Thermic effect of feeding carbohydrate, fat, protein and mixed meal in lean and obese subjects1 2

Transcription

1 Original Research Communications-general Thermic effect of feeding, fat, protein and meal in lean and obese subjects1 2 R Swaminathan, RFGJ King, J Holmfield, RA Siwek, M Baker and JK Wales ABSTRACT The thermic effect of 1.67 MJ (400 kcal) of (glucose), fat,protein and meal were examined in 11 lean and 11 obese subjects by indirect calorimetry. The changes in metabolic rate in response over 90 mm period ( mm after the meal) to the different meals were compared with that seen after a similar volume of low calorie drink. The thermic effects of glucose and protein were not significantly different between lean and obese subjects. Obese subjects showed very little increase in metabolic rate following ingestion of fat (-0.9 ± 2.0%, 1 ± SEM) and this was significantly different from that seen in lean subjects (14.4 ± 3.4%). The thermogenic response to meal was also significantly lower in obese subjects when expressed as percentage change (12.9 ± 2.3% compared to 25.0 ± 4.8%). There was no evidence for delay in gastric emptying times for glucose and fatty meal in the six obese subjects in whom these were measured. We conclude that obese subjects show a reduced thermogenic response to fat. AmJClinNuzr l985;42:l KEY WORDS Thermogenesis, obesity, diet-induced thermogenesis, gastric emptying Introduction For many years, it has been known that some subjects do not gain weight easily even when overfed (1, 2). This inability to gain weight has been attributed in part to an increased thermogenic effect of feeding (3, 4). On the other hand, some subjects tend to gain weight easily and a reduced thermic effect of feeding has been suggested as a possible factor for this (3). Wang and Strouse (5) demonstrated a reduced increment in metabolic rate after a meal in obese subjects. Pittet et al (6) using direct and indirect calorimetry showed a reduced thermic effect to a 50-g glucose load in obese subjects, when only a 5% rise in metabolic rate was observed compared to 13% in lean subjects. Shetty et al (7) and Kaplan and Leveille (8) have confirmed this finding. However, others have not been able to confirm a reduced thermic effect of food in obese subjects (9-13). In order to clarify whether any change in the thermic effect of food in obesity may be due to the composition of the food, we have examined the effect of isocaloric amounts of, protein, fat and meal on postprandial metabolic rate in lean and obese subjects. Materials and Methods Eleven healthy subjects (8F and 3M) and 11 obese subjects (8F and 3M) were studied. Their age, height, From the Departments of Chemical Pathology, Surgery (RFGJK, JH), Medical Physics (RAS, MB), Nuclear Medicine (RAS, MB) and Medicine (JW), University of Leeds, eds. 2Ade reprint requests to: Dr R Swaminathan, Department of Chemical Pathology, Prince of Wales Hospital, Sha Tin, Hong Kong. Received vember 24, Accepted for publication February 12, The American Journal of Clinical Nutrition 42: AUGUST 1985, pp Printed in USA 1985 American Society for amid Nutrition 177

2 178 SWAMINATHAN ET AL body weight, Th2 shown in Table 1. ne of the obese subjects were diabetic or hypertensive, and their thyroid functions were normal. All subjects were studied as outpatients and they came to the hospital at 8:30 AM after an overnight fast for their measurements. The subjects lay supine on a bed and remained at bed rest throughout the study period. After an acclimatization period of 30 mm, the resting metabolic rate was measured for 30 mm. Following this period the test meal was eaten and the metabolic rate was measured for a further 120 mm. The test meals consisted of either a low-calorie lemon drink, 1.67 Mi glucose (100 g crystalline glucose) dissolved in low-calorie lemon juice, 1.67 Mi fat (as 38 g of corn oil and 70 g of pure tomato juice), 1.67 Mi protein (as 114 g of Maxipro (Scientific Hospital Supplies Ltd. Liverpool, UK)-a commercial protein supplement] or 1.67 Mi of meal (cheese, cream cracker, grapefruit juice and sugar, containing protein, fat and in the ratio 1:2.8:4.6). Each test meal was given on a separate day, at the same time of day. Due to the length of time involved in the study, all the subjects could not take part in the full study. All subjects (11 lean and 11 obese) took part in the basal (lemon drink) measurement. All 11 lean and 9 obese subjects drank the glucose meal, 7 lean and 10 obese subjects took part in the fat study, 5 lean and 8 obese subjects took the protein and 8 lean and 10 obese subjects took the meal. Metabolic rate was continuously measured by monitoring oxygen consumption and carbon dioxide production by the ventilated hood technique (14). Oxygen