excretion (Du Bois 1924) The lsquonon-proteinrsquo RQ was then determined and the tables of Lusk (1924) were used to compute the amount of carbohydrate and lipid (as fatty acids) oxidized during the experiment Total heat losses were measured by gradient-layer direct calorimetry (Spinnler JCquier Favre Dolivo amp Vannotti 1973) which enabled the separate measurement of lsquo dry rsquo heat losses (radiation + convection) and lsquo evaporative rsquo heat losses (insensible perspiration and sweating) to be made

Statistical analyses were done using Studentrsquos t test for paired and unpaired results The threshold of significance was P lt 005

R E S U L T S

Metabolic rate and heat losses After the glucose load the metabolic rate measured over a period of 150 min

increased significantly in both groups (Table 2) In the control group this increase was 130 of the initial (fasting) value whereas in the obese group it was only 52 ie about half the thermic effect found in the control group This difference in metabolic rate stimulation was highly significant (Table 3)

During the 150 min after the glucose load neither the total nor the lsquoevaporativersquo heat losses (which represent about 29 of the total heat loss) were changed in either group

However it is important to note that during the fasting period the total heat losses were significantly higher in the control group (Table 3) and that they remained higher throughout the postprandial period

Oxidation rate of the d$ferent substrates The mean oxidation rates of the different substrates are shown in Fig I and Table 4 During the fasting period the control group oxidized mainly carbohydrate (904 mg

min) whereas the principal energy source utilized by the obese group was lipid (103lsquo7 mgmin) The oxidation rates (mglmin) of the other substrates were control group 688 and 348 for lipid and protein respectively obese group 502 and 268 for carbohydrate and protein respectively

After the glucose load the oxidation rate of this substrate (carbohydrate) increased significantly in the control group lipid utilization decreased slightly and protein metabolism remained virtually unchanged Although there were similar over-all changes in substrate metabolism in the obese group the fasting level for the oxidation rate of carbohydrate was approximately half that in the control group for the fasting period and after the glucose load the value for the oxidation rate of carbohydrate was similar to that in the control group for the fasting period

Another point of interest is that the high level of utilization of lipid by the fasting obese subjects remained high after the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

