Treatment of obesity is difficult, and the success rate in the medium-and long-term is discouraging in children as well as in adults (1,2). Different strategies of therapy have been proposed, but energy-intake restriction combined with a program of exercise is the most common treatment in clinical practice. Previous research has been directed to finding the best diet to achieve weight loss, while the role of exercise has received less attention. In particular, the role of exercise in inducing and maintaining weight loss in obese children has to be defined further. Estimations of the level of activity of obese children and of the rate of total energy expenditure (TEE) they devote to physical activity every day are necessary prerequisites to approaching this problem.
Scanty data are available on the level of activity and the energy expenditure for physical activity of obese children. The purpose of the present study was to estimate the TEEs and the activity patterns in a group of 32 obese and nonobese 8-10-year-old children.
MATERIALS AND METHODS
Investigations were carried out in 32 prepubertal children divided into two groups of 16 obese and 16 nonobese children. Obesity was defined as a body mass index (BMI) >97th percentile of the reference values for age and sex. Control children were selected based on BMIs ≤97th percentile of the reference values for age and sex (3). Medical history, physical examination, and routine laboratory tests, including urine standard test, fasting plasma glucose, total cholesterol, Thyroid Stimulating Hormone (TSH) (n = 1), and oral glucose tolerance test (n = 2), were performed and reasonably excluded health problems other than obesity. Pubertal stage was assessed according to Tanner (4). None of the subjects reported significant changes in body weight during the month preceding the study. No child was taking any drugs.
Anthropometric measurements (weight, height, and skin-fold thicknesses) were carried out by the same investigator. Height was measured to the nearest 0.5 cm on a standardized, wall-mounted height board. Weight was determined to the nearest 0.1 kg by standard beam scale with the child dressed only in light underwear. BMI was calculated as weight (kg) divided by height squared (m2). Skin-fold thicknesses were measured in triplicate to the nearest millimeter at the triceps and subscapular sites by means of an Harpenden skin-fold calliper (CMS Weighing Equipment Ltd., London, U.K.). The Lohman formulae were used to calculate the percentage of body fat (FM%) from the sum of the two skin-folds (5). Fat mass (FM) was obtained by multiplying the percentage of body fat by body weight. Fat-free mass (FFM) was obtained by subtracting body fat from body weight.
Prior to the postabsorptive metabolic rate measurement, children consumed an unrestricted diet. The day before the test they did not perform any intense physical exercise. The children arrived by car at the Department of Paediatrics (University of Verona Polyclinic, Verona, Italy) at 7:30 a.m.; their last meal had been taken at 8 p.m. the day before. After 30 min of rest on a hospital bed in a comfortable temperature-controlled environment, continuous respiratory exchange was initiated by indirect calorimetry. During the 30-min measurement period, children rested quietly watching cartoons. Throughout the postabsorptive metabolic rate (PMR) measurement, a technician observed the child to verify that he or she was motionless and relaxed. After PMR measurement, a light meal was taken, and ≅2 h later the child took a treadmill test. Prior to the treadmill test, the children were instructed about the protocol and carefully familiarized with the experimental apparatus; in particular, they were taught about breathing through the mouthpiece while running on the treadmill.
Postabsorptive Metabolic Rate
The PMR was measured by respiratory gas exchange for 30 min as previously described (6). An open-circuit computerized indirect calorimeter (Deltatrac TM, Datex Inc., Finland) connected to a transparent hood system was used. The instrument was calibrated before each test with a gas mixture (O2 95.2% and CO2 4.8%). Oxygen consumption and carbon dioxide production values were printed out at 1-min intervals. The mean of the final 20 min was used to evaluate PMR. The energy expenditure was derived from oxygen uptake (VO2) and carbon dioxide output (VCO2) using the formula of Lusk (7).
Heart Rate-Based Energy Expenditure Calculation
The relationship between heart rate (HR) and VO2 was established in each child, as previously described (8). Briefly, to determine the individual HR-VO2 regression line, the subject performed an intermittent and incremental physical exercise test on a treadmill (9). VO2 and HR were simultaneously measured under standardized conditions. VO2 and VCO2 were measured by a standard open-circuit method. The expired air was collected in a Douglas bag by a two-way nonrebreathing valve and nose clip and was analyzed for volume (Gas Meter MCG, SIM, Rome, Italy) and for gas composition (Oxinos 100 and Binos C gas analyzers, Leybold-Heraeus Hanau, Germany). HR was measured by a small portable HR cardiotachometer (Polar Sport Tester, Polar Electro, Finland), which consisted of an electrode-belt transmitter and a wrist microcomputer receiver. Resting VO2 and HR were obtained when the subject was lying, sitting, and standing. The resting energy expenditure (REE) was calculated as the mean of the VO2 for three resting activities. At least six calibration points for the nonresting activities were taken during walking and running on the treadmill (PV Rolling Belt, Beta, Milan, Italy). Starting at 2 km/h, the speed was increased by steps of 1 km/h until the maximal work load was reached or a heart rate of 200 beats/min was obtained (peak oxygen uptake or peak VO2). Measurements of VO2 and HR were made during the final 3 min of each walking or running period. Between 6 and 10 min of recovery was allowed between two consecutive steps. VO2 was converted into energy expenditure by means of the simplified Weir formula, which assigns a value of 20.5 kJ to each liter of oxygen consumed (10).
