Low relative resting metabolic rate and body weight gain in adult Caucasian Italians

Transcription

1 (2005) 29, & 2005 Nature Publishing Group All rights reserved /05 $ PAPER Low relative resting metabolic rate and body weight gain in adult Caucasian Italians S Buscemi 1 *, S Verga 1, G Caimi 1 and G Cerasola 1 1 Department of Internal Medicine and Cardiovascular Diseases, University of Palermo, Italy OBJECTIVE: To investigate the relationship between resting metabolic rate (RMR) and subsequent changes in body size and degree of fatness in a group of adult Caucasian Italians. DESIGN: Prospective, longitudinal, observational study. SUBJECTS: In total, 155 subjects (72 males and 83 females, age range: y; BMI: kg/m 2 ) were evaluated. In total, 43 (26 m and 17 f; BMI: kg/m 2, mean7s.e.m.) of them were reassessed y later. MEASUREMENTS: Anthropometric and body composition (bioimpedance analysis) parameters and RMR (indirect calorimetry) were taken at baseline and after y. RESULTS: Subjects (15 m, 8 f) who gained body weight (arbitrarily defined as a change in body weight Z5 kg) had baseline BMI ( vs ; P ¼ NS) and body composition in terms of fat mass (FM%) and fat-free mass (FFM kg) comparable to those of the subjects (11 m, 9 f) whose body weight remained stable. Baseline RMR was significantly lower in subjects who gained weight than in those who did not ( vs kj/kg-ffm 24 h; Po0.001), although it did not differ significantly between the two groups (11972 vs kj/kg-ffm 24 h; P ¼ NS) y later. Baseline RMR was inversely correlated to both change in body weight (r ¼ 0.57; Po0.001) and FM (r ¼ 0.50; Po0.001). CONCLUSION: A low RMR normalized for FFM appears to be associated with body weight gain in the long run in adult Caucasian Italians. (2005) 29, doi: /sj.ijo Published online 11 January 2005 Keywords: resting metabolic rate; energy expenditure; body composition Introduction Obesity results from a chronic imbalance between energy intake and expenditure. The prospect that some individuals may be predisposed to excessive body weight gain on account of constitutionally low-energy expenditure is still controversial. Cross-sectional studies have shown that energy expenditure is higher in obese subjects than in nonobese ones. The resting metabolic rate (RMR) contributes to about 70 80% of total daily energy expenditure in adult sedentary subjects. When RMR is normalized for the size of fat-free mass (FFM), namely the metabolically active body compartment, no difference is observed between obese and nonobese subjects. 1,2 Compensatory mechanisms, first of all FFM expansion, may become active as body weight increases; their effect of increasing energy expenditure probably *Correspondence: Dr S Buscemi, Via D Cimarosa 5c, I Palermo, Italy. silbus@tin.it Received 13 January 2004; revised 13 October 2004; accepted 4 November 2004; published online 11 January 2005 contributes to limit further body weight gain and may play a confounding role in cross-sectional energy expenditure studies. There is thus general agreement that only longitudinal studies can disclose whether or not low-energy expenditure is associated with body weight gain. Although previous studies have noted this association in adult Pima Indians, 3 infants 4 and 5-y-old girls, 5 there have been no convincing data regarding adult Caucasians. 6 8 The present study consists in a longitudinal investigation as to (follow-up 8 12 y) whether low relative energy expenditure is associated with body weight gain in a group of adult Caucasian Italians. Methods Selection of subjects In total, 155 Caucasian Italian subjects (72 males and 83 females; age: y; BMI: kg/m 2 ) were recruited between 1987 and They included healthy volunteer subjects and obese outpatients seen at the Institute of Internal Medicine of the University of Palermo (I). On the

2 288 basis of clinical examination, routine blood tests and standard ECG, all subjects were normally glucose tolerant and otherwise healthy, apart from their obesity. Body composition, RMR, biochemical and hormonal blood values were measured. The same group was subsequently contacted for re-evaluation between 1997 and A total of 43 subjects (26 males, 17 females) were reinvestigated and completed the study; 41 subjects were excluded for various reasons (12 had undergone obesity surgery, two were extreme outliers, 26 had developed diabetes or hypertension or both, one had become anorexic); 67 subjects were not available (five had died, 22 refused to participate in the study, 34 could not be reached, six had emigrated). On both occasions, body weight had been stable for at least the last 3 months in all subjects. Evaluations were performed in the morning, after overnight fasting and resting. All subjects were asked to maintain their habitual