Inflammatory markers as potential mediators on the negative association between training load and bone mineral density in adolescent competitive swimmers: ABCD-growth study

Introduction: Competitive swimming during adolescence has been linked to poor bone development, potentially influenced by training load, inflammation, hormones, and bone markers. However, this influence has been poorly investigated in the literature.

Objective: To compare whether competitive adolescent swimmers present differences in inflammatory, immunological, anabolic, and bone markers compared with non-sport group and to analyse whether inflammatory variables mediate the association between training load and areal bone mineral density (aBMD) in the swimmers group.

Methods: This cross-sectional study included 61 adolescents (20 females, 15.4±2.3 years), of which 30 were adolescent swimmers and 31 did not participate in sports (non-sports group). The daily training load was obtained by multiplying the perceived exertion score by the training volume of the swimmers. Lean soft tissue, fat mass and aBMD were estimated from whole-body scans using dual-energy X-ray absorptiometry and peak height velocity considering stature and body mass. Blood samples were collected to assess bone markers (calcium, 25-hydroxy vitamin D, C-terminal telopeptide, and osteocalcin), growth hormones, insulin-like growth factor 1 (IGF-1), lymphocytes, leukocytes, and inflammatory markers (interleukin-6 [IL-6] and C-reactive protein [CRP]). Statistical analyses applied a significance level at p<0.05.

Results: Besides lower values in BMD (expect in upper limbs), swimmers had higher calcium (10.0 ± 0.30 vs 9.7 ± 0.44, p=0.007), vitamin D (42.6 ± 10.4 vs 24.2 ± 5.9, p<0.001), and IGF-1 (397.2 ± 115.1 vs 220.3± 73.9, p<0.001) concentrations than their non-sports peers. The mediation analysis found no indirect associations between training load and aBMD through inflammatory markers. Nevertheless, training load was directly and negatively associated with aBMD in the lower limbs (β=-0.1533, 95%CI:-0.2875, -0.0191) and total body less head (β=-0.0978, 95%CI:-0.1880, -0.0076) through IL-6 and directly and negatively associated with aBMD at all sites through CRP.

Conclusion: The swimming and non-sports groups did not show differences in bone, inflammatory, or immunological markers. In contrast, swimmers had higher concentrations of IGF-1, calcium, and 25-hydroxy vitamin D. Although the training load was negatively associated with aBMD, inflammatory markers (IL-6 and CRP) did not mediate this association. Reinforcing the hypothesis that swimmers have lower aBMD due to the hypogravitational environment.

1 Introduction

The literature shows that sports with different mechanical impacts can distinctly affect the bone mineral density (BMD) (1). Sports with a higher mechanical impact and muscle activity, i.e. basketball and gymnastic, seem to be more effective for areal bone mineral density (aBMD) accrual (2), mainly during adolescence, when most of the bone mass observed in adulthood is accumulated (3). In addition, sports without a mechanical impact, such as swimming, are not beneficial for bone health, even with high muscle activity (4). Recent studies have shown that swimming can similarly or negatively affect the bone health of adolescents, even when compared to adolescents not engaged in sports, as observed in lower limbs and whole-body aBMD (4, 5).

Therefore, it is important to understand whether other factors, such as inflammation, could explain the negative association between training load and aBMD, in addition to hypogravity. Evidence indicates that the training load and previous sports practice time are negatively correlated with aBMD in adolescent swimmers (6, 7). Furthermore, training with very high physical effort (e.g., ~74% VO_2max or >64% VO_2max) may favour the increased circulation of immunological variables (8, 9), cytokines, and proteins related to the inflammatory response, such as interleukin-6 (IL-6), interleukin-1 (IL-1), tumour necrosis factor alpha (TNF-α), and C-reactive protein (CRP) (10). However, an intense bout of exercise in adolescents can lead to an acute reduction of anabolic markers such as insulin-like growth factor 1 (IGF-1) immediately after exercise (11), which is important for bone growth (12). Among the pro-inflammatory markers, IL-6 cytokine deserves special attention due to its increased expression on the activator of nuclear factor-kB ligand (RANKL), an essential protein for osteoclastogenesis (bone resorption) (13, 14). The increase in this process resultant of osteoclast activity may reduce bone mass accrual, resulting in decreased BMD. Additionally, evidence indicates that IL-6 leads to the increased production of CRP (15), which is a marker of inflammation in clinical practice (16).

