Department of Foods and Nutrition, Purdue University, West Lafayette, Indiana 47907; and Indiana University School of Medicine, Indianapolis, Indiana 46223
Address all correspondence and requests for reprints to: Connie M. Weaver, Ph.D., Department of Foods and Nutrition, Purdue University, 1264 Stone Hall, West Lafayette, Indiana 47907-1264. E-mail: weavercm{at}cfs.purdue.edu
The adolescent years are a window of opportunity to influence lifelong bone health. Approximately 40% of peak bone mass in girls is accumulated in this short stage of the life cycle (1). Bone acquisition is accelerated within the genetic potential by lifestyle choices, including diet, eating behavior, exercise, and smoking. The focus of this review is on the role of diet and individual nutrients in building a healthy skeleton in females. The major raw materials of bone are minerals and bone matrix nutrients. The macrominerals include calcium, phosphorus, and magnesium. The skeleton serves a structural function and also as a reservoir for calcium to maintain serum levels within normal range during periods of inadequacy. Bone matrix nutrients include the energy nutrients and a variety of nutritional cofactors.
It is commonly accepted that development of a higher peak bone mass during adolescent years protects against postmenopausal osteoporosis. There is a strong inverse relationship between bone mineral density and incidence of fracture in postmenopausal women (2, 3). This was demonstrated in a study of two communities in Yugoslavia with different levels of dairy food consumption (4). In this cross-sectional study, adults from the high dairy food-consuming community had higher initial bone density and fewer fractures despite a rate of bone loss similar to that in the low dairy food-consuming community. Bone loss in a postmenopausal woman with low peak bone density will have greater consequences for fracture than that in a woman with higher peak bone mineral density.
Recommending good nutritional habits to adolescents for bone health is not particularly controversial, although as will be discussed, there is some debate on the appropriate intake levels of required nutrients. More problematic is how to influence the behavior of this particular group, the adolescent. Teenage girls are notoriously concerned about body image and frequently respond more to peer pressure than to good advice. Therefore, it is wise to instill in children good eating habits before adolescence and to be armed with facts for every teachable moment for adolescents.
Peak bone mass
Puberty is a period of enormous skeletal growth, although individuals differ in skeletal acquisition rates and the age at which they attain peak bone mass. In a longitudinal study of Canadian girls, mean peak skeletal accretion occurred at age 13 yr, and the peak bone mineral content velocity was 240 g/yr (5). In British girls, the peak velocity was associated with menarche, which occurs more than 2 yr after peak height velocity (6). In 247 American women, aged 1132 yr, the average age of attainment of 90% total body bone mineral content was 16.9 ± 1.3 yr; for 99%, it was 26.2 ± 3.7 yr (7). Different skeletal sites achieve peak bone mass at very different times. For example, the hip achieves a peak earliest at age 1618 yr, whereas the vertebrae continue to be responsive to diet and exercise for several more years (1).
Not only is low bone mineral density associated with fractures late in life, but new evidence shows that bone mineral density predicts fractures in children. In 100 girls, aged 315 yr, with recent distal forearm fractures and 100 controls, the odds ratios for low bone density in the fracture group compared to the control group were 2.2 for the ultradistal radius, 2.6 for the lumbar spine, and 2.0 for the trochanter (8). Thus, the importance of attaining a high peak bone mass is not exclusively pertinent to the postmenopausal years.
Nutrition
Life-long adequate nutrition, especially calcium, is associated with good bone health. The role of calcium in adolescent growth has been reviewed by Peacock (9). For example, in the higher dairy food-consuming region in Yugoslavia already mentioned (3), bone density was greater than that in a region that consumed few dairy products. The association of bone density with dairy product consumption has also been observed in China (10). Retrospective studies in postmenopausal women show that bone density is associated with childhood and adolescent milk consumption (11). A low consumption of dairy products is not only associated with decreased intakes of calcium, but also with vitamin A, folate, riboflavin, vitamin B6, magnesium, and potassium (12).
A positive energy balance from macronutrients is important during
growth for synthesis of bone and also the muscles that exert
contractile forces on bone, thereby influencing modeling. Recently, the
Food and Nutrition Board of the National Academy of Science revised
requirements for those nutrients considered most related to bone
health, i.e. calcium, vitamin D, phosphorus, and magnesium.