D b

$

80TO

I -

s I 0

EI +

s0 + 912+ s0 + 601+

1E

-

LL + 1s

-

661 +

n

5 5 k

20 + 601 +

LPl+

10 -

10 -

8+ 0E + E

II+

9E - SEI + s1

-

6L +

96 0061

65 8002

61 9881

8E S661

11 L661

2s

E1

02

80 0161

01

zsoz

12

ES61

E6 S9Ez

80 E281

SE 9161

6E Z

LI-2

LI SS91

98 98

61 96L1

E6 0661

22

1161

0E L2

22

61 9012

60 9881

81 0O

LI

as uaam

95 8681

9L L661

01

6261 09

9861 1

2

oE61

01

0161

21

L161

06 9LEz

SE 6611

61 2681

61

16S1 E6

66Sr L0

EgL1

2E

2681

1E 8602

60

6SSI

86 0

80

2

66 8602

81 E881

E6 L991

8E L

Oamp

Z

9E 1802

06 2061

LS 8661

61 0161

oE 6602

61 1861

8L 8S

OZ

50 s681

PE 6S

oz 2

1

EE61 9E

28

22

L0 sE

81 66

S881 L1

6881 96

06

12

61 6891

62 S281

6S 6261

81 zS91

62 6602

61 6E81

11 8981

66 6812

BS uaam

v

6

L6 E161

65 8902

21

0L81

66 SEoz

60 6961

L2 6

21

2

60 1661

60 Looz

62 L802

06 9zPz

L0 s181

L1 8661

$I Z

ampI

66 E

822 61

6891 E6

6961

6S 2661

0s

oSL1 Sz

6622 61

0LOZ

01 0681

61 0681

6E 6881

29 2902

6S

SSoz

LI 0E

61 1L

2002

26

9202

21

L061

60 0181

81 E

061

0s

6862

L0 5181

8E 0002

91 888

19 S

ozz 11

LS91

62

S6L1

09 2661

Sz

0661 E6

1ZLI

22

0912

89 0

20

2

06 9E

lZ

zS E061

8S 0S61

E1 ISg1

L2 0581

LI 0E61

E9 6E

oz 81

E061 5

21

8E

02 01

00

02

22

L

SZ

Z

SE 2S61

9I 888

11 LS91

62 IZLI

11 StL

1 62

06L1

22

2691

11 E

10

2

L0 s181

69 8612

2E

0861

1s 1922

as ueapq

I

ES 0261

0s

8LLI

E1

9L

gI 6E

0P91 5

2

8E6I

s2

6281

SI 1L61

66 S061

60 6Soz

zE

EL

61 L0

0281 61

8ZLI

91 888

11

LSg1

L6 6291

11

S6LI

62 C

6LI

6E E681

60 6861

8s 6661

1s

1922 62

1S61

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 2 c

ontin

ued

Subj

ect

Bs

M

F

M

DA

L

S

B

SA

M

H

M

GJ

M

S

T

L

B G

C

P M

Mea

n

Cal

orim

etry

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Perio

d be

fore

gl

ucos

e lo

ad

(fas

ting)

Peri

od a

fter

glu

cose

load

I

__h_7

Mea

n

150

148

214

180

181

133

I75

181

184

I77

166

140

162

163

I59

171

167

166

147

I59

178

I 62

212

20

0

SE I I

07

86

6-5

32

29

43

29

14

68

18

I 8

4

0 50

I 4

32

25

29

I 4

71

45

22

22

11

2 3

7- ----

4 5

M

ean

SE

Mea

n

(b)

Obe

se s

ubje

cts

148

158

184

I77

142

207

I74

I75

185

I 86

16

6

148

I59

161

I 63

163

161

I 60

I4

3

I 60

176

163

210

20

0

I 8

1 4

5

4 2

2

11

25

40

3

2 4

7 5

4 4

3 1

8

I 4

104

2

9 2

5

43

22

32

I 8

6

8 4

3

11

11

SE

J 4

0

4

25

I 4

43

25

1 4

5

4

25

65

I I

65

11

11

32

29

36

07

I 8

I 4

29

1 4

6

1 3

7

Mea

n

148

I55

181

171

I44

20

I

20 I

176

176

I74

175

162

144

I 60

163

162

167

164

156

142

I97

I 60

I73

162

SE

07

32

68

32

32

43

I 4

47

3 2

1 4

1 4

0

7 3

2 I 4

1

4 2

9 2

2

1 4

61

3 7

36

22

22

11

Mea

n (

initi

al

rate

)+

+ 2

5 +

64

+ 11

8 -

35

+ 3

9 -

68

+ 81

+ 101

f11

4 - 5

5

+ 5

6 - 2

6

+ 4

0

+ 4

2 +

62

- 1

7 +

56

- 7

6 +

38

0

- 22

- 2

2

+ 5

2k

I3

-

05k

17

Fo

r de

tails

of

expe

rim

enta

l pro

cedu

res

see

p 282

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing

150

min

aft

er g

luco

se lo

ad a

s a

perc

enta

ge of

the

fast

ing

met

abol

ic ra

te (b

efor

e gl

ucos

e lo

ad)

w

cn Y

P F N

h

a N 3

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 4: Thermic effect of glucose in obese subjects studied by

