The exact timing of peak bone mass attainment in young women is unclear; it has been speculated to be as early as late adolescence(3,15,31,47), or as late as the end of the third to beginning of the fourth decade(40,45). Regardless of the exact age at which peak bone mass is reached, it is recognized as a determinant of subsequent risk of osteoporotic fracture (28,33). Therefore, attainment and maintenance of peak bone mass play very important roles in the prevention of osteoporosis.
In a twin study, Pocock et al. (39) found that genetics was the principal determinant of bone mass in adults. However, several other factors play a role in the attainment of peak bone mass, including physical activity, body composition, hormonal status, and nutrition(11,19). These other factors can also modify the rate of loss during middle and old age, and thereby affect osteoporosis risk(42).
Body weight is thought to affect bone mineral density (BMD) in pre- and postmenopausal women by affecting the mechanical stress placed on the skeleton(26,27,41). The greater the body weight, the greater the mechanical forces placed on the skeleton in weight-bearing and, therefore, the greater the stimulus for osteogenesis(1,26,27). However, composition of the weight may be more important than body weight alone. Several theories regarding how body composition might influence BMD have been advanced, including both fat and lean mass. Some researchers(27,41) have shown the strongest relationship between fat mass and BMD, but these studies also demonstrated a relationship between body weight alone and BMD. Increased body weight is often associated with increased fat mass, and since fat may act as a peripheral site for the conversion of androgens to estrogens(1,22,41,46), the association between BMD and fat mass may be a result of hormonal as well as mechanical factors. Other investigators (35,46) have found stronger relationships between BMD and lean mass. Bone mass and BMD may be associated with lean mass because of the muscular forces leading to localized stresses on the bone, which then result in osteogenesis(27,28). It is unclear if BMD is most influenced by body weight alone, or by some component of body composition.
Women with regular, normal length menstrual cycles who have been physically active generally have a higher bone mass than their sedentary counterparts(45). It has been hypothesized that the stress placed on the bone through weight-bearing exercise, such as running or jumping, leads to osteogenesis and therefore increased BMD(4,13,36). College athletes involved in weight-bearing sports have been shown to have increased BMD at various sites compared with their nonathletic counterparts(20,21,35), even when matched for age, height, and weight (25). Swimmers have been shown to have lower lumbar spine BMD compared with weight-bearing athletes(42), lending further credibility to the notion that weight-bearing exercise stimulates osteogenesis(13,23,42,43).
Pocock et al. (38) and Snow-Harter et al.(44) observed site-specific adaptation of bone to increased local muscular forces. Others (35,46) have found significant correlations between muscular strength, lean tissue mass, and BMD, although these relationships were not always site-specific. It remains unclear, however, whether bone adapts specifically to increased local muscular forces because of increased strength, or if increased strength is more indicative of an increased lean body mass (LBM) and, therefore, increased total body BMD rather than site-specific increases.
The purpose of this study was to examine more closely the relationships between body weight, body composition, muscular strength, physical activity, and BMD among eumenorrheic, college-aged women. More specifically, it was hypothesized that low-body-weight female athletes involved in weight-bearing collegiate sports would have BMD values similar to a group of average-body-weight sedentary women, but greater than a group of low-body-weight sedentary women.
Subjects. Sixty females, ages 18-26 yr, gave their written, informed consent to participate in this study. (The study was approved by the Human Subjects Research Committee and Radiation Use Committee at the University of California, Davis) All subjects had at least three out of four grand-parents of Caucasian descent. Twenty of the women were engaged in weight-bearing athletic training, with all being current or past (N= 2) members of intercollegiate athletic teams. They were of low body weight relative to height, with a BMI value of less than 20.3 kg·m-2. Weight-bearing sports included gymnastics (N = 3), soccer(N = 5), volleyball (N = 3), track and field (N= 4), and cross-country (N = 5). The remaining 40 subjects ranged from sedentary to moderately active (i.e., physical activity less than 3 times per week, 30 min per session at light to moderate intensity), with none of them following any consistent weight-bearing exercise training program. Half of these subjects were of low body weight relative to height, i.e., BMI of 20.3 kg·m-2 or less; the remaining 20 comprised the average weight group, with BMI values between 20.4 and 25.2 kg·m-2. All groups were matched for age, age at menarche, and height, with the two low-body-weight groups also matched for weight.