physical activity and energy intake especially during the 3 days preceding measurements. Table 1 shows the physical characteristics of the subjects studied, when separated in greater or less than 5 kg body weight change. The ethical committee of the Institute approved the study protocol, and all subjects gave their written voluntary consent according to the Helsinki declaration. Table 1 Anthropometric characteristics, energy expenditure and respiratory quotient data on subjects divided into two groups according to body weight change Body weight change o5kg Z5kg P m/f 11/9 15/8 Baseline data Age NS Body weight (kg) NS Body mass index (kg/m 2 ) NS Fat mass (%) NS Fat-free mass (kg) NS Waist to hip ratio NS Resting metabolic rate kj/24 h NS kj/kg-ffm 24 h o0.001 Respiratory quotient (VCO 2 /VO 2 ) NS Final data Follow-up (y) NS Body mass index (kg/m 2 ) NS Waist to hip ratio NS D body weight (kg) o D fat mass (%) o D fat-free mass (kg) o Resting metabolic rate kj/24 h NS kj/kg-ffm 24 h NS D resting metabolic rate kj/24 h o kj/kg-ffm 24h o Respiratory quotient (VCO 2 /VO 2 ) NS FFM ¼ fat-free mass; NS ¼ not significant. Measurements A blood venous sample was obtained to determine chemical and hormonal data (radioimmunoassay for insulin and leptin); leptin concentrations were measured only at the second encounter (Human Leptin RIA KIT; Linco Res Inc., St Charles, MO, USA). Body composition was determined by means of the bioimpedance analysis method. Briefly, total body resistance and reactance were measured with an 800 ma, 50 khz, tetrapolar impedance plethysmograph (BIA 103, RJL, Detroit, USA/Akern, Florence); percentage of fat mass (FM%) and fatfree mass size (FFM kg) were obtained according to the factory predictive equations, as described elsewhere. 9 Repeated measurements in a group of 20 subjects with a wide range of BMI ( ) gave a BIA variability of 0.2% on the same day and of 0.5% on 2 consecutive days. Body circumferences were measured at the umbilicus (waist circumference) and at the most prominent buttock level (hip circumference), and their ratio (waist to hip ratio, WHR) was used as an indirect index of body-fat distribution. RMR and fasting respiratory quotient (RQ: an indirect measurement of the mixture of carbohydrate and lipid oxidation) were obtained by means of the indirect calorimetry method as described elsewhere. 10 At the first measurement session, a mouthpiece system of indirect calorimetry was employed (9000 IV, Gould electronics; The Netherlands). For technical reasons, at the second evaluation, the device was substituted with a new one utilizing a more evolved ventilated hood system (2900 MMC, Sensormedics; Yorba Linda, CA, USA/Bilthoven, The Netherlands). Both devices were equipped with an infra-red analyzer for carbon dioxide measurement (VCO 2 ) and a zirconium cell analyzer for oxygen measurement (VO 2 ). Analyzers were calibrated before each test by use of gases with known percentages of O 2 and CO 2. Briefly, when the mouthpiece system was employed, respiratory gas exchanges were measured three times at 20-min intervals for about 10 min each time. During each collection, the subjects, while resting in the supine position, breathed through a mouthpiece and wore a noseclip. All subjects were familiarized with the equipment before measurements were recorded. The previously measured CV on obese and nonobese subjects averaged 4.5 and 5.2% for RMR and RQ respectively. When the ventilated hood system was employed, respiratory gas exchanges were continuously measured for about 1 h, data were obtained from at least 30 min of stable measurements, and the average intrasubject variability was 3.9 and 4.8% for RMR and RQ, respectively. Comparative tests were performed with both devices in seven subjects. Results showed a coefficient of variation of 3.2% for RMR and of 4.6% for RQ. The RMR was expressed both in absolute terms (kj/24 h) and normalized for FFM size (kj/kg-ffm 24 h). Alternatively, due to the positive intercept in the relationship between RMR and FFM, a regression-based approach was considered in order to adjust RMR for FFM size. 11