D b

$

80TO

I -

s I 0

EI +

s0 + 912+ s0 + 601+

1E

-

LL + 1s

-

661 +

n

5 5 k

20 + 601 +

LPl+

10 -

10 -

8+ 0E + E

II+

9E - SEI + s1

-

6L +

96 0061

65 8002

61 9881

8E S661

11 L661

2s

E1

02

80 0161

01

zsoz

12

ES61

E6 S9Ez

80 E281

SE 9161

6E Z

LI-2

LI SS91

98 98

61 96L1

E6 0661

22

1161

0E L2

22

61 9012

60 9881

81 0O

LI

as uaam

95 8681

9L L661

01

6261 09

9861 1

2

oE61

01

0161

21

L161

06 9LEz

SE 6611

61 2681

61

16S1 E6

66Sr L0

EgL1

2E

2681

1E 8602

60

6SSI

86 0

80

2

66 8602

81 E881

E6 L991

8E L

Oamp

Z

9E 1802

06 2061

LS 8661

61 0161

oE 6602

61 1861

8L 8S

OZ

50 s681

PE 6S

oz 2

1

EE61 9E

28

22

L0 sE

81 66

S881 L1

6881 96

06

12

61 6891

62 S281

6S 6261

81 zS91

62 6602

61 6E81

11 8981

66 6812

BS uaam

v

6

L6 E161

65 8902

21

0L81

66 SEoz

60 6961

L2 6

21

2

60 1661

60 Looz

62 L802

06 9zPz

L0 s181

L1 8661

$I Z

ampI

66 E

822 61

6891 E6

6961

6S 2661

0s

oSL1 Sz

6622 61

0LOZ

01 0681

61 0681

6E 6881

29 2902

6S

SSoz

LI 0E

61 1L

2002

26

9202

21

L061

60 0181

81 E

061

0s

6862

L0 5181

8E 0002

91 888

19 S

ozz 11

LS91

62

S6L1

09 2661

Sz

0661 E6

1ZLI

22

0912

89 0

20

2

06 9E

lZ

zS E061

8S 0S61

E1 ISg1

L2 0581

LI 0E61

E9 6E

oz 81

E061 5

21

8E

02 01

00

02

22

L

SZ

Z

SE 2S61

9I 888

11 LS91

62 IZLI

11 StL

1 62

06L1

22

2691

11 E

10

2

L0 s181

69 8612

2E

0861

1s 1922

as ueapq

I

ES 0261

0s

8LLI

E1

9L

gI 6E

0P91 5

2

8E6I

s2

6281

SI 1L61

66 S061

60 6Soz

zE

EL

61 L0

0281 61

8ZLI

91 888

11

LSg1

L6 6291

11

S6LI

62 C

6LI

6E E681

60 6861

8s 6661

1s

1922 62

1S61

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 2 c

ontin

ued

Subj

ect

Bs

M

F

M

DA

L

S

B

SA

M

H

M

GJ

M

S

T

L

B G

C

P M

Mea

n

Cal

orim

etry

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Perio

d be

fore

gl

ucos

e lo

ad

(fas

ting)

Peri

od a

fter

glu

cose

load

I

__h_7

Mea

n

150

148

214

180

181

133

I75

181

184

I77

166

140

162

163

I59

171

167

166

147

I59

178

I 62

212

20

0

SE I I

07

86

6-5

32

29

43

29

14

68

18

I 8

4

0 50

I 4

32

25

29

I 4

71

45

22

22

11

2 3

7- ----

4 5

M

ean

SE

Mea

n

(b)