A modified Brownell et al. (5) questionnaire administered to the subjects asked information regarding family and personal history of bone fracture, exercise training history, menstrual history, vitamin and mineral supplementation, estimations of weekly intake of certain foods, dietary preference (vegetarian or mixed diet), and eating pattern history. All individuals accepted as subjects were eumenorrheic (10-12 menses in the previous 12 months); women were not excluded for taking oral contraceptives (OC), unless taken for medical reasons. None were suffering from any disease known to affect bone metabolism. Women who were pregnant or had given birth were excluded. No subjects were strict vegan vegetarians, suffering from any eating disorder, taking megadoses of calcium (i.e., more than 2000 mg·d-1) or other minerals, smoked cigarettes, or reported any recreational drug use.
Protocol. Testing took place on two separate occasions. The first was conducted at the Human Performance Laboratory, U.C. Davis, in which body weight, height, and other anthropometric data, including body composition, as well as maximal strength measurements, were obtained. The second session took place at the USDA Western Human Nutrition Research Center, Presidio of San Francisco. The subjects' total body BMD and bone mineral content (BMC), L2-L4 BMD, and femoral neck BMD were measured by dual-energy x-ray absorptiometry(DXA) using a Lunar bone densitometer. Both sessions were completed within a 10-d period for all but three subjects.
Procedures. Body density (Db) was determined by hydrostatic weighing with the highest underwater weight of three to six trials used to calculate Db (6). Residual volume was determined by a modification of Wilmore's oxygen-rebreathing nitrogen(N2) dilution technique (49), employing a 9-L Collins respirometer and an Ohio 700 N2 analyzer. Skinfolds at four sites (triceps, subscapular, suprailiac, and mid-thigh), circumferences at three sites (mid-upper arm, mid-thigh, and head), measured using a spring-retractable tape measure, and five skeletal widths (biacromial, elbow, bitrochanteric, biiliac, and knee), taken using a sliding wooden caliper, were obtained using standard techniques (30).
Total body BMD and BMC, L2-L4 BMD, and femoral neck BMD were measured using a Lunar dual-energy x-ray absorptiometer (model DPX-L with version 3.6R software, Lunar Radiation Corp., Madison, WI). The amount of absorbed energy from the x-ray source is directly proportional to BMD, which is measured with a precision of better than 1% in vivo (9). The BMD is an area density, expressed in g·cm-2, which represents bone mass per unit of projected bone area (32). The three scans were completed in succession for each subject and took approximately 35 min. All scans were performed by the same experienced technician. Precision in this laboratory, expressed as the coefficient of variation, was less than 1% for L2-L4 and total body BMD and 2% for femoral neck BMD.
A LIDO Active Isokinetic Rehabilitation System was used to measure peak isometric torque. The accuracy and test-retest reliability of the LIDO system in this laboratory has been reported (7,34). All subjects were tested for maximal isometric torque of the torso, dominant arm elbow joint, and dominant leg knee joint. Subjects were familiarized with the LIDO apparatus and strength testing procedures before testing. A total of six maximal tests were completed on each subject. Torso flexion and extension were measured with the hip joint as the point of rotation and the angle set at 0° (i.e., sitting, upright posture). Elbow flexion and extension were performed at 60° and 80°, respectively (0° = full extension). Knee flexion and extension were completed at 20° and 60°, respectively(0° = full extension). To ensure proper muscle warm-up, subjects performed several submaximal isometric efforts before the maximal tests. Strength testing consisted of three maximal isometric contractions with 45-50 s rest between each contraction. Peak torque of each contraction was recorded by the LIDO software system. The highest torque value of the three contractions was taken as the maximal isometric torque. Flexor and extensor torques of the arm and leg were summed to produce a total torque for each limb. The torso flexion and extension values were evaluated separately.
