The Association Between Measures of Fitness and Metabolic Health in Treatment-Seeking Youth with Obesity

Abstract

Background:

Both cardiorespiratory fitness (CRF) and measures of muscular fitness are associated with metabolic syndrome in adults. However, limited information exists about these relationships in youth with severe obesity who are at increased risk of metabolic dysfunction. The purpose of this study was to examine the relationship between fitness and metabolic health in treatment-seeking youth with obesity.

Methods:

Data for this analysis were collected at the time of baseline visits at a stage 3 pediatric weight management center. Maximal voluntary contractions were obtained by using isometric hand-grip dynamometry, and CRF was obtained from a maximal treadmill test. Resting blood pressure and fasting measures of blood lipids, glucose, and insulin were used to calculate a continuous metabolic syndrome score (cMetS); homeostasis model assessment of insulin resistance (HOMA-IR) was calculated from fasting insulin and glucose. Relationships between measures of fitness and metabolic health were evaluated by using partial correlations adjusted for age.

Results:

Sixty-nine participants (21 boys, 48 girls) were included in this analysis. Of these, 46% (n = 32) met the criteria for metabolic syndrome. No differences were found between boys and girls for any variable analyzed. Muscular strength was positively associated with cMetS (r = 0.35), though this association weakened after adjustment for body mass index percentile. CRF was inversely associated with homeostasis model assessment of insulin resistance (HOMA-IR) (r = −0.26) and fasting insulin (r = −0.27). Body fat percentage was positively associated with insulin (r = 0.36). No significant relationship was found between CRF and cMetS.

Conclusion:

Contrary to previous studies, CRF was not associated with metabolic syndrome in this group. Muscular strength, however, was associated with cMetS. Notably, CRF was associated with elevated HOMA-IR, which may be seen as a precursor to metabolic syndrome. These results suggest that CRF and muscular strength influence metabolic function independently.

Introduction

Despite extensive research into the causes and treatment of pediatric obesity, rates of severe obesity and childhood onset of traditionally “adult” diseases continue to increase. The metabolic syndrome (MetS), a clustering of adverse cardiovascular disease risk factors, is generally considered to provide a composite picture of metabolic risk among adults and, more recently, has been used in this capacity in children and adolescents. However, dichotomous classification of MetS among youth presents a number of limitations. Thus, many researchers have begun using a continuous metabolic syndrome score (cMetS) to provide a more statistically robust indicator of metabolic risk.¹

Physical activity is often prescribed to treat obesity. However, aerobic fitness can be accurately measured more easily than physical activity and is highly influenced through intensity and duration of exercise (i.e., physiological stress), which are features that are necessary to promote improvement in metabolic factors. These factors may make physical fitness a stronger predictor of metabolic health than is physical activity. Many studies have focused on the role of cardiorespiratory fitness (CRF) in metabolic health^2

–6; fewer have considered the independent contributions of muscular strength, which may be independently associated with metabolic health. In addition, no studies have evaluated these relationships in a sample of youth with obesity and severe obesity. Therefore, the purposes of this study were to describe fitness and metabolic characteristics of children seeking treatment for obesity, and to evaluate the relationship between measures of muscular and CRF and cardiometabolic health in this population.

Materials and Methods

This study was a retrospective analysis of data collected at the time of baseline visits at a stage 3 pediatric weight management center.⁷ The sample included boys and girls between the ages of 6 and 18 years who had a body mass index (BMI) at or above the 95th percentile for age and sex at the time of the initial clinic visit. An initial sample of 130 participants was identified over a 9-month period. The Institutional Review Board approved the study protocol.

Anthropometry

All anthropometric variables were assessed at the time of the baseline clinic visit according to standard procedures. Briefly, body mass was assessed to the nearest 0.1 kg by using a beam balance (Scaltronix, White Plains, NY). Standing height was measured without socks or shoes to the nearest 0.1 cm by using a wall-mounted stadiometer (Harpenden, Great Britain). BMI was calculated by dividing body mass into height in meters squared; BMI percentile was determined according to the 2000 CDC grown charts.⁸ Waist circumference (WC) was measured to the nearest 0.1 cm at the level of the superior border of the iliac crest by using a Gullick tape. Body composition was assessed by using bioelectrical impedance analysis (RJL Systems Quantum II, Clinton Township, MI).

