Few studies have examined the influence of pre-exercise meals with different glycaemic indices (GIs) on substrate oxidation and non-homeostatic appetite (i.e. food reward) in adults of various ages and ethnicities. We aimed to examine the effects of pre-exercise high and low GI meals on substrate oxidation and food reward in middle-aged Japanese women. This randomised crossover trial included fifteen middle-aged women (aged 40⋅9 ± 6⋅5 years, mean ± sd). The participants consumed a high or low GI breakfast at 09.00 and rested until 11.00. Thereafter, participants performed a 60-min walk at 50 % of their estimated maximum oxygen uptake (11.00–12.00) and rested until 13.00. Expired gas samples were collected every 30 min prior to walking, and samples were collected continuously throughout the walking and post-walking periods. Blood samples and subjective appetite ratings were collected every 30 min, except during walking. The Leeds Food Preference Questionnaire in Japanese (LFPQ-J) was used to assess food reward at 09.00, 10.00, and 13.00 h. The cumulative fat oxidation during exercise was higher in the low GI trial than in the high GI trial (P = 0⋅03). The cumulative carbohydrate oxidation during walking was lower in the low GI trial than in the high GI trial (P = 0⋅01). Trial-by-time interactions were not found for any food-reward parameters between trials. Low GI meals elicited enhanced fat oxidation during a subsequent 60-min walk in middle-aged women. However, meals with different GIs did not affect food reward evaluated over time in the present study.

The present study investigated the potential role of the composition of dietary fatty acids in the regulation of biological rhythms, such as the sleep architecture, core body temperature and leukocyte clock gene expression, in subjects fed meals rich in palmitic acid (PA) or oleic acid (OA). Eleven males participated in two sessions of indirect calorimetry in a whole-room metabolic chamber. In each session, subjects consumed three meals rich in PA (44·3 % of total fat as PA and 42·3 % as OA) or OA (11·7 % of total fat as PA and 59·3 % as OA) in the metabolic chamber. The ratio of PA to OA in plasma was significantly lower and fat oxidation was significantly higher during 24 h of indirect calorimetry in the session with meals rich in OA than in that with meals rich in PA. The duration of slow wave sleep (SWS) was shorter, the latency of SWS was longer and the nadir of core body temperature after bedtime was later in the session with meals rich in PA than in that with meals rich in OA. The peak in CRY1 gene expression was earlier and its amplitude was higher in the session with meals rich in PA than in that with meals rich in OA. In healthy young males, meals rich in PA decreased fat oxidation and disrupted biological rhythms, particularly the sleep architecture and core body temperature during sleep, more than meals rich in OA.

Prebiotic supplements and high-protein (HP) diets reduce body weight and modulate intestinal microbiota. Our aim was to elucidate the combined effect of an inulin/oligofructose (FOS) and HP diet on body weight gain, energy metabolism and faecal microbiota. Forty male C57BL/6NCrl mice were fed a control (C) diet for 2 weeks and allocated to a C or HP (40 % protein) diet including no or 10 % inulin/FOS (C + I and HP + I) for 4 weeks. Inulin/FOS was added in place of starch and cellulose. Body weight, food intake, faecal energy and nitrogen were determined. Indirect calorimetry and faecal microbiota analysis were performed after 3 weeks on diets. Body weight gain of HP-fed mice was 36 % lower than HP + I- and C-fed mice (P < 0⋅05). Diet digestibility and food conversion efficiency were higher in HP + I- than HP-fed mice (P < 0⋅01), while food intake was comparable between groups. Total energy expenditure (heat production) was 25 % lower in HP + I- than in C-, HP- and C + I-fed mice (P < 0⋅001). Carbohydrate oxidation tended to be 24 % higher in HP- than in HP + I-fed mice (P < 0⋅05). Faecal nitrogen excretion was 31–45 % lower in C-, C + I- and HP + I- than in HP-fed mice (P < 0⋅05). Faecal Bacteroides–Prevotella DNA was 2⋅3-fold higher in C + I- and HP + I- relative to C-fed mice (P < 0⋅05), but Clostridium leptum DNA abundances was 79 % lower in HP + I- than in HP-fed mice (P < 0⋅05). We suggest that the higher conversion efficiency of dietary energy of HP + I but not C + I-fed mice is caused by higher digestibility and lower heat production, resulting in increased body mass.

