Association between Short Sleeping Hours and Physical Activity in Boys Playing Ice Hockey

Association between Short Sleeping Hours and Physical Activity in Boys Playing Ice Hockey URS EIHOLZER, MD, UDO MEINHARDT, MD, VALENTIN ROUSSON, PHD, RENATO PETRÒ, MICHAEL SCHLUMPF, GERHARD FUSCH, PHD, CHRISTOPH FUSCH, MD, THEO GASSER, MD, AND FELIX GUTZWILLER, MD

Objectives To determine physical activity in healthy boys and how physical activity relates to training and daily awake hours. Study design In 66 boys (5 to 15 years) affiliated with an ice-hockey club, we measured total daily energy expenditure (TDEE, doubly-labeled water) and basal metabolic rate (ventilated-hood method). Physical activity energy expenditure for the whole day (DAEE), during training, and during spontaneous physical activity was measured by accelerometry and activity protocols. Univariate (UA) and multivariate (MA) correlation analysis was applied. Results Physical activity level, DAEE, and TDEE for prepubertal (2.0 and 2.2 Mcal/d) and pubertal (bone age >13 years; 1.8 and 2.8 Mcal/d) boys were matched to literature data from normal boys of equal age. In prepubertal boys DAEE correlated positively with awake hours (rUA ⴝ 0.55, rMA ⴝ 0.39, P < .01). In pubertal boys this correlation was not significant, the slopes between the 2 groups being significantly different (P ⴝ .025). In prepubertal boys spontaneous physical activity expenditure correlated significantly positively with training activity expenditure (rUA ⴝ 0.72, rMA ⴝ 0.52, P < .001). Conclusion Contrary to findings in adults, where short sleepers had lower physical activity and intensive training was negatively compensated reducing spontaneous physical activity, in physically active prepubertal boys, total daily and spontaneous physical activity relate positively to awake hours and training; suggesting child-specific control of physical activity. (J Pediatr 2008;153:640-5)

hysical activity energy expenditure (PAEE) plays a central role in the regulation of energy balance. For an even energy balance, PAEE plus basal metabolic rate (BMR) has to equal total energy intake (nutrition). Irrespective of age, a reduction in physical activity and the related reduction in energy consumption is associated with a great number of negative health consequences and an increased number of obese people.1-3 Over the last decades there is a continuing reduction in physical activity and physical fitness of children and adolescents.4,5 We consider this trend to be a social phenomenon; however, little is known about the intrinsic control of physical activity in children, and successful efforts to improve levels of physical activity in the population are contingent on an accurate understanding of how physical activity is controlled. This may lead up to new strategies for childhood obesity prevention. Energy regulation is a central feature to secure survival of the species and the individual. As such it is centrally regulated by systems with great redundancy.6 The following model of how physical activity may be controlled is not new:7 Intrinsic and extrinsic From the Center for Pediatric Endocrinology (PEZZ) (U.E., U.M., R.P., M.S.), and the factors do interact to control physical activity. Intrinsic regulation of physical activity is believed Department of Biostatistics, Institute for Soto be located in the hypothalamus, being tightly linked with the control of hunger and satiety.6 cial and Preventive Medicine, University of Zurich (V.R., T.G., F.G.), Zurich, Switzerland, Play instinct in children reflects the intrinsic drive of physical activity. Extrinsic factors and the Department of Neonatology, Ernstmainly depend on the social environment such as family, school, and workplace. For Moritz-Arndt-University, University Chilexample, an increasing workload at school or limited space because of urbanization can act dren’s Hospital Greifswald (G.F., C.F.), Greifswald, Germany. as negative extrinsic factors.8 The authors declare no potential conflicts The profound reluctance toward physical activity typically seen in children with of interest, real or perceived. Prader-Willi syndrome (PWS) suggests this syndrome to be a model of perturbed Submitted for publication Jul 5, 2007; last revision received Apr 11, 2008; accepted intrinsic control of physical activity.9 Our previous work in these children exemplifies

