Clin Auton Res DOI 10.1007/s10286-015-0277-y
RESEARCH ARTICLE
Heart rate variability, adiposity, and physical activity in prepubescent children Andre Filipe Santos-Magalhaes • Luisa Aires • Clarice Martins • Gustavo Silva • Ana Maria Teixeira Jorge Mota • Luis Rama
•
Received: 23 August 2014 / Accepted: 4 December 2014 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study aimed at examining the associations between weight status, body fat mass, and heart rate variability in prepubescent children, adjusting for physical activity levels. Methods A cross-sectional investigation in which a total of 50 Caucasian pre-pubertal children (21 normal weight; 8 overweight; 21 obese), aged 6–10 years (8.33 ± 1.14), including both boys (n = 24) and girls (n = 26), were
A. F. Santos-Magalhaes (&) School of Health, University Fernando Pessoa, Porto, Portugal e-mail: [email protected] A. F. Santos-Magalhaes Health and Exercise Research Group, School of Sport and Exercise Sciences, University of Kent, Chatham, Kent, UK e-mail: [email protected] L. Aires C. Martins G. Silva J. Mota Research Centre in Physical Education Activity, Health and Leisure Time, Faculty of Sports – CIAFEL, University of Porto, Porto, Portugal e-mail: [email protected] C. Martins e-mail: [email protected] G. Silva e-mail: [email protected] J. Mota e-mail: [email protected] A. M. Teixeira L. Rama Research Centre for Sport and Physical Activity, Faculty of Sport Sciences and Physical Education, University of Coimbra, Coimbra, Portugal e-mail: [email protected] L. Rama e-mail: [email protected]
recruited from local schools. Total body fat and trunk fat were evaluated through dual-energy X-ray absorptiometry. Free-living physical activity levels were evaluated by accelerometer. Short-term heart rate variability acquisition was performed; time- and frequency-domain parameters were analysed. Logarithmic transformations of the lowfrequency (LnLFnu), high-frequency (LnHFnu) normalized units and low-frequency/high-frequency (LnLFnu/HFnu) ratio were computed. Results Adjusting for age, Tanner stage, and moderate to vigorous physical activity levels, obese children compared to normal weight children showed a significant decreased LnHfnu (3.8 ± 0.2 vs 4.1 ± 0.2 %) and both higher LnLFnu (4.0 ± 0.4 vs 3.7 ± 0.3 %) and LnLFnu/LnHFnu ratio (1.1 ± 0.1 vs 0.9 ± 0.1). LnHFnu showed significant negative correlation with waist circumference (r = -0.598; P = 0.000), total body fat (r = -0.409; P = 0.011) and trunk fat (r = -0.472; P = 0.003). Both LnLFnu and LnLFnu/LnHFnu ratio showed positive correlations with waist circumference (r = 0.455; r = 0.513) and trunk fat (r = 0.370; r = 0.415). Conclusions A higher amount of body fat mass, particularly central fat, was shown to be related to decreased parasympathetic modulation in time-domain heart rate variability. This finding highlights the potential cardiovascular risk that excessive fat mass may represent even at very young age. Keywords Heart rate variability Children Fat mass DXA Physical activity
Introduction The prevalence of obesity in children in the past two decades has rapidly increased, not only in developed
123
Clin Auton Res
countries but also in the developing world [1, 2]. Despite stabilizing and even decreasing prevalence rates in some countries, obesity is still a problem in most places [3]. Hypertension, dyslipidemia, insulin resistance, and type 2 diabetes are becoming more frequent in overweight and obese children [4], even among the very young children [5, 6]. The early onset of these co-morbidities is associated with both increased mortality and increased morbidity in adulthood [7–9]. It has been shown that these disorders have a greater relationship to fat distribution rather than to total weight [10, 11]. Obesity in children is also related to impairment of cardiovascular autonomic function, which reflects an impairment of the autonomic nervous system. Heart rate variability (HRV) is a noninvasive and reliable method to quantify the autonomic nervous system modulation of cardiovascular function, both in adults and in the pediatric population [12]. A decreased HRV in childhood is associated with several pediatric conditions such as congenital heart disease [13], acute traumatic brain injuries [14], diabetes mellitus [15] and obesity. Several studies have shown that obese children and adolescents have a decreased HRV, manifested by an altered sympathovagal balance, mainly due to a decrease in vagal modulation [16–20]. However, a vast majority of these investigations were conducted in pubescent children and adolescents and have used only indirect measures of obesity such as the BMI cut-off points. To date, very few studies have investigated the relationship between the quantity and distribution of fat mass as measured directly with dual-energy X-ray absorptiometry (DXA) and HRV [21, 22]. In a cohort of pre-pubescent and pubescent children Kaufman et al. [21] found an association between total body fat and decreased HRV vagal modulation, although the possible confounding effect of physical activity on HRV was not acknowledged, as in most of the studies done on obese children. In adolescents, favorable HRV profiles have been shown to be associated with higher amounts of moderate and vigorous physical activity, even after controlling for the body fat mass [22]. Moreover, exercise training programs have been shown to induce a positive effect on HRV vagal modulation in healthy children [23– 25]. To our knowledge, no data are available about the interaction between adiposity, HRV, and daily physical activity levels in prepubescent children. Therefore, the purpose of this study was to examine the relationship between weight status, body fat content as measured by dual-energy X-ray absorptiometry (DXA), and HRV in pre-pubertal children, adjusting for their physical activity levels. We hypothesized that even in prepubescent children there is an inverse relationship between body fat, particularly trunk fat, and parasympathetic HRV modulation.
123
Methods Participants This study was conducted as part of a school-based research program carried out in five public primary schools. A total of 50 Caucasian pre-pubertal children aged 6–10 years (8.33 ± 1.14), both genders (24 boys and 26 girls), were recruited from local schools. According to the body mass index (BMI) percentiles for age and gender, 21 children were categorized as normal weight, 8 were overweight, and 21 were obese. The Regional Education Board approved the study, and written informed consent from participants’ parents/guardians was obtained. The informed consent including the purpose of the research and the risks and benefits of participation were fully explained to each child and their respective parent(s), consistent with the Helsinki Declaration. The evaluation methods and procedures were reviewed and approved by the Scientific Board of the Faculty of Sport Sciences and Physical Education from the University of Coimbra. Medical history was ascertained via questionnaire and consultation with a medical doctor. Any children with abnormal cardiovascular function and/or abnormal cardiac electrophysiology and/or taking medication at the time were excluded. Procedures The evaluations were performed in the morning, between 8:00 am and 10:00 am, in the faculty laboratories. The same trained technician performed all of the evaluations to reduce potential measurement errors between observers. Children were asked to avoid strenuous exercise in the 24-h prior to the evaluation. They were also asked to do a 12-h fast and were only allowed to drink water after dinner on the day before the evaluation. After the anthropometric measurements, body composition, and HRV evaluation procedures, children were allowed to have breakfast and received accelerometers for physical activity assessment, together with instructions for use. Anthropometry and body composition Height was measured to the nearest 0.1 cm in bare or stocking feet with the children standing upright against a Holtain Stadiometer. Weight was measured to the nearest 0.1 kg, with the children lightly dressed using an electronic weight scale (Seca 708 portable digital beam scale). BMI was calculated from the ratio of body weight (kg)/body height (m2). BMI percentiles for age and gender were calculated and the cut-off points established by the International Obesity Task Force [26] were applied to categorize children as normal weight (BMI \ 85th percentile),
Clin Auton Res
overweight (85th percentile B BMI \ 95th percentile) or obese (BMI C 95th percentile). Waist circumference was evaluated following the NHANES protocol [27] using a circumference measuring tape (Seca 203). Percentage of total body fat and trunk fat was obtained through DXA (Explorer QDR 4500, Hologic, Bedford, MA) with whole body protocol. This assessment procedure lasted 20 min. Children were placed in the supine position with their arms in extension near the trunk and lower limbs in extension, with a slight abduction of the feet. Children removed clothes and all metallic objects (earrings, watches, etc.) and wore a gown. Blood pressure Blood pressure was measured with Colin Press Mate NonInvasive Blood Pressure Monitor (Model BP 8800p, Colin Medical Instruments Corporation, San Antonio, TX, USA)]. Systolic blood pressure and diastolic blood pressure were measured in the right arm with the subjects in the fasting state. The participants were in the sitting position (without their legs crossed), with the right arm at heart level. Three standard pressure cuffs of correct size (9 9 18, 12 9 23, 14 9 28 cm) were used according to the published guidelines for blood pressure assessment in children [28]. The first and second measurements were taken after 5 min and 10 min resting, considering the mean
of these measurements for statistical purposes. If these two measurements differed by 2 mmHg, the protocol was repeated to obtain two new measurements that could not exceed a 2 mmHg difference. Maturational stage Each child self-assessed his/her own stages of secondary sex characteristics: stage of breast development in females, genital development in males, and pubic hair development for both genders [29]. Only self-assessed pubic hair development was used in the study and that is considered to be a valid and reliable indicator of maturational status [30]. Mota el al. [30] found a high correlation (r = 0.73) between these ratings on two separate occasions (3-day interval) in a sample of 50 Portuguese youths. Physical activity Physical activity was objectively assessed by accelerometers (Actigraph GT3X, Pensacola, FL, USA) for seven consecutive days to obtain a reliable picture of the usual physical activity of the study participants [31]. Data were recorded in 15 s sampling periods (epochs). The participants used the accelerometer in an elastic waistband with the device positioned on the right hip during the daytime, except while sleeping, bathing, and during other