Requirements for adolescents are given in Table 1. Many nutrients in more microquantities
may also influence bone health, including vitamin K, zinc, manganese,
copper, boron, and vitamin C.
|
Protein, fat, and carbohydrates are the energy nutrients. Energy balance has trended toward being excessive in children in recent years. Although bone health may not be compromised by being overweight, and indeed, bone mass and density are positively related to body weight, perhaps through mechanical loading, it would be irresponsible for the sake of skeletal health to advocate an unhealthy energy balance. On the other hand, bone health is compromised by excessive thinness whether it be due to eating disorders, excess exercise, or insufficient dietary fat. Excessive thinness provides less weight-bearing load on the skeleton, and if accompanied by menstrual dysfunction, estrogen is also deficient, which further compromises skeletal growth.
Macronutrients also have nonenergy roles in skeletal growth. For example, fat regulates PG synthesis, an essential cytokine in regulating bone metabolism. The effect of biologically active lipids on bone health is an area of current investigation.
Protein comprises most of the nonmineral composition of bone, and an adequate intake is essential for the synthesis of bone matrix. Protein intake and bone mass gains are positively correlated in children. The mechanism is unknown, but may relate to insulin-like growth factor (IGF) production. Eighteen months of milk supplementation increased serum IGF-I in adolescent girls relative to that in the control group (35% vs. 25%; P = 0.02), presumably through the increase in protein consumption (6). IGF-I stimulates bone growth by enhancing chondrocyte proliferation and differentiation in the growth plate, osteoblast proliferation and differentiation, and collagen synthesis.
Bone mineral nutrients
Calcium. By far the most studied nutrient for its role in bone health is calcium. A large body of literature, including epidemiological evidence, randomized clinical trials, and metabolic balance studies, indicates that dietary calcium is a determinant of skeletal, i.e. calcium, accretion. This is a logical consequence of calcium being the dominant mineral of the skeleton, existing as hydroxyapatite.
Randomized clinical trials in children and adolescents all show greater increments in bone mass and bone density with calcium supplementation whether in the form of supplements, fortified foods, or dairy products (13, 14, 15, 16, 17). The skeletal advantage is not maintained (at least significantly) on follow-up visits after cessation of the supplementation, suggesting that higher calcium intakes must be maintained to achieve higher peak bone mass. To determine the calcium intake that maximizes retention during adolescence, 12- to 14-yr-old girls were studied on a range of calcium intakes in a metabolic study (18). The calcium intake above which little further increase in calcium retention occurred was 1300 mg/day. The 95% confidence interval was large; 75% of the variation could be explained by postmenarcheal age. Thus, calcium accretion is not uniform across puberty. Because of the large variation, it may be useful to individualize calcium requirements. Regression models showed that approximately 75% of the variation in calcium retention could be explained by a model containing a measure of sexual maturity or physical growth and a measure of rate of bone turnover (19). The model using postmenarcheal age (PMA) and serum osteocalcin (OC) was as follows: calcium retention (mg/day) = 137 + 3.99 (OC, pg/L) - 15.0 (PMA, yr). Thus, it is clinically possible to estimate calcium retention and, therefore, desirable calcium intakes for each stage of growth on an individual basis.
Less understood is the long term effect on bone health of suboptimal calcium intake or supplementation during a restricted period, particularly during adolescence. Does the skeleton "catch up" if adequate calcium is restored after a period of restriction? One study in rats demonstrated a nonreversible, deleterious effect on peak bone mass due to calcium restriction throughout adolescence. After rats were restricted in calcium from weanling to 20 weeks (through adolescence) and then restored to recommended levels of calcium until 37 weeks of age, their adult tibia bone volume remained lower than that in the control group and was similar to that in rats kept on the calcium-deficient diet throughout (20). The deficient diets contained half the calcium requirements of rats. This is an extreme deficiency but comparable to the mean calcium intakes of American female adolescents. Approximately 69% of the recommended intake is consumed by 9- to 13-yr-old girls, and 55% of the recommended intake is consumed by 14- to 18-yr-old girls. The effects of more marginal deficiencies and shorter periods of calcium deprivation are probably more subtle.
