Research Department of Human Nutrition, Centre for Advanced Food Studies, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark

Correspondence to: T H Vasilaras, Research Department of Human Nutrition, Centre for Advanced Food Studies, The Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark. E-mail: thv@kvl.dk

OBJECTIVE: To investigate 24 h energy expenditure (24 h EE) and substrate oxidations in overweight and obese subjects before and after 6 months' ad libitum intake of a low-fat, high-simple carbohydrate diet (SCHO), a low-fat, high-complex carbohydrate diet (CCHO), or a habitual control diet (CD).

SUBJECTS: Twenty-four healthy overweight and obese subjects (11 males and 13 females; body mass index 30.7±0.6 kg/m²; age 42.2±1.8 y).

MEASUREMENTS: Twenty-four hour EE, substrate oxidation rates and spontaneous physical activity (SPA) measured in a respiration chamber, and food intake.

RESULTS: After the intervention no differences were seen in 24 h EE, postprandial thermogenesis, basal metabolic rate or SPA. Carbohydrate oxidation, adjusted for energy balance, increased on both carbohydrate-rich diets (SCHO 13.0%, CCHO 11.5%) and decreased on the CD diet (6.5%); however, the changes were not significantly different between diets. The opposite pattern was seen for fat oxidation, which increased by 2.9% on the CD diet and decreased by 17.1 and 25.6% on the SCHO and CCHO, respectively. The changes only differed between the CD and CCHO diet (P=0.03).

CONCLUSION: Six months' ad libitum intake of a diet rich in simple or complex carbohydrates or a habitual diet induced a shift in the oxidation pattern to closely reflect the diet composition in overweight and obese subjects. No differences between diets were seen in 24 h EE.

International Journal of Obesity (2001) 25, 954-965

fat oxidation; carbohydrate oxidation; indirect calorimetry; obesity; dietary control

Introduction

Maintaining a stable body weight requires that energy intake matches energy expenditure.¹ In order to achieve weight loss it is therefore necessary to induce an imbalance in the equilibrium, either by lowering energy intake or by enhancing energy expenditure.

Numerous long-term intervention studies have shown that ad libitum intake of a high-carbohydrate, low-fat diet induces a spontaneous weight loss.^2,3,4,5,6 A high-carbohydrate, low-fat diet is therefore useful in the treatment and prevention of obesity. The Nordic dietary guidelines recommend a diet with a maximum of 30% of the energy (E%) coming from fat and 55-60 E% carbohydrate with an intake of dietary fibre >3 g/MJ. Furthermore, adults with low energy intakes (<8 MJ/day) and children are advised to keep their intake of refined sugars below 10 E% to prevent vitamin and mineral deficiency and minimize the risk of dental caries.⁷

The weight loss induced by the high-carbohydrate, low-fat diet, can in some^2,8 but not all ^3,4,5 studies be ascribed to a reduction in energy intake. A diet rich in carbohydrates may also induce a higher diet-induced thermogenesis as compared to a fat-rich diet,^9,10,11 which suggests that the weight loss obtained by a carbohydrate-rich diet might partly be ascribed to an increase in energy expenditure.¹²

In contrast to the role of fat and carbohydrate intake in the regulation of body weight, less attention has been given to the long-term effect of different carbohydrate types, such as simple (mono- and disaccharides) vs complex carbohydrates (polysaccharides). Several short-term studies have shown different thermogenic responses following intake of different types of carbohydrate given either as a liquid single-load^13,14,15 or as mixed meals.^16,17,18 Thus, a greater thermogenic effect of fructose and sucrose than both glucose^{13,14,15,16,17} and starch (a polymer of glucose)^15,18 has been reported. In one longer-term study a higher 24 h energy expenditure was also seen after 14 days ad libitum intake of a sucrose-rich, low-fat diet compared to a starch-rich, low-fat diet.¹⁹ However, this study was still not long enough to allow adaption to the diets and the effects may have been temporary.

The aim of the present study was therefore to investigate the long-term effects (6 months) of two low-fat diets rich in either simple or complex carbohydrates on 24 h energy expenditure and macronutrient oxidation rates in a group of overweight and obese subjects.

Methods

Study design

This centre-specific study was a part of the European multi-centre 6 months' intervention trial CARMEN (CArbohydrate Ratio Management in European National diets), described in detail elsewhere.²⁰ Participants in the present study represented a subgroup of the CARMEN experimental group. A schematic presentation of the study design is given in Figure 1. Briefly, subjects were randomly allocated to either a seasonal control group (a quarter of the subjects), or an experimental group (three-quarters of the subjects) stratified by age, gender and body mass index (BMI).

Intervention diets

Prior to the intervention, subjects in the experimental group completed a 5 week run-in period with habitual dietary intakes to get accustomed to the experimental shop system, which is described below. Subjects were then randomly assigned to the three intervention groups, stratified by gender, age, tertile of simple to complex carbohydrate and fat intake as measured during the last week of the run-in period by a 7 day food record.

The subjects in the experimental group consumed one of three diets ad libitum for 6 months: a low-fat, high-simple carbohydrate diet (SCHO), a low-fat, high-complex carbohydrate diet (CCHO), or a control diet (CD) corresponding to average national intake.