Table 1 Descriptive characteristics according to gender Variables
Boys (n = 26)
Girls (n = 24)
Total sample (n = 50)
P value boys vs girls
Age (years)
8.2 ± 1.2
8.4 ± 1.0
8.3 ± 1.1
0.386
Tanner stage
1.3 ± 0.6
1.5 ± 0.6
1.4 ± 0.6
0.215
Height (cm)
133.5 ± 9.7
133.6 ± 6.8
133.5 ± 8.3
Weight (kg)
35.8 ± 12.8
36.0 ± 8.7
35.9 ± 10.8
0.975 0.948
BMI (kg/m2)
19.5 ± 4.7
20.0 ± 3.5
19.7 ± 4.1
0.702
WC (cm)
67.2 ± 14.3
68.3 ± 9.7
67.7 ± 12.0
0.755
TBF (%)
31.4 ± 9.2
37.3 ± 6.4
34.5 ± 8.3
0.011*
TF (%) SBP (mmHg) DBP (mmHg)
27.6 ± 10.1 104.7 ± 8.5
34.3 ± 7.8
31.2 ± 9.5
0.012*
108.0 ± 13.5
106.4 ± 11.5
0.326
58.0 ± 4.4
59.1 ± 7.5
58.6 ± 6.2
0.557
Total PA (min/day) SEDPA (min/day)
726.2 ± 74.7 325.6 ± 79.7
755.2 ± 53.4 347.1 ± 66.5
741.7 ± 75.1 337.1 ± 72.9
0.148 0.342
LIGPA (min/day)
339.9 ± 61.4
356.1 ± 53.2
348.5 ± 57.0
0.360
MODPA (min/day)
47.7 ± 20.1
40.2 ± 12.3
43.7 ± 16.6
0.140
VIGPA (min/day)
13.0 ± 9.7
11.9 ± 6.2
12.4 ± 7.9
0.649
MVPA (min/day)
60.7 ± 27.2
52.1 ± 16.5
56.1 ± 22.3
0.207
Data are mean ± standard deviation P values reported from T test for independent means BMI body mass index, WC waist circumference, TBF total body fat, TF trunk fat, SBP systolic blood pressure, DBP diastolic blood pressure, Total PA total physical activity, SEDPA time in sedentary physical activity, LIGPA time in light physical activity, MODPA time in moderate physical activity, VIGPA time in vigorous physical activity, MVPA time in moderate to vigorous physical activity * P \ 0.05 for statistical significance
123
Clin Auton Res
count intervals [32] were considered: 0–100 for SEDPA, 101–2295 for LIGPA, 2296–4011 for MODPA, C4012 for VIGPA and C2296 for MVPA. These cut-off points seem to provide acceptable classification accuracy for all four levels of physical activity intensity and perform well among children of all ages [33]. Heart rate variability
Fig. 1 Gender differences in time- (a) and frequency-domain (b) heart rate variability. General linear model adjusted for age, Tanner stage and time in moderate to vigorous physical activity. Notes: RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LnLFnu natural logarithm of low-frequency normalized units, LnFHnu natural logarithm of high-frequency normalized units
aquatic activities. A data sheet was given to the participants, who were instructed to record the time when the monitor was attached in the morning, detached in the evening, and when it was removed for a bath. The standard software PROPERO (developed by the Centre of Research in Childhood Health of the University of Southern Denmark, 2011) was used to reduce the raw activity data from the accelerometers into daily physical activity. Time periods with at least 10 consecutive minutes of zero counts recorded were excluded from analysis assuming that the monitor was not worn. A minimum recording of 8 h/day was the criteria to accept daily physical activity data as valid. Individual subject data were only accepted for analysis if recordings of least 2 weekdays and 1 weekend day were successfully assessed. The main outcomes of reduced data were total physical activity (counts/min/day), time in sedentary physical activity [SEDPA (min/day)], light physical activity [LIGPA (min/day)], moderate physical activity [MODPA (min/day)], vigorous physical activity [VIGPA (min/day)] and moderate to vigorous physical activity [MVPA (min/day)]. To determine the time spent in physical activity of different intensities, the following
123
The HRV acquisition and analysis were conducted according to the international recommendations [34]. Prior to the HRV acquisition the proceedings were explained to the children, emphasizing the importance of staying still and relaxed during the procedure. The HRV acquisition was done in a quiet room with controlled temperature (21–23 °C) and humidity (40–60 %). To avoid any interference in the signal, cell phones or other electromagnetic devices were not allowed in the room during the heart rate recording. The equipment used was a Polar S810 (Polar Electro OY, Kempele, Finland) heart rate monitor receiver and strap, which has previously been validated for the HRV assessment in children [35]. The heart rate monitor strap was placed on each participant’s chest over the lower third of the sternum and the heart rate receiver was placed on the wrist. HRV was recorded for 15 min while participants were lying in the supine position. They were also asked to breathe at a constant respiratory rate of 0.25 Hz (14–16 breaths/min) with the help of a metronome. Importantly, prior to HRV assessment, the children were given a brief training on how to breathe at a constant rate. Only the last 5 min of the heart rate recording were utilized for the HRV analysis. The software Polar Performance SW was used to transfer data from the monitors and to analyze and correct for ectopic and missing beats. The HRV analysis was done with the Kubios HRV Analysis Software (Biosignal Analysis and Medical Image Group, Department of Physics, University of Kuopio, Finland). The successive interbeat (R–R) intervals, the standard deviation of normal R– R intervals (SDRR) and the square root of the mean of the squares of successive R–R interval differences (RMSSD) were computed as time-domain measures. The frequency domain analysis was performed through fast Fourier transformation algorithm, allowing the computation of the three-frequency bands: very low frequency (VLF 0–0.04 Hz), low frequency (LF 0.04–0.15 Hz), and high frequency (HF 0.15–0.4 Hz). However, only LF and HF were considered in our analysis, since the VLF component is not as reliable because of its lack of physiological meaning and interpretation. The normalized units of the LF (LFnu), HF (HFnu), and respective ratio (LFnu/HFnu) were computed to avoid the influence of the total power within the power spectrum [34].
Clin Auton Res Table 2 Descriptive characteristics according to weight status Variables
Age (years)
Normal weight (n = 21)
8.0 ± 1.2
Over weight (n = 8)
8.5 ± 0.9
Obese (n = 21)
8.5 ± 1.0
P values Overall
NW vs OW
NW vs OB
OW vs OB
0.263
0.846
0.373
1.000
Tanner stage
1.3 ± 0.7
1.5 ± 0.8
1.3 ± 0.5
0.777
1.000
1.000
1.000
Height (cm)
129.0 ± 8.1
134.5 ± 5.5
137.9 ± 6.9
0.001*
0.221
0.001*
0.821 0.007*
Weight (kg)
26.9 ± 5.6
35.8 ± 4.0
45.0 ± 8.7
0.000**
0.010*
0.000**
BMI (kg/m2)
16.0 ± 1.7
19.8 ± 1.2
23.5 ± 3.0
0.000**
0.001*
0.000**
0.001*
WC (cm)
58.0 ± 5.8
65.8 ± 4.7
78.3 ± 9.7
0.000**
0.053
0.000**
0.001*
TBF (%) TF (%)
27.6 ± 6.2 22.9 ± 6.1
35.8 ± 3.3 32.4 ± 4.7
40.7 ± 6.1 38.6 ± 6.7
0.000** 0.000**
0.004* 0.002*
0.000** 0.000**
0.150 0.061
SBP (mmHg)
100.3 ± 10.0
109.0 ± 10.2
111.1 ± 11.0
0.006*
0.160
0.005*
1.000
DBP (mmHg)
56.9 ± 7.1
58.0 ± 4.8
60.4 ± 5.4
0.179
1.000
0.205
1.000
Total PA (min/day)
740.5 ± 55.2
763.8 ± 58.2
733.9 ± 78.3
0.600
1.000
1.000
0.952
SEDPA (min/day)
314.0 ± 60.3
387.9 ± 58.0
342.0 ± 82.5
0.064
0.064
0.709
0.451
LIGPA (min/day)
360.3 ± 51.3
329.0 ± 35.6
343.4 ± 68.8
0.421
0.667
1.000
1.000
51.1 ± 16.9
33.0 ± 12.2
39.7 ± 14.8
0.018*
0.034*
0.098
1.000
MODPA (min/day) VIGPA (min/day)
15.1 ± 7.5
13.9 ± 11.6
8.8 ± 5.4
0.051
1.000
0.054
0.424
MVPA (min/day)
66.2 ± 22.5
47.0 ± 22.0
48.6 ± 18.2
0.026*
0.130
0.046*
1.000
Data are mean ± standard deviation P values reported from one-way analysis of variance (ANOVA) with Bonferroni adjustments for post hoc comparisons BMI body mass index, WC waist circumference, TBF total body fat, TF trunk fat, SBP systolic blood pressure, DBP diastolic blood pressure, Total PA total physical activity, SEDPA time in sedentary physical activity, LIGPA time in light physical activity, MODPA time in moderate physical activity, VIGPA time in vigorous physical activity, MVPA time in moderate to vigorous physical activity * P \ 0.05 for statistical significance **
P \ 0.001 for statistical significance
Statistical analysis
Results
Descriptive data are presented as mean and standard deviation (mean ± SD), unless noted differently. Distribution was checked for normality and skewness with the entire sample. LFnu and HFnu were transformed with the natural logarithm (LnLFnu; LnHFnu) and the LnLFnu/LnHFnu ratio calculated. Categorical variables were described as absolute and relative frequencies. Chi-square analysis was used to test for frequency differences between genders for weight status. T test for independent means and one-way analysis of variance (ANOVA) were used to explore differences between gender and BMI groups. General Linear Model (Analysis of Covariance—ANCOVA) was used to analyse differences between BMI groups in HRV parameters, with adjustments for covariates. ANCOVA was adjusted for age, Tanner stage, and MVPA. Multiple comparisons were carried out with Bonferroni’s post hoc adjustments. Partial correlations were used to analyse the relationships between waist circumference, body fat percentage, and HRV parameters, adjusting for the same variables used in the General Linear Model. All analyses were completed with a statistical software package (IBM SPSS 20.0) and significance level was set at 0.05.
Girls presented significantly higher values of total body fat and trunk fat compared to boys (Table 1). No significant differences were found between genders in any HRV parameter (Fig. 1). The proportions of boys and girls across BMI groups were similar (v2 = 0.015; P = 0.992). Eleven girls were classified as normal weight, 4 were overweight and 11 were obese. Ten of the boys were classified as normal weight, 4 were overweight, and 10 were obese. A summary of demographic variables and physical activity levels according to the weight status is presented in Table 2. Obese children presented higher waist circumference (78.3 ± 9.7 cm), total body fat (40.7 ± 6.1 %), and trunk fat (38.6 ± 6.7 %) compared to their normal weight and overweight counterparts. Obese children also showed significant higher systolic blood pressure compared to normal weight (111.1 ± 11 vs 100.3 ± 10 mmHg) but not different from the overweight children (109 ± 10.2 mmHg). Regarding the physical activity levels, normal weight children presented significantly higher levels of MODPA (51.1 ± 16.9 vs 39.7 ± 14.8 min/day) and MVPA (66.2 ± 22.5 vs 48.6 ± 18.2 min/day) compared to the obese children.
