Secondary Logo

Journal Logo

Physical Activity Increases Bone Mineral Density in Children with Type 1 Diabetes


Medicine & Science in Sports & Exercise: July 2012 - Volume 44 - Issue 7 - p 1206–1211
doi: 10.1249/MSS.0b013e3182496a25

Introduction Osteoporosis is a growing health problem in patients with type 1 diabetes mellitus (T1DM). The aim of this study was to determine the effects of a 9-month weight-bearing physical activity program on bone mineral density (BMD) and bone biomarkers in T1DM compared with healthy children.

Methods This was a randomized controlled trial including 27 diabetic and 32 healthy children (mean age = 10.5 ± 2.5 yr). Both T1DM and healthy participants were randomized to either an exercise or a control group (i.e., four groups). At baseline and 9 months, total body (TB), lumbar spine (LS2–LS4), femoral neck, and greater trochanter areal BMD (aBMD) and serum bone biomarkers (osteocalcin, type 1 collagen cross-linking) were measured. The intervention consisted of two 90-min sessions per week of weight-bearing physical activity (ball games, jumping, rope skipping, and gymnastics).

Results Baseline variables were similar among groups. At 9 months, changes in TB (T1DM = 0.035 ± 0.022 g·cm−2, healthy = 0.031 ± 0.017 g·cm−2) and LS2–LS4 (T1DM = 0.046 ± 0.038 g·cm−2, healthy = 0.063 ± 0.034 g·cm−2) aBMD were statistically significant in the intervention groups and of similar magnitude between T1DM and healthy subjects. The level of type 1 collagen cross-linking (T1DM = −0.12 ± 0.32 ng·mL−1, healthy = −0.36 ± 0.11 ng·mL−1) decreased in the intervention groups but was not associated with TB aBMD changes.

Conclusions Regular weight-bearing physical activity (180 min·wk−1, including ball games, jumping activities, and gymnastics) improves total and LS2–LS4 bone mineral accretion in children with T1DM, in a similar magnitude to healthy subjects. We conclude that children with T1DM should be encouraged to practice regular physical activity to enhance peak bone mass and prevent osteoporosis later in life.

1Pediatric Sports Medicine and Obesity Care Program, Service of Pediatric Specialties, Department of Child and Adolescent, University Hospital of Geneva, Geneva, SWITZERLAND; 2Service of Bone Diseases, Department of Rehabilitation and Geriatrics, University Hospital of Geneva, Geneva, SWITZERLAND; 3Pediatric Cardiology Unit, Service of Pediatric Specialties, Department of Child and Adolescent, University Hospital of Geneva, Geneva, SWITZERLAND; and 4University of Geneva, Geneva, SWITZERLAND

Address for correspondence: Nathalie Jacqueline Farpour-Lambert, M.D., Pediatric Sports Medicine and Obesity Care Program, Service of PediatricSpecialties, Department of Child and Adolescent, University Hospital ofGeneva, 6 rue Willy-Donzé, 1211 Geneva 14, Switzerland; E-mail:

Submitted for publication April 2011.

Accepted for publication October 2011.

Osteoporosis is a growing health problem with high morbidity (27). In children, low bone mineral density (BMD) has been shown to be associated with a higher fracture rate (12). Bone mass is acquired mainly during childhood until young adulthood (30) and influenced by physical activity (17). Weight-bearing and high-impact physical activities have been shown to increase bone mineral mass (10,22,32) and may prevent osteoporosis (4,28). Prepubertal children have a better response to exercise than older children (32), and one cross-sectional study showed that, in this age group, a higher local muscle mass is associated with a higher BMD (7). However, the dose and type of exercise to optimize bone mineral acquisition are still not well defined (6).

Children and adolescents with type 1 diabetes mellitus (T1DM) may have diminished BMD because of insulin deficiency and consequent low insulin-like growth factor 1 production (19,20,29). These findings remain controversial in this population probably because of different methods of investigation, sites of measurements, and disease duration (8,14). We recently demonstrated that BMD was normal in T1DM subjects, whereas bone formation markers were reduced (23), as shown also in other studies (14,26). Biochemical markers of bone turnover are an easy tool to assess bone formation (osteocalcin, procollagen type 1 amino terminal propeptides) and resorption (type 1 collagen C-terminal cross-linking telopeptides) and may be the first hint of the disease repercussion on bone modeling and remodeling (24).

