Sustained Benefits from Previous Physical Activity on Bone Mineral Density in Males

0021-972X/06/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 91(7):2600 –2604 Copyright © 2006 by The Endocrine Society doi: 10.1210/jc.2006-0151

Sustained Benefits from Previous Physical Activity on Bone Mineral Density in Males Anna Nordstro¨m, Tommy Olsson, and Peter Nordstro¨m Department of Community Medicine and Rehabilitation (A.N.), Rehabilitation Medicine; Department of Surgical and Perioperative Sciences (A.N., P.N.), Sports Medicine; and Department of Public Health and Clinical Medicine (T.O., P.N.), Medicine, Umeå University, 90185 Umeå, Sweden Context: The effect of physical activity on bone mineral density (BMD) is not well investigated longitudinally after puberty in men. Objective: Our objective was to evaluate the effect of exercise and reduced exercise on BMD after puberty in men. Design: We conducted a longitudinal study. Participants: Sixty-three healthy young athletes and 27 male controls, both with a mean age of 17 yr at baseline, participated. Also, 136 of the participants’ parents were investigated to evaluate heritable influences. Main Outcome Measures: Total body, total hip, femoral neck, and humerus BMD (grams per square centimeter) were measured at baseline and after mean periods of 27, 68, and 94 months in the young cohort.

W

EAK AND OSTEOPOROTIC bones in the elderly are an increasing cause of mortality and painful physical impairment, especially in the Western world (1). The incidence of osteoporosis is higher in women, but with incidence in men expected to triple over the next 50 yr (2), preventive measures for both sexes are crucial. Physical activity has many positive effects on the musculoskeletal system, including increased bone mineral density (BMD). The activity should be performed regularly, be weight bearing, and involve tension from various angles to evoke a high osteogenic response (3–5). It is not clear at what age the bone is most sensitive to weight-bearing loading, but childhood to early puberty may be the period most sensitive to physical activity (6). Less is known of the effect of physical activity during the postpubertal period, especially in men. This period is of interest because humans achieve peak BMD after puberty (7). In a recent study, we found significant and sustained effects of ice hockey training on BMD in postpubertal males during 6 yr of follow-up (8). A key question is whether a high BMD resulting from physical activity is sustained despite decreased activity later in life. In studies involving shorter follow-up periods and elderly subjects with low exercise-induced BMD benefits, First Published Online April 24, 2006 Abbreviations: BMD, Bone mineral density; CV, coefficient of variation. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

Results: BMDs of control parents and athlete parents were equal, suggesting absence of selection bias. The 23 athletes that remained active throughout the study increased BMD at all sites when compared with controls (mean difference, 0.04 – 0.12 g/cm2; P ⬍ 0.05) during the study period. After an average of 3 yr, 27 athletes ended their active careers. Although this group initially lost BMD at the hip compared with active athletes, the former athletes still had higher BMD than controls at the femoral neck (0.12 g/cm2; P ⫽ 0.007), total hip (0.11 g/cm2; P ⫽ 0.02), and humerus (0.10 g/cm2; P ⫽ 0.02) at the final follow-up. Conclusions: High sensitivity to physical loading persists after puberty in men. Reduced physical activity is associated with BMD loss in the first 3 yr in weight-bearing bone. Sustained benefits in BMD are preserved 5 yr after intensive training ends. (J Clin Endocrinol Metab 91: 2600 –2604, 2006)

BMD declined to pretraining levels (9 –12). However, other studies have reported that increased bone mass resulting from training persists despite reduced activity (13, 14). There have been no long-term prospective studies in which athletes with high BMD resulting from previous training have been monitored with repeated measurements after intensive training ended. The aim of this 8-yr prospective study was to monitor BMD over time in postpubertal males and investigate the effect of physical activity and reduced activity levels on BMD. The BMD of parents of this young cohort were also measured to investigate the influence of heritable factors on BMD and evaluate the risk of selection bias. Subjects and Methods Subjects We recruited 117 healthy adolescent Caucasian males from high schools and badminton and ice hockey clubs in Umeå in northern Sweden for this longitudinal study, also called the Northern Osteoporosis and Obesity Study. The experimental group (athletes) included actively training ice hockey players and badminton players. After a mean period of 27 months, 112 subjects participated in a first follow-up examination. One subject in the control group was diagnosed with anorexia nervosa and depression and was therefore excluded. Twelve ice hockey players had stopped training before the first follow-up and were therefore excluded from the study. A second follow-up was conducted after a mean period of 68 months; 94 of the original participants (66 athletes and 28 controls) could be located and participated in this follow-up. A third follow-up was conducted after a mean period of 94 months; 90 of the original subjects (63 athletes and 27 controls) could be located for this follow-up and were included in the present study. During the first study period, the athletes were enrolled at local ice