123
Clin Auton Res Table 3 General linear model testing for trends in heart rate variability parameters Variables
Normal weight (n = 21)
Over weight (n = 8)
Obese (n = 21)
Trend F
P values Partial g2
Overall
NW vs OW
NW vs OB
OW vs OB
ANCOVA with adjustments for age, Tanner stage, MVPA RR (ms) RMSSD (ms) SDRR (ms) LF (ms2) 2
725.2 ± 102.9
719.3 ± 101.1
796.2 ± 107.8
2.671
0.129
0.083
1.000
0.201
0.194
54.8 ± 28.9 61.7 ± 32
71.5 ± 28.3 81.4 ± 31.5
57.1 ± 30.2 82 ± 33.6
0.987 2.201
0.052 0.109
0.382 0.125
0.190 0.482
0.844 0.191
0.225 1.000
692.7 ± 79.3
656.1 ± 126.2
962.8 ± 83.5
3.486
0.162
0.041*
1.000
0.092
0.135
HF (ms )
1075.8 ± 102.6
1175.2 ± 163.4
781.8 ± 108
2.905
0.139
0.068
1.000
0.199
0.141
LFnu (%)
39.9 ± 11.8
33.8 ± 11.6
55.4 ± 12.4
13.967
0.437
0.000**
0.265
0.000*
0.000**
HFnu (%)
60.1 ± 11.8
65.8 ± 11.6
44.6 ± 12.4
13.596
0.430
0.000**
0.804
0.001*
0.000**
0.7 ± 0.4
0.5 ± 0.4
1.3 ± 0.4
21.367
0.543
0.000**
0.335
0.000**
0.000**
LF/HF LnLFnu (%)
3.7 ± 0.3
3.5 ± 0.3
4.0 ± 0.4
8.787
0.328
0.001*
0.212
0.005*
0.001*
LnHFnu (%)
4.1 ± 0.2
4.2 ± 0.2
3.8 ± 0.2
17.500
0.493
0.000**
0.329
0.000**
0.000**
LnLFnu/ LnHFnu
0.9 ± 0.1
0.8 ± 0.1
1.1 ± 0.1
13.582
0.430
0.000
0.254
0.000**
0.000**
Data are mean ± standard deviation P values reported from one-way analysis of variance (ANOVA) with Bonferroni adjustments for post hoc comparisons RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LF low frequency, HF high frequency, LFnu low-frequency normalized units, LnLFnu natural logarithm of low-frequency normalized units, HFnu high-frequency normalized units, LnHFnu natural logarithm of high-frequency normalized units, LF/HF low frequency to high frequency ratio * **
P \ 0.05 for statistical significance P \ 0.001 for statistical significance
Fig. 2 Partial correlation plot between LnHFnu and WC, controlling for age, Tanner stage, and MVPA (regression residuals of LnHFnu on the controlling variables vs regression residuals of waist circumference on the controlling variables). Notes: LnHFnu natural logarithm of high-frequency normalized units, WC waist circumference, MVPA moderate to vigorous physical activity
Table 3 shows the differences between groups in HRV parameters adjusting for age, Tanner stage, and MVPA. In HRV time-domain parameters, there were no significant differences found between groups. In frequency-domain HRV obese children showed a significant higher LnLFnu
123
(4.0 ± 0.4 vs 3.7 ± 0.3 %), higher LnLFnu/LnHFnu ratio (1.1 ± 0.1 vs 0.9 ± 0.1) and lower LnHFnu (3.8 ± 0.2 vs 4.1 ± 0.2 %) in comparison with their normal weight counterparts. No differences were found between the normal weight and overweight children in frequency-domain HRV. In regard to the association between fat mass content and HRV, using the same adjustment as in the group´s comparison, LnHFnu presented a significant negative correlation with waist circumference (r = -0.598; P = 0.000) (Fig. 2), total body fat (r = -0.409; P = 0.011) and trunk fat (r = -0.472; P = 0.003). The LnLFnu showed a significant positive correlation with waist circumference (r = 0.455; P = 0.004) and trunk fat (r = 0.370; P = 0.022). The LnLFnu/LnHFnu ratio presented positive correlations with waist circumference (r = 0.513, P = 0.001), total body fat (r = 0.357; P = 0.028) and trunk fat (r = 0.415; P = 0.010) (Table 4). No associations between body fat mass and time-domain HRV variables were found.
Discussion The main finding from the present study was that the correlation between body fat content and decreased vagal
Clin Auton Res Table 4 Correlation between waist circumference, fat mass and heart rate variability Heart rate variability parameters
WC (cm) r
TBF (%) P
r
TF (%) P
r
P
RR (ms)
0.182
0.267
0.061
0.717
0.068
SDRR (ms)
0.071
0.669
0.168
0.315
0.148
0.377
-0.131
0.428
0.081
0.629
0.047
0.780
RMSSD (ms)
0.685
LF (ms2)
0.339
HF (ms2)
-0.315
LFnu (%)
0.548
0.000**
0.380
0.019*
0.437
0.006*
HFnu (%) LF/HF
-0.548 0.633
0.000** 0.000**
-0.381 0.424
0.018* 0.008*
-0.439 0.493
0.006* 0.002*
LnLFnu(%) LnHFnu (%) LnLFnu/LnHFnu
0.455 -0.598 0.513
0.037* 0.054
0.004* 0.000** 0.001*
0.301
0.067
0.309
0.059
-0.119
0.475
-0.192
0.249
0.370
0.022*
-0.409
0.316
0.011*
0.054
-0.472
0.003*
0.357
0.028*
0.415
0.010*
Values in Pearson’s correlation coefficient adjusting for age, Tanner stage and moderate to vigorous physical activity RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LF low frequency, HF high frequency, LFnu low-frequency normalized units, LnLFnu natural logarithm of low-frequency normalized units, HFnu high-frequency normalized units, LnHFnu natural logarithm of high-frequency normalized units, LF/HF low frequency to high frequency ratio, WC waist circumference, TBF total body fat, TF trunk fat * P \ 0.05 for statistical significance ** P \ 0.001 for statistical significance
modulation in prepubescent children, accounting for the potential confounding effect of MVPA. Previous research showed that intense physical activity is associated with higher HRV vagal-related indexes in preadolescents [25]. In prepubescent lean children, a long-duration moderateintensity exercise program showed to have a positive effect on HRV [23]. However, high-intensity training is not proven to have an effect on HRV despite inducing a significant increase in aerobic fitness [36]. In our study, the accelerometer data showed no differences in either the total amount or the intensity of physical activity performed by boys and girls, which is unexpected since boys are usually more physically active [37]. This similar physical activity profile could partially explain the non-differences observed in HRV between genders, as MVPA has been included in the covariates, alongside with age, and Tanner stage. Previous research in children has shown that HRV is to some extent dependent on gender and age, increasing gradually with age, peaking between 6 and 9 years of age, then followed by a decrease or stagnation [38, 39]. Nevertheless, other studies reported no gender or age dependence at all [40, 41]. A more recent study [42] with a large cohort of young children has shown that in general, boys present higher HRV and at times age-related wave-like changes both in parasympathetic increases and sympathetic decreases. The same study suggests that physical fitness is a potent HRV determinant. In the BMI´ group analysis, obese children showed reduced MODPA and MVPA compared to their normal weight counterparts. This may be explained by
the negative correlation between the amount of time spent at moderate and vigorous physical activity and body fatness in children [37]. In terms of HRV, our results showed a distinctive relationship between lower parasympathetic modulation in frequency-domain HRV and weight status. Obese children presented with lower parasympathetic modulation and vagal withdrawal shown as a decreased LnHFnu, as similarly reported by the majority of the previous research [16–20]; although a recent study, using quantitative pupillography, showed lower parasympathetic modulation in obese children and adolescents that has not been observed in the LnHFnu [43]. It is difficult to interpret the higher LnLFnu values observed in our cohort of obese children since the LF component expresses not only the sympathetic modulation [44, 45] but also the parasympathetic modulation [46, 47]. This dual modulation of the LF also complicates the interpretation of the sympathovagal balance using the LF/HF ratio [47]. Previous research is inconsistent in terms of obese children having an increased or decreased LnLFnu. These conflicting results may arise from differences in cohort’ demographics; as highlighted previously, HRV in children is partially gender, age, and maturation stage dependent, and also highly influenced by physical activity and physical fitness levels [38, 39, 42]. The HRV in overweight children showed no differences in comparison to their normal weight counterparts, as observed by Kaufman et al. [16] in a sample of prepubescent and pubescent children. The predictable less time of exposure to obesity of the overweight children might explain
123
Clin Auton Res
this similarity with their normal weight counterparts [48]. Nevertheless, the small number of overweight children in our sample makes any kind of assumptions difficult. The correlation between body fat content and HRV was performed adjusting for the same covariates as in gender and BMI groups’ comparison. As seen in groups’ comparison, no associations between time-domain HRV and body fat content were found, similar to the findings of Kaufman et al. [16]. However, previous studies reported lower RMSSD values in pubescent obese children and adolescents [18, 48]. Our results suggest that time-domain measures might be less sensitive to detect obesity-related cardiac autonomic impairment in younger children, with possibly less time of exposure to obesity. In the frequencydomain, higher body fat mass, particularly trunk fat, was shown to be inversely correlated with vagal modulation, seen in the LnHFnu and positively correlated with LnLFnu and LnLFnu/LnHFnu ratio. Interestingly, the stronger correlations were observed with the waist circumference. This finding highlights the utility of this simple measurement in the risk assessment of autonomic impairment in young children. In a previous study [21], analogous significant correlations were found for DXA total body fat without adjustments, but not after adjusting for age, Tanner stage, insulin, and C-reactive protein levels. Gutin et al. [22], in a study with a large sample of adolescents, found that higher levels of visceral adipose tissue and subcutaneous adipose tissue were related to higher LFnu/HFnu ratio and were independent of race and sex. The relationship between central adiposity and cardiac autonomic dysfunction is complex and not clearly understood [49]. However, the overproduction of inflammatory cytokines by the fat cells is seen as the pathological trigger that leads to autonomic dysfunction [50]. The main limitation of this research was the small sample size, in particular the number of overweight children in the group, which did not allow us to run a more complex statistical analysis. In addition, we did not perform certain specialized autonomic tests such as the cold-pressor test, valsava maneuver, or deep-breathing. This is another important limitation since it might have provided more insight when analyzed in coordination with the resting HRV data. The lack of information about the time of exposure to obesity and the fact that metabolic parameters such as insulin resistance were not measured are also limitations in this study. The respiration rate was also not monitored during the HRV data collection so we cannot be completely sure that all of the participants maintained the defined 14–16 breaths/min during the procedure. Therefore, we cannot exclude a potential interference of respiration on vagal modulation. However, we did make every effort to ensure that all of
123
the children were still, relaxed, and breathing at the established frequency.
Conclusion Controlling for the possible influence of physical activity, this study shows a clear relationship between frequencydomain HRV impairment and obesity in prepubescent children. Despite the young age of the cohort, a higher amount of body fat mass, predominantly central fat, was related to decreased parasympathetic modulation. These findings highlight the potential cardiovascular risk that higher levels of body fat mass may represent even at a very young age. Further research is needed to clarify the obesity related autonomic dysfunction pathophysiology and the autonomic benefits of different types of exercise programs in young children.