Subjects in the intervention groups were told to eat as much as they liked until they felt pleasantly satisfied. Once or twice a week all subjects were provided with natural, commercially available food items from a specially designed shop at our department. Food items could be chosen on an ad libitum basis within the frame of the dietary design. At least 70% of each individuals' fat intake and 50% of his or her carbohydrate intake should be supplied by the shop foods. Subjects were allowed to collect between 75 and 125% of their daily energy needs from the shop, and if necessary to buy other foods from conventional supermarkets. The individual dietary goals during the intervention were based on predicted energy expenditure (EEpred) (kJ/day), calculated as: predicted basal metabolic rate (kJ/day)²¹´physical activity level as determined by an activity questionnaire.²² Habitual intakes of energy (EIrec), fat (FATrec) and carbohydrate (CHOrec) from the 7 day food record from the end of the run-in period as well as national fat (FATnat) and carbohydrate (CHOnat) intakes²³ were used as factors in the following way: final level of energy intake (kJ/day)=(2 EEpred (kJ/day)+EIrec (kJ/day))/3, final level of energy percentage fat (E%)=(FATrec (E%)+2 FATnat (E%))/3, final level of energy percentage carbohydrate (E%)=(CHOrec (E%)+2 CHOnat (E%))/3.

All individual purchases were recorded by bar codes and a computerised system, previously developed and validated at our department²⁴ and adapted to the specific needs in the present study. Targeted and actual energy and macronutrient contents in the intervention diets for the 24 completers are shown in Table 1.

Subjects

Thirty overweight and obese, otherwise healthy, subjects (15 males and 15 females) were recruited by news media or personal communication for participation in the present study. Six subjects (four males and two females) did not complete the study; three subjects were excluded due to lack of compliance, one left because of pregnancy, and two withdrew from the study for occupational reasons. The six subjects who did not complete the study did not differ significantly from the 24 completers with respect to initial anthropometrical measures (measured in the morning of the first test day) and were omitted from all statistical analyses.

Twenty-four overweight and obese subjects (mean±s.e.m.; BMI 30.7±0.6 kg/m²) of both gender, all Caucasians (age 42.2±1.8 y), and with no history of health problems other than being obese, completed the study. Among those were four smokers. None of the subjects had undergone surgical obesity treatment. Prior to participation all subjects underwent a screening examination including medical history, physical examination and clinical blood and urine biochemistry. The blood tests included haematological and endocrine parameters, as well as parameters on liver and kidney function, whereas proteinuria, haematuria and glucoseuria were tested by sticks. Weight change between recruitment and study start was less than 5 kg. Exclusion criteria were current slimming attempts, intensive sporting activities (>7 h/week), medication use suspected to affect appetite or metabolic parameters, an alcohol intake>28 drinks/week for men and >21 drinks/week for women (1 drink ~12 g of pure alcohol),²⁵ and for women pregnancy, or lactation as well as a wish to become pregnant during the study period. Furthermore, any absence from the 6 month study period (due to holiday, illness, work or missing shop appointments) for more than 28 days resulted in an immediate exclusion from the study.

All subjects agreed not to change their habitual physical activity during the study, and the only woman who followed a hormone replacement therapy was requested not to change it. The physical characteristics of the 24 subjects completing the study are given in Table 2. None of the differences between the three intervention groups were significant. Self-reported habitual energy and macronutrient intakes, measured at screening by 3 day weighed food records, were also not significantly different between groups (Table 3).

All subjects gave written consent after having received verbal and written information about the study. The study was approved by the Municipal Ethical Committee of Copenhagen and Frederiksberg to be in accordance with the Helsinki II declaration.

Respiratory measurements

Twenty-four hour energy expenditure (24 h EE) and substrate oxidation rates were measured by indirect whole-body calorimetry in one of our two 14.7 m³ open-circuit respiratory chambers at our department, as described in detail previously.²⁶ Subjects were measured twice, the first time before the intervention period, when subjects consumed their habitual diet and the second time after 6 months' intake of the diet. The two test days were separated by at least 23 weeks and not more than 37 weeks (NS between groups). The subjects arrived at the department at 22:00 h the evening prior to the test day and stayed overnight to get accustomed to the chambers, and thereby minimize stress responses.²⁷ Anthropometric measurements took place after voiding at 08:00 h the following morning after an overnight fast from 21:00 h the evening before. Body weight was measured to the nearest 0.05 kg by a Lindeltronic 8000 (Copenhagen, Denmark). Body composition was measured by bio-electrical impedance²⁸ using an Animeter (HTS Engineering Aps, Odense, Denmark) and calculated as previously reported.²⁹ Height was measured to the nearest 0.5 cm and blood pressure was measured by an automatically inflating cuff and reported as an average of three measurements (A & D, Digital Blood Pressure Meter model UA-743, A & D Company Ltd, Tokyo, Japan). Subsequently a portable EKG device was attached to the subjects (Dialogue 2000 type 2070-14 XTNJ, Danica Elektronics, Rödovre, Denmark). The chambers were closed at 08:30 h. The respiratory measurements began at 09:00 h and continued for 24 h. Postprandial values were obtained at 18:00-22:00 h after consumption of dinner, the most energy-containing meal of the day, during which time subjects were sedentary (ie no physical activity). During the last hour of the chamber stay (08:00-09:00 h the second morning), basal metabolic rate (BMR) was measured, followed by anthropometric measurements. The room temperature was kept at 24°C during the day (08:00-23:00 h) and lowered gradually within an hour to 18°C, the temperature kept at night (23:00-08:00 h) and during BMR measurements (08:00-09:00 h). A standard protocol with fixed sessions of physical activity was followed, including two sessions of cycling (15 min, at a work rate of 75 W) on a static cycle ergometer (Monark 814E, Monark AB, Varberg, Sweden) and two sessions of walking back and forth 25 times in the chamber; otherwise, only sedentary work was allowed. Three meals were given during each chamber stay, breakfast at 9:00 h, lunch at 13:00 h and dinner at 18:00 h. Subjects were under constant surveillance by a laboratory technician during the day-time and by a trained medical student during night-time. The within-subject day-to-day coefficient of variation for repeated 24 h measurements in our chambers has previously been found to be 1.5% for 24 h energy expenditure, 6.7% for fat oxidation and 5.6% for carbohydrate oxidation.³⁰

