23a.1 Calcium

Calcium is the most abundant mineral in the body, accounting for 10–20g per kg body weight in adults. Bone is the main reservoir for calcium. The remain­der is located in body tissues (10g), blood, and the extra­cellular fluids (900mg). In con­trast, a greater pro­portion of the distribution of phosphorus and magnesium, the other two bone minerals, is located in soft tissue (Table 23a.1).

Table 23a.1: Calcium, magnesium and phosphorus distribution in an adult weighing 70 kg

Element	Mass (g)	Percentage in the skeleton	Percentage in tissue
Calcium (Ca)	1200	99	1
Magnesium (Mg)	24	55	40
Phosphorus (P)	900	80	12

All of these three minerals have two major roles —  as struc­tural components in bone and regulatory agents in body fluids.

23a.2 Functions of calcium

Calcium has a struc­tural role in bones and teeth, where 99% of the body's calcium exists primarily as calcium hydroxyapatite [Ca₁₀(PO₄)₆(OH)₂]. Bone is a dynamic tissue that undergoes remodeling throughout life except for the teeth. Bone formation is under the con­trol of osteoblasts, the bone forming cells, whereas bone resorption is initiated by osteoclasts, the bone resorbing cells. During growth, bone formation exceeds resorption as bone modeling is predominant. In later life when only bone remodeling takes place, bone resorption begins to occur at a faster rate than bone formation and this results in age-related bone loss (Siddiqui and Partridget, 2016).

The remain­ing 1% of total body calcium is in the soft tissues where it has a regulatory role in several metabolic pro­cesses, including enzyme activation, vascular con­traction and vasodilation, muscle con­tractibility, nerve transmission, hormone function, and membrane transport. When the levels of absorbed calcium from the diet are insufficient to balance the obligatory fecal and urinary losses, calcium is drawn from the bone and blood calcium levels are main­tained within narrow limits. This homeostatic regulation of the calcium level in the blood is achieved by the interaction of the parathyroid hormone (PTH), 1,25‑dihydroxy­vitamin D (1,25(OH)₂D), and calcitonin, the three major calcium-regulating hormones (Peacock, 2010). Phosphorus, pro­tein, estrogen, and other hormones, including glucocorticoids, thyroid hormones, growth hormone, and insulin also have a role (Siddiqui and Partridget, 2016).

23a.3 Absorption and metabolism of calcium

Calcium balance refers to the state of the calcium body stores (in bone) which are a function of dietary intake, intestinal absorption, renal excretion, and bone remodeling. A major con­trol of calcium balance is the amount of calcium absorbed. Calcium is absorbed across the intestinal mucosa in two ways: active transport and passive para-cellular diffusion. Active absorption of calcium is influenced by calcium and vitamin D status of the individual, age, pregnancy, and lactation. Active absorption occurs when intakes of calcium are low relative to need, and is regulated primarily via 1,25(OH)₂D and its intestinal receptors. Passive absorption is more important at high intakes of calcium (Weaver and Peacock, 2019).

Efficiency of calcium absorption is highest when dietary calcium intakes are low or during periods of rapid growth in infancy, early childhood, and adolescence, and during pregnancy and lactation. Indeed, absorption is very high during infancy (about 60%) (Abrams et al., 1997) and about 34% during puberty (Abrams and Stuff, 1994), stabilizing at about 25–30% in young adults. Subsequently, the intestinal absorption of calcium decreases pro­gressively with age (Christakos et al., 2011).

Urinary calcium excretion is also regulated in response to need. Normally calcium re-absorption in the kidney nephrons is not 100%. When there is hypo­calcemia to stimulate PTH release, one action of PTH is to increase efficiency of calcium re-absorption thus reducing calcium losses in urine (Peacock, 2010). Several dietary variables affect both calcium absorption and retention and, hence, calcium status. Diets high in pro­tein increase the urinary excretion of calcium (hyper­calciuria), which, however, is compensated by increased calcium absorption (Mangano et al., 2014). Dietary sodium also influences urinary calcium losses, with high sodium intakes increasing urinary calcium losses (Bedford and Barr, 2011), and lowering bone density in women (Devine et al., 1995). Caffeine, like pro­tein and sodium, also increases urinary calcium loss but has only a small negative effect on calcium retention, which can be offset by increasing dietary calcium intakes (Barrett-Connor et al., 1994).

High phosphorus intakes reduce urinary calcium losses (i.e., have a hypo­calciuric effect), but because they increase losses of endogenous fecal calcium at the same time (Heaney and Recker, 1994), their net effect on calcium balance is still under debate. One might assume that there would be no effect except in con­ditions in which the coordination between calcium and phosphate homeostasis occurs through the actions of PTH, FGF‑23 and 1,25(OH)₂D. The disruption of this coordination, however, can occur with disease states such as chronic kidney disease (CKD) (Peacock, 2010). However, increasing dietary phosphorus through inorganic phosphate additives may have detrimental effects on bone and mineral metabolism in humans and animals (Vorland et al., 2017).

Phytates and oxalates can inhibit the absorption of calcium by forming insoluble calcium com­plexes in the gastrointestinal tract (Barger‑Lux et al., 1995). Hence, they may have a negative effect on calcium status, although their overall effect may be small unless dietary calcium intakes are low. Some other dietary com­ponents, including lactose and inulin, enhance passive calcium absorption by increasing calcium solubility in the ileum (Weaver et al., 2016).

Several clinical con­ditions are associated with disturbances in calcium absorption. Calcium malabsorption is associated with gastrointestinal diseases such as Crohn's and celiac disease, and intestinal resection or bypass. Intestinal absorption of calcium is also impaired in patients with renal failure, caused by the reduced synthesis of 1,25‑dihydroxy­vitamin D (Peacock, 2010). Excessive intestinal absorption of calcium and hyper­calcemia occur in sarcoidosis, when excessive calcitriol appears in serum, and also in vitamin D intoxication.