2600

Nordstro¨m et al. • Physical Activity and Bone

J Clin Endocrinol Metab, July 2006, 91(7):2600 –2604

hockey or badminton clubs and trained regularly with their team. At baseline, the average amount of weight-bearing physical activity of the athletes amounted to 9.4 ⫾ 2.6 h/wk. This activity consisted mainly of training or matches and some additional weight and aerobic training. Data were recorded about starting age and years of active athletic training. At baseline, the control group’s total average amount of weightbearing physical activity was estimated to be 2.1 ⫾ 2.7 h/wk during their spare time. This group’s weight-bearing physical activity consisted of playing soccer and floor ball, distance running, and some weight training. None of these subjects participated in any regular organized training. All participants had reached at least Tanner stage 4 at the beginning of our study. Weight and height were measured using standardized equipment. A standardized questionnaire was used at baseline and at follow-ups to record smoking habits, known illnesses, medication intake, and type and amount of reported physical activity. None of the subjects suffered from any disease or took any medication known to affect bone metabolism. At the first visit, the groups were not significantly different in terms of age, pubertal stage, height, or weight. All data were collected at the Sports Medicine Unit at Umeå University. To evaluate the influence of heritable factors on BMD, the BMDs of parents of the cohort were measured. Of a possible 186 parents, a total number of 136 were included; 35 were parents of the control group (73% follow-up rate), and 101 were parents of the athlete group (77% follow-up rate). Of the 50 who were not included, four were excluded because of diseases known to affect bone metabolism, one was excluded because of extreme obesity, and the rest could not be reached or did not want to participate. The parents completed a questionnaire, including information about daily physical activities, diseases, current medication, and lifestyle characteristics. The athletes’ parents were predominantly measured in 1996, and the controls’ parents were predominantly measured in 2005, resulting in a significantly different age at time of measurement. Informed consent was given by all the participants, and the study protocol was approved by the Ethical Committee of the Medical Faculty, Umeå University, Umeå, Sweden.

BMD The BMD (g/cm2) of the total body was measured using a Lunar DPX-L (Lunar Co., Waukesha, WI) dual-energy x-ray absorptiometer, software version 4.6e. The BMD and bone area of the right femoral neck and the total hip were measured using the same equipment and the corresponding program. BMD of the lumbar spine (L2–L4) was measured in the parents using the specific spine program. The coefficient of variation (CV, i.e. sd/mean) was determined by scanning one person seven times the same day with repositioning between each scan. Accordingly, the CV values were 0.7% for the total body scan, approximately 1% for the femoral neck/total hip scan, 0.6% for the lumbar spine scan, 0.7% for total hip area, and 0.9% for the neck area. BMD of dominant humerus was derived from the total body scan. All analyses were made by the same investigator (A.N). To maximize precision, the po-

2601

sition of the underarm of all subjects was standardized, and the scaling option was used and set to 200. To evaluate the precision of the regionof-interest program, two different persons were scanned. The first person was scanned seven times the same day, with repositioning between each scan. The CV value was then calculated to be 1.6%. The second person was scanned 10 times during different days. Accordingly, the CV value increased to 2.4%. The equipment was calibrated each day using a standardized phantom to detect drifts in bone density measurements. All scans were made on the same Lunar DPX-L.

Statistical analysis Differences in age, anthropometric data, physical activity, and changes in BMD among three groups or more were investigated using ANOVA with Bonferroni’s post hoc test for multiple comparisons. Differences between two groups were investigated using Student’s t test for independent samples. Differences in BMD between several groups were investigated using analysis of covariance with body weight as a covariate. The SPSS package, version 9.0 for PC, was used for the statistical analyses. A P value of ⬍0.05 was considered statistically significant. Data are presented as means ⫾ sd in the text; in the figures, they are presented as means ⫾ sem.

