Short-term, mixed-diet overfeeding in man: no evidence for "luxuskonsumption"

Am J Physiol Endocrinol Metab 249:E470-E477, 1985. ; You might find this additional info useful... This article has been cited by 5 other HighWire-hosted articles: http://ajpendo.physiology.org/content/249/5/E470#cited-by Updated information and services including high resolution figures, can be found at: http://ajpendo.physiology.org/content/249/5/E470.full

E. Ravussin, Y. Schutz, K. J. Acheson, M. Dusmet, L. Bourquin and E. Jequier

Additional material and information about American Journal of Physiology - Endocrinology and Metabolism can be found at: http://www.the-aps.org/publications/ajpendo This information is current as of May 11, 2013.

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

American Journal of Physiology - Endocrinology and Metabolism publishes results of original studies about endocrine and metabolic systems on any level of organization. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright 1985 the American Physiological Society. ISSN: 0193-1849, ESSN: 1522-1555. Visit our website at http://www.the-aps.org/.

Short-term, mixed-diet overfeeding in man: no evidence for luxuskonsumption

E. RAVUSSIN, Y. SCHUTZ, AND E. JEQUIER
Institute
of Physiology,

K. J. ACHESON,
of Lausanne,

M. DUSMET,

L. BOURQUIN,

University

CH-1005 Lausanne, Switzerland

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

RAVUSSIN, E., Y. SCHUTZ, K. J. ACHESON, M. DUSMET, L. dxed-diet ouerfeeding in man: no evidence for luxuhnsumption. Am. J. Physiol. 249 (Endocrinol. Metab. 12): E470-E477,1985.-After 13 days of weight maintenance diet (13,720 t 620 kJ/day, 40% fat, 15% protein, and 45% carbohydrate), five young men (71.3 t 7.1 kg, 181 t 8 cm; means t SD) were overfed for 9 days at 1.6 times their maintenance requirements (i.e., +8,010 kJ/day). Twentyfour-hour energy expenditure (24-h EE) and basal metabolic rate (BMR) were measured on three occasions, once after 10 days on the weight-maintenance diet and after 2 and 9 days of overfeeding. Physical activity was monitored throughout the study, body composition was measured by underwater weighing, and nitrogen balance was assessed for 3 days during the two experimental periods. Overfeeding caused an increase in body weight averaging 3.2 kg of which 56% was fat as measured by underwater weighing. After 9 days of overfeeding, BMR increased by 622 kJ/day, which could explain one-third of the increase in 24-h EE (2,038 kJ/day); the remainder was due to the thermic effect of food (which increased in proportion with excess energy intake) and the increased cost of physical activity, related to body weight gain. This study shows that approximately one-quarter of the excess energy intake was dissipated through an increase in EE, with 75% being stored in the body. Under our experimental conditions of mixed overfeeding in which body composition measurements were combined with those of energy balance, it was possible to account for all of the energy ingested in excess of maintenance requirements.

BOURQUIN, AND E. J~QUIER. Short-term,

results and noticed that his body weight was influenced by energy intake (EI) and changed in a totally predictable manner. Webb (31) has proposed a form of unmeasured energy, but Garrow (9) has claimed that the techniques for measuring energy balances are sufficiently inaccurate to allow for these apparent errors in energy balance studies. Most of the cited studies were designed to investigate the effect of overfeeding on basal metabolic rate (BMR), physical activity, the thermic effect of a meal, and the cost of physical activity. However, total 24-h EE during overfeeding has rarely been measured (1, 3, 26, 32). Therefore this study was undertaken, by using a respiration chamber, to measure 24-h EE during weight maintenance and after 2 and 9 days of overfeeding at 160% of weight maintenance EI. BMR, sleeping metabolic rate, spontaneous daily physical activity, and the thermic effect of the meals were also measured. This study suggests that all the excess EI is accounted for by an increase in energy stores and that the increase in total EE can be explained by an increase in BMR in relationship with the increase in fat-free mass (FFM), an increased thermic effect of food proportional to the excess EI, and an increased cost of physical activity related to the body weight gain.
METHODS

twenty-four-hour physical activity;

energy expenditure; thermic effect of food

indirect

calorimetry;

AT THE BEGINNING of the century, Neumann (16) observed that the increase in body weight, in response to hypercaloric diets, was not proportional to the excess energy ingested. To explain this discrepancy he hypothesized that some of the energy ingested in excess of normal requirements was dissipated as heat or luxuskonsumption. This concept was revived later by Miller et al. (15) and was supported by the findings of the Vermont studies where it was observed that lean volunteers who gained weight while overeating required 50% more energy to maintain their new body weights than their previous maintenance intake (29). However, controversy still exists, because Glick et al. (10) failed to find an increase in energy expenditure (EE) in response to overfeeding and others who have observed this phenomenon deny that it is due to luxuskonsumption (17). Furthermore, Forbes (6) recently reexamined Neumanns E470