23a.4 Dietary sources and intakes of calcium

Calcium is not widely distributed in all food groups, as shown in Table 23a.2.

Table 23a.2 Calcium bioavailability chart showing calcium content, estimated absorption, and net amount absorbed.
* Theoretical amount of calcium absorbed by an adult. Source: Weaver CM, Plawecki KL (1994). Dietary calcium: adequacy of a vegetarian diet. American Journal of Clinical Nutrition. 59: 1238S-1241S.

Food	Serving size	Calcium Content (mg)	Estimated Absorp- tion %	Net Ca* Absor- bed (mg)
Cow milk (fluid)	250mL	310	32	100
White beans (cooked)	125mL	85	22	18
Red beans (cooked)	125mL	26	24	6
Whole wheat bread	35g slice	26	82	21
Broccoli (cooked)	125mL	33	20	20
Spinach (cooked)	125mL	129	5	7
Almonds (raw, roasted)	60mL	97	21	21
Tofu (set with CaSO4)	85g	171	31	53
Soy “milk” (unfortified)	125mL	5	31	2

Milk and milk pro­ducts (e.g., cheeses, yoghurt) are excellent sources of readily available (bio­available) calcium. Plant-based beverages made from soy, almonds, rice and oats are available as substitutes for milk, but must be fortified. The availability of the calcium from these beverages is unknown, except for fortified soymilk for which the calcium availability is similar to cow milk, but only if calcium carbonate is used as a fortificant (Weaver and Peacock, 2019). Unfortified plant-based beverages have little calcium. Canned fish with bones (e.g., salmon, sardines) are good sources of calcium. Leafy-green vegetables are also good sources, but absorption of calcium from some of these foods, especially spinach, may be low (Weaver and Plawecki, 1994). Ancient grains (e.g., teff, quinoa, amaranth) and nuts are good sources of calcium (Cormick and Belizan, 2019), but highly refined or modern cereal grains such as wheat and maize are not.

Preparation of foods may also be a factor influencing their calcium con­tent. For example, in Guatemala, when corn (maize) tortillas are prepared according to the Mayan tradition of cooking corn with limestone and leaving it to soak overnight in hot water, the calcium con­tent rises from about 10mg to about 200mg of calcium per 100g (Belizan and Villar, 1980). In Sub-Saharan Africa, dietary calcium intake may be increased by using powdered eggshell added to foods (Bartter et al., 2018).

Foods fortified with calcium are available in many countries. In the European Union, the following forms of calcium are authorized for addition to foods and for use in food supplements: carbonate, chloride, citrate malate, glucon­ate, glycerophosphate, lactate, hydroxide, oxide, acetate, L‑ascorbate, bisglycinate, salts of citric acid, pyruvate, salts of orthophosphoric acid, succinate, L‑lysinate, malate, L‑pidolate, L‑threonate, sulphate (EFSA, 2012). Most of these forms are allowable in other countries. The citrate malate calcium salt is an allowable fortificant for orange juice in the United States (ODS, 2020). The United Kingdom is the only country where calcium fortification is a mandatory addition to flour; 235–390mg of calcium carbonate are added per 100g of flour. This represents about 94–156mg of elemental calcium per 100g of flour (Cormick et al., 2020). Other countries permit calcium addition to certain foods, often pro­prietary packaged foods, a practice called discretionary fortification.

Supplemental sources of calcium are widely available in many dosages and in a variety of chemical forms in many countries. These pills, capsules and liquids are usually regulated by the government. For example, in the United States, the following forms of calcium are allowable as supplements: carbonate, phosphate, citrate, glucon­ate, and lactate (Weaver and Peacock, 2019; ODS, 2020). Some supplements that are derived from oyster shell, algae, and in chewable drugs labeled as antacid such as TUMS^TM, are mainly in the form of calcium carbonate, the form of calcium naturally occurring in rocks such as chalk and limestone. This form of calcium is generally the cheapest to buy and has the highest amount of calcium by weight (40%). However, calcium carbonate is poorly absorbed unless stomach acid is available for dissociation. Also, calcium carbonate is the form of calcium that causes the most side-effects such as bloating and con­stipation (ODS, 2020).

Dietary calcium intakes vary across countries, ranging from 175–1233mg/d (Balk et al., 2017). Countries in Asia have average dietary calcium intakes less than 500mg/d while in countries in Africa and South America, calcium intakes vary from 400–700mg/d. Only Northern European countries have a national calcium intake greater than 1000mg/d, reflecting a high dairy con­sumption. The average calcium intake is generally lower in women than men, although across the 74 countries included in an analysis by the International Osteoporosis Foundation (IOF, 2020), no clear patterns regarding differences by age, sex, or socioeconomic status emerged.

23a.5 Nutrient reference values for calcium

New EAR, RDA and UL levels for calcium were published by the Institute of Medicine in 2011, as shown in Table 23a.3.

Table 23a.3 Estimated Average Requirement (EAR), Recommended Dietary Allowance (RDA, or Adequate Intake equivalent) and Upper level (UL) for calcium. * Applies to pregnant and lactating women. Source: Institute of Medicine (2011) and WHO (2018).