Metabolic syndrome

Fasting blood lipids and glucose were determined by blood work routinely completed as part of the first clinic visit. This included fasting measures of cholesterol [total, high-density lipoprotein (HDL), and low-density lipoprotein (LDL)], triglycerides, glucose, and insulin. All blood analysis was performed according to standard procedures by a CLIA-accredited hospital laboratory. Resting blood pressure was obtained on the right arm after 10 min of seated rest by using an automated sphygmomanometer (Datascope, Accutorr Plus). Mean arterial pressure was calculated by multiplying pulse pressure, systolic blood pressure−diastolic blood pressure (SBP−DBP), by 1/3 and adding this value to DBP. Metabolic syndrome was classified according to the criteria outlined by Cook et al.⁹ as follows: (1) WC ≥90th percentile; (2) triglyceride ≥110 mg/dL; (3) HDL-C < 40 mg/dL; (4) blood pressure ≥90th percentile; and (5) fasting glucose ≥110 mg/dL. To allow for analysis of cardiometabolic health as a continuous variable, we calculated a cMetS by regressing WC, blood pressure, fasting lipids (HDL and triglycerides), and fasting glucose values onto age- and sex-specific percentiles derived from NHANES III data, similarly to the method described by Eisenmann et al.,¹⁰ where greater values indicate more adverse cardiometabolic risk factor profiles. Importantly, the use of nationally representative data for construction of the cMetS score allows for greater generalizability than do sample-specific scores previously reported. Insulin resistance was evaluated by the homeostatic model assessment (HOMA-IR).¹¹

Fitness

Muscular strength was assessed by maximal voluntary contraction using isometric hand-grip dynamometry. CRF, expressed as maximal oxygen consumption (VO_2max), was obtained from a maximal treadmill test.¹²

Statistical analysis

Descriptive statistics were calculated for the total sample and by sex. Differences between the sexes and between participants with and without the metabolic syndrome were examined by using independent-samples t-tests. Relationships between measures of fitness and metabolic health were evaluated by using partial correlations adjusted first for age and sex. Hand-grip dynamometry was further adjusted for BMI percentile to account for the influence of body size on strength. Significance was evaluated at the α = 0.05 level. Analyses were performed by using SPSS vs. 22.0.

Results

Subject characteristics

Participants without complete data for fasting insulin (n = 38), components of the metabolic syndrome [HDL cholesterol, triglycerides, fasting blood glucose, resting blood pressure (n = 13)], and CRF (n = 10) were excluded from this analysis. A total of 69 participants (21 boys, 48 girls; mean age 12.3 ± 2.9 years) were included in the final sample. Age and anthropometric characteristics of excluded and included participants did not differ (all P > 0.40). The majority of participants (n = 53; 77%) were classified as having severe obesity (BMI >120% of 95th percentile). Male participants exhibited greater BMI z-score, fat free mass, and strength when compared with female participants (Table 1). No other characteristics differed between the sexes. Approximately 28% (n = 19) of the final sample self-identified as African American, whereas 59.4% (n = 41) identified as white, 5.8% (n = 4) as Hispanic, and 7.2% (n = 5) did not report ethnicity.

Table 1.

Descriptive Characteristics of the Sample

	Male (n = 21), mean (SD)	Female (n = 48), mean (SD)	Total (n = 69), mean (SD)	P
Age (years)	12.5 (2.6)	12.2 (3.0)	12.3 (2.9)	0.641
Height (cm)	161.3 (16.6)	154.4 (12.2)	156.5 (14.0)	0.059
Weight (kg)	100.4 (41.6)	82.9 (25.4)	88.2 (32.0)	0.036^a
BMI (kg/m²)	37.2 (11.6)	33.9 (7.0)	34.9 (8.7)	0.142
BMI percentile (%)	99.3 (0.6)	98.8 (1.0)	99.0 (0.9)	0.023^a
% of 95th BMI percentile	150.6 (36.0)	135.1 (20.6)	139.8 (27.0)	0.027^a
BMI z-score	2.6 (0.4)	2.4 (0.3)	2.4 (0.3)	0.023^a
Height percentile	73.0 (29.6)	73.7 (24.5)	73.5 (25.9)	0.923
Weight percentile	99.1 (1.8)	99.1 (1.2)	99.1 (1.4)	0.948
% Body fat	37.2 (11.5)	36.0 (9.2)	36.4 (9.9)	0.649
Fat mass (kg)	40.7 (28.0)	31.8 (16.7)	34.5 (21.0)	0.107
Fat free mass (kg)	61.1 (17.6)	51.1 (10.7)	54.2 (13.9)	0.023^a
Strength (kg)	64.0 (28.8)	48.8 (15.0)	53.4 (21.2)	0.005^a
VO_2max (mL/kg/min)	26.4 (4.1)	24.2 (5.2)	24.9 (5.0)	0.088

Values are mean (SD), unless otherwise stated.