However, to our knowledge, no studies have analysed whether, in fact, a competitive swimming training routine can cause changes in inflammatory, immunological, anabolic, and bone markers in adolescents compared with a non-sports group. Furthermore, whether inflammatory markers mediate the negative association between training load and BMD in swimmers has not yet been analysed. We hypothesized that adolescent competitive swimmers exhibit differences in physiological markers compared to controls, with inflammatory markers potentially mediating the negative association between training load and BMD. Thus, this cross-sectional study firstly aimed to compare whether competitive adolescent swimmers present differences in inflammatory, immunological, anabolic, and bone markers compared to adolescents who do not perform organised sports. Secondly, we analyzed whether inflammatory variables mediate the association between training load and aBMD in adolescent swimmers.

2 Methods

2.1 Study design

This was a cross-sectional study, and the data were collected as part of ongoing research, titled “Analysis of Behaviors of Children During Growth: ABCD-Growth Study”. The data from this study were collected as part of investigations by the Laboratory of InVestigation in Exercise (LIVE) conducted in Presidente Prudente, São Paulo, Brazil, from 2017 to 2018. The ABCD Growth Study aims to identify the impact of physical activity and sports participation on health variables, including bone tissue. This study was approved by the Ethical Research Committee of the São Paulo State University-UNESP (process number: 02891112.6.0000.5402).

2.2 Sample

The sports group comprised adolescents from a swimming team classified in Tier 2: Trained/Developmental, from Participant Classification Framework proposed by McKay et al. (2022) (17). While adolescents not engaged in organised sports were invited to join the non-sports group. To participate in the study, the adolescents were required to provide informed consent signed by their parents or guardians in according with the Declaration of Helsinki.

For adolescents who participated in sports, the following inclusion criteria were adopted: i) previous involvement for at least six months in the current sport (in the present study sample, the minium period of engagement observed in swimming was 24 months), ii) signed written consent form, iii) absence of clinical or metabolic disorders (previously diagnosed), iv) no regular use of medications that may affect the development and growth of the adolescent, and v) age of 10–19 years. For the adolescents in the non-sports group, the following criteria were applied: i) no regular participation in sports, ii) signed written consent, iii) absence of clinical or metabolic disorders (previously diagnosed), iv) no regular use of medications, and iv) age of 10–19 years.

The sample size was estimated based on the study by Agostinete et al. (6), using a correlation (two tailed bivariate correlation) coefficient of R=0.563 between the training load and lower limb aBMD, an alpha error of 0.05, a statistical power of 80%. The minimum estimated sample size was 22 participants, which was calculated using G*Power software. Thus, the final sample comprised 61 adolescents (41 males and 20 females) divided into two groups: 30 adolescent swimmers (21 males and 9 females) and 31 adolescents in the non-sports group (20 males and 11 females).

2.3 Outcomes

Height was measured using a fixed stadiometer (accurate to 0.1 cm; Sanny model, American Medical Do Brazil Ltda, Brazil), and body mass was measured using an electronic scale (accurate to 0.1 kg; Filizola PL 150 model; Filizzola Ltda, Brazil).

Biological maturation was estimated by the maturity offset (MO) using mathematical models based on anthropometric measures (18) for males = -7.999994+ [0.0036124 * (Age * Stature)] and females = -7.709133+ [0.0036124 * (Age * Stature)]. Thus, peak height velocity (PHV), the period of age where the fastest growth is seen, is given by the difference between chronological age and maturity offset (PHV = chronological age – MO).

To assess the whole-body aBMD (g/cm²), lean soft tissue (kg), and fat mass (kg and %), dual-energy x-ray absorptiometry (DXA) scanning (models Lunar DPX-NT and Lunar Prodigy advance; General Electric Healthcare, Little Chalfont, Buckinghamshire, UK) was performed at the university laboratory in a temperature-controlled room using GE Medical System Lunar software. Each day, a trained researcher performed the scans and tested the scanner quality before the first examination. The scans were performed using a standardised protocol, with the participants remaining in the supine position and wearing light clothing without shoes. Regional BMD analysis according to the definitions of the regions of interest (ROIs), including the upper limbs, lower limbs, spine, and whole body (less the head), occurred offline after the scans were performed, as previously described (19).

