original research Pediatric Exercise Science, 2014, 26, 221-230 http://dx.doi.org/10.1123/pes.2013-0109 © 2014 Human Kinetics, Inc.

The Relationship Between Objectively Measured Physical Activity, Salivary Cortisol, and the Metabolic Syndrome Score in Girls Katrina D. DuBose East Carolina University

Andrew J. McKune University of KwaZulu-Natal The relationship between physical activity levels, salivary cortisol, and the metabolic syndrome (MetSyn) score was examined. Twenty-three girls (8.4 ± 0.9 years) had a fasting blood draw, waist circumference and blood pressure measured, and wore an ActiGraph accelerometer for 5 days. Saliva samples were collected to measure cortisol levels. Previously established cut points estimated the minutes spent in moderate, vigorous, and moderate-to-vigorous physical activity. A continuous MetSyn score was created from blood pressure, waist circumference, high-density-lipoprotein (HDL), triglyceride, and glucose values. Correlation analyses examined associations between physical activity, cortisol, the MetSyn score, and its related components. Regression analysis examined the relationship between cortisol, the MetSyn score, and its related components adjusting for physical activity, percent body fat, and sexual maturity. Vigorous physical activity was positively related with 30 min post waking cortisol values. The MetSyn score was not related with cortisol values after controlling for confounders. In contrast, HDL was negatively related with 30 min post waking cortisol. Triglyceride was positively related with 30 min post waking cortisol and area under the curve. The MetSyn score and many of its components were not related to cortisol salivary levels even after adjusting for physical activity, body fat percentage, and sexual maturity. Keywords: physical activity, salivary cortisol, metabolic syndrome, children The metabolic syndrome (MetSyn) is a clustering of specific cardiovascular risk factors, including high blood pressure, high glucose, dyslipidemia (high triglycerides and low high-density-lipoprotein cholesterol), and central obesity. According to Ford et al. (18), 34.3% of adults living in the United States have the MetSyn. The MetSyn has also been observed in 8.6% of adolescents (35) and 5% of children (12). Adults with the MetSyn have an increased risk for type II diabetes and cardiovascular disease (CVD; 19, 36). The development of CVD can begin in youth (48). Therefore, youth with the MetSyn may have a higher risk for either CVD or Type 2 diabetes as young adults. The etiology for the MetSyn is currently unknown; dysregulation of the hypothalamicpituitary-adrenal (HPA) axis is one likely factor (4,5). A DuBose is with the Dept. of Kinesiology, East Carolina University, Greenville, NC. McKune is with the Discipline of Biokinetics, Exercise and Leisure Sciences, University of KwaZulu-Natal, Durban, South Africa. Address author correspondence to Katrina D. DuBose at [email protected].

dysregulation of the HPA axis increases cortisol levels. Increased cortisol is associated with increased abdominal body fat accumulation, triglyceride storage in the adipose cell, insulin resistance, and promotes hypertension (28). Cortisol is very responsive to exercise and physical activity in adults (24). Moreover, physical activity can improve HPA axis functioning. The impact physical activity has on cortisol levels in children is unclear. After an acute bout of exercise in children, the findings have shown increased serum cortisol levels (9), decreased salivary cortisol levels (25), or no change in salivary cortisol levels (50). Further, a paucity of knowledge exists regarding the effect of total daily physical activity on cortisol levels in children. Martikainen et al. (39) reported that in children, neither total daily physical activity nor vigorous physical activity was related to salivary cortisol levels. In contrast, 12 weeks of exercise training resulted in decreased serum cortisol levels in obese boys (32). Limited research has examined the relationship between HPA axis dysregulation and the MetSyn in children. Depending on the marker used to examine this relationship, mixed results have been reported between

221

Downloaded by Univ of Hong Kong Libraries on 09/16/16, Volume 26, Article Number 3

