Resting metabolic rate in young Polynesian and Caucasian women

Transcription

1 International Journal of Obesity (1997) 21, 1071±1075 ß 1997 Stockton Press All rights reserved 0307±0565/97 $12.00 Resting metabolic rate in young and women EC Rush 1, LD Plank 2 and SM Robinson 3 1 Department of Applied Science, Auckland Institute of Technology, and Departments of 2 Surgery and 3 Physiology, University of Auckland, Auckland, New Zealand OBJECTIVE: To investigate whether resting metabolic rate (RMR) differs between and women. DESIGN: Cross-sectional comparison. SUBJECTS: Eighty-two (42, 40 ) healthy women aged between 18 and 27 y. MEASUREMENTS: RMR (indirect calorimetry) and body composition (fat-free mass and fat mass derived from oxygen- 18 dilution measurement of total body water). RESULTS: RMR was similar in the ( (s.d.) kj/d) and ( kj/d) groups while fat-free mass was signi cantly lower in the group ( vs kg, P < 0.002). After adjustment for fat-free mass and fat mass, RMR was lower in the than the groups ( vs kj/d, P ˆ 0.023). CONCLUSION: The signi cantly lower relative RMR observed in compared to women may predispose women to eventual onset of obesity. Keywords: resting metabolic rate; ethnicity; body composition; fat-free mass; indirect calorimetry; isotope dilution Introduction A low resting metabolic rate for a given size and body composition is a familial characteristic 1 and is a risk factor for body weight gain. 2 In New Zealand, the overall incidence of obesity, de ned as a body mass index (BMI) > 30 kg/m 2 is increasing. 3 people in New Zealand have twice the incidence of obesity 4 and diabetes mellitus 5 of that seen in people of origin. Physical activity is reduced and diversity in diet increased upon migration from the islands to New Zealand and with `modernisation' resulting from a shift from a rural to urban environment. Reduced physical activity and increased availability of food interacting with a genetic predisposition to obesity may lead to the increased weight gain. 6 The present study examined the hypothesis that the higher prevalence of obesity in s, compared to s, might, in part, be attributed to a low relative resting metabolic rate (RMR). We tested this hypothesis by comparing the RMR of young women with women of similar age and range of body size. Methods Subjects The study was approved by the University of Auckland Human Subjects Ethics Committee and the Auckland Institute of Technology Ethics Committee. Study subjects were 82 healthy female volunteers aged 18±27 y, selected for ethnicity and body size by personal contact and advertisement. All subjects gave their free and informed consent. Subjects were excluded who were pregnant or lactating, currently dieting or not maintaining stable body weight. Measured blood pressure and fasting blood glucose were within normal limits for all subjects, that is, the diastolic blood pressure was not greater than 90 mmhg and the fasting blood glucose was less than 5.7 mm. Forty-two identi ed themselves as, 40 as (22 Samoan, 12 Maori, 3 Tongan, 2 Nuiean and 1 Cook Islander). Twenty-one and 20 in these respective ethnic groups had a BMI > 30 kg/m 2. Average age s.d. for the group was y (range 18±27 y) and for the group was y (18±25 y). Correspondence: Elaine Rush, Department of Applied Science, Auckland Institute of Technology, Private Bag 92006, Auckland, New Zealand Received 20 August 1997; revised 24 March 1997; accepted 18 July 1997 Study protocol Two weeks prior to the present investigation, measurements were made of resting blood pressure and fasting blood glucose and Oxygen-18 labelled water was administered orally. At this time, standing height was measured to 0.1 cm with a stadiometer, weight was measured to kg on a beam balance with the subject in minimal clothing and skinfold thicknesses