Blood samples were collected by a certified nurse. Adolescents were instructed to refrain from exercise training on the day previous to blood collection. After an overnight fast, venous blood was collected from the antecubital vein in vacuum collection tubes with anticoagulant separating gel. Blood collection was carried out in a private laboratory, where it was also analyzed, and all the quality control standards adopted by the Brazilian Health Ministry were followed. Bone turnover markers, such as calcium, 25-hydroxy vitamin D, C-terminal telopeptide (CTX), osteocalcin, growth hormone, insulin, and IGF-1, were analysed using an electrochemiluminescence kit (Roche), Cobas e170, and Modular Analytics E170 equipment (chemiluminescence assay kit [VITROS] with the VITROS^® XT 7600 equipment). Lymphocytes (CD3+, CD4+, and CD8+ cells) and leukocytes were analysed by flow cytometry using a Beckman Coulter kit and the Aquios-Beckman Coulter equipment. Finally, IL-6 and CRP levels were analysed using a Siemens kit and Immulite 2000 XPI equipment (Siemens Healthcare GmbH, Erlanger-Germany). Due to the sensitivity of the kit used, IL-6 values <1.5 were defined as 1.4 for the statistical analyses.

The training load was measured in the swimming group. The adolescents reported the intensity using the rating of perceived exertion proposed by Borg in 1982 and adapted by Foster (20) and the volume of training sessions (in hours and minutes) during the entire month (considering the days of training) prior to data collection. This method has been applied previously in studies involving adolescent swimmers, (6) and the daily training load was obtained by multiplying the rating of the perceived exertion score by the training volume (in minutes). The sum of the daily training loads was used to obtain the monthly training load used in this study.

2.4 Statistical analysis

Descriptive statistics included means, standard deviations, and confidence intervals (95% CIs). A generalized estimating equation (GEE) was used to compare variables between groups by applying linear or gamma scale responses, when appropriate, and considering poshoc of Bonferroni. All comparisions were performed in SPSS (version 24.0). Since the adjusted GEE analysis do not provide standard deviations (SD), these values were calculated using the following formula: $SD=SEM x\sqrt{n}$ (21), where SEM was obtained from stardard error (std. error). In sequence, it was calculated the SD pooled applying the formula: $S{D}_{\text{pooled}}=\sqrt{\frac{\left(n_1-1\right)S{D}_1^2+\left(n_2-1\right)S{D}_2^2}{n_1+n_2-2}}$. Lastly, the cohen’s d effect sizes were calculated following the equation: $d=\frac{{\text{mean}}_1-{\text{mean}}_2}{S{D}_{\text{pooled}}}$. The magnitude of effect size was calculated considering the following cut-off points: trivial ≤ 0.20, small = 0.20–0.59, moderate = 0.60–1.19, large =1.20–1.99, and very large ≥ 2.00 (22). Mediation analyses were performed using the model proposed by Valeri and Vanderweele (23) and applying the “med4way” command in Stata (version 14.0) and the theorical model is presented in Figure 1. This method decomposed the total association between training load and aBMD into direct (i.e., the direct effect of training load on aBMD through pathways not related to inflammatory markers) and pure indirect (i.e. the effect of training load on aBMD mediated by inflammatory markers) effects. In addition, we analysed the reference (i.e. the effect of training load on aBMD due to interactions with inflammatory markers) and mediated (i.e. the effect of training load on aBMD due to both mediation by and interaction with inflammatory markers) interactions. All mediation analyses were adjusted for chronological age and APHV. For all analyses, the significance level was set at p<0.05.

Figure 1

Figure 1. Theoretical framework of the mediation model. Independent variable: training load; dependent variable: areal bone mineral density (aBMD); mediator variables: inflammatory markers (IL-6 or CRP). Confounders: chronological age and APHV.

