SEARCH

SEARCH BY CITATION

Keywords:

  • Artistic Gymnastics;
  • bone mineral content;
  • childhood;
  • Dual-Energy X-ray Absorptiometry;
  • Recreational

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Competitive female gymnasts have greater bone mineral measures than nongymnasts. However, less is known about the effect of recreational and/or precompetitive gymnastics participation on bone development. The purpose of this study was to investigate whether the differences previously reported in the skeleton of competitive female gymnasts are also demonstrated in young children with a current or past participation history in recreational or precompetitive gymnastics. One hundred and sixty-three children (30 gymnasts, 61 ex-gymnasts, and 72 nongymnasts) between 4 and 6 years of age were recruited and measured annually for 4 years (not all participants were measured at every occasion). Total-body (TB), lumbar spine (LS), and femoral neck (FN) bone mineral content (BMC) were measured by dual-energy X-ray absorptiometry (DXA). Multilevel random-effects models were constructed and used to predict differences in TB, LS, and FN BMC between groups while controlling for differences in body size, physical activity, and diet. Gymnasts had 3% more TB and 7% more FN BMC than children participating in other recreational sports at year 4 (p < .05). No differences were found at the LS between groups, and there were no differences between ex-gymnasts' and nongymnasts' bone parameters (p > .05). These findings suggest that recreational and precompetitive gymnastics participation is associated with greater BMC. This is important because beginner gymnastics skills are attainable by most children and do not require a high level of training. Low-level gymnastics skills can be implemented easily into school physical education programs, potentially affecting skeletal health. © 2011 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Gymnastics training results in unique high mechanical loading to the skeleton and therefore provides an excellent model for assessing the effects of weight-bearing physical activity on bone mineral development. Most studies have demonstrated that competitive adolescent, collegiate, and retired female gymnasts have greater areal bone mineral density (aBMD, g/cm2) and bone mineral content (BMC, g) than other athletic and nonathletic populations.1–6 Furthermore, total and regional aBMD is greater in gymnasts with higher exposure to gymnastics, that is, greater hours or years of training.7, 8 Studies examining gymnasts' bones generally have been cross-sectional and focused on adolescent female competitive athletes who had been systematically training for at least 2 years and who trained a minimum of 15 hours per week.3–5, 9 Although competitive adolescent gymnasts are known to have greater bone strength in adolescence, little is known about their bone properties in young childhood. In addition, competitive gymnastics is a high-level sport, and participation is limited to a select number of skilled individuals. Recreational gymnastics, on the other hand, is attainable by most children, starts at a very young age, and does not require a high level of training. However, less is known about the effect of recreational and/or precompetitive gymnastics participation (ie, low-level gymnastics exposure) on male and female bone mineral accrual during childhood.

To our knowledge, only one study has examined the effect of recreational gymnastics participation on bone accrual. Laing and colleagues8 found that 4- to 8-year-old girls participating in 1 hour of recreational gymnastics per week gained more aBMD at the lumbar spine and bone area at the forearm over 2 years than children participating in nongymnastic activities. However, it is well documented that aBMD does not adjust adequately for bone size, which is particularly problematic when examining growing children and may lead to underestimation of the impact of gymnastics participation on bone mineral accrual.10, 11 There is also a paucity of research examining the effect of gymnastics participation at both the competitive and recreational levels on male bone mineral accrual. Therefore, the aim of this study was to investigate whether the differences previously reported in the skeleton of competitive adolescent female gymnasts are also demonstrated in young children with a current or past participation history in recreational or precompetitive gymnastics. We hypothesized that young male and female gymnasts with low levels of past or current gymnastics exposure (on average 1.5 hours per week at baseline) would have greater bone mineral accrual than children involved in nongymnastic recreational sports.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Study design

Participants were part of a mixed longitudinal study performed at the University of Saskatchewan between 2006 and 2010. At study entry, three cohorts were identified: 4, 5, and 6 years of age. Data were collected annually for 4 years. Additional participants were recruited during the second and third years of data collection to increase participant numbers at different ages. Because there were overlaps in ages between the clusters, it was possible to estimate a consecutive 6-year developmental pattern (4 to 10 years) over a shorter 4-year period.

Participants were excluded if they had any condition that prevented them from performing exercise safely or any condition known to affect bone development (eg, heart conditions or neurologic or musculoskeletal problems). Informed consent was obtained from all parents or guardians, and verbal assent was obtained from all children. This study was approved by the University of Saskatchewan's Biomedical Research Ethics Board (Bio 06–111).

