点击显示 收起
Creighton University, 601 North 30th Street, Suite 4841, Omaha, NE 68131, E-mail:rheaney{at}creighton.edu
Introduction
In theory, dietary guidelines are instruments of national nutritional policy rather than statements of nutrient requirements (1). Dietary guidelines are about diets rather than nutrients. There is, therefore, a built-in resistance to incorporating nutrient-specific issues into the guidelines. However, that line in the sand has already been crossed for 4 nutrients [alcohol, fat (and cholesterol), sugar, and sodium] and there are compelling, diet-specific reasons for adding calcium to that list or, perhaps, for substituting calcium for 1 of the 4.
I will summarize here, but not attempt to reargue, the importance of an adequate calcium intake. This point has been satisfactorily dealt with in several nutritional policy-related official statements (2–5) and recent reviews (6) and must be considered firmly established. There are still dissenters, of course, but their stance seems based either on a highly selective reading of the evidence or on premises or preferences that have little or no credible evidential base.
Multisystem involvement of calcium
Adequate calcium intakes have been convincingly shown to protect the skeleton (5), to lower blood pressure (7–9), to reduce the risk of colon cancer (10, 11), to lessen the symptoms of premenstrual syndrome (12), and to reduce the risk of renal stone formation (13, 14). The evidence is strong for both osteoporosis and hypertensive disorders. In the former, the size of the effect is large, whereas with the latter, the effect at a general population level is smaller (8). The evidence is persuasive for the other disorders as well, but less massive than for osteoporosis and hypertension, and the size of the effect at the population level is still uncertain. For all the disorders concerned, optimum benefit occurs at intakes above both prevailing intakes and the dietary reference intakes of virtually every industrialized nation. These seemingly diverse effects of calcium have a largely dietary rather than a biochemical basis (discussed below), which is in itself a reason for a dietary guideline for calcium.
It is widely recognized that the calcium ion plays an essential role as an intracellular second messenger and that it mediates processes as diverse as muscle contraction, interneuronal synaptic signal transmission, glandular secretion, cell division, and blood clotting. These biochemical functions of calcium are exceedingly well protected, first by intracellular calcium stores and by the sheer size of the extracellular nutrient reserve (the skeleton), and second by an elaborate endocrine control system (the parathyroid hormone–vitamin D axis and calcitonin). As a consequence of these protections, nutritional calcium deficiency virtually never compromises, or even threatens, the essential biochemical functions of the mineral. Calcium is unique among the nutrients in that deficiency relates not to impairment of its biochemical roles, but instead to 3 groups of effects that are a consequence of low intake: 1) reduction in the size of the calcium reserve, 2) reduction in the quantity of unabsorbed calcium in food residues, and 3) collateral effects on the other body systems of the regulatory apparatus that protects the organism from hypocalcemia.
The skeletal effect of dietary calcium is straightforward. Skeletal mass (ie, the size of the calcium nutrient reserve) is a direct function of intake up to age-specific thresholds, both during and after growth. It is now clear that contemporary calcium intakes support neither full realization of the genetic potential for skeletal mass nor its maintenance. Roughly 50 studies of investigator-controlled increases in calcium intake have been published, most of which were randomized controlled trials published since 1990 (6). All but 2 studies showed greater skeletal mass gain during growth, reduced bone loss with age, or reduced osteoporotic fracture risk. The sole exceptions among these studies were a supplementation trial in men in which the calcium intake of the control group was already high (nearly 1200 mg/d) (15) and a study confined to early postmenopausal women in whom bone loss is predominantly due to estrogen deficiency (16).
Complementing this primary evidence are 80 observational studies testing the association of calcium intake with bone mass, bone loss, or fracture (6). It was shown elsewhere (17) that such observational studies are inherently weak, not only for the generally recognized reason that uncontrolled or unrecognized factors may produce or obscure associations between the variables of interest, but because the principal variable in this case, lifetime calcium intake, cannot be measured directly and must be estimated by dietary recall methods. The errors of such estimates have been abundantly documented (18, 19). Nevertheless, more than three-fourths of those studies showed a calcium benefit. In the face of the inaccuracies of the method, the fact that the results of most of these observational studies are positive emphasizes the strength of the association.