3 Results

The characteristics of the participants are shown in Table 1. Swimmers and control groups presented similar values for all anthropometric, maturational, and body composition variables in crude comparisons, except for body mass (p=0.022), fat mass in kilograms (p=0.004), and fat mass percentage (p=0.040). Table 2 presents the adjusted comparisons of aBMD, bone markers, and hormone/inflammatory variables between the swimmers and the control group. Swimmers showed lower aBMD in the lower limbs (p=0.011, d=0.59, small effect size), spine (p<0.001, d=1.09, moderate effect size), and total body less the head (TBLH; p<0.001, d=0.80, moderate effect size) than the control group. Regarding the bone markers, swimmers had higher calcium (p=0.002, d=0.74, moderate effect size) and vitamin D (p<0.001, d=2.18, very large effect size) concentrations. The same pattern was observed for hormonal variables such as IGF-1 (p<0.001, d=1.84, large effect size).

Table 1

Table 1. Descriptive characteristics of the participants stratified by swimming participation (n=61).

Table 2

Table 2. Comparison of markers by swimming participation (n=61).

Finally, the mediation analyses presented in Table 3 were performed to confirm the absence of a mediating effect of inflammatory markers on the association between training load and aBMD among swimmers. In the mediation models of IL-6, a direct effect was found between training load and aBMD in the lower limbs (β=-0.153, p=0.025) and TBLH (β=-0.098, p=0.034), while no indirect effect of IL-6 was found (p>0.05). The mediation models of CRP showed no indirect effect (p>0.05), while the training load directly affected aBMD in the upper limbs (β=-0.013, p=0.019), lower limbs (β=-0.024, p=0.009), total spine (β=-0.015, p=0.034), and TBLH (β=-0.017, p=0.005).

Table 3

Table 3. Mediation models of the association between training load and aBMD by inflammatory markers (IL-6 and CRP).

4 Discussion

This cross-sectional study aimed to analyse whether adolescent swimmers present different concentration patterns of inflammatory, immunological, anabolic, and bone markers compared with adolescents who do not practice organised sports and whether inflammatory variables mediate the association between training load and aBMD in adolescent swimmers. The main findings showed that although swimmers present lower BMD values in most body regions, they have higher concentrations of calcium, 25-hydroxy vitamin D, and IGF-1. Finally, in addition to the groups showing similar values for bone metabolism, immunological markers, and inflammatory variables, these inflammatory markers did not mediate the negative association between training load and BMD in swimmers.

The literature consistently states that swimming does not appear to be an effective sport for improving BMD (4), and our findings corroborate this statement. However, while some studies have shown a neutral effect of sports compared to controls (24) or a gain but with lower magnitude (25), others have reported a considerable negative effect, (5, 26) including the present study, where swimmers presented lower BMD compared to the non-sports group, except for in the upper limbs. To support this statement, a recent meta-analyses (27), demonstrated that in human models, swimming can present a neutral or negative effect on bone depending to anatomical region. The negative effect seems to occurs mainly in studies that include swimmers with a higher frequency of weekly training and consequently, a higher training load. Therefore, it was speculated that the physiological and inflammatory consequences of training could partially explain this phenomena, in addition to the environment without mechanical stimulation. Nonetheless, our results do not confirm these hypotheses and reinforce the fact that swimmers have lower aBMD, especially in the hypogravitational environment (4).

Competitive swimmers have a higher training load and consequently, higher daily training volume (i.e., time in water) (2). Therefore, the longer the time in the water (hypogravitational environment), the shorter the daily duration of mechanical stimulation from weight-bearing activities on land, even those not directly linked to sports practice. Consequently, less stimulation of bone tissue and lower BMD were observed in adolescents who did not practice sports. This result is interesting because it reinforces the idea that despite the high muscle activity during swimming, which is a significant predictor of BMD (28, 29), these muscle contractions do not seem to be effective in bone tissue, mainly in the lower limbs and whole body. In fact, studies have found that the relationship between the muscle and bone unit is different in swimmers (24), because the sport does not involve eccentric muscle contractions, in addition to low strain magnitude/strain during practice (30–32). Thus, dryland training should be recommended for swimmers, which can assist performance and is beneficial for the accumulation of bone mass (4, 33).