Participants

One hundred and seventy-eight participants were recruited. To be included in this analysis, participants required complete anthropometric, body composition, and lifestyle data; 163 participants (92%) fulfilled these requirements and are presented here. Table 1 provides a breakdown of eligible participants by age and sex for each testing year; it should be noted that not all participants were present at every testing occasion. Of the 163 participants, 95% were white, 2% Asian, and 3% other (biracial). Gymnasts were recruited from the recreational and precompetitive programs of three competitive gymnastics clubs in Saskatoon, Saskatchewan, and the University of Saskatchewan's recreational gymnastics program. Gymnasts had participated in gymnastics for 45 minutes or more per week for at least one term (4 months) at study initiation. The study was designed to examine the influence of gymnastics participation on bone development; however, since some of the gymnasts who were participating in gymnastics at study initiation subsequently did not participate further (no participation in the previous 4 months), a subgroup of ex-gymnasts was identified. Therefore, the ex-gymnast group includes individuals who were classified as gymnasts at baseline but were not participating in subsequent follow-up years. Parents of gymnasts and ex-gymnasts reported how many hours per week their children participated in gymnastics. This represents the current participation hours for gymnasts; however, since ex-gymnasts had not participated in gymnastics for the previous 4 months, their hours of training are a representation of participation prior to cessation of gymnastics. Nongymnasts were recruited from other recreational sport programs, such as swimming lessons and summer soccer, basketball, T-ball, and “sports ‘r’ fun” sport camps at the University of Saskatchewan. Nongymnasts had no exposure to gymnastics stimulus. Thus three gymnastic status groups were identified: gymnasts, ex-gymnasts, and nongymnasts.

Table 1. Mixed Longitudinal Study Design With Number of Males (Females) Measured at Each Test Year by Age Category
AgeTest yearTotal
2006–20072007–20082008–20092009–2010
422(23) 4(2) 26(25)
523(22)15(15)7(5)6(2)51(44)
613(21)23(20)12(15)7(8)55(64)
75(11)16(18)20(16)18(15)59(60)
8 4(12)15(17)21(17)40(46)
9  6(16)13(16)19(32)
10   4(13)4(13)
Total63(77)58(65)64(71)69(71)254(284)

Chronologic age and anthropometrics

The chronologic age of each child was recorded to the nearest 0.01 year by subtracting the decimal year of the participant's date of birth from the decimal year of the day of testing. Anthropometric measurements included standing height and weight. Heights were recorded to the nearest millimeter using a wall-mounted stadiometer (Holtain Limited, Crymmych, UK) and body mass to the nearest 0.5 kg using a digital scale (Model 1631, Tanita Corp., Tokyo, Japan). All measures were performed twice, and if the difference was greater than 0.4, a third measure was recorded. The mean or median then was reported depending on whether two or three measures were recorded, respectively.12 All measures were performed by the same Canadian Society for Exercise Physiology–certified exercise physiologist.

Physical activity and dietary assessment

Physical activity was assessed using the previously validated Netherlands Physical Activity Questionnaire (NPAQ).13, 14 Parents were asked to report their child's current physical activity level. The NPAQ proxy report includes items about activity preferences and everyday activity choices rather than a specific recall of physical activity.14 Questionnaire responses range from 7 (low physical activity) to 35 (high physical activity). Calcium and vitamin D intakes were assessed through the use of a 24-hour recall questionnaire. Dietary data were analyzed using the Food Processor and Nutritional Software Version 8.5 (ESHA Research Software, Salem, OR, USA). The 24-hour recall has been suggested as a suitable method to assess individual nutrient intakes of children.15

Dual energy X-ray absorptiometry

Body composition measurements were performed using a Hologic Discovery Wi dual-energy X-ray absorptiometry (DXA) scanner (Hologic, Inc., Waltham, MA, USA). Three different scans were performed: total-body (TB), lumbar spine (LS), and femoral neck (FN). The TB scans are presented as TB and TB less head (TBLH). The International Society for Clinical Densitometry recommends that the head be excluded when calculating bone mass measurements of the whole body in children and adolescents.16 However, the majority of pervious research has reported TB BMC with the head included. Therefore, both methods are presented here. Bone mineral content (BMC, g), lean mass (kg), and fat mass (kg) were derived from the scans. All scans were administered and analyzed by a certified radiology technologist. Quality-control phantom scans were performed daily. The coefficients of variation (CV, %) for these measures from our laboratory, based on duplicate measures in 30 young healthy female university students (20 to 30 years of age), were 0.5% for whole-body BMC, 0.7% for lumbar spine BMC, and 1.0% for the proximal femur BMC. Fat and lean tissue mass was assessed from the whole-body scans. Our laboratory has determined CVs for these measures to be 3.0% and 0.5%, respectively.

Statistical analysis

Variables are presented as means and SDs. Group differences (gymnasts versus ex-gymnasts versus nongymnasts) for height, weight, total-body fat mass, total-body lean mass, calcium, vitamin D, and physical activity were assessed in each age category by analysis of variance. Group differences (gymnasts versus nongymnasts) were assessed using ANCOVA (covariates: age, sex, height, weight, physical activity, calcium, and vitamin D) at the first testing occasion (2006–2007). Statistical analysis was performed using SPSS software Version 18.0 (SPSS, Inc., Chicago, IL, USA), and α was set at 0.05.