Body weight control is thought to be improved when physical activity and energy intake are both high (high energy turnover (ET)). The aim of the present study was to investigate the short-term impact of ET on fat balance during zero energy balance (EB), energetic restriction (ER) and overfeeding (OF). In a randomised crossover study, nine healthy men (BMI: 23·0 (SD 2·1) kg/m2, 26·6 (SD 3·5) years) passed 3 × 3 d in a metabolic chamber: three levels of ET (low, medium and high; physical activity level = 1·3−1·4, 1·5−1·6 and 1·7−1·8) were performed at zero EB, ER and OF (100, 75 and 125 % of individual energy requirement). Different levels of ET were obtained by walking (4 km/h) on a treadmill (0, 165 and 330 min). Twenty-four-hour macronutrient oxidation and relative macronutrient balance (oxidation relative to intake) was calculated, and NEFA, 24-h insulin and catecholamine secretion were analysed as determinants of fat oxidation. During EB and OF, 24-h fat oxidation increased with higher ET. This resulted in a higher relative fat balance at medium ET (EB: +17 %, OF: +14 %) and high ET (EB: +23 %, OF: +17 %) compared with low ET (all P < 0·05). In contrast, ER led to a stimulation of 24-h fat oxidation irrespective of ET (no differences in relative fat balance between ET levels, P > 0·05). In conclusion, under highly controlled conditions, a higher ET improved relative fat balance in young healthy men during OF and EB compared with a sedentary state.

This study aimed to characterise lean and obese phenotypes according to diet and body composition, and to compare fasting and postprandial appetite and metabolic profiles following a high-fat test meal. A total of ten lean (BMI<25 kg/m2) high-fat (LHF), ten lean low-fat (LLF; >40 and <30 % energy from fat) and ten obese (BMI>30 kg/m2) high-fat consumers (OHF; >40 % energy from fat) were recruited. Before and following the test meal (4727 kJ (1130 kcal), 77 % fat, 20 % carbohydrate (CHO) and 3 % protein), fasting plasma glucose, insulin, leptin, ghrelin, peptide YY (PYY), RER, RMR and subjective appetite ratings (AR) were measured for 6 h. Thereafter, subjects consumed a self-selected portion of a standardised post-test meal (40 % fat, 45 % CHO and 15 % protein) and reported AR. Fasting (P=0·01) and postprandial (P<0·001) fat oxidation was significantly higher in LHF than in LLF but was not different between LHF and OHF. Although similar between the lean groups, fasting and postprandial energy expenditures were significantly higher in OHF compared with LHF (P<0·01). Despite similar AR across groups, LLF consumed a relatively greater quantity of the post-test meal than did LHF (7·87 (sd 2·96) v. 7·23 (sd 2·67) g/kg, P=0·013). The lean groups showed appropriate changes in plasma ghrelin and PYY following the test meal, whereas the OHF group showed a blunted response. In conclusion, the LHF phenotype had a greater capacity for fat oxidation, which may be protective against weight gain. OHF individuals had a blunted appetite hormone response to the high-fat test meal, which may subsequently increase energy intake, driving further weight gain.

The high prevalence of obesity is a major public health issue and contributes to the ‘double burden’ of disease in developing countries. Early exposure to poor nutrition may cause metabolic adaptations that, when accompanied by exposure to ‘affluent’ nutrition, may increase the risk for obesity and other metabolic disorders. The aim of this study was to determine differences in energy metabolism and nutritional status between normal-height and growth-retarded North Korean children living in South Korea. A total of 29 children were recruited and underwent measurements of resting energy expenditure (REE), respiratory quotient (RQ), anthropometrics and dietary intake. There was no difference in REE or any assessment of obesity between the growth-retarded and normal-height children. Children who were classified as growth retarded (HAZ<−1.0) or stunted (HAZ<−2.0) had a significantly higher RQ (β=0.036 or 0.060, respectively, P=0.018 or 0.016), independent of sex, age, fat-free mass, fat mass and food quotient, compared with children with normal height. The results from this study, the first from an Asian population, add to the growing body of literature suggesting that undernutrition early in life results in adaptations in energy metabolism that favor fat deposition, increasing the risk of stunted children becoming overweight or obese later in life. Continued research on this topic is warranted, given the continued rise in the prevalence of the double burden in transitional countries.