Furthermore, calcium, vitamin D, and IGF-1 were the only blood variables that showed concentration differences in the swimmers. Young athletes at a competitive level usually receive nutritional monitoring of calcium and vitamin D to improve the athlete’s health performance and avoid growth and developmental delays (34). Therefore, although we did not control for food intake, a diet rich in these nutrients justifies these results (34). In contrast, high IGF-1 concentrations appear to be associated with swimming practice. The concentrations of peptide and protein hormones, such as IGF-1, increase after physical exercise (35, 36); therefore, it is expected that swimmers will have higher concentrations of these hormones. However, despite its anabolic effect on different tissues such as muscle and bone, swimmers had lower BMD.

Despite the innovative aspects of the study, such as comparisons of different blood markers and testing the possible mediating effect of inflammatory markers, some limitations need to be acknowledged. First, the cross-sectional design prevented cause-and-effect interpretations. Second, the study did not analyse the food intake of the participants, which could indirectly affect any of the analysed markers. In addition, the sample size did not allow for analysis stratified by sex and age groups. Future studies are encouraged to explore the association between training load and BMD, as well as its potential mediators, in adolescents of different age groups (e.g., 10–14 and 15–19 years), stratified by sex and considering athletes with higher levels of sport engagement. The IL-6 assessment also had limitations, as the kit sensitivity did not detect values ​​below 1.5. Lastly, the method used to estimate the training load was subjective, and therefore presents greater limitations compared to objective methods previously applied in studies involving swimmers (37).

In summary, although adolescent swimmers at the competitive level had lower BMD in most body segments than those in the non-sports group, they did not show differences in bone turnover, inflammatory, or immunological markers. In contrast, swimmers presented with higher concentrations of IGF-1, calcium, and 25-hydroxy vitamin D. Furthermore, although the training load was negatively associated with the aBMD, IL-6 and CRP levels did not mediate this association. These results have two possible explanations. First, other biological markers need to be checked for an association between the training load and bone to confirm the findings. In contrast, this could reinforce the hypothesis that swimmers have lower aBMD, especially due to the hypogravitational environment, and should be encouraged to include dryland training in their training routines.

Data availability statement

The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

Ethics statement

The studies involving humans were approved by Ethical Research Committee of the São Paulo State University-UNESP. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

Author contributions

RA: Writing – review & editing, Writing – original draft. PN: Writing – review & editing, Writing – original draft. MC-e-S: Writing – review & editing. RB: Writing – review & editing. LG: Writing – review & editing. BT: Writing – review & editing. RF: Writing – review & editing. DV: Writing – original draft, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by the São Paulo Research Foundation-FAPESP (2017/09182-5, 2018/24164-6, and 2015/19710-3). Pedro Henrique Narciso received a grant from the FAPESP (2022/08875-5).

Acknowledgments

The authors gratefully acknowledge São Paulo Research Foundation (FAPESP), Coordenação de Aperfeicoamento de Pessoal de Nível Superior- Brasil (CAPES), and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). Finally, the authors acknowledge the efforts of the participants and their parents, coaches, and all researchers from the Laboratory of Investigation in Exercise (LIVE) for their support in data collection.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that Generative AI was used in the creation of this manuscript. The Generative AI was used only to check English grammar when necessary and to perform equations. All content was written by the authors.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

1. Yan C, Moshage SG, and Kersh ME. Play during growth: the effect of sports on bone adaptation. Curr Osteoporos Rep. (2020) 18:684–95. doi: 10.1007/s11914-020-00632-0, PMID: 33084999

PubMed Abstract | Crossref Full Text | Google Scholar

2. Agostinete RR, Fernandes RA, Narciso PH, Maillane-Vanegas S, Werneck A, and Vlachopoulos D. Categorizing 10 sports according to bone and soft tissue profiles in adolescents. Med Sci Sports Exerc. (2020) 52:2673–81. doi: 10.1249/MSS.0000000000002420, PMID: 32735110