For the longitudinal analyses, hierarchical (multilevel) random-effects models were constructed using a multilevel modeling approach (MlwiN Version 1.0, Multilevel Models Project; Institute of Education, University of London, London, UK). A detailed description of multilevel modeling is presented elsewhere.17 In brief, bone mineral accrual and bone area were measured repeatedly in individuals (level 1 of hierarchy) and between individuals (level 2 of hierarchy). Analysis models that contain variables measured at different levels of hierarchy are known as multilevel regression models. Specifically, the following additive random-effects multilevel regression models were adopted to describe the developmental changes in bone mineral accrual and bone area:

  • equation image

where y is the bone mineral content or bone area parameter on measurement occasion i in the jth individual, α is a constant, βjxij is the slope of the time component (age centered around 7 years) for the jth individual, and k1 to kn are coefficients of various explanatory variables (eg, height, physical activity, hours of training, etc.) at assessment occasion i in the jth individual. Dummy variables were created for gymnastic groups with nongymnasts as the reference category. These are the fixed parameters in the model. Both µj and εij are random quantities whose means are equal to zero; they form the random parameters in the model. They are assumed to be uncorrelated and follow a normal distribution, and thus their variances can be estimated; µj is the level 2 (between-subjects variance) and εij is the level 1 residual (within-individual variance) for the ith assessment of bone mineral content in the jth individual. Models were built in a stepwise procedure; that is, predictor variables (k fixed effects) were added one at a time, and the log-likelihood ratio statistics were used to judge the effects of including further variables.17

The predictor variable coefficients (fixed variables in Table 4) were used to predict BMC (g) accrual with age for total body, femoral neck, and lumbar spine (Fig. 3). Height, weight, and physical activity scores were controlled in the prediction equations using the sex-specific averages shown in Table 2.

Table 2. Descriptive Statistics of Chronologic Age–Related Anthropometric, Body Composition, and Lifestyle Data for Gymnasts, Nongymnasts, and Ex-gymnasts
Age (years)
 45678910
  • Variable mean ± SD; Ht= height; Wt= weight; TB= total body; Vit D= vitamin D; PA score = Netherlands physical activity score.

  • a

    Gymnasts significantly different than nongymnasts.

  • b

    Gymnasts significantly different than ex-gymnasts.

Nongymnasts
 N2240544937267
 Ht (cm)105.4 ± 4.1110.0 ± 4.7116.7 ± 5.5121.5 ± 5.8126.3 ± 5.6133.7 ± 6.9134.9 ± 4.7
 Wt (kg)17.6 ± 1.319.8 ± 2.421.8 ± 2.823.7 ± 3.126.1 ± 3.530.7 ± 4.834.0 ± 4.0
 TB fat (kg)4.3 ± 0.84.7 ± 1.44.5 ± 1.34.6 ± 1.35.1 ± 1.76.3 ± 2.38.4 ± 2.8
 TB lean (kg)12.2 ± 1.013.8 ± 1.815.8 ± 2.217.4 ± 2.519.4 ± 2.722.5 ± 3.623.4 ± 3.0
 Calcium (mg)906.4 ± 231.91038.4 ± 91.31030.9 ± 444.11091.1 ± 605.2919.4 ± 549.9903.9 ± 330.11021.1 ± 775.4
 Vit D (IU)188.9 ± 129.2231.3 ± 152.3200.1 ± 140.2189.6 ± 136.5185.3 ± 179.1211.8 ± 129.4202.0 ± 150.5
 PA score24.4 ± 3.924.8 ± 3.524.9 ± 3.925.5 ± 3.824.8 ± 4.124.8 ± 5.025.6 ± 2.5
Gymnasts
 N91318171083
 Ht (cm)103.5 ± 3.9107.1 ± 5.0113.4 ± 4.2a,b119.9 ± 5.2124.9 ± 5.1128.9 ± 5.7134.0 ± 5.3
 Wt (kg)18.0 ± 3.118.7 ± 2.721.0 ± 3.323.8 ± 4.626.8 ± 5.528.9 ± 9.028.5 ± 3.1
 TB fat (kg)4.1 ± 1.34.3 ± 0.74.4 ± 1.24.6 ± 1.45.4 ± 2.36.1 ± 4.84.4 ± 0.7
 TB lean (kg)12.7 ± 1.813.2 ± 1.815.5 ± 2.417.7 ± 3.219.6 ± 3.221.0 ± 3.922.1 ± 1.9
 Calcium (mg)908.2 ± 379.71097.0 ± 425.9974.7 ± 370.81041.4 ± 417.4761.0 ± 357.9937.7 ± 207.7773.3 ± 103.9
 Vit D (IU)193.9 ± 121.6197.5 ± 109.6200.3 ± 151.9158.6 ± 122.2129.5 ± 95.7115.8 ± 91.4318.3 ± 138.0
 PA score25.6 ± 3.025.7 ± 2.925.4 ± 3.025.8 ± 2.726.0 ± 5.024.4 ± 3.826.7 ± 4.2
Ex-gymnasts
 N2042475339177
 Ht (cm)104.4 ± 5.1111.2 ± 6.1117.9 ± 6.2122.7 ± 6.6129.6 ± 7.0133.6 ± 7.4134.2 ± 7.3
 Wt (kg)18.1 ± 2.720.5 ± 3.623.4 ± 4.524.9 ± 4.128.2 ± 5.230.1 ± 5.529.9 ± 4.8
 TB fat (kg)4.2 ± 1.24.7 ± 1.75.1 ± 2.24.8 ± 2.05.7 ± 2.46.0 ± 2.75.3 ± 2.0
 TB lean (kg)12.9 ± 1.714.5 ± 2.416.8 ± 2.618.5 ± 3.020.6 ± 3.522.2 ± 3.622.5 ± 3.3
 Calcium (mg)1145.5 ± 467.81089.7 ± 478.61010.7 ± 424.31013.5 ± 447.21029.7 ± 409.61119.0 ± 532.6666.13 ± 321.3
 Vit D (IU)244.0 ± 176.3221.6 ± 137.4227.9 ± 202.6208.6 ± 155.7196.3 ± 152.2190.3 ± 79.0158.7 ± 78.8
 PA score24.8 ± 2.924.9 ± 2.625.0 ± 3.225.8 ± 3.426.0 ± 3.526.2 ± 2.927.6 ± 3.3