Most of the investigator-controlled studies used supplements as the source of calcium, but at least 8 used dairy sources; the results of all these studies were positive. Additionally, essentially all the observational studies involved natural food sources (principally dairy products) and the food sources produced effects comparable with those of supplements. Hence, no further distinction needs to be made between dietary and supplemental sources of calcium. Skeletal effects, at least, depend mainly on total calcium intake.
The effect of calcium intake on colon cancer risk has a different, but equally straightforward basis. In individuals with hereditary or acquired oncogenic factors predisposing to colon cancer, constituents of the chyme residue (ie, unabsorbed fatty acids and bile acids) act as cancer promoters by stimulating colonic mucosal proliferation and mitotic activity. Dietary calcium, precisely because it is poorly absorbed, is also a part of the food residue that reaches the colon. By forming calcium soaps with the fatty acids and salts with the bile acids, dietary calcium renders the fatty acids and bile acids inert; ie, calcium functions as an antipromoter. Calcium's ability to do this depends on the relative quantities of the reactants in the food residue. With high-calcium diets there is an excess of calcium in the chyme and the promoters are fully complexed; with low-calcium diets the opposite is the case. (Incidentally, this imbalance with low calcium intakes is made worse by another dietary feature, the relatively high fat content of modern diets, which leads to a higher concentration of cancer promoters in the residue, ie, more unabsorbed fatty acids and bile acids.)
The mechanisms for protection in the hypertensive disorders and in premenstrual syndrome are less well understood, but appear to be related to the chronically high blood concentrations of parathyroid hormone, 1,25-dihydroxyvitamin D, or both in persons with low calcium intakes. These hormones, which evolved to sustain extracellular fluid Ca2+ concentrations during periods of low environmental calcium availability, also increase cytosolic Ca2+ concentrations; in sensitive tissues such as vascular smooth muscle, this effect thereby increases vascular tone.
Under primitive conditions, with a normally high calcium intake, parathyroid hormone secretion would have been episodic and confined largely to periods of fasting or famine. Under modern dietary conditions, however, parathyroid hormone secretion is continuously high. Presumably, a sensitive subset of the population with less redundancy in their control systems develops autonomic dysregulation as a consequence of this sustained exposure, much as fava beans unmask glucose-6-phosphate 1-dehydrogenase deficiency in certain persons of Mediterranean ancestry. Interestingly, this is a dietary, or foods issue, rather than just a calcium issue, because diets high in potassium and magnesium, among other nutrients, appear to potentiate the calcium effect. In the Dietary Approaches to Stop Hypertension (DASH) Study (8), the blood pressure benefit produced by the addition of nonfat dairy products was approximately twice as great as was reported for calcium supplements alone.
Finally, protection from kidney stones has a basis similar to protection from colon cancer. Unabsorbed dietary calcium forms complexes not only with fatty acids, but also with dietary oxalate, thereby preventing its absorption. Although oxalate of dietary origin normally accounts for less than one-fourth of the renal oxalate burden, any reduction in urinary oxalate will lower the risk of calcium stone formation. Furthermore, because urinary oxalate is a stronger risk factor for kidney stones than is urinary calcium, reduction in urinary oxalate excretion produces a net reduction in renal stone risk. (This effect of oral calcium has long been recognized and exploited in the management of the syndrome of intestinal hyperoxalosis, in which the intestinal hyperproduction of oxalate leads to massive kidney calcification and for which the standard therapy is large oral doses of calcium carbonate.)