Dietary Reference Intakes Ca (mg/d) IOM  (2011)
Age/sex groups	EAR	RDA	UL
0–6 mo	–	200 (AI)	1000
6–12 mo	–	260 (AI)	1500
1–3y F & M	500	700	2500
4–8y F & M	800	1000	2500
9–18y F* & M	1100	1300	3000
19–50y F* & M	800	1000	2500
51–70y F	1000	1200	2000
51–70y M	800	1000	2000
71+y M & F	1000	1200	2000
World Health Organization  (2018)
Pregnancy		1500

Calcium balance studies pro­vided the evidence for values for ages 1–50y, whereas for those over 50y, values were set to prevent bone loss. For infants age 0–6mos, an AI is given based on the average intake of calcium from human milk, whereas at age 7–12mos, the AI assumes 120mg calcium from human milk plus 140mg calcium from com­plementary food. The values set for the AIs for infants pro­vide enough absorbed calcium to meet the calcium requirements estimated by the factorial method. For children, estimates of desirable bone accrual were con­sidered; EAR values, which start at 500mg for toddlers (age 1–3y), rise to 800mg for children, then peak at 1100mg for adolescents, were based on a new calculation of calcium retention data (Vatanparast et al., 2010). For adults, main­tenance of bone through attaining neutral calcium balance is the goal. The EAR values are 800mg for men age 19–70y and for women 19–50y, and the RDA for these age groups is 1000mg. The difference in age cut-offs for men (i.e., 70y) and women (i.e., 50y) reflects data showing that women at 50y experience abrupt bone loss at menopause in con­trast to men at this age where bone loss is slow. For women over 50y and men over 70y, EAR values are similar at 1000mg which translates into an RDA value of 1200mg. This higher level set for women over 50y and men over 70y was con­sidered a justified public health measure in the face of supportive yet incon­clusive data suggesting a higher intake of calcium would prevent bone loss and pro­tect against fractures. Values during pregnancy and lactation are not different from other women in the same age group in view of the well documented ability of pregnant and lactating women to increase calcium absorption, and hence compensate for their increased needs (Kovacs, 2016).

The IOM Upper Level (UL) for calcium for infants 0–1y was derived from feeding data that indicated no adverse effects on calcium excretion when calcium intakes as high as 1000–1500mg were pro­vided. For all other age groups, the UL for calcium (Table 23a.3) was based on data related to the incidence of kidney stones, largely from the research con­ducted with post-menopausal women in trials where calcium supplements were pro­vided. In con­trast, the European Food Safety Authority (EFSA, 2012) retained their calcium UL of 2500mg for adults, including pregnant and lactating women, on the basis that the new data did not alter their previous adult UL estimates. Further, EFSA was not con­vinced there were sufficient data to set a UL for infants, children, or adolescents.

23a.6 Calcium deficiency

23a.6.1 Hormonal responses to calcium deficiency

Calcium homeostasis is under systemic (i.e., hormonal) con­trol together with additional local regulating factors through intestinal absorption, influx and efflux from bone, and calcium excretion and re-absorption by the kidney. Serum calcium con­centrations must be kept con­stant at approximately 2.5mmol/L, a con­centration that is con­trolled by the interrelated action of three hormones: parathyroid hormone (PTH), 1,25(OH)₂D, and calcitonin. The first two hormones determine how much calcium is retained in the body through con­trol of absorption, bone turnover, and renal excretion. Calcitonin has a role in how calcium moves between the extra­cellular fluid and bone (EFSA, 2012).

Fasting, which represents acute calcium deficiency, causes a decrease in serum calcium con­centrations, inducing the release of PTH via the calcium-sensing receptor (CaSR) located on the cell surface of the parathyroid glands (EFSA, 2012). In hypo­calcemia, PTH has three actions which together bring serum calcium levels back to normal. The immediate response is bone resorption to release calcium into the blood. The second response is to increase renal re-absorption of calcium in kidney tubules in order to retain calcium. Finally, PTH stimulates 1,25(OH)₂D synthesis in the kidney (Figure 23a.1)

which serves to allow active absorption of calcium in enterocytes. Feedback con­trol of this rise in serum calcium occurs as a normal serum con­centration of calcium inhibits PTH secretion. In addition, higher than normal serum calcium stimulates calcitonin secretion by the para-follicular C cells of the thyroid gland, causing calcium to move into bone. Calcitonin's actual role in bone formation remains unclear, but it is a recognized treatment for hyper­calcemia (Felsenfeld and Levine, 2015).

23a.6.2 Bone health

Chronic calcium deficiency, arising from an habitually inadequate intake or poor absorption of calcium, is one of the many important factors associated with reduced bone mass. Bone mass in later life depends on the peak bone mass achieved during growth and the rate of the subsequent age-related bone loss. The age span over which peak bone mass is achieved is between the early teenage years and age 30, during which time 95% of bone mass is achieved. By 30y further gains are unlikely (Weaver et al., 2016), although exact timing varies with the skeletal site and sex. Optimization of bone mass is important because it reduces the risk of osteoporosis in later life.

Calcium intake and its association with bone health has been the subject of intense debate as epidemiological studies have not always found a con­sistent relationship between calcium intake and bone density. One reason is that calcium is a threshold nutrient so that when present in sufficient amounts, a higher calcium intake does not result in any further improvement in bone mineral accrual (Heaney, 2007). In most countries the Nutrient Reference Values (NRVs) have been set for main­tenance of bone health. For example, in the United States and Canada, the Estimated Average Requirement (EAR) and Recommended Dietary Allowance (RDA) for calcium for all ages is based on main­taining a positive calcium balance in children and preventing a negative calcium balance in adults (IOM, 2011); more details are described below.

Calcium-deficiency rickets and osteomalacia occur in many parts of the world. Case reports from the United States, South Africa, and Nigeria have described infants and children with radiological rickets, growth retardation, and bio­chemical signs of hyper­parathyroidism but with normal vitamin D status. In all these cases of rickets, habitual calcium intakes were low but phosphorus intakes were adequate or high. Moreover, the children in South Africa responded to calcium-rich hospital diets (Thandrayon and Pettifor, 2018), whereas the Nigerian children responded more to calcium or a com­bination of calcium and vitamin D, and not to vitamin D alone (Thacher et al., 1999). Children with severe acute malnutrition experience hypo­calcemia and many have accompanying rickets (Smilie et al., 2020).

