点击显示 收起
Ethel Austin Martin Program in Human Nutrition, Box 2204, South Dakota State University, Brookings, SD 57007, E-mail:speckerb{at}ur.sdstate.edu
In preparing this manuscript I took the privilege of using the minutes of the September 1998 meeting of the Dietary Guidelines Advisory Committee, which was held in Washington, DC (1). During that meeting, Cutberto Garza, in his introductory remarks to the committee, mentioned a list of 5 C's: confusion, change, complexity, controversy, and challenge. These same words often came to mind when I was synthesizing the information for this article. The five C's relative to calcium as given below provide a format for the discussion of why there should not be a specific guideline for calcium within the 2000 edition of the dietary guidelines:
1) confusion as to why there should be a separate guideline for calcium,
2) change in the recommended intakes within the guidelines,
3) complexity of the role of diet in explaining bone mass and fracture risk,
4) controversy surrounding fracture data, and
5) challenges in maximizing bone mass and preventing osteoporotic fractures.
Confusion as to why there should be a separate guideline for calcium
There is some confusion as to why there is a discussion about whether there should be a guideline specific for calcium because the dietary guidelines were not meant to be nutrient specific. The dietary guidelines describe food choices that will help the American people meet the recommended dietary allowances (RDAs) (2). The first of the 7 dietary guidelines in the 1995 edition is "Eat a variety of foods" (3). It is within this guideline that calcium appears. Within the main text of this one guideline, the words milk, calcium, vitamin D, and cheese are mentioned 18 times. In addition, a text box highlights the recommended serving sizes of various dairy products and an additional box summarizes good sources of calcium other than dairy products. The food guide pyramid is also depicted within this guideline.
Change in the recommended intakes within the guidelines
The food guide pyramid, developed by the US Department of Agriculture (USDA), is presented within the dietary guidelines. The food guide pyramid is an educational tool that recommends the kinds and amounts of foods to eat each day. The food groups and numbers of servings depicted in the food guide pyramid are based on recommendations contained within the RDAs, or dietary reference intakes (DRIs), and the dietary guidelines as well as on national survey data of food consumption patterns. To satisfy the 1989 RDA for calcium and to emphasize the need to eat a variety of foods, the current food guide pyramid contains dairy products as 1 of the 5 major food groups, with a recommendation to consume 2–3 servings of dairy products/d.
The change in calcium recommendations between the 1989 RDA and the 1997 adequate intake (AI) (4) is shown in Figure 1. Across all age groups, the maximum recommended intake increased by only 100 mg/d, from 1200 mg/d in the 1989 RDA to 1300 mg/d in the 1997 AI. Because the food guide pyramid is currently designed for persons aged >2 y, the current recommended number of servings in each of the 5 major food groups, including the dairy group, should be sufficient to cover all age groups. If it is assumed that most of the calcium in the diet is obtained from dairy products, the calcium intake of a diet that includes 2–3 servings of dairy products/d could range from 570 to 1350 mg/d depending on the dairy product consumed [from a low of 285 mg Ca in a cup (240 mL) of buttermilk to a high of 450 mg in a serving of plain, nonfat yogurt]. This amount would increase by an additional 94 mg to 180 mg/d if 2 of the 3–5 servings of vegetables included a high-calcium source such as broccoli (47 mg Ca in 92 g or one-half cup) or kale (90 mg Ca in 65 g or one-half cup). Adding one more serving of dairy products per day, or encouraging the consumption of vegetables containing high amounts of calcium, should cover the additional 100-mg/d maximum increase in the recommended calcium intake. However, whether the number of servings within the food guide pyramid needs to be changed to satisfy the new AI for calcium will need to be determined by the USDA. It is not within the authority of the Dietary Guidelines Advisory Committee to change the food guide pyramid, although it is likely that they can make recommendations to the USDA.
Complexity of the role of diet in explaining bone mass and fracture risk
There is agreement among the scientific community that calcium is a critical nutrient for bone health. However, the amount of calcium needed to optimize bone health and the role of calcium in bone mass and fracture risk is controversial. This controversy is likely due in part to the complexity of how various factors interact with each other in their effect on bone mass.
There are numerous examples of interactions among factors that may affect bone, most of which are just now being realized. Differences in loss of bone density at the femoral neck in postmenopausal women were shown to be associated with the vitamin D receptor genotype (5). Calcium intake, however, appears to modify the relation between bone loss and vitamin D receptor genotype so that a benefit of calcium supplementation is observed only in women with the BB genotype and not in women with the bb or Bb genotype (Figure 2).