222  DuBose and McKune

the dysregulated HPA axis and the MetSyn. In boys, stress level (measured by surveys) was positively related with the MetSyn score, which is a continuous score composed of the MetSyn components (27). Others have reported that serum cortisol is positively related with the MetSyn in obese youth (46,51). However, DuBose and McKune (11) reported that salivary cortisol was not related to the MetSyn score in children. Regardless of the mechanism involved with the development of the MetSyn, physical activity is recommended as part of the treatment in youth (43). Yet, only a few studies have investigated the effect of physical activity on the MetSyn score in youth (3,6,14,15,29). The MetSyn score was found to be inversely related to total physical activity in children and adolescents (3,6,14,15). Hsu et al. (29) reported that youth with the MetSyn spent less time in moderate-to-vigorous physical activity (MVPA) than those without the MetSyn. To the best of our knowledge only one study has investigated the relationship between physical activity participation, a dysregulated HPA axis (via a stress score), and the MetSyn in children (27). The authors reported participation in MVPA was inversely related to the MetSyn score regardless of the stress score. Physical activity participation appears to reduce the risk of MetSyn development. It is important to consider the marker used to measure HPA axis functioning when evaluating the role that the HPA axis has in MetSyn development. Commonly used measures included either self-reported stress levels (27), or a physiologic marker like cortisol (9,29,46,50,51). Cortisol can be measured via serum or saliva. Serum cortisol reflects bound cortisol levels in the body, whereas salivary cortisol reflects the amount of active hormone levels in the body making it more reflective of its potential biological effects (1,23). Physical activity has been recommended as a treatment of the MetSyn and to reduce stress levels, but the relationship between physical activity, cortisol, and the MetSyn in children is currently unclear. Therefore, the purpose of this study was to investigate the relationship between physical activity, salivary cortisol, and the MetSyn score in young girls. A secondary purpose was to examine the relationship between physical activity, salivary cortisol, and components of the MetSyn in this population.

Methods Participant Selection Participants were recruited via local after-school programs, churches, elementary schools, and newspaper advertisements. Those interested in participating in the study needed to be girls between the ages of 7–10 years old and both the parent and child provided written consent and assent, respectively, in accordance with the university’s Institutional Review Board. Children could not participate in the study if they had any of the following conditions: 1) known cardiovascular disease; 2) diabetes

(Type 1 or Type 2); or 3) any condition that limited their ability to perform physical activity.

Study Design The participants completed two study visits that were about 8 days apart from each other. The following procedures were completed at the study visits: 1) informed consent/assent, 2) resting blood pressure, 3) anthropometrics (height, weight, skinfolds, and circumferences), 4) fasting blood draw, 5) sexual maturity assessment, 6) distribution and explanation of activity monitors, physical activity log book, and saliva collection, and 7) return activity monitors, physical activity log book, and saliva.

Measures Anthropometric Data.  Height was measured to

the nearest 0.1 cm using a wall mounted stadiometer (Perspective Enterprises, Portage, MI, USA). Weight was measured to the nearest 0.1 kg using a portable electronic scale (Model # 68987, Befour Inc., Saukville, WI, USA). Both height and weight were measured in duplicate with shoes off, but wearing light weight clothing. Age and sex adjusted body mass index (BMI) was calculated as kg/ m2 according to the Centers for Disease Control (CDC) guidelines (33). The BMI percentile was calculated for each child to determine normal weight (> 5th—< 85th %tile), overweight (85th %tile—< 95th %tile), and obese (³ 95th %tile; 33). Total percent body fat was determined by skinfold measurements from the triceps and calf (CRE100 Lange, Beta Technology Inc., Sana Cruz CA, USA). Trained personnel took the measurements in duplicate using procedures outlined by the American College of Sports Medicine (2). Age and sex appropriate equations were used to calculate percent body fat (38). Waist circumference measurements were taken at the narrowest portion of the waist underneath the clothing using a Gulick tape measure (2). Waist circumference was measured in duplicate to the nearest 0.1 mm and the average value was calculated. Sexual Maturity.  Sexual maturity (secondary sex

characteristics) was determined by Tanner staging via parent proxy report. Parents of the participants viewed sketches of the stages of breast and pubic hair development for sexual maturation, identified their child’s development, and then placed the sheets in an envelope and sealed them before returning them to the investigator. For Tanner stages I and II, this method has been shown to be as valid as Tanner staging by a physician without the increased burden placed on the child (44). Both breast development and pubic hair development have been linked to obesity in children (7,31). Measuring both also provides a clearer picture as to the participant’s sexual development.

Blood Pressure.  Resting blood pressure was obtained on two different study visits. The child rested quietly

Physical Activity, Cortisol, and MetSyn   223

for five minutes before blood pressure was measured in duplicate by trained personnel using a random-zero sphygmomanometer. The first and last Korotkoff sounds were recorded as systolic and diastolic blood pressure, respectively (2). All systolic and diastolic measurements were used to calculate the average systolic and diastolic blood pressure. These average values were then used to calculate mean arterial pressure using the following formula: MAP = ((systolic blood pressure-diastolic blood pressure)/3) + diastolic blood pressure.

Downloaded by Univ of Hong Kong Libraries on 09/16/16, Volume 26, Article Number 3

Blood Collection and Analyses.  Blood samples

were obtained from the antecubital vein after an 8 hr fast using standard venipuncture methods by a trained phlebotomist. Glucose, total cholesterol, and triglyceride concentrations were measured enzymatically using a Cobas Mira Chemistry System (Roche Diagnostic Systems, Indianapolis, IN). High-density-lipoprotein cholesterol (HDL-C) concentrations were also measured enzymatically using a Cobas Mira Chemistry System (Diagnostic Chemicals Ltd, Oxford, CT).