Results

Table 1 shows the age, anthropometric measures, and hours of physical activity of the controls and athletes at baseline. BMDs at the various follow-ups are presented in Fig. 1. Differences in body weight, height, and physical activity during the study are presented in Table 2. At baseline, the athletes and controls differed significantly in the amount of physical activity (P ⬍ 0.001), without differences in anthropometric measures (Table 1). The athletes had significantly higher femoral neck, total hip, and humerus BMDs than the controls (P ⬍ 0.05). To evaluate genetic influences on BMD and evaluate the risk of selection bias at baseline also, BMDs of the athletes’ and controls’ parents were measured. There were no significant differences between the two groups at any site before or after converting BMD to z-scores adjusting for age, weight, and physical activity (data not shown). There were also no significant differences in terms of anthropometric measures, hours of physical activity, or BMD between those athletes (n ⫽ 23) who continued training throughout the study (active athletes) and those athletes (n ⫽ 40) who ceased active training at some point during the study period (Table 1). The total group of athletes (n ⫽ 63) gained significantly

TABLE 1. Baseline data for age, anthropometric data, physical activity, bone density at different sites, and bone area of the femoral neck and hip in controls and all athletes and in the group of athletes who remained active throughout the study (active) and athletes who ended their active career during the study (former) Controls (n ⫽ 27)

Age (yr) Weight (kg) Height (cm) Passed puberty (%) Physical activity (h/wk) Starting age of training (yr) BMD (g/cm2) Total body Femoral neck Total hip Humerus Neck area (cm2) Hip area (cm2)

All athletes (n ⫽ 63)

17.1 ⫾ 1.1 72.5 ⫾ 13.0 180 ⫾ 5 81 2.6 ⫾ 2.9

17.2 ⫾ 2.0 70.9 ⫾ 8.7 178 ⫾ 5 75 9.1 ⫾ 2.8 7.6 ⫾ 1.9

1.20 ⫾ 0.11 1.17 ⫾ 0.15 1.22 ⫾ 0.15 1.21 ⫾ 0.15 5.56 ⫾ 0.50 36.6 ⫾ 2.46

1.23 ⫾ 0.07 1.26 ⫾ 0.14 1.33 ⫾ 0.14 1.28 ⫾ 0.11 5.61 ⫾ 0.44 36.9 ⫾ 2.53

Significance (P value)

0.81 0.54 0.13 0.59 ⬍0.001 0.20 0.01 0.001 0.04 0.38 0.22

Athletes Active (n ⫽ 23)

Former (n ⫽ 40)

Significance (P value)

17.0 ⫾ 1.7 69.2 ⫾ 7.7 178 ⫾ 5 71 8.8 ⫾ 2.7

17.2 ⫾ 2.2 71.8 ⫾ 9.3 178 ⫾ 6 77 9.3 ⫾ 2.8

0.73 0.25 0.80 0.60 0.48

1.21 ⫾ 0.07 1.25 ⫾ 0.12 1.33 ⫾ 0.12 1.26 ⫾ 0.10 5.69 ⫾ 0.49 36.1 ⫾ 2.62

1.24 ⫾ 0.07 1.26 ⫾ 0.15 1.33 ⫾ 0.15 1.29 ⫾ 0.12 5.56 ⫾ 0.41 37.3 ⫾ 2.39

0.21 0.82 0.94 0.31 0.28 0.06

2602

J Clin Endocrinol Metab, July 2006, 91(7):2600 –2604

Nordstro¨m et al. • Physical Activity and Bone

FIG. 1. Changes in BMD (grams per square centimeter) in the femoral neck (A), total hip (B), total body (C), and humerus (D), during the study period. Four subgroups were studied: athletes that were active throughout the study (active athletes), athletes that stopped training between the first and the second follow-up (former athletes 1), athletes that stopped training between the second and third follow-up (former athletes 2), and the control group.