Subjects. Five healthy male subjects with an average age of 24 yr (range 22-27 yr), height 181 cm (range 172190 cm), weight 71.3 kg (range 65.3-83.5 kg), percent body fat 14.6% (range 9.3-23.4%), and body mass indexes [wt (kg)/ht2 (m2)] of 22.2, 21.8, 22.5, 23.9, and 19.0 volunteered for the study after reading a detailed account of the protocol and discussing it with the investigators. The five subjects were known to have maintained their body weight essentially constant over 2 yr before the study. Besides having no family history of diabetes mellitus or obesity, the subjects were found to be healthy by a full medical examination before being considered for the experiment. None of the subjects were smokers and none were taking any medication before or during the experiment. The protocol had previously been reviewed and accepted by the hospital ethical committee. Experimental protocol. Basically the protocol consisted of studying the subjects at the end of a 2-wk period during which their body weights had been maintained
0193-1849/85 $1.50 Copyright
0

1985 the American

Physiological

Society

SHORT-TERM,
Day

MIXED-DIET

OVERFEEDING

IN MAN

E471
TABLE 1. Mean daily energy and nutrient intak,es in five volunteers during a period of 14 days of weight maintenance and 9 days of mixed diet overfeeding
Weight-Maintaining Diet Overfeeding

(MJ)

Diet

Body Body Activity Urine

weight composition indexes collections

1 ;

24 EE + BMR

cl

cl

. 0

FIG. 1. Experimental metabolic rate.

design. EE, energy expenditure;

BMR, basal

Free ranging Energy, kJ Protein, g Fat, g Carbohydrate, g Respiration chamber Energy, kJ Protein, g Fat, g Carbohydrate, g

13,720f620 123k6 146k7 369t17 10,940+520 98t5 116,t5 294k14

21,730*970 195t9 231tlO 584k26 17,410+880 156&B 184,tlO 467zk24

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

stable with a balanced diet composed of 40% fat, 15% protein, and 45% carbohydrate energy. They were then studied on two further occasions after their EI had been increased by 60% above their maintenance energy requi .rements, i.e., after 2 and 9 days of overfeeding (Fig. 1) . On the 1st day of the experiment the subjects spent 24 h in a respiration chamber during which they received a balanced diet providing 172 kJ/kg FFM and their 24h EE was measured. This preliminary study was used, not only to determine each individuals free-ranging energy requirements, calculated as EE within the chamber plus an estimated 25% for physical activity while outside the chamber (unpublished observation), but also to accustom them to the chamber and the conditions of the experiment. During the weight-maintenance period, it was necessary to increase the food intake of two subjects (P.V. and M.K.) by -400 kJ/day to maintain their body weights. On average the subjects energy requirements were 13,720 t 620 kJ/day while free ranging and 10,940 t 520 kJ/ day while in the chamber (Table 1). During overfeeding the subjects free-ranging EI was 21,730 t 970 kJ and 17,410 t 880 kJ/day while in the chamber. During the entire experi *mental period the subjects received three meals a day composed of natural foods, which were prepared by a trained dietitian. The breakfast was composed of bread, butter and marmalade, or breakfast cereals served with decaffeinated coffee or chocolate and milk. The lunch and dinner varied from day to day and were composed of red meats, poultry, fish, or eggs and fresh or frozen green vegetables, pulses, white and whole wheat bread, dairy products (cheese and yogurt), and fresh and canned fruits. Drinks consisted of fruit juices with or without added sucrose, tea ., and milk. Water was allowed ad libitum. All . meals were taken at the Insti tute of Physiology under supervision to ensure that all the food was eaten. Body weight was measured on a Florenz torsion balance (Braunau-Wien, Austria, 200 kg t 50 g) every morning after voiding. Urine was collected continuously for 3 consecutive days during weight maintenance (days -5 to -3 inclusive) and overfeeding (days 6-8 inclusive). Body composition was estimated by underwater weighing with simultaneous underwater residual volume determination (helium dilution) on the last day of the weight

maintenance and overfeeding phases of the experiment (11). During the 3 wk of the study physical activity was monitored and recorded twice a day using a pedometer worn at the waist and an accelerometer on the nondominant wrist. Energy expenditure measurements. Measurements of 24-h EE were performed in a comfortable room (5 m long, 2.5 m wide, and 2.5 m high with a net vol of 30,600 1) constructed as a large open-circuit indirect calorimeter (13, 22). Measurements were made continuously for 22.5 h from 8:00 to 6:30 A.M. the next day. Although no vigorous physical activity was permitted in the chamber, spontaneous physical activity was estimated by radar (28). The following morning postaborptive resting metabolic rate was measured for 1 h, using an open-circuit ventilated hood, the last 30 min of which were used in the results that we have referred to as BMR in the remainder of the text, in spite of the conditions of overfeeding. At the end of this period, 25 ml of blood was taken for substrate and hormone analyses. Analyses. The blood samples were analyzed for glucose using the glucose oxidase method (Beckman glucose analyzer II, Beckman, Fullerton, NY) and free fatty acids on the Dole extract. Hormone analyses included insulin (IRI), catecholamines (norepinephrine and epinephrine) using an high-pressure liquid chromatography technique, and thyroid hormones (T3, radioimmunoassay). The 24-h urine collections were divided into day and night samples, each of which was analyzed for nitrogen using the Kjeldahl method and catecholamine excretion. Data analysis. Energy intake and food composition was determined from food tables prepared by a large Swiss food company and by values provided by the manufacturers (Nutritive Values of food, Migros) and entered onto a Hewlett-Packard desktop computer (HP-9830, Hewlett-Packard, Palo Alto, CA). The metabolizable energy of the n utrients was taken to be 4.0 kcal/g for carbohydrate, 4.0 kcal /g f 0 r protei n, and 9.0 kcal/g for fat. All foods used in the experiment were purchased at their outlets. Nitrogen balance was assessed during a 3day nitrogen intake and excretion period during both the weight-maintenance and overfeeding periods (Fig. 1). A value of 5 mg/kg body wt per day for integumental losses was taken into account in the total nitrogen losses (2) . In the respiration chamber oxygen consumption ., car-