References 1. Wang Y, Lobstein T (2006) Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes 1(1):11–25 2. de Onis M, Blossner M (2000) Prevalence and trends of overweight among preschool children in developing countries. Am J Clin Nut 72(4):1032–1039 3. OECD (2012) Obesity update 2012. OECD Health Division. http:// www.oecd.org/health/49716427.pdf. Accessed 30 July 2013 4. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS (1999) The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 103(6 Pt1):1175–1182 5. Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, Steer C, Sherriff A (2005) Early life risk factors for obesity in childhood: cohort study. BMJ 330(7504):1357. doi:10.1136/bmj. 38470.670903.E0 6. Falaschetti E, Hingorani AD, Jones A, Charakida M, Finer N, Whincup P, Lawlor DA, Davey Smith G, Sattar N, Deanfield JE (2010) Adiposity and cardiovascular risk factors in a large contemporary population of pre-pubertal children. Eur Heart J 31(24):3063–3072. doi:10.1093/eurheartj/ehq355 7. Engeland A, Bjorge T, Sogaard AJ, Tverdal A (2003) Body mass index in adolescence in relation to total mortality: 32-year followup of 227,000 Norwegian boys and girls. Am J Epidemiol 157(6):517–523 8. Must A, Strauss RS (1999) Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord 23(Suppl 2):S2–S11 9. Power C, Lake JK, Cole TJ (1997) Measurement and long-term health risks of child and adolescent fatness. Int J Obes Relat Metab Disord 21(7):507–526 10. Despres JP, Lemieux I, Prud’homme D (2001) Treatment of obesity: need to focus on high risk abdominally obese patients. BMJ 322(7288):716–720 11. Teixeira PJ, Sardinha LB, Going SB, Lohman TG (2001) Total and regional fat and serum cardiovascular disease risk factors in lean and obese children and adolescents. Obes Res 9(8):432–442. doi:10.1038/oby.2001.57
Clin Auton Res 12. Dietrich A, Rosmalen JG, Althaus M, van Roon AM, Mulder LJ, Minderaa RB, Oldehinkel AJ, Riese H (2010) Reproducibility of heart rate variability and baroreflex sensitivity measurements in children. Biol Psychol 85(1):71–78. doi:10.1016/j.biopsycho. 2010.05.005 13. Massin M, von Bernuth G (1998) Clinical and haemodynamic correlates of heart rate variability in children with congenital heart disease. Eur J Pediatr 157(12):967–971 14. Biswas AK, Scott WA, Sommerauer JF, Luckett PM (2000) Heart rate variability after acute traumatic brain injury in children. Crit Care Med 28(12):3907–3912 15. Javorka M, Javorkova J, Tonhajzerova I, Javorka K (2005) Parasympathetic versus sympathetic control of the cardiovascular system in young patients with type 1 diabetes mellitus. Clin Physiol Funct Imaging 25(5):270–274 CPF623 16. Kaufman CL, Kaiser DR, Steinberger J, Kelly AS, Dengel DR (2007) Relationships of cardiac autonomic function with metabolic abnormalities in childhood obesity. Obesity 15(5):1164–1171. doi:10.1038/oby.2007.619 17. Latchman PL, Mathur M, Bartels MN, Axtell RS, De Meersman RE (2011) Impaired autonomic function in normotensive obese children. Clin Auton Res 21(5):319–323. doi:10.1007/s10286011-0116-8 18. Martini G, Riva P, Rabbia F, Molini V, Ferrero GB, Cerutti F, Carra R, Veglio F (2001) Heart rate variability in childhood obesity. Clin Auton Res 11(2):87–91 19. Nagai N, Matsumoto T, Kita H, Moritani T (2003) Autonomic nervous system activity and the state and development of obesity in Japanese school children. Obes Res 11(1):25–32. doi:10.1038/ oby.2003.6 20. Tascilar ME, Yokusoglu M, Boyraz M, Baysan O, Koz C, Dundaroz R (2011) Cardiac autonomic functions in obese children. J Clin Res Ped Endo 3(2):60–64. doi:10.4274/jcrpe.v3i2. 131 21. Kaufman CL, Kaiser DR, Steinberger J, Dengel DR (2007) Relationships between heart rate variability, vascular function, and adiposity in children. Clin Auton Res 3:165–171. doi:10.1007/ s10286-007-0411-6 22. Gutin B, Howe C, Johnson MH, Humphries MC, Snieder H, Barbeau P (2005) Heart rate variability in adolescents: relations to physical activity, fitness, and adiposity. Med Sci Sports Exer 37(11):1856–1863 doi:00005768-200511000-00005 23. Nagai N, Hamada T, Kimura T, Moritani T (2004) Moderate physical exercise increases cardiac autonomic nervous system activity in children with low heart rate variability. Childs Nerv Syst 20(4):209–214. doi:10.1007/s00381-004-0915-5 24. Mandigout S, Melin A, Fauchier L, N’Guyen LD, Courteix D, Obert P (2002) Physical training increases heart rate variability in healthy prepubertal children. Eur J Clin Invest 32(7):479–487 25. Buchheit M, Platat C, Oujaa M, Simon C (2007) Habitual physical activity, physical fitness and heart rate variability in preadolescents. Int J Sport Med 28(3):204–210. doi:10.1055/s2006-924296 26. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320(7244):1240–1243 27. Kuczmarski RJ (1996) Bioelectrical impedance analysis measurements as part of a national nutrition survey. Am J Clin Nut 64(3 Suppl):453S–458S 28. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents (2004) The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents: part 1. Pediatrics 114(2 Suppl 4):555–576 29. Tanner JM (1962) Growth at adolescence. Charles C Thomas, Springfield
30. Mota J, Guerra S, Leandro C, Pinto A, Ribeiro JC, Duarte JA (2002) Association of maturation, sex, and body fat in cardiorespiratory fitness. Am J Hum Biol 14(6):707–712. doi:10. 1002/ajhb.10086 31. Trost SG, McIver KL, Pate RR (2005) Conducting accelerometer-based activity assessments in field-based research. Med Sci Sport Exer 37(11 Suppl):S531–S543 32. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG (2008) Calibration of two objective measures of physical activity for children. J Sports Sci 26(14):1557–1565. doi:10.1080/ 02640410802334196 33. Trost SG, Loprinzi PD, Moore R, Pfeiffer KA (2011) Comparison of accelerometer cut points for predicting activity intensity in youth. Med Sci Sport Exer 43(7):1360–1368. doi:10.1249/MSS. 0b013e318206476e 34. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996) Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 93(5):1043–1065 35. Gamelin FX, Baquet G, Berthoin S, Bosquet L (2008) Validity of the polar S810 to measure R-R intervals in children. Int J Sport Med 29(2):134–138. doi:10.1055/s-2007-964995 36. Gamelin FX, Baquet G, Berthoin S et al (2009) Effect of high intensity intermittent training on heart rate variability in prepubescent children. Eur J Appl Physiol 105(5):731–738. doi:10. 1007/s00421-008-0955-8 37. Ekelund U, Sardinha LB, Anderssen SA, Harro M, Franks PW, Brage S, Cooper AR, Andersen LB, Riddoch C, Froberg K (2004) Associations between objectively assessed physical activity and indicators of body fatness in 9- to 10-y-old European children: a population-based study from 4 distinct regions in Europe (the European Youth Heart Study). Am J Clin Nut 80(3):584–590 38. Silvetti MS, Drago F, Ragonese P (2001) Heart rate variability in healthy children and adolescents is partially related to age and gender. Int J Cardiol 81(2–3):169–174 [pii]: S016752730100537X 39. Galeev AR, Igisheva LN, Kazin EM (2002) Heart rate variability in healthy six- to sixteen-year-old children. Hum Physiol 28(4):428–432. doi:10.1023/A:1016529931519 40. Fukuba Y, Sato H, Sakiyama T, Yamaoka Endo M, Yamada M, Ueoka H, Miura A, Koga S et al (2009) Autonomic nervous activities assessed by heart rate variability in pre- and postadolescent Japanese. J Physiol Anthropol 28(6):269–273. doi:10. 2114/jpa2.28.269 41. Faulkner MS, Hathaway D, Tolley B (2003) Cardiovascular autonomic function in healthy adolescents. Heart Lung 32(1):10–22. doi:10.1067/mhl.2003.6 42. Michels N, Clays E, De Buyzere M, Huybrechts I, Marild S, Vanaelst B, De Henauw S, Sioen I (2013) Determinants and reference values of short-term heart rate variability in children. Eur J Appl Physiol 113(6):1477–1488. doi:10.1007/s00421-0122572-9 43. Baum P, Petroff D, Classen J, Kiess W, Bluher S (2013) Dysfunction of autonomic nervous system in childhood obesity: a cross-sectional study. PLoS One 8(1):e54546. doi:10.1371/jour nal.pone.0054546 44. Malliani A, Pagani M, Lombardi F (1994) Importance of appropriate spectral methodology to assess heart rate variability in the frequency domain. Hypertension 24(1):140–142 45. Montano N, Porta A, Cogliati C et al (2009) Heart rate variability explored in the frequency domain: a tool to investigate the link between heart and behavior. Neurosci Biobehav Rev 33(2):71–80. doi:10.1016/j.neubiorev.2008.07.006 46. Eckberg DL (2007) Sympathovagal balance: a critical appraisal. Circulation 96(9):3224–3232
123
Clin Auton Res 47. Reyes del Paso GA, Langewitz W, Mulder LJ, van Roon A, Duschek S (2013) The utility of low frequency heart rate variability as an index of sympathetic cardiac tone: a review with emphasis on a reanalysis of previous studies. Psychophysiology 50(5):477–487. doi:10.1111/psyp.12027 48. Rabbia F, Silke B, Conterno A, Grosso T, De Vito B, Rabbone I, Chiandussi L, Veglio F (2003) Assessment of cardiac autonomic modulation during adolescent obesity. Obes Res 11(4):541–548. doi:10.1038/oby.2003.76
123
49. Tentolouris N, Argyrakopoulou G, Katsilambros N (2008) Perturbed autonomic nervous system function in metabolic syndrome. Neuromolecular Med 10(3):169–178. doi:10.1007/ s12017-008-8022-5 50. Grassi G, Dell’Oro R, Facchini A, Quarti Trevano F, Bolla GB, Mancia G (2004) Effect of central and peripheral body fat distribution on sympathetic and baroreflex function in obese normotensives. J Hypertens 22(12):2363–2369 doi:00004872200412000-00019
RESEARCH ARTICLE
Heart rate variability, adiposity, and physical activity in prepubescent children Andre Filipe Santos-Magalhaes • Luisa Aires • Clarice Martins • Gustavo Silva • Ana Maria Teixeira Jorge Mota • Luis Rama
•
Received: 23 August 2014 / Accepted: 4 December 2014 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study aimed at examining the associations between weight status, body fat mass, and heart rate variability in prepubescent children, adjusting for physical activity levels. Methods A cross-sectional investigation in which a total of 50 Caucasian pre-pubertal children (21 normal weight; 8 overweight; 21 obese), aged 6–10 years (8.33 ± 1.14), including both boys (n = 24) and girls (n = 26), were