Significant difference between sexes (P < 0.05).

BMI, body mass index.

Mean VO_2max among adolescents fell more than 3 standard deviations below the mean (boys mean z = −3.3; girls mean z = −3.8), based on nationally representative FITNESSGRAM values.¹³ Nationally representative norms for children under age 12 years are not currently available for comparison. At this time, nationally representative percentile norms for grip strength are not available.

Cardiometabolic health

To determine the degree of cardiometabolic risk, participant biomarkers were compared with recommended levels of blood lipids, glucose, blood pressure, and WC; mean values for the sample are shown in Table 2. Briefly, all participants demonstrated elevated WC, most of them had adverse levels of HDL cholesterol (85.5%), and few (1.4%) exhibited adverse fasting glucose. Forty-six percent (n = 32) of the sample (50% of girls vs. 38% of boys) met criteria for MetS, whereas the mean cMetS score was 4.0 ± 2.5 (girls 3.8 ± 2.5; boys 4.6 ± 2.5).

Table 2.

Features of the Metabolic Syndrome in the Study Sample

	Male (n = 21), mean (SD)	Female (n = 48), mean (SD)	Total (n = 69), mean (SD)	P	Percent adverse %
Waist circumference (cm)	115.2 (23.7)	106.4 (16.4)	109.1 (19.1)	0.079	100.0
Total cholesterol (mg/dL)	157.2 (29.3)	159.5 (28.6)	158.8 (28.7)	0.766	14.5
HDL cholesterol (mg/dL)	38.0 (9.4)	39.8 (10.0)	39.2 (9.8)	0.490	85.5
Triglycerides (mg/dL)	115.4 (63.4)	102.4 (62.7)	106.4 (62.7)	0.434	18.8
Glucose (mg/dL)	90.5 (7.1)	88.6 (9.1)	89.2 (8.6)	0.418	1.4
Insulin (mIU/L)	23.1 (14.4)	25.0 (17.8)	24.4 (16.7)	0.670	—
HOMA-IR	5.2 (3.2)	5.6 (4.3)	5.5 (4.0)	0.719	—
Systolic BP (mmHg)	114.8 (16.6)	111.1 (15.0)	57.9 (28.7)	0.369	24.6^a
Diastolic BP (mmHg)	69.1 (9.7)	67.2 (10.3)	63.8 (2.6)	0.492	11.6^a
cMetS	4.6 (2.5)	3.8 (2.5)	4.0 (2.5)	0.214	—
MetS (y/n)	—	—	—	—	46.4

Values are mean (SD) unless otherwise indicated.

A combined 29% of the sample had at least one BP value ≥90th percentile.

cMetS, continuous metabolic syndrome score; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance.

Relationships between fitness and cardiometabolic health

When unadjusted, body fat percentage, fat free mass (FFM), and fat mass (FM) were positively associated with HOMA-IR (Table 3). After adjusting for age and sex, associations between body fat percentage, FM, and HOMA-IR remained significant, whereas the link between FFM and HOMA-IR weakened to non-significance.

Table 3.

Correlations Between Measures of Fitness, Body Composition, and Metabolic Health

	cMetS		HOMA
	r^a	r^b	r^a	r^c
% Body fat	0.12	0.03	0.33^d	0.30^d
Fat mass (kg)	0.22	0.11	0.32^d	0.33^d
Fat free mass (kg)	0.33^d	0.31	0.30^d	0.28
Strength (kg)	0.38^d	0.30^d	0.15	−0.03
VO_2max (mL/kg/min)	0.00	−0.02	−0.30^d	−0.26^d

Coefficients are unadjusted.

Adjusted for age and sex.

Strength coefficients are adjusted for age, sex, and BMI percentile.

Statistical significance (P < 0.05).

Neither muscular strength nor CRF was related to blood pressure, blood lipids, or fasting blood glucose (data not shown). Strength was positively correlated with cMetS and showed no relationship with HOMA-IR (Table 3). The positive relationship with cMetS remained after adjusting for age and sex, and BMI percentile. CRF was inversely associated with HOMA-IR before adjusting for age and sex (Table 3), and these relationships remained after adjusting for covariates.