P

BMR DAEE PAEE PWS SpAEE TDEE

640

Basal metabolic rate Daily activity energy expenditure Physical activity energy expenditure Prader-Willi Syndrome Spontaneous activity energy expenditure Total daily energy expenditure

TDEEacc TDEEDLW TrAEE

Total daily energy expenditure measured by combining accelerometry and calorimetry Total daily energy expenditure measured by the doubly labeled water method Training activity energy expenditure

May 8, 2008. Reprint requests: Urs Eiholzer, MD, Center for Pediatric Endocrinology Zurich (PEZZ), Möhrlistrasse 69, CH-8006 Zurich, Switzerland. E-mail: [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.05.015

successful modulation of extrinsic factors: Specifically, a daily high-intensity training of just 4 minutes induced a significant increase in spontaneous daily walking distance from 3.7 to 8 km.10 The regulatory interplay of factors controlling physical activity may be different in children compared with adults. Contrary to what we described in children with PWS, in adults exercise interventions (in particular high-intensity exercise) do not affect spontaneous physical activity.11,12 In obese adults induced exercise was even shown to be negatively compensated by a reduction of spontaneous physical activity outside of training hours.13 There is growing evidence that sleep duration impacts energy balance.14 Reduction in daily sleeping hours was identified as a risk factor for obesity both in children and adults.15,16 This may be explained by specific endocrine factors regulating appetite.17 However, sleep duration or, inversely, daily awake hours could affect not only appetite but also regulation of physical activity and therefore exert parallel effects on energy balance. To further explore the hypothesis that decreased daily awake hours and increased training promote physical activity in children, we studied the relationship between training, daily awake hours, and total daily physical activity in a group of healthy pubertal and prepubertal boys.

METHODS Subjects We approached 72 boys aged between 5 and 15 years (mean age 10.5 years) affiliated with a local ice-hockey club (GCK Lions). Only boys younger than 15 years were included to minimize heterogeneity of extrinsic factors influencing physical activity, 15 years being the age when adolescents are entering high school or starting an apprenticeship. Of the 72 boys participating, 6 were excluded because of invalid activity measurements. We thus obtained activity data for 66 boys. Another 7 boys did not train during the measurement’s period, reducing the sample to 59 for the analyses involving measurements of training. Protocol During 4 randomly assigned consecutive days (96 hours, Monday to Sunday) physical activity was objectively measured by a body-fixed triaxial accelerometer (RT3; Stayhealthy Inc., Monrovia, California) in combination with an activity protocol. All boys and their parents were instructed on an individual basis by a single person (R.P.) in the Center for Pediatric Endocrinology Zurich (PEZZ), Switzerland, before starting measurements. The accelerometer was worn at all times, except during sleeping hours and when taking a bath or a shower. On the activity protocol the exact time of start and end of every physical activity (training session, school-sport lesson, period of free play), as well as sleeping hours were noted. The number of sleeping hours recorded on the protocol corresponded with the accelerometer readings except when the accelerometer was taken off while sleeping.