Osteopenia and osteoporosis are con­ditions characterized by a reduced bone mass. In osteopenia, the T‑score is low, at −1. In osteoporosis, reduction in bone mass is greater, i.e., the T‑score is −2 and there is increased bone fragility and risk of fracture. The T‑score is a relative rating expressed in standard deviation units describing the bone mineral density (BMD) measured at the spine, hip, or forearm. A T‑score is similar to a Z‑score except it is not measured against a same age person but rather against a young adult with normal BMD. Measurement is done using a variety of methods described in Section 23a.9. Osteoporosis diagnosis, according to the International Osteoporosis Foundation, (IOF), is done most often using BMD tests. For example, there is a Fracture Risk Assessment Tool for physicians to use with patients of both sexes called FRAX®. It is a web-based calculator to assess the ten-year risk of osteoporosis fracture based on specific individual risk factors, with or without BMD values (Kanis et al., 2018). There are separate country models for more than 60 individual countries, and it is available in more than 30 languages. The IOF supports the main­tenance, development, and education about the use of FRAX® worldwide. The factors used in FRAX® to predict 10y fracture risk are listed in Table 23a.4.

Table 23.4: Risk factors for osteoporosis used in the FRAX® Fracture Risk Assessment Tool for physicians to assess 10-year risk of developing osteoporosis. Source: www.iofbonehealth.org

Risk factors	Risk factors used in FRAX® to predict 10y fracture risk
Age	Increasing risk 40–90y
Sex	Female is risk factor
Weight and height	Low BMI is risk factor
Previous fracture	Previous fracture as adult is risk factor
Parent fractured hip	Indicates genetic risk
Smoking	Current use of tobacco is risk factor
Glucocorticoids	≥ 3mos prednisolone (≥ 5mg daily or equivalent) is a risk factor
Rheumatoid arthritis	Confirmed diagnosis of rheumatoid arthritis is a risk factor
Secondary causes of osteoporosis	Type I diabetes, adult osteogenesis imperfecta, untreated hyperthyroid- ism, hypogonadism or premature menopause, chronic malnutrition, malabsorption, chronic liver disease
Alcohol	≥ 3 or more units of alcohol daily is a risk factor
Bone mineral density (BMD)	Risk increases with decreased femoral neck BMD (in g/cm2).

Dietary patterns and lifestyle factors such as sex, body composition, genetics, and physical activity con­tribute to bone health (Weaver et al., 2016). Dietary factors include high intakes of pro­tein and phosphorus (both as positive and negative influences) and suboptimal intakes of vitamin D as well as other micro­nutrients including vitamin K, vitamin C, magnesium, and zinc. These micro­nutrients are related to calcium metabolism and/or bone or con­nective tissue metabolism. However, based on a recent systematic review, the Level of Evidence was designated as “Inadequate Evidence” of the benefit of micro­nutrients (with the exception of calcium and vitamin D) on bone during acquisition of peak bone mass (Weaver et al., 2016). In examining dietary patterns, Moderate Evidence was found to support a beneficial role for dairy pro­ducts on bone, whereas for dietary fiber and fruit and vegetable intake the evidence on bone was deemed “Limited”. Likewise, the evidence for a detrimental effect of cola and caffeinated beverages on bone acquisition was also designated as “Limited”. For bone loss in older adults, a Canadian study con­cluded that the “Prudent” dietary pattern con­sisting of fruit, vegetables, whole grains, fish, and legumes reduced bone turnover com­pared to the Western diet of soft drinks, potato chips, french fries, meats, and desserts (Langsetmo et al., 2016).

23a.6.3 Non-bone effects of calcium deficiency

Calcium deficiency has been implicated in effects on health that are not bone-related. These include: reduction in hyper­tensive disorders of pregnancy, reduction in high blood pressure, reduction in risk of colorectal adenomas, and improvements in blood cholesterol in those at risk (Cormick and Belizan, 2019). Further, calcium has been implicated in weight management (Illich et al., 2009). The evidence for or against these effects is presented below. None of these non-bone effects were used to set Nutrient Reference Values (NRVs), yet each pro­vides additional reasons for being con­cerned about calcium adequacy. Blood pressure and calcium intake are inversely related in studies in humans as well as animal models, suggesting a role for calcium in main­taining normal blood pressure (Cormick and Belizan, 2019). Intracellular calcium regulates blood pressure in vascular smooth muscle cells, directly through vasocon­striction and indirectly through vascular volume con­trol (Villa-Etchegoyen et al., 2019). Hyper­tensive disorders of pregnancy, particularly eclampsia and pre-eclampsia, are a significant cause of severe morbidity, long-term disability and death among both mothers and their babies (WHO, 2018). Using moderate-certainty evidence from intervention studies, WHO has made the following recommendation: In populations with low dietary calcium intake, daily calcium supplementation (1.5–2.0g oral elemental calcium) is recommended for pregnant women to reduce the risk of pre-eclampsia. Additional con­siderations included in this recommendation are that all of the calcium may be from foods, if possible; that the intake should be obtained in divided doses, not all at once; and that iron supplements should be taken at a different time from calcium supplements.

Reduction in risk of colorectal adenomas with dietary calcium was first postulated in 1984, whereby in the presence of calcium ions, the toxic effects of free fatty acids on colon epithelial cells could be reduced by con­version to insoluble calcium soaps (Newmark et al., 1984). Since that time, numerous studies have refined this hypothesis, and pro­tective effects have been noted in animal studies. A systematic review and meta-analysis of randomized con­trolled trials regarding calcium supplementation for the prevention of colorectal adenomas found a modest pro­tective effect of calcium (1200–2000mg doses of supplemental calcium) in prevention of adenomas (Relative Risk (RR) = 0.89, 95% CI: 0.82–0.96) (Bonovas et al., 2016).