A. F. Santos-Magalhaes (&) School of Health, University Fernando Pessoa, Porto, Portugal e-mail: [email protected] A. F. Santos-Magalhaes Health and Exercise Research Group, School of Sport and Exercise Sciences, University of Kent, Chatham, Kent, UK e-mail: [email protected] L. Aires C. Martins G. Silva J. Mota Research Centre in Physical Education Activity, Health and Leisure Time, Faculty of Sports – CIAFEL, University of Porto, Porto, Portugal e-mail: [email protected] C. Martins e-mail: [email protected] G. Silva e-mail: [email protected] J. Mota e-mail: [email protected] A. M. Teixeira L. Rama Research Centre for Sport and Physical Activity, Faculty of Sport Sciences and Physical Education, University of Coimbra, Coimbra, Portugal e-mail: [email protected] L. Rama e-mail: [email protected]
recruited from local schools. Total body fat and trunk fat were evaluated through dual-energy X-ray absorptiometry. Free-living physical activity levels were evaluated by accelerometer. Short-term heart rate variability acquisition was performed; time- and frequency-domain parameters were analysed. Logarithmic transformations of the lowfrequency (LnLFnu), high-frequency (LnHFnu) normalized units and low-frequency/high-frequency (LnLFnu/HFnu) ratio were computed. Results Adjusting for age, Tanner stage, and moderate to vigorous physical activity levels, obese children compared to normal weight children showed a significant decreased LnHfnu (3.8 ± 0.2 vs 4.1 ± 0.2 %) and both higher LnLFnu (4.0 ± 0.4 vs 3.7 ± 0.3 %) and LnLFnu/LnHFnu ratio (1.1 ± 0.1 vs 0.9 ± 0.1). LnHFnu showed significant negative correlation with waist circumference (r = -0.598; P = 0.000), total body fat (r = -0.409; P = 0.011) and trunk fat (r = -0.472; P = 0.003). Both LnLFnu and LnLFnu/LnHFnu ratio showed positive correlations with waist circumference (r = 0.455; r = 0.513) and trunk fat (r = 0.370; r = 0.415). Conclusions A higher amount of body fat mass, particularly central fat, was shown to be related to decreased parasympathetic modulation in time-domain heart rate variability. This finding highlights the potential cardiovascular risk that excessive fat mass may represent even at very young age. Keywords Heart rate variability Children Fat mass DXA Physical activity
Introduction The prevalence of obesity in children in the past two decades has rapidly increased, not only in developed
123
Clin Auton Res
countries but also in the developing world [1, 2]. Despite stabilizing and even decreasing prevalence rates in some countries, obesity is still a problem in most places [3]. Hypertension, dyslipidemia, insulin resistance, and type 2 diabetes are becoming more frequent in overweight and obese children [4], even among the very young children [5, 6]. The early onset of these co-morbidities is associated with both increased mortality and increased morbidity in adulthood [7–9]. It has been shown that these disorders have a greater relationship to fat distribution rather than to total weight [10, 11]. Obesity in children is also related to impairment of cardiovascular autonomic function, which reflects an impairment of the autonomic nervous system. Heart rate variability (HRV) is a noninvasive and reliable method to quantify the autonomic nervous system modulation of cardiovascular function, both in adults and in the pediatric population [12]. A decreased HRV in childhood is associated with several pediatric conditions such as congenital heart disease [13], acute traumatic brain injuries [14], diabetes mellitus [15] and obesity. Several studies have shown that obese children and adolescents have a decreased HRV, manifested by an altered sympathovagal balance, mainly due to a decrease in vagal modulation [16–20]. However, a vast majority of these investigations were conducted in pubescent children and adolescents and have used only indirect measures of obesity such as the BMI cut-off points. To date, very few studies have investigated the relationship between the quantity and distribution of fat mass as measured directly with dual-energy X-ray absorptiometry (DXA) and HRV [21, 22]. In a cohort of pre-pubescent and pubescent children Kaufman et al. [21] found an association between total body fat and decreased HRV vagal modulation, although the possible confounding effect of physical activity on HRV was not acknowledged, as in most of the studies done on obese children. In adolescents, favorable HRV profiles have been shown to be associated with higher amounts of moderate and vigorous physical activity, even after controlling for the body fat mass [22]. Moreover, exercise training programs have been shown to induce a positive effect on HRV vagal modulation in healthy children [23– 25]. To our knowledge, no data are available about the interaction between adiposity, HRV, and daily physical activity levels in prepubescent children. Therefore, the purpose of this study was to examine the relationship between weight status, body fat content as measured by dual-energy X-ray absorptiometry (DXA), and HRV in pre-pubertal children, adjusting for their physical activity levels. We hypothesized that even in prepubescent children there is an inverse relationship between body fat, particularly trunk fat, and parasympathetic HRV modulation.
123
Methods Participants This study was conducted as part of a school-based research program carried out in five public primary schools. A total of 50 Caucasian pre-pubertal children aged 6–10 years (8.33 ± 1.14), both genders (24 boys and 26 girls), were recruited from local schools. According to the body mass index (BMI) percentiles for age and gender, 21 children were categorized as normal weight, 8 were overweight, and 21 were obese. The Regional Education Board approved the study, and written informed consent from participants’ parents/guardians was obtained. The informed consent including the purpose of the research and the risks and benefits of participation were fully explained to each child and their respective parent(s), consistent with the Helsinki Declaration. The evaluation methods and procedures were reviewed and approved by the Scientific Board of the Faculty of Sport Sciences and Physical Education from the University of Coimbra. Medical history was ascertained via questionnaire and consultation with a medical doctor. Any children with abnormal cardiovascular function and/or abnormal cardiac electrophysiology and/or taking medication at the time were excluded. Procedures The evaluations were performed in the morning, between 8:00 am and 10:00 am, in the faculty laboratories. The same trained technician performed all of the evaluations to reduce potential measurement errors between observers. Children were asked to avoid strenuous exercise in the 24-h prior to the evaluation. They were also asked to do a 12-h fast and were only allowed to drink water after dinner on the day before the evaluation. After the anthropometric measurements, body composition, and HRV evaluation procedures, children were allowed to have breakfast and received accelerometers for physical activity assessment, together with instructions for use. Anthropometry and body composition Height was measured to the nearest 0.1 cm in bare or stocking feet with the children standing upright against a Holtain Stadiometer. Weight was measured to the nearest 0.1 kg, with the children lightly dressed using an electronic weight scale (Seca 708 portable digital beam scale). BMI was calculated from the ratio of body weight (kg)/body height (m2). BMI percentiles for age and gender were calculated and the cut-off points established by the International Obesity Task Force [26] were applied to categorize children as normal weight (BMI \ 85th percentile),
Clin Auton Res
overweight (85th percentile B BMI \ 95th percentile) or obese (BMI C 95th percentile). Waist circumference was evaluated following the NHANES protocol [27] using a circumference measuring tape (Seca 203). Percentage of total body fat and trunk fat was obtained through DXA (Explorer QDR 4500, Hologic, Bedford, MA) with whole body protocol. This assessment procedure lasted 20 min. Children were placed in the supine position with their arms in extension near the trunk and lower limbs in extension, with a slight abduction of the feet. Children removed clothes and all metallic objects (earrings, watches, etc.) and wore a gown. Blood pressure Blood pressure was measured with Colin Press Mate NonInvasive Blood Pressure Monitor (Model BP 8800p, Colin Medical Instruments Corporation, San Antonio, TX, USA)]. Systolic blood pressure and diastolic blood pressure were measured in the right arm with the subjects in the fasting state. The participants were in the sitting position (without their legs crossed), with the right arm at heart level. Three standard pressure cuffs of correct size (9 9 18, 12 9 23, 14 9 28 cm) were used according to the published guidelines for blood pressure assessment in children [28]. The first and second measurements were taken after 5 min and 10 min resting, considering the mean
of these measurements for statistical purposes. If these two measurements differed by 2 mmHg, the protocol was repeated to obtain two new measurements that could not exceed a 2 mmHg difference. Maturational stage Each child self-assessed his/her own stages of secondary sex characteristics: stage of breast development in females, genital development in males, and pubic hair development for both genders [29]. Only self-assessed pubic hair development was used in the study and that is considered to be a valid and reliable indicator of maturational status [30]. Mota el al. [30] found a high correlation (r = 0.73) between these ratings on two separate occasions (3-day interval) in a sample of 50 Portuguese youths. Physical activity Physical activity was objectively assessed by accelerometers (Actigraph GT3X, Pensacola, FL, USA) for seven consecutive days to obtain a reliable picture of the usual physical activity of the study participants [31]. Data were recorded in 15 s sampling periods (epochs). The participants used the accelerometer in an elastic waistband with the device positioned on the right hip during the daytime, except while sleeping, bathing, and during other