The RT3 consists of 3 uniaxial piezoresistive accelerometers, and it registers acceleration in 3 orthogonal directions. PAEE is derived from the magnitude of the vectors of the 3 axes (x-, y-, and z-axis) using the RT3 software package. PAEE does not include BMR. The accelerometer’s measurements of PAEE has been validated against the indirect calorimetry method in children aged 8 to 18 years.18,19 Total daily energy expenditure (TDEE) was measured over a period of 14 days with the doubly labeled water (DLW) method. Each boy was given an oral preweighed DLW dose (0.5 g H2 18O and 0.3 g 2H2O/kg body weight). The parent received labeled sampling containers and detailed instructions to collect timed urine samples. The specimens were taken before administration of the dose, in the morning after dosing, and for the following 14 days. Samples were frozen until assayed in triplicate for 2H and 18O by isotope-ratio mass spectrometry with a PDZ Europa ANCA 20-20 (University of Greifswald, Greifswald, Germany).20 The multipoint approach to derive TDEE was used with an assumed food quotient of 0.845.21 The data from urine samples collected in the first morning were not included routinely because analysis indicates that the isotopes are not fully equilibrated in urine at this time. BMR was measured with an open-circuit, ventilatedhood indirect calorimeter (Deltatrac II; Datex-Engstrom, Helsinki, Finland). Each subject fasted overnight for a minimum of 12 hours and engaged in minimal activity before the determination of metabolic rate. A 30-minute rest and a 5-minute equilibration period proceeded the 30-minute measurement period. Weir’s equation was used to calculate BMR.22 To validate measurements of PAEE, we directly compared TDEE measured by the DLW method (TDEEDLW, gold standard) and total daily energy expenditure (TDEEacc) derived from the combined accelerometric and calorimetric data (TDEEacc ⫽ (PAEE/0.9) ⫹ BMR). Physical activity level was calculated on the basis of the above measures of TDEEDLW and BMR: physical activity level ⫽ TDEE/ BMR.23 Biologic age was assessed on the basis of skeletal maturity before activity measurements were started. Skeletal age (bone age) was read by a single investigator (U.E.) using the method of Greulich and Pyle.24 For boys a skeletal age of 13 years and more refers to pubertal or postpubertal developmental status, allowing classification into a prepubertal and pubertal group without any clinical examination. A single investigator (R.P.) measured height with a Harpender stadiometer (accuracy of measurement, 1 mm). For weight measurements we used an electronic balance (SECA708; SECA, Hamburg, Germany; d ⫽ 0.1 kg; coefficient of variation, 0.055%).

Data Modeling and Data Analysis On the basis of information from the activity protocol total daily activity energy expenditure (DAEE) was separated into spontaneous activity energy expenditure (SpAEE) and

Association between Short Sleeping Hours and Physical Activity in Boys Playing Ice Hockey

641

training activity energy expenditure (TrAEE); each was then expressed as amount (kcal/measured period) and intensity (kcal/min) of PAEE. Amount of DAEE reflected the total PAEE recorded over the measurement’s period. Intensity of DAEE was derived by dividing the amount of DAEE by the individually recorded total awake time. Intensity of SpAEE was calculated by dividing the measured amount of SpAEE by the time when boys were not training or sleeping. To correct for interindividual variation in recorded non-training time, the amount of SpAEE was calculated as the sum of the PAEE recorded outside of training plus the intensity of SpAEE (as defined above) multiplied by the recorded minutes of training (Table I; available at www.jpeds.com). Intensity of TrAEE was calculated by dividing the recorded amount of TrAEE by the time when boys were actually training. An adjustment of the amount of TrAEE was necessary because of the interindividual variations in number and duration of training sessions included in the measurement period, which did not necessarily reflect the real differences in amount of training. For example 2 boys, 1 having a 1-hour training session once a week on a Monday, the other having training sessions of 90 minutes duration 3 times per week on Monday, Thursday, and Saturday, both being measured from Monday to Wednesday, the recorded measurement period will not reflect their usual number and duration of training sessions. Therefore the amount of TrAEE was derived by multiplying the mean duration for any 96-hour time period during the week for that individual with the intensity of TrAEE as defined above (Table I). Descriptive statistics are presented as mean ⫾ SD. Activity measurements were log transformed to achieve nearnormal distributions. To study the relationships between the various activity measurements and awake hours, Pearson correlations (referred to as univariate analyses), and partial Pearson correlations adjusted for the effect of biologic age (referred to as multivariate analyses) have been calculated. These analyses have been carried out separately for prepubertal boys (with biologic age less than 13 years) and for pubertal boys (with biological age greater than or equal to 13 years). Regression analyses that included all boys (prepubertal and pubertal) were also calculated and made it possible to test whether the effects (the slopes) of a given explanatory variable (for example the intensity of daily activity) on a given response (for example the daily awake hours) were significantly different for prepubertal and pubertal boys. To explore whether periods of high physical activity (training) relate negatively to physical activity during the rest of the day as reported for adults, data were analyzed as follow: THE RELATIONSHIP BETWEEN AMOUNT AND INTENSITY OF DAEE. Because these 2 variables are intrinsically related, a clear positive correlation is expected. However, in the case of a negative compensational process, such as high intensity of daily activity corresponding to a relatively smaller amount of daily activity, this correlation should not be very large. 642