E472

RAVUSSIN

ET AL.

bon dioxide production, and spontaneous physical activity were continuously measured, and the variables were processed using a Hewlett-Packard data acquisition system (HP-3052A and HP-9825A). For data presentation these values have been integrated over l-h periods. EE was calculated from the overall respiratory quotient (RQ) and oxygen consumption (Vo2) by using the formula EE (kcal/min) (RQ - 0.707) 0.293

includes not only changes in BMR and the thermic effect of food but also the increased cost of physical activity due to a greater body weight and possible modifications of activity. It also represents the inefficiency of energy retention. The data are presented as means t SE, and paired t test analyses were used to compare the weight-maintenance period with that of overfeeding.
RESULTS Body weight and activity indexes. Figure 2 shows that body weight was maintained within 0.5 kg during the 14 days of the base-line period. The average weight for the last 2 days of this weight-maintenance period was 71.32 t 3.17 kg of which 14.6 t 2.7% was fat. Overfeeding produced an initial rapid weight gain that became progressively slower. Total weight gain after 8 days of overfeeding was 3.21 t 0.26 kg, i.e., an increase of 4.5% in body weight (Table 2). At this time body fat, measured 4 I

1 1
x 0.361

x voz

where 4.686 kcal/l is the energy value of 1 liter 02 at a nonprotein RQ of 0.707; RQ is the measured respiratory quotient; 0.707 is the RQ when only fat is oxidized; 0.293 is the difference between the RQ for carbohydrate and fat oxidation; 0.361 is the difference in energy value of a liter of oxygen between an RQ of 1 and that of 0.707; and VOW (l/min) is the rate of oxygen consumption at STPD conditions. Spontaneous physical activity measured by radar in the chamber was averaged over the 24-h period and represents the proportion of time during which movements were detected over the 24-h period (28). Overall thermogenic response to the three meals (breakfast, lunch, and dinner) was calculated as previously described (27). In brief, the mean resting EE for the whole day (16 h), which includes the thermic effect of the meals, was obtained by the intercept of the regression line for EE (indirect calorimetry) versus percent physical activity (radar system). The difference between this value and basal metabolic rate represents an estimate of the thermic effect of the three meals; the latter was computed for a 16-h period. Fraction of excess energy intake dissipated. One other way to assess the overall response to overfeeding is to express the percent of the excess EI (AEI) which is expended over 24 h, i.e., (AEE/AEI) X 100. This value
TABLE 2. Body weight, 24-h energy expenditure,

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

n-5

I I I w-e--

1~1 - 0

-e--v

------------

I .

-15

-10

-5

5 Days

10
and 9

FIG. 2. Body wt change during 14 days of wt maintenance days of mixed diet overfeeding (means t SE).

and basal metabolic rate in five lean subjects

Subjects Weight Maintenance Z-Day Overfeeding g-Day Overfeeding

Body weight, kg

K.J. P.V. T.V. JP.G. M.K. K.J. P.V. T.V. JP.G. M.K. K.J. P.V. T.V. JP.G. M.K.

Mean SE Basal metabolic rate, kJ/day

Mean SE
24-h

Energy expenditure,

kJ/day

Mean SE * Statistically different from weight-maintenance

65.3 71.1 68.3 83.7 68.4 71.3 3.2 7,410 7,473 7,289 9,761 7,711 7,929 463 8,673 9,552 9,422 11,251 9,858 9,751 423

66.5 72.7 69.3 84.6 69.3 72.5t 3.2 7,711 8,075 7,230 9,941 7,774 8,146 469 9,096 9,916 10,192 12,602 10,309 10,423* 584

69.0 74.9
71.4 85.9 71.4 74.5t 2.9 8,736 8,494 7,832 9,761 8,134 8,591* 330 12,715 10,962 10,661 13,213 11,397 11,789* 500

period, P

0.05.

t Statistically

different from weight-maintenance

period, P <

0.01.