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

At the first measurement occasion, gymnasts participated approximately 1.5 ± 1.2 hours per week in gymnastics and had been training for 1.0 ± 1.1 years. At the last measurement occasion (2010), gymnasts, on average, were participating in gymnastics for 4.6 ± 4.2 hours per week and had been training for 4.6 ± 1.3 years. Ex-gymnasts had participated in gymnastics for approximately 1.6 ± 1.7 hours per week for approximately 2.3 ± 1.1 years. Anthropometric, body composition, and dietary data, as well as physical activity scores, are presented in Table 2. There were no significant differences (p > .05) between groups for the variables presented, with the exception of gymnasts, who were significantly shorter than both ex-gymnasts and nongymnasts at 6 years of age (p < .05). Table 3 summarizes the results from the first year of testing (2006–2007). There were no differences between gymnasts and nongymnasts unadjusted (data not shown) and adjusted bone mineral content at any site measured (p > .05).

Table 3. Adjusted Bone Mineral Content Values for Gymnasts and Nongymnasts at the First Testing Occasion (2006–2007) (Covariates: Age, Sex, Height, Weight, Physical Activity, Calcium, Vitamin D)
 Gymnasts (n = 77)Nongymnasts (n = 63)
  1. Adjusted marginal mean ± SEE (standard error of the estimate) of bone mineral content in grams. TB = total body; BMC = bone mineral content; TBLH = total body less head; FN = femoral neck; LS = lumbar spine.

TB BMC (g)686.3 ± 5.5675.0 ± 6.0
TBLH BMC (g)429.2 ± 3.3421.4 ± 3.7
FN BMC (g)8.4 ± 0.28.2 ± 0.2
LS BMC (g)15.8 ± 0.215.3 ± 0.2

Table 4 summarizes the results from the multilevel models for TB, TBLH, FN, and LS bone mineral content development. The model for total-body BMC indicated that once age centered (1 year predicts 16.6 ± 4.3 g of BMC), height (1 cm predicts 7.6 ± 0.9 g of BMC), weight (1 kg predicts 7.1 ± 1.2 g of BMC), vitamin D (1 IU predicts 0.03 ± 0.1 g of BMC), and sex (females have 17.5 ± 7.7 g less BMC than males) were controlled, there was a significant independent gymnastic group effect (Table 4 and Fig. 1). Gymnasts had, on average, 27.9 ± 10.9 g more TB BMC than nongymnasts (p < .05); there was no significant difference between ex-gymnast and nongymnast TB BMC (Table 4 and Fig. 1). There also were no significant gymnastic group by age-centered interactions (p > .05). The model for total-body BMC with the head excluded (TBLH) resulted in similar findings: When age-centered, height and weight were considered, gymnasts had, on average, 22.0 ± 6.3 g more BMC than nongymnasts (p < .05). However of note, sex was not a significant predictor for the TBLH model. Similar findings were noted at the FN when age centered, height, weight, sex, and physical activity were controlled. Gymnasts had 0.15 ± 0.06 g more BMC than nongymnasts (p < .05); no significant differences were found between ex-gymnasts and nongymnasts (p > .05) (Table 4 and Fig. 1). No significant difference was found at the lumbar spine between the groups (p > .05) when age, height, and weight were controlled (Table 4 and Fig. 1). The multilevel models for TB, TBLH, FN, and LS bone area revealed no significant differences between groups in bone area development when adjusted for age, height, and weight (p > .05; data not shown).