more femoral neck BMD than did the controls between baseline and the first follow-up (0.06 vs. 0.02 g/cm2; P ⫽ 0.009). At the first follow-up, what was to be the three different groups of athletes had significantly higher total body, femoral neck, and total hip BMD vs. controls (Fig. 1) (P ⬍ 0.05). The active athletes also had higher BMD of the humerus. Between the first and second follow-ups, 27 athletes (former athletes 1) ceased active training at a mean age of 22.2 ⫾ 2.9 yr. This group lost significantly more BMD at the femoral neck and total hip than the active athletes between the first and second follow-ups (P ⬍ 0.01). The active athletes also lost significantly less BMD at the femoral neck compared with the controls (P ⫽ 0.02, data not shown). At the second follow-up, the active athletes were found to have significantly higher BMD at all measured sites compared with the controls (P ⬍ 0.05) (Fig. 1). Between the second and third follow-ups, 13 athletes (former athletes 2) ceased active training at a mean age of 22.7 ⫾ 1.5 yr. This group lost significantly more BMD at the femoral neck than all other groups during this time period (P ⬍ 0.05, data not shown). The active athletes and the controls lost significantly less BMD of the total hip than the former athletes 2 (P ⬍ 0.05). At the third follow-up, the active athletes had significantly higher BMD at all measured sites vs. the controls (P ⬍ 0.05) (Fig. 1). The former athletes 1 still had higher BMD of the

femoral neck, total hip, and humerus than did the controls (P ⬍ 0.05) (Fig. 1). During the whole study period, the active athletes gained more BMD of the total body (0.14 vs. 0.09 g/cm2; P ⫽ 0.03), femoral neck (0.05 vs.⫺0.07 g/cm2; P ⫽ 0.001), total hip (0.05 vs.⫺0.05 g/cm2; P ⫽ 0.002), and humerus (0.13 vs. 0.05 g/cm2; P ⫽ 0.02) vs. the controls (n ⫽ 27). Discussion

The unique design of the present study included measurements also of the parents’ BMD, with no significant differences between parents of controls and parents of athletes. This would suggest absence of selection bias, because we also found similar pubertal stage at baseline and similar development of height when comparing athletes and controls during the study period. We report sustained benefits from training on BMD during the postpubertal period. These benefits amounted to at most 0.12 g/cm2 at the femoral neck. We also report that although some of these benefits were lost at the proximal femur with reduced training in former athletes, sustained benefits were seen at several sites 5 yr after the end of an active career. The inference would be that the higher BMD found in former athletes would be a result of previous training and not genetic factors. Determining the age at which the bone is most sensitive to

Nordstro¨m et al. • Physical Activity and Bone

J Clin Endocrinol Metab, July 2006, 91(7):2600 –2604

2603

FIG. 1. Continued

physical loading is an important issue. Prospective studies have shown that high-impact loading has clear effects on various aspects of bone mass in children (15, 16). MacKelvie et al. (16) evaluated the effect of a physical education program on 75 girls, aged 9 –12 yr, over the course of 20 months. BMC of the femoral neck and spine in the intervention group increased more than in the control group. Based on this and some other intervention studies (16 –18), it has been suggested that bone mass may be especially sensitive to exercise during early puberty, at least in girls (6). In the present study, the athletes had already trained for a mean of 10 yr at the beginning of the study, and at baseline this cohort had BMDs 0.03– 0.11 g/cm2 higher than the con-

trols. During the study period, the differences in BMD gains were 0.04 – 0.12 g/cm2 between the active athletes and controls. Thus, it seems that in addition to the pre- and peripubertal periods, the postpubertal period is important in maximizing peak BMD. Interestingly, at the proximal femur, the activity seemed to preserve BMD after the age of 19, whereas at other sites, such as the total body, BMD increased during the whole study period. This may reflect the fact that peak BMD occurs at different ages at different sites (7, 19). A key issue in preventing osteoporotic fractures is determining whether a high BMD from previous physical activity is sustained after intensive training ends. Inconclusive results from previous studies may be related to different and

TABLE 2. Changes in weight, height, and physical activity in the controls and subgroups of athletes who remained active throughout the study (active, n ⫽ 23), athletes who ended their active career between the first and second follow-up (former 1, n ⫽ 27), and athletes who ended their active career between the second and third follow-up (former 2, n ⫽ 13) First follow-up

Controls Active cohort Former 1 cohort Former 2 cohort a

Second follow-up

Height (cm)

Activity (h/wk)

Weight (kg)

Height (cm)

Activity (h/wk)

Weight (kg)

Height (cm)

Activity (h/wk)