High-fibre diets offer several beneficial health effects. The objective of the present study was to investigate whether replacement of 30 % of the available carbohydrates with polydextrose (PDX) or soluble maize fibre (SCF) at breakfast and lunch would result in an increased fat oxidation rate and satiety, which may be of relevance for body weight control and diabetes prevention. In a single-blind, randomised cross-over study, eighteen overweight men and women underwent four different dietary interventions, which consisted of a PDX diet, a SCF diet and two control diets (full energetic and isoenergetic, comparable with PDX with respect to g or energy percentage of macronutrients, respectively). Glycaemic profile, energy expenditure and substrate oxidation were measured for 24 h in a respiration chamber. Circulating insulin, NEFA and TAG concentrations were determined over a 14 h period during daytime. Appetite ratings were assessed using visual analogue scales. The replacement of available carbohydrates with PDX or SCF reduced the peak glucose response, which was accompanied by reduced postprandial insulin responses. Moreover, higher concentrations of circulating NEFA were observed after consumption of both fibre diets, which were accompanied by an increased fat oxidation over 24 h. This effect was mainly attributed to the lower energetic value of the fibre diets and not to the fibres per se. Besides increasing fat oxidation, PDX exerted a pronounced suppressive effect on appetite ratings. The replacement of available carbohydrates with PDX may be of special interest because of its beneficial effects on metabolic profile and it may affect body weight control in the long term.

The present study examined the impact of breakfast and exercise on postprandial metabolism, appetite and macronutrient balance. A sample of twelve (blood variables n 11) physically active males completed four trials in a randomised, crossover design comprising a continued overnight fast followed by: (1) rest without breakfast (FR); (2) exercise without breakfast (FE); (3) breakfast consumption (1859 kJ) followed by rest (BR); (4) breakfast consumption followed by exercise (BE). Exercise was continuous, moderate-intensity running (expending approximately 2·9 MJ of energy). The equivalent time was spent sitting during resting trials. A test drink (1500 kJ) was ingested on all trials followed 90 min later by an ad libitum lunch. The difference between the BR and FR trials in blood glucose time-averaged AUC following test drink consumption approached significance (BR: 4·33 (sem 0·14) v. FR: 4·75 (sem 0·16) mmol/l; P= 0·08); but it was not different between FR and FE (FE: 4·77 (sem 0·14) mmol/l; P= 0·65); and was greater in BE (BE: 4·97 (sem 0·13) mmol/l) v. BR (P= 0·012). Appetite following the test drink was reduced in BR v. FR (P= 0·006) and in BE v. FE (P= 0·029). Following lunch, the most positive energy balance was observed in BR and least positive in FE. Regardless of breakfast, acute exercise produced a less positive energy balance following ad libitum lunch consumption. Energy and fat balance is further reduced with breakfast omission. Breakfast improved the overall appetite responses to foods consumed later in the day, but abrogated the appetite-suppressive effect of exercise.

Endurance training improves exercise performance and insulin sensitivity, and these effects may be in part mediated by an enhanced fat oxidation. Since n-3 and n-9 unsaturated fatty acids may also increase fat oxidation, we hypothesised that a diet enriched in these fatty acids may enhance the effects of endurance training on exercise performance, insulin sensitivity and fat oxidation. To assess this hypothesis, sixteen normal-weight sedentary male subjects were randomly assigned to an isoenergetic diet enriched with fish and olive oils (unsaturated fatty acid group (UFA): 52 % carbohydrates, 34 % fat (12 % SFA, 12 % MUFA, 5 % PUFA), 14 % protein), or a control diet (control group (CON): 62 % carbohydrates, 24 % fat (12 % SFA, 6 % MUFA, 2 % PUFA), 14 % protein) and underwent a 10 d gradual endurance training protocol. Exercise performance was evaluated by measuring VO2max and the time to exhaustion during a cycling exercise at 80 % VO2max; glucose homeostasis was assessed after ingestion of a test meal. Fat oxidation was assessed by indirect calorimetry at rest and during an exercise at 50 % VO2max. Training significantly increased time to exhaustion, but not VO2max, and lowered incremental insulin area under the curve after the test meal, indicating improved insulin sensitivity. Those effects were, however, of similar magnitude in UFA and CON. Fat oxidation tended to increase in UFA, but not in CON. This difference was, however, not significant. It is concluded that a diet enriched with fish- and olive oil does not substantially enhance the effects of a short-term endurance training protocol in healthy young subjects.