Table 1 Descriptive characteristics according to gender Variables
Boys (n = 26)
Girls (n = 24)
Total sample (n = 50)
P value boys vs girls
Age (years)
8.2 ± 1.2
8.4 ± 1.0
8.3 ± 1.1
0.386
Tanner stage
1.3 ± 0.6
1.5 ± 0.6
1.4 ± 0.6
0.215
Height (cm)
133.5 ± 9.7
133.6 ± 6.8
133.5 ± 8.3
Weight (kg)
35.8 ± 12.8
36.0 ± 8.7
35.9 ± 10.8
0.975 0.948
BMI (kg/m2)
19.5 ± 4.7
20.0 ± 3.5
19.7 ± 4.1
0.702
WC (cm)
67.2 ± 14.3
68.3 ± 9.7
67.7 ± 12.0
0.755
TBF (%)
31.4 ± 9.2
37.3 ± 6.4
34.5 ± 8.3
0.011*
TF (%) SBP (mmHg) DBP (mmHg)
27.6 ± 10.1 104.7 ± 8.5
34.3 ± 7.8
31.2 ± 9.5
0.012*
108.0 ± 13.5
106.4 ± 11.5
0.326
58.0 ± 4.4
59.1 ± 7.5
58.6 ± 6.2
0.557
Total PA (min/day) SEDPA (min/day)
726.2 ± 74.7 325.6 ± 79.7
755.2 ± 53.4 347.1 ± 66.5
741.7 ± 75.1 337.1 ± 72.9
0.148 0.342
LIGPA (min/day)
339.9 ± 61.4
356.1 ± 53.2
348.5 ± 57.0
0.360
MODPA (min/day)
47.7 ± 20.1
40.2 ± 12.3
43.7 ± 16.6
0.140
VIGPA (min/day)
13.0 ± 9.7
11.9 ± 6.2
12.4 ± 7.9
0.649
MVPA (min/day)
60.7 ± 27.2
52.1 ± 16.5
56.1 ± 22.3
0.207
Data are mean ± standard deviation P values reported from T test for independent means BMI body mass index, WC waist circumference, TBF total body fat, TF trunk fat, SBP systolic blood pressure, DBP diastolic blood pressure, Total PA total physical activity, SEDPA time in sedentary physical activity, LIGPA time in light physical activity, MODPA time in moderate physical activity, VIGPA time in vigorous physical activity, MVPA time in moderate to vigorous physical activity * P \ 0.05 for statistical significance
123
Clin Auton Res
count intervals [32] were considered: 0–100 for SEDPA, 101–2295 for LIGPA, 2296–4011 for MODPA, C4012 for VIGPA and C2296 for MVPA. These cut-off points seem to provide acceptable classification accuracy for all four levels of physical activity intensity and perform well among children of all ages [33]. Heart rate variability
Fig. 1 Gender differences in time- (a) and frequency-domain (b) heart rate variability. General linear model adjusted for age, Tanner stage and time in moderate to vigorous physical activity. Notes: RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LnLFnu natural logarithm of low-frequency normalized units, LnFHnu natural logarithm of high-frequency normalized units
aquatic activities. A data sheet was given to the participants, who were instructed to record the time when the monitor was attached in the morning, detached in the evening, and when it was removed for a bath. The standard software PROPERO (developed by the Centre of Research in Childhood Health of the University of Southern Denmark, 2011) was used to reduce the raw activity data from the accelerometers into daily physical activity. Time periods with at least 10 consecutive minutes of zero counts recorded were excluded from analysis assuming that the monitor was not worn. A minimum recording of 8 h/day was the criteria to accept daily physical activity data as valid. Individual subject data were only accepted for analysis if recordings of least 2 weekdays and 1 weekend day were successfully assessed. The main outcomes of reduced data were total physical activity (counts/min/day), time in sedentary physical activity [SEDPA (min/day)], light physical activity [LIGPA (min/day)], moderate physical activity [MODPA (min/day)], vigorous physical activity [VIGPA (min/day)] and moderate to vigorous physical activity [MVPA (min/day)]. To determine the time spent in physical activity of different intensities, the following
123
The HRV acquisition and analysis were conducted according to the international recommendations [34]. Prior to the HRV acquisition the proceedings were explained to the children, emphasizing the importance of staying still and relaxed during the procedure. The HRV acquisition was done in a quiet room with controlled temperature (21–23 °C) and humidity (40–60 %). To avoid any interference in the signal, cell phones or other electromagnetic devices were not allowed in the room during the heart rate recording. The equipment used was a Polar S810 (Polar Electro OY, Kempele, Finland) heart rate monitor receiver and strap, which has previously been validated for the HRV assessment in children [35]. The heart rate monitor strap was placed on each participant’s chest over the lower third of the sternum and the heart rate receiver was placed on the wrist. HRV was recorded for 15 min while participants were lying in the supine position. They were also asked to breathe at a constant respiratory rate of 0.25 Hz (14–16 breaths/min) with the help of a metronome. Importantly, prior to HRV assessment, the children were given a brief training on how to breathe at a constant rate. Only the last 5 min of the heart rate recording were utilized for the HRV analysis. The software Polar Performance SW was used to transfer data from the monitors and to analyze and correct for ectopic and missing beats. The HRV analysis was done with the Kubios HRV Analysis Software (Biosignal Analysis and Medical Image Group, Department of Physics, University of Kuopio, Finland). The successive interbeat (R–R) intervals, the standard deviation of normal R– R intervals (SDRR) and the square root of the mean of the squares of successive R–R interval differences (RMSSD) were computed as time-domain measures. The frequency domain analysis was performed through fast Fourier transformation algorithm, allowing the computation of the three-frequency bands: very low frequency (VLF 0–0.04 Hz), low frequency (LF 0.04–0.15 Hz), and high frequency (HF 0.15–0.4 Hz). However, only LF and HF were considered in our analysis, since the VLF component is not as reliable because of its lack of physiological meaning and interpretation. The normalized units of the LF (LFnu), HF (HFnu), and respective ratio (LFnu/HFnu) were computed to avoid the influence of the total power within the power spectrum [34].
Clin Auton Res Table 2 Descriptive characteristics according to weight status Variables
Age (years)
Normal weight (n = 21)
8.0 ± 1.2
Over weight (n = 8)
8.5 ± 0.9
Obese (n = 21)
8.5 ± 1.0
P values Overall
NW vs OW
NW vs OB
OW vs OB
0.263
0.846
0.373
1.000
Tanner stage
1.3 ± 0.7
1.5 ± 0.8
1.3 ± 0.5
0.777
1.000
1.000
1.000
Height (cm)
129.0 ± 8.1
134.5 ± 5.5
137.9 ± 6.9
0.001*
0.221
0.001*
0.821 0.007*
Weight (kg)
26.9 ± 5.6
35.8 ± 4.0
45.0 ± 8.7
0.000**
0.010*
0.000**
BMI (kg/m2)
16.0 ± 1.7
19.8 ± 1.2
23.5 ± 3.0
0.000**
0.001*
0.000**
0.001*
WC (cm)
58.0 ± 5.8
65.8 ± 4.7
78.3 ± 9.7
0.000**
0.053
0.000**
0.001*
TBF (%) TF (%)
27.6 ± 6.2 22.9 ± 6.1
35.8 ± 3.3 32.4 ± 4.7
40.7 ± 6.1 38.6 ± 6.7
0.000** 0.000**
0.004* 0.002*
0.000** 0.000**
0.150 0.061
SBP (mmHg)
100.3 ± 10.0
109.0 ± 10.2
111.1 ± 11.0
0.006*
0.160
0.005*
1.000
DBP (mmHg)
56.9 ± 7.1
58.0 ± 4.8
60.4 ± 5.4
0.179
1.000
0.205
1.000
Total PA (min/day)
740.5 ± 55.2
763.8 ± 58.2
733.9 ± 78.3
0.600
1.000
1.000
0.952
SEDPA (min/day)
314.0 ± 60.3
387.9 ± 58.0
342.0 ± 82.5
0.064
0.064
0.709
0.451
LIGPA (min/day)
360.3 ± 51.3
329.0 ± 35.6
343.4 ± 68.8
0.421
0.667
1.000
1.000
51.1 ± 16.9
33.0 ± 12.2
39.7 ± 14.8
0.018*
0.034*
0.098
1.000
MODPA (min/day) VIGPA (min/day)
15.1 ± 7.5
13.9 ± 11.6
8.8 ± 5.4
0.051
1.000
0.054
0.424
MVPA (min/day)
66.2 ± 22.5
47.0 ± 22.0
48.6 ± 18.2
0.026*
0.130
0.046*
1.000
Data are mean ± standard deviation P values reported from one-way analysis of variance (ANOVA) with Bonferroni adjustments for post hoc comparisons BMI body mass index, WC waist circumference, TBF total body fat, TF trunk fat, SBP systolic blood pressure, DBP diastolic blood pressure, Total PA total physical activity, SEDPA time in sedentary physical activity, LIGPA time in light physical activity, MODPA time in moderate physical activity, VIGPA time in vigorous physical activity, MVPA time in moderate to vigorous physical activity * P \ 0.05 for statistical significance **
P \ 0.001 for statistical significance
Statistical analysis
Results
Descriptive data are presented as mean and standard deviation (mean ± SD), unless noted differently. Distribution was checked for normality and skewness with the entire sample. LFnu and HFnu were transformed with the natural logarithm (LnLFnu; LnHFnu) and the LnLFnu/LnHFnu ratio calculated. Categorical variables were described as absolute and relative frequencies. Chi-square analysis was used to test for frequency differences between genders for weight status. T test for independent means and one-way analysis of variance (ANOVA) were used to explore differences between gender and BMI groups. General Linear Model (Analysis of Covariance—ANCOVA) was used to analyse differences between BMI groups in HRV parameters, with adjustments for covariates. ANCOVA was adjusted for age, Tanner stage, and MVPA. Multiple comparisons were carried out with Bonferroni’s post hoc adjustments. Partial correlations were used to analyse the relationships between waist circumference, body fat percentage, and HRV parameters, adjusting for the same variables used in the General Linear Model. All analyses were completed with a statistical software package (IBM SPSS 20.0) and significance level was set at 0.05.
Girls presented significantly higher values of total body fat and trunk fat compared to boys (Table 1). No significant differences were found between genders in any HRV parameter (Fig. 1). The proportions of boys and girls across BMI groups were similar (v2 = 0.015; P = 0.992). Eleven girls were classified as normal weight, 4 were overweight and 11 were obese. Ten of the boys were classified as normal weight, 4 were overweight, and 10 were obese. A summary of demographic variables and physical activity levels according to the weight status is presented in Table 2. Obese children presented higher waist circumference (78.3 ± 9.7 cm), total body fat (40.7 ± 6.1 %), and trunk fat (38.6 ± 6.7 %) compared to their normal weight and overweight counterparts. Obese children also showed significant higher systolic blood pressure compared to normal weight (111.1 ± 11 vs 100.3 ± 10 mmHg) but not different from the overweight children (109 ± 10.2 mmHg). Regarding the physical activity levels, normal weight children presented significantly higher levels of MODPA (51.1 ± 16.9 vs 39.7 ± 14.8 min/day) and MVPA (66.2 ± 22.5 vs 48.6 ± 18.2 min/day) compared to the obese children.