Measurements in the two available respiration chambers were planned so that each subject was measured in the same chamber during both stays. Furthermore, measurements during a test day were always done on subjects from two different diet groups.

Energy expenditure and oxidation of carbohydrate, fat and protein were calculated from the gas exchange and urinary excretion measurements using the formulas of Elia and Livesey,³¹ as described in Flint et al.³² For smokers, CO₂ production and O₂ consumption were corrected in the subsequent analyses, as previous findings revealed a CO₂ production of 0.31 l/cigarette and an O₂ consumption of 0.29 l/cigarette (John Lind, personal communication 2000). Energy balances were determined by subtracting 24 h energy expenditures from 24 h energy intakes and likewise macronutrient balances were determined by subtracting 24 h substrate oxidations from 24 h intakes of the respective macronutrients.

Spontaneous physical activity (SPA) was assessed by two microwave radar detectors (Sisor Mini-Radar, Static Input System SA, Lausanne, Switzerland), which continuously emit and receive a signal. When the radar signal is disturbed by movements in the chamber it is recorded by the receiver. The SPA measurements indicate the percentage of time in which the subjects are active to a detectable degree.

Respiration chamber diets

During the two chamber stays the subjects were fed a weight-maintenance diet based on the following equation:

Energy intake (kJ/24 h)

=387.8+116.2 FFM (kg)

+190.5 SPA(%)+29.2 FM (kg)+41.04 DE (min)

+140.4 sex-4.48 age (y)³³

where spontaneous physical activity (SPA) for females=5.6% and for males=5.8%, duration of exercise (DE)=30 min, and sex=0 for females and 1 for men (FM=fat mass; FFM=fat-free mass). Diets were prepared according to each subjects' individual energy needs, adjusted to the nearest 0.5 MJ.

The diet included three meals, each eaten within 1 h. At the first occasion subjects were given a standard diet with 47.8 E% carbohydrate (CHO) (11.4 E% simple CHO), 36.9 E% fat, and 15.4 E% protein (Appendix 1 and Table 4). The energy intakes were calculated based on body weight measurements from the beginning of the run-in period, and averaged 10.6±0.4, 10.9±0.5 and 10.5±0.6 MJ/day in the SCHO, CCHO and CD groups, respectively (NS). The energy content in the breakfast amounted to 24.7%, in the lunch to 35.4%, and in the dinner to 39.9% of each individuals' total daily energy intake. At the second occasion subjects consumed a diet corresponding to their respective intervention groups (Appendix 2 and Table 4) with a macronutrient composition matching the target (Table 1). The absolute intakes of simple carbohydrates were, however, lower than targeted in all three groups (due to computational error). However, the simple/complex carbohydrate ratio was still twice as high on the SCHO diet as compared to the other two diets, as intended. Energy needs during the second stay were estimated from changes in body weight as a result of 4 months' intervention.³³ The energy content in the breakfast amounted to 24.3-26.4%, the lunch to 29.9-34.5%, and the dinner to 41.2-43.7% of each individuals' total daily energy intake. Subjects were requested to finish their meals and consume nothing else except tap water, coffee and tea. Cigarette smoking, for the four included smokers, as well as drinking amounts were recorded during the first stay and replicated at the second stay (smoking amounts: 9.9±1.6 cigarettes per stay).

Subjects' energy and nutrient intakes were calculated using a computerized version of the Danish Nutrient Database, DanKostÒ version 2.0 (National Food Agency of Denmark, S�borg, Denmark).

Statistics

All results are given as means±standard error of means (s.e.m.). The significance level was set at P<0.05. All changes in measured variables are computed as time 1 (before intervention) subtracted from time 2 (after intervention) so that positive values indicate increases and negative values indicate decreases of variables.

Differences between groups in anthropometric, metabolic and energy intake data were tested by one-way analysis of variance (ANOVA), whereas hour-by-hour data on energy expenditure and RQnp was analysed by repeated measures ANOVA using the GLM procedure with diet group and time as factors. When ANOVA indicated a significant group effect, post hoc comparisons were made, with Bonferroni adjustment of significance levels for the pairwise comparisons, using unpaired t-tests. Adjustments for differences between diet groups in FM, FFM or energy balances were performed as described by Ravussin and Bogardus.³⁴

Statistical analysis were performed with Statistical Analysis Package 6.12, SASÒ (SAS Institute, Cary, NC, USA) and SPSSÒ Software version 10.0 for WindowsÒ (SPSS Inc. Headquarters, Chicago, Illinois, USA).