concentration was measured by a paramagnetic analyzer (Taylor Servomex, Sussex, UK)-and carbon dioxide concentration by an infrared analyzer (PK Morgan Ltd. Kent, UK). At each session a standard gas mixture (British Oxygen Co. London) was introduced into the ventilated hood and the analyzers were calibrated. Metabolic rate was calculated from the oxygen consumption and respiratory quotient neglecting protein metabolism (15). Metabolic rate during the period from 30 to 120 mm after each meal was compared to a similar period after the low-calorie drink (basal), thirty minutes after ingestion being allowed for equilibration. The day to day variation in resting metabolic rate for each subject was assessed by calculating the coefficient of variation of the metabolic rate during the 30-mm period before each meal when more than three values were available. The mean coefficient of variation in metabolic rate was 4.9% (±0.9%)(n = 8) in lean subjects and 4.1% (±0.8)(n = 9) in obese subjects. In eight lean and nine obese subjects, total body potassium was measured by a whole-body counter as described by Burkinshaw (16). In 6 of the 11 obese subjects, gastric emptying was measured on two separate occasions, once after a glucose meal and once after fatty meal (corn oil and tomato juice). Each test meal containing technetium-99m ( T) sulphur colloid was given and gastric emptying was measured using a gamma camera. Results are given as metabolic rate per minute for the study period and as mean ± SEM. Metabolic rate after a test meal was compared to the metabolic rate after a low-calorie drink in each subject. Changes in metabolic rate due to a test meal within a group (either lean or obese) was compared by paired test for paired data. Differences between lean and obese groups were compared by Students: test and Mann-Whitney U test for unpaired data. Protocol for the study was approved by the Research Ethics Committee of the General Infirmary at Leeds. Results Resting metabolic rate Resting metabolic rate in the obese subjects was significantly higher (p <0.05) than in lean subjects (Table 1). Total body potassium (TBK) was higher in obese subjects than in lean subjects and there was a highly significant correlation between TBK and resting metabolic rate (r = 0869; p <0.01). Response to meal The metabolic rate during the period mm and at the end of 120 mi after low-calorie drink,, fat, protein, and meal for lean and obese subjects is shown in Table 2. In lean subjects all the meals tested produced a significant elevation in metabolic rate during the study period (Table 3). The response was greatest for the meal and then for protein, and fat. The metabolic rate at the end of the study period remained elevated (Table 2). In obese subjects,, protein and meal produced a significant increase in metabolic rate and the greatest response was seen after protein meal. There was no significant change in metabolic rate after fat. The metabolic rate remained elevated at the end of the study period for, protein and meal. When TABLE 1 Characteristics of subjects studied Lean subjects (n11) Obese subjects (n)l) Age (yr) Weight (kg) l44t Height (cm) W/H2 (kg/rn2) 20.5 ± ± 2.39t Resting metabolic rate (ki/min) 3.56 ± ± O.22 Total body potassium 2462± ±515 (mmol) (n=8) (n=9) Results for age, weight and height are given as range and the rest as mean ± SEM. Significanceof by Student I test and Mann- Whitney U test. * p <0.05. t p <0.01.

3 DIET-INDUCED THERMOGENESIS AND OBESITY 179 TABLE 2 Metabolic rate after lo w calorie drink or test meals (glucose, fat, protein or meal) in lean and o bese subjects Metabolic rate (ki/min) Lean subjects Obese subjects Mean metabolic rate Mean metabolic rate during the period 30- Metabolic rate at during the period 30- Metabolic rate at l20minalterameal theendofstudy l20minafterameal theendofstudy Low-calorie drink 3.56 ± 0.23 (11) 3.37 ± ± 0.22 (11) 4.70 ± 0.29 Glucose 4.08 ± 0.28 (11) 4.00 ± ± 0.28 (9) 5.21 ± 0.33 Fat 3.84 ± 0.29 (7) 3.68 ± ± 0.21 (10) 4.63 ± 0.30 Protein 4.22 ± 0.40 (5) 4.07 ± ± 0.26 (8) 5.56 ± 0.35 Mixed meal 4.47 ± 0.31 (8) 4.37 ± ± 0.25 (10) 5.01 ± 0.30 Results are mean ± SEM. Number in parenthesis indicate the number of subjects studied for each meal. the changes in metabolic rate in lean and obese subjects are compared, the response for fat was less in obese, whether expressed either in absolute terms or as percentage (p <0.01). When the change in metabolic rate is expressed as a percentage, obese subjects showed a reduced response to meal as well (p <0.05) (Table 3). The change in metabolic rate observed in response to the meal was related to that seen in response to fat when obese and lean subjects are included in the analysis (r = 0.620; n = 15, p <0.05). There was no correlation between