Eiholzer et al

THE

DAEE AND Boys with high intensity of total daily activity may negatively compensate with more sleeping hours; in this case the correlation between the 2 variables would be negative. RELATIONSHIP BETWEEN INTENSITY OF

DAILY AWAKE HOURS.

THE

TrAEE (INTENSITY AND SpAEE. If training activity is negatively compensated, an increased TrAEE would lead to a decreased amount of SpAEE, and the correlation between the 2 would be negative. RELATIONSHIP

BETWEEN

AMOUNT) AND AMOUNT OF

Ethics Informed consent was obtained from all participants and their parents. The study was approved by the ethical committee of the University Hospital of Zurich, Switzerland.

RESULTS TDEE measured by the DLW method (gold standard) was significantly (P ⬍ .001) higher than when measured by combining accelerometry and calorimetry, mean TDEEDLW and mean TDEEacc, being 2297 kcal/d and 1938 kcal/d, respectively. This may be explained by ice hockey (sliding movement) being the predominant physical activity in our study. However, individual TDEEDLW and TDEEacc were highly positively and significantly correlated (r ⫽ 0.61, P ⬍ .001).

Clinical Characteristics and Descriptive Statistics For prepubertal boys (n ⫽ 46) mean (SD) age, biologic age, height SDS, and weight SDS was 9.43 (2.18), 9.13 (2.09), 0.25 (0.77), and 0.42 (1.17). For pubertal boys (n ⫽ 20) mean (SD) age, biologic age, height SDS, and weight SDS were 13.79 (2.18), 13.82 (0.77), 0.64 (1.33), and 0.75 (1.36). There was no significant difference in the number of daily awake hours and physical activity level; however, all other energy and activity measurements were significantly higher in the pubertal compared with the prepubertal group (Table II; available at www.jpeds.com). TDEEDLW, BMR, and physical activity level in both groups were within the range of published data for boys of equal age (Table III).25-30 Relationship Between the Intensity and Amount of Total Daily Activity Amount and intensity of DAEE were extremely highly and positively correlated in both prepubertal and pubertal boys in the univariate analysis (rprepub ⫽ 0.97, Pprepub ⬍ .0001; rpubertal ⫽ 0.96, Ppubertal ⬍ .0001) and in the multivariate analysis including the covariate biologic age (rprepub ⫽ 0.95, Pprepub ⬍ .0001; rpubertal ⫽ 0.96, Ppubertal ⬍ .0001). Relationship Between Intensity of Daily Activity and Daily Awake Hours For prepubertal boys the univariate analysis showed a significant positive correlation between intensity (Figure 1, A) The Journal of Pediatrics • November 2008

Table III. Literature data for TDEEDLW, BMR and PAL of healthy normal prepubertal and pubertal boys are listed, choosing 4 age groups closest to the mean age of our group Author Pubertal Bandini et al25 Livingstone et al30 Black et al26 Bratteby et al27 Mean/sum Our group Prepubertal Black et al26 Ekelund et al28 Goran et al29 Livingstone et al30 Mean/sum Our group

N

Age (y)

TDEEDLW (kcal/d)

BMR (kcal/d))