SHORT-TERM,

MIXED-DIET

OVERFEEDING

IN

MAN

E473

by underwater weighing, represented 16.4 t 2.7% of body weight, thus indicating an increase in fat mass of 1.8 kg, i.e., 57% of the total body weight gain. During weight maintenance daily nitrogen intake was 19.7 t 0.9 g/day and urinary nitrogen output was 16.3 t 1.6 g/day (3-day measurement). During overfeeding nitrogen intake was 31.2 t 1.4, whereas urinary output increased to 22.7 t 1.6 g/day. Figure 3 shows that activity, measured by pedometers (top) or accelerometers (bottom), was not altered during overfeeding in the free-ranging conditions (7 days of overfeeding vs. last 7 days of wt maintenance). Energy expenditure. BMR was increased in all but one subject after 2 days of overfeeding (Table 2; avg 1.7%, NS). After 9 days of overfeeding there was a significant increase in BMR from 5.52 t 0.33 to 5.94 t 0.21 kJ/min (P < 0.05), i.e., a 7.6% increase. Twenty-four-hour EE was increased after 9 days of overfeeding (Table 2) (Figs. 4 and 5). Twenty-four-hour EE was already increased by 7% on the 2nd day of overfeeding (10,423 t 584 vs. 9,751 t 423 kJ/day; P < 0.05; Table 2) and by 21% on the 9th day of overfeeding (11,789 k 500 kJ/day; P < 0.05, Table 2, Fig. 5). This corresponded to an increase in diurnal (8 A.M. to bedtime) EE averaging 7.60 t 0.37, 8.16 t 0.46, and 9.23 t 0.40 kJ/min and in sleeping EE 4.80 t 0.29, 5.10 t 0.46, and 5.53 t 0.32 kJ/min during weight maintenance and after 2 and 9 days of overfeeding, respectively (Table 3).

meal

meal

I
1 I I

meal

I
I1

:::. j: 1:: ::: . ... . ::. :: ::::: ._~.,, ::::: 1.:. .:::: .:.: .:.: :: : .:_ ..~....._ : :.:.:.:.:.: :... ::. .:.._ :::::.:j::.:.:. :,~ :,:, :...:. ,_,. ,__. b
-sleep -

ot 8

10

12

14

16

18

20

11

22

24

Time FIG. 4. Time course of oxygen and night. Dotted area represents overfeeding over weight-maintenance

( hours

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

consumption (VO& throughout day excess Voz after 9 days of mixed-diet base line.

-Overfeeding

65
24 EE 24EE

75
BODY

80

I 85
(Kg)

WEIGHT

5. Individual changes in 24-h (A- --A) and basal (C--O) energy expenditures (EE) expressed in kJ/min as a function of changes in body wt caused by 9 days of overfeeding. The 2 regression lines were calculated for preoverfeeding values of basal metabolic rate (BMR) (dashed line) and 24-h EE (continuous line). Note that when BMR is considered, slope of preoverfeeding values (0) are consistent with changes induced by overfeeding, whereas it is not the case for 24-h EE (A), i.e., EE increased more than body wt.
FIG.

-8

-5

-2

2
TIME

5
( days 1

FIG. 3. Average pattern of activity index throughout study obtained with pedometers (top) and accelerometers (bottom). Note low-activity value obtained in chamber with pedometer (24-h energy expenditure).

The relationships between body weight and the preoverfeeding values of BMR and 24-h EE are shown in Fig. 5. The values of BMR after 9 days of overfeeding fit the preoverfeeding regression line well, whereas this is not the case for 24-h EE after overfeeding, all values being above the regression line. The average 24-h RQ was 0.847 -t 0.010 during weight maintenance and increased to 0.902 t 0.013 (P < 0.01) and 0.918 t 0.007 (P < 0.01) after 2 and 9 days of overfeeding, respectively. Spontaneous physical activity, as measured by radar in the chamber, was unaltered by overfeeding (7.1 t 0.9% maintenance, 7.8 t 0.6% 2nd day, and 9.8 t 1.8% 9th day of overfeeding, NS). The tendency to an increase was accounted for by a single subject (K.J.) whose activity increased from 4.2 to 16.6%. After interview this subject admitted that he had danced in the chamber for 2 h during the evening, elated by the termination of the study. The estimated cost of physical activity as measured by the slope of the regression line between energy expenditure (kJ/min) and spontaneous physical activity

E474
TABLE 3. Mean energy expenditures, thermic effect of food, and cost of physical activity during weight maintenance and after 2 or 9 days of mixed diet overfeeding
Weight Maintenance 2-Day Overfeeding

RAVUSSIN

ET

AL.

g-Day

Overfeeding

Energy intake, kJ/day Diurnal energy expenditure, kJ/min Sleeping energy expenditure, kJ/min Basal metabolic rate, kJ/min Energy expenditure without activity, kJ / min Thermic effect of food, kJ/16 h Thermic effect of food / energy intake, % Cost of spontaneous physical activity, kJ/min

10,937-r-502

per %

7.60t0.37 4.80k0.29 5.52t0.33 6.73t0.29 1,176+152 10.8t1.4 0.026t0.003

17,451+853t 8.16kO.46" 5.10t0.46 5.65t0.33 7.20k0.44 1,477+159t 8.520.8 0.026t0.004

17,359+887t 9.23kO.40" 5.53t0.32" 5.94t0.21" 7.8OkO.41" 1,757+284t 10.2k1.6 0.031t0.004

Diurnal energy expenditure was measured from 8 A.M. UntiLbedtime. Energy expenditure without activity was calculated as the y-intercept of the regression line of 64 points (15-min periods) between the amount of activity (%, x-axis) measured by radar and energy expenditure (kJ/min, y-axis) (28). Thermic effect of food was calculated as the difference between the energy expenditure without activity and the basal metabolic rate over 16 h (28). Cost of spontaneous physical activity was slope of the regression line between the amount of activity (%) and energy expenditure (kJ/min) (64 points). * Statistically different from weight maintenance period, P < 0.05. t Statistically different from weightmaintenance period, P < 0.01.