Approximately three fourths of the calcium in the American diet comes from dairy products. One 8-oz glass of milk provides 300 mg calcium. To achieve the recommendation of 1300 mg calcium, the adolescent requires at least four glasses of milk each day or the equivalent, such as one cup of yogurt or 1.5 oz of cheese. The mean calcium intake for girls aged 913 yr is 889 ± 42 mg/day, which falls to 713 ± 42 mg/day at age 1418 yr according to the 1994 USDA Continuing Surveys of Food Intakes by Individuals data. The low intake of calcium by adolescent girls is at least partially attributed to the trend of replacing milk as the beverage of choice with soft drinks. Soft drink and fruit drink daily consumption increased by 27% and 26%, respectively, from 19891991 to 19941996 as determined by the Continuing Surveys of Food Intakes by Individuals (21). Adolescent girls may avoid dairy products for fear of gaining weight. However, girls did not gain more weight or percent body fat when given dairy products compared to a calcium supplement in a 4-yr longitudinal study (22). For adolescents who avoid dairy products, calcium needs must be met through fortified foods or supplements.
Magnesium. Approximately 60% of the magnesium in the body is in bone, although there is a paucity of evidence that dietary magnesium influences the development of peak bone mass. Magnesium is required for matrix and mineral metabolism in the bone through its indispensable role in metabolism of ATP and as a cofactor for over 300 enzymes. It may also increase bone quality by decreasing hydroxyapatite crystal size, thereby preventing larger, more perfect mineral crystals that result in brittle bone.
Adolescents who avoid dairy products are less likely to consume the
recommended intakes of magnesium (Table 1), as dairy products provide
20% of the U.S. dietary magnesium. Other sources include nuts, whole
grains, and green leafy vegetables. The mean intake of magnesium for
girls aged 913 yr is 223 ± 8 mg/day or about 93% of the
Recommended Daily Allowance. The situation for girls aged 1418 yr
deteriorates, as their intake of 217 ± 11 mg/day is only about
60% of the requirement.
Phosphorus and vitamin D. Almost 85% of the bodys
phosphorus is in bone. Phosphorus requirements for adolescents (Table 1) were calculated from the need to meet expanding lean body mass and
bone mass, adjusted for urinary losses and absorption efficiency.
During puberty, the efficiency of conversion of 25-hydroxyvitamin D to
1,25-dihydroxyvitamin D increases, which enhances calcium absorption to
meet the needs of the rapidly growing skeleton (23). Therefore, vitamin
D requirements are not higher at this age.
Dietary phosphorus and vitamin D are not of great concern in adolescent girls. Mean phosphorus intakes approximate the requirements. Phosphorus depletion by diet is unusual, and when it occurs, rickets ensues. Most adolescents are able to synthesize sufficient vitamin D by brief exposure to sunlight. Vitamin D stores accumulated during summer generally suffice throughout the winter months. However, children who live in the far northern or far southern latitudes may require a vitamin D supplement (24).
Bone matrix nutrients. Bone matrix contains collagen, elastin, and other proteins whose synthesis requires an array of nutrients, including vitamins C, D, and K, and several minerals, including copper, manganese, and zinc. The relationships of the status of these nutrients to bone health are not well studied, and data in adolescents are scarce. Examination of monthly diet records over 3 yr in 6- to 14-yr-old identical twins (13) failed to show any association between dietary variables and bone mineral density of the spine, hip, or radius (our unpublished data). In a study of 456 Caucasian girls, aged 813 yr, the most significant predictors of bone mass included dietary calcium (25). Protein or phosphorus could be substituted for calcium in the model, probably because these nutrients are frequently consumed together in dairy foods. Other nutrients evaluated in this cross-sectional analysis did not contribute to the model. Failure to find relationships between diet and bone mass in a generally well nourished population does not mean that adequate intakes of these nutrients are unimportant for bone health. Several of these nutrients have not been studied because they have not been available in food composition databases. Vitamins K and D and boron are not available, and other nutrients, such as copper, magnesium, and manganese, are incompletely available in some databases and not available at all in others.
Although many nutrients are involved in the general health and growth
of all tissues, including bone, there are some functions that are worth
discussing with specific reference to bone matrix. Ascorbic acid is
necessary for hydroxylation of protein, and copper is a cofactor for
lysyl oxidase, an enzyme required in forming collagen cross-links.
Vitamin K is a cofactor of -carboxylase, an enzyme necessary for the
-carboxylation of glutamic acid residues in proteins, including
osteocalcin, the principal noncollagenous protein of bone.