PubMed Abstract | Crossref Full Text | Google Scholar

3. Baxter-Jones AD, Faulkner RA, Forwood MR, Mirwald RL, and Bailey DA. Bone mineral accrual from 8 to 30 years of age: An estimation of peak bone mass. J Bone Mineral Res. (2011) 26:1729–39. doi: 10.1002/jbmr.412, PMID: 21520276

PubMed Abstract | Crossref Full Text | Google Scholar

4. Gomez-Bruton A, Montero-Marín J, González-Agüero A, García-Campayo J, Moreno LA, Casajús JA, et al. The effect of swimming during childhood and adolescence on bone mineral density: A systematic review and meta-analysis. Sports Med. (2016) 46:365–79. doi: 10.1007/s40279-015-0427-3, PMID: 26607734

PubMed Abstract | Crossref Full Text | Google Scholar

5. Agostinete RR, Vlachopoulos D, Werneck AO, Maillane-Vanegas S, Lynch KR, Naughton G, et al. Bone accrual over 18 months of participation in different loading sports during adolescence. Arch Osteoporos. (2020) 15:64. doi: 10.1007/s11657-020-00727-2, PMID: 32335776

PubMed Abstract | Crossref Full Text | Google Scholar

6. Agostinete RR, Maillane-Vanegas S, Lynch KR, Turi-Lynch B, Coelho-e-Silva MJ, Campos EZ, et al. The impact of training load on bone mineral density of adolescent swimmers: A structural equation modeling approach. Pediatr Exerc Sci. (2017) 29:520–8. doi: 10.1123/pes.2017-0008, PMID: 28605234

PubMed Abstract | Crossref Full Text | Google Scholar

7. Ribeiro-Dos-Santos MR, Lynch KR, Agostinete RR, Maillane-Vanegas S, Turi-Lynch B, Ito IH, et al. Prolonged practice of swimming is negatively related to bone mineral density gains in adolescents. J Bone Metab. (2016) 23:149–55. doi: 10.11005/jbm.2016.23.3.149, PMID: 27622179

PubMed Abstract | Crossref Full Text | Google Scholar

8. Lancaster GI, Halson SL, Khan Q, Drysdale P, Wallace F, Jeukendrup AE, et al. Effects of acute exhaustive exercise and chronic exercise training on type 1 and type 2 T lymphocytes. Exerc Immunol Rev. (2004) 10:91–106., PMID: 15633589

PubMed Abstract | Google Scholar

9. Walsh NP, Gleeson M, Pyne DB, Nieman DC, Dhabhar FS, Shephard RJ, et al. Position statement. Part two: Maintaining immune health. Exerc Immunol Rev. (2011) 17:64–103., PMID: 21446353

PubMed Abstract | Google Scholar

10. Cerqueira É, Marinho DA, Neiva HP, and Lourenço O. Inflammatory effects of high and moderate intensity exercise-A systematic review. Front Physiol. (2020) 10:1550. doi: 10.3389/fphys.2019.01550, PMID: 31992987

PubMed Abstract | Crossref Full Text | Google Scholar

11. Nemet D, Oh Y, Kim H-S, Hill M, and Cooper DM. Effect of intense exercise on inflammatory cytokines and growth mediators in adolescent boys. Pediatrics. (2002) 110:681–9. doi: 10.1542/peds.110.4.681, PMID: 12359780

PubMed Abstract | Crossref Full Text | Google Scholar

12. Locatelli V and Bianchi VE. Effect of GH/IGF-1 on bone metabolism and osteoporsosis. Int J Endocrinol. (2014) 2014:235060. doi: 10.1155/2014/235060, PMID: 25147565

PubMed Abstract | Crossref Full Text | Google Scholar

13. Hashizume M, Hayakawa N, and Mihara M. IL-6 trans-signalling directly induces RANKL on fibroblast-like synovial cells and is involved in RANKL induction by TNF-alpha and IL-17. Rheumatol (Oxford). (2008) 47:1635–40. doi: 10.1093/rheumatology/ken363, PMID: 18786965

PubMed Abstract | Crossref Full Text | Google Scholar

14. Rincon M. Interleukin-6: from an inflammatory marker to a target for inflammatory diseases. Trends Immunol. (2012) 33:571–7. doi: 10.1016/j.it.2012.07.003, PMID: 22883707