PAL

13 5 31 25 74 12

14.4 12.3 14.5 15 14.05 13.21

3103.88 2552.34 3366.52 3299.66 3080.60 2800.65

1733.40 1504.19 1933.96 1745.34 1729.22 1544.78

1.79 1.7 1.74 1.89 1.78 1.81

32 15 11 5 63 46

9.8 9.1 9.3 9.3 9.38 9.43

2339.85 2117.80 2070.05 2332.69 2215.10 2165.55

744.93 1243.94 1411.07 1134.11 1133.51 1093.52

1.72 1.7 1.47 2.05 1.74 2

significant covariate in this group. In pubertal boys there was a nonsignificant trend toward a negatively correlation between intensity of DAEE and daily awake hours both in the univariate (r ⫽ ⫺0.16) (Figure 1, B) and in the multivariate analysis (r ⫽ ⫺0.18). Importantly in the multivariate analysis including all boys, where we allowed different slopes (different beta coefficients) for characterizing the influence of intensity of daily activity on the daily awake hours, the difference between the slopes for prepubertal and pubertal boys was statistically significant (P ⫽ .025). Combining the 2 groups, the relationship between intensity of DAEE and daily awake hours in the univariate analysis was still significantly positive (P ⫽ .005, r ⫽ 0.34), whereas in the multivariate analysis r was .13 (P ⫽ .29).

Figure 1. Simple linear regression (Pearson correlation) between intensity of DAEE (log scale) and daily awake hours A, for prepubertal boys and B, for pubertal boys. Correlation coefficient (r) and significance levels (P) are indicated.

of DAEE and daily awake hours (r ⫽ 0.55, P ⫽ .0001). This correlation remained clearly positive and significant (r ⫽ 0.39, P ⬍ .01) in the multivariate analysis, biologic age not being a

Spontaneous Activity Versus Training Activity For prepubertal boys, the univariate correlation between amount of SpAEE and TrAEE intensity (Figure 2, a) and amount of SpAEE and TrAEE was highly and significantly (P ⬍ .0001) positive, the correlation coefficient being 0.72 and 0.58, respectively. The multivariate analysis showed (1) that intensity of TrAEE was a strong and highly significant predictor of the amount of SpAEE (r ⫽ 0.52, P ⬍ .001), biologic age not being a significant covariate (Table IV, A; available at www.jpeds.com) and (2) that the amount of TrAEE was a somewhat weaker positive predictor of the amount of SpAEE (r ⫽ 0.34, P ⫽ .03), biologic age being a significant covariate in this model (r ⫽ 0.38, P ⫽ .02) (Table IV, B; available at www.jpeds.com). For pubertal boys the univariate and multivariate correlation between the amount of SpAEE and TrAEE (intensity and amount) showed positive trends without reaching significance (Figure 2, B, and Table IV). In the multivariate analysis between SpAEE and TrAEE (intensity and amount) including all boys, differences between slopes of prepubertal and pubertal boys were not

Association between Short Sleeping Hours and Physical Activity in Boys Playing Ice Hockey

643

Figure 2. Simple linear regression (Pearson’s correlation) between intensity of TrAEE (log scale) and amount of SpAEE (log scale) A, for prepubertal and B, for pubertal boys. Correlation coefficient (r) and significance levels (P) are indicated.

significant (P ⫽ .22 and P ⫽ .58 for intensity and amount of TrAEE, respectively).

DISCUSSION We investigated in healthy boys how spontaneous physical activity relates to daily awake hours and induced physical activity (training). This is of particular interest because successful efforts to prevent childhood obesity by improving levels of physical activity in the population depend on an accurate understanding of how physical activity is controlled. In prepubertal boys intensity of DAEE and the number of daily awake hours, the 2 factors determining the amount of DAEE were significantly positively correlated even after correction for biologic age, indicating that in prepubertal boys high-intensity physical activity is not compensated by more resting hours. Interestingly, in pubertal boys we were unable to find a positive correlation between intensity of DAEE and daily awake hours. In other words, at least in this group of boys, daily awake hours— or short sleeping hours—seem to be positively related to physical activity before the age of puberty. This finding in prepubertal boys is new. On the basis of cross-sectional studies in children and adults showing a “doseresponse” relationship between short sleep duration and over644