3 Data analysis The subjects who completed the study were divided into two groups according to their body weight change. A bodyweight increase of less than 3 kg in 10 y is commonly considered physiological in adult subjects; 12 therefore, an arbitrary cutoff value of 5 kg was utilized to distinguish subjects who gained weight from those who did not. Results were also analyzed by subdividing subjects according to the tertile (from the lowest to the highest) of initial RMR normalized for FFM. Since age, sex and body composition are known to be major determinants of energy expenditure, the relationship between body weight change and absolute RMR was also explored by use of a multiple regression procedure. All data are expressed as mean7s.e.m. SPSS statistical software (release 6.0, SPSS Inc., Chicago, IL, USA) was utilized to explore differences between groups and relations between variables. Results Anthropometric characteristics, energy expenditure (expressed both in absolute value and as the RMR/FFM ratio) and RQ data of subjects who gained body weight and those who did not are reported in Table 1. The RMR, adjusted for FFM according to the Ravussin procedure, was significantly lower in the subjects who gained weight than in those who did not (respectively: vs kj/24 h; Po0.01); no significant difference between the two groups was found at the end of follow-up ( vs kj/24 h; P ¼ NS). Subjects were also divided according to the tertile of initial RMR normalized for FFM. Figure 1 reports absolute changes in the above-mentioned data. A significant inverse correlation (r ¼ 0.57; Po0.001) was observed between the initial RMR normalized for FFM (kj/kg- FFM 24 h) and the amount of body weight change. Similarly, the initial normalized RMR was inversely correlated to the fat mass (FM) change expressed both in percentage (r ¼ 0.50; Po0.001) and in absolute value (r ¼ 0.44; Po0.01). The change in body weight was also significantly correlated to the change in normalized RMR (r ¼ 0.59; Po0.001). Multiple regression analysis showed that the change in body weight was related to the initial RMR (Po0.001) independent of FFM according to the following relationship: Figure 1 Absolute changes (D) in body weight and FM of subjects grouped according to the tertile of initial RMR normalized for fat-free mass (kj/kg-ffm 24 h). Figure 2 Linear correlation between initial resting metabolic rate (RMR) residuals (measured minus predicted) and body weight change (r ¼ 0.39; Po0.01). body weight change (kg) ¼ RMR (kj/24 h) þ 0.39 FFM (kg); age, sex and FM were excluded. The residuals of initial RMR (measured minus predicted according to the Ravussin approach) were significantly and inversely correlated (r ¼ 0.39; Po0.01) to the change in body weight (Figure 2). The follow-up duration was significantly related only to the change in percentage of FM (r ¼ 0.34; Po0.03). Insulin and leptin levels did not differ significantly between groups. In particular, the average leptin blood levels were and mg/l, respectively, in the group that gained weight at the end of follow-up and the one that did not (P ¼ NS). Discussion The results of this study are in agreement with the hypothesis that a low relative RMR is associated with body weight gain in the long run, in adult Caucasian Italians. Although subjects whose body weight increased had initial physical characteristics comparable to those of subjects who did not, their RMR normalized for FFM was significantly lower. It increased at the end of follow-up, after body weight gain, until it became comparable to that of subjects whose body weight remained stable. Furthermore, the baseline normalized RMR was inversely and linearly correlated to the change in both body weight and FM. Similarly, there was a significant positive linear correlation between the change in normalized RMR and the change in body weight. Since RMR normalized for FFM increases in subjects who gain weight and decreases in those who lose it, we can conclude that some energy expenditure changes independently of body composition modifications. An interesting hypothesis is that a number of regulatory mechanisms influence energy expenditure to some extent, independently of FFM change, in order to preserve body weight, thus limiting its fluctuations. The ability of some people to increase energy 289

4 290 expenditure in response to an increased energy intake and vice versa is probably a genetic characteristic that might be mediated by different mechanisms such as sympathetic activity or b 3 receptors, uncoupling protein, brown adipose tissue activity or other systems Ravussin suggested normalizing RMR for FFM, taking into consideration the positive intercept in the linear regression between the two variables. 11 However, the same conclusions were reached, even when the procedure of normalizing RMR for FFM was followed. Interestingly, the residuals of initial RMR (measured minus predicted) were inversely correlated to the change in body weight (Figure 2). Another approach was to consider the tertiles of initial relative RMR. Similarly, at the end of follow-up, a higher increase in body weight and FM were observed in the lowest tertile of initial relative RMR. It is well known that body weight increases and RMR decreases over time, 19 thus the long follow-up period of this study might be a confounding factor. However, the length of follow-up was significantly correlated only to the change in percentage of FM, not in body weight, FFM or RMR. Furthermore, it was observed that relative RMR increased in subjects who gained body weight, while the time effect is in the sense of reduction. It thus seems plausible to exclude any time bias in the relationships between body weight, FFM and RMR. Other possible biasing influences due to methods for body composition and energy expenditure determination must be taken into consideration. Although BIA analysis is a reproducible method (both interobserver and intrasubject CV is less than 1%), it is not the gold standard for body composition determination. In particular, this method is known to underestimate the FM and to overestimate the FFM in obese subjects. 