To our knowledge, no study has evaluated the influence of exercise training on bone accretion and biochemical markers in T1DM children.

The aim of this study was to determine the effects of a weight-bearing physical activity program on bone mineral mass and bone turnover markers in T1DM children, compared with healthy children.

Back to Top | Article Outline


Study design and subjects.

This was a randomized controlled trial including 27 children with T1DM (mean age = 10.5 ± 2.4 yr, mean disease duration = 3.2 ± 2.3 yr [range = 0.4 to 7.5 yr]) and 32 healthy age-matched children (mean age = 10.5 ± 2.5 yr). Baseline data were recently published (23). Diabetic children were recruited from the Unit of Pediatric Endocrinology and Diabetology of the University Hospitals of Geneva and were excluded for any of the following reasons: 1) presence of other chronic diseases, including thyroid or gastrointestinal diseases; 2) medications, hormones other than insulin, or calcium preparations, which may influence bone accretion, taken in the preceding 6 months; 3) presence of nephropathy; 4) systemic disease or hospitalization for more than 2 wk in the preceding year; and 5) participation in competition sport, which may be more prevalent in the healthy group and further increase BMD.

Healthy children were recruited from peers of the diabetic subjects or from local schools and did not participate in any competition sport. Healthy subjects were excluded according to the preceding criteria.

Patients with T1DM and healthy subjects were randomized by sex (1:1) to two groups: 1) exercise (Ex) (diabetic [ExD, n = 15] and healthy [ExH, n = 15]) and 2) control (Con) (diabetic [ConD, n = 12] and healthy [ConH, n = 17]). We used closed envelopes containing 50% of the Ex and 50% of Con group in each disease group. The randomization process resulted in a similar girls–boys ratio in each group.

Back to Top | Article Outline


All children, healthy and diabetic, randomized to the Ex group attended together two 90-min exercise sessions per week during 9 months (excluding holidays). Sessions were supervised by two physical education teachers and a pediatrician who had experience in T1DM management. The training program included warm-up for 10 min, followed by 10 min of drop jump, with the height of the platform increasing from 20 cm (first 3 months) to 40 cm (last 6 months). Then, one or two different weight-bearing activities (rope skipping, jumping, ball games, and gymnastics) were performed during 60 min, and cool-down exercises were performed during the last 10 min. During the first 8 wk, HR was monitored to make sure that the intensity was at least 140 min−1 during the jumping activities and games. Controls were relatively inactive during the intervention and were not involved in any competition sports. The study flow diagram is presented in Figure 1.



All children were tested during the same period at baseline (August to September) and at 9 months (May to June), reducing seasonal variation among groups for leisure time physical activity and vitamin D concentrations.

The Mother and Child Ethics Committee of the University Hospitals of Geneva approved this study, and informed written consent was obtained from both the parent and the child.

Back to Top | Article Outline


Healthy and T1DM subjects visited the Children’s Hospital between 8 and 12 p.m. and underwent identical testing at baseline and at 9 months. Observers were blinded to subject grouping. Testing procedures have been described in a previous publication (23) and are outlined below.

Back to Top | Article Outline

Anthropometric measurements.

We measured body weight in light clothes to the nearest 0.1 kg using an electronic scale (Seca 701, Hamburg, Germany) and height to the nearest 0.1 cm using a Harpenden stadiometer (Harpenden, Birmingham, UK). We calculated body mass index as weight / height squared (kg·m−2). Pubertal development (Tanner stages) was determined by clinical examination. We measured lean body mass (LBM [kg]) using dual-energy x-ray absorptiometry (GE Lunar Prodigy; Lunar Corp., Madison, WI). The intraclass correlation for repeated measurements of LBM was 0.998 in our laboratory.

Back to Top | Article Outline

BMD measurements.

Areal BMD (aBMD [g·cm−2]) was assessed at the total body (TB), lumbar spine (LS2–LS4), right femoral neck (FN), and greater trochanter (GT) using dual-energy x-ray absorptiometry (GE Lunar Prodigy; Lunar Corp.). The coefficient of variation for repeated measurements was <1% for TB aBMD, 1% for LS2–LS4 aBMD, 2% for GT aBMD, and 2% for FN aBMD.