Major nutrient interactions
Putative adverse effects of high calcium. Soft tissue accumulation of calcium as a result of dietary intake should not be of concern in healthy children. Concern has been expressed over recommending high levels of calcium intake causing interference with absorption of other essential nutrients, including iron, magnesium, and zinc. Therefore, upper safe levels of calcium intake were set at 2.5 g/day for all ages by the National Academy of Science. Single meal studies show reduced iron absorption in the presence of calcium up to loads of 300 mg (26). However, chronic feeding of high calcium has not been demonstrated to have adverse effects on iron status, presumably due to up-regulation of iron absorption. Girls of mean age 10.8 yr given 1000 mg calcium daily for 4 yr had changes in serum ferritin and red blood cell indexes similar to those in girls assigned to the placebo group (27). In this same population, zinc retention was not adversely affected by calcium supplementation (28). High calcium intakes have not been associated with altered magnesium retention or body pool sizes. In a cross-over design in adolescent girls, aged 1214 yr, magnesium metabolism was studied using metabolic balance and magnesium stable isotopic tracers for kinetic analysis (29). Parameters of magnesium metabolism were not different for calcium intakes of 1800 vs. 800 mg/day.
Dietary factors impacting calcium. In contrast to the lack of evidence of an adverse effect of chronically consuming high calcium levels, there is some evidence that several dietary factors inhibit calcium utilization. However, with few exceptions, these dietary relationships have not been studied in adolescents.
Some dietary constituents inhibit calcium absorption, and others
increase calcium excretion. Oxalic acid, which is found in spinach,
beans, sweet potatoes, and rhubarb, is a potent inhibitor of calcium
absorption. Phytic acid, the storage form of phosphorus in seeds, is a
modest inhibitor of calcium absorption. Sodium chloride is a major
determinant of urinary calcium excretion. For every additional gram of
salt consumed, an extra 26 mg Ca are excreted in the urine of
adolescent girls (30). For comparison, for every additional gram of
protein consumed, only 1 mg additional Ca is excreted in the urine.
Thus, calcium requirements for an individual depend on intakes of other
dietary constituents. The calcium requirements given in Table 1 were
developed using data on subjects consuming typical western diets.
Discussion and Summary
Many gaps exist in our understanding of the role of nutrition and other lifestyle variables during the growing years on lifelong bone health. The difficulty in measuring bone changes during growth is a major barrier to our understanding of environmental effects. Changes in bone size are difficult to monitor with two-dimensional images such as those obtained by bone densitometry. Apparent bone density changes may reflect changes in bone size rather than bone density. Volumetric bone density measures, if they can be made at appropriate sites with acceptable radiation exposure, will provide the next generation of outcome measures for assessing interventions. Quantitative ultrasound measures are being explored for use in children as a potential measure of bone quality, but the interpretation of the measures is as of yet unknown.
It is difficult to follow the effects of adolescent nutrition into adulthood. Most work in children is observational in nature, rather than involving a rigorous intervention. Published randomized placebo-controlled trials are relatively short in duration. It is unknown whether short term increases in the rate of bone accretion due to dietary interventions such as calcium result in greater increases in peak bone mass or if the peak is merely achieved earlier. During the pubertal growth spurt, hormonal changes may overwhelm any lifestyle interventions, as seems to be the situation during menopause, a much more studied phenomenon.
We know so little about interactions of nutrients with other parameters during growth. Important areas for further research include nutrient-gene interactions, nutrient-exercise interactions, and nutrient-nutrient interactions. How do genetics determine who will respond to dietary supplementation? Will exercise increase bone size without sufficient calcium? Does an active lifestyle modify nutrient requirements?
When considering the role of childhood and adolescent nutrition in lifelong bone health, we may not know precise requirements, but the risks lie with insufficient nutrient intakes, with the exception of energy. Therefore, clinicians should take an active role in assessing the diets of all patients, but especially for adolescents whose skeletal demands are so great. Research on behavior modification to achieve appropriate lifestyle choices in this age group is urgent. Strategies to instill healthy habits before this age are equally important, and they may be more effective. The benefits are potentially great. Variations in calcium nutrition in early life may account for as much as 50% of the difference in hip fracture rates in postmenopausal years (31). Taking calcium-rich products with every meal goes a long way toward ensuring that requirements are met for many bone-related nutrients, including calcium, magnesium, phosphorus, and vitamin D. However, we need more alternative strategies for many individuals, including fortified foods and palatable supplements. Evidence is compelling that inadequate nutrition during puberty results in suboptimal peak bone mass, which, in turn, increases the risk of fracture in childhood and later in life.
Received January 7, 1999.
Accepted January 29, 1999.
References