Obe

se s

ubje

cts

148

158

184

I77

142

207

I74

I75

185

I 86

16

6

148

I59

161

I 63

163

161

I 60

I4

3

I 60

176

163

210

20

0

I 8

1 4

5

4 2

2

11

25

40

3

2 4

7 5

4 4

3 1

8

I 4

104

2

9 2

5

43

22

32

I 8

6

8 4

3

11

11

SE

J 4

0

4

25

I 4

43

25

1 4

5

4

25

65

I I

65

11

11

32

29

36

07

I 8

I 4

29

1 4

6

1 3

7

Mea

n

148

I55

181

171

I44

20

I

20 I

176

176

I74

175

162

144

I 60

163

162

167

164

156

142

I97

I 60

I73

162

SE

07

32

68

32

32

43

I 4

47

3 2

1 4

1 4

0

7 3

2 I 4

1

4 2

9 2

2

1 4

61

3 7

36

22

22

11

Mea

n (

initi

al

rate

)+

+ 2

5 +

64

+ 11

8 -

35

+ 3

9 -

68

+ 81

+ 101

f11

4 - 5

5

+ 5

6 - 2

6

+ 4

0

+ 4

2 +

62

- 1

7 +

56

- 7

6 +

38

0

- 22

- 2

2

+ 5

2k

I3

-

05k

17

Fo

r de

tails

of

expe

rim

enta

l pro

cedu

res

see

p 282

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing

150

min

aft

er g

luco

se lo

ad a

s a

perc

enta

ge of

the

fast

ing

met

abol

ic ra

te (b

efor

e gl

ucos

e lo

ad)

w

cn Y

P F N

h

a N 3

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 5: Thermic effect of glucose in obese subjects studied by

Tab

le 2 c

ontin

ued

Subj

ect

Bs

M

F

M

DA

L

S

B

SA

M

H

M

GJ

M

S

T

L

B G

C

P M

Mea

n

Cal

orim

etry

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Ind

Dir

In

d D

ir

Perio

d be

fore

gl

ucos

e lo

ad

(fas

ting)

Peri

od a

fter

glu

cose

load

I

__h_7

Mea

n

150

148

214

180

181

133

I75

181

184

I77

166

140

162

163

I59

171

167

166

147

I59

178

I 62

212

20

0

SE I I

07

86

6-5

32

29

43

29

14

68

18

I 8

4

0 50

I 4

32

25

29

I 4

71

45

22

22

11

2 3

7- ----

4 5

M

ean

SE

Mea

n

(b)

Obe

se s

ubje

cts

148

158

184

I77

142

207

I74

I75

185

I 86

16

6

148

I59

161

I 63

163

161

I 60

I4

3

I 60

176

163

210

20

0

I 8

1 4

5

4 2

2

11

25

40

3

2 4

7 5

4 4

3 1

8

I 4

104

2

9 2

5

43

22

32

I 8

6

8 4

3

11

11

SE

J 4

0

4

25

I 4

43

25

1 4

5

4

25

65

I I

65

11

11

32

29

36

07

I 8

I 4

29

1 4

6

1 3

7

Mea

n

148

I55

181

171

I44

20

I

20 I

176

176

I74

175

162

144

I 60

163

162

167

164

156

142

I97

I 60

I73

162

SE

07

32

68

32

32

43

I 4

47

3 2

1 4

1 4

0

7 3

2 I 4

1

4 2

9 2

2

1 4

61

3 7

36

22

22

11

Mea

n (

initi

al

rate

)+

+ 2

5 +

64

+ 11

8 -

35

+ 3

9 -

68

+ 81

+ 101

f11

4 - 5

5

+ 5

6 - 2

6

+ 4

0

+ 4

2 +

62

- 1

7 +

56

- 7

6 +

38

0

- 22

- 2

2

+ 5

2k

I3

-

05k

17

Fo

r de

tails

of

expe

rim

enta

l pro

cedu

res

see

p 282

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing

150

min

aft

er g

luco

se lo

ad a

s a

perc

enta

ge of

the

fast

ing

met

abol

ic ra

te (b

efor

e gl

ucos

e lo

ad)

w

cn Y

P F N

h

a N 3

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 6: Thermic effect of glucose in obese subjects studied by

286 PH PITTET AND OTHERS I976

Table 3 Comparison of the metabolic rates and of the total heat losses before and after an oral loadof 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectst

(Mean values with their standard errors)