2 1072 were obtained. Weight was checked for stability seven days later. Subjects were asked not to change their normal pattern of exercise the day before RMR measurement and to refrain from exercise on the morning of the measurement. Prior to RMR measurement, subjects arrived by car at the laboratory at h after an overnight fast. Water was allowed ad libitum. RMR was measured by indirect calorimetry approx 30 min after arrival. All measurements on all subjects were made by the same investigator. The laboratory was airconditioned and at a constant temperature of 21.5 C. Barometric pressure and humidity were measured before each procedure. Anthropometry and body composition Skinfold thicknesses (triceps, biceps, subscapular and suprailiac) were measured as the average of triplicate measures using Harpenden calipers and standard techniques. 7 Total body water (TBW) was determined by stable isotope (oxygen-18) dilution using the multipoint slope/intercept method as described by Coward. 8 An initial dose of 1.5 g of 10.16% H 2 18 O (Enritech, Weizmann Institute of Science, Rehovot, Israel) per kg of fat free mass predicted from the anthropometric measurements was given. Timed urine samples were collected 4 h and 5 h after dose administration and then 2,5, 7 and 14 d later (the day of RMR measurement). These were the points used to determine the oxygen-18 dilution space which was divided by 1.01 to give TBW. Fat mass (FM in kg), fat-free mass (FFM in kg) and percentage fat, were determined using the assumption that FFM is 73% water. Resting metabolic rate (RMR) Subjects were familiarised with the required procedure immediately on arrival at the laboratory. Heart rate was monitored using a portable cardiotachometer (Sport Tester PE3000, Polar Electro, Kempele, Finland). The subject sat on a chair and listened to quiet music with a noseclip and mouthpiece in place. The measurement protocol started when the heart rate had been stable for at least 10 min. Four further minutes of stable heart rate were recorded before the rst veminute expired air collection into a Douglas bag was obtained, followed one minute later by another veminute sample. The subjects wore a noseclip and breathed through a mouthpiece connected via a twoway Koegel respiratory valve (Ewald Koegel Co., TX, USA) to an open respiratory system. Room air was inspired through the valve and expired gas was directed through a manifold, with taps to the Douglas bags. Immediately the procedure was completed the collected expired gas was sampled sequentially from each bag via a small tube containing granulated anhydrous magnesium perchlorate. The dried gas was then passed at 200 ml/min through, in turn, a paramagnetic oxygen analyser (Servomex OA570, Taylor Instrument Analytics Ltd, Crowborough, Sussex, England) and then an infrared carbon dioxide analyser (Medical Gas Analyser LB2, Beckman Instruments Ltd, IL, USA). The gas analysers were calibrated against known gas standards (5.00% CO 2, and 14.90% O 2 ), pure nitrogen (0.00% CO 2 and 0.00% O 2 ) and fresh air (0.03% CO 2 and 20.94% O 2 ). The volume of gas in each Douglas bag was measured by drawing the gas through a coal gas meter. This meter was calibrated and checked for linearity against a Tissot bell spirometer (W.E. Collins, MA, USA). Volumes samples for gas analyses were added to the volumes measured. Oxygen uptake (VO 2 ) and carbon dioxide consumption (VCO 2 ) were calculated and the respiratory exchange ratio (R) derived. Resting oxygen uptake was taken as the average of the two ve-minute samples and converted to energy consumption by multiplying by the thermal equivalent of oxygen for the calculated respiratory quotient. 9 The coef cient of variation of the duplicate measures of oxygen uptake for a subject was 5.9%. Statistical analysis Statistical analyses were carried out using SAS (version 6.04, SAS Institute, Cary, NC, USA). Statistical comparisons of groups were made by unpaired t-test. Linear and multiple regression analysis was utilised to investigate relationships between selected variables. Regression relationships were compared using analysis of variance (ANOVA) to test for homogeneity of slopes and analysis of covariance for differences in elevation. Statistical signi cance was set at P < Results are expressed as mean s.d. unless otherwise stated. Results Physical characteristics of the subjects are shown in Table 1. The women had signi cantly lower (P ˆ ) FFM than the women. The similar FM in the two groups appeared to be distributed more centrally in the group as judged by the higher subscapular/triceps skinfold ratio (STR) in this group (P ˆ ). Fasting blood glucose whilst not abnormally elevated in any subject, was signi cantly higher in the group (P ˆ ). Table 2 summarises the results of the indirect calorimetry measurements. The fasting respiratory exchange ratio was similar for the two ethnic groups. Absolute RMR was similar in the two groups. FFM explained 58% (P ˆ ) of the variation in RMR in the group and 55% (P ˆ ) in the group. The addition of FM explained a further 3% (P ˆ 0.1) in both groups. Using both FFM and FM in multiple regression