Metabolic Syndrome Score.  Due to the low prevalence

of the MetSyn and the lack of a standard definition for the MetSyn, a continuous MetSyn score was created from the following factors: waist circumference, HDL-C, triglycerides, glucose, and MAP (16,17,22,30). These components were chosen as factor analyses has indicated that lipids, glucose, blood pressure, and adiposity are important for cardiovascular risk in the pediatric population (21,22,37). The standardized residuals (z-score) were calculated for each component and then the standardized score for HDL-C was multiple by -1 since it is inversely related to metabolic risk. The z-scores for waist circumference, glucose, triglyceride, HDL-C and MAP were them summed together to generate the MetSyn score. The higher the MetSyn score the less favorable metabolic profile the child had. Saliva Collection.  The children, together with a parent/

guardian, were trained in the saliva collection procedure. They were requested to adhere as closely as possible to a standardized collection guideline (41). Saliva samples were collected three times at home over one day (waking, 30 min post waking and evening [~8 p.m.]).The parent/guardian recorded the time each saliva sample was collected to ensure the procedures were being followed. Saliva samples were collected via unstimulated, passive drool over a time period of 3 min. While seated the children were asked to lean slightly forward and tilt their heads down and accumulate saliva in the floor of the mouth for a minute. At the end of the minute the saliva was swallowed and they then had to accumulate saliva for a further three minutes. During the three minutes they could dribble the saliva through a 5 cm plastic straw into a preweighed polypropylene cryovial (2 ml capacity) at any time. Care was taken to allow saliva to dribble into the collecting tubes with minimal orofacial movement. Samples were refrigerated immediately after collection in home freezers (-20 °C) and kept frozen until reaching the laboratory, upon which they were stored at -70 °C until analysis.

Salivary Cortisol.  Salivary cortisol was measuring using an enzyme-linked immunoabsorbent assay (ELISA) kit (Salimetrics, State College, PA, USA). All samples were analyzed in duplicate with the inter- and intraassay coefficients of variation averaging 4.8% and 5.5%, respectively. Area under the curve with respect to ground was calculated to estimate the total free salivary cortisol within the 1st 30 min upon waking. Physical Activity Assessment.  Participants wore a pedometer ActiGraph accelerometer (GT1M, ActiGraph LLC, Pensacola, FL, USA) for five days, including two weekend days. The activity monitors were worn during waking hours around the waist. To increase compliance, participants called the laboratory each morning and indicated they were wearing the activity monitors. Study staff followed-up with children who did not call the laboratory each day. Data from the ActiGraph determined the amount of time spent in moderate, vigorous, and moderate-to-vigorous physical activity per day. Physical Activity Data Reduction.  The epoch length

was set at 1 min intervals. Data were included if the ActiGraph was worn for at least 8 hr and 4 days (including at least 1 weekend day). Amount of time spent in moderate and vigorous physical activity was determined using Evenson’s cut points (49). Moderate physical activity was classified as being between 4–6.9 METs and vigorous was ³ 7 METs. Moderate-to-vigorous physical activity (MVPA) was defined as activities ³ 4 METs.

Statistical Analysis Means and standard deviations were calculated for demographic characteristics. Student’s t test was used to examine differences in demographic characteristics racial group (Caucasians and non-Caucasians). Cortisol values for 30-min post waking, evening, and AUC were not normally distributed, so they were transformed using log transformation. Pearson’s correlations were used to examine the unadjusted relationship between waking, 30-min post, evening and AUC cortisol values and physical activity levels. Pearson’s correlations were used to also examine the unadjusted relationship between the physical activity levels, MetSyn score, and its related components. Regression analyses were used to examine the relationship between cortisol factors and the MetSyn score and its related components adjusted for body fat percentage, and breast and pubic hair development. SAS (version 9.3, Research Triangle, NC) was used for all analysis. Significance level was set at p < .05.