Calculations. Percent body fat was computed from Db using a modified formula that corrects for individual variations in BMC via a three-component model originally derived by Lohman(29):%BF = (6.0/Db + 3.649bo -5.661 × 100, where bo = bone mineral as a decimal fraction of body weight(10).
Upper arm and thigh muscle plus bone cross-sectional area (M + B CSA) were calculated using an anthropometric formula incorporating limb circumference and skinfolds (17). Since these anthropometric formulas cannot distinguish between individual muscles, strength measurements for flexors and extensors of the upper arm and thigh, respectively, were added together and then divided by M+B CSA. Since measurement of CSA of trunk extensor and flexor muscles involved was not feasible, torso flexion and extension values were expressed per unit of total LBM.
Statistics. A one-factor ANOVA was performed to test for significant differences among the three groups. Upon obtaining a significantF-ratio, post hoc Scheffe F tests were performed to determine which mean values were significantly different from each other. Pearson product-moment correlations between variables were also calculated and their significance from zero was tested. Finally, stepwise multiple regressions, using LBM (or leg M + B CSA), body weight, height, fat mass, and strength measurements to predict total body, L2-L4, and femoral neck BMD, respectively, were performed. Statistical significance was accepted at theP < 0.05 level.
Subject characteristics and body composition of the three groups, referred to hereafter as the athletes, low weight sedentary (LWS), and average weight sedentary (AWS), are given in Table 1. There were no significant differences among the groups in age at menarche, nor for the number of subjects in each group taking OC. Among those taking OC (athletes = 10, LWS = 12, AWS = 9), there was no significant difference among the groups for the average length of time the subjects had been taking OC. Reflecting their differences in body weight and/or activity levels, the athletes' mean daily caloric intake was 2,543 kcal, whereas that for the AWS group was 1,908 kcal and that for the LWS group was 1,674 kcal. Nearly the same number of subjects in each group reported eating a standard mixed diet (athletes = 14; LWS = 16; and AWS = 15), with three subjects in each group consuming a semivegetarian diet (i.e., includes some chicken or fish, but little or no beef), and the remainder an ovo-lacto vegetarian diet.
All but three of the athletes participated in weight training, averaging 2.4 h·wk-1. The athletes began training at an average age of 12.4 ± 1.9 yr and had been training four or more d·wk-1 for 7.4 ± 3.5 yr. Weekly training varied depending on the sport, but averaged 13-17 h·wk-1. In the sedentary groups, no subject had been involved in regular, significant weight-bearing activity (i.e., walking or jogging > 1.5 h·wk-1) in the past 12 months, but three of the LWS subjects had participated in significant activity after 18 yr of age(sports or dancing, including aerobic dance ≈10 h·wk-1), with two of these three also participating in weight-bearing activity during adolescence. In the AWS group, three of the subjects participated in weight-bearing activity during adolescence, with one subject also participating in drill team dancing (≈10 h·wk-1) for 1 yr after 18 yr of age. None of the sedentary subjects reported any significant weight-bearing activity before high school (i.e., age 14 yr).
Each of the three groups differed in percent body fat corrected for BMC, with the athletes being the lowest (13.8%), followed by the LWS group (23.2%) and the AWS group (27.1%). The athletes and the AWS group did not differ in LBM, but both were significantly greater than the LWS group. Fat mass was lowest for the athletes and highest for the AWS group, with the three groups significantly different from each other. The arm and leg M+B CSA values revealed that the athletes were not significantly different from the AWS group in either the arm or leg. The athletes were significantly greater than the LWS group in leg M+B CSA, but not arm M+B CSA. The AWS group had significantly greater arm and leg M+B CSA than the LWS group.
Group skeletal widths and bone mineral data are shown inTable 2. The athletes and LWS group did not differ significantly in any skeletal width. Except for biacromial width, the AWS group had significantly greater skeletal widths than both of the other groups. The athletes did not differ significantly from the AWS group in total body BMD or L2-L4 BMD; however, their femoral neck BMD was significantly greater than that of the AWS group. Further, despite similar body weight, height, and skeletal widths, all of the athletes' BMD values were significantly greater than those of the LWS group. The LWS group's total body BMD was significantly less than that for the AWS group. Total body BMC values as a decimal fraction of LBM did not differ significantly among groups.