PubMed Abstract | Crossref Full Text | Google Scholar

15. Bermudez EA, Rifai N, Buring J, Manson JE, and Ridker PM. Interrelationships among circulating interleukin-6, C-reactive protein, and traditional cardiovascular risk factors in women. Arterioscler Thromb Vasc Biol. (2002) 22:1668–73. doi: 10.1161/01.ATV.0000029781.31325.66, PMID: 12377747

PubMed Abstract | Crossref Full Text | Google Scholar

16. Cayres SU, Werneck AO, Urban JB, Turi-Lynch BC, Barbosa MF, and Fernandes RA. Sports participation is inversely associated with C-reactive protein levels in adolescents: ABCD Growth Study. Scand J Med Sci Sports. (2019) 29:1000–5. doi: 10.1111/sms.13418, PMID: 30883960

PubMed Abstract | Crossref Full Text | Google Scholar

17. McKay AKA, Stellingwerff T, Smith ES, Martin DT, Mujika I, Goosey-Tolfrey VL, et al. Defining training and performance caliber: A participant classification framework. Int J Sports Physiol Perform. (2022) 17:317–31. doi: 10.1123/ijspp.2021-0451, PMID: 34965513

PubMed Abstract | Crossref Full Text | Google Scholar

18. Moore SA, McKay HA, Macdonald H, Nettlefold L, Baxter-Jones ADG, Cameron N, et al. Enhancing a somatic maturity prediction model. Med Sci Sports Exerc. (2015) 47:1755–64. doi: 10.1249/MSS.0000000000000588, PMID: 25423445

PubMed Abstract | Crossref Full Text | Google Scholar

19. Narciso PH, Werneck AO, Luiz-de-Marco R, Ventura Faustino-da-Silva YDS, Maillane-Vanegas S, Agostinete RR, et al. Influential role of lean soft tissue in the association between training volume and bone mineral density among male adolescent practitioners of impact-loading sports: ABCD Growth study. BMC Pediatr. (2020) 20:496. doi: 10.1186/s12887-020-02402-4, PMID: 33115423

PubMed Abstract | Crossref Full Text | Google Scholar

20. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, et al. A new approach to monitoring exercise training. J Strength Cond Res. (2001) 15:109–15., PMID: 11708692

PubMed Abstract | Google Scholar

21. Higgins J and Deeks J. Chapter 7: Selecting studies and collecting data. In: Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons, Chichester, UK (2008).

Google Scholar

22. Hopkins WG, Marshall SW, Batterham AM, and Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. (2009) 41:3–13. doi: 10.1249/MSS.0b013e31818cb278, PMID: 19092709

PubMed Abstract | Crossref Full Text | Google Scholar

23. Valeri L and VanderWeele TJ. Mediation analysis allowing for exposure–mediator interactions and causal interpretation: Theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods. (2013) 18:137–50. doi: 10.1037/a0031034, PMID: 23379553

PubMed Abstract | Crossref Full Text | Google Scholar

24. Gomez-Bruton A, Gonzalez-Aguero A, Matute-Llorente A, Gomez-Cabello A, Casajus JA, and Vicente-Rodríguez G. Longitudinal effects of swimming on bone in adolescents: a pQCT and DXA study. Biol Sport. (2017) 34:361–70. doi: 10.5114/biolsport.2017.69824, PMID: 29472739

PubMed Abstract | Crossref Full Text | Google Scholar

25. Ferreira FA, Santos CC, Palmeira AL, Fernandes RJ, and Costa MJ. Effects of swimming exercise on early adolescents’ Physical conditioning and physical health: A systematic review. J Funct Morphol Kinesiol. (2024) 9:158. doi: 10.3390/jfmk9030158, PMID: 39311266

PubMed Abstract | Crossref Full Text | Google Scholar

26. Maillane-Vanegas S, Agostinete RR, Lynch KR, Ito IH, Luiz-de-Marco R, Rodrigues-Junior MA, et al. Bone mineral density and sports participation. J Clin Densitometry. (2020) 23:294–302. doi: 10.1016/j.jocd.2018.05.041, PMID: 30068490