Results Table 1 shows the demographic characteristics for all participants. The average age was 8 years old and overall the girls had normal lipid and glucose levels. The average amount of time spent in MVPA was about 30 min per day. The majority of the participants were Caucasian (65%),

224  DuBose and McKune

17% were African American, 4% were Asian, and 13% were other. Demographic characteristics were similar across the racial groups (Caucasians versus non-Caucasians). Eight-seven percent and 100% of the children were at a Tanner Stage 1 or 2 for breast and pubic hair development, respectively. Four percent (n = 1) were at a Tanner stage 3 and 9% (n = 2) at a Tanner Stage 4 for breast development. Table 2 shows the univariate analysis between cortisol variables, physical activity levels, the MetSyn score, and MetSyn components. 30-min post waking cortisol

levels had a strong to moderate positive association with vigorous physical activity and MVPA, respectively. Vigorous physical activity was also positively related to AUC. No other relationships were observed between physical activity levels and cortisol variables. Physical activity levels were not associated with either the MetSyn score or its components, except vigorous physical activity was negatively related with mean arterial pressure. The relationship between cortisol variables and physical activity levels adjusting for body fat percentage and either breast or pubic hair development are presented in

Table 1  Demographic Characteristics of the Participants

Downloaded by Univ of Hong Kong Libraries on 09/16/16, Volume 26, Article Number 3

Measures

N = 23 (Mean ± SD)

Age (years)

8.4 ± 0.9

Anthropometric Weight (kg)

40.8 ± 13.7

Height (cm)

13.8.8 ± 8.4

Body mass index (kg/m2)

20.8 ± 5.1

Waist circumference (cm)

66.8 ± 13.6

Body fat (%)

29.0 ± 10.8

Metabolic Systolic blood pressure (mmHg)

107.6 ± 9.1

Diastolic blood pressure (mmHg)

69.4 ± 4.8

Mean arterial pressure

82.1 ± 4.7

HDL (mg/dl)

46.8 ± 8.3

Triglycerides (mg/dl)

65.1 ± 27.6

Glucose (mg/dl)

80.4 ± 11.4

Metsyn score

–0.04 ± 2.64

Salivary Cortisol Waking cortisol (pg/ml)

0.35 ± 0.2

30-min post waking cortisol (pg/ml)

0.38 ± 0.41

Evening cortisol (pg/ml)

0.14 ± 0.49

Area under the curve

–0.37 ± 0.25

30-min post waking cortisol

(pg/ml)a

–1.21 ± 0.63

Evening cortisol (pg/ml) a

–3.20 ± 1.08

Area under the curve

–1.16 ± 0.55

a

Physical Activity Sedentary (min/day)

937.8 ± 56.2

Light PA (min/day)

383.4 ± 48.9

Moderate PA (min/day)

25.7 ± 11.3

Vigorous PA (min/day)

7.1 ± 7.6

MVPA (min/day)

32.8 ± 17.1

Wear Time (min/day)

811.9 ± 40.5

Note. HDL = high-density-lipoprotein cholesterol; Metsyn = score for metabolic syndrome; PA = physical activity; MVPA = moderate-to-vigorous physical activity. a =log transformed variables. *p < .05

Physical Activity, Cortisol, and MetSyn   225

Table 2  Correlations Between Cortisol Variables, Physical Activity Variables, the Metabolic Syndrome Score, and Metabolic Syndrome Components Waking Waking

Evening

AUC

Mod

Vig

MVPA

MetSyn score

HDL

Glucose

Trig

Waist

0.55*

1.0

Evening a

0.61*

0.37

1.0

AUC

0.81*

0.90*

0.46*

1.0

0.06

0.23

–0.25

0.17

1.0

0.14

0.52*

–0.12

0.46*

0.63*

1.0

MVPA

0.14

0.39*

–0.22

0.32

0.94*

0.86*

1.0

MetSyn score

–0.04

0.11

–0.07

0.08

–0.30

–0.16

–0.27

1.0

–0.22

–0.63*

–0.55*

-0.53*

0.04

–0.38

–0.015

–0.25

1.0

–0.19

0.55*

–0.05

1.0

–0.05

0.72*

–0.31

0.11

1.0

–0.16

0.56*

0.46

0.15

0.25

1.0

–0.34

0.64*

0.30

0.15

0.25

0.68*

a

Vig

HDL Glucose Trig Waist MAP

MAP

1.0

30 min post waking a

Mod

Downloaded by Univ of Hong Kong Libraries on 09/16/16, Volume 26, Article Number 3

30-min post wake

0.09 0.31 0.10 –0.12

–0.17 0.47 –0.08 –0.06

0.21

–0.21

–0.24

0.51*

–0.14 –0.08

–0.03 –0.16

–0.29 –0.15

0.00 0.09

–0.02

–0.31

–0.19

-0.48*

1.0

Note. AUC = area under the curve; Mod = moderate physical activity; Vig = vigorous physical activity; MVPA = moderate-to-vigorous physical activity; MetSyn score= score for metabolic syndrome; HDL = high-density-lipoprotein cholesterol; Trig = triglycerides; waist = waist circumference; MAP = mean arterial pressure. a =log transformed variables * p