Results

Energy and macronutrient intake

During the intervention the average ad libitum energy intake was similar among treatments (Table 1). No group effect was seen in total carbohydrate intake, which on the contrary was seen for intake of both simple (P=0.002) and complex (P=0.016) carbohydrates (Table 1). As expected, subjects on the SCHO diet consumed more simple carbohydrates (vs CCHO P=0.002, vs CD P=0.036), whereas the CCHO group had the highest intake of complex carbohydrates (vs SCHO P=0.016; Table 1). The deviation in the intakes of the two carbohydrate types led to a highly significant difference in the simple to complex carbohydrate ratio between diet groups (P<0.001), being highest in the SCHO group compared to the CD group (P=0.017) and lowest in the CCHO group (P<0.001; Table 1). There was a tendency to a difference in fat intake between treatments (P=0.098; Table 1). Protein intake differed between groups (P=0.047) and was higher in the CCHO group as compared to the SCHO group (P=0.043; Table 1), whereas self-reported alcohol intake did not differ between treatments (Table 1).

Energy expenditure (EE)

Total unadjusted 24 h EE, as well as 24 h EE, basal metabolic rate (BMR) and postprandial thermogenesis (18:00-22:00 h), all adjusted for differences between groups in FM and FFM, were similar between groups both before and after 6 months' intake of the experimental diets (Table 5). The changes were not significantly different either (Table 5). Furthermore, no differences were found in the hour-by-hour profiles across the two 24 h EE measurement periods among diet groups (Figure 2).

On average, subjects in the two carbohydrate groups were in a slightly negative energy balance and the CD group in a slightly positive energy balance during both chamber stays (mean over the two stays±s.e.m.: -251±200, -23±189 and 213±147 kJ/day for SCHO, CCHO and CD, respectively). This was, however, not significantly different between groups at any stay (data not shown), and subjects were therefore assumed to be in energy balance.^35,36 On an individual level the balances were highly reproducible between the two respiration chamber stays.

Macronutrient oxidation rates

On the standard diet, prior to the intervention, macronutrient oxidations, adjusted for group differences in 24 h energy balances, were not significantly different between any of the treatments (Table 5). However, after the three ad libitum diets alterations in the substrate oxidation patterns were observed.

As for the adjusted carbohydrate oxidations we found a group effect (P=0.028), with no difference between the two carbohydrate diets, but with the CCHO diet being higher than the CD diet (P=0.025). Carbohydrate oxidation rates increased on both carbohydrate-rich diets, by 13.0% on SCHO and by 11.5% on CCHO, and decreased by 6.5% on the CD diet (Table 5), but the changes were not significantly different between groups (P=0.127).

An overall group effect was also seen regarding the adjusted fat oxidations (P=0.040), which were significantly higher in the CD group as compared to the CCHO group (P=0.037). Fat oxidation increased by 2.9% on the CD diet and decreased by 17.1% and 25.6% on the SCHO and CCHO diet, respectively (group effect: P=0.032). The difference was significant between the CD and the CCHO group (P=0.033; Table 5). Protein oxidations had increased on all three diets after the intervention, but were similar between diet groups (Table 5).

The unadjusted hour-by-hour profiles of 24 h non-protein respiratory quotient (RQnp) did not differ between diet groups before the intervention period (Figure 3). However, after the intervention a significant time´group interaction (P<0.001) was found, reflecting the differences in time course between groups. A greater postprandial RQnp was seen in both carbohydrate groups, as compared to the CD group. This was entirely attributed to a difference during the day period (9:00-23:00 h) although most clearly following the evening meal (time´group interaction: P=0.002), whereas no change was seen during the night period (23:00-8:00 h); (time´group interaction, P=0.375; group effect, P=0.029). Macronutrient balances were not significantly different between groups at any stay (data not shown).

Spontaneous physical activity, heart rate and blood pressure

Total 24 h SPA and changes in 24 h SPA were similar between groups (Table 5). The same applied to heart rate, fasting diastolic and systolic blood pressure (data not shown).

Body weight and composition

After 6 months' ad libitum intake of the experimental diets, the CCHO group had lost an average of 2.8±1.2 kg (corresponding to 3.0% of initial body weight), whereas the CD group had gained 2.8±0.6 kg (3.2% of initial body weight), and the SCHO group had gained 1.4±2.0 kg (1.5% of initial body weight; group effect, P=0.026). The changes were only significantly different between the CCHO and the CD group (P=0.033). There were no differences between the changes in the three diet groups in FM or FFM.

Discussion

To our knowledge the present study is the first, controlled ad libitum study which has investigated 24 h energy expenditure and substrate oxidations after 6 months' intake of different types of carbohydrates. A 6 months' intervention period has the advantage of allowing adaption to the diets and thereby giving a more realistic picture of the long-term effects of such dietary manipulations.

The present study showed that 6 months' ad libitum intake of a fat-reduced, high-carbohydrate diet led to no differences in energy expenditure (Table 5 and Figure 2), but to stimulation of carbohydrate oxidation and suppression of fat oxidation compared to a normal-fat diet. However, when substrate oxidations were adjusted for group differences in 24 h energy balance, only the CCHO diet differed from the CD diet (Table 5). These findings were reflected in the hour-by-hour profiles of 24 h RQnp, which showed the lowest RQnp values in the CD group and the highest values in the CCHO group (Figure 3). Looking separately at day and night it was apparent that the differences occurred mainly during the day-time, ie during periods associated with food intake and physical activity.