Critical heart rate (flex-HR)—above which the calibration curve was used to estimate the energy expenditure corresponding to HR and below which the REE value was used to determine the sedentary energy expenditure—was identified for each child (11). Flex-HR was calculated as the mean of the highest HR for the resting activities (lying, sitting, and standing) and the lowest HR during the lightest imposed exercise.
The HR monitoring technique was previously validated using the doubly labeled water (2H2-18O) method (11) and indirect chamber calorimetry (12) as references.
Heart Rate Monitoring
The HR was recorded continuously in free-living conditions for 3 weekdays during normal daily activities. The HR transmitter—directly integrated into the electrode belt—was secured on the chest by means of an elastic band fixed with sticking plasters to the skin to reduce the possibility of detachment. Parents were instructed to fix the band with the electrodes and the HR transmitter on the chest and to turn the recorder on and off. The times at which the recorder was turned on and off were automatically recorded. Pulse was recorded at 1-min intervals up to 16 h. Stored data were retrieved daily, and the memory was reprogrammed at the subject's home via an interface unit and personal computer. An additional program was designed by one of us (M.Z.) to compute TEE from HR. The monitoring started when the child woke up and continued until bedtime, when the parent and/or child removed the instrument.
TEE was calculated by adding the sleeping energy expenditure, the sedentary (resting) energy expenditure, and the activity energy expenditure (EE-Act) (9). Sleeping energy expenditure (SEE) was assessed by multiplying the sleeping time in minutes by the PMR in kilojoules per minute. Sleeping time was assessed by recording the time between going to bed and waking up, assessed by the children and/ or parents in a notebook. In the same notebook, the times of beginning and ending the main physical activities of the day, with their nature and duration, were also recorded. Sedentary energy expenditure was calculated by multiplying the nonsleeping time with flex-HR during the day by REE. Sedentary activities consisted of sitting and standing activities (writing, reading, studying, playing quietly, watching TV, or sitting in the car, in the bus, at the cinema, in the church, etc.). Activity energy expenditure was calculated by applying each individual calibration line to HR in excess of flex-HR.
The results are expressed as means ± SD. A Student unpaired t test was used to compare obese and nonobese children. In those instances in which the sample was not normally distributed, the Mann-Whitney test was used. The Systat statistical package, version 5.0 (SPSS Inc., Chicago, IL, U.S.A.), was used to analyze the data.
Informed consent was obtained from the parents of all subjects. The protocol was approved by the Ethics Committee of the University Hospital of Verona (Verona, Italy).
We were not able to obtain complete HR recordings for one boy and one girl. Another girl was sick after the first day of recording, and it was not possible to obtain her HR monitoring after convalescence. Therefore, the data of 29 children were evaluated.
Age and height were not different in the two groups. Weight, BMI, relative weight percentage, FM, and FFM were significantly higher (p < 0.01) in the obese than in the nonobese children (Table 1). All the children were in the prepubertal phase, but two obese and two nonobese children had a Tanoer pubertal stage of II.
Mean resting HR, flex-HR, and daytime HR were not different in obese and nonobese children (Table 2). The HR values at 50% peak VO2 and at 70% peak VO2 were not different in the two groups of children.
Duration of sleeping was not different in obese and nonobese children (Table 3), while time spent in sedentary activity (HR below flex-HR) was significantly higher (p < 0.05) in the obese group. In contrast, time spent at HR higher than flex-HR was significantly lower in the obese than in the nonobese children. Time spent in moderate or vigorous activity—i.e., time spent with HR at 50-70% of peak VO2 and with HR >70% of peak VO2, respectively—was not different in the two groups (Table 3).