Table 4. Multilevel Regression Analysis of Total-Body, Total Body Less Head, Femoral Neck, and Lumbar Spine BMC Development (g)
VariableTotal-body BMCTB less head BMCFemoral neck BMCLumbar spine BMC
Fixed effects
 Constant−304.2 ± 92.8−346.0 ± 62.0−1.7 ± 0.6−16.1 ± 2.9
 Age centered16.6 ± 4.313.2 ± 2.90.04 ± 0.03−0.06 ± 0.1
 Age centered21.0 ± 0.62.0 ± 0.50.01 ± 0.006NS
 Sex−17.5 ± 7.7NS−0.09 ± 0.04NS
 Height7.6 ± 0.95.7 ± 0.60.02 ± 0.0060.25 ± 0.03
 Weight7.1 ± 1.27.4 ± 0.90.02 ± 0.0070.17 ± 0.04
 Vitamin D0.03 ± 0.01NSNSNS
 CalciumNSNSNSNS
 Physical ActivityNSNS0.01 ± 0.004NS
 Hours trainedNSNSNSNS
 Gym vs NGym27.9 ± 10.922.0 ± 6.30.15 ± 0.06NS
 XGym vs NGymNSNSNSNS
  1. Note: Fixed-effect values are estimated mean coefficients ± SEE (standard error estimate) of bone mineral content in grams. Random-effects values are estimated mean variance ± SEE (bone mineral content) in grams2. Age centered (age – 7) years; sex (0 = male; 1 = female); height, cm; weight, kg; vitamin D, IU; calcium, mg; physical activity (7 = low to 35 = high). TB less head = total-body BMC – head BMC; Gym = gymnasts group; NGym = nongymnasts group; XGym = ex-gymnasts group; Age cen. = age centered; NS (p > .05) = not significant and variable removed from the final model.

Random effects
Level 1
Constant788.7 ± 70.9454.1 ± 40.90.06 ± 0.0050.89 ± 0.08
Level 2ConstantAge centeredConstantAge centeredConstantAge centeredConstantAge centered
Constant2984.1 ± 383.1431.3 ± 88.41428.1 ± 187.9274.6 ± 49.70.05 ± 0.0090.007 ± 0.0032.7 ± 0.360.41 ± 0.09
Age centered431.3 ± 88.482.7 ± 32.8274.6 ± 49.746.8 ± 19.20.007 ± 0.0030.007 ± 0.0020.41 ± 0.090.07 ± 0.03
thumbnail image

Figure 1. Predicted total-body, femoral neck, and lumbar spine BMC development in gymnasts, nongymnasts, and ex-gymnasts from the multilevel regressions (Table 4). The triangle on the graph represents gymnasts, the square is ex-gymnasts, and the asterisk is nongymnasts. Height, weight, and physical activity were held constant using values obtained from Table 2.

Download figure to PowerPoint

Age centered was added as both a fixed and random coefficient. The random-effects coefficients describe the two levels of variance [within individuals (level 1 of the hierarchy) and between individuals (level 2 of the hierarchy)]. For all three BMC models, the significant variances at level 1 of the models indicate that BMC was increasing significantly at each measurement occasion within individuals (estimate > 2 × SEE; p < .05; Table 4). The between-individuals variance matrix (level 2) for each model indicated that individuals had significantly different BMC growth curves in terms of both their intercepts (constant/constant, p < .05) and the slopes of their lines (age/age, p < .05). The variances of these intercepts and slopes were positively and significantly correlated (constant/age, p < .05) in all models. The variance between individuals therefore was different at different ages.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

This is the first prospective study to examine the effect of low-level gymnastics exposure (on average, 1.5 hours per week at baseline) on bone mineral accrual in young males and females. The aim was to investigate the development of BMC to determine if the advantages reported in the skeleton of competitive adolescent female gymnasts with high-level gymnastics exposure also were apparent in young children with a current or past participation history in recreational or precompetitive gymnastics. The main finding was that recreational and precompetitive male and female gymnasts had greater total-body and femoral neck BMC than children engaged in other recreational sports.

A cross-sectional comparison of the gymnasts versus nongymnasts at the first year revealed no differences between groups for any bone parameter. However, in the longitudinal analysis, it was found that by year 4, recreational and precompetitive gymnasts had 3% more total-body and 7% more femoral neck BMC than children participating in other recreational sports when body size, physical activity, and diet were considered. This suggests that 1 year of gymnastics participation at the recreational level was not sufficient to change these bone parameters (ie, a longer duration of stimulus is required in the present group of children). Most ex-gymnasts ceased participating in gymnastics between the first and second measurement occasions, which may be contributing to the lack of difference observed between ex-gymnasts and nongymnasts in the current cohort. Conversely, it may be that DXA is not a sensitive enough measure to detect differences from 1 year of stimulus versus a longer duration. This is supported by the cross-sectional analysis of peripheral quantitative computed tomography (pQCT) data in this cohort, which found that both gymnasts and ex-gymnasts had greater BMC at the distal radius at year 3.26 There also was no difference in bone area between groups once adjusted for age, height, and weight, suggesting that individuals have an appropriate area for their size. This finding is consistent with the previously reported pQCT data in this cohort, which also found no differences in bone area among gymnasts, nongymnasts, and ex-gymnasts.26 Furthermore, this indicates that gymnastics training did not appear to be altering bone area, only mineralization within the area.