Eiholzer et al

weight and overweight being related to self-reported physical inactivity, we would have speculated short sleepers as being less active.15,16 To our knowledge this is the first study on the relationship between sleeping hours and objectively measured physical activity in children. How sleeping hours impact on energy balance is a field of growing interest. Life style factors, parental factors, as well as (neuro)endocrine factors seem to be important.6,31 Our results suggest that before puberty in a group of competitive boys short sleep, at least up to a certain point, relates to changes in energy balance counteracting obesity. In prepubertal boys, independent of their biologic age, training activity was found to be positively correlated with spontaneous physical activity, this correlation being stronger with the intensity than with the amount of training. Prepubertal boys who trained more, and in particular, more intensely, had the highest spontaneous physical activity. On the basis of the study design, it is not possible to differentiate whether this relationship is causal or associative. However, the results are in keeping with our previous work in a group of intrinsically hypoactive children with PWS, where addition of a short but intensive daily training significantly increased spontaneous physical activity.10 Having studied boys selected from a group of ice-hockey players, a group selection bias may explain the positive relationship—athletes would not be athletes without having a spontaneous preference for physical activity. The relationship between intensity of training and spontaneous physical activity being highest in prepubertal boys, becoming weaker in pubertal boys, and even inverse in adults suggests a child-specific regulation of physical activity.11,12 Prepubertal and pubertal boys were separated on the basis of their biologic age (bone age), suggesting that intrinsic factors were responsible for the differences in relationship between sleeping hours and daily physical activity, and between spontaneous physical activity and training intensity. When puberty begins, many extrinsic factors such as time availability and social activities do change simultaneously (adolescence). For example, a tighter timetable leading to active time management in adolescent girls was shown to be associated with physical inactivity.32,33 Further studies are needed to explore the effect of puberty on the control of physical activity. The presented data provide additional evidence that there is child-specific control of physical activity and that possibly after the age of onset of puberty, this child-specific pattern converges to what is seen in adults. Whether this is due to intrinsic or extrinsic factors is unknown. The authors thank the Ice Hockey Club GCK Lions and all the volunteers who participated in the study.

REFERENCES 1. Andersen LB, Harro M, Sardinha LB, Froberg K, Ekelund U, Brage S, et al. Physical activity and clustered cardiovascular risk in children: a cross-sectional study (The European Youth Heart Study). Lancet 2006;368:299-304. 2. Brage S, Wedderkopp N, Ekelund U, Franks PW, Wareham NJ, Andersen LB, et al. Features of the metabolic syndrome are associated with objectively measured