Discussion

This article provides data regarding fitness and metabolic characteristics of children seeking treatment for obesity, and it presents an analysis of the relationships between fitness and cardiometabolic risk factors in this important population. Just less than half of the participants (46%) met the criteria for metabolic syndrome as outlined by Cook et al.,⁹ which is comparable to the prevalence reported in previous studies of youth with obesity (38%–50%).¹⁴ Characteristics of the metabolic syndrome were regressed onto age- and sex-specific values from NHANES¹⁵ and were used to construct the cMetS. According to this method, a cMetS score equal to 0.0 indicates that the participant's cardiovascular risk is similar to the median of children of the same age and sex. Thus, a negative score indicates more favorable risk, whereas an increasingly positive score indicates increased risk. Despite the relatively high prevalence of metabolic syndrome, the minimum and maximum values for the cMetS reported in this study (mean = 4.0 ± 2.5; min −1.8, max 11.4) indicate that cardiovascular risk status can vary widely even among “high risk” youth with obesity or severe obesity.

The CRF of this sample was quite poor when compared with nationally representative values, falling more than 3 standard deviations below the mean reported from FITNESSGRAM data (boys mean z = −3.3; girls mean z = −3.8).¹³ Unfortunately, there are not any published norms that allow for a comparison of VO_2max scaled for FFM, which may be a more appropriate method for a comparison between children with obesity and those with healthy weight.¹⁶ When comparing handgrip strength in 1-year age groups in the current study with NHANES values,¹⁷ boys ≥11 years and girls ≥7 years were generally stronger than average, likely due to the influence of body mass. Published data allowing for determination of grip strength percentiles are not currently available.

In regards to the relationships between fitness variables and metabolic health, our findings are generally consistent with other studies of normal weight and overweight youth, and these findings are extended to children with obesity. Similar relationships between body fat and markers of insulin resistance have been shown in primarily normal weight children from the European Heart Study.¹⁸ This same study also showed inverse relationships between CRF and HOMA, similar to our findings. In the Ruiz study, the inverse relationship between CRF and insulin resistance was the most pronounced in the highest tertiles of WC and body fat, which is supported by our findings in children with obesity and severe obesity. Benson et al.¹⁹ found that children in the highest tertile for CRF were 95% less likely to demonstrate high insulin resistance when compared with the lowest tertile of CRF. Although the relationships found in the current study are weaker than those previously shown, this may be explained by the comparatively low fitness level and lack of variability in our sample. Taken together, these findings suggest that increasing CRF could reduce the negative metabolic consequences associated with high adiposity.

After adjusting for potential confounders, muscular strength was not associated with fasting HOMA-IR in our study. In contrast to our findings, Benson et al.¹⁹ showed that moderate and high muscular strength was associated with significantly less likelihood of insulin resistance when compared with children with low muscular strength. Both the Benson paper and ours report associations that are adjusted for body composition; our analyses are adjusted for BMI percentile, whereas theirs are adjusted for WC and BMI. We did not adjust for WC, since this variable is included in the cMetS score; however, 100% of our sample exhibited an elevated WC and body fat percentage. As such, we may not have wide enough variation in central adiposity or body composition to detect a difference in this regard. Contrary to previous studies,⁵ our results do not support a relationship between CRF and cMetS in this population. Again, this may be related to the relative homogeneity of our sample with regard to obesity and central adiposity.

Our data support the notion that high adiposity and low fitness are related to insulin resistance represented by HOMA. However, our data do not support a relationship between CRF and cardiovascular risk as described by a cMetS score. Insulin resistance is hypothesized to be the central, driving factor in the development of the metabolic syndrome. Although our sample was drawn from a population of children seeking treatment for pediatric obesity, and thus likely at increased cardiovascular disease risk, it may be that the insulin resistance seen in this sample has not progressed to the point that other cardiovascular risk factors are impacted to the same degree. Furthermore, we were not able to statistically adjust for pubertal maturation and the influence of sex hormones; future studies should include assessment of maturational differences. An extended study of these children into later adolescence and early adulthood may provide additional insight into the timing and progression of insulin resistance and the metabolic syndrome and their potential for prevention with weight management treatment.