Group strength (peak isometric torque) data are given inTable 3. Significant differences in absolute total arm strength (sum of flexion and extension) and in absolute total leg strength were observed between the athletes and LWS group. However, when expressed per unit of M+B CSA, the athletes and LWS group did not differ significantly in either total arm or total leg strength. The AWS group did not differ significantly from the other groups in absolute total arm strength or in absolute total leg strength, but when corrected for M+B CSA the AWS group had significantly lower relative strength values than did the athletes. The absolute strength values for torso flexion revealed a significant difference only between the athletes and LWS group, which remained when strength was normalized for LBM. The athletes had significantly greater absolute torso extension strength than both sedentary groups, which did not differ significantly from each other. When torso extension strength was normalized for LBM, the athletes maintained a significantly greater value over both sedentary groups, which did not differ from each other.
Correlation coefficients between strength and BMD for all groups combined are given in Table 4. Correlations between absolute total(flexion plus extension) limb strength and total body BMD were significant, as were those for femoral neck BMD. Absolute total arm strength was significantly correlated with L2-L4 BMD which, however, was not significantly correlated with absolute total leg strength. When absolute total arm and absolute total leg strength values were corrected for M+B CSA, correlations with BMD variables were consistently lower, and no relationship achieved significance.
Stepwise multiple regression prediction of total body BMD from seven variables, including LBM, body weight, height, fat mass, and absolute strength measurements, resulted in LBM being selected first (r = 0.49) and trunk extension strength second (r = 0.55; SE = 0.06). No other variable significantly improved the predictive precision of total body BMD. In the case of lumbar BMD, LBM was selected first (r = 0.51; SE = 0.10), with none of the other six variables significantly improving predictive precision. Six variables, including leg M + B CSA, were used to predict femoral neck BMD. Again, LBM was selected first (r = 0.52), with fat mass second (r = 0.56; SE = 0.11); none of the other variables significantly improved predictive precision.
The results of this cross-sectional study demonstrate that a group of athletes, although categorized as low body weight (i.e., BMI < 20.3 kg·m-2), had levels of BMD equal to and, at the femoral neck, greater than age-matched AWS subjects. Compared with age- and height-matched LWS subjects, the athletes had significantly greater BMD at all sites tested. These results demonstrate that, in addition to the influence of body weight on BMD, body composition must also be considered. Results of this study also suggest that prolonged, high level physical activity, such as that required for weight-bearing sports and which begins before menarche, may increase bone osteogenesis and achievement of a higher peak BMD. No consistent association between relative strength (i.e., per unit of M + B CSA or LBM) and BMD at specific sites was observed in the present study.
Body weight and composition. Several investigators(1,26,27,41) have observed a positive relationship between body weight and BMD, although it may be influenced by subject age, menopausal status, and body composition(1,11). Reid et al. (41) found a significant correlation between body weight and total body BMD (r = 0.69; P < 0.001) in premenopausal women. These authors also observed that total fat mass (r = 0.60) was a better predictor of total body BMD than LBM (r = 0.55) in premenopausal women. However, their subjects were older (mean age = 33 ± 8 yr) and had a greater percentage of fat (mean= 31 ± 8%) than did our subjects (21.4%).
As was observed by Reid et al. (41), a positive relationship between body weight and total body BMD (r = 0.48; P< 0.01) was found for the sedentary groups in this study (AWS and LWS,N = 40). A significant relationship between body weight and femoral neck BMD (r = 0.41; P < 0.01) also existed in the sedentary groups. However, when the athletes were included in the correlation analyses(N = 60), the strength of all correlations decreased, and the relationship between body weight and femoral neck BMD became nonsignificant (r= 0.21).
Similar to the analyses of body weight and BMD in the present study, analysis of the sedentary groups alone (N = 40) revealed significant correlations between fat mass and total body BMD, L2-L4 BMD, and femoral neck BMD (P < 0.01). However, inclusion of the athletes (N = 60) led to nonsignificant correlations between fat mass and total body BMD, L2-L4 BMD, and femoral neck BMD. These results indicate that the relationship between body weight and BMD and between fat mass and BMD are not always statistically significant but are affected and, in the case of femoral neck BMD, negated when weight-bearing athletes are included with sedentary young women.