Back to Top | Article Outline

Laboratory measurements.

Morning blood samples were collected during the same period for all children via venipuncture, after a 10-h overnight fast. Biochemical markers of bone turnover were measured on serum samples that have been kept frozen (−80°C). Type 1 collagen C-terminal cross-linking telopeptides (CTX, osteocalcin (OC), procollagen type-1 amino-terminal propetides (PINP), and 25-OH vitamin D (25-OH-D) were assessed in serum with electrochemiluminescence immunoassays on the automated an alyzer Elecsys 2010 (Roche Diagnostics, Rotkreuz, Switzerland). The intra- and interassay variation was 2.4%–7.2% for CTX, 1.1%–5.9% for OC, 1.7%–4.0% for PINP, and 4.2%–6.1% for 25-OH-D, respectively. In addition, glycosylated hemo globin (HbA1c [%]) was determined using a quantitative automotive technique (Synchron LX20). The intra- and interassay coefficients of variation were 2.8% and 2.7%, respectively. Calibration was performed every 30 d using the HbA1c Synchron.

Back to Top | Article Outline

Physical activity.

We determined the proportion of attended sessions for each subject and the proportion of subjects who completed all training and testing sessions. Leisure time physical activity level for the previous 12 months was assessed by a recall questionnaire (the Modifiable Activity Questionnaire for Adolescents) (1), and the mean number of hours per week was calculated.

Back to Top | Article Outline

Sample size and statistical analysis.

Calculation of sample size was based on an observed exercise-associated effect size for TB and site-specific measures of aBMD (g·cm−2) ranging from 0.01 to 0.06 in cross-sectional and longitudinal studies involving healthy children and adolescents (5,13). We did expect and accept a moderate effect size of 0.02 between the Ex and Con groups for posttraining aBMD outcomes. With this anticipated effect size, using a one-way ANOVA, a sample size of 10 subjects in each of the Ex and Con groups could detect statistically significant differences at P < 0.05 with a statistical power of 80% (β = 0.20).

Statistical analyses were performed using SPSS 15.0 (Chicago, IL). Data were initially screened for normality, using skewness and kurtosis tests. Data are presented as mean and SD. We present the results as an intention-to-treat analysis. Baseline statistical differences between the diabetic and healthy groups were analyzed using the independent Student’s t-test and chi-square test, and the difference between the four groups was analyzed using an ANOVA. A two-way ANOVA was performed to explore the effect of the disease and the intervention on the change scores. Effect size was expressed with partial η2 to evaluate the relative magnitude of the difference between means. We performed linear regression analysis to determine the association between biochemical markers and TB BMD changes and Pearson correlation to look at the relationship between LBM and TB BMD. However, because we performed multiple separate statistical analyses, we must acknowledge that the chance of a type I error is at approximately 90%. Differences were considered significant if P < 0.05.

Back to Top | Article Outline


Baseline results.

Baseline characteristics of subjects are presented in Table 1. There were no significant differences among T1DM and healthy children for anthropometric, body composition, or physical activity values. The mean HbA1c level was higher in T1DM than in healthy subjects. Baseline aBMD and metabolism results are shown in Table 2. The aBMD values were not different between groups, whereas OC, PINP, and CTX levels were significantly lower in T1DM children compared with healthy subjects. Finally, baseline characteristics and bone parameters were not different among the Ex and Con groups in T1DM or healthy subjects.





Back to Top | Article Outline

Participation rate.

Eighty-three percent of the 64 scheduled exercise sessions were attended by the Ex groups, and 24 of the subjects (80%) participated in two sessions per week. The participation rate was similar for the diabetic and healthy children. There was no dropout in any of the groups, and no diabetic patients had adverse events, such as hypoglycemia, during the training sessions.

Back to Top | Article Outline

Changes during the exercise intervention.

Pubertal stages at the end of the 9-month intervention were not different among groups (P = 0.794). Pubertal stage changed in two in the ConD group (stages I/II/III/IV–V = 6/2/1/3), four in the ExD group (4/2/6/3), four in the ConH group (5/4/4/4), and eight in the ExH group (4/4/3/4).