It may be helpful to point out that all these disorders are multifactorial and that inadequate calcium intakes explain only a part of the respective problems. If there is any residual significant uncertainty in the scientific community about the importance of a high calcium intake, it may be precisely because of the multifactorial character of these disorders. One's individual scientific experience with osteoporosis or hypertension, for example, may be so dominated by the effects of other equally real factors (eg, female hormones, fall patterns, or ethnicity in the case of osteoporosis) that calcium effects are pushed into the background. This is one of the reasons randomized controlled trials are so crucial. In addition to the strong causal inference they permit, they effectively factor out, for investigational purposes, the other important variables and thereby serve to establish the reality of the calcium effect, not as the sole cause of the disorders concerned, but as one of several.
Why a calcium guideline?
Contemporary diets typically contain less calcium than is needed to ensure the foregoing benefits. Moreover, the disorders concerned relate to several components of contemporary diets, not just to calcium. Because these problems transcend single-nutrient issues, they are fundamentally dietary problems, not nutrient problems. A calcium guideline is needed to round out the current dietary guidelines.
At a total diet level, it is worth recalling that the primitive human diet, the one that prevailed during the millenniums of hominid evolution and to which our physiologies were adapted, had a high calcium density, estimated to be 2.9–3.3 mg Ca/kJ (70–80 mg Ca/100 kcal) from vegetable sources alone and substantially higher if, as was often the case, the diet included insect grubs or the bones of small prey or fish (20). In contrast with foods accessible to industrialized populations, calcium was widely distributed in the plant foods available to hominids and other primates (eg, roots, tubers, and greens). In foods such as coccinia root, for example, the calcium density is >63 mg/kJ (>1500 mg/100 kcal). The only contemporary food that approaches that density is Chinese cabbage at 31 mg/kJ (750 mg/100 kcal). Skim milk, the dairy food with the highest calcium density, has a density of 15 mg/kJ (350 mg/100 kcal). Wild chimpanzees, our closest primate relative, have a diet with a calcium density of 3.3–4.2 mg/kJ (80–100 mg Ca/100 kcal) and the diets we feed our laboratory and household companion animals have higher calcium densities still, ranging from 11.1 to 18.9 mg/kJ (266 to 452 mg/100 kcal). These comparatively high densities apply equally to the diets of herbivores, carnivores, and omnivores. Laboratory feed for primates has an intermediate density: 12.6 mg/kJ (300 mg/100 kcal). For comparison, the median calcium density of the diet of women in the third National Health and Nutrition Examination Survey was only 1.5 mg/kJ (36 mg/100 kcal) (21), and the 1997 adequate intake recommendations compute to a total dietary calcium density of 2.1 mg/kJ (50 mg/100 kcal) (5). There is some evidence that the calcium density of laboratory animal feeds may be higher than the minimum needed for full skeletal development, perhaps by as much as a factor of 2. However, even if one discounts these animal diet densities by 50%, the resulting values of 556–946 mg/kJ (133–226 mg Ca/100 kcal) are still much higher than those of the diets humans consume today.
The reason for the change from the high calcium densities under primitive conditions to the low contemporary values has a dietary origin and is a consequence of the agricultural revolution, which was based in the domestication and cultivation of seed plants such as cereals and legumes, neither of which figured significantly in the diets of evolving hominids (20). Seeds are life-support packages for the plant embryos they contain and provide essential nutrients such as phosphorus and the B vitamins until the embryo develops its own synthetic apparatus and root system. Calcium, the fifth most abundant element in the biosphere, would have been present in most soils and fresh waters. There was, therefore, no evolutionary need to add calcium to most seeds. Hence, cereals, legumes, and fruit tend to be low in calcium and diets based on them are low as well. However, for the first 7000 y after the shift to farming, human grain-processing practices would have added substantial amounts of calcium to flour because all milling was done with soft stones, such as limestone, which abraded in use and contributed calcium to the flour (22). Only after the Iron Age (and the corresponding ability to make harder millstones), which began barely 3000 y ago in Eurasia, did dietary calcium density fall toward values inherent in the cereals themselves.