Both the TEE and each of its components (SEE, REE, and EE-Act) were significantly higher in obese than in nonobese children (Fig. 1). Obese children had higher PMRs (5.49 ± 0.66 vs. 4.65 ± 0.54 MJ/day; p < 0.001) and nonpostprandial energy expenditures (TEE - PMR; 3.98 ± 1.30 vs. 2.94 ± 1.39 MJ/day; p < 0.01)—i.e., the energy expenditure for activity plus thermogenesis—than did the nonobese children. The proportion of TEE due to physical activity (plus thermogenesis) was not statistically different in the two groups, as revealed by the TEE/PMR ratio (obese vs. nonobese children, 1.72 ± 0.25 vs. 1.62 ± 0.28; p = NS).
The spontaneous physical activities most frequently performed by the children were walking and cycling. Programmed physical activities reported by the boys included football (two obese and three nonobese children), tennis (one obese child), judo (one obese and one nonobese child), and swimming (two obese and three nonobese children). In girls the activities included dancing (three nonobese children), volleyball (one obese and two nonobese children), and swimming (two obese and three nonobese children).
The results of this study show that TEE is greater in obese than in nonobese children. The proportion of TEE due to PMR was ≅60%, and PMR was able to explain ≅50% of the variance of TEE (R2 = 0.482; p < 0.01). The nonpostabsorptive metabolic rate, which is the sum of thermogenesis and the energy expenditure for activity, was slightly but significantly greater in the obese children. Therefore, obese children spent more energy for activity than nonobese children did. In addition, a gross index of activity, the TEE/PMR ratio, was not statistically different in the two groups, and it was comparable with that extrapolated by the U.S.A. Recommended Dietary Allowances for 8-10-year-old children (13). These results confirm data reported in obese and nonobese adolescents in whom TEE was measured by means of the doubly labeled water (2H218O) technique (14).
The evaluation of daily HR patterns of obese and nonobese children offers the opportunity to get an objective estimation of the activity pattern of these children. Overweight children spent significantly more time at lower HRs (below flex-HR—i.e., performing sedentary activities) and less time at higher HRs (above flex-HR—i.e., performing nonsedentary activities) than nonobese children. In other words, obese children were more sedentary than nonobese children. It is not possible to know from the results of the present cross-sectional study whether the lower time devoted to physical activity in obese children was a consequence of excess weight or was a characteristic present before the onset of obesity. In adult Pima Indians, low spontaneous activity observed in the confinement of a respiration chamber was associated with subsequent weight gain (15). In free-living conditions, this association remains to be investigated but may be stronger.
Surprisingly, the time spent at HRs higher than the HR at 50% of peak VO2 (≅1.5 h/day) was not statistically different in the two groups. In spite of their greater sedentariness, the energy expenditure devoted to physical activity (plus thermogenesis)—i.e., TEE — PMR—in the obese children was significantly greater than that of the nonobese children. The higher energy expenditure necessary to move the heavier body mass during weight-bearing activities (16) and the greater PMR of obese children (17) help explain the apparent paradox of a shorter time devoted to activity yet a higher absolute activity-induced energy expenditure in these children. The higher PMR of obese children is consistent with the significantly higher FFM found in these children than in nonobese children. The higher PMR explains the significantly higher SEE found in obese children, even though the time spent sleeping was the same in both groups. Similarly, the higher PMR and the greater time spent in sedentary activities explain the significantly higher REE found in obese children.
From the results of this study, a “threshold level” might be ideally hypothesized after which the increase in PMR and EE-Act following body mass gain cannot compensate for the reduction in energy expenditure due to the lower spontaneous activity induced by overweight itself. In particular, excess weight makes body movements (in particular weight-bearing activities) progressively more difficult and promote a self-reduction in activity by the obese individual (18). After the threshold level is reached, obesity is more difficult to treat. Certainly, several other mechanisms can interact to favor weight gain, such as energy intake, the proportion, quantity, and quality of macronutrient intake, postprandial thermogenesis, protein turnover, thyroid hormone activity, and body weight composition and distribution (19).