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

(%) tended to increase from 0.026 during weight maintenance to 0.031 kJ/min per percent activity after 9 days of overfeeding (Table. 3). When the effect of activity was subtracted, i.e., the EE at 0% activity obtained by regression analysis, the mean resting energy expenditure from 8 A.M. to 12 P.M. increased significantly from 6.73 t 0.29 (wt maintenance) to 7.20 t 0.44 and 7.80 t 0.41 kJ/min after 2 and 9 days of overfeeding, i.e., increases of 6.9 and 15.8%) respectively. The thermic effect of the three meals increased significantly (P < 0.05) from 1,176 -t 152 to 1,477 t 159 and 1,757 t 284 kJ/16 h after 2 and 9 days of mixed overfeeding, respectively (Table 3). However, when expressed as a percentage of energy intake, the total daily thermic effect of food was unchanged (10.8 t 1.4, 8.5 t 0.8, and 10.2 t 1.6%, respectively, Table 3). Fraction of excess intake dissipated. After 9 days of overfeeding, the mean percentage of the excess EI oxidized [ (AEE/AEI) x 1001 was 33 t lo%, indicating that 67% of the excess EI was stored, whereas 33% was dissipated. The individual values were 73,22,22,26, and 24% for K.J., P.V., T.V., JP.G., and M.K., respectively. If the atypical activity of subject K.J. is excluded, the average value for the four other subjects is -24%, which is similar to that of K.J. when corrected for his increased activity. Blood and urinary parameters are presented in Table 4. Plasma glucose concentrations were unaffected by overfeeding, whereas insulin concentrations were significantly increased (P < 0.05) already after 2 days of overfeeding (Table 4). Plasma free fatty acid concentrations decreased significantly (P < 0.001) after 2 days of overfeeding. Thyroid hormone concentrations and norepinephrine and epinephrine concentrations were unaffected by 9 days of overfeeding. Similarly, urinary excretion rates of norepinephrine and epinephrine were unchanged, although there was a tendency for them to increase with 1verfeeding during the day as well as during the night.
DISCUSSION

TABLE 4. Effect of mixed-diet overfeeding on postabsorptive circulating levels of substrates and hormones and on urinary norepinephrine and epinephrine excretion rates
Weight Maintenance After 2 Days of Overfeeding After 9 Days of Overfeeding

Plasma levels Free fatty acids, pmol/l Glucose, mg/dl Insulin, pU/ml Total Td, nmol/l Free T1, pmol/l Uptake TB, % Total TB, nmol/l Free TI, pmol/l Reverse TB, nmol/l TSH, $J/ml Norepinephrine, pg/ml Epinephrine, pg/ml Urinary excretion rates, M/h Norepinephrine
Day

418-t47 92t2 8.5t0.7 98,t4 23.7k3.1 31tl 2.62AO.20 6.8kO.l 0.36kO.02 2.3k0.4 164k5 68&6

289&20-f 94t2

11.3k1.5' 97k5 21.5zk2.0 30&l 2.59kO.27 6.7kO.2 0.31kO.02 2.5kO.4 152t9 69t7

2952153: 94&l 11.7&2.0* 100t5 21.423.1 3Okl 2.73zk0.22 6.8k0.3 0.28t0.03 3.OkO.6 177kll 65k7

Night Epinephrine
Day

1.67kO.25 0.72t0.05 0.35kO.04 0.06kO.01

1.71t0.20 0.82t0.08
0.30&0.09

1.90k0.28 0.98kO. 12
0.4ot,o.o9

Night TSH, < 0.05. thyrotropin. 7 Statistically

0.10t0.04

O.llIkO.04

* Statistically different from

different weight maintenance P weight maintenance P < 0.01.

Numereous

studies of overfeeding

in man (4, 12, 14-

16,29) have s lown that the body weight gain of overfed

subjects was much smaller than the minimum expected weight gain if all the excess energy intake was stored as fat. In the classical study of Miller et al. (15) small weight gains were observed in the overfed volunteers. This study revived Neumanns (16) old concept of luxuskonsumption, suggesting that thin people, in response to overfeeding, can induce some metabolically inefficient mechanism that burns off part or all of the excess ingested energy. In most of the cited studies it can, however, be questioned whether physical activity increased simultaneously with overfeeding. In the present study body weight gain was 3.2 kg after 9 days of overfeeding with an excess energy intake of -74 MJ over maintenance requirements (Table 5). Furthermore, physical activity was unaltered during overfeeding as indicated by the relatively crude techniques

