Cardiac adaptation to long-term endurance exercise is characterized by increased chamber size and wall thickness, which in turn is associated with increased aerobic capacity, in both adults and adolescents [
Moreover, in normal puberty, release of hormones with the primary function to regulate bodily growth changes dramatically with maturation in both sexes [
This study, a part of the Exercise Project at Jönköping University, Sweden, used a cross-sectional and comparative design. Subjects in the active group consisted of 24 adolescents 13–19 years old, which were recruited from orienteering and cross-country ski clubs in an area of southern Sweden. They were practicing endurance exercise (orienteering, running, cross-country skiing, and cycling) and competed at an elite level in their sports for at least two years prior to study enrollment; on average, they exercised 5 days a week for 30–120 minutes on each occasion in moderate to vigorous intensity, in addition to participation in compulsory physical education in school. The control group was recruited from public schools in the same area, consisted of 24 individually age- and sex-matched adolescents not engaged in regular exercise, except for mandatory physical education at school. Each group included 14 boys and 10 girls, respectively. All subjects were instructed to refrain from exercise on the day the test was performed. They also completed a questionnaire regarding exercise habits and medical conditions. Body surface area (BSA) was calculated according to the DuBois formula.
Written informed consent was obtained from all subjects, and the study was approved by the Regional Ethical Review Board in Linköping, Sweden (Dnr 2013/89–31). The study was executed in accordance with the Declaration of Helsinki.
A resting 12-lead electrocardiogram (ECG) was obtained with the MAC 5500HD version 10 (GE Healthcare, Milwaukee, WI, USA). Right arm systolic and diastolic blood pressure was acquired with the subjects in a supine position after at least 5 minutes of rest.
Nonfasting, resting venous blood samples were drawn between 3.45 and 4.15 pm into sample tubes without anticoagulants. The samples were separated, and serum was stored at −80°C until analysis. Nonfasting cortisol in serum was determined with chemiluminescence microparticle immunoassay (CMIA) using Architect i2000 (Abbott, Chicago, Illinois, USA). Cortisol in the sample was bound to anticortisol coated microparticles. After incubation, cortisol acridinium-labeled conjugate was added to the sample of mixture, which was bounded to available binding sites on the anticortisol coated microparticles. The resulting chemiluminescent was measured as relative light units. The detection limit for the assay was 25 nmol/L.
IGF-1, IGF-2, FSH, GH, LH, prolactin, and TSH were analyzed in serum with multiplex fluorochrome technique (Luminex, Bio-Rad Laboratories, USA). All hormones were analyzed by using a human metabolic assay (Bio-Rad Laboratories), and a Bio-Plex 200TM system (Luminex xMAPTM Technology) was used for identification and quantification of each hormone. Median fluorescence intensity (MFI), for each sample was registered and analyzed with Bio-Plex ManagerTM Software 5.0. The sample concentration values were estimated from a five-parameter logistic equation based standard curve. The cutoff value for minimum detectable concentrations for each metabolic marker was as follows: IGF-1 (0.29 ng/mL), IGF-2 (0.05 ng/mL), FSH (0.04 mLU/mL), GH (0.03 ng/mL), LH (0.13 mLU/mL), prolactin (0.04 ng/mL), and TSH (0.02
All participants underwent echocardiographic examination at rest, before the CPET, which previously has been described in detail by Rundqvist et al. [
To assess peak VO2, CPET on a treadmill was performed by all subjects, where exhaled air was analysed on a breath-by-breath basis for O2 and CO2 content with a Jaeger Oxycon Pro (VyAir Inc, Mettawa, Ill, USA). A non-rebreathing valve was connected to a mouthpiece to prevent mixing of inspired and expired air. The CPET was according to the modified Bruce protocol [
Statistics were performed with SPSS Statistics software version 21 (IBM software, Armonk, New York, USA). Variables are presented as median value and range (min-max). Differences between the active group and controls were tested with the nonparametric Wilcoxon matched-pairs signed-rank test, with a probability level of <0.05 as significant. Correlation between hormones and cardiac dimensions was analysed by bivariate correlations, and when separated into an active group and controls, linear regression analysis with and without controlling for BSA was performed.
All resting ECG were normal, none of the subjects reported a history of cardiovascular disease, and all subjects were nonsmokers. The only medical treatment reported was the regular use of bronchodilators to combat asthma in five active and one control participant. Characteristics were similar between the groups except for higher peak VO2 and lower resting heart rate (HR) in the active group (Table
Characteristics of the study population.
Active group ( |
Controls ( |
| |
---|---|---|---|
Age (years) | 15.5 (13–19) | 15.4 (13–19) | 0.934 |
BSA (m2) | 1.69 (1.37–1.98) | 1.72 (1.26–2.23) | 0.578 |
SBP (mmHg) | 120 (105–155) | 115 (105–130) | 0.449 |
DBP (mmHg) | 65 (50–85) | 65 (55–80) | 0.610 |
HR at rest (beats/min) | 62 (42–85) | 69 (49–88) |
|
Peak VO2 (mL/kg/min) | 62 (53–79) | 44 (27–62) |
|
Data are presented as median with range (min-max). Bold styling denotes statistical significance. BSA, body surface area; SBP and DBP, systolic and diastolic blood pressure; HR, heart rate; VO2, oxygen uptake.