Increased carbohydrate oxidation rates after carbohydrate ingestion, as compared to a fat-rich diet, have been reported in several previous studies,^{8,11,37,38,39,40} and is ascribed to the oxidative autoregulatory properties characterising carbohydrate metabolism.⁴¹

It is well-established, that differences exists in the digestion and absorption pattern of mono-, di- and polysaccharides.^42,43 The high intake of simple carbohydrates, mainly as fructose and sucrose (the disaccharide of fructose and glucose), in the SCHO group was therefore expected to introduce a delay in the RQnp profiles as compared to the CCHO group with the high intake of complex (starchy) carbohydrates. However, no delay was observed, as the hour-by-hour profiles of 24 h RQnp of the two carbohydrate-rich diets followed the same pattern after meal intake (Figure 3). Figure 3 shows increased postprandial values on both carbohydrate diets as compared to the CD diet. These differences did not reach significance, but this may be due to lack of sufficient statistical power.

The findings of a similar oxidation pattern on the SCHO and CCHO diet in the present study, is in contrast with a number of short-term studies, lasting from 3 to 6 h after meal-intake, where different oxidation rates were observed following intake of different carbohydrate types.^{13,14,15,16,17,18} In a longer-term study over 14 days, Raben et al described increased carbohydrate oxidation and suppressed fat oxidation following 14 day controlled ad libitum intake of a sucrose-rich diet (23 E% sucrose) as compared to a starch-rich diet (2 E% sucrose).¹⁹ However, in that study energy intake was higher on the sucrose-rich diet compared with the starch-rich diet, which may explain the higher RQnp on the sucrose-rich diet.

The present study, however, differed from the above-mentioned short-term studies in that our subjects were overweight and obese. Thus, Blaak et al found that carbohydrate oxidation, 6 h after ingestion of either a liquid single-load of sucrose or glucose, increased significantly more after sucrose than after glucose in normal-weight subjects, whereas there were no differences in the obese between ingested carbohydrate sources (unpublished data). Moreover, carbohydrate oxidation after sucrose intake increased significantly less in obese as compared to normal-weight subjects. Finally, suppression of lipid oxidation 2-6 h postprandially was weaker in obese as compared to normal-weight subjects on both sucrose and glucose (Blaak et al, unpublished data).

It is, however, also likely that discrepancies between previous studies and ours with regard to oxidation rates following intake of different carbohydrates, may be attributed to an oxidative regulatory adaption during the 6 months' dietary intervention.

This hypothesis is supported by findings in another sub-sample of the three intervention groups from the CARMEN study after 6 months' of intervention. In these subjects no group differences were seen in RQnp (assessed by an open circuit ventilated hood system) during 4 h following oral ingestion of 75 g of sucrose.⁴⁴

When it comes to 24 h energy expenditure (24 h EE) the results are more diverging. Hill et al³⁹ found no differences in total EE following 7 days' intake of either a high-carbohydrate (60 E%), a mixed (35 E% carbohydrate, 45 E% fat) or a high-fat (60 E%) diet of fixed energy intakes. A similar study using isoenergetic diets of varying carbohydrate content (high carbohydrate (79 E%), medium carbohydrate (48 E%), low-carbohydrate (9 E%)) drew the same conclusions after measurements of 48 h EE.⁴⁵ Moreover, Thomas et al⁴⁰ observed no changes in 24 h EE after 7 days' intake of a high-carbohydrate (62 E%) vs a high-fat (52 E%) diet, although a higher total energy intake was seen on the high-fat diet.

On the other hand, 24 h EE was increased on the sucrose-rich diet as compared to the fat-rich diet as observed in the study by Raben et al.¹⁹ This was supported by findings of increased catecholamine concentrations, indicating increased activity in the sympathetic nervous system on the sucrose-rich diet.

It is energetically more costly to metabolize fructose than glucose.^14,46 The high intake of simple carbohydrates, mainly as fructose and sucrose (the disaccharide of fructose and glucose), in the SCHO group was therefore expected to result in a higher thermogenic response as compared to the CCHO group. However, we saw no differences in the thermogenic response between the two carbohydrate-rich diets (Table 5).

Several short- and longer-term studies have reported differences in EE depending on which carbohydrate types were tested, both 3 and 5 h postprandially¹⁸ and over 24 h.¹⁹ A postprandial increase in metabolic rate, ie diet-induced thermogenesis (DIT), might be reflected in an increased total energy expenditure over the day or perhaps over a couple of days, but the effect might be levelled off in the longer term, when adaption to the diet has occurred. Against this is, however, a study in long-term male vegetarians showing 11% higher BMR and increased noradrenaline concentrations compared to non-vegetarians. These differences were ascribed to a higher carbohydrate intake in the vegetarians, leading to a greater stimulation of the sympathetic nervous system.⁴⁷

In the present study we saw no differences in stimulation of postprandial energy expenditure (defined as measurements between 18:00 and 22:00 h) between groups, which also applied to the longer-term study by Raben et al.¹⁹ This may be ascribed to the fact that the measurements were done in a respiration chamber, and not as in most other studies with a ventilated hood system. In a chamber, the subjects are generally in a non-resting situation, which makes the proposed values of postprandial EE or DIT less precise than with a ventilated hood.

