Bone is a living tissue and, as such, requires all essential nutrients for growth and maintenance. A host of micronutrients are required to create and maintain the elaborate structure of bone mineral crystals deposited in connective tissue fibrils. Because bone turns over through continuous modeling and remodeling processes to shape bones during growth, to repair microarchitecture damage, and to respond to an ever-changing environment as we age, there are obligatory daily losses of bone constituents. As essential micronutrients, by definition, bone cannot be synthesized by the body (with the exception of vitamin D through Ultraviolet B exposure, they must be supplied through diet). Unfortunately, the Dietary Guidelines for Americans for several iterations1-3 have identified several of the most important nutrients to bone as shortfall nutrients, including calcium, vitamin D, and magnesium. Low intakes of calcium trigger the parathyroid hormone-vitamin D axis to increase calcium absorption at the gut and to decrease excretion at the kidney. Unfortunately, these processes are limited and cannot maintain serum calcium within the highly regulated narrow range essential for vital life functions such as neurotransmitter release and muscle contraction for prolonged periods of inadequate calcium from the diet. The large reservoir of calcium in this situation is the bone, and persistent bone resorption (elevated in menopause in a state of estrogen deficiency) leads to bone loss and fracture risk.
The critical, functional role of dietary calcium is not uncertain. The current debate then is how much do we need. At what stages in life? What form should we consume to build and maintain bone? How much is too much to cause harm? The evidence base is limited. There are no meta-analyses of randomized clinical trials (RCTs) for milk intake and fracture because controlling milk intake for sufficiently long periods in a large-enough population to measure fracture as an outcome is unfeasible. Randomized controlled trials of diet and chronic disease outcomes are scarce, in part because of the lack of financial incentive with low profit margins and difficulty in claiming intellectual property for foods.
Our highest level of evidence for the relationship of calcium and bone comes from calcium supplement studies. Unfortunately, there are many limitations with the evidence. Almost all RCTs have used calcium and vitamin D in combination. As with many nutritional supplement studies, the literature base suffers from poor compliance, design flaws, and weak methodology. This leads to equivocal and conflicting results.
Weaknesses in study design and methodology for assessing dietary calcium and bone outcomes include lack of consideration of baseline status, poor ability to assess nutrition intake generally, and failure to consider subpopulation effects. Calcium is a threshold nutrient, and intakes above the amount needed for maximal retention are excreted. Threshold intakes are known for several age-sex-race groups.4 We have no good measure of determining calcium status currently, so only controlled feeding studies can be used to determine threshold intakes. Moreover, for RCTs of supplements on bone measures or fracture, even if compliance with the supplement is adequately monitored, the ability to know calcium intakes of the background self-selected diet is weak. To illustrate the effect of baseline status and compliance in the largest RCT of calcium (and vitamin D) on disease outcomes and mortality, examine the Figure. Panel A shows that the effect of calcium supplementation on hip fracture and all the other outcomes except heart disease in more than 68 000 postmenopausal women was not significant.5 When the noncompliant (<80% compliant) women were removed and when women taking their own calcium and vitamin D supplements (presumably those above the threshold) were removed, the hazard ratio for hip fracture was an impressive 0.2.6
Some are concerned about calcification of the arteries from calcium supplements from selected prospective studies, retrospective analyses of RCTs, and observational studies. Notice in the Figure that there was no risk for heart disease by either analysis in the largest RCT. The assumption is that an acute elevation in serum calcium can translate to calcification of soft tissue. Yet, 1 g of calcium as calcium citrate, which produced a rise in PTH, resulted in a positive influence on cardiovascular function.7
Only in an animal model appropriate for studying calcification by a mechanism comparable to humans for sufficiently long periods to observe calcification can diets be controlled and soft tissue be invasively examined. We studied high-calcium intakes up to the equivalent of the upper level from nonfat dry milk or calcium carbonate in a pig model of metabolic syndrome fed with an atherogenic diet to induce atherosclerosis.8 There was no increased uptake because of high-calcium intake into coronary arteries of a calcium tracer, 41Ca, which can be measured at 10-18M concentrations by accelerator mass spectrometry compared to the control diet. Moreover, high-calcium intakes had also no effect of on vascular function or calcification of arteries as measured by histology or various imaging techniques.
A recent position paper by the National Osteoporosis Foundation and the American Society for Preventative Cardiology concluded that there was moderate-quality evidence that calcium has no relationship to risk of cardiovascular disease, mortality, or all-cause mortality at this time.9 To ensure an adequate intake of calcium without risking harm, it is prudent to follow the Dietary Guidelines, which recommends 3 servings of dairy per day. For every serving missed, take 300 mg calcium as a supplement or from fortified food.
REFERENCES