A simple hypothetical example of how these interactions can affect the interpretation of study results and conclusions related to the role of various factors in determining an outcome is illustrated in Figure 3. This example is based on an animal model described by Hogben in 1933 and is often used to illustrate a gene-environment interaction (6). For a chicken to have yellow shanks, both a yellow shank gene and the consumption of yellow corn is required. If one investigator conducts a study in a population of chickens, some of which have the yellow shank gene and some of which do not, and all the chickens are fed yellow corn, the investigator will conclude that the presence of yellow shanks is 100% attributed to genetics. However, if another investigator studies chickens that only have the yellow shank gene and half of the chickens are fed yellow corn and half are fed white corn, this investigator will conclude that the presence of yellow shanks is 100% attributed to diet. This interaction between genetics and diet leads the 2 investigators to 2 very different conclusions. This is an extreme example of how the study population, study design, and interactions between 2 factors can have significant effects on the results of a study and the conclusions reached.
Other interactions within the context of bone density may also exist, including the possible interactions of physical activity and dietary calcium (7, 8), physical activity and estrogen (9), and dietary calcium and estrogen (10). The results of a meta-analysis indicate that there may be an interaction between physical activity and calcium intake on bone loss, as shown in Figure 4 (7). The data presented are the mean annualized rates of change in bone mineral density at different calcium intakes for groups of individuals assigned to either an exercise group or a control group. Each data point represents groups of individuals, with sample sizes ranging from 7 to 34 per group. As illustrated in this figure, there appears to be a benefit of exercise in preventing loss of bone density only at calcium intakes greater than 1000 mg/d. Another way of interpreting these data is that the benefit of a high calcium intake on preventing loss of bone density appears to occur only in the exercise groups.
If these interactions among individual factors are not considered in study designs or in the statistical analyses of data pertaining to bone density or fracture, estimates of the amount of variance attributed to these factors are not valid (11). Rather than being able to separate the amount of variation in bone mass within a population into individual factors whose variances sum to 100% as shown in Figure 5A (12), the relations are far more complex, with the individual factors interacting with each other in their influence on bone mass (Figure 5B). Therefore, part of the variation in bone mass within a population that is attributed to diet may be a function of how diet interacts with the specific genetic composition, physical activity patterns, or other characteristics specific to that population. The complexity of these interactions may be one of the reasons for inconsistent findings among various studies with regard to the role of diet in determining bone mass and fracture risk. The presence of interactions and lack of independence among the different factors may explain why some investigators reported that genetic effects may explain up to 75–80% of the variation in bone density (13, 14), whereas the same investigators reported that both physical activity (15) and diet (16) may each explain up to 40% of the variation in bone density. Eisman (17) recognized the possible influence of other factors on estimates of the variance attributed to genetic effects by stating that these estimates were made assuming that other known environmental and medical factors affecting bone density were controlled for.
The extent of the confusion, controversy, and complexity of the role of calcium intake in maximizing peak bone mass, preventing osteoporosis, and reducing fractures can be illustrated by the decision of the Food and Nutrition Board (FNB) of the National Academy of Sciences to take a new approach to the RDAs. Rather than recommending one level of intake, or the RDA, the FNB agreed that 4 different recommendations would exist in the form of the dietary recommended intakes, or DRIs (4). The new DRIs consist of an estimated average requirement (EAR), a recommended dietary allowance (RDA), an adequate intake (AI), and a tolerable upper intake limit (UL). The FNB decided to set AIs rather than EARs for calcium on the basis of the following factors (4): "(1) uncertainties in the methods inherent in the balance studies that form the basis of the maximal retention model, (2) the lack of concordance between observational and experimental data (mean calcium intakes in the United States and Canada are much lower than are experimentally-derived values required to achieve maximal calcium retention), and (3) the lack of longitudinal data that could be used to verify the association of the experimentally derived calcium intakes for maximal retention with the rate and extent of long-term bone loss and its clinical sequelae, such as fracture."
The interactions of various factors, both environmental and genetic, that affect bone mass are complex. These complex interactions likely explain much of the uncertainties in the estimates of calcium requirements.
Controversy surrounding fracture data
The role of high dietary calcium intake or consumption of dairy products in determining fracture risk is controversial. The results of retrospective studies are conflicting: some studies reported a protective effect (18, 19), whereas others reported no association between calcium intake and fractures (20, 21). Retrospective studies that collect dietary intake data after the fracture has occurred are subject to dietary recall bias (22), which often makes the results difficult to interpret. In addition, women who know they are at risk of fracture because of their family history may be more likely to consume more calcium than are women without a family history of fracture or osteoporosis. Prospective studies have the benefit of collecting dietary information before fractures occur, thereby reducing the possibility of recall bias.