Several investigators have found LBM to be a better predictor of BMD than fat mass (1,35,46). Sowers et al.(46) found that the greater the amount of LBM in premenopausal women, the greater the BMD, regardless of fat mass. That is, large amounts of lean mass led to greater femoral neck BMD in women with both low and high amounts of fat mass. Further, subjects with high fat mass but low lean mass had lower BMD. Nichols et al. (35) observed, in subjects of similar age and height to those of the present study, that regional lean tissue mass, as opposed to regional fat mass, was a better predictor of regional BMD. In the present study, the correlations between LBM and BMD variables in the sedentary groups alone (N = 40) were not statistically significant (r = 0.22 to 0.28). Inclusion of the athletes in the analysis with the sedentary subjects (N = 60) led to significant correlations between LBM and all BMD variables (r = 0.43 to 0.45; P< 0.001). Indeed, LBM for the athletes (N = 20) was well related to BMD at all sites: (1) 0.64 for total body BMD, (2) 0.69 for L2-L4 BMD, and(3) 0.82 for femoral neck BMD (P < 0.001). Thus, it appears that LBM in young adult women varying substantially in body weight and physical activity level is more closely related to BMD than is fat mass.
Muscular strength. Research has shown muscular strength to be significantly related to BMD at several sites in premenopausal women(22,38,44). However, in these studies absolute strength was used without measuring body composition and correcting for LBM, muscle CSA, or M+B CSA. Going and Lohman (16) observed that once strength measurements were corrected for muscle mass, the apparent independent effect of strength on BMD and BMC failed to reach statistical significance.
In the present study, total arm and total leg strength corrected for M+B CSA did not show consistent differences among groups (Table 3). For example, the athletes and AWS group differed significantly in only femoral neck BMD, yet differed significantly in relative strength of both the arm and leg (Table 3). Further, the LWS group did not differ significantly from the athletes in relative strength of the arm or leg(Table 3), yet had significantly lower total body, L2-L4, and femoral neck BMD. If relative strength was a true independent predictor of BMD, these similar relative strength values should not exist between two groups of subjects with dissimilar BMD values. The LWS group had significantly lower absolute strength in both the arm and leg than the athletes. These results merit further investigation of the relationship between relative strength and BMD, rather than absolute strength, which may reflect primarily the amount of M+B CSA instead of being an independent influence on BMD.
Block et al. (2) observed no significant relationship between back strength and spine and hip BMD in a group of young men that included weight training participants, water polo athletes, and sedentary controls. In the present study, several significant correlations were observed between torso strength and BMD (Table 4), but when torso strength measurements were corrected for LBM (i.e., relative strength), correlations with all BMD variables were consistently reduced. However, both relative torso flexion and torso extension remained significantly related to total body BMD, although neither was significantly correlated to L2-L4 BMD(Table 4).
In this study, stepwise multiple regression analysis of the prediction of BMD from LBM (or leg M + B CSA), height, body weight, fat mass, and absolute strength measurements resulted in LBM being selected first at all sites. None of the strength variables added significantly to predictive precision for total body, L2-L4, or femoral neck BMD. These results strongly support our contention that LBM is more closely associated with BMD than absolute strength and that the latter does not have an independent influence on BMD.
Physical activity. The significantly greater total body BMC and BMD, L2-L4 BMD, and femur BMD in the athletes compared with their weight-matched sedentary counterparts strongly suggests that prolonged, intense weight-bearing exercise has a substantial impact on bone remodeling in eumenorrheic young women. However, it is possible that the disparate group BMD results in this cross-sectional study are mainly a result of self-selection, in that young girls who are genetically predisposed to acquire greater BMD may be more likely to choose and continue participation in weight-bearing sports, while those who may be genetically predisposed to lower BMD drop out (perhaps because of injury) or choose not to initiate such activity. However, Vuori(48) contends that side-to-side comparison of the bones of athletes who play sports that cause unilateral loading, such as tennis and squash (24), strongly suggests that the high BMD values found in athletes are caused mainly by bone-loading exercise and are not the result of self-selection or other confounding factors.