Another explanation for the different outcome between studies could be the differences between intervention diets with respect to sucrose intake, both regarded as percentage of energy supplied and percentage of total carbohydrate intake.

Finally, we found that the CCHO diet induced a weight loss (2.8 kg), whereas the weight gain on the SCHO diet (1.4 kg) did not differ from the weight gain on the CD diet (2.8 kg). These results should, however, not be given too much weight since the outcome of the entire CARMEN trial on 236 subjects showed that the CCHO group lost 1.8 kg, the SCHO group lost 0.9 kg, and the CD group gained 0.8 kg.²⁰

In conclusion, we observed no differences in 24 h energy expenditure, BMR or postprandial energy expenditure in overweight and obese subjects after 6 months' ad libitum intake of a diet rich in simple or complex carbohydrates or a habitual diet. However, the diet composition did affect the macronutrient oxidation rates by shifting the oxidation pattern to closely reflect the macronutrient composition of the intervention diets.

Acknowledgements

The authors are grateful to the EU-FAIR program (PL 95-809), the European Sugar Industries, the Danish Medical Research Council, and the Danish Research and Development Programme for Food Technology for supporting this study. We commemorate the essential contribution by Foppe ten Hoor, MD, PhD and thank Wim Saris and Arnold Balfort with whom the idea of the CARMEN project was conceived. We are indebted to the participants for their commitment and great cooperation throughout the study. We also thank the project staff of the Research Department of Human Nutrition: Christina Cuthbertson, Majken Ege, Bente Knap, Charlotte Kostecki, Jannie M Larsen, John Lind, Per Mikkelsen, Dorthe Mortensen, Lars Paaske, Ulla Pedersen, Tina Petersen and Inge Timmermann, as well as the involved students, Rikke Jacobsen and Tina S Lindel�v. Finally, we thank the food sponsors: A/S Hatting Bageri, Coca-Cola Danmark A/S, Daloon A/S, Danisco Foods, Danisco Sugar, Danish Crown/Den gr�nne Slagter, Danish Prime Food Company K/S, Delimo A/S, Flensted A/S, Haribo Lakrids A/S, Havnem�llerne A/S, Kelsen The International Bakery A/S, Malaco A/S, Master Foods a.s, Mayo A/S, Nordisk Kellogg's A/S, Pingvin Lakrids/Galle & Jessen A/S, Rynkeby Foods A/S, Schulstad Br�d A/S, Stryhn's Leverpostej ApS, Van den Bergh Foods A/S and Vestjyske slagterier AmbA for kindly contributing to the food selection.

1 Flatt JP. Dietary fat, carbohydrate balance, and weight maintenance: effects of exercise. Am J Clin Nutr 1987; 45: 296-306, MEDLINE

2 Kendall A, Levitsky DA, Strupp BJ, Lissner L. Weight loss on a low-fat diet: consequence of the imprecision of the control of food intake in humans. Am J Clin Nutr 1991; 53: 1124-1129, MEDLINE

3 Prewitt TE, Schmeisser D, Bowen PE, Aye P, Dolecek TA, Langenberg P, Cole T, Brace L. Changes in body weight, body composition, and energy intake in women fed high- and low-fat diets. Am J Clin Nutr 1991; 54: 304-310, MEDLINE

4 Raben A, Jensen ND, Marckmann P, Sandström B, Astrup A. Spontaneous weight loss during 11 weeks' ad libitum intake of a low fat/high fiber diet in young, normal weight subjects. Int J Obes Relat Metab Disord 1995; 19: 916-923, MEDLINE

5 Siggaard R, Raben A, Astrup A. Weight loss during 12 weeks' ad libitum carbohydrate-rich diet in overweight and normal-weight subjects at a Danish work site. Obes Res 1996; 4: 347-356, MEDLINE

6 Astrup A, Ryan L, Grunwald GK, Storgaard M, Saris W, Melanson E, Hill JO. The role of dietary fat in body fatness: evidence from a preliminary meta-analysis of ad libitum low-fat dietary intervention studies. Br J Nutr 2000; 83: (Suppl 1) S25-S32, MEDLINE

7 Nordic Committee on Foods, Nordic Council of Ministers. Nordic nutrition recommendations. Report. 1996, 28: Nordic Council of Ministers: Copenhagen 1996, pp. 11-23.

8 Stubbs RJ, Harbron CG, Murgatroyd PR, Prentice AM. Covert manipulation of dietary fat and energy density: effect on substrate flux and food intake in men eating ad libitum. Am J Clin Nutr 1995; 62: 316-329, MEDLINE

9 Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt J-P, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr 1988; 48: 240-247, MEDLINE

10 Lean MEJ, James WPT. Metabolic effects of isoenergetic nutrient exchange over 24 hours in relation to obesity in women. Int J Obes Relat Metab Disord 1988; 12: 15-27,

11 Horton TJ, Drougas H, Brachey A, Reed GW, Peters JC, Hill JO. Fat and carbohydrate overfeeding in humans: different effects on energy storage. Am J Clin Nutr 1995; 62: 19-29, MEDLINE