123
Clin Auton Res Table 3 General linear model testing for trends in heart rate variability parameters Variables
Normal weight (n = 21)
Over weight (n = 8)
Obese (n = 21)
Trend F
P values Partial g2
Overall
NW vs OW
NW vs OB
OW vs OB
ANCOVA with adjustments for age, Tanner stage, MVPA RR (ms) RMSSD (ms) SDRR (ms) LF (ms2) 2
725.2 ± 102.9
719.3 ± 101.1
796.2 ± 107.8
2.671
0.129
0.083
1.000
0.201
0.194
54.8 ± 28.9 61.7 ± 32
71.5 ± 28.3 81.4 ± 31.5
57.1 ± 30.2 82 ± 33.6
0.987 2.201
0.052 0.109
0.382 0.125
0.190 0.482
0.844 0.191
0.225 1.000
692.7 ± 79.3
656.1 ± 126.2
962.8 ± 83.5
3.486
0.162
0.041*
1.000
0.092
0.135
HF (ms )
1075.8 ± 102.6
1175.2 ± 163.4
781.8 ± 108
2.905
0.139
0.068
1.000
0.199
0.141
LFnu (%)
39.9 ± 11.8
33.8 ± 11.6
55.4 ± 12.4
13.967
0.437
0.000**
0.265
0.000*
0.000**
HFnu (%)
60.1 ± 11.8
65.8 ± 11.6
44.6 ± 12.4
13.596
0.430
0.000**
0.804
0.001*
0.000**
0.7 ± 0.4
0.5 ± 0.4
1.3 ± 0.4
21.367
0.543
0.000**
0.335
0.000**
0.000**
LF/HF LnLFnu (%)
3.7 ± 0.3
3.5 ± 0.3
4.0 ± 0.4
8.787
0.328
0.001*
0.212
0.005*
0.001*
LnHFnu (%)
4.1 ± 0.2
4.2 ± 0.2
3.8 ± 0.2
17.500
0.493
0.000**
0.329
0.000**
0.000**
LnLFnu/ LnHFnu
0.9 ± 0.1
0.8 ± 0.1
1.1 ± 0.1
13.582
0.430
0.000
0.254
0.000**
0.000**
Data are mean ± standard deviation P values reported from one-way analysis of variance (ANOVA) with Bonferroni adjustments for post hoc comparisons RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LF low frequency, HF high frequency, LFnu low-frequency normalized units, LnLFnu natural logarithm of low-frequency normalized units, HFnu high-frequency normalized units, LnHFnu natural logarithm of high-frequency normalized units, LF/HF low frequency to high frequency ratio * **
P \ 0.05 for statistical significance P \ 0.001 for statistical significance
Fig. 2 Partial correlation plot between LnHFnu and WC, controlling for age, Tanner stage, and MVPA (regression residuals of LnHFnu on the controlling variables vs regression residuals of waist circumference on the controlling variables). Notes: LnHFnu natural logarithm of high-frequency normalized units, WC waist circumference, MVPA moderate to vigorous physical activity
Table 3 shows the differences between groups in HRV parameters adjusting for age, Tanner stage, and MVPA. In HRV time-domain parameters, there were no significant differences found between groups. In frequency-domain HRV obese children showed a significant higher LnLFnu
123
(4.0 ± 0.4 vs 3.7 ± 0.3 %), higher LnLFnu/LnHFnu ratio (1.1 ± 0.1 vs 0.9 ± 0.1) and lower LnHFnu (3.8 ± 0.2 vs 4.1 ± 0.2 %) in comparison with their normal weight counterparts. No differences were found between the normal weight and overweight children in frequency-domain HRV. In regard to the association between fat mass content and HRV, using the same adjustment as in the group´s comparison, LnHFnu presented a significant negative correlation with waist circumference (r = -0.598; P = 0.000) (Fig. 2), total body fat (r = -0.409; P = 0.011) and trunk fat (r = -0.472; P = 0.003). The LnLFnu showed a significant positive correlation with waist circumference (r = 0.455; P = 0.004) and trunk fat (r = 0.370; P = 0.022). The LnLFnu/LnHFnu ratio presented positive correlations with waist circumference (r = 0.513, P = 0.001), total body fat (r = 0.357; P = 0.028) and trunk fat (r = 0.415; P = 0.010) (Table 4). No associations between body fat mass and time-domain HRV variables were found.
Discussion The main finding from the present study was that the correlation between body fat content and decreased vagal
Clin Auton Res Table 4 Correlation between waist circumference, fat mass and heart rate variability Heart rate variability parameters
WC (cm) r
TBF (%) P
r
TF (%) P
r
P
RR (ms)
0.182
0.267
0.061
0.717
0.068
SDRR (ms)
0.071
0.669
0.168
0.315
0.148
0.377
-0.131
0.428
0.081
0.629
0.047
0.780
RMSSD (ms)
0.685
LF (ms2)
0.339
HF (ms2)
-0.315
LFnu (%)
0.548
0.000**
0.380
0.019*
0.437
0.006*
HFnu (%) LF/HF
-0.548 0.633
0.000** 0.000**
-0.381 0.424
0.018* 0.008*
-0.439 0.493
0.006* 0.002*
LnLFnu(%) LnHFnu (%) LnLFnu/LnHFnu
0.455 -0.598 0.513
0.037* 0.054
0.004* 0.000** 0.001*
0.301
0.067
0.309
0.059
-0.119
0.475
-0.192
0.249
0.370
0.022*
-0.409
0.316
0.011*
0.054
-0.472
0.003*
0.357
0.028*
0.415
0.010*
Values in Pearson’s correlation coefficient adjusting for age, Tanner stage and moderate to vigorous physical activity RR beat-by-beat interval, SDRR standard deviation of normal R–R intervals, RMSSD square root of the mean squared differences of successive R–R intervals, LF low frequency, HF high frequency, LFnu low-frequency normalized units, LnLFnu natural logarithm of low-frequency normalized units, HFnu high-frequency normalized units, LnHFnu natural logarithm of high-frequency normalized units, LF/HF low frequency to high frequency ratio, WC waist circumference, TBF total body fat, TF trunk fat * P \ 0.05 for statistical significance ** P \ 0.001 for statistical significance
modulation in prepubescent children, accounting for the potential confounding effect of MVPA. Previous research showed that intense physical activity is associated with higher HRV vagal-related indexes in preadolescents [25]. In prepubescent lean children, a long-duration moderateintensity exercise program showed to have a positive effect on HRV [23]. However, high-intensity training is not proven to have an effect on HRV despite inducing a significant increase in aerobic fitness [36]. In our study, the accelerometer data showed no differences in either the total amount or the intensity of physical activity performed by boys and girls, which is unexpected since boys are usually more physically active [37]. This similar physical activity profile could partially explain the non-differences observed in HRV between genders, as MVPA has been included in the covariates, alongside with age, and Tanner stage. Previous research in children has shown that HRV is to some extent dependent on gender and age, increasing gradually with age, peaking between 6 and 9 years of age, then followed by a decrease or stagnation [38, 39]. Nevertheless, other studies reported no gender or age dependence at all [40, 41]. A more recent study [42] with a large cohort of young children has shown that in general, boys present higher HRV and at times age-related wave-like changes both in parasympathetic increases and sympathetic decreases. The same study suggests that physical fitness is a potent HRV determinant. In the BMI´ group analysis, obese children showed reduced MODPA and MVPA compared to their normal weight counterparts. This may be explained by
the negative correlation between the amount of time spent at moderate and vigorous physical activity and body fatness in children [37]. In terms of HRV, our results showed a distinctive relationship between lower parasympathetic modulation in frequency-domain HRV and weight status. Obese children presented with lower parasympathetic modulation and vagal withdrawal shown as a decreased LnHFnu, as similarly reported by the majority of the previous research [16–20]; although a recent study, using quantitative pupillography, showed lower parasympathetic modulation in obese children and adolescents that has not been observed in the LnHFnu [43]. It is difficult to interpret the higher LnLFnu values observed in our cohort of obese children since the LF component expresses not only the sympathetic modulation [44, 45] but also the parasympathetic modulation [46, 47]. This dual modulation of the LF also complicates the interpretation of the sympathovagal balance using the LF/HF ratio [47]. Previous research is inconsistent in terms of obese children having an increased or decreased LnLFnu. These conflicting results may arise from differences in cohort’ demographics; as highlighted previously, HRV in children is partially gender, age, and maturation stage dependent, and also highly influenced by physical activity and physical fitness levels [38, 39, 42]. The HRV in overweight children showed no differences in comparison to their normal weight counterparts, as observed by Kaufman et al. [16] in a sample of prepubescent and pubescent children. The predictable less time of exposure to obesity of the overweight children might explain
123
Clin Auton Res
this similarity with their normal weight counterparts [48]. Nevertheless, the small number of overweight children in our sample makes any kind of assumptions difficult. The correlation between body fat content and HRV was performed adjusting for the same covariates as in gender and BMI groups’ comparison. As seen in groups’ comparison, no associations between time-domain HRV and body fat content were found, similar to the findings of Kaufman et al. [16]. However, previous studies reported lower RMSSD values in pubescent obese children and adolescents [18, 48]. Our results suggest that time-domain measures might be less sensitive to detect obesity-related cardiac autonomic impairment in younger children, with possibly less time of exposure to obesity. In the frequencydomain, higher body fat mass, particularly trunk fat, was shown to be inversely correlated with vagal modulation, seen in the LnHFnu and positively correlated with LnLFnu and LnLFnu/LnHFnu ratio. Interestingly, the stronger correlations were observed with the waist circumference. This finding highlights the utility of this simple measurement in the risk assessment of autonomic impairment in young children. In a previous study [21], analogous significant correlations were found for DXA total body fat without adjustments, but not after adjusting for age, Tanner stage, insulin, and C-reactive protein levels. Gutin et al. [22], in a study with a large sample of adolescents, found that higher levels of visceral adipose tissue and subcutaneous adipose tissue were related to higher LFnu/HFnu ratio and were independent of race and sex. The relationship between central adiposity and cardiac autonomic dysfunction is complex and not clearly understood [49]. However, the overproduction of inflammatory cytokines by the fat cells is seen as the pathological trigger that leads to autonomic dysfunction [50]. The main limitation of this research was the small sample size, in particular the number of overweight children in the group, which did not allow us to run a more complex statistical analysis. In addition, we did not perform certain specialized autonomic tests such as the cold-pressor test, valsava maneuver, or deep-breathing. This is another important limitation since it might have provided more insight when analyzed in coordination with the resting HRV data. The lack of information about the time of exposure to obesity and the fact that metabolic parameters such as insulin resistance were not measured are also limitations in this study. The respiration rate was also not monitored during the HRV data collection so we cannot be completely sure that all of the participants maintained the defined 14–16 breaths/min during the procedure. Therefore, we cannot exclude a potential interference of respiration on vagal modulation. However, we did make every effort to ensure that all of
123
the children were still, relaxed, and breathing at the established frequency.
Conclusion Controlling for the possible influence of physical activity, this study shows a clear relationship between frequencydomain HRV impairment and obesity in prepubescent children. Despite the young age of the cohort, a higher amount of body fat mass, predominantly central fat, was related to decreased parasympathetic modulation. These findings highlight the potential cardiovascular risk that higher levels of body fat mass may represent even at a very young age. Further research is needed to clarify the obesity related autonomic dysfunction pathophysiology and the autonomic benefits of different types of exercise programs in young children.