Longitudinal studies of young women have shown much lower increases in BMD with exercise training than those observed in cross-sectional studies of athletes compared with sedentary controls, with the greatest increases being 4-6% consequent to strength and impact training(8,11,14). In the present study, the athletes had 8.7%, 8.5%, and 15.1% greater mean values than the LWS group for total body, L2-L4, and femoral neck BMD, respectively. Vuori(48) contends that there are two primary reasons for larger BMD differences observed in cross-sectional studies than in exercise training studies: 1) the intensity, duration, and total amount of training of the athletes greatly exceeds those used in training studies; and 2) a great majority of athletes begin training in childhood or early adolescence, a period when rapid bone mineral accretion normally occurs(15,31,47).
Haapasalo et al. (18) found significantly greater values of BMD and BMC of the humerus in the playing arms compared with the nonplaying arms of female squash players who began their career before or during menarche compared with those players who had begun training 1 yr or more after menarche. The athletes in the present study were much more physically active during their adolescent and early adult years compared with the sedentary groups. On average, the athletes began consistent physical training at a mean age of 12.4 yr and reached menarche at 13.4 yr(Table 1). Further, only two of the 20 athletes started training more than 1 yr following menarche.
Eisman et al. (12) observed in their co-twin control studies that femoral neck BMD was more sensitive to environmental factors, such as physical loading, than was lumbar BMD, which was more strongly influenced by genetics. Wolman et al. (50) found that elite female runners had significantly higher mid-shaft femur BMD than did rowers, dancers, and sedentary controls, who had similar mean values. They contended that this was because of the considerable cyclical loading of the lower body experienced in the weight-bearing exercise of running up to 70 miles/wk, while the dancers spent most of their work time in movements involving coordination, balance, and flexibility, with less than 10% involving jumping. In the present study, the athletes' femoral neck BMD was the only site that was significantly greater than that for both the AWS group (11.8%) and the LWS group (15.1%). The athletes were all involved in weight-bearing sports, supporting the contention that physical activity influenced the accumulation of mineral in the femoral neck via intense cyclical loading(37,50). While these cross-sectional data strongly suggest that prolonged, intense weight-bearing activity typical of training and participation in competitive sports can significantly increase BMD in young women, longitudinal study of the influence of weight-bearing activity over several years on BMD, beginning just before or at menarche, is warranted.
The authors gratefully acknowledge the technical assistance of Richard Fadling of the Human Performance Laboratory, University of California, Davis, and Patrick Mayclin of the Bioenergetics Research Unit, USDA Western Human Nutrition Research Center, Presidio of San Francisco, CA. The authors also acknowledge the loan of the LIDO back attachment device arranged by Dr. Stanley Geel, Director of the Physical Therapy Program, California State University, Sacramento. Sincere appreciation is extended to the subjects for their contribution of time and effort.
Address for correspondence: William C. Adams, Department of Exercise Science, University of California, Davis, CA 95616.
1. Aloia, J. F., A. Vaswani, R. Ma, and E. Flaster. To what extent is bone mass determined by fat-free or fat mass? Am. J. Clin. Nutr.
2. Block, J. E., A. L. Friedlander, G. A. Brooks, P. Steiger, H. A. Stubbs, and H. K. Genant. Determinants of bone density among athletes engaged in weight-bearing and non-weight-bearing activity. J. Appl. Physiol.
3. Bonjour, J. P., G. Theintz, B. Buchs, D. Slosman, and R. Rizzoli. Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J. Clin. Endocrinol. Metab.
4. Brewer, V., B. M. Meyer, M. S. Keele, S. J. Upton, and R. D. Hagan. Role of exercise in prevention of involutional bone loss.Med. Sci. Sports Exerc.
5. Brownell, K. D., S. N. Steen, and J. H. Wilmore. Weight regulation practices in athletes: analysis of metabolic and health effects.Med. Sci. Sports Exerc.
6. Buskirk, E. R. Underwater weighing and body density: a review of procedures. In: Techniques for Measuring Body Composition
. J. Brozek and A. Henschel (Eds.). Washington, DC: NASNRC, 1961, pp. 90-106.