The effects of the disease and exercise training on physical characteristics and bone parameters are shown in Table 3. As expected, the intervention had a moderate to large effect on LBM but not on other anthropometry variables.



Back to Top | Article Outline


TB and LS2–LS4 aBMD changes were higher in the intervention groups. These changes were not different between healthy and T1DM subjects because the P value for the interaction of the disease and the intervention was not significant. We did not find any difference for changes among gender (P > 0.05). Interestingly, there was a positive correlation between LBM and TB BMD accrual in the ExD group only (r = 0.681, P = 0.007).

The influence of puberty was then examined by adjusting the two-way ANOVA analysis for this variable. We found that puberty development had a significant influence on TB, FN, and LS2–LS4 aBMD changes (TB: P value for puberty = 0.002, partial η2 = 0.180; FN: P = 0.001, partial η2 = 0.189; and LS2–LS4: P < 0.001, partial η2 = 0.322). However, puberty did not influence the intervention effects.

Back to Top | Article Outline

Bone biochemical markers.

The effects of the disease and exercise training on biochemical markers of bone turnover are presented in Table 3. During the intervention period, within-group CTX levels decreased equally in ConH and ConD, as well as in the ExH group; however, treatment effects were not significant. We observed a lower 25-OH-D value in the exercising group. Bone biomarkers were not associated with TB BMD changes (P > 0.05 for all).

Back to Top | Article Outline


To our knowledge, this is the first randomized controlled trial that determined the effects of weight-bearing physical activity on BMD and metabolism in T1DM children. We observed that T1DM children who performed a 9-month exercise training program (180 min·wk−1) improved their TB and LS2–LS4 BMD compared with diabetic controls, and the magnitude of changes induced by exercise was similar between T1DM and healthy subjects. This improvement was not associated with variation in bone biochemical markers.

Physical activity is known to be one of the main determinants of bone development in healthy children (4,10,21); however, its effect has not been well studied during growth in T1DM patients. Children with this condition may have reduced bone mineral acquisition compared with healthy subjects, due to insulin deficiency and consequent low insulin-like growth factor 1 production (19,29).

We found that regular weight-bearing physical activity improves TB and LS BMD accrual in children with T1DM, as much as in healthy children. Our results are in concordance with several studies performed in healthy children populations. Some authors showed that jumping exercises specifically improve LS2–LS4 and FN BMD in prepubertal children (10) or BMD at all sites for pre- and peripubertal subjects (15). Physical activity has a different influence on various bone sites among gender and pubertal stages and is dependent on many factors such as the type of exercise, training duration, or volume (21,22). Because, in our study, the sample size is very small, it is difficult to determine whether BMD changes are due to puberty, gender, or the intensity or volume of training. These last two factors may be different among children even if they attended the same training program.

The latest pediatric position statement indicates that the hip is not a reliable site for measurements because of significant variability in skeletal development and lack of reproducible regions of interest (3). TB and spine aBMD are considered as the most accurate and reproducible sites, and LS2–LS4 is mainly made of trabecular bone and highly sensitive to metabolic changes. Interestingly, we found significant TB aBMD changes, as well as LS2–LS4 aBMD, with exercise in both diabetic and healthy subjects.

Few studies have evaluated the influence of exercise training on bone turnover biomarkers in children. In adults, two studies have shown an increase in bone formation markers after 1 month of training in premenopausal women, as well as in formation and resorption markers after 2 months in young military recruits (2,9). The duration and the repetition of the activity seem important because bone resorption has been shown to decrease immediately after a 45-min moderate-intensity resistance exercise session, with a normalization of CTX level by 24 h after cessation of training (31). When performed regularly, one study in adult soccer players has shown that 6 h or less per week of physical activity was sufficient to sustain an increase in bone turnover markers (18). On the contrary, one study showed no effect of 12-wk high-intensity training on bone turnover markers in young women (18–29 yr) (25). These conflicting results may be explained by diverse populations studied, type of exercise, or timing of blood examinations. Even if little is known in humans, a study on rats demonstrated that uninterrupted and interrupted resistance training programs were equally effective in stimulating bone remodeling (11). We have no explanation for the lower 25-OH-D values in the exercising group at the end of the intervention.