Capsinoids are non-pungent compounds with molecular structures similar to capsaicin, which has accepted thermogenic properties. To assess the acute effect of a plant-derived preparation of capsinoids on energy metabolism, we determined RMR and non-protein respiratory quotient (NPRQ) after ingestion of different doses of the capsinoids. Thirteen healthy subjects received four doses of the capsinoids (1, 3, 6 and 12 mg) and placebo using a crossover, randomised, double-blind trial. After a 10 h overnight fast as inpatients, RMR was measured by indirect calorimetry for 45 min before and 120 min after ingesting capsinoids or placebo. Blood pressure and axillary temperature were measured before ( − 55 and − 5 min) and after (60 and 120 min) dosing. Before dosing, mean RMR was 6247 (se 92) kJ/d and NPRQ was 0·86 (se 0·01). At 120 min after dosing, metabolic rate and NPRQ remained similar across the four capsinoids and placebo doses. Capsinoids also had no influence on blood pressure or axillary temperature. Capsinoids provided in four doses did not affect metabolic rate and fuel partitioning in human subjects when measured 2 h after exposure. Longer exposure and higher capsinoids doses may be required to cause meaningful acute effects on energy metabolism.

A gorging pattern of food intake has been shown to enhance lipogenesis and increase body weight, which may be due to large fluctuations in storage and mobilisation of nutrients. In a state of energy balance, increasing meal frequency, and thereby decreasing inter-meal interval, may prevent large metabolic fluctuations. Our aim was to study the effect of the inter-meal interval by dividing energy intake over two or three meals on energy expenditure, substrate oxidation and 24 h satiety, in healthy, normal-weight women in a state of energy balance. The study was a randomised crossover design with two experimental conditions. During the two experimental conditions subjects (fourteen normal-weight women, aged 24·4 (sd 7·1) years, underwent 36 h sessions in energy balance in a respiration chamber for measurements of energy expenditure and substrate oxidation. The subjects were given two (breakfast, dinner) or three (breakfast, lunch, dinner) meals per d. We chose to omit lunch in the two meals condition, because this resulted in a marked difference in inter-meal-interval after breakfast (8·5 h v. 4 h). Eating three meals compared with two meals had no effects on 24 h energy expenditure, diet-induced thermogenesis, activity-induced energy expenditure and sleeping metabolic rate. Eating three meals compared with two meals increased 24 h fat oxidation, but decreased the amount of fat oxidised from the breakfast. The same amount of energy divided over three meals compared with over two meals increased satiety feelings over 24 h. In healthy, normal-weight women, decreasing the inter-meal interval sustains satiety, particularly during the day, and sustains fat oxidation, particularly during the night.

Human beings are susceptible to sustained weight gain in the modern environment. Although both men and women can get fat, they get fat in different ways, and suffer different consequences. We review differences between men and women in the incidence of obesity, fat deposition patterns, fat metabolism, and the health consequences of obesity, and examine potential evolutionary explanations for these differences. Women generally have a larger proportion of body mass as fat, and are more likely to deposit fat subcutaneously and on their lower extremities; men are more likely to deposit fat in the abdominal region. Excess adipose tissue in the abdominal region, especially visceral fat, is associated with more health risks. Women have higher rates of reuptake of NEFA into adipose tissue; however, they also have higher rates of fat oxidation during prolonged exercise. Oestrogen appears to underlie many of these differences. Women bear higher nutrient costs during reproduction. Fat and fertility are linked in women, through leptin. Low leptin levels reduce fertility. Ovarian function of adult women is associated with their fatness at birth. In our evolutionary past food insecurity was a frequent occurrence. Women would have benefited from an increased ability to store fat in easily metabolisable depots. We suggest that the pattern of central obesity, more commonly seen in men, is not adaptive, but rather reflects the genetic drift hypothesis of human susceptibility to obesity. Female obesity, with excess adiposity in the lower extremities, reflects an exaggeration of an adaptation for female reproductive success.