SHORT-TERM,

MIXED-DIET

OVERFEEDING

IN

MAN

E475

ferent possibi1itie.s. Both BMR and 24-h EE were increased after 9 days of overfeeding: BMR increased by +%, whereas body weight increased by 4.5% and FFM, the most metabolically active tissue, by only -2.5%. The Energy intake for wt maintenance, MJ/9 day 112.4 observed increase in BMR with overfeeding is in good Energy intake during overfeeding, MJ/9 day 186.9 agreement with numerous studies (1, 14, 17, 19, 20, 26, Excess energy intake above maintenance, 74.4 MJ/9 day 30) except one (10). However, when expressed on the Estimated increase in energy expenditure in 17.8 basis of FFM, BMR did not change significantly with response to overfeeding, MJ/9 day* overfeeding: 130 t 3 during weight maintenance vs. 138 Excess energy intake above energy expenditure, 56.6 t 4 kJ/kg FFM x day during overfeeding. We can MJ/9 day therefore conclude that the increase in BMR is mainly Body wt gain, kg 3.21 related to the increase in the metabolically active tissue Energy equivalent of wt gain, kJ/g 17.6 mass. It is, however, impossible to exclude a residual * From interpolation of the relative increase in energy expenditure thermogenic effect of the last meal even if the latter was measured in the respiration chamber to free-living conditions. ingested 14 h before the BMR determination. Whether the sympathetic nervous system plays a role during overfeeding is still controversial (18, 33). In the present using pedometers and accelerometers (Fig. 3). This weight gain was much greater than the expected 1.9 kg study neither plasma norepinephrine concentrations nor calculated if all of the excess energy intake was stored as urinary norepinephrine excretion supported the involvefat, neglecting the net energy cost of lipid storage or 1.4 ment of the sympathetic nervous system in response to overfeeding (Table 4). In the absence of significant inkg if the cost of fat deposition is assumed to be -55 kJ/ (Table 4), it is g (21). By th e underwater weighing method it can be creases of free or total triiodothyronine estimated that 56% of this 3.2 kg weight gain was fat also difficult to believe that thyroid hormones play an important role in regulating EE during short-term overand 44% FFM. Another independent approach for determining the composition of weight gain is nitrogen bal- feeding with a mixed diet. The recycling of carbohydrate through 3-carbon compounds has been proposed as an ance. From our 3-day nitrogen balance one can calculate that an average of -8.0 g/day nitrogen was retained in energy dissipative process (5), but its measurement is the body if an estimate of integumental nitrogen losses not yet feasible. The small increase in postabsorptive of 5 mg/kg body wt per day is taken into account in the plasma insulin concentrations during overfeeding is of total nitrogen losses. If one extrapolates the 3-day bal- interest, because insulin may participate in increasing ance to 9 days this corresponds to a gain of -50 g of EE (23), possibly by stimulating protein turnover rate. The 21% increase in 24-h EE in response to overfeedprotein/day x 9 days = 450 g of protein or -2 kg of FFM ing was significantly greater than that in BMR, i.e., 2,038 assuming that protein represents 23% of FFM. The fat mass gain therefore represents 38% of the weight gain vs. 662 kJ/day, respectively (Fig. 6). It represented -33% of the excess energy intake, whereas 67% was stored. compared with 56% when estimated by underwater However, this value of 33% was artificially inflated by weighing. It is, however, well known that nitrogen balance (as well as underwater weighing) has limitations when estimating the composition of small body weight changes. From the results of energy balance and body weight gain (Table 5), one can obtain a crude estimate of the composition of the gain calculated from the energy equivalent of the tissue mass gained that averaged -18 activity kJ/g. If one assumes that 39 kJ are stored per gram of fat mass and 5 kJ/g of FFM (7), it can be calculated that I roughly 39% of weight gain was fat, whereas 61% was FFM. In an ideal situation in which a complete energy TEF balance can be assessed this calculation of the energy equivalent of weight gain may represent an adequate way of assuming the composition of relatively small changes in body weight. Once again in the present study, because EE has not been measured continuously throughout the entire study, one cannot exclude systematic errors on the 9753 calculated stored energy value. 2 The fact that body weight gain was rather important 423 r does not imply that EE was not increased. An increase / 24h EE 24h EE in overall EE can result from four different mechanisms: Weight Overfeeding 1) increased basal metabolic rate; 2) increased thermic Maintenance Day 9 effect of food; 3) decreased work efficiency (1, 15); and FIG. 6. Partition of excess 24-h energy expenditure (EE) induced 4) increased amount of physical activity or increased by overfeeding into 3 components: BMR, basal metabolic rate; TEF, body weight-related cost of physical activity (28). In the thermic effect of food (3 meals); activity, increase in EE due to both changes in activity and increased cost of physical activity. present study we have therefore investigated these difTABLE

5. Nine-day cumulated energy balance and energy equivalent of weight gain in five volunteers during mixed-diet overfeeding

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

E476

RAVUSSIN

ET

AL.