3 Table 1 Physical characteristics of the subjects a (n ˆ 42) (n ˆ 40) 1073 Height (cm) (153±180) (153±178) Weight (kg) (48.1±138.3) (56.2±131.7) Age (y) 22 2 (18±27) 22 2 (18±25) BMI (kg/m 2 ) (16.4±48.0) (19.8±51.8) Body fat (%) (22.1±59.3) (26.0±54.0) Fat mass (kg) (10.9±82.0) (15.6±71.1) Fat-free mass (kg) (30.5±57.5) (37.0±65.5)* Biceps skinfold (cm) (0.37±3.79) (0.42±2.60) Triceps skinfold (cm) (1.03±5.21) (1.03±4.85) Subscapular skinfold (cm) (0.78±5.91) (1.05±5.74) Suprailiac skinfold (cm) (0.68±6.43) (0.72±6.05) Subscapular/triceps ratio (0.38±1.45) (0.64±1.77)* Fasting glucose (mm) (2.8±5.1) (2.9±5.3)* Systolic blood pressure (mmhg) (105±140) (105±130) Diastolic blood pressure (mmhg) 78 7 (65±90) 77 6 (65±90) Resting heart rate (bpm) (51±100) (51±96) a Mean s.d.; range in parentheses. BMI, body mass index. * P < between ethnic groups by two sample independent t-test. Table 2 Results of indirect calorimetry measurements a (n ˆ 42) (n ˆ 40) Respiratory exchange ratio (0.69±0.98) (0.69±0.92) Oxygen uptake (ml/min STPD) (168±349) (168±367) RMR (kj/d) (4827±10293) (4866±10828) a Mean s.d.; range in parentheses. RMR, resting metabolic rate. equations gave the following results: RMR kj=d ˆ 1281:8 111:8 FFM 18:0 FM; R 2 ˆ 0:61; SEE standard error of estimate ˆ 823 kj=d RMR kj=d ˆ 500:5 114:7 FFM 22:3 FM; R 2 ˆ 0:58; SEE ˆ 855 kj=d ANOVA showed that the regression slopes of RMR, on each of the variates FFM and FM, are statistically indistinguishable for the two ethnic groups. The common slopes regression equation is: RMR kj=d ˆ 1145:6 113:7 FFM 19:4 FM 497:5 group Figure 1 Relationship between resting metabolic rate (RMR) and fat-free mass (FFM) for 42 (open symbols) and 40 (closed symbols) women. The linear regressions are RMR ˆ FFM (r 2 ˆ 0.59, SEE 842 kj/d) for the s (dashed line) and RMR ˆ FFM (r 2 ˆ 0.55, SEE 872 kj/d) for the s (solid line). R 2 ˆ 0:60; SEE ˆ 829kJ=d where group is coded as 0 for s and 1 for s. Covariance analysis showed that RMR, after adjustment for both FFM and FM, was signi cantly lower in the group ( vs kj/d, P ˆ 0.023). Using FFM alone as the covariate, also yielded a signi cant difference in RMR between the groups, and kj/d for s and s, respectively (P ˆ ). Figure 1 shows the relationships between RMR and FFM for the two groups. When STR was included as a covariate in addition to FFM and FM, the difference in RMR between the

4 1074 two ethnic groups was eliminated. The RMR adjusted for these three covariates was kj/d for the s and kj/d for the s (P ˆ 0.26). The multiple regression equation for all subjects is: RMR kj=d ˆ 1958:8 107:9 FFM 30:3 FM 1161:2 STR R 2 ˆ 0:62; SEE ˆ 805 kj=d An increase in STR, at xed FFM and FM, is associated with a decrease in RMR. Discussion The present study demonstrates that the women have a lower relative RMR than the women. Genetic differences in energy requirements have been shown in twin and family studies. 10,11 Our study provides further con rmation that subjects from different ethnic backgrounds with apparently similar physical characteristics (weight and height) may require more or less energy to maintain their body weights. A recent study has been reported 12 where the RMR of 21 African-American women was compared with 18 women of similar physical characteristics. After adjusting for FFM, measured by dual-energy x-ray absorptiometry, the African women had a signi cantly lower RMR than the women. This difference was not related to FM or body fat distribution. The difference was 523 kj/d (125 kcal/d) compared to 676 kj/d found in the present study. Ravussin and Bogardus 13 have reviewed the relationship of genetics to energy expenditure and conclude that although an environmental component cannot be excluded, a genetic determinant of RMR can be demonstrated. About 20 molecular markers for determining human obesity phenotypes and fat distribution have been demonstrated in studies reviewed by Bouchard. 14 A number of reports in the literature reveal that measurements of resting and basal metabolic rate vary with different populations and techniques of measurement. There is a decline in RMR with age and an increase with greater physical activity and energy intake. 