obesity index and the thermogenic response to (r = -0.06) or protein (r = 0.05) or meal (r = 0.06). But there was a significant negative relationship between obesity index and fat response (r = -0.50; p <0.05). Gastric emptying time Gastric emptying time was 99 ± 27 mm after glucose and 58 ± 15 mm after fat in the six obese subjects. The range seen in normal subjects is mm. The gastric emptying time for fat was not greater than that for glucose. Discussion This study shows that the increase in metabolic response ie the thermogenic response, to a meal or fat meal is lower ( mm after the meal) in obese subjects than in lean subjects. This agrees with the results of Shetty et al (7) and Kaplan and Leveille (8) but does not agree with that of Nair et al (11), and Welle and Campbell (12). The data from some of these studies are shown in Table 4. It can be seen in most of the studies that either only (glucose) or a meal was used to examine thermogenesis. When glucose only was used, no between obese and lean subjects was observed (12, 13, 17) except in the study of Pittet et al (6). In this careful study, both direct and indirect calorimetry were used to TABLE 3 Changes in metabolic rate in response to 1.67 kj (400 kcal) of, fat, protein and meal in lean and obese subjects Meal Absolute cbs nges (ki/min) Percentage changes (%) Lean subjects Obese subjects Lean subjects Obese subjects Carbohydrate ± 0.094* (11) ± (9) 13.9 ± ± 3.4 Fat ± 0.102* (7) ± 0.099f (10) 14.4 ± ± 2.Ot Protein ± 0.133* (5) ± 0.172* (8) 22.5 ± ± 3.8 Mixed meal ± 0.162* (8) ± 0.099* (10) 25.0 ± ± 2.3t Results are mean ± SEM and the number in parenthesis are the number of subjects studied. Results shown are the changes in metabolic rate/mm during the period mm after a meal compared to similar period after a low-calorie drink. * Significantlydifferentby paired I test (p <0.05). t Significantly different between lean and obese subjects by unpaired I test and Mann-Whitney U test (p <0.05).

4 180 SWAMINATHAN ET AL TABLE 4 Analysis of data from literature Rekesnce Kaplan and Leveille (1976) (Ref 8) Meal (high protein) value 800 kcal Pittet et al (1976) (Ref 6) 200 kcal Shetty et al (1981) (Ref 7) kcal Morgan et al (1982) (Ref 4) 500 and 1000 kcal Reduced in low energy intake, group of normal Sharief and MacDonald (1982) (Ref 17) 350 kcai Nair et al (1983) (Ref 11), fat,protein 300 kcal Schwartz et al (1983) (Ref 25) (high protein) 800 kcal Welle and Campbell (1983) (Ref 12) 400 kcal Present study, fat,protein 400 kcal for fat and meal study the response to glucose. A significant reduction in thermogenic response was noted in obesity in those studies where a meal of greater than 500 kcal (2.08 MJ)was used. Nairet al (11) studied thermic response to 300 kcal (1.25 MJ) of, fat and protein separately and failed to find a between obese and normal subjects. One possible explanation is the lower energy value of their meal used, 300 kcal (1.25 MJ) compared to 400 kcal (1.67 MJ) or more used in all other studies. When a meal was employed to test the thermic effect, a between obese and lean subjects was noted (Table 4). In the present study, we observed a correlation between the thermic response to fat and thermic response to meal. We would like to support the suggestion of Zed and James (18) that the lower thermic effect in obese subjects in response to a meal is due to its fat content. One of the problems in the comparison of thermogenesis using a test meal is that the fasted period can be taken in sequence to the meal, or comparison can be made between similar times on different days. We have adapted the latter approach to avoid the possibility of changes in metabolic rate during the day. However, the disadvantage of this method is the day to day variability in metabolic rate. The coefficient of variation in resting metabolic rate in each subject was 4.9% for lean and 4.1% for obese subjects. The changes due to test meal were about 3-6 times higher. The metabolic rate was measured only for 120 mm after a meal and itcould be argued that the s between lean and obese groups in response to a test meal could disappear ifthe measurements had been continued for longer period. However, the dif ferences in metabolic rate still persisted at the end of 120 mm, and the results from other groups support our conclusion. Using the respiration chamber, the Lausanne group studied the overall thermogenic response to three meals over a 14-h period in lean and obese subjects (19-21) and found that the mean increase in energy expenditure during the 14 h was 35% in lean and 20% in obese subjects. The observed poor thermic effect of fat in obese subjects could be due to delayed gastric emptying of fat in obese subjects. We could not demonstrate a delay in gastric emptying of fat in obesity. In fact, some authors have reported that in obese subjects, emptying of solid meal is faster (22). It has been argued that a reduced thermogenic response to a meal is not relevant in the etiology of obesity based on measure-