In children, it is more challenging to demonstrate such changes because bone markers vary with age and puberty (23). The effects of exercise on bone biomarkers in diabetic children have never been studied. We found a significant reduction of CTX concentrations after the intervention, with no difference between the two diseased and healthy groups. However, the clinical importance of the changes in biomarkers is difficult to assess because we did not find any association between these markers and TB aBMD changes. However, larger sample size and multiple blood examinations may be needed to demonstrate such associations.

Our study has some strengths and limitations. The quality of the study design, comparison of treatment effects between T1DM and healthy children, and the absence of dropout add strengths to our study. However, the small sample size does not allow a subanalysis to evaluate the influence of age and puberty on the different variables. Furthermore, because of wide biological variations during growth, we may hypothesize that a much greater number of subjects would be needed to detect changes in bone biomarkers.

To our knowledge, this is the first study that shows positive effects of weight-bearing physical activity on TB and LS2–LS4 bone mineral acquisition in children with T1DM, similarly to healthy children. We conclude that weight-bearing sports, including ball games, jumping activities, or gymnastics, should be encouraged in this population to optimize bone mineral acquisition during growth and potentially prevent the development of osteoporosis later in life.

This study was supported financially by the Swiss National Science Foundation, the Sir Jules Charitable Overseas Trust, and the Mimosa grant of the University of Geneva.

This study is registration trial NCT01220479 in

The authors declare no conflicts of interest.

The authors thank the subjects for volunteering for the study, Marius Kraenzlin, Didier Hans, Giulio Conicella, and the staff of the Pediatric Policlinic for their assistance.