15 The limited age range of our subjects means that we could not test an age effect. The higher metabolic rate of the subjects may re ect a more active lifestyle, but given the young age and that most of the subjects were drawn from student and employed backgrounds in urban Auckland, this is not a likely cause. It may also re ect different oxidative capacities and proportions of the various tissues of the FFM between the two groups. Clark and coworkers 16,17 have demonstrated in 18 women but not in 13 men, that the RMR (adjusted for FFM) of `small-eating' women is lower by 310 kj/d (74 kcal/d) than that of `large-eating' women of the same ethnicity. They have also shown 18 that `small-eaters' burn more carbohydrate in the fasted state than the `large-eaters'. Thus within an ethnic group there have been differences, measured in metabolic rate, fuel and appetite. We did not select for appetite in this study, but we have no reason to believe that the mix of `small' and `large' eaters in our groups is dissimilar. Resting skeletal muscle metabolism has been shown to be related to RMR 19 with thyroid hormone levels, muscle bre types, muscle tone, sympathetic innervation and catecholamine levels being postulated as possible in uential factors. Ravussin 20 reviewed studies of the role of the sympathetic nervous system (SNS) in the genesis of a low RMR as a risk factor for weight gain. Studies in s indicate that SNS activity is related to RMR, but in the Pima Indians who have low SNS activity, the RMR is not correlated with SNS activity. The Tokelau study 21 presents evidence that in a limited study, women on two different atolls, showed no signi cant difference in 24 h urine catecholamine levels, but men in the same study did. This raises the questions of whether s have lower SNS activity than s and whether there are differences between the genders. SNS activity will raise blood glucose, but in our group the mean fasting blood glucose was signi cantly higher. Bouchard and coworkers, in twin studies, have shown a genetic effect for both RMR 11 and fasting blood glucose. 10 Skeletal muscle is a large fraction of FFM. As expected, in this study FFM was the best single predictor of resting energy expenditure. We found that FM and ethnicity also contribute to differences in RMR. Ethnic differences in body fat distribution as measured by the STR was a stronger predictor than ethnicity. STR is a measure of upper body, central to peripheral fat distribution. 22 In our study, an increase in upper truncal fat is associated with a decrease in RMR. Young South Asians in the UK have a higher STR than Europids matched for age and BMI. 23 The Asian group had relative hyperglycaemia and hypercholesterolaemia and a high prevalence of noninsulin-dependent diabetes and ischaemic heart disease. Our women also had a higher blood glucose and a higher STR. Bivariate genetic analysis of genetic and environmental correlations of skinfold measures in 408 male and female Mexican-Americans 24 has demonstrated that patterns of central vs peripheral fat distribution, are largely a function of genetics. A relatively high STR and fasting blood glucose have been shown to be associated with `Syndrome X' in Mexican-Americans, studied over eight years as part of the San Antonio Heart Study. 22,25 Syndrome X describes a condition consisting of insulin resistance, compensatory hyperinsulinaemia, glucose intolerance, hypertriglyceridaemia, low levels of high density lipoprotein (HDL) cholesterol and hypertension. There is an increased preva-

5 lence of these risk factors in New Zealand people of origins compared to s It is simplistic to imply a low RMR is due to having parents. What we have measured are a number of phenotypic differences among women probably due to expression of a combination of genes, coupled with a relatively af uent Western lifestyle. It does however lead to speculation that a `thrifty genotype' as proposed by Neel 29 may have some validity. A predisposition to weight gain in women may be related to a low RMR. Further prospective studies are required to look at the associations of body fat distribution, ethnicity, blood glucose and RMR as risk factors for increased noninsulin-dependent diabetes and ischaemic heart disease in people in New Zealand. Acknowledgements This work was supported by grants from the Auckland Institute of Technology Contestable Research Fund and the Health Research Council of New Zealand. References 1 Bogardus C, Lillioja MB, Ravussin E, Abbott W, Zawadzki JK, Young A, Knowler WC, Jacobowitz R, Moll PP. Familial dependence of the resting metabolic rate. N Engl J Med 1986; 315: 96± Ravussin E, Lillioja S, Knowler WC, Christin L, Freymond D, Abbott WGH, Boyce V, Howard BV, Bogardus C. Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med 1988; 318: 467± Simmons G, Jackson R, Swinburn B, Yee RL. The increasing prevalence of obesity in New Zealand: is it related to recent trends in smoking and physical activity. NZ Med J 1996; 109: 90±92. 4 Dryson E, Metcalf P, Baker J, Scragg R. The relationship between body mass index and socioeconomic status in New Zealand: ethnic and occupational factors. NZ Med J 1992; 105: 233± Simmons D, Gatland B, Flemming C. Prevalence of known diabetes in a multiethnic community. NZ Med J 1994; 107: 219± McGarvey S. Obesity in Samoans and a perspective on its etiology in s. Am J Clin Nutr 1991; 53(suppl): 1586S±1594S. 7 Durnin JVGA, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 1974; 32: 77±97. 8 Coward WA. Calculation of pool sizes and ux rates. In: Prentice AM, (ed.) The doubly-labeled water method for measuring energy expenditure. Technical recommendations for use in humans. International Atomic Energy Agency: Vienna, 1990, pp 48±68. 9 McArdle W, Katch F, Katch V. Exercise Physiology: Energy, Nutrition and Human Performance. Lea and Febiger: Philadelphia, 1981, p Bouchard C, Tremblay A, Nadeau A, DespreÂs J-P, Theriault G, Boulay MR, Lortie G, Leblanc C, Fournier G. Genetic effect in resting and exercise metabolic rates. Metabolism 1989; 38: 364± Bouchard C, Perusse L, Deriaz O, DespreÂs J-P, Tremblay A. Genetic in uences on energy expenditure in humans. Crit Rev Food Sci Nutr 1993; 33: 345± Albu J, Shur M, Curi M, Murphy L, Heyms eld S, Pi-Sunyer FX. Resting metabolic rate in African American women [Abstract]. FASEB J 1996; 10: A Ravussin E, Bogardus C. Relationship of genetics, age, and physical tness to daily energy expenditure and fuel utilisation. Am J Clin Nutr 1989; 49: 968± Bouchard C. Genetics of obesity: an update on molecular markers. Int J Obes 1995; 19(Suppl): S10±S Mole PA. Impact of energy intake and exercise on resting metabolic rate. Sports Med 1990; 10: 72± Clark D, Tomas F, Withers RT, Brinkman M, Chandler C, Phillips J, Ballard FJ, Berry MN, Nestel P. Differences in energy metabolism between normal weight `large-eating' and `small-eating' women. Br J Nutr 1992; 68: 31± Clark D, Tomas F, Withers RT, Neville SD, Noland SR, Brinkman M, Chandler C, Clark C, Ballard FJ, Berry M, Nestel P. No major differences in energy metabolism between matched and unmatched groups of `large eating' and `small eating' men. Br J Nutr 1993; 70: 393± Clark DG, Tomas FM, Withers RT, Brinkman M, Berry MN, Oliver JR, Owens PC, Butler RN, Ballard FJ, Nestel PJ. Differences in substrate metabolism between self-perceived `large-eating' and `small-eating' women. Int J Obes 1995; 19: 245± Zurlo F, Larson K, Bogardus C, Ravussin E. Skeletal muscle metabolism is a major determinant of resting energy expenditure. J Clin Invest 1990; 86: 1423± Ravussin E. Low resting metabolic rate as a risk factor for weight gain: role of the sympathetic nervous system. Int J Obes 1995; 19(Suppl): S8±S9. 21 Jenner DA, Harrison GA, Prior IA, Leonetti DL, Fujimoto WY, Kabuto M, Fujimoto W. Inter-population comparisons of catecholamine excretion. Ann Hum Biol 1987; 14: 1±9. 22 Stern MP, Morales PA, Haffner SM, Valdez RA. Hyperdynamic circulation and the insulin resistance syndrome (`Syndrome X'). Hypertension 1992; 20: 802± Potts J, Simmons D. Sex and ethnic group differences in fat distribution in young United Kingdom South Asians and Europids. J Clin Epidemiol 1994; 47: 837± Comuzzie AG, Blangero J, Mahaney MC, Mitchell BD, Stern MP, MacCluer JW. Genetic and environmental correlations among skinfold measures. Int J Obes 1994; 18: 413± Haffner SM, Valdez RA, Hazuda HP, Mitchell BD, Morales PA, Stern MP. Prospective analysis of the insulin-resistance syndrome (Syndrome X). Diabetes 1992; 41: 715± Scragg R, Baker J, Metcalf P, Dryson E. Prevalence of diabetes mellitus and impaired glucose tolerance in a New Zealand multiracial workforce. NZ Med J 1991; 104: 395± Scragg R, Baker J, Metcalf P, Dryson E. Hypertension and its treatment in a New Zealand multiracial workforce. NZ Med J 1993; 106: 147± Scragg R, Baker J, Metcalf P, Dryson E. Serum lipid levels in a New Zealand multiracial workforce. NZ Med J 1993; 106: 96± Neel JV. A `thrifty' genotype rendered detrimental by `progress'? Am J Hum Genet 1962; 14: 353±