20 However, such an error, if present in this study, could have influenced the value of normalized RMR in the sense of a falsely lower increase at the end of follow-up in subjects with increased body weight, thus it would not lead to different conclusions but would instead reenforce them. Multiple regression analysis showed that body weight change was independently correlated to the initial FFM size. A possible explanation is that the BIA-derived FFM is probably preferentially influenced by the limb lean body mass and to a lesser extent by the trunk lean body mass (BIA assumes that the body is a cylinder; measurements are obtained by positioning the electrodes distally at the upper and lower extremities, and limb lengths are significantly greater than trunk lenghts). It has recently been shown that the distribution pattern of lean body mass influences RMR and could explain, at least in part, racial differences in RMR. 21 In fact, trunk lean body mass, reflecting a higher organ mass, has a higher energy expenditure than limb lean body mass. However, since, in this study, no measurement of lean body mass distribution was made, the potential role it plays in RMR and body weight change remains speculative. Indirect calorimetry has an intrasubject CV of about 3%; another CV of 3.2% is due to the fact that the first measurement was made with equipment that employed the mouthpiece respiratory system, while at the second examination, the canopy system was utilized. At worst, if the 2 CV are summed, it is possible to hypothesize an error of about 6% in RMR measurement. However, the group of subjects who gained body weight increased its RMR, on average, by about 30% in absolute value and 10.5% when normalized for FFM, values that are well above such a possibility of error. Subjects who gained weight had, on average, an initial RMR of about 750 kj/24 h lower than subjects whose body weight remained stable. If we assume that about kj are required to increase body weight by 1 kg (made up of 70% fat and 30% fat-free mass) and that the corresponding RMR rise is about 121 kj/24 h for each kg of FFM expansion (based on the present study results), we can calculate that an average difference in RMR of 750 kj/24 h accounts for a body weight increase of about 14 kg in 8 y. This is much greater than what was observed in the study (about 9 kg in 10 y). A possible explanation for such a difference is that, similarly to energy expenditure, even energy intake is regulated in order to try to maintain body weight stability. Therefore, subjects whose body weight increased may have reduced their level of habitual energy intake due to neuroendocrine adaptations. However, this hypothesis remains speculative inasmuch as this study does not provide data that support it. The present study is not in agreement with a previous report by Seidell et al 6 that did not evidence an association between low RMR and future body weight gain in Caucasian subjects. However, different issues in the above-mentioned study might explain the reasons for such negative results. First, in the study by Seidell et al, the out-of-date Haldane system was initially employed to perform indirect calorimetry. Furthermore, body composition was obtained by means of plicometry, a method with very low reproducibility. The group studied was not homogeneous, particularly for age range (18 98 y), and subjects were examined after they had spent more than 2 days in a confined environment. These factors might explain the discrepancies between our study and the Seidell results. Similarly, the wide range of age (18 68 y), the shorter follow-up period (5.5 y) and the use of plicometry might explain the discrepancies between our data and those reported by Katzmarzyk et al. 7 The RQ is an index of the mixture of whole-body carbohydrate and lipid oxidized. The value of RQ generally varies within the range , implying that at 0.70 lipids oxidation is 100% and that only carbohydrates are oxidized at Thus, the higher the RQ, the lower the lipid oxidation. Previous studies have shown that in Pima Indians, a high RQ is a risk factor for future body weight gain as well. 22,23 Our study failed to show this an association. However, here, only the postabsorptive RQ was obtained, while in previous studies 24 h-rq was measured. Furthermore, the coefficient of variation of RQ is higher than that of RMR, so it cannot be excluded that the number of subjects enrolled in the study was too small to evidence such a characteristic. Further evidence for the need of a larger