12 Dirlewanger M, di Vetta V, Guenat E, Battilana P, Seematter G, Schneiter P, Jéquier E, Tappy L. Effects of short-term carbohydrate or fat overfeeding on energy expenditure and plasma leptin concentrations in healthy female subjects. Int J Obes Relat Metab Disord 2000; 24: 1413-1418, MEDLINE

13 Macdonald I. Differences in dietary-induced thermogenesis following the ingestion of various carbohydrates. Ann Nutr Metab 1984; 28: 226-230, MEDLINE

14 Tappy L, Randin J-P, Felber J-P, Chiolero R, Simonson DC, Jequier E, DeFronzo RA. Comparison of thermogenic effect of fructose and glucose in normal humans. Am J Physiol 1986; 250: E718-E724, MEDLINE

15 Blaak EE, Saris WHM. Postprandial thermogenesis and substrate utilization after ingestion of different dietary carbohydrates. Metabolism 1996; 45: 1235-1242, MEDLINE

16 Schwarz J-M, Schutz Y, Froidevaux F, Acheson KJ, Jeanprêtre N, Schneider H, Felber J-P, Jéquier E. Thermogenesis in men and women induced by fructose vs glucose added to a meal. Am J Clin Nutr 1989; 49: 667-674, MEDLINE

17 Schwarz J-M, Schutz Y, Piolino V, Schneider H, Felber J-P, Jéquier E. Thermogenesis in obese women: effect of fructose vs glucose added to a meal. Am J Physiol 1992; 262: E394-E401, MEDLINE

18 Raben A, Kiens B, Richter EA. Differences in glycaemia, hormonal response and energy expenditure after a meal rich in mono- and disaccharides compared to a meal rich in polysaccharides in physically fit and sedentary subjects. Clin Physiol 1994; 14: 267-280, MEDLINE

19 Raben A, Macdonald I, Astrup A. Replacement of dietary fat by sucrose or starch: effects on 14 d ad libitum energy intake, energy expenditure and body weight in formerly obese and never-obese subjects. Int J Obes Relat Metab Disord 1997; 21: 846-859, MEDLINE

20 Saris WHM, Astrup A, Prentice AM, Zunft HJF, Formiguera X, Verboeket-van de Venne WPHG, Raben A, Poppitt SD, Seppelt B, Johnston S, Vasilaras TH, Keogh GF. Randomised controlled trial of changes in dietary carbohydrate/fat ratio and simple vs complex carbohydrates on body weight and blood lipids: the CARMEN study. Int J Obes Relat Metab Disord 2000; 24: 1310-1318, MEDLINE

21 FAO/WHO/UNU. Energy and protein requirements. Report of a joint FAO/WHO/UNU Expert Consultation. Technical Report Series no. 724. World Health Organization: Geneva, 1985,

22 Baecke JAH, Burema J, Frijters JER. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36: 936-942, MEDLINE

23 Andersen NL, Fagt S, Groth MV, Hartkopp HB, Möller A, Ovesen L, Warming DL. Danish dietary habits 1995 (in Danish with an English summary). Publication no. 235 . Copenhagen: National Food Agency, 1996, pp. 80-111.

24 Skov AR, Toubro S, Raben A, Astrup A. A method to achieve control of dietary macronutrient composition in ad libitum diets consumed by free-living subjects. Eur J Clin Nutr 1997; 51: 667-672, MEDLINE

25 Grønbæk MN, Iversen L, Olsen J, Becker PU, Hardt F, Sørensen TIA. Genstandsgrænser. Ugeskr læger 1997; 159: 5939-5945,

26 Astrup A, Thorbek G, Lind J, Isaksson B. Prediction of 24 h energy expenditure and its components from physical characteristics and body composition in normal-weight humans. Am J Clin Nutr 1990; 52: 777-783, MEDLINE

27 De Boer JO, Van Es AJH, Vogt JE, Van Raaij JMA, Hautvast JGAJ. Reproducibility of 24 h energy expenditure measurements using a human whole body indirect calorimeter. Br J Nutr 1987; 57: 201-209, MEDLINE

28 Lukaski HC, Johnson PE, Bolonchuk WW, Lykken GI. Assessment of fat-free mass using bioelectrical impedance measurements of the human body. Am J Clin Nutr 1985; 41: 810-817, MEDLINE

29 Lukaski HC, Bolonchuk WW, Hall CB, Siders WA. Validation of tetrapolar bioelectrical impedance method to assess human body composition. J Appl Physiol 1986; 60: 1327-1332, MEDLINE

30 Toubro S, Christensen NJ, Astrup A. Reproducibility of 24 h energy expenditure, substrate utilization and spontaneous physical activity in obesity measured in a respiration chamber. Int J Obes Relat Metab Disord 1995; 19: 544-549, MEDLINE

31 Elia M, Livesey G. Energy expenditure and fuel selection in biological systems: The theory and practice of calculations based on indirect calorimetry and tracer methods. In: Simopoulos AP (ed). Metabolic control of eating, energy expenditure and the bioenergetics of obesity. World review of nutrition and diet. Vol 70: Karger: Basel, 1992, pp. 68-131.