Control Obese Periodof amp

measurement Mean SE Mean SE

Metabolic rate (kJm2 per h) B 1778 50 1641 48Ns A 2008 59 1726 6-1

Total heat losses (kJm2 per h) B 1920 53 1626 39 A 1899 46 1618 37

Change in metabolic rate (yo initial rate)l f13o 15 f 52 I3

Change in total heat losses ( initial rate)$ -10 15 -0 5 1 7 ~ ~

Ns not significant B before glucose load (fasting period) A during 150 min after glucose load Differences between mean values for the two groups were statistically significant (unpaired t test)

t For details of subjects see Table I and for details of experimental procedures see p 282 1 Mean change during 150 min after glucose load as a percentage of the fasting value (before glucose

Pltooog PltOOOI

load)

Temperatures The values for the internal and skin temperatures are given in Table 5 After the

glucose load the internal temperature of the control subjects tended to increase (+ o - I S O ) whereas the mean skin temperature remained unchanged

In the obese subjects the glucose load did not cause any significant increase in either the internal or the skin temperature

The results given in Table 5 indicate that the internal temperature of the control subjects was significantly higher than that of the obese subjects during the post- prandial period Furthermore the mean skin temperature of the control group was significantly higher than that of the obese group and remained higher throughout the postprandial period

Thermal balance All control subjects were in negative thermal balance (heat production minus heat

losses) during the fasting period (- 142 +_ 50 kJm2 per h) ie their heat losses were greater than their heat production (Fig 2) After the glucose load the amount of heat stored increased significantly and the thermal balance became positive ( + 108 56 kJmz per h) However the obese subjects were already in slight positive balance in the preprandial state (+ 14 k $8 kJm2 per h) This positive balance was further re- inforced after the glucose load ( + 108 amp $8 kJm2 per h)

Energy-free meal The possibility that the effects we have found were caused by the glucose carrier

were minimal However three control experiments were done using three of the lean control subjects to study the effect of an oral load of the same volume but without glucose ie 150 ml water + lemon juice No significant changes in metabolic rate or

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 7: Thermic effect of glucose in obese subjects studied by

VOl 35 Thermic effect of glucose in obese subjects

C L P C L P - - Fasting After glucose

E 150 E v

p 100 c 0 - 48

50 9 t

0 C L P C L P - -

Fasting After glucose

Fig I Mean oxidation rates (mglmin) of different substrates (C carbohydrates L lipids P proteins) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for (a) group of ten female control subjects and for (b) group of eleven female obese subjects The increase in the glucose oxidation rate after the glucose load was statistically significant for both groups P lt O ~ O I Vertical bars represent the standard errors of the mean values

total heat losses were found (Table 6) The negative thermal balance obtained in the fasting period persisted throughout the experiment Mean skin temperature was not modified but the internal temperature decreased significantly during the experiment (Table 7) confirming the negative thermal balance The fasting oxidation rates of the substrates were similar to those obtained in the first part of this experiment and were not significantly modified after ingestion of the energy-free meal (Table 7)

On the basis of these results we can conclude that an energy-free meal of 150 ml water + lemon juice had no significant effect on metabolic rate Therefore the findings reported in the first part of this study were due to the glucose load

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 8: Thermic effect of glucose in obese subjects studied by

288 PH PITTET AND OTHERS I976

Table 4 Mean oxidation rate (mgimin) of carbohydrate fa t ty ac ih (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after an oral load of 50 g glucose for a group of ten female control subjects and for a group of eleven female obese subjectsf

(Mean values with their standard errors)

Metabolites oxidized A

f Carbohydrate Fatty acids Amino acids

Subjects measurement Mean SE Mean SE Mean SE Periodof pA------- amp

Control B 904 34 688 61 348 30 A 1591 86 518 5 ~ ~ 296 3gN8

ttt ttt NS

Obese B 502 75 1037 60 26 44 A 956 II~ 886 70~ 300 59N8

N8 not significant B before glucose load (fasting period) A during 150 min after glucose load Changes in mean oxidation rates for each metabolite for period A were statistically significant

Differences between mean values for oxidation rates for the two groups for period B and for period A

$ For details of subjects see Table I and for details of experimental procedures see p 282