Blood cholesterol levels may be influenced by calcium intake. A systematic review reported that calcium supplementation reduced LDL cholesterol and increased HDL cholesterol through mechanisms involving suppression of calcium-regulating hormones (e.g., PTH) that subsequently reduced intracellular calcium in adipocytes, thus stimulating lipogenesis and lipid storage (Cormick and Belizan, 2019). However, as described below in Section 23a.7, excess calcium intake may be implicated in pro­moting cardio­vascular disease. Hence, to date there is no con­clusion regarding the role of calcium intake in reducing heart disease risk through cholesterol metabolism. Body weight con­trol by calcium has been debated for many decades. The initial suggestion arose from the finding that dairy intake facilitated weight loss better than low calorie diets without dairy, an effect that was attributed to calcium (Zemel et al., 2005). A recent trial com­pared the effect of supplements and dairy-con­taining foods and con­cluded that con­suming 4 to 5 servings of low-fat dairy foods per day or taking calcium and vitamin D supplements in persons with borderline dietary intakes of these two nutrients is beneficial for weight loss and/or main­tenance of weight loss in postmenopausal women (Ilich et al., 2009).

23a.7 Effects of high intakes of calcium

There are several health con­cerns regarding excess calcium intake over long periods of time. The risk of renal stones, risk of cardio­vascular disease, and inhibition of iron absorption are current con­cerns (Cormick and Belizan, 2019). Previously, recognition that high intakes of calcium caused calcification of soft tissue leading to “milk-alkali syndrome” (Whiting and Wood, 1995), was the rationale for setting the first Tolerable Upper Intake Level (UL) for calcium at 2500mg/d for all persons older than 1y, including pregnant and lactating women (IOM, 1997). More recently, the U.S. Food and Nutrition Board has set the Tolerable Upper Intake Level (UL) at 2000mg/d for adults over 50y due to reports of the increased incidence of kidney stones occurring in calcium supplementation trials (IOM, 2011). A higher UL is set for other age groups in the United States and Canada as the risk of stone occurrence in younger ages is much lower. The European Food Safety Authority (EFSA) however, as noted earlier, ruled that there was not sufficient data to change from the previous UL of 2500mg derived as the amount showing no adverse effects with chronic intake (EFSA, 2012).

One con­cern that remains con­troversial is the effect of excess calcium on risk of cardio­vascular disease (CVD). Many calcium trials were re‑analyzed for heart effects even though none were specifically designed to assess this effect. The most recent systematic review and meta-analysis of cohort studies and randomized con­trolled trials (Yang et al., 2020) found that calcium intake from dietary sources did not increase the risk of CVD including coronary heart disease (CHD) and stroke. However, there was evidence that calcium supplements might raise CHD risk, especially myocardial infarction. It should be noted that many of these trials pro­vided subjects (most often older women) with daily calcium supplements of 1000–1500mg without measuring dietary calcium intakes or accounting for the habitual calcium supplement use by subjects. Hence, calcium intakes may have exceeded 2000mg/d, the UL level set for adults over 50y by the Institute of Medicine (IOM, 2011).

23a.8 Biochemical indices of calcium status

As calcium homeostasis is tightly regulated, there is no satisfactory single test to assess calcium status directly on a routine basis. Calcium is an important extra­cellular electrolyte (Ca⁺⁺) and can be measured in serum or plasma. Normally values are reported as total calcium con­centration, but ionized calcium can be an important measure­ment. Hypo­calcemia and hyper­calcemia (reduced and elevated serum calcium con­centrations, respectively) are not likely to be caused by dietary intakes alone except on a transient basis. Urinary calcium excretion (calciuria) may give some clues as to calcium metabolism but as levels are equally likely to be from bone turnover or from dietary intakes, calciuria is difficult to interpret on its own. Instead, bio­chemical markers of bone formation and bone resorption are used as measures of calcium status as they reflect need for calcium (shown as excess bone resorption) and calcium adequacy (shown as appro­priate bone formation) (Bonjour et al., 2014).

23a.8.1 Serum calcium and its regulation

Serum calcium is not a test of nutritional status but rather reflects hormonal regulation of calcium, not calcium balance. Calcium is present in the blood in three different forms: as free Ca²⁺ ions, bound to pro­tein (about 45%), and com­plexed to citrate, phosphate, sulfate and carbonate (about 10%). Calcium in the blood (and in extra­cellular fluid) is kept con­stant at 2.5mmol/L (range 2.25–2.6mmol/L), but it is ionized calcium (between 1.1–1.4mmol/L) that is the form that is actively transported across membranes (EFSA, 2012). Values decrease with age in men. Females usually have slightly lower con­centrations (2.20–2.54mmol/L), associated with small differences in serum albumin con­tent and the calcium-reducing effect of estrogens. Serum calcium decreases by 5–10% up to the end of the third trimester of pregnancy, after which con­centrations rise. This decrease in total serum calcium may not be “true” hypo­calcemia, as it is not due to a decrease in ionized calcium. However, changes in the serum chemistries and calcium-regulating hormones can be mistaken as a disorder of calcium (Almaghamsi et al., 2018).

Hypo­calcemia (low serum calcium) and hyper­calcemia (elevated serum calcium) indicate serious disruption of calcium homeostasis but generally do not reflect calcium balance. In severe hypo­calcemia, blood calcium is < 2.12mmol/L, and in severe hyper­calcemia blood calcium is > 2.72mmol/L. Serum calcium levels must be corrected for the serum albumin con­centration before con­firming the diagnosis of hyper­calcemia or hypo­calcemia (Fong and Khan, 2012).

Hypo­calcemia is most com­monly a con­sequence of vitamin D inadequacy or hypo­para­thyroidism, or a resistance to either of these hormones. Hypo­calcemia has also been associated with the use of certain drugs including bisphos­phonates, cisplatin, anti­epileptics, diuretics, and pro­ton pump inhibitors. Symptoms of hypo­calcemia may include muscle spasms, cramps, tetany, numb­ness, and seizures, but also can be asymp­tomatic (Fong and Khan, 2012). Hypo­calcemia is com­mon in the critically ill patient but the interpretation of the low serum calcium level is complicated by the co-existence of hypo­albuminemia and disorders of acid-base balance, thus in these patients hypo­calcemia is defined in terms of ionized calcium con­centration rather than total calcium con­centration. Mechanisms causing hypo­calcemia in critically ill patients include: hypo­para­thyroidism, severe vitamin D deficiency, hyper­phos­phatemia, infusion of chelating agents such as citrate, alkalosis caused by hyper­ventilation, and pancreatitis that causes a rise of free fatty acids (FFAs) that result in increased binding of calcium to albumin (Kelly and Levine, 2013).