There were no differences observed between the groups regarding resting levels of circulating hormones except for prolactin, which was slightly higher in the active group compared with controls (Table
Resting levels of circulating hormones in the active group vs controls and at subgroups of boys (b) and girls (g).
Active group ( |
Control group ( |
| |
---|---|---|---|
Cortisol (nmol/L) | 136 | 120 | 0.421 |
b | 125 (40–277) | 120 (33–286) | 0.541 |
g | 149 (48–305) | 108 (37–330) | 0.971 |
IGF-1 (ng/mL) | 25 | 23 | 0.288 |
b | 27 (11–40) | 25 (14–48) | 0.427 |
g | 21 (11–43) | 19 (4–29) | 0.529 |
IGF-2 (ng/mL) | 1.0 | 0.0 | 0.134 |
b | 1.1 (0.0–2.8) | 0.0 (0.0–8.0) | 0.454 |
g | 0.5 (0.0–5.0) | 0.0 (0.0–3.0) | 0.315 |
FSH (mLU/mL) | 7.4 | 5.7 | 0.177 |
b | 6.3 (1.8–12.1) | 4.1 (1.4–9.1) | 0.137 |
g | 7.8 (3.6–17.9) | 6.3 (3.8–11.9) | 0.684 |
GH (ng/mL) | 1.4 | 2.0 | 0.327 |
b | 1.5 (0.1–26.1) | 3.2 (0.2–28.6) | 0.571 |
g | 1.1 (0.6–6.4) | 1.8 (0.5–13.8) | 0.315 |
LH (mLU/mL) | 2.1 | 1.4 | 0.095 |
b | 2.2 (0.7–11.5) | 0.7 (0.0–4.6) |
|
g | 2.1 (1.2–28.0) | 2.0 (0.6–6.1) | 0.796 |
Prolactin (ng/mL) | 1.5 | 0.9 |
|
b | 1.6 (0.3–4.8) | 0.9 (0.0–4.3) |
|
g | 1.2 (0.3–5.5) | 0.9 (0.0–2.8) | 0.481 |
TSH ( |
2.6 | 2.8 | 0.877 |
b | 2.7 (1.1–10.6) | 2.8 (1.2–12.0) | 0.910 |
g | 2.6 (0.8–11.0) | 2.5 (1.3–9.8) | 0.796 |
Data are presented as median with range (min-max). Bold styling denotes statistical significance. IGF, insulin-like growth factor; FSH, follicle-stimulating hormone; GH, growth hormone; LH, luteinizing hormone; TSH, thyroid-stimulating hormone.
Cardiac dimensions of the active group versus controls are presented in Table
Cardiac parameters of the active group and controls.
Active group ( |
Controls ( |
| |
---|---|---|---|
LVID (mm) | 49 (44–56) | 45 (41–54) |
|
LVPWT (mm) | 7.6 (6.1–11.8) | 7.0 (5.3–9.3) |
|
LVM (g) | 109 (75–180) | 77 (54–143) |
|
LVEDV (mL) | 97 (78–153) | 80 (58–140) |
|
LV stroke volume (mL) | 59 (50–94) | 48 (33–87) |
|
LA diameter (mm) | 38 (28–45) | 33 (25–43) |
|
LA volume (mL) | 45 (34–70) | 31 (21–59) |
|
RVD1 (mm) | 39 (32–47) | 35 (27–46) |
|
RA diameter (mm) | 38 (27–48) | 35 (26–48) |
|
RA area (cm2) | 16 (10–22) | 12 (8–19) |
|
Data are presented as median with range (min-max). Bold styling denotes statistical significance. LVID, left ventricular internal diameter in diastole; LVPWT, left ventricular posterior wall thickness in diastole; LVM, left ventricular mass; LVEDV, left ventricular end-diastolic volume; LA, left atrium; RVD1, right ventricular basal internal diameter in diastole; RA, right atrium.
Analysing all participants together, IGF-1 was positively associated to cardiac dimensions such as LVID, LVM, LA diameter, LA volume, RA diameter, and RA area, as presented in Table
Bivariate correlation of cardiac dimensions vs hormones, all participants.