Though the attainment of a healthy BMI is an optimal outcome for pediatric weight management, significant change in adiposity takes a substantial amount of time and maintenance of improved weight trajectories is often unsuccessful. Improvements in insulin sensitivity related to CRF and muscular strength may occur before any significant changes in adiposity, presenting a more timely method to improve metabolic health. Furthermore, as demonstrated by our data, children with obesity tend to be stronger than their normal weight peers. Incorporation of strength training in pediatric weight management may provide additional benefits in the form of improved self-efficacy and adherence to the weight management intervention. Improved understanding of the mechanisms that contribute to insulin resistance and poor metabolic health will allow for the development of effective, evidence-based early interventions to improve metabolic health in children with obesity.

Footnotes

Acknowledgments

The authors wish to acknowledge the contributions of staff at the Healthy Weight Center at Helen DeVos Children's Hospital and undergraduate and graduate students research assistants in the University of Wyoming Pediatric Physical Activity Lab.

Author Disclosure Statement

No conflicting financial interests exist.

References

Eisenmann

. On the use of a continuous metabolic syndrome score in pediatric research. Cardiovasc Diabetol, 2008; 7:17.

Steele

, Brage

, Corder

, et al. Physical activity, cardiorespiratory fitness, and the metabolic syndrome in youth. J Appl Physiol, 2008; 105:342–351.

Eisenmann

. Secular trends in variables associated with the metabolic syndrome of North American children and adolescents: A review and synthesis. Am J Hum Biol, 2003; 15:786–794.

Rizzo

, Ruiz

, Hurtig-Wennlöf

, et al. Relationship of physical activity, fitness, and fatness with clustered metabolic risk in children and adolescents: The European youth heart study. J Pediatr, 2007; 150:388–394.

Martínez-Gómez

, Eisenmann

, Moya

, et al. The role of physical activity and fitness on the metabolic syndrome in adolescents: Effect of different scores. The AFINOS Study. J Physiol Biochem, 2009; 65:277–289.

Countryman

, Saab

, Llabre

, et al. Cardiometabolic risk in adolescents: Associations with physical activity, fitness, and sleep. Ann Behav Med, 2013; 45:121–131.

Barlow

. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: Summary report. Pediatrics, 2007; 120 Suppl:S164–S192.

Kuczmarski

, Ogden

, Guo

, et al. 2000 CDC Growth Charts for the United States: Methods and Development. National Center for Health Statistics. Vital Health Stat, 2002; 11.

Cook

, Weitzman

, Auinger

, et al. Prevalence of a metabolic syndrome phenotype in adolescents: Findings from the third National Health and Nutrition Examination Survey, 1988–1994. Arch Pediatr Adolesc Med, 2003; 157:1988–1994.

10.

Eisenmann

, Laurson

, DuBose

, et al. Construct validity of a continuous metabolic syndrome score in children. Diabetol Metab Syndr, 2010; 2:8.

11.

Avignon

, Boegner

, Mariano-Goulart

, et al. Assessment of insulin sensitivity from plasma insulin and glucose in the fasting or post oral glucose-load state. Int J Obes Relat Metab Disord, 1999; 23:512–517.

12.

Eisenmann

, Guseman

, Morrison

, et al. Graded exercise testing in a pediatric weight management center: The DeVos protocol. Child Obes, 2015; 11:657–663.

13.

Eisenmann

, Laurson

, Welk

. Aerobic fitness percentiles for U.S. adolescents. Am J Prev Med, 2011; 41:S106–S110.

14.

Weiss

, Dziura

, Burgert

, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med, 2004; 350:2362–2374.

15.

Cook

, Auinger

, Huang

TT-K

. Growth curves for cardio-metabolic risk factors in children and adolescents. J Pediatr, 2009; 155:S6.e15–S6.e26.

16.

McMurray

, Hosick

, Bugge

. Importance of proper scaling of aerobic power when relating to cardiometabolic risk factors in children. Ann Hum Biol, 2011; 38:647–654.

17.

Ervin

, Fryar

, Wang

C-Y

, et al. Strength and body weight in US children and adolescents. Pediatrics, 2014; 134:e782–e789.

18.

Ruiz

, Ortega

, Warnberg

, et al. Associations of low-grade inflammation with physical activity, fitness and fatness in prepubertal children; the European Youth Heart Study. Int J Obes (Lond), 2007; 31:1545–1551.

19.

Benson

, Torode

, Fiatarone Singh

. Muscular strength and cardiorespiratory fitness is associated with higher insulin sensitivity in children and adolescents. Int J Pediatr Obes, 2006; 1:222–231.