7. Castro, M. J., D. J. McCann, J. D. Shaffrath, and W. C. Adams. Peak torque per unit cross-sectional area differs between strength-trained and untrained young adults. Med. Sci. Sports Exerc.
8. Chilibeck, P. D., D. G. Sale, and C. E. Webber. Exercise and bone mineral density. Sports Med.
9. Cullum, I. D., P. J. Ell, and J. P. Ryder. X-ray dual photon absorptiometry: a new method for the measurement of bone density.Br. J. Radiol.
10. Deck-Coté, K. and W. C. Adams. Effect of bone density on body composition estimates in young black and white women.Med. Sci. Sports Exerc.
11. Drinkwater, B. L. Does physical activity play a role in preventing osteoporosis? Res. Q. Exerc. Sport
12. Eisman, J. A., P. J. Kelly, N. A. Pocock, J. Birmingham, and P. N. Sambrook. Exercise and its interaction with genetic effects on bone: co-twin control studies of exercise and bone density.Abstracts of the 3rd International Conference on Physical Activity, Aging and Sports
, Jyvaskla, May 31-June 4, 1992, p. 38.
13. Fehling, P. C., L. Alekel, J. Clasey, A. Rector, and R. J. Stillman. A comparison of bone mineral densities among female athletes in impact loading and active loading sports. Bone
14. Friedlander, A. L., H. K. Genant, S. Sadowsky, N. N. Byl, and C. C. Gluer. A two-year program of aerobics and weight training enhances bone mineral density of young women. J. Bone Min. Res.
15. Gilsanz, V., D. T. Gibbens, M. Carlson, M. I. Boechat, C. E. Cann, and E. E. Schulz. Peak trabecular vertebral density: a comparison of adolescent and adult females. Calcif. Tiss. Int.
16. Going, S. and T. Lohman. Letters to the editor-in-chief. Med. Sci. Sports Exerc.
17. Guerney, J. M. and D. B. Jelliffe. Arm anthropometry in nutritional assessment: nomogram for rapid calculation of muscle circumference and cross-sectional muscle and fat areas. Am. J. Clin. Nutr.
18. Haapasalo, H., P. Kannus, H. Sievänen, A. Heinonen, P. Oja, and I. Vuori. Long-term unilateral loading and bone mineral density and content in female squash players. Calcif. Tiss. Int.
19. Halioua, L. and J. J. B. Anderson. Lifetime calcium intake and physical activity habits: independent and combined effects on the radial bone of healthy premenopausal Caucasian women. Am. J. Clin. Nutr.
20. Heinonen, A., P. Oja, P. Kannus, H. Sievänen, A. Mänttäri, and I. Vuori. Bone mineral density of female athletes in different sports. Bone Min.
21. Heinrich, C. H., S. B. Going, R. W. Pamenter, C. D. Perry, T. W. Boyden, and T. G. Lohman. Bone mineral content of cyclically menstruating female resistance and endurance trained athletes. Med. Sci. Sports Exerc.
22. Hyakutake, S., S. Goto, M. Yamagata, and H. Moriya. Relationship between bone mineral density of the proximal femur and lumbar spine and quadriceps and hamstrings torque in healthy Japanese subjects.Calcif. Tiss. Int.
23. Jacobsen, P. C., W. Beaver, S. A. Grubb, T. N. Taft, and R. V. Talmage. Bone density in women: college athletes and older athletic women. J. Orthop. Res.
24. Kannus, P., H. Haapasalo, H. Sievanen, P. Oja, and I. Vuori. The site-specific effects of long-term unilateral activity on bone mineral density and content. Bone
25. Kirchner, E. M., R. D. Lewis, and P. J. O'Connor. Bone mineral density and dietary intake of female college gymnasts. Med. Sci. Sports Exerc.
26. Lanyon, L. E. Control of bone architecture by functional load bearing. J. Bone Min. Res.
27. Lindsay, R., F. Cosman, B. S. Herrington, and S. Himmelstein. Bone mass and body composition in normal women. J. Bone Min. Res.