78.5 ⫾ 16.4 75.0 ⫾ 8.6 77.7 ⫾ 11.3 78.6 ⫾ 10.6

182 ⫾ 6 181 ⫾ 5 179 ⫾ 6 181 ⫾ 6

3.0 ⫾ 2.3a 10.1 ⫾ 3.1 8.0 ⫾ 2.6 8.1 ⫾ 2.8

82.7 ⫾ 15.9 78.6 ⫾ 9.3 81.7 ⫾ 12.2 80.0 ⫾ 9.4

183 ⫾ 6 182 ⫾ 5 180 ⫾ 6 181 ⫾ 5

2.5 ⫾ 3.3 10.0 ⫾ 3.6b 3.8 ⫾ 2.5 7.6 ⫾ 3.4

83.9 ⫾ 15.4 79.2 ⫾ 8.7 82.6 ⫾ 11.0 83.3 ⫾ 10.3

183 ⫾ 6 182 ⫾ 5 180 ⫾ 6 181 ⫾ 5

1.7 ⫾ 1.7 9.2 ⫾ 4.0c 3.1 ⫾ 1.9 3.1 ⫾ 1.6

Controls less than all other groups. Active cohort, former 2 cohort more than controls and former 1 cohort. c Active cohort more than all other groups. b

Third follow-up

Weight (kg)

2604

J Clin Endocrinol Metab, July 2006, 91(7):2600 –2604

relatively short follow-up times, potentially different reaction patterns to unloading in weight-bearing and nonweight-bearing bone, different activity levels after the cessation of intensive training, and/or different study designs (8, 13, 20 –23). In the present study, 27 athletes (former athletes 1) stopped intensive athletic training between the first and second follow-up. Between the second and third followup, this cohort did not lose BMD at any site in comparison with either the controls or the active athletes. At the final follow-up, the former athletes still had higher BMD of the proximal femur and humerus compared with the control group. Sustained benefits from previous training with no signs of losses have not before been demonstrated 5 yr after cessation of training. Based on prospective studies in older men, such benefits would reduce the risk of future fractures by at least 50% (24). Supporting this hypothesis is a study in which we evaluated the risk of fragility fractures in 400 male former athletes and 800 controls, aged 60 yr or more. The former athletes had fewer fragility fractures than the controls (2.0 vs. 4.2%; P ⬍ 0.05) and fewer distal radius fractures (0.75 vs. 2.5%; P ⬍ 0.05) (25). High-impact physical activity may also have an effect on bone size in children (15, 26). Greater bone size resulting from athletic training would increase the breaking strength of the bone, and greater bone size maintained throughout life could also reduce the risk of fracture. In the present study, we evaluated the bone area of the proximal femur because BMD of this site was most affected by training and detraining. However, clearly the type of weight-bearing loading the athletes were engaged in influenced only the BMD and not bone size. In conclusion, the novel design of the present study included also measurement of the parents’ BMD and suggested absence of selection bias. We report continuous BMD gains from rather constant physical loading after puberty in men at both weight-bearing and non-weight-bearing sites. Importantly, former athletes retain significantly higher BMD at the clinically important proximal femur and humerus 5 yr after a reduction in activity level. These results may suggest that a high peak BMD resulting from previous training may reduce the risk of osteoporotic fractures in men. Acknowledgments Received January 24, 2006. Accepted April 10, 2006. Address all correspondence and requests for reprints to: Peter Nordstro¨m, M.D., Ph.D., Sports Medicine Unit, Department of Surgical and Perioperative Sciences, Umeå University, 901 85 Umeå, Sweden. E-mail: [email protected]. This work was supported by grants from the La¨nsfo¨rsa¨kringar insurance company (project no. P4/01) and from the Swedish National Center for Research in Sports (project no. 112/01).

References 1. 1993 Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 94:646 – 650 2. Gullberg B, Johnell O, Kanis JA 1997 World-wide projections for hip fracture. Osteoporos Int 7:407– 413