Further dietary and lifestyle changes contributing to low contemporary calcium intakes (but of much more recent origin) are the decreased energy expenditure and energy intake associated with post World War II expansion in the use of private automobiles and other labor-saving devices, together with the development and aggressive marketing of low-nutrient-density beverages and snack or convenience foods. Thus, we do less work and eat less food than our grandparents did. Eating less makes it harder to meet our full nutrient requirements and the problem is made worse by the fact that we increasingly fill up on nutrient-poor but tasty foods.
It might be argued that the public is confused on this issue, but I doubt that. The media batten on controversy and there is no dearth of coverage of fringe positions that aggressively promote or decry certain calcium sources, but there is general acceptance of the need for calcium. The Food and Drug Administration has allowed a health claim for calcium-rich foods for the past 7–8 y; Jane Brody, a major opinion-shaper among nutrition journalists, characterized calcium as a "superstar mineral," devoting 2 issues of Science Times to the topic in 1998 (23); and Newsweek magazine in its millennial medicine issue summarized calcium's many benefits under the headline "The Little Mineral That Could" (24). Over the past 17 y of my own extensive interaction with science writers, the focus of their questions to me has shifted from whether we need calcium to how we can best get all we need.. Thus, confusion is not a reason to postpone a calcium guideline. In fact, in the face of the public information now available, not having a calcium guideline could well be a source of confusion.
Moreover, adding a calcium guideline would complement 2 of the other guidelines, namely, the recommendations to consume a diet low in fat and a diet moderate in sugars. Failure on both of these counts either contributes to the problems of low calcium intake (eg, colon cancer) or itself further lowers calcium intake. On the other hand, "eating a diet rich in calcium" (or however a calcium guideline might be expressed) would complement the other guidelines and help improve the total diet. This is because most readily available high-calcium sources (dairy foods and vegetable greens) are either naturally low in fat and sugars or are widely available in low-fat varieties. At the same time, both are high in many other essential nutrients, thereby substantially enhancing overall diet quality (25, 26). Finally, adding a calcium guideline [or possibly substituting calcium for the problematic sodium guideline (27)] would help to encourage a rational calcium fortification policy, in accord with the Surgeon General's 1988 report on nutrition and health (28).
Conclusion
In the final analysis, the desirability of a calcium guideline depends on pragmatic considerations: Will it help Americans consume a better overall diet? Will it help policymakers improve the nutrition of all Americans? My assessment of the situation leads me to answer "yes" to these questions. Whatever the final decision, we cannot lose sight of the need to improve calcium nutrition for the majority of Americans whose low calcium intakes place them at increased risk of osteoporosis, colon cancer, hypertension, and renolithiasis.
Note added in proof
Recent reports have both added new disorders to the list of conditions associated with low calcium intake and shed new light on the general mechanism behind several of the known effects. Zemel et al (29), in an analysis of the third National Health and Nutrition Examination Survey database, showed that the risk of being obese increases 6-fold as one proceeds from the highest to the lowest quartile of calcium intake. Thys-Jacobs et al (30) recently reported an unprecedented reversal by calcium and vitamin D of polycystic ovary syndrome, a leading cause of infertility in women of childbearing years and a disorder not heretofore linked with the calcium economy.
In both cell culture and transgenic mouse model systems, Zemel et al (29) showed that the high serum 1,25-dihydroxyvitamin D concentrations evoked by low calcium intake increase cytosolic free calcium ion concentrations in many tissues, and that, in the adipocyte, this change switches the cell from lipolysis to lipogenesis. In mice overexpressing the agouti gene, low calcium intake lowers core body temperature and increases body fat. This seemingly paradoxical effect of low calcium intakes on cytosolic [Ca2+] was previously shown for platelets in patients with hypertension (31), as well as for smooth muscle cells. Presumably, it is partly responsible for increased vascular tone and thus contributes to hypertension. Thys-Jacobs et al (30) also explicitly propose that it is the effect of cytosolic [Ca2+] on oocyte maturation that is the trigger for polycystic ovary syndrome in otherwise sensitive individuals.
Obesity is the most common dietary disorder in the United States today. To the extent that low calcium intakes contribute to the population burden of this disorder, a guideline for a high calcium diet make ever greater sense.
REFERENCES