References 1. Wang Y, Lobstein T (2006) Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes 1(1):11–25 2. de Onis M, Blossner M (2000) Prevalence and trends of overweight among preschool children in developing countries. Am J Clin Nut 72(4):1032–1039 3. OECD (2012) Obesity update 2012. OECD Health Division. http:// www.oecd.org/health/49716427.pdf. Accessed 30 July 2013 4. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS (1999) The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 103(6 Pt1):1175–1182 5. Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, Steer C, Sherriff A (2005) Early life risk factors for obesity in childhood: cohort study. BMJ 330(7504):1357. doi:10.1136/bmj. 38470.670903.E0 6. Falaschetti E, Hingorani AD, Jones A, Charakida M, Finer N, Whincup P, Lawlor DA, Davey Smith G, Sattar N, Deanfield JE (2010) Adiposity and cardiovascular risk factors in a large contemporary population of pre-pubertal children. Eur Heart J 31(24):3063–3072. doi:10.1093/eurheartj/ehq355 7. Engeland A, Bjorge T, Sogaard AJ, Tverdal A (2003) Body mass index in adolescence in relation to total mortality: 32-year followup of 227,000 Norwegian boys and girls. Am J Epidemiol 157(6):517–523 8. Must A, Strauss RS (1999) Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord 23(Suppl 2):S2–S11 9. Power C, Lake JK, Cole TJ (1997) Measurement and long-term health risks of child and adolescent fatness. Int J Obes Relat Metab Disord 21(7):507–526 10. Despres JP, Lemieux I, Prud’homme D (2001) Treatment of obesity: need to focus on high risk abdominally obese patients. BMJ 322(7288):716–720 11. Teixeira PJ, Sardinha LB, Going SB, Lohman TG (2001) Total and regional fat and serum cardiovascular disease risk factors in lean and obese children and adolescents. Obes Res 9(8):432–442. doi:10.1038/oby.2001.57
Clin Auton Res 12. Dietrich A, Rosmalen JG, Althaus M, van Roon AM, Mulder LJ, Minderaa RB, Oldehinkel AJ, Riese H (2010) Reproducibility of heart rate variability and baroreflex sensitivity measurements in children. Biol Psychol 85(1):71–78. doi:10.1016/j.biopsycho. 2010.05.005 13. Massin M, von Bernuth G (1998) Clinical and haemodynamic correlates of heart rate variability in children with congenital heart disease. Eur J Pediatr 157(12):967–971 14. Biswas AK, Scott WA, Sommerauer JF, Luckett PM (2000) Heart rate variability after acute traumatic brain injury in children. Crit Care Med 28(12):3907–3912 15. Javorka M, Javorkova J, Tonhajzerova I, Javorka K (2005) Parasympathetic versus sympathetic control of the cardiovascular system in young patients with type 1 diabetes mellitus. Clin Physiol Funct Imaging 25(5):270–274 CPF623 16. Kaufman CL, Kaiser DR, Steinberger J, Kelly AS, Dengel DR (2007) Relationships of cardiac autonomic function with metabolic abnormalities in childhood obesity. Obesity 15(5):1164–1171. doi:10.1038/oby.2007.619 17. Latchman PL, Mathur M, Bartels MN, Axtell RS, De Meersman RE (2011) Impaired autonomic function in normotensive obese children. Clin Auton Res 21(5):319–323. doi:10.1007/s10286011-0116-8 18. Martini G, Riva P, Rabbia F, Molini V, Ferrero GB, Cerutti F, Carra R, Veglio F (2001) Heart rate variability in childhood obesity. Clin Auton Res 11(2):87–91 19. Nagai N, Matsumoto T, Kita H, Moritani T (2003) Autonomic nervous system activity and the state and development of obesity in Japanese school children. Obes Res 11(1):25–32. doi:10.1038/ oby.2003.6 20. Tascilar ME, Yokusoglu M, Boyraz M, Baysan O, Koz C, Dundaroz R (2011) Cardiac autonomic functions in obese children. J Clin Res Ped Endo 3(2):60–64. doi:10.4274/jcrpe.v3i2. 131 21. Kaufman CL, Kaiser DR, Steinberger J, Dengel DR (2007) Relationships between heart rate variability, vascular function, and adiposity in children. Clin Auton Res 3:165–171. doi:10.1007/ s10286-007-0411-6 22. Gutin B, Howe C, Johnson MH, Humphries MC, Snieder H, Barbeau P (2005) Heart rate variability in adolescents: relations to physical activity, fitness, and adiposity. Med Sci Sports Exer 37(11):1856–1863 doi:00005768-200511000-00005 23. Nagai N, Hamada T, Kimura T, Moritani T (2004) Moderate physical exercise increases cardiac autonomic nervous system activity in children with low heart rate variability. Childs Nerv Syst 20(4):209–214. doi:10.1007/s00381-004-0915-5 24. Mandigout S, Melin A, Fauchier L, N’Guyen LD, Courteix D, Obert P (2002) Physical training increases heart rate variability in healthy prepubertal children. Eur J Clin Invest 32(7):479–487 25. Buchheit M, Platat C, Oujaa M, Simon C (2007) Habitual physical activity, physical fitness and heart rate variability in preadolescents. Int J Sport Med 28(3):204–210. doi:10.1055/s2006-924296 26. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320(7244):1240–1243 27. Kuczmarski RJ (1996) Bioelectrical impedance analysis measurements as part of a national nutrition survey. Am J Clin Nut 64(3 Suppl):453S–458S 28. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents (2004) The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents: part 1. Pediatrics 114(2 Suppl 4):555–576 29. Tanner JM (1962) Growth at adolescence. Charles C Thomas, Springfield
30. Mota J, Guerra S, Leandro C, Pinto A, Ribeiro JC, Duarte JA (2002) Association of maturation, sex, and body fat in cardiorespiratory fitness. Am J Hum Biol 14(6):707–712. doi:10. 1002/ajhb.10086 31. Trost SG, McIver KL, Pate RR (2005) Conducting accelerometer-based activity assessments in field-based research. Med Sci Sport Exer 37(11 Suppl):S531–S543 32. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG (2008) Calibration of two objective measures of physical activity for children. J Sports Sci 26(14):1557–1565. doi:10.1080/ 02640410802334196 33. Trost SG, Loprinzi PD, Moore R, Pfeiffer KA (2011) Comparison of accelerometer cut points for predicting activity intensity in youth. Med Sci Sport Exer 43(7):1360–1368. doi:10.1249/MSS. 0b013e318206476e 34. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996) Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 93(5):1043–1065 35. Gamelin FX, Baquet G, Berthoin S, Bosquet L (2008) Validity of the polar S810 to measure R-R intervals in children. Int J Sport Med 29(2):134–138. doi:10.1055/s-2007-964995 36. Gamelin FX, Baquet G, Berthoin S et al (2009) Effect of high intensity intermittent training on heart rate variability in prepubescent children. Eur J Appl Physiol 105(5):731–738. doi:10. 1007/s00421-008-0955-8 37. Ekelund U, Sardinha LB, Anderssen SA, Harro M, Franks PW, Brage S, Cooper AR, Andersen LB, Riddoch C, Froberg K (2004) Associations between objectively assessed physical activity and indicators of body fatness in 9- to 10-y-old European children: a population-based study from 4 distinct regions in Europe (the European Youth Heart Study). Am J Clin Nut 80(3):584–590 38. Silvetti MS, Drago F, Ragonese P (2001) Heart rate variability in healthy children and adolescents is partially related to age and gender. Int J Cardiol 81(2–3):169–174 [pii]: S016752730100537X 39. Galeev AR, Igisheva LN, Kazin EM (2002) Heart rate variability in healthy six- to sixteen-year-old children. Hum Physiol 28(4):428–432. doi:10.1023/A:1016529931519 40. Fukuba Y, Sato H, Sakiyama T, Yamaoka Endo M, Yamada M, Ueoka H, Miura A, Koga S et al (2009) Autonomic nervous activities assessed by heart rate variability in pre- and postadolescent Japanese. J Physiol Anthropol 28(6):269–273. doi:10. 2114/jpa2.28.269 41. Faulkner MS, Hathaway D, Tolley B (2003) Cardiovascular autonomic function in healthy adolescents. Heart Lung 32(1):10–22. doi:10.1067/mhl.2003.6 42. Michels N, Clays E, De Buyzere M, Huybrechts I, Marild S, Vanaelst B, De Henauw S, Sioen I (2013) Determinants and reference values of short-term heart rate variability in children. Eur J Appl Physiol 113(6):1477–1488. doi:10.1007/s00421-0122572-9 43. Baum P, Petroff D, Classen J, Kiess W, Bluher S (2013) Dysfunction of autonomic nervous system in childhood obesity: a cross-sectional study. PLoS One 8(1):e54546. doi:10.1371/jour nal.pone.0054546 44. Malliani A, Pagani M, Lombardi F (1994) Importance of appropriate spectral methodology to assess heart rate variability in the frequency domain. Hypertension 24(1):140–142 45. Montano N, Porta A, Cogliati C et al (2009) Heart rate variability explored in the frequency domain: a tool to investigate the link between heart and behavior. Neurosci Biobehav Rev 33(2):71–80. doi:10.1016/j.neubiorev.2008.07.006 46. Eckberg DL (2007) Sympathovagal balance: a critical appraisal. Circulation 96(9):3224–3232
123
Clin Auton Res 47. Reyes del Paso GA, Langewitz W, Mulder LJ, van Roon A, Duschek S (2013) The utility of low frequency heart rate variability as an index of sympathetic cardiac tone: a review with emphasis on a reanalysis of previous studies. Psychophysiology 50(5):477–487. doi:10.1111/psyp.12027 48. Rabbia F, Silke B, Conterno A, Grosso T, De Vito B, Rabbone I, Chiandussi L, Veglio F (2003) Assessment of cardiac autonomic modulation during adolescent obesity. Obes Res 11(4):541–548. doi:10.1038/oby.2003.76
123
49. Tentolouris N, Argyrakopoulou G, Katsilambros N (2008) Perturbed autonomic nervous system function in metabolic syndrome. Neuromolecular Med 10(3):169–178. doi:10.1007/ s12017-008-8022-5 50. Grassi G, Dell’Oro R, Facchini A, Quarti Trevano F, Bolla GB, Mancia G (2004) Effect of central and peripheral body fat distribution on sympathetic and baroreflex function in obese normotensives. J Hypertens 22(12):2363–2369 doi:00004872200412000-00019