32 Flint A, Raben A, Rehfeld JF, Holst JJ, Astrup A. The effect of glucagon-like peptide-1 on energy expenditure and substrate metabolism in humans. Int J Obes Relat Metab Disord 2000; 24: 288-298, MEDLINE

33 Klausen B, Toubro S, Astrup S. Age and sex effects on energy expenditure. Am J Clin Nutr 1997; 65: 895-907, MEDLINE

34 Ravussin E, Bogardus C. Relationship of genetics, age, and physical fitness to daily energy expenditure and fuel utilization. Am J Clin Nutr 1989; 49: 968-975, MEDLINE

35 Garrow JS. Energy balance and obesity in man. 1st edn. North Holland Publishing Company: Amsterdam, 1974, pp. 1-46.

36 Schrauwen P, Lichtenbelt WDM, Westerterp KR. Energy balance in a respiration chamber: individual adjustment of energy intake to energy expenditure. Int J Obes Relat Metab Disord 1997; 21: 769-774, MEDLINE

37 Acheson KJ, Schutz Y, Bessard T, Ravussin E, Jéquier E. Nutritional influences on lipogenesis and thermogenesis after a carbohydrate meal. Am J Physiol 1984; 246: E62-E70, MEDLINE

38 Flatt JP, Ravussin E, Acheson KJ, Jéquier E. Effects of dietary fat on postprandial substrate oxidation and on carbohydrate and fat balances. J Clin Invest 1985; 76: 1019-1024, MEDLINE

39 Hill JO, Peters JC, Reed GW, Schlundt DG, Sharp T, Greene HL. Nutrient balance in humans: effects of diet composition. Am J Clin Nutr 1991; 54: 10-17, MEDLINE

40 Thomas CD, Peters JC, Reed GW, Abumrad NN, Sun M, Hill JO. Nutrient balance and energy expenditure during ad libitum feeding of high-fat and high-carbohydrate diets in humans. Am J Clin Nutr 1992; 55: 934-942, MEDLINE

41 Schutz Y, Flatt JP, Jéquier E. Failure of dietary fat intake to promote fat oxidation: a factor favoring the development of obesity. Am J Clin Nutr 1989; 50: 307-314, MEDLINE

42 Southgate DAT. Digestion and metabolism of sugars. Am J Clin Nutr 1995; 62: (Suppl) 203S-211S, MEDLINE

43 Wolever TMS, Jenkins DJA, Collier GR, Lee R, Wong GS, Josse RG. Metabolic response to test meals containing different carbohydrate foods: 1. relationship between rate of digestion and plasma insulin response. Nutr Res 1988; 8: 573-581,

44 Verboeket-van de Venne WPHG, Erckens DTH, Saris WHM. Sucrose-induced thermogenesis and substrate utilisation in obese subjects. Int J Obes Relat Metab Disord 2000; 24: (Suppl) 428,

45 Shetty PS, Prentice AM, Goldberg GR, Murgatroyd PR, McKenna APM, Stubbs RJ, Volschenk PA. Alterations in fuel selection and voluntary food intake in response to isoenergetic manipulation of glycogen stores in humans. Am J Clin Nutr 1994; 60: 534-543, MEDLINE

46 Tappy L, Jéquier E. Fructose and dietary thermogenesis. Am J Clin Nutr 1993; 58: (Suppl) 766S-770S, MEDLINE

47 Toth MJ, Poehlman ET. Sympathetic nervous system activity and resting metabolic rate in vegetarians. Metabolism 1994; 43: 621-625, MEDLINE

Table A1

Table A2

Figure 1 Schematic presentation of the study design including measurements performed in the present study.

Figure 2 Energy expenditure (EE) over a 24 h period measured before (top) and after (bottom) 6 months' ad libitum intake of a diet rich in simple (SCHO) or complex carbohydrates (CCHO) or a habitual diet (CD). Data are means (± s.e.m.). Left panel (repeated measures ANOVA): before¾time effect, P<0.001, group effect, P=0.747, time´group interaction, P=0.812; after¾time effect, P<0.001, group effect, P=0.750, time´group interaction, P=0.870. Right panel (unpaired t-test, Bonferroni adjusted): before¾NS; after¾NS. Meals were given at 9:00, 13:00 and 18:00 h and physical activity was performed at 9:30, 11:00, 14:30 and 16:00 h.

Figure 3 Non-protein respiratory quotient (RQnp) over a 24 h period measured before (top) and after (bottom) 6 months' ad libitum intake of a diet rich in simple (SCHO) or complex carbohydrates (CCHO) or a habitual diet (CD). Data are means (±s.e.m.). Left panel (repeated measures ANOVA): before¾time effect, P<0.001, group effect, P=0.415, time´group interaction, P=0.721; after¾time effect, P<0.001, group effect, P=0.015, time´group interaction, P<0.001. Right panel (unpaired t-test, Bonferroni adjusted): before¾NS; after¾*P<0.05, different from CCHO. Meals were given at 9:00, 13:00 and 18:00 h and physical activity was performed at 9:30, 11:00, 14:30 and 16:00 h.

Table 1 Targeted and actual^a macronutrient composition and energy content in the intervention diets

Table 2 Physical characteristics of subjects in the three intervention groups^a

Table 3 Self-reported habitual diet^a

Table 4 Energy and macronutrient intake during the first^a and the second^b respiration chamber stay

Table 5 Twenty-four hour energy expenditure, postprandial energy expenditure, BMR and substrate oxidations

Table A1 Menu during the first respiration chamber stay, standard diet

Table A2 Menu during the second respiration chamber stay according to intervention group