Hormone-sensitive lipase (HSL) is believed to play a regulatory role in initiating the degradation of intramuscular triacylglycerol (IMTG) in skeletal muscle. A series of studies designed to characterise the response of HSL to three stimuli: exercise of varying intensities and durations; adrenaline infusions; altered fuel supply have recently been conducted in human skeletal muscle. In an attempt to understand the regulation of HSL activity the changes in the putative intramuscular and hormonal regulators of the enzyme have also been measured. In human skeletal muscle at rest there is a high constitutive level of HSL activity, which is not a function of biopsy freezing. The combination of low adrenaline and Ca2+levels and resting levels of insulin appear to dictate the level of HSL activity at rest. During the initial minute of low and moderate aerobic exercise HSL is activated by contractions in the apparent absence of increases in circulating adrenaline. During intense aerobic exercise, adrenaline may contribute to the early activation of HSL. The contraction-induced activation may be related to increased Ca2+and/or other unknown intramuscular activators. As low- and moderate-intensity exercise continues beyond a few minutes, activation by adrenaline through the cAMP cascade may also occur. With prolonged moderate-intensity exercise beyond 1–2 h and sustained high-intensity exercise, HSL activity decreases despite continuing increases in adrenaline, possibly as a result of increasing accumulations of free AMP, activation of AMP kinase and phosphorylation of inhibitory sites on HSL. The existing work in human skeletal muscle also suggests that there are numerous levels of control involved in the regulation of IMTG degradation, with control points downstream from HSL also being important. For example, it must be remembered that the actual flux (IMTG degradation) through HSL may be allosterically inhibited during prolonged exercise as a result of the accumulation of long-chain fatty acyl-CoA.

Dairy Ca intake has been shown to be superior to elemental Ca in increasing the loss of body fat during energy restriction. We questioned whether the mechanisms involved an increase in postprandial energy expenditure, fat oxidation and/or a greater lipolysis. The acute effects of different sources of Ca were examined in eight subjects, aged 47–66 years and BMI 27·6–36·1kg/m2, in a three-way cross-over study. Subjects were randomly provided breakfast meals either low in dairy Ca and vitamin D (LD; control), high in non-dairy Ca (calcium citrate) but low in vitamin D (HC) or high in dairy Ca and vitamin D (HD). Diet-induced thermogenesis, fat oxidation rates (FOR), carbohydrate oxidation rates (COR), insulin, glucose, ΔNEFA and glycerol were measured hourly over a 6h postprandial period. Postprandial data were calculated as a change (Δ) from the fasting value. Results showed that ΔNEFA was significantly different between meals (LD −1·50 (sem 0·26), HC −1·22 (sem 0·32), HD −0·94 (sem 0·27) mmol/l×6h; P=0·035), with a lesser suppression following both high-Ca meals. ΔFOR was significantly higher following the two high-Ca meals (LD −6·5 (sem 2·2), HC 2·93 (sem 2·34), HD 3·3 (sem 2·5) g×6;h; P=0·005), while reciprocally ΔCOR was significantly lower. ΔGlycerol was less suppressed following the high-Ca meals but statistical significance was not achieved. No differences in diet-induced thermogenesis, insulin or glucose were observed. Regardless of source, Ca intake acutely stimulated postprandial fat oxidation; and there was a lesser suppression of NEFA following these meals.

The present study investigated the effects of mixed high-carbohydrate (CHO) meals (breakfast and lunch) with different glycaemic indices (GI) on substrate metabolism during rest throughout the postprandial periods and during subsequent exercise. Nine recreationally active males completed two trials, high glycaemic index (HGI) and low glycaemic index (LGI), separated by 7 d in a randomised crossover design. In each trial, participants consumed breakfast and lunch, both of which were followed by a 3 h resting postprandial period. Following this, participants completed a 60 min run at 70 % of V˙O2max. The plasma glucose and serum insulin concentrations following both meals were significantly higher in the HGI trial than in the LGI trial (P<0·05). Serum insulin concentrations remained higher throughout the postprandial period following lunch in the HGI trial compared with the LGI trial (P<0·05). The total amount of fat oxidised was higher during the 3 h rest following lunch in the LGI trial than in the HGI trial (P<0·01) and subsequently CHO oxidation was lower (P<0·005). No significant differences in substrate utilisation were observed throughout the subsequent run. At 45 and 60 min, plasma glucose concentrations were higher in the LGI trial v. the HGI trial (P<0·05). The results of the present study provide further support that the GI concept can be successfully applied to mixed meals. The results also suggest that meals composed of LGI CHO may be more beneficial for maintaining a favourable metabolic milieu during the postprandial periods. Furthermore, during subsequent exercise, plasma glucose concentrations were better maintained following the LGI CHO meals.