Findings from the Nurses' Health Study have generated a controversy over the role of dairy consumption in the risk of fracture (23). This 12-y prospective study of >77000 women found no benefit of increased dairy consumption, or increased calcium consumption, on the risk of fracture. Although not emphasized by the authors, a slight trend of increasing hip fracture risk with increasing dairy consumption was observed. Although these findings are receiving a significant amount of attention in the public press (24), a paper released 1 y later by the same group of investigators is not being referred to in the discussions surrounding this controversy (25). The more recent report examined the role of the vitamin D receptor gene in predicting hip fracture risk in a subset of the women from the Nurses' Health Study for whom blood samples were available. In this report, hip fracture risk was higher in women with the BB genotype than in women with the bb genotype, with the risk being even greater for those women who had low calcium intakes. Because genotypes were available for only a subset of the women in the original report, it is difficult to interpret how these findings may have influenced the original results. Thus, at this time it is difficult to conclude whether high calcium intake is associated with fracture risk.
Two randomized trials of increased calcium and vitamin D intakes reported a reduction in the incidence of nonvertebral fractures among elderly persons whose mean calcium intakes were low at baseline (<800 mg/d) (26, 27). It is not clear whether calcium alone would have led to the reduction in fracture risk or whether a similar effect would be observed in younger persons. However, additional evidence should be available soon. Ongoing studies such as the Women's Health Initiative should provide evidence as to whether a high calcium intake with supplemental vitamin D prevents fracture. The Women's Health Initiative is a large, randomized trial of 40000 women, half of whom are receiving calcium and vitamin D and half of whom are receiving placebo. One of the main outcome variables in this large, clinical trial sponsored by the National Institutes of Health is fracture.
Additional studies sponsored by the National Institutes of Health involving randomized trials of increased calcium intake at various ages are also ongoing in an effort to determine the role and importance of increasing calcium intake in an effort to improve bone health. If there were already overwhelming evidence for a role of high calcium intakes in preventing fractures, such trials would be considered unethical and would be stopped.
Challenges in maximizing bone mass and preventing osteoporotic fractures
There is agreement among the scientific community that calcium is a critical nutrient for bone health. The controversy is in the amount of calcium needed to optimize bone health and prevent fractures and the extent to which that amount varies in the population. It is likely that the complexity among factors that affect bone have led to some of the inconsistent findings among studies and to the confusion and controversy surrounding the individual effects of these factors on bone. Determining whether and how these factors interact with each other has just begun, and defining such interactions represents a major challenge in bone research. The prevention of osteoporotic fractures is far more challenging than the simple addition of higher amounts of calcium to the diet. By including a dedicated dietary guideline focusing specifically on calcium, are we leading people to believe that they are doing enough about their health, rather than finding genuine solutions?
Summary
The following are the main reasons for not including a specific dietary guideline for calcium:
1) The dietary guidelines were not originally meant to be nutrient-specific recommendations, but rather general recommendations on food choices.
2) The current dietary guidelines already contain numerous recommendations regarding adequate calcium intake.
3) The relation between calcium intake and bone health is complex and poorly understood.
4) Including a specific dietary guideline for calcium may prematurely lead people to believe they are doing enough about their health.
FIGURE 1. . Recommended calcium intakes by age according to the 1989 recommended dietary allowance [- - - (2)] and the 1997 adequate intake [— (4)].
FIGURE 2. . Adjusted rates of change in femoral neck bone mineral density (BMD) by vitamin D receptor genotype and calcium supplement status (, placebo group; , calcium-supplemented group). Each mean was adjusted for baseline BMD, log-years postmenopausal, dietary calcium intake, vitamin D intake, and smoking status. The interaction between calcium intake and vitamin D receptor genotype on femoral neck BMD was significant (P = 0.09). The BB placebo group was significantly different from the bb placebo group, P < 0.01. Data are from reference 5 with permission.
FIGURE 3. . A simple, hypothetical example of how different investigators can come to different conclusions with the same basic truth. In this example, chicks must both have the YY gene and be fed yellow corn to have yellow shanks.
FIGURE 4. . Results of a meta-analysis showing the interaction between calcium intake and physical activity on mean percentage change in spine bone mineral density (BMD) in the exercise groups () and control groups (). Data are from reference 7 with permission.
FIGURE 5. . It is often assumed that the variation in bone mass within a population can be attributed to individual factors (A). However, the variation in bone mass within the population is likely to be a function of how the different factors interact with each other (B). If these interactions occur, estimates of the amount of variance attributed to individual factors is not valid.
REFERENCES