Pltoor Plt0005 I Pltooor

were statistically significant (unpaired t test) ttt P lt 0001

Table 5 Internal temperature (Tint) () and mean skin temperature (TCut) () measured during the 40 min controlperiod (fasting) (initial T) and during theJina110 min ($nu1 T) of the 150 min period after an oral dose of 5 0 g glucose for a group of ten female control subjects and for a group of eleven female obese subjects

(Mean values with their standard errors)

Initial T Final T Change in T -amp-

Mean SE Mean SE Mean SE

3680 010 3698 010 018 005

3671 005 3672 004 001 004 Tint NS

3332 008 3372 009 000 0 0 2

I ) Tcut

Control

Obese Control

Obese 3329 011 3341 011 012 005

The change in T for Ts and for Tt for both groups was not statistically significant (paired t test) NS not significant Differences between mean values for Tjat and for Tcut for the two groups were statistically sig-

t For details of subjects see Table I and for details of experimental procedures see p 282 nificant (unpaired t test) Pio05 PltOOZ

D I S C U S S I O N

Obesity is not always the consequence of an absolute excess of energy intake this excess can be relative if it is accompanied by a decreased energy expenditure The main factors contributing to decreased energy expenditure in obese subjects are (a) de- creased physical activity ( 6 ) increased energy efficiency of the utilization of nutrients or (c) a lack of thermogenic stimulation or both b and c

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 9: Thermic effect of glucose in obese subjects studied by

VOl 35 Thermic effect of glucose in obese subjects

Fasting After glucose

+8

+4 h c N E - 0 s m 0 -4

aJ bn

Y YI

c1

aJ I -8

-12

-16

289

Fig 2 Mean thermal balance (heat storage (kJm2 per h) ie heat production minus heat losses) during a 40 min control period (fasting) and during 150 min after an oral dose of 50 g glucose for a group of ten female control subjects (0) and for a group of eleven female obese subjects (El) The change in thermal balance after the glucose load was statistically sig- nificant for control group P lt 0001 for obese group P lt 0-05 Vertical bars represent the standard errors of the mean values

It is not the aim of this paper to discuss factors a and b extensively Decreased physical activity in the obese individual has often been described Bjorntorp (1966) has found that lipogenesis from glucose is twice as high in the obese as in normal-weight subjects There are a few biochemical pathways in intermediary metabolism which can modify energy efficiency ie synthesis of fatty acids from glucose and oxidation of the glucose produced (Ball amp Jungas 1964) the esterification-hydrolysis cycle for fatty acids (Ball I 965) and the glycerophosphate-dihydroxyacetone cycle (Galton amp

Lack of thermogenic stimulation in the obese individual can be accounted for by a weak dietary-induced thermic effect and by a low metabolic response to cold

The results of this study indicate that at an ambient temperature of 2 8 O obese subjects have smaller heat losses than control subjects These results are in agreement with those of Gygax Pittet amp JCquier (1972) and JCquier Gygax Pittet amp Vannotti (1974) who reported similar findings at ambient temperatures of 28 and 20 after 2 h exposure at 20deg the control subjects tended to stimulate their metabolic rate but the obese subjects did not

Our results indicate that although both obese and control subjects stimulate their metabolic heat production after an oral load of 50 g glucose this thermic effect was lower in the obese group (only half that of the control group)

The fasting oxidation rates of the different substrates were different in the obese subjects when compared with the control subjects The latter oxidized more carbo-

Bray 1967)

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 10: Thermic effect of glucose in obese subjects studied by