The influence of the source of dietary fat on postprandial thermogenesis and substrate oxidation rates, was examined in twelve postmenopausal women aged 57–73 years, with BMI 21·9–38·3 kg/m2. A single blind, randomised, paired comparison of two high-fat, isoenergetic, mixed test meals was conducted. The major source of fat was either cream (CREAM) or extra virgin olive oil (EVOO). RMR, diet-induced thermogenesis (DIT) and substrate oxidation rates over 5 h were measured by indirect calorimetry. There were no differences in body weight, RMR, fasting carbohydrate or fat oxidation rates between the two occasions. DIT (EVOO 97 (sd 46) v. CREAM 76 (sd 69) kJ/5 h and EVOO 5·2 (sd 2·5) v. CREAM 4·1 (sd 3·7)% energy) did not differ between the two test meals. The postprandial increase in carbohydrate oxidation rates, relative to their respective fasting values (ΔCOX), was significantly lower following the EVOO meal (EVOO 10·6 (sd 8·3) v. CREAM 17·5 (sd 10) g/5 h; paired t test, P=0·023), while postprandial fat oxidation rates (ΔFOX) were significantly higher (EVOO 0·0 (sd 4·4) v. CREAM -3·6 (sd 4·0) g/5 h; P=0·028). In the eight obese subjects, however, DIT was significantly higher following the EVOO meal (EVOO 5·1 (sd 2·0) v. CREAM 2·5 (sd 2·9) %; P=0·01). This was accompanied by a significantly lower ΔCOX (EVOO 10·9 (sd 9·9) v. CREAM 17·3 (sd 10·5) g/5 h; P=0·03) and significantly higher ΔFOX (EVOO 0·11 (sd 4·4) v. CREAM −4·1 (sd 4·5) g/5 h, P=0·034). The present study showed that olive oil significantly promoted postprandial fat oxidation and stimulated DIT in abdominally obese postmenopausal women.

The present study was designed to examine the effects of mixed high-carbohydrate meals with different glycaemic indices (GI) on substrate utilization during subsequent exercise. Nine healthy male recreational runners (age 26·8 (sem 1·1) years, body mass 74·7 (sem 2·4) kg, V˙O2max 58·1 (sem 1·7) ml/kg per min) completed three trials: high-glycaemic-index meal (HGI), low-glycaemic-index meal (LGI) and fasting (FAST), separated by 7 d. The test meals contained 2 g carbohydrate/kg body mass, they were isoenergetic and the GI values were 77·4, 36·9 and 0·0 respectively. In each trial, subjects consumed the test meal 3 h before performing a 60 min run at 65 % V˙O2max on a motorized treadmill. Ingestion of the HGI and LGI resulted in hyperglycaemia and hyperinsulinaemia during the postprandial period compared with the FAST (P<0·05). The incremental area under the curve for plasma glucose was 2-fold higher for HGI compared with LGI (108·7 v. 48·9 mmol/l per min). In contrast, plasma non-esterified fatty acid concentrations were significantly lower following HGI and LGI compared with FAST (P<0·05). During the subsequent submaximal exercise, plasma glucose declined to below the fasting value in HGI compared with LGI and FAST (P<0·05). The estimated total fat oxidation was significantly higher for the LGI than the HGI during exercise (P<0·05). In summary, both pre-exercise carbohydrate meals resulted in lower rates of fat oxidation during subsequent exercise than when subjects performed exercise in the fasting state. However, the LGI resulted in a higher rate of fat oxidation during exercise than following the consumption of the HGI.