The greater total-body and FN BMC values observed in these young male and female gymnasts are consistent with, though on a smaller magnitude, the findings previously reported in competitive adolescent, collegiate, and retired female gymnasts.1–6, 18, 19 The lower-magnitude response is not unexpected because total and regional aBMD has been shown to be greater in gymnasts with higher exposure to gymnastics (ie, greater hours or years of training), suggesting a dose-response relationship between loading and bone mass.7, 8 However, despite the fact that the current cohort is young and had a low level of gymnastics exposure, subjects had greater TB and FN BMC, suggesting that recreational and precompetitive gymnastics participation has a beneficial effect at this young age. The greater BMC also supports the assertion of Laing and colleagues8 that the stimuli received during introductory classes is sufficient to increase bone mass compared with other recreational sports.

Laing and colleagues8 were the first to examine the effect of beginner gymnastics classes on bone accrual in females. They found that 4- to 8-year-old girls participating in recreational gymnastics classes showed a significantly greater increase in lumbar spine aBMD and forearm bone area over 2 years than girls not participating in gymnastics.8 The authors suggested that beginning-level gymnastics skills performed in introductory classes seem to be adequate stimuli for enhancing gains in both bone mineral density and size. However, it is well documented that aBMD does not adjust adequately for bone size, which is particular problematic when assessing growing children.11 Therefore, in this study, a size correction was applied directly to the DXA-measured BMC value by adjusting for the participants' height and weight at each measurement occasion in the multilevel model and assessed male as well as female recreational and precompetitive gymnasts.

Most of the pervious literature has focused on females; there is limited research focusing on the effect of gymnastics training on male bone development. Daly and colleagues20 were the first to assess the effect of competitive gymnastics training on male bone parameters; they found that male gymnasts had a greater increase in calcaneal bone parameters than controls over 18 months (12.8% versus 7.2%, respectively). However, they used ultrasound and assessed broadband ultrasonic attenuation (BUA). The precise skeletal properties reflected by BUA have not been well established, and ultrasound does not measure either bone structure or material properties directly; therefore, caution should be taken when interpreting these results. To our knowledge, only one study has examined the effect of competitive gymnastics training on male and female bone parameters using DXA. Zanker and colleagues21 found that females had 8% to 10% greater aBMD at the TB, LS, and forearm; however, there was no significant difference between male gymnasts and nongymnasts at any site. The authors did observe a trend toward a higher TB and forearm aBMD in males.21 This is in contrast to this study, where male as well as female gymnasts had greater TB and FN BMC. Zanker and colleagues21 suggested that the lack of a significant difference between male gymnasts and nongymnasts may be related to their lower level of cumulative high-impact weight-bearing activities compared with female gymnasts. However, the male gymnasts in the study by Zanker and colleagues21 trained 4 to 6 hours per week and had been training for 1 to 2 years, which is greater than the current cohort at baseline and comparable with the hours of training after 4 years of participation. Our study findings, as well as those by Laing and colleagues,8 would suggest that the loading received in the study by Zanker and colleagues21 should have been sufficient to produce significant positive effects on the skeleton. The discrepancy in results may be related to the fact that Zanker and colleagues21 examined only 10 male gymnasts; thus they simply may have been underpowered to observe the effect of lower gymnastics exposure on bone parameters.

As stated previously, this study observed greater TB and FN BMC in the young male and female gymnasts than in children engaged in other sports; however, there was no difference in LS BMC between the groups. This is in contrast to the cross-sectional comparisons of competitive adolescent, collegiate, and retired female gymnasts, who have been reported to have higher total-body as well as regional bone mineral measures than other athletic and nonathletic populations of similar age, height, and weight.1–6, 19, 22 Laing and colleagues22 reported that female gymnasts 8 to 13 years of age had higher aBMD at the total body, femoral neck, and lumbar spine (6%, 14%, and 15%, respectively). The lack a significant difference between groups at the lumbar spine may be related to the young age of the gymnasts in the current cohort. The appendicular skeleton (ie, femoral neck) has been found to grow more rapidly and accrue more bone before puberty, whereas axial skeleton (ie, lumbar spine) growth is accelerated during puberty.23 Therefore, differences in LS BMC in the current cohort may become apparent as they approach puberty and axial skeletal growth accelerates. Alternatively, it also may be that the loading experienced from low-level gymnastics exposure is insufficient to increase BMC at the lumbar spine compared with other recreational activities. The gymnasts in the previously described study were competitive gymnasts who trained, on average, 11.7 hours per week and had been training for approximately 5.9 years.22

Competitive gymnasts have greater total and regional aBMD and BMC, and this has been found to persist after retirement from training and competition in former collegiate gymnasts.1, 2 As such, competitive gymnastics training and high-level gymnastics exposure potentially may delay or prevent osteoporosis and related fractures. However, competitive gymnastics is a high-level sport, and participation is limited to a select number of skilled individuals. Zanker and colleagues21 stated that owing to the great skill and physical and mental demands of competitive gymnastics, it would be unrealistic to prescribe this activity to children as a possible prophylactic to osteoporosis in adulthood. Recreational gymnastics or low-level gymnastics skills, on the other hand, are attainable by most children and do not require a high level of training. Recreational gymnastics involves the development of spatial and body awareness, muscular strength, and neuromuscular coordination. Low-level gymnastics skills can be integrated easily into school physical education programs; thus most children may benefit from this training.