From a therapeutic point of view, the results of this study might lead to some interesting suggestions. In particular, we might suppose that a progressive increase of the level of activity of obese children—for instance, to reach the same ratio between resting time and activity time showed by nonobese children—will significantly increase the TEE. In this study, the difference in resting time between obese and nonobese children was 105 min/day (400 - 295 min/day). Since the EE-Act of obese children was 11.9 kJ/min and the REE of obese children was 4.6 kJ/min, the net energy expenditure for activity was 7.3 kJ/min (11.9 - 4.6 kJ/min). Multiplying the difference in time spent on sedentary activity between obese and nonobese children by the net energy cost of nonsedentary activity (105 min × 7.3 kJ/min), a total of 766.5 kJ/day, which represent 8% of TEE (766.5/9,460 kJ/day), is found. Therefore, a change in the lives of the obese children by their reducing inactivity and performing long-duration and low-intensity physical exercises, such as walking, might significantly contribute to weight control. Continuous encouragement to a more active lifestyle, together with a reduction of the fat/complex carbohydrate intake ratio, should be generally advised to the entire pediatric population for preventive purposes. In children with a BMI >95th percentile, a specific medical treatment should be planned (20). In contrast, in younger children or in children with modest excess weight (<50% of weight predicted for height), encouragement to maintain body weight rather than to promote weight loss seems to be preferable, since the natural growth in height occurring at this age will progressively decrease excess weight. A more structured intervention procedure should be planned for heavy adolescents or fatter children (excess weight, ≥50%), who have an higher risk of being obese, in adulthood (21).
In conclusion, the obese children expend more energy per day (≅26%) than the nonobese children. This is mainly due to the significantly higher PMR of obese children, explained principally by their larger FFM—i.e., metabolically active tissue. The index of activity (TEE/PMR) was comparable in obese and nonobese children although the time spent in activity was shorter in obese than in nonobese children. The higher absolute energy cost of activity in obese children explains this apparent discrepancy. A progressive increase in the time devoted to physical activity may constitute a helpful strategy for increasing total energy expenditure and weight loss in obese prepubertal children.
Acknowledgment: This study was supported by the National Research Council (NRC) Target Project “Prevention and Control of Disease Factors,” subproject no. 7, contract no. 94.00.164PF41.
1. Epstein LH, Valoski A, Wing RR, McCurley J. Ten-year follow-up of behavioural, family-based treatment for obese children. JAMA
2. Bennet W. Dietary treatments of obesity. Ann NY Acad Sci
3. Rolland-Cacherà MF, Sempè M, Guillaud-Bataille M, et al. Adiposity indices in children. Am J Clin Nutr
4. Tanner JM. Growth at adolescence
. 2nd ed. Oxford: Black-well, 1962.
5. Lohman TG. Applicability of body composition techniques and constants for children and youth. Exerc Sport Sci Rev
6. Maffeis C. Schutz Y, Pinelli L. Effect of weight loss on resting energy expenditure in obese prepubertal children. Int J Obes
7. Lusk G. The elements of science of nutrition
. 4th ed. Philadelphia: Saunders, 1928.
8. Bradfield R. A technique for determination of usual daily energy expenditure in the field. Am J Clin Nutr
9. Maffeis C, Schutz Y, Zaffanello M, Piccoli R, Pinelli L. Elevated energy expenditure and reduced energy intake in obese prepubertal children: paradox of poor dietary reliability in obesity? J Pediatr
10. Weir BDV. New method for calculating metabolic rate with special references to protein metabolism. J Physiol
11. Livingstone MB, Prentice AM, Coward WA, et al. Simultaneous measurement of free-living energy expenditure by the doubly-labelled water method and heart-rate monitoring. Am J Clin Nutr
12. Spurr GB, Prentice Am, Murgatroyd PR, et al. Energy expenditure from minute-by-minute heart-rate recording comparison with indirect calorimetry. Am J Clin Nutr
13. Subcommittee on the Tenth Edition of the RDAs, Food and Nutrition Board, Commission on Life Sciences, National Research Council. Recommended dietary allowances
. 10th ed. Washington, DC: National Academy Press, 1989:33.
14. Bandini L, Schoeller DA, Dietz WH. Energy expenditure in obese and nonobese adolescents. Pediatr Res
15. Zurlo F, Ferraro RT, Fontvieillie AM, Rising R, Bogardus C. Ravussin E. Spontaneous physical activity and obesity: cross sectional and longitudinal studies in Pima Indians. Am J Physiol
16. Maffeis C, Schutz Y, Schena F, Zaffanello M, Pinelli L. Energy expenditure during walking and running in obese and nonobese prepubertal children. J Pediatr
17. Brunch H. Obesity in childhood. II. Basal metabolism and serum cholesterol of obese children. Am J Dis Child
18. Ferraro R, Boyce VL, Swinburn B, De Gregorio M, Ravussin E. Energy cost of physical activity on a metabolic ward in relationship to obesity. Am J Clin Nutr
19. Ravussin E, Swinburn BA. Pathophysiology of obesity. Lancet
20. Himes JH, Dietz WH. Guidelines for overweight in adolescent preventive services: recommendations from an expert committee. Am J Clin Nutr
21. Mossberg HO. Forty year follow-up of overweight children. Lancet