Hyper­calcemia occurs in association with hyper­para­thyroidism, hyper­thyroidism, and sarcoid­osis and when large parts of the body are immobilized (e.g., when a patient is hospitalized after spinal cord injury or major limb-bone fractures). In the latter cases, calcium from the rapidly atrophying bone is released into the circulating body fluids. In addition, hyper­calcemia is one of the characteristic features of vitamin D intoxication. It arises from hyper­absorption of intestinal calcium and to a lesser degree from the release of calcium from bone (Peacock, 2010). Severe chronic hyper­calcemia can result in nephrolithiasis (renal stones) and impairment of kidney function, as well as in calcification of soft tissues (e.g., nephrocalcinosis and vascular calcification), the latter when phosphorus con­centrations in the blood are also high, as in renal insufficiency (EFSA, 2012).

Hyper­calcemia may occur in advanced stages of the following cancers: multiple myeloma, breast cancer, parathyroid cancer, lung cancer, kidney cancer, lymphoma, leukemia, and bone metastases (Canadian Cancer Society, 2020). This hyper­calcemia causes many unpleasant clinical symptoms such as vomiting, anorexia, con­stipation, frequent and increased urination, a patient may seek medical treatment at which time blood levels may be measured along with urinalysis and kidney function tests. Treatment would involve initially reducing hyper­calcemia through fluid replacement followed by use of calcium-losing diuretics such as furosemide. As well, anti-resorptive medications such as bisphosphonates and denosumab, that reduce bone turnover, may be used to prevent calcium release from bone resorption.

23a.8.2 Measurement of serum calcium

Serum calcium levels have traditionally been determined by flame atomic absorption spectro­photometry (AAS) (Zettner and Seligson, 1964) or by automated pro­cedures that use the o‑cresolphthalein com­plexone method (Bourke and Delaney, 1993). Earlier methods used flame photometry. For surveying the American population, the National Health and Nutrition Examination Survey (NHANES) uses a DxC800 system with ion-selective electrode methodology to measure calcium con­centrations in serum, plasma, or urine. Only NHANES III pro­vides ionized calcium values for the American population. Considerable differences in the quality of serum calcium measure­ments among laboratories may occur, emphasizing the necessity for rigorous quality-con­trol practices. In general, the coefficient of variation for the AAS method is less than that for flame photometry (Linko et al., 1998). Analytical variation for serum calcium via AAS ranges from 0.9%–3.04% (Gallagher et al., 1989).

As such a large amount of calcium in serum is bound to albumin, it follows that con­centrations can be influenced by changes in serum albumin levels. As a result, when serum albumin levels are outside the normal limits (Iqbal et al., 1988), as may occur in liver disease or malnutrition, serum calcium values should be corrected, as noted earlier. The correction is approximate, and if there is uncertainty in the interpretation of the results, serum ionized calcium should be directly determined.

In many methods. serum rather than plasma should be used for calcium analysis as the former has no anticoagulant added. Most anticoagulants, with the notable exception of lithium heparin, interfere with the determination of calcium: they com­plex with or precipitate the calcium in the sample.

23a.8.3 Measurement of serum ionized calcium

Fasting blood samples are recommended for the measure­ment of ionized calcium in serum, to eliminate the effect of a recent meal. The samples should be stored anerobically at 0°C prior to the assay. Serum ionized calcium can be measured using com­mercially available ion-specific electrodes (Wandrup and Kvetny, 1985). Care must be taken to ensure that the calcium standards used for the assay con­tain sodium and chloride at the same levels as in the test samples. Further details on preparing blood for ionized calcium are published (Hamroun et al., 2020). Factors affecting serum ionized calcium levels include physical activity and circadian rhythm.

Many clinicians estimate ionized calcium using formulas that “adjust” total calcium with albumin and/or pro­tein levels, and sometimes including other blood con­stituents such as phosphate (Hamroun et al., 2020). There are over 30 published formulas, with the most widely used one being the Payne Formula: \[\small \mbox{Adjusted calcium (mmol/L) = Total calcium (mmol/L) + 0.02 (40 − serum albumin (g/L))}\] However, caution is advised before applying this formula as it is not applicable in many situations.

The clinical utility of ionized calcium ranges from con­ditions with moderate evidence (drug-induced calcium disorders, cardio­vascular outcomes), good evidence (post-thyroidectomy hypo­calcemia), and very good to strong evidence (calcium disorders in critically ill patients, primary hyper­para­thyroidism, malignancy-related hyper­calcemia). There is poor to fair evidence for use of ionized calcium in pre-eclampsia (Hamroun et al., 2020).

Interpretive criteria

The con­centration of serum ionized calcium is tightly maintained within a physiologic range of 4.4–5.4mg/dl (1.10–1.35mmol/L) (Peacock, 2010). Reported levels of serum ionized calcium in adults vary. In NHANES III, mean age-adjusted serum ionized calcium values were presented for male and female non-Hispanic Caucasians, African Americans, and Mexican Americans for two age groups: 25–59y and 60–89y; the reported mean age-adjusted value for serum ionized calcium was 1.237mmol/L for males and 1.232mmol/L for females (Vargas et al., 1998). Mean age-adjusted values for African Americans were slightly higher.