The results of this present study do not represent endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Aaron DJ, Kriska AM, Dearwater SR, Cauley JA, Metz KF, LaPorte RE. Reproducibility and validity of an epidemiologic questionnaire to assess past year physical activity in adolescents. Am J Epidemiol. 1995; 142: 191–201.
2. Adami S, Gatti D, Viapiana O, et al.. Physical activity and bone turnover markers: a cross-sectional and a longitudinal study. Calcif Tissue Int. 2008; 83: 388–92.
3. Baim S, Leonard MB, Bianchi ML, et al.. Official Positions of the International Society for Clinical Densitometry and executive summary of the 2007 ISCD Pediatric Position Development Conference. J Clin Densitom. 2008; 11: 6–21.
4. Bradney M, Pearce G, Naughton G, et al.. Moderate exercise during growth in prepubertal boys: changes in bone mass, size, volumetric density, and bone strength: a controlled prospective study. J Bone Miner Res. 1998; 13: 1814–21.
5. Cassell C, Benedict M, Specker B. Bone mineral density in elite 7- to 9-yr-old female gymnasts and swimmers. Med Sci Sports Exerc. 1996; 28 (10): 1243–6.
6. Daly RM. The effect of exercise on bone mass and structural geometry during growth. Med Sport Sci. 2007; 51: 33–49.
7. Daly RM, Saxon L, Turner CH, Robling AG, Bass SL. The relationship between muscle size and bone geometry during growth and in response to exercise. Bone. 2004; 34: 281–7.
8. De Schepper J, Smitz J, Rosseneu S, Bollen P, Louis O. Lumbar spine bone mineral density in diabetic children with recent onset. Horm Res. 1998; 50: 193–6.
9. Evans RK, Antczak AJ, Lester M, Yanovich R, Israeli E, Moran DS. Effects of a 4-month recruit training program on markers of bone metabolism. Med Sci Sports Exerc. 2008; 40 (11 suppl): S660–70.
10. Fuchs RK, Bauer JJ, Snow CM. Jumping improves hip and lumbar spine bone mass in prepubescent children: a randomized controlled trial. J Bone Miner Res. 2001; 16: 148–56.
11. Goettsch BM, Smith MZ, O’Brien JA, Gomez GV, Jaque SV, Sumida KD. Interrupted vs. uninterrupted training on BMD during growth. Int J Sports Med. 2008; 29: 980–6.
12. Goulding A, Jones IE, Taylor RW, Williams SM, Manning PJ. Bone mineral density and body composition in boys with distal forearm fractures: a dual-energy x-ray absorptiometry study. J Pediatr. 2001; 139: 509–15.
13. Grimston SK, Willows ND, Hanley DA. Mechanical loading regime and its relationship to bone mineral density in children. Med Sci Sports Exerc. 1993; 25 (11): 1203–10.
14. Gunczler P, Lanes R, Paz-Martinez V, et al.. Decreased lumbar spine bone mass and low bone turnover in children and adolescents with insulin dependent diabetes mellitus followed longitudinally. J Pediatr Endocrinol Metab. 1998; 11: 413–9.
15. Gunter K, Baxter-Jones AD, Mirwald RL, et al.. Jump starting skeletal health: a 4-year longitudinal study assessing the effects of jumping on skeletal development in pre and circum pubertal children. Bone. 2008; 42: 710–8.
16. Deleted in proof.
    17. Hind K, Burrows M. Weight-bearing exercise and bone mineral accrual in children and adolescents: a review of controlled trials. Bone. 2007; 40: 14–27.
    18. Karlsson KM, Karlsson C, Ahlborg HG, Valdimarsson O, Ljunghall S. The duration of exercise as a regulator of bone turnover. Calcif Tissue Int. 2003; 73: 350–5.
    19. Kemink SA, Hermus AR, Swinkels LM, Lutterman JA, Smals AG. Osteopenia in insulin-dependent diabetes mellitus; prevalence and aspects of pathophysiology. J Endocrinol Invest. 2000; 23: 295–303.
    20. Lettgen B, Hauffa B, Mohlmann C, Jeken C, Reiners C. Bone mineral density in children and adolescents with juvenile diabetes: selective measurement of bone mineral density of trabecular and cortical bone using peripheral quantitative computed tomography. Horm Res. 1995; 43: 173–5.
    21. Macdonald HM, Kontulainen SA, Petit MA, Beck TJ, Khan KM, McKay HA. Does a novel school-based physical activity model benefit femoral neck bone strength in pre- and early pubertal children? Osteoporos Int. 2008; 19: 1445–56.
    22. MacKelvie KJ, Petit MA, Khan KM, Beck TJ, McKay HA. Bone mass and structure are enhanced following a 2-year randomized controlled trial of exercise in prepubertal boys. Bone. 2004; 34: 755–64.
    23. Maggio AB, Ferrari S, Kraenzlin M, et al.. Decreased bone turnover in children and adolescents with well controlled type 1 diabetes. J Pediatr Endocrinol Metab. 2010; 23: 697–707.
    24. Miller PD, Baran DT, Bilezikian JP, et al.. Practical clinical application of biochemical markers of bone turnover: consensus of an expert panel. J Clin Densitom. 1999; 2: 323–42.
    25. Mullins NM, Sinning WE. Effects of resistance training and protein supplementation on bone turnover in young adult women. Nutr Metab (Lond). 2005; 2: 19.
    26. Olmos JM, Perez-Castrillon JL, Garcia MT, Garrido JC, Amado JA, Gonzalez-Macias J. Bone densitometry and biochemical bone remodeling markers in type 1 diabetes mellitus. Bone Miner. 1994; 26: 1–8.
    27. Riggs BL, Melton LJ 3rd. The worldwide problem of osteoporosis: insights afforded by epidemiology. Bone. 1995; 17: 505S–11S.
    28. Saggese G, Baroncelli GI, Bertelloni S. Osteoporosis in children and adolescents: diagnosis, risk factors, and prevention. J Pediatr Endocrinol Metab. 2001; 14: 833–59.
    29. Thrailkill KM. Insulin-like growth factor-I in diabetes mellitus: its physiology, metabolic effects, and potential clinical utility. Diabetes Technol Ther. 2000; 2: 69–80.
    30. van der Sluis IM, de Ridder MA, Boot AM, Krenning EP, de Muinck Keizer-Schrama SM. Reference data for bone density and body composition measured with dual energy x ray absorptiometry in white children and young adults. Arch Dis Child. 2002; 87: 341–7; discussion 341–7.
    31. Whipple TJ, Le BH, Demers LM, et al.. Acute effects of moderate intensity resistance exercise on bone cell activity. Int J Sports Med. 2004; 25: 496–501.
    32. Zanker CL, Gannon L, Cooke CB, Gee KL, Oldroyd B, Truscott JG. Differences 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: 1043–50.


    ©2012The American College of Sports Medicine