PubMed Abstract | Crossref Full Text | Google Scholar

27. Freitas L, Bezerra A, Boppre G, Amorim T, Fernandes RJ, and Fonseca H. Does swimming exercise impair bone health? A systematic review and meta-analysis comparing the evidence in humans and rodent models. Sports Med. (2024) 54:2373–94. doi: 10.1007/s40279-024-02052-x, PMID: 38900358

PubMed Abstract | Crossref Full Text | Google Scholar

28. Ubago-Guisado E, Vlachopoulos D, Fatouros IG, Deli CK, Leontsini D, Moreno LA, et al. Longitudinal determinants of 12-month changes on bone health in adolescent male athletes. Arch Osteoporos. (2018) 13:106. doi: 10.1007/s11657-018-0519-4, PMID: 30306385

PubMed Abstract | Crossref Full Text | Google Scholar

29. Vlachopoulos D, Ubago-Guisado E, Barker AR, Metcalf BS, Fatouros IG, Avloniti A, et al. Determinants of bone outcomes in adolescent athletes at baseline: the PRO-BONE study. Med Sci Sports Exerc. (2017) 49:1389–96. doi: 10.1249/MSS.0000000000001233, PMID: 28166122

PubMed Abstract | Crossref Full Text | Google Scholar

30. Gomez-Bruton A, Gonzalez-Aguero A, Matute-Llorente A, Lozano-Berges G, Gomez-Cabello A, Moreno LA, et al. The muscle-bone unit in adolescent swimmers. Osteoporos Int. (2019) 30:1079–88. doi: 10.1007/s00198-019-04857-3, PMID: 30729250

PubMed Abstract | Crossref Full Text | Google Scholar

31. Hawkins SA, Schroeder ET, Wiswell RA, Jaque SV, Marcell TJ, and Costa K. Eccentric muscle action increases site-specific osteogenic response. Med Sci Sports Exerc. (1999) 31:1287–92. doi: 10.1097/00005768-199909000-00009, PMID: 10487370

PubMed Abstract | Crossref Full Text | Google Scholar

32. Meakin LB, Price JS, and Lanyon LE. The contribution of experimental in vivo models to understanding the mechanisms of adaptation to mechanical loading in bone. Front Endocrinol (Lausanne). (2014) 5:154. doi: 10.3389/fendo.2014.00154, PMID: 25324829

PubMed Abstract | Crossref Full Text | Google Scholar

33. Agostinete RR, Werneck AO, Narciso PH, Ubago-Guisado E, Coelho-e-Silva MJ, Bielemann RM, et al. Resistance training presents beneficial effects on bone development of adolescents engaged in swimming but not in impact sports: ABCD Growth Study. BMC Pediatr. (2024) 24:247. doi: 10.1186/s12887-024-04634-0, PMID: 38594697

PubMed Abstract | Crossref Full Text | Google Scholar

34. Smith JW and Jeukendrup A. Performance nutrition for young athletes. In: Bagchi D, Nair S, and Sen CK, editors. Nutrition and Enhanced Sports Performance. London: Elsevier (2013). p. 523–9. doi: 10.1016/B978-0-12-396454-0.00055-2

Crossref Full Text | Google Scholar

35. Moghetti P, Bacchi E, Brangani C, Donà S, and Negri C. Metabolic effects of exercise. Front Horm Res. (2016) 47:44–57. doi: 10.1159/000445156, PMID: 27348753

PubMed Abstract | Crossref Full Text | Google Scholar

36. Rojas Vega S, Knicker A, Hollmann W, Bloch W, and Strüder HK. Effect of resistance exercise on serum levels of growth factors in humans. Horm Metab Res. (2010) 42:982–6. doi: 10.1055/s-0030-1267950, PMID: 21053157

PubMed Abstract | Crossref Full Text | Google Scholar

37. Morgado JM, Rama L, Silva I, de Jesus Inácio M, Henriques A, Laranjeira P, et al. Cytokine production by monocytes, neutrophils, and dendritic cells is hampered by long-term intensive training in elite swimmers. Eur J Appl Physiol. (2012) 112:471–82. doi: 10.1007/s00421-011-1966-4, PMID: 21584686

PubMed Abstract | Crossref Full Text | Google Scholar