The Journal of Pediatrics • November 2008

physical activity and fitness in Danish children: the European Youth Heart Study (EYHS). Diabetes Care 2004;27:2141-8. 3. Rizzo NS, Ruiz JR, Hurtig-Wennlof A, Ortega FB, Sjostrom M. Relationship of physical activity, fitness, and fatness with clustered metabolic risk in children and adolescents: the European youth heart study. J Pediatr 2007;150:388-94. 4. Ekelund U, Sardinha LB, Anderssen SA, Harro M, Franks PW, Brage S, et al. Associations between objectively assessed physical activity and indicators of body fatness in 9- to 10-year-old European children: a population-based study from 4 distinct regions in Europe (the European Youth Heart Study). Am J Clin Nutr 2004;80:584-90. 5. Tomkinson GR, Leger LA, Olds TS, Cazorla G. Secular trends in the performance of children and adolescents (1980-2000): an analysis of 55 studies of the 20m shuttle run test in 11 countries. Sports Med 2003;33:285-300. 6. Schwartz MW, Woods SC, Porte D, Jr., Seeley RJ, Baskin DG. Central nervous system control of food intake. Nature 2000;404:661-71. 7. Thorburn AW, Proietto J. Biological determinants of spontaneous physical activity. Obes Rev 2000;1:87-94. 8. Cohen DA, Ashwood JS, Scott MM, Overton A, Evenson KR, Staten LK, et al. Public parks and physical activity among adolescent girls. Pediatrics 2006;118:e1381-e9. 9. Davies PS, Joughin C. Using stable isotopes to assess reduced physical activity of individuals with Prader-Willi syndrome. Am J Ment Retard 1993;98:349-53. 10. Eiholzer U, Nordmann Y, l’Allemand D, Schlumpf M, Schmid S, KromeyerHauschild K. Improving body composition and physical activity in Prader-Willi Syndrome. J Pediatr 2003;142:73-8. 11. Westerterp KR. Alterations in energy balance with exercise. Am J Clin Nutr 1998;68:970S-4S. 12. Westerterp KR. Pattern and intensity of physical activity. Nature 2001;410:539. 13. Kempen KP, Saris WH, Westerterp KR. Energy balance during an 8-week energy-restricted diet with and without exercise in obese women. Am J Clin Nutr 1995;62:722-9. 14. Van Cauter E, Holmback U, Knutson K, Leproult R, Miller A, Nedeltcheva A, et al. Impact of sleep and sleep loss on neuroendocrine and metabolic function. Horm Res 2007;67(Suppl):2-9. 15. Sekine M, Yamagami T, Handa K, Saito T, Nanri S, Kawaminami K, et al. A dose-response relationship between short sleeping hours and childhood obesity: results of the Toyama Birth Cohort Study. Child Care Health Dev 2002;28:163-70. 16. Chaput JP, Brunet M, Tremblay A. Relationship between short sleeping hours and childhood overweight/obesity: results from the ’Quebec en Forme’ Project. Int J Obes (Lond) 2006;30:1080-5. 17. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med 2004;141:846-50. 18. McMurray RG, Baggett CD, Harrell JS, Pennell ML, Bangdiwala SI. Feasibility

of the Tritrac R3D accelerometer to estimate energy expenditure in youth. Pediatr Exerc Sci 2004;16:219-30. 19. Rowlands AV, Thomas PW, Eston RG, Topping R. Validation of the RT3 triaxial accelerometer for the assessment of physical activity. Med Sci Sports Exerc 2004;36:518-24. 20. Tobias HJ, Goodman KJ, Blacken CE, Brenna JT. High-precision D/H measurement from hydrogen gas and water by continuous-flow isotope ratio mass spectrometry. Anal Chem 1995;67:2486-92. 21. Black AE, Prentice AM, Coward WA. Use of food quotients to predict respiratory quotients for the doubly-labeled water method of measuring energy expenditure. Hum Nutr Clin Nutr 1986;40:381-91. 22. Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 1949;109:1-9. 23. Energy and protein requirements. Report of a joint FAO/WHO/UNU Expert Consultation. World Health Organ Tech Rep Ser 1985;724:1-206. 24. Greulich WW, Pyle SJ. Radiographic Atlas of the Development of the Hand and Wrist. Stanford CA, USA: Stanford University Press; 1959. 25. Bandini LG, Schoeller DA, Dietz WH. Energy expenditure in obese and nonobese adolescents. Pediatr Res 1990;27:198-203. 26. Black AE, Coward WA, Cole TJ, Prentice AM. Human energy expenditure in affluent societies: an analysis of 574 doubly-labeled water measurements. Eur J Clin Nutr 1996;50:72-92. 27. Bratteby LE, Sandhagen B, Fan H, Enghardt H, Samuelson G. Total energy expenditure and physical activity as assessed by the doubly labeled water method in Swedish adolescents in whom energy intake was underestimated by 7-d diet records. Am J Clin Nutr 1998;67:905-11. 28. Ekelund U, Sjostrom M, Yngve A, Poortvliet E, Nilsson A, Froberg K, et al. Physical activity assessed by activity monitor and doubly labeled water in children. Med Sci Sports Exerc 2001;33:275-81. 29. Goran MI, Gower BA, Nagy TR, Johnson RK. Developmental changes in energy expenditure and physical activity in children: evidence for a decline in physical activity in girls before puberty. Pediatrics 1998;101:887-91. 30. Livingstone MB, Coward WA, Prentice AM, Davies PS, Strain JJ, McKenna PG, et al. Daily energy expenditure in free-living children: comparison of heart-rate monitoring with the doubly labeled water (2H2(18)O) method. Am J Clin Nutr 1992;56:343-52. 31. Crisalli J, Amin RS. The systemic effects of short sleep period. J Pediatr 2007;150:331-2. 32. Neumark-Sztainer D, Story M, Hannan PJ, Tharp T, Rex J. Factors associated with changes in physical activity: a cohort study of inactive adolescent girls. Arch Pediatr Adolesc Med 2003;157:803-10. 33. Rushovich BR, Voorhees CC, Davis CE, Neumark-Sztainer D, Pfeiffer KA, Elder JP, et al. The relationship between unsupervised time after school and physical activity in adolescent girls. Int J Behav Nutr Phys Act 2006;3:20.