Nordstro¨m et al. • Physical Activity and Bone

3. Raab-Cullen DM, Akhter MP, Kimmel DB, Recker RR 1994 Bone response to alternate-day mechanical loading of the rat tibia. J Bone Miner Res 9:203–211 4. Rubin CT, Lanyon LE 1987 Kappa Delta Award paper. Osteoregulatory nature of mechanical stimuli: function as a determinant for adaptive remodeling in bone. J Orthop Res 5:300 –310 5. Lanyon LE, Rubin CT 1984 Static vs dynamic loads as an influence on bone remodelling. J Biomech 17:897–905 6. MacKelvie KJ, Khan KM, McKay HA 2002 Is there a critical period for bone response to weight-bearing exercise in children and adolescents? A systematic review. Br J Sports Med 36:250 –257; discussion, 257 7. Bonjour JP, Theintz G, Buchs B, Slosman D, Rizzoli R 1991 Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab 73:555–563 8. Gustavsson A, Olsson T, Nordstrom P 2003 Rapid loss of bone mineral density of the femoral neck after cessation of ice hockey training: a 6-year longitudinal study in males. J Bone Miner Res 18:1964 –1969 9. Winters KM, Snow CM 2000 Detraining reverses positive effects of exercise on the musculoskeletal system in premenopausal women. J Bone Miner Res 15:2495–2503 10. Snow CM, Shaw JM, Winters KM, Witzke KA 2000 Long-term exercise using weighted vests prevents hip bone loss in postmenopausal women. J Gerontol A Biol Sci Med Sci 55:M489 –M491 11. Iwamoto J, Takeda T, Ichimura S 2001 Effect of exercise training and detraining on bone mineral density in postmenopausal women with osteoporosis. J Orthop Sci 6:128 –132 12. Michel BA, Lane NE, Bjorkengren A, Bloch DA, Fries JF 1992 Impact of running on lumbar bone density: a 5-year longitudinal study. J Rheumatol 19:1759 –1763 13. Kontulainen S, Kannus P, Haapasalo H, Sievanen H, Pasanen M, Heinonen A, Oja P, Vuori I 2001 Good maintenance of exercise-induced bone gain with decreased training of female tennis and squash players: a prospective 5-year follow-up study of young and old starters and controls. J Bone Miner Res 16:195–201 14. Bass S, Pearce G, Bradney M, Hendrich E, Delmas PD, Harding A, Seeman E 1998 Exercise before puberty may confer residual benefits in bone density in adulthood: studies in active prepubertal and retired female gymnasts. J Bone Miner Res 13:500 –507 15. Fuchs RK, Bauer JJ, Snow CM 2001 Jumping improves hip and lumbar spine bone mass in prepubescent children: a randomized controlled trial. J Bone Miner Res 16:148 –156 16. MacKelvie KJ, McKay HA, Khan KM, Crocker PR 2001 A school-based exercise intervention augments bone mineral accrual in early pubertal girls. J Pediatr 139:501–507 17. Heinonen A, Sievanen H, Kannus P, Oja P, Pasanen M, Vuori I 2000 Highimpact exercise and bones of growing girls: a 9-month controlled trial. Osteoporos Int 11:1010 –1017 18. Petit MA, McKay HA, MacKelvie KJ, Heinonen A, Khan KM, Beck TJ 2002 A randomized school-based jumping intervention confers site and maturityspecific benefits on bone structural properties in girls: a hip structural analysis study. J Bone Miner Res 17:363–372 19. Henry YM, Fatayerji D, Eastell R 2004 Attainment of peak bone mass at the lumbar spine, femoral neck and radius in men and women: relative contributions of bone size and volumetric bone mineral density. Osteoporos Int 15:263–273 20. Nordstrom A, Olsson T, Nordstrom P 2005 Bone gained from physical activity and lost through detraining: a longitudinal study in young males. Osteoporos Int 16:835– 841 21. Snow CM, Williams DP, LaRiviere J, Fuchs RK, Robinson TL 2001 Bone gains and losses follow seasonal training and detraining in gymnasts. Calcif Tissue Int 69:7–12 22. Karlsson MK, Linden C, Karlsson C, Johnell O, Obrant K, Seeman E 2000 Exercise during growth and bone mineral density and fractures in old age. Lancet 355:469 – 470 23. Vico L, Collet P, Guignandon A, Lafage-Proust MH, Thomas T, Rehaillia M, Alexandre C 2000 Effects of long-term microgravity exposure on cancellous and cortical weight-bearing bones of cosmonauts. Lancet 355:1607–1611 24. De Laet CE, Van Hout BA, Burger H, Weel AE, Hofman A, Pols HA 1998 Hip fracture prediction in elderly men and women: validation in the Rotterdam study. J Bone Miner Res 13:1587–1593 25. Nordstrom A, Karlsson C, Nyquist F, Olsson T, Nordstrom P, Karlsson M 2005 Bone loss and fracture risk after reduced physical activity. J Bone Miner Res 20:202–207 26. Morris FL, Naughton GA, Gibbs JL, Carlson JS, Wark JD 1997 Prospective ten-month exercise intervention in premenarcheal girls: positive effects on bone and lean mass. J Bone Miner Res 12:1453–1462

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.