23a.8.4 Calcium balance

Calcium is lost daily through urinary and fecal excretion. In the gastrointestinal tract, calcium is lost through sloughing of cells as well as unabsorbed calcium in the feces. A small amount of calcium is lost in sweat. Replacement of these losses must occur through the diet (Weaver and Peacock, 2019). A zero calcium balance means all losses have been replaced, and there is no net bone loss to pro­vide serum calcium during periods of fasting. A positive calcium balance indicates that losses are less than intake, and implies calcium accretion into bones, an expected finding during growth and during pregnancy. A negative calcium balance means losses exceed intake, and the additional calcium lost is from bone. Over time a negative calcium balance indicates bone formation is less than bone resorption. Table 23a.5

Table 23a.5: Calcium balance, bone, and the life stages.

Ca Balance	Bone	Life Stage
Positive	Accretion Formation > Resorption	Childhood Adolescence Pregnancy
Neutral (or zero)	Maintenance Formation = Resorption	Adults
Negative	Loss Formation < Resorption	Post-menopausal Women Older Adults

pro­vides a summary of these assumptions.

Measurement of calcium balance was a main­stay of nutrition research prior to the introduction of bone imaging equipment. These early experiments required facilities (metabolic units) where participants could receive all meals, sometimes for several months. A typical experimental pro­tocol would adhere to the following steps:

1. participants ingest the experimental diet for at least one week adaptation;
2. in each 15d cycle participants would con­sume only those foods and beverages pro­vided, and every day collect 24h urine samples;
3. additionally participants would collect all feces along with con­sumption of a fecal marker during these 15d;
4. study research personal prepare com­posites of food, urine, and feces representing each of three 5d cycles; and
5. calcium was measured in the food and fecal com­posites after dry ashing to remove the organic material, followed by quantitative dilution, while urine was acidified to prevent precipitation of calcium (Anand and Linkswiler, 1974).

Balance is calculated as Ca-intake minus Ca-losses. Unless participants resided in a metabolic unit, calcium losses through sweat were not taken into con­sideration. Overall, this method is expensive and requires exceptional participant cooperation.

23a.8.5 Urinary calcium as a bio­marker

The amount of urinary calcium excreted is highly variable as it is dependent on the diet and the efficiency of intestinal absorption. Under con­ditions of bone accretion, one expects less calcium to be excreted as bones are forming, while during bone loss, more excretion might be observed. Measure of 24h urine calcium may pro­vide some indication of calcium intake as extremely low excretion rates may indicate low calcium intake. However, extremely high excretion rates, (i.e., hyper­calciuria), may be indicative of bone loss or clinical con­ditions unrelated to dietary intake.

For calcium balance, a 24h collection of urine is necessary to obtain the estimate of calcium excretion, yet this pro­cedure requires enormous participant com­pliance. To gauge an indication of urinary calcium rate, some studies measure calcium in spot urine samples, usually after an overnight fast, when urinary calcium is expressed per mg or mmol of urinary creatinine. Using urinary creatinine in the denominator corrects for variability of urinary excretion as the amount of creatinine excreted in 24 hours is con­stant for individuals. Further, creatinine excretion can be used as an measure of muscle mass as it is made from creatine. Hence, when expressing urinary calcium excretion as a ratio of urinary creatinine excretion, one can equalize excretion between participants of dissimilar body mass. There is good correlation between excretion ratios of calcium expressed per mg or mmol of urinary creatinine in 24h urine collections and those based on spot urine collections (Ilich et al., 2009).

23a.8.6 Biomarkers of bone formation and resorption

Bone remodeling is an important pro­cess that con­tinuously renews bone throughout life. It repairs micro-damage, main­tains mineral homeostasis, and ensures mechanical pro­ficiency by modifying the micro-architecture. Bone remodeling is regulated by a variety of systemic and local factors. Bone remodeling involves the following cycle of events: osteocyte signaling, recruitment of osteoclasts to begin resorption, degradation and removal of bone, reversal, formation of new bone by osteoblasts, and then a period of resting (Bonjour et al., 2014). Each of the stages in bone remodeling can pro­duce molecules, most of which are pro­tein-derived, that enter the plasma. These bio­chemical markers of bone remodeling (see Table 23a.6)

Table 23.6 Proteins or protein fragments released from osteoclasts measure bone resorption while those from osteoblasts measure bone formation. Source: Bonjour et al., Nutrition Research Reviews, 2014, 27: 252–267.

Bone Resorption (Osteoclast activity)
Osteoclast number Tartrate-resistant acid phosphatase (TRAP)
Bone matrix absorption Pyridium cross-links (PYR) Carboxy terminal telopeptide (CTX) terminal propeptide (PINP)
Bone Formation (Osteoblast activity)
Osteoblast protein synthesis Procollagen type
Osteoblast activity Bone alkaline phosphatase (BALP) Osteocalcin (OC)

can be used to measure subtle changes in the rate of formation or degradation of the bone matrix brought about by dietary influences, such as availability of calcium, phosphate, and pro­tein. When remodeling rates are changing, a com­bination of bio­markers, such as one formation marker and one resorption marker can pro­vide more information than a single marker (Watts, 1999). In addition, they can pro­vide an earlier estimate on the effect of dietary interventions or drug treatments com­pared to measures of BMD. However, because of increases in bone turnover during the night, bone bio­marker con­centrations are affected by circadian rhythm. For all the bone bio­markers shown in Table 23a.6, their measure­ment is generally recommended using automated platforms which have improved technical performance (Bonjour et al., 2014).

A description of the com­monly used bone bio­markers shown in Table 23a.6 is outlined below. Of these, use of only two, namely P1NP (for formation) and CTX‑1 (for resorption) are recommended by the International Osteoporosis Foundation (IOF), and the International Federation of Clinical Chemistry (IFCC) because they are well characterized for use in fracture risk prediction and for monitoring osteoporosis treatment (Kim et al., 2020).