one subject who considerably increased his physical ac- received an average of 74 MJ energy intake above weight tivity. When corrections were made for activity, the value maintenance requirements. In response to this overfeedaveraged -24%, which is rather consistent from one ing, energy expenditure was estimated to increase by -18 subject to another. It also shows that physical acti .vity MJ, resulting in a positive energy balance of -56 MJ. can be the most potent effector of thermogenesis and Consequently, as reported above the energy equivalent this must be especially true in free-living conditions. The of weight gain was found to be 18 kJ/g, which is well increase in 24-h EE, which was not accounted for by the accounted fo r by the results on the cha nges in body increase in BMR, can be related to two different mech- compositio n, especially when measured by nitrogen balanisms: the increased thermic effect of food and the ance. higher activity-related EE. Because intake was markedly In conclusion, this study shows that a short g-day increased during overfeeding, 17,410t 880 vs. 10,940t period of overfeeding with a nutrient-balanced diet re520 kJ/day in the respiration chamber, it is logical to sults in a weight gain that can be entirely explained by expect an overall higher thermic effect of food expressed the computation of energy balance and the assessment in absolute terms. The thermic effect of food over a day of the composition of the weight gain. These results (3 meals) increased from 1,176to 1,757 kJ/day in re- confirm the earlier studies performed by Passmore et al. sponse to overfeeding, i.e., an increase of -580 kJ/day. (19,20). Therefore our data do not support the presence Thus when calculated on the basis of EI, the thermic of luxuskonsumption. It can, however, be argued that the effect of food was unaffected by overfeeding. It is worth duration of the overfeeding period was too short to noting that after 9 days of balanced-diet overfeeding, the promote inefficient mechanisms of EE as suggested by average 24-h RQ was still below 1.0 (0.918 t 0.007vs. the long-term Vermont study (29) and as pointed out by 0.847 t 0.010 during overfeeding and wt maintenance, Garrow (8) in his review. The main question that remains respectively), indicati ng the absence of net de novo unanswered is the magnitude of the inefficiency of energy lipogenesis. In contrast, 7 days of progressive carbohyretention during a more prolonged overfeeding period. In drate overfeeding has been shown to be accompanied by our conditions -25% of the excess EI was expended. an increase in the mean 24-h RQ reaching an average of This implies that -75% of the excess EI above mainte1.12in three volunteers (26). In the latter study net de nance was stored, which represents a substantial positive novo lipogenesis was strongly stimulated resulting in an energy balance. Although long periods of overfeeding are extra cost of storing the excess carbohydrate as fat difficult to study in man, the recent development of the (obligatory the rmogenesis) as well as an extra facu ltative doubly labeled water (D2180) technique for measuring thermogenesis related to sympathetic nervous system EE may prove to be useful because it allows an integrated activation as evidenced by a significant increase in urimeasurement over periods of weeks (25). Another point nary norepinephrine excretion. From the present results to be considered is the age of the subjects; experiments we can therefore conclude that the amount of carbohyin rats show tha t adaptive thermogenesis is mainly acdrate ingested was not sufficient to increase either the tivated in young growing animals, whereas adults have a obligatory or the facultative thermogenic components. lower thermogenic capacity (24).It may be of interest to As shown in Fig. 6 the increase in 24-h EE (2,038 kJ/ study adolescents before the end of growth to ascertain day) can be divided into three major components, each whether their thermogenic capacity is greater than that of them accounting for approximately one-third of the of adult individuals. overall increase: an increase in BMR (-660 kJ/day), a The authors thank the volunteers for their participation in and higher absolute thermic effect of food (-580 kJ/day), contribution to this study, Professor LeMarchand, E. Maeder, and D. and a higher activity-related EE calculated by difference Kock for their expert technical assistance, and J. Braissant for her (.-SO0 kJ/day). secretarial help. They also thank the Nestle Co., Switzerland, for its In Table 5 we have attempted to calculate a complete financial support. K. J. Acheson is on secondment from NESTEC Research Departg-day energy balance by interpolating our relative inment, 1800 Vevey, Switzerland creases in EE measured in the chamber to the free-living conditions. It can be seen that over 9 days animals Received 17 December 1984; accepted in final form 24 June 1985.
REFERENCES 1. APFELBAUM, M., J. BOSTSARRON, AND D. LACATIS. Effect of caloric restriction and excessive caloric intake on energy expenditure. Am. J. Clin. Nub. 24: 1404-1409, 1971. 2. CALLOWAY, D. H., A. C. F. ODELL, AND S. MARGEN. Sweat and miscellaneous nitrogen losses in human balance studies. J. Nutr. 101: 775-786,197l. 3. DALLOSSO, H. M., AND W. P. T. JAMES. Whole body calorimetry studies in adult men. 1. The effect of fat overfeeding on 24-h energy expenditure. Br. J. Nub. 52: 49-64, 1984. 4. DANFORTH, E., JR., R. J. DANIELS, H. L. KATZEFF, E. RAVUSSIN, AND J. S. GARROW. Thermogenic responsiveness in Pima Indians. Clin. Res. 29: 663A, 1981. 5. DEFRONZO, R. A., D. THORIN, J. P. FELBER, D. C. SIMONSON, D. THI$BAUD, E. JEQUIER, AND A. GOLAY. Effect of beta- and alphaadrenergic blockade on glucose-induced thermogenesis in man. J.

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

Clin. Invest. 73: 633-639, 1984.

Energy intake and body weight: a re-examination studies. Am. J. Clin. Nutr. 39: 349-350, 1984. Energy balance and obesity in man. Amsterdam: 1974. The regulation of energy expenditure in man. In: Recent Advances in Obesity Research, edited by G. A. Bray. London: Newman, 1978, vol. 2, p. 200-210. 9. GARROW, J. S. Thermogenesis and obesity in man. In: Recent Advances in Obesity Research, edited by P. Bjorntorp, M. Cairella and A. N. Howard. London: John Libbey, 1981, vol. 3, p. 208-213. 10. GLICK, Z., E. SCHVARTZ, A. MAGAZANIK, AND M. MODAN. Absence of increased thermogenesis during short-term overfeeding in normal and overweight women. Am. J. Clin. Nutr. 30: 1026-1035,1977. 11. GOLDMAN, R. F., AND E. R. BUSKIRK. Method for underwater 6. FORBES, G. B. of two classic 7. GARROW, J. S. North-Holland, 8. GARROW, J. S.