Assessment of BMC with exercise studies is limited because important changes in the structural properties of bone may occur and go undetected. The primary outcome in this study, BMC, does not account for changes in the shape or structure of bone. This is important because bone strength has been shown to improve significantly independent of mineralization.24 Modeling during growth can alter endosteal and periosteal dimensions, measures of the structural properties of bone, in addition to bone mineral, and may have provided valuable additional information.25 In a cross-sectional analysis of this cohort using DXA and pQCT, it was shown that there was a modest 5% difference in total-body BMC between gymnasts and nongymnasts, with a 25% greater estimated strength at the distal radius.26 pQCT allows for an assessment of bone structure (ie, spatial distribution of bone mineral). Therefore, the effect of recreational and/or precompetitive gymnastics participation on bone strength may be underestimated without measurements of bone structure. This is supported by Jarvinen and colleagues,27 who suggest that while DXA provides a reasonable overall description of bone status, it overlooks structural bone alterations that largely and independently can influence bone strength. Longitudinal investigations of bone structural adaptation are required to better understand the effect of both competitive and recreational gymnastics training on bone development.

There are some other limitations to this study. The mixed longitudinal design does not allow for a cause-effect assessment of current or past gymnastics participation on bone parameters. Well-controlled prospective studies starting before the initiation of gymnastics participation are required to answer definitively whether the greater bone mineral measures are the result of cumulative exposure to gymnastics or present before the initiation of participation. Detailed information about the types of exercises undertaken by the gymnasts at the four different centers was not available; therefore, it was not possible to quantify the loading experienced by these young recreational and precompetitive gymnasts. The high activity level of the nongymnast group may be reducing or even masking the positive effect of gymnastics participation on bone mineral accrual. The nongymnasts in this cohort were participating in other recreational sports (such as soccer, T-ball, basketball, and karate), which may lead to increased loading in a similar manner as gymnastics at the lumbar spine and femoral neck. This is supported by similar and high physical activity scores across the groups. More active children may emerge from adolescence with 5% to 10% greater bone mass depending on the skeletal site28; therefore, if the nongymnasts in this cohort have greater activity levels and bone mass than average, the effect of recreational and precompetitive gymnastics participation on bone parameters actually may be greater than the difference reported in this cohort.

In summary, when compared with other physically active children, recreational and precompetitive gymnasts had greater total-body and femoral neck BMC. These findings are important because recreational gymnastics skills are attainable by most children and do not require a high level of training. Low-level gymnastics skills can be implemented easily into school physical education programs, and thus most children may benefit from this training, potentially developing greater total-body and femoral neck BMC. This training could have a potential impact on primary osteoporosis and fracture prevention. However, randomized, controlled trials are required to substantiate our findings.

Disclosures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

All the authors state that they have no conflicts of interest.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