Cortisol | IGF-1 | IGF-2 | FSH | GH | LH | Prolactin | TSH | |
---|---|---|---|---|---|---|---|---|
Peak VO2 | 0.01 | 0.23 | 0.08 | 0.06 | −0.12 | 0.03 | 0.20 | 0.02 |
LVID | 0.01 |
|
−0.001 | 0.24 | −0.21 | 0.16 | 0.05 | −0.09 |
LVPWT | 0.05 | 0.27 | 0.03 | 0.05 | −0.13 | 0.01 | 0.23 | 0.10 |
LVM | 0.02 |
|
0.02 | 0.12 | −0.17 | 0.11 | 0.12 | −0.04 |
LA diameter | 0.001 |
|
−0.07 |
|
−0.13 | 0.11 | −0.05 | 0.003 |
LA volume | 0.01 |
|
0.004 | 0.18 | −0.26 | −0.04 | 0.14 | 0.05 |
RVD1 | 0.07 | 0.22 | −0.15 | 0.00 | −0.22 | 0.02 | −0.06 | −0.11 |
RA diameter | 0.03 |
|
−0.07 | −0.09 | −0.14 | −0.14 | −0.04 | −0.14 |
RA area | −0.01 |
|
−0.08 | −0.03 | −0.12 | −0.12 | −0.02 | −0.11 |
Data are expressed with Pearson’s correlation coefficient. The asterisk
Linear regression between cardiac variables and IGF-1 for active group and controls, respectively.
Cardiac variables | Beta ( |
Beta ( |
||
---|---|---|---|---|
Active group | Controls | Active group | Controls | |
Peak VO2 | 0.19 (0.238) | 0.11 (0.565) | 0.18 (0.260) | 0.10 (0.651) |
LVID | 0.10 (0.185) |
|
0.09 (0.116) | 0.06 (0.313) |
LVPWT | 0.01 (0.769) |
|
0.01 (0.861) | 0.02 (0.276) |
LVM | 0.40 (0.617) |
|
0.24 (0.638) | 0.63 (0.103) |
LA diameter | 0.14 (0.168) | 0.11 (0.294) | 0.12 (0.156) | −0.02 (0.810) |
LA volume | 0.32 (0.207) | 0.27 (0.205) | 0.27 (0.094) | −0.04 (0.821) |
RVD1 | −0.02 (0.879) | 0.17 (0.089) | −0.03 (0.706) | 0.04 (0.660) |
RA diameter | 0.09 (0.493) |
|
0.07 (0.519) | 0.20 (0.088) |
RA area | 0.07 (0.377) |
|
0.06 (0.321) | 0.06 (0.220) |
Beta, unstandardized coefficient.
This is the first study to our knowledge that has analysed the aspects of circulating hormones at rest and their association to cardiac dimensions in endurance-trained adolescents as well as in untrained age- and sex-matched controls. Our findings showed that resting levels of the analysed hormones, including IGF-1 and GH, uncontrolled for any confounding factors, could not be related to greater cardiac dimensions in endurance-trained adolescents. Notably, an association between resting levels of IGF-1 and cardiac dimensions was found in the controls, but it failed to reach significance when controlled for BSA. We found similar levels between the study groups regarding hormones associated to growth and metabolism at rest.
It has previously been confirmed that long-term endurance exercise leads to remodelling of all four cardiac chambers with respect to myocardial mass, wall thickness, and internal diameter in athletic adolescents and adults [
Although IGF-1 is mainly synthesized and secreted by the liver, other tissues such as the myocardium have the capacity to produce IGF-1, which then acts locally as an autocrine and a paracrine hormone [
The response of IGF-1 to exercise is fast and peaks approximately 10 minutes after onset of exercise [
Even if circulating level of IGF-1 has been reported to decrease during periods of heavy training, sports with similar training intensity throughout the season have not found changes in IGF-1 levels over time [
The period of adolescence is difficult to define in chronological years because it varies in onset and termination. The present research was performed on adolescents aged 13–19 years, which theoretically means that some of the subjects may have been in a prepubertal phase. Thus, a weakness of our investigations was that we did not take into account maturity status or Tanner stages during the data collection. Obviously, adjustments to these factors would have contributed to a deeper understanding of our results. However, hormones responsible for inducing puberty (e.g., LH and FSH) did not differ significantly at group levels between the endurance-trained and the control group, which may suggest comparable pubertal status between the two groups. There is a large degree of interindividual variation in hormone response to exercise, even when the subjects are matched for age, sex, body composition, and so on [
In conclusion, increased cardiac dimensions in endurance-trained adolescents were not associated with resting circulating levels of growth factors, including IGF-1 and GH, which indicate that other mechanisms and triggers are of greater importance to physiological cardiac hypertrophy in endurance-trained adolescents. In addition, resting levels in serum of hormones associated with growth and metabolism did not differ between the active group and the controls, where the active group had greater cardiac size compared with the controls. Our results may contribute to the knowledge about factors that may (or may not) trigger cardiac hypertrophy in adolescent athletes.
The data used to support the findings of this study are available from the corresponding author upon request.
No conflicts of interest, financial or otherwise, are declared by the authors.
The authors thank Johan Wedenfeldt at Region Jönköping County, Division of Medical Diagnostics, Clinical Physiology, Ryhov County Hospital, Jönköping, Sweden, for collecting echocardiography data at rest and after exercise. The authors also thank Ingemar Kåreholt at the Institute of Gerontology, School of Health and Welfare, Jönköping University, Sweden, for helping us with statistical analysis. This study was supported by grants from the Medical Research Council of Southeast Sweden (FORSS), Grant number 651971.