28. Lloyd, T., N. Rollings, M. B. Andon, et al. Determinants of bone density in young women. I. Relationships among pubertal development, total body bone mass, and total body bone density in premenarcheal females. J. Clin. Endocrinol. Metab.
29. Lohman, T. G. Applicability of body composition techniques and constants for children and youths. Exerc. Sports Sci. Rev.
30. Lohman, T. G., A. F. Roche, and R. Martorell.Anthropometric Standardization Reference Manual
. Champaign, IL: Human Kinetics, 1988, pp. 55-68.
31. Matkovic, V., T. Jelic, G. M. Wardlaw, et al. Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. J. Clin. Invest.
32. Mazess, R. B. and H. S. Barden. Measurement of bone by dual photon absorptiometry and dual-energy x-ray absorptiometry. Ann. Chir. Gynaecol.
33. Mazess, R. B. and H. S. Barden. Bone density in premenopausal women: effects of age, dietary intake, physical activity, smoking, and birth-control pills. Am. J. Clin. Nutr.
34. McCrory, M. A., S. G. Aitkens, C. M. Avery, and E. M. Bernauer. Reliability of concentric and eccentric measurements on the LIDO active isokinetic rehabilitation system (Abstract). Med. Sci. Sports Exerc.
35. Nichols, D. L., C. F. Sanborn, S. L. Bonnick, B. Gench, and N. Dimarco. Relationship of regional body composition to bone mineral density in college females. Med. Sci. Sports Exerc.
36. Nilsson, B. E. and N. E. Westlin. Bone density in athletes. Clin. Orthop. Rel. Res.
37. Norkin, C. C. and P. K. Levangie. Joint Structure and Function: A Comprehensive Analysis
. 2nd Ed. Philadelphia: F.A. Davis, 1992, pp. 300-336.
38. Pocock, N., J. Eisman, T. Gwinn, P. Sambrook, P. Kelly, J. Freund, and M. Yeates. Muscle strength, physical fitness and weight but not age predict femoral neck bone mass. J. Bone Min. Res.
39. Pocock, N. A., J. A. Eisman, J. L. Hoper, M. G. Yeates, P. N. Sambrook, and S. Eberl. Genetics determinants of bone mass in adults: a twin study. J. Clin. Nutr.
40. Recker, R. R., M. Davies, S. M. Hinders, R. P. Heany, M. R. Stegman, and D. B. Kimmel. Bone gain in young adult women.JAMA
41. Reid, I. R., L. D. Plank, and M. C. Evans. Fat mass is an important determinant of whole body bone density in premenopausal women but not in men. J. Clin. Endocrinol. Metab.
42. Riggs, B. L. Pathogenesis of osteoporosis. Am. J. Obstet. Gynecol.
43. Risser, W. L., E. J. Lee, A. Leblanc, H. B. W. Poindexter, J. M. H. Risser, and V. Schneider. Bone density in eumenorrheic female college athletes. Med. Sci. Sports Exerc.
44. Snow-Harter, C., M. Bouxsein, B. Lewis, S. Charette, P. Weinstein, and R. Marcus. Muscle strength as a predictor of bone mineral density in young women. J. Bone Min. Res.
45. Snow-Harter, C. and R. Marcus. Exercise, bone mineral density, and osteoporosis. Exerc. Sport Sci. Rev.
46. Sowers, M. R., A. Kshirsagar, M. M. Crutchfield, and S. Updike. Joint influence of fat and lean body composition compartments on femoral bone mineral density in premenopausal women. Am. J. Epidemiol
. 136:257-265, 1992.
47. Theintz, G., B. Buchs, R. Rizzoli, et al. Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J. Clin. Endocrinol. Metab.
48. Vuori, I. Exercise and physical health: musculoskeletal health and functional capabilities. Res. Q. Exerc. Sport
49. Wilmore, J. H. A simplified method for determination of residual lung volume. J. Appl. Physiol.
50. Wolman, R. L., L. Faulmann, P. Clark, R. Hesp, and M. G. Harries. Different training patterns and bone mineral density of the femoral shaft in elite, female athletes. Ann. Rheum. Dis.