Tab

le 6

Hea

t pro

duct

ion

(kJ

m2 p

er h

) obt

aine

d by

ind

irec

t cal

orim

etry

(Id

) and

hea

t lo

sses

(kJ

m2 p

er h

) m

easu

red

by d

irect

cal

orim

etry

(D

ir) d

urin

g a

40 m

in c

ontro

l per

iod

(fas

ting

) and

dur

ing J

ive

30 rn

in p

erio

ds a

fter

the i

nges

tion

of a

n en

ergy

-free

mea

l (150 m

l wat

er + l

emon

ju

ice)

for

thre

e fem

ale

cont

rol s

ubje

ctse

(M

ean

valu

es w

ith t

heir

sta

ndar

d er

rors

)

cd

Peri

od b

efor

e en

ergy

-fre

e Pe

riod

aft

er e

nerg

y-fr

ee m

eal

I

mea

l (fa

stin

g)

I 2

3 4

5 M

ean

F F

2 --------

( in

itial

Su

bjec

ts

Cal

orim

etry

M

ean

SE

Mea

n SE

Mea

n SE

M

ean

SB

Mea

n SE

M

ean

SE

Mea

n SE

ra

te)

J In

d D

D

ir

Bo

Ind

F

Dir

B

u In

d F

D

ir

Mea

n In

d D

ir

The

rmal

ba

lanc

e1

173

8 0

6 17

56

38

171

7 20

178

4 28

180

3 4

4 1808

45

177

4 1

7 +

21

17

99

08

179

9 08

178

4 0

7 18

06

10

176

4 0

8 17

56

16

178

1 0

9 -

10

169

5 2

3 17

15

38

168

4 4

6

172

7 7

1 17

68

26

172

7 4

4 17

24

13

+ 1

7 3

173

4 20

174

1 21

174

1 10

1820

09

177

3 05

182

8 20

178

1 1

7 +

27

156

2 8

1 1565

89

151

5 2

3 15

64

50

1552

32

153

7 3

7 15

47

09

- 10

3 17

77

20

17

67

25

175

1 1

7 17

92

09

178

5 11

176

9 08

177

3 0

7 -

02

i3 16

65

53

167

9 5

8 16

39

63

169

2 6

6 17

08

78

169

1 80

1682

69

f 10

E 17

70

19

176

9 1

7 17

59

13

180

6 08

177

4 0

6 17

84

22

177

8 0

3 +

05

-105

55

-97

6

6

F

or d

etai

ls o

f ex

peri

men

tal

proc

edur

es

see

p 2

82

t M

ean

incr

ease

in m

etab

olic

rate

dur

ing 15

0 min

aft

er e

nerg

y-fr

ee m

eal

as a

per

cent

age

of th

e fa

stin

g m

etab

olic

rat

e (b

efor

e en

ergy

-fre

e m

eal)

$

The

dif

fere

nce

betw

een

heat

pro

duct

ion

and

heat

los

ses

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 11: Thermic effect of glucose in obese subjects studied by

VOl 34 Thermic eflect of glucose in obese subjects 291

Table 7 Internal temperature (Tint) () and mean skin temperature (Tamp () measured during the 40 min control period (fasting) and during the Jinal 10 min of the 150 min period after the ingestion of an energy-free meal (150 ml water + lemonjuice) and mean oxidation rate ( ~ m i n ) of carbohydrate fatty acids (lipid) and amino acids (protein) obtained by indirect calorimetry and from urinary nitrogen measurements before and after the meal for three female control subjectst

(Mean values with their standard errors)

Period of measurement lt

Before meal After meal Change in T ampamp-

Mean SE Mean SE Mean SE

Tint Tcut

3671 003 3658 002 -013 0 0 2 ~

3335 016 3340 015 +005 OOINS

Mean oxidation rates Carbohydrate 957 34 936 83NB Fatty acids 764 121 838 1 0 7 ~ ~ Amino acids 468 2 2 310 3SNB

Changes in mean oxidation rates for each metabolite after the energy-free meal were not statistically

NS not significant The change in T for Tint was statistically significant Pltoor t For details of experimental procedures see p z8z

significant

hydrate (90-4 mglmin) than lipid (68-8 mgmin) whereas the fasting obese subjects oxidized more lipid (1037 mgmin) than carbohydrate (502 mglmin) After the glucose load there was an increase in glucose metabolism in both groups but it was much lower in the obese than in the control subjects