The aim of the present study was to investigate whether capsaicin assists weight maintenance by limiting weight regain after weight loss of 5 to 10%. In this randomized double-blind placebo-controlled study, ninety-one moderately overweight subjects were randomly assigned to an intensive group that underwent all the measurements, and an extensive group that underwent the same measurements except the metabolism measurements. After a 4-week very-low-energy diet (VLED) intervention, a 3-month weight-maintenance period followed. During weight maintenance, subjects were divided into a capsaicin (135 mg capsaicin/d) and a placebo group. Body mass was measured before and after the VLED and after 1, 2 and 3 months of weight maintenance. The mean body-mass loss during the VLED was 6·6 (sd 2·0) kg (7·8 (sd 1·8)% initial body mass), and was not different between the subsequent treatment and placebo group. During weight maintenance, mean % regain during treatment was not significantly different compared with placebo (33·3 (sd 35·7) v. 19·2 (sd 41·8)%, P=0·09). RQ was significantly less increased during weight maintenance in the treatment group compared with placebo (0·04 (sd 0·06) v. 0·07 (sd 0·05), P < 0·05), indicating a relatively more sustained fat oxidation. Fat oxidation (g/h) after weight maintenance was higher in the capsaicin group compared with placebo (4·2 (sd 1·1) v. 3·5 (sd 0·9), P < 0·05). These results indicate that capsaicin treatment caused sustained fat oxidation during weight maintenance compared with placebo. However, capsaicin treatment has no limiting effect on 3-month weight regain after modest weight loss.

The prevalence of obesity is increasing in westernized societies. In the USA the age-adjusted prevalence of BMI ≫30 kg/m2 increased between 1960 and 1994 from 13 % to 23 % for people over 20 years of age. This increase in the prevalence of obesity has been attributed to an increased fat intake and a decreased physical activity. However, the role of the impact of the level of dietary fat intake on human obesity has been challenged. High-fat diets, due to their high energy density, stimulate voluntary energy intake. An increased fat intake does not stimulate its own oxidation but the fat is stored in the human body. When diet composition is isoenergetically switched from low to high fat, fat oxidation only slowly increases, resulting in positive fat balances on the short term. Together with a diminished fat oxidation capacity in pre-obese subjects, high-fat diets can therefore be considered to be fattening. Another environmental factor which could explain the increasing prevalence of obesity is a decrease in physical activity. The percentage of body fat is negatively associated with physical activity and exercise has pronounced effects on energy expenditure and substrate oxidation. High-intensity exercise, due to a lowering of glycogen stores, can lead to a rapid increase in fat oxidation, which could compensate for the consumption of high-fat diets in westernized societies. Although the consumption of high-fat diets and low physical activity will easily lead to the development of obesity, there is still considerable inter-individual variability in body composition in individuals on similar diets. This can be attributed to the genetic background, and some candidate genes have been discovered recently. Both leptin and uncoupling protein have been suggested to play a role in the prevention of diet-induced obesity. Indeed, leptin levels are increased on a high-fat diet but this effect can be attributed to the increased fat mass observed on the high-fat diet. No effect of a high-fat diet per se on leptin levels is observed. Uncoupling proteins are increased by high-fat diets in rats but no data are available in human subjects yet. In conclusion, the increased intake of dietary fat and a decreasing physical activity level are the most important environmental factors explaining the increased prevalence of obesity in westernized societies.

Short-term (2 weeks) effects of the consumption of a high-energy (2920 kJ (700 kcal)) or low-energy (418 kJ (100 kcal)) breakfast on dietary patterns, blood variables and energy expenditure (indirect calorimetry) were compared in ten free-living healthy young men in a crossover study. During the high-energy breakfast, total energy intake was increased, the intake of protein and lipids was unchanged but the intake of carbohydrates was increased. Thus, 48 (SD 4) % OF ENERGY CAME FROM CARBOHYDRATES IN THE HIGH-ENERGY BREAKFAST COMPARED WITH 42 (sd 5) % in the low-energy breakfast. Excluding breakfast, the macronutrient composition of the diet remained identical in the two situations. After the high-energy breakfast, fasting serum triacylglycerol concentration was higher and HDL-cholesterol concentration was lower than after the low-energy breakfast. A high glycaemic response was observed in the morning after the high-energy breakfast period, while there was a peak of free fatty acids after the low-energy breakfast. The high-energy breakfast induced a strong inhibition of fat oxidation throughout the day. Although long-term adaptation to a high-energy breakfast cannot be excluded, the high-energy breakfast in this study did not appear to be favourable to health. Our results do not support the current advice to consume more energy at breakfast.