SHORT-TERM,

MIXED-DIET

OVERFEEDING

IN MAN
573,1982. 23. ROTHWELL,

E477
moderately obese and control subjects. Am. J. C&n. Nutr. 35: 566N., AND M. STOCK. A role for insulin in the dietinduced thermogenesis of cafeteria fed rats. Metabolism 30: 673-

12. 13.

14.

15.

weighing and the determination of body density. In: Techniques for Measuring Body Composition, edited by J. Brozek and A. Herschel. Washington, DC: Nat. Res. Count., 1961. GULLICK, A. A study of weight regulation in the adult human body during over nutrition. Am. J. Physiol. 60: 371-395, 1922. J$QUIER, E., AND Y. SCHUTZ. Long-term measurements of energy expenditure in humans using a respiration chamber. Am. J. Clin. Nutr. 38: 989-998,1983. KATZEFF, H. L., AND E. DANFORTH, JR. The thermogenesis response to norepinephrine, food and exercise in lean man during overnutrition. Clin. Res. 29: 663A, 1981. MILLER, D. S., P. MUMFORD, AND M. J. STOCK. Gluttony. 2. Thermogenesis in overeating man. Am. J. Clin. Nutr. 20: 1223R. 0. Experimentelle Beitrage zur Lehre von dem taglichen Nahrungsbedarf des Menschen unter besonderer Berucksichtigung der notwendigen Eiweissmenge. Arch. Hyg. 45: l-87,

678,198l. 24. ROTHWELL,

25.

26.

1229,1967. 16. NEUMANN,

27.

N. G., AND J. V. G. A. DURNIN. The effect of 6 weeks of overfeeding on the body weight, body composition, and energy metabolism of young men. Am. J. Clin. Nutr. 33: 978-988,198O. 18. ODEA, K., M. ESLER, P. LEONARD, J. R. STOCKIGT, AND P. NESTEL. Noradrenaline turnover during under and over-eating in normal weight subjects. Metabolism 31: 896-899,1982. 19. PASSMORE, R., A. P. MEIKLEJOHN, A. D. DEWAR, AND R. K. THOW. Energy utilization in overfed thin men. Br. J. Nutr. 9: 2020. 26, 1955. PASSMORE,

1902. 17. NORGAN,

28.

N., AND M. J. STOCK. Thermogenesis induced by cafeteria feeding in young growing rats (Abstract). Proc. Nutr. Sot. 39: 5A, 1980. SCHOELLER, D. A., AND E. VAN SANTEN. Measurements of energy expenditure in humans by doubly labeled water method. J. Appl. Physiol. 53: 955-959,1982. SCHUTZ, Y., K. J. ACHESON, AND E. JJ~QUIER. Twenty-four-hour energy expenditure and thermogenesis: response to progressive carbohydrate overfeeding in man. Int. J. Obesity, 1985. In press. SCHUTZ, Y., T. BESSARD, AND E. JEQUIER. Thermogenesis measured over a whole day in obese and nonobese women. Am. J. Clin. Nutr. 40: 542-552,1984. SCHUTZ, Y., E. RAVUSSIN, R. DIETHELM, AND E. JEQUIER. Spontaneous physical activity measured by radar in obese and control subjects studied in a respiration chamber. Int. J. Obesity 6: 23-28, H., E. DANFORTH, JR., E. S. HORTON, G. A. BRAY, J. AND L. B. SALANS. Endocrine and metabolic effects of experimental obesity in man. Rec. Progr. Horn. Res. 29: 457-

Downloaded from http://ajpendo.physiology.org/ at University of Maryland on May 11, 2013

1982. 29. SIMS, E. A. A. GLENNON, 486,1973. 30. STRONG,

R., J. A. STRONG, Y. E. SWINDELLS, AND N. EL DIN. The effect of overfeeding on two fat young women. Br. J. Nutr. 17:

373-383,1963. 21. PULLAR, J.

D., AND A. J. F. WEBSTER. The energy cost of fat and protein deposition in the rat. Br. J. Nutr. 37: 355-363, 1977. 22. RAVUSSIN, E., B. BURNAND, Y. SCHUTZ, AND E. J~QUIER. Twentyfour-hour energy expenditure and resting metabolic rate in obese,

J. A., D. SHIRLING, AND R. PASSMORE. Some effects of overfeeding for four days in man. Br. J. Nutr. 21: 909-919,1967. 31. WEBB, P. The measurement of energy exchange in man: an analysis. Am. J. Clin. Nutr. 33: 1299-1310, 1980. 32. WEBB, P., AND J. F. ANNIS. Adaptation to overeating in lean and overweight men and women. Human Nutr. 37C: 117-131,1983. 33. WELLE, S., AND R. G. CAMPBELL. Stimulation of thermogenesis by carbohydrate overfeeding. Evidence against sympathetic nervous system mediation. J. Clin. Invest. 71: 916-925, 1983.