Felber Moody amp Vannotti (1965) reported a decrease in the glucbse tolerance of non-obese subjects when the concentration of circulating free fatty acids is increzsed by an infusion of lipids preceding the glucose load Gomez et al (1972) suggested that this glucose intolerance was probably due to inhibition of carbohydrate oxidation rate by metabolites of free fatty acids as described by Randle Hales Garland amp Newsholme (1963) in muscle tissue A similar mechanism could account for our results in obese subjects

In our ambient conditions the thermal balance obtained indicated that the two groups were in a different starting state The fasting control subjects were in negative thermal balance their heat losses were greater than their metabolic heat production and their temperatures tended to decrease On the other hand fasting obese subjects were in thermal balance their metabolic heat production was slightly higher than their heat losses and their body temperature was stable

The ingestion of the glucose induced an important modification in the thermic state of the control subjects from a negative level ie from a deficit of heat they reached a positive thermal balance

In the obese subjects ingestion of the glucose did not cause such an important

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at

Page 12: Thermic effect of glucose in obese subjects studied by

292 PH PITTET AND OTHERS I976 change in their thermic state Glucose ingestion only induced a reinforcement of a pre-existing equilibrated thermal balance

Thus the results of this study suggest that an inefficient stimulation of thermo- genesis might be a factor contributing to the low energy expenditure of the obese individual This decreased dietary-induced thermic effect may be due in part to in- hibition of glucose oxidation by fatty acid metabolites

It remains to be determined whether the equilibrated thermal balance of the obese individual is involved in inhibiting the thermic effect found after glucose ingestion The thermic effect of protein and lipid should also be carefully studied in the obese individual

The authors thank Mrs M Marmier and M Oberti for their skilful technical assistance Financial support from the Nest16 Company Vevey Switzerland is grate- fully acknowledged

R E F E R E N C E S

Ball E G (1965) Ann N Y Acad Sci 131 225 Ball E G amp Jungas R L (1964) Recent Prog Horm Res 20 183 Bjorntorp P (1966) Acta med scand 179 229 Cabanac M amp Duclaux R (1970) Science N Y 168 496 Du Bois E F (1924) Basal Metabolism in Health and Disease 1st ed p 23 Philadelphia Lea and

Felber J-P Moody A J amp Vannotti A (1965) Helv med Acta 32 323 Galton D J amp Bray G A (1967) J clin Endocr Metab 27 1573 Gomez F JCquier E Chabot V Biiber V amp Felber J-P (1972) Metabolism 21 381 Gygax P-H Pittet Ph amp Jtquier E (1972) Experientia 28 728 Hardy J D amp Du Bois E F (1937) J Nutr 15 5 Jtquier E Gygax P-H Pittet Ph amp Vannotti A (1974) J appl Physiol 36 674 Johnson M-L Burke B S amp Mayer J (1956) Am J clin Nutr 4 37 Linton P-H Conley M Kuechenmeister C amp McClusky H (1972) Am J clin Nutr 25 368 Lusk G (1924) J biol Chem 5g41 Miller D S amp Mumford P (1966) Proc Nutr Sac 25 100 Pittet Ph Gygax P-H amp Jkquier E (1974) Br J Nutr 31 343 Randle P J Hales C N Garland P B amp Newsholme E A (1963) Lancet i 785 Spinnler G Jtquier E Favre R Dolivo M amp Vannotti A (1973) J appl Physiol 35 158 Stefanik P-A Heald F-P amp Mayer J (1959) Am J din Nutr 7 5 5 Stirling J L amp Stock M J (1968) Nature Lond 220 801

Febiger

Printed in Great Britain

httpswwwcambridgeorgcoreterms httpsdoiorg101079BJN19760033Downloaded from httpswwwcambridgeorgcore University of Basel Library on 30 May 2017 at 201922 subject to the Cambridge Core terms of use available at