Association between Short Sleeping Hours and Physical Activity in Boys Playing Ice Hockey

645

Table I. Definitions of physical activity variables Variable

Unit

Definition

Amount of DAEE Intensity of DAEE Amount of SpAEE Intensity of SpAEE Amount of TrAEE Intensity of TrAEE

kcal kcal/min kcal kcal/min kcal kcal/min

Total PAEE recorded over 96 hours Amount of DAEE/total awake time Total PAEE recorded outside of training ⫹ (Intensity of SpAEE ⫻ Recorded total training time) Total PAEE outside of training/(Total awake time ⫺ Total training time) Mean duration of training over 96 hours according to individual training schedule ⫻ Intensity of TrAEE Total PAEE recorded during training/Total training time

PAEE, Physical activity energy expenditure; DAEE, total daily activity energy expenditure; SpAEE, spontaneous physical activity energy expenditure; TrAEE, Training activity energy expenditure.

Table II. Clinical characteristics (mean [SD]) of the 46 prepubertal and 20 pubertal children

Chronological age (y) Biological age (y) Height SDS Weight SDS Awake hours/day TDEEDLW (kcal/day) BMR (kcal/day) PAL (kcal/kcal) Amount DAEE (kcal/4 days) Intensity DAEE (kcal/min) Amount SpAEE (kcal/4 days) Intensity SpAEE kcal/min) Amount TrAEE (kcal/4 days) Intensity TrAEE (kcal/min)

Prepubertal

Pubertal

9.43 (2.18) 9.13 (2.09) 0.25 (0.77) 0.42 (1.17) 12.67 (0.79) 2166 (501) 1095 (179) 2.00 (0.44) 2536 (843) 0.83 (0.24) 2058 (693) 0.73 (0.21) 511 (316) 2.16 (0.77)

13.79 (1.15)* 13.82 (0.77)* 0.64 (1.33) 0.75 (1.36) 13.06 (1.04) 2801 (525)* 1544 (177)* 1.81 (0.26) 3421 (909)* 1.09 (0.30)* 2796 (736)* 0.97 (0.27)* 781 (357)* 3.30 (0.82)*

P values were calculated from 2-sample t tests. Amounts of physical activity energy expenditure reflect the total during the 4-day measurement period. *P ⬍ .05 for prepubertal boys.

645.e1

Eiholzer et al

Table IV. Multivariate analysis in prepubertal (n ⴝ 43) and pubertal children (n ⴝ 16) on the amount of spontaneous physical energy expenditure

(a) A model including intensity of training activity energy expenditure (TrAEE) and biologic age R2 Intensity of TrAEE Biologic age (b) A model including amount of TrAEE and biologic age R2 Amount of TrAEE Biologic age

Prepubertal children

Pubertal children

0.53 0.52 (.0005) 0.12 (.46)

0.07 0.22 (.42) 0.11 (.70)

0.43 0.34 (.027) 0.38 (.013)

0.10 0.29 (.29) 0.19 (.49)

The R2 value together with partial correlation coefficients and corresponding P values (in parentheses).

The Journal of Pediatrics • November 2008