We gratefully acknowledge the study participants and their families for their enthusiasm and commitment to the project. This study was supported in part by funding from the Canadian Institute of Health Research (CIHR), the Saskatchewan Health Research Foundation (SHRF), and the CIHR doctoral regional partnership program.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  • 1
    Pollock NK, Laing EM, Modlesky CM, O'Connor PJ, Lewis RC. Former college artistic gymnasts maintain higher BMD: a nine-year follow-up. Osteoporos Int. 2006; 17: 16911697.
  • 2
    Zanker CL, Osbrone C, Cooke CB, Oldroyd B, Truscott JG. Bone density, body composition and menstrual history of sedentary female former gymnasts, aged 20–32. years. Osteoporos Int. 2004; 15: 145154.
  • 3
    Proctor KL, Adams WC, Shaffrath JD, Van Loan MD. Upper-limb mineral density of female collegiate gymnasts versus controls. Med Sci Sport Exerc. 2002; 34: 18301835.
  • 4
    Robinson TL, Snow-Harter C, Taaffe DR, Gillis D, Shaw J, Marcus R. Gymnasts exhibit higher bone mass than runners despite similar prevalence of amenorrhea and oligmenorrhea. J Bone Miner Res. 2005; 10: 2635.
  • 5
    Bass S, Pearce G, Bradney M, et al. Exercise before puberty may confer residual benefits in bone density in adulthood: studies of active pre-pubertal and retired female gymnasts. J Bone Miner Res. 1998; 13: 500507.
  • 6
    Nickols-Richardson SM, O'Coner PJ, Sharpese SA, Lewis RD. Longitudinal bone mineral density changes in female child artistic gymnasts. J Bone Miner Res. 1999; 14: 9941002.
  • 7
    Scerpella TA, Davenport M, Morganti CM, Kanaley JA, Johnson LM. Dose related association of impact activity and bone mineral density in pre-pubertal girls. Calcif Tissue Int. 2003; 72: 2431.
  • 8
    Laing EM, Wilson AR, Modlesky CM, O'Conner PJ, Hall DB, Lewis RD. Initial years of recreational artistic gymnastics training improves lumbar spine bone mineral accrual in 4- to 8- year old females. J Bone Miner Res. 2005; 20: 509519.
  • 9
    Daly RM, Caine D, Bass SL, Pieter W, Broekhoff J. Growth of highly vs moderately trained competitive female artistic gymnasts. Med Sci Sport Exerc. 2005; 37: 10531060.
  • 10
    Prentice A, Parsons TJ, Cole TJ. Uncritical use of bone mineral density in absorptiometry may lead to size-related artifacts in the identification of bone mineral determinants. Am J Clin. Nutr. 1994; 60: 837842.
  • 11
    Faulkner RA, Forwood MR, Beck TJ, Mafukidze JC, Russell K, Wallace W. Strength indices of the proximal femur and shaft in prepubertal female gymnasts. Med Sci Sports Exerc. 2003; 35: 513518.
  • 12
    International Standards for Anthropometric Assessment. 2001; International Society for the Advancement of Kinanthropometry, Australia.
  • 13
    Montoye HJ, Kemper HCG, Saris WHM, Washburn RA. Measuring physical activity and energy expenditure. Human Kinetics, Champagin, IL: 1996.
  • 14
    Janz KF, Broffitt B, Levy SM. Validation evidence for the Netherlands Physical Activity Questionnaire for young children: The Iowa Bone Development. Study Res Quart Exerc Sport. 2005; 76: 363369.
  • 15
    Whiting SJ, Shrestha RK. Dietary assessment of elementary school-aged children and adolescents. J Can Diet Assoc. 1993; 54: 193196.
  • 16
    Gordon CM, Bachrach LK, Carpenter TO, Crabtree N, El-Hajj Fuleihan G, Kutilek S, Lorenc RS, Tosi LL, Ward KA, Kalkwarf HJ. Dual energy X-ray absorptiometry interpretation and reporting in children and adolescents: the 2007 ISCD Pediatric Official Positions. J Clin Densitom. 2008; 11: 4358.
  • 17
    Baxter-Jones ADG, Mirwald R. Multilevel modeling. In: Hauspie RC, Cameron N, Molinari L, eds. Methods in Human Growth Research. Cambridge University Press, Cambridge, UK: pp. 306330. 2004.
  • 18
    Dowthwaite JN, Flowers PPE, Spadaro JA, Scerpella TA. Bone geometry, density and strength indices of the distal radius reflect loading via childhood gymnastic activity. J Clin Densitom. 2007; 10: 6575.
  • 19
    Nickols-Richardson SM, Modlesky CM, O'Coner PJ, Lewis RD. Premenarcheal gymnasts possess higher bone mineral density than controls. Med Sci Sports Exerc. 2000; 32: 6369.
  • 20
    Daly RM, Rich PA, Klein R, Bass S. Effects of high-impact exercise on ultrasonic and biochemical indices of skeletal status: A prospective study in young male gymnasts. J Bone Miner Res. 1999; 14: 12221230.
  • 21
    Zanker CL, Gannon L, Cooke CB, Gee KL, Oldroyd B, Truscott JG. Difference in bone density, body composition, physical activity, and diet between child gymnasts and untrained children 7–8 years of age. J Bone Miner Res. 2003; 18: 10431050.
  • 22
    Laing EM, Massoni JA, Nickolas-Richardson SM, Modlesku CM, O'Conner PJ, Lewis RD. A prespective study of bone mass and body composition in female adolescent gymnasts. J Pediatr. 2002; 141: 211216.
  • 23
    Bass S, Pierre DD, Pearce G, Hendrich E, Taensky A, Seeman E. The differing tempo of growth in bone size, mass, and density in girls is region-specific. J Clin Invest. 1999; 104: 795804.
  • 24
    Adami S, Gatti D, Braga V, Bainchini D, Rossini M. Site-specific effects of strength training on bone structure and geometry of ultradistal radius in postmenopausal women. J Bone Miner Res. 1999; 14: 120124.
  • 25
    Petit MA, McKay HA, MacKelvie KJ, Heinonen A, Khan KM, Beck TJ. A randomized school-based jumping intervention confers site and maturity-specific benefits on bone structural properties in girls: A hip structural analysis study. J Bone Miner Res. 2002; 17: 363372.
  • 26
    Erlandson MC, Kontulainen SA, Baxter-Jones ADG. Precompetitive and recreational gymnasts have greater bone density, mass, and estimated strength at the distal radius in young childhood. Osteoporos Int. 2011; 22: 7584.
  • 27
    Jarvinen TL, Kannus P, Sievanen H. Have the DXA-based exercise studies seriously underestimated the effects of mechanical loading on bone? J Bone Miner Res. 1999; 14: 16341635.
  • 28
    Slemenda CW, Miller JZ, Hui SL, Reister TK, Johnston CC. Role of physical activity in the development of skeletal mass in children. J Bone Miner Res. 1991; 6: 1122711233.