5 subject group to detect differences in RQ is that the CV for RQ between the two systems of indirect calorimetry utilized here (canopy and mouthpiece) was well above (4.6%) that for RMR. Obesity is a multifactorial disorder in which genetic and environmental factors play a role. Measuring RMR could be a useful tool in order to select people who are at higher risk of becoming obese, as they could benefit from greater research and better targeted therapeutic intervention. Acknowledgements We express their gratitude to professor Yves Schutz, University of Lausanne (CH) for his critical comments during the revision of this manuscript and to Nancy W Birch, BA who edited the English. This study was supported in part by the Italian Ministry of Education (MURST funds 60% project 2001). References 1 Ravussin E, Burnand B, Schutz Y, Jequier E. Twenty-four hour energy expenditure and resting metabolic rate in obese, moderately obese and control subjects. Am J Clin Nutr 1982; 35: Verga S, Buscemi S, Caimi G. Resting energy expenditure and body composition in morbidly obese, obese and control subjects. Acta Diabetol 1994; 31: 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: Roberts SB, Savage J, Coward WA, Chew B, Lucas A. Energy expenditure and intake in infants born to lean and overweight mothers. N Engl J Med 1988; 318: Griffiths M, Payne PR, Stunkard AJ, Rivers JP, Coc M. Metabolic rate and physical development in children at risk of obesity. Lancet 1990; 336: Seidell JC, Muller DC, Sorkin JD, Andres R. Fasting respiratory exchange ratio and resting metabolic rate as predictors of weight gain: the Baltimore Longitudinal Study on aging. Int J Obesity Relat Metab Disord 1992; 16: Katzmarzyk PT, Perusse L, Tremblay A, Bouchard C. No association between resting metabolic rate or respiratory exchange ratio and subsequent changes in body mass and fatness: 5.5 year follow-up of the Québec family study. Eur J Clin Nutr 2000; 54: Weinsier RL, Nelson KM, Hensrud DD, Darnell BE, Hunter GR, Schutz Y. Metabolic predictors of obesity. Contribution of energy expenditure, thermic effect of food, and fuel utilization to four year weight gain of post-obese and never-obese women. J Clin Invest 1995; 95: Khaled MA, McCutcheon MJ, Reddy S, Pearman PL, Hunter GR, Weinsier RL. Electrical impedance in assessing human body composition. J Appl Physiol 1986; 60: Jéquier E, Acheson K, Schutz Y. Assessment of energy expenditure and fuel utilization in man. Ann Rev Nutr 1987; 7: Ravussin E, Bogardus C. Relationship of genetics, age, and physical fitness to daily energy expenditure and fuel utilization. Am J Clin Nutr 1989; 49: Rossner S. Defining success in obesity management. Int J Obes Relat Metab Disord 1997; 21s1: S2 S4. 13 Weigle DS, Kuijper JL. Obesity genes and the regulation of body fat content. Bioessays 1996; 18: Jequiér E, Tappy L. Regulation of body weight in humans. Physiol Rev 1999; 79: York D, Bouchard C. How obesity develops: insights from the new biology. Endocrine 2000; 13: Kimm SY, Glynn NW, Aston CE, Damcott CM, Poehlman ET, Daniels SR, Ferrell RE. Racial differences in the relation between uncoupling protein genes and resting energy expenditure. Am J Clin Nutr 2002; 75: Arner P, Hoffstedt J. Adrenoceptor genes in human obesity. J Intern Med 1999; 245: Clement K, Vaisse C, Manning BS, Basdevant A, Guy-Grand B, Ruiz J, Silver KD, Shuldiner AR, Froguel P, Strosberg AD. Genetic variation in the beta 3-adrenergic receptor and an increased capacity to gain weight in patients with morbid obesity. N Engl J Med 1995; 333: Vaughan L, Zurlo F, Ravussin E. Aging and energy expenditure. Am J Clin Nutr 1991; 53: Hogdon JA, Fitzgerald PI. Validity of impedance prediction at various levels of fatness. Hum Biol 1987; 59: Byrne NM, Weinsier RL, Hunter GR, Desmond R, Patterson MA, Darnell BE, Zuckerman PA. Influence of distribution of lean body mass on resting metabolic rate after weight loss and weight regain: comparison of responses in white and black women. Am J Clin Nutr 2003; 77: Zurlo F, Lillioja S, Del Puente A, Nyomba BL, Raz I, Saad MF, Swinburn BA, Knowler WC, Bogardus C, Ravussin E. Low ratio of fat to carbohydrate oxidation as predictor of weight gain: study of 24 h RQ. Am J Physiol 1990; 259: E650 E Swinburn B, Ravussin E. Energy balance or fat balance. Am J Clin Nutr 1993; 57: 766s 771s. 291