5 DIET-INDUCED THERMOGENESIS AND OBESITY 181 ments made in obese subjects (7). However, this mechanism may be important during the early stages viz the dynamic phase of obesity. Iftwo subjects (both of normal weight initially),one having normal, and the other having subnormal thermogenic response to a meal, increase their food intake, the former will not put on as much weight as the latter. Thus the demonstration of a reduced thermogenic response to a meal in obese subjects may identify a component of metabolic problem. It could be argued that the changes we have observed are the result of obesity, and not related to the etiology of obesity. In animal experiments, Tobin and Hervey (23) showed that rats, overfed for a period of time to make them obese, showed a reduced but prolonged thermogenic response to a meal. On the other hand, Shetty et al (7) observed that, in post-obese as well as obese subjects, postprandial thermogenesis is reduced. Recently Schutz et al (24) and Bessard et al (21) have reported that diet-induced thermogenesis was sub-normal even after weight loss. These studies therefore suggest that the reduced thermic effect of food is in fact due to obesity. We conclude that thermogenic response to fat and meal is lower in obese subjects, and to demonstrate the probably 1.67 Mi or more is required. 0 The financial support of the Yorkshire Regional Health Authority is gratefully acknowledged. References 1. Miller AS, Mumford P. Gluttony 1. An experimental study of overeating low or high-protein diets. Am J Clin Nutr l96720: Sims EAH. Experimental obesity, dietary intake and their clinical implications. Clin Endocrinol Metab l976;5: Miller DS, Mumford P. Stock Mi. Gluttony 2. Thermogenesis in overeating man. Am 3 Clin Nutr l967, Morgan JB, York DA, Wasilewska A, Portman 3. A study of the thermic responses to a meal and to a syinpathomimetic drug (epheclrine) in relation to energy balance in man. Br 3 Nutr l982;47:2l Wang CC, Strouse S. Studies on the metabolism of obesity III. The specific dynamic action of food. Arch Intern Med l924;34: Pittet PH, Chappuis PH, Acheson K, De Techtermann F, J#{233}quierE. Thermic effect of glucose in obese subjects studied by direct and indirect calorimetry. Br J Nutr l976;35:28l Shetty PS, Jung RT, James WPT, Barrand MA, Callingham BA. Postprandial thermogenesis in obesity. Clin Sci l98l;60.5l Kaplan ML, Leveille GA. Calorigenic response in obese and nonobese women. Am 3 Clin Nutr :l Clough DP, Durnin SVGA. The rise in metabolic rate following the ingestion of single large meals by thin and average men and women. J Physiol (Lond) l97o,207:89p. 10. Bradfiekl RB, Jourdan MH. Relative importance of specific dynamic action in weight-reduction diets. Lancet l973;2: II. Nair KS, Halliday D, Garrow is. Therinic response to isoenergetic protein, or fat meals in lean and obese subjects. Clin Sd l983;65:307-l Welle SL, Campbell RG. rmal thermic effect of glucose in obese women. Am 3 Clin Nutr l983;37: Segal KR, Gutin B. Thermic effects of food and exercise in lean and obese women. Metabolism 1983;32:58l Kappagoda CT, Stoker JB, Linden RT. A method for the continuous measurement of oxygen consumption. J Appl Physiol l974;37: De Weir JB. New method for calculating metabolic rate with special reference to protein metabolism. 3 Physiol l949;l09:l Burkinshaw L Sex-dependent calibration factor of a whole-body radiation counter. mt 3 AppI Rad and Isotopes l978;29: Sharief NN, Macdonald I. Differences in dietaryinduced thermogenesis with various s in normal and overweight men. Am 3 Clin Nutr 1982;35: Zed CA, James WPT. Thermic response to fat feeding in lean and obese subjects. Proc Nutr Soc 1982;4l:32A. 19. Schutz Y, Bessard T, J#{233}quier E. Diet-induced thermogenesis measured over a whole day in obese and nonobese woman. Am J Clin Nutr l984;40: J#{233}quierE, Schutz Y. Long-term measurements of energy expenditure in human using a respiration chamber. Am J Clin Nutr l983;38: Bessard T, Schutz Y, J#{233}quierE. Energy expenditure and postprandial thermogenesis in obese woman before and after weight loss. Am 3 Clin Nutr l983;38: Wright PA, Kiinsky 5, Fleeman C, Trujillo 3, Teague E. Gastric emptying and obesity. Gastroenterology 1983;84: Hervey GR, Tobin G. The part played by variation of energy expenditure in the regulation of energy balance. Proc Nutr Soc 1982;4l:l Schutz Y, Golay A, Felber J, J#{233}quierE. Decreased glucose-induced thermogenesis after weight loss in obese subjects a predisposing factor for relapse of obesity? Am 3 Clin Nutr l984;