23a.8.6.1 Biomarkers of bone formation

Bone formation is determined by measuring pro­teins that increase in serum as a result of osteoblast activity during bone formation. These include bone alkaline phosphatase (BALP), which is measured as enzymatic activity, and the vitamin K‑dependent pro­tein osteocalcin (OC). A third marker, pro­collagen type I N‑terminal pro­peptide (PINP) that reflects the pro­tein synthesis capacity of the osteoblast can also be used. Early measure­ment of these three bone formation markers (OC, BALP and particularly PINP) has shown positive correlations with subsequent bone mineral density improvements in trials of teriparatide, a drug pro­moting bone formation (Bonjour et al., 2014). Bone alkaline phosphatase (BALP), an isoenzyme also known as bone-specific alkaline phosphatase or skeletal alkaline phosphatase, is measured in serum, and is essential for normal bone mineralization. Several techniques have been developed to measure the activity of this enzyme. In adults approximately 50% of serum total alkaline phosphatase is present as bone alkaline phosphatase, whereas during childhood, the bone isoenzyme predominates. BALP may also be measured when assessing vitamin D deficiency (see chapter 18b.12.2) during which osteoblast activity is suppressed due to low calcium absorption. Thus, BALP is not specific to calcium deficiency and interpretation of its activity must be accompanied by other measure­ments.

Osteocalcin (OC), measured in serum, is also called a bone γ‑carboxy­glutamic acid–con­taining pro­tein because it con­tains up to three γ‑carboxy­glutamic acid residues. This small non-collagenous pro­tein is specific for bone tissue and dentin and com­prises about 1–2% of total bone pro­tein. During synthesis of osteocalcin, a small fraction is released directly into the blood. Osteocalcin is cleared by the kidneys so that serum levels depend on renal function. The half-life of serum osteocalcin is short (15–70min). Uncertainties persist as to the exact role of osteocalcin. Indeed, osteocalcin is secreted solely by osteoblasts, yet has only minor effects on bone mineralization. Instead, osteocalcin acts as a hormone-like factor in glucose homeostasis, brain development, cognition, and male fertility (Moser and van der Eerden, 2019).

Procollagen type I N‑terminal pro­peptide (PINP) measured in serum, is the preferred marker to measure bone formation (Bonjour et al., 2014). Most of the organic matrix of bone is type I collagen associated with non-collagenous pro­teins, and Type I collagen is rich in the amino acid hydroxy-pro­line. During the con­version of pro­collagen to collagen, which occurs extra­cellularly, the pro­collagen pro­peptides, pro­collagen type I carboxy- terminal pro­peptide (PICP) and pro­collagen type I amino­terminal pro­peptide (PINP) are released into the circulation. Although both bone alkaline phosphatase and PINP are markers of bone formation, PINP is the reference bio­marker for bone formation (Kim et al., 2020).

23a.8.4.2 Biomarkers of bone resorption

Bone resorption markers include the pyridinium cross-links (PYR) subdivided into pyridinoline and deoxypyridinoline together with the associated telopeptides, i.e., carboxy terminal telopeptide (CTX) and N‑terminal telopeptide (NTX), all of which are released during collagen breakdown ( Table 23a.6). A second marker of bone resorption is tartrate-resistant acid phosphatase (TRAP), a measure of osteoclast number (Bonjour et al., 2014).

Pyridinium cross-links (PYD) measured in urine were among the first bone bio­markers of resorption, but are rarely measured today. The crosslinks are hydroxylysylpyridinoline (known as pyridinoline, PYD) and lysylpyridinoline (known as deoxypyridinoline, DPD) and are important for the struc­tural integrity of the collagen. Pyridinoline is widely distributed in both type I collagen of bone and type II collagen of cartilage, as well as in smaller amounts in the other con­nective tissues, and is thus con­sidered non-specific. In con­trast, deoxypyridinoline is found almost exclusively in type I collagen of bone (Kuo and Chen, 2017). Both the cross-links are released into the circulation after mature tissue collagen is degraded. They are not re-utilized or metabolized in the liver but, instead, are excreted in the urine unchanged, in peptide-bound and free forms. Neither PYD or DPD are com­monly measured (Bonjour et al., 2014).

Cross-linked N‑telopeptides (NTX) and C‑telopeptides (CTX) measured in serum or urine are now the most commonly measured bone bio­markers of resorption. C‑telopeptides (CTX ) are better characterized than NTX and are the preferred bone resorption marker, although both have similar pro­perties and origins. CTX is the reference marker for bone resorption (Kim et al., 2020). CTX and NTX are carboxy-terminal and amino-terminal fragments of collagen, respectively which are released into blood and subsequently excreted in the urine with cross-links attached (Bonjour et al., 2014). However, serum CTX levels are decreased by food intake so blood withdrawal must take place in the fasting state (Kuo and Chen, 2017). Measurement of bio­markers can be con­ducted using more than one detection method for each bio­marker. A com­plete list of all the bio­markers discussed above as well as others still under con­sideration is available (Kuo and Chen, 2017), com­bining the bio­markers to generate a bio­marker risk score is under development to enhance the accuracy of predicting fractures (Kim et al., 2020).

23a.9 Measurement of bone mineral con­tent and bone mineral density

Noninvasive techniques can be used to pro­vide an indirect assessment of calcium balance as well as calcium “stores” in bone by measuring the bone mineral con­tent (BMC) or bone mineral density (BMD). These methods include dual X‑ray absorptiometry (DXA, also called DEXA) and quantitative com­puted tomography (QCT), both major diagnostic tools for osteoporosis. Ultrasound (US), although less accurate for medical diagnosis, can serve as a tool for screening and for field work. These technologies measure the mineral con­tent of bone in the appendicular skeleton, the axial skeleton, or the total skeleton, measure­ments that relate directly to the calcium con­tent of the skeleton. As almost all of the calcium in the body is stored in the skeleton, it follows that skeletal mineral con­tent can be used as an indirect measure of body calcium stores. These same methods can also be used to monitor the response to changes in calcium intakes over relatively long time periods, i.e., months or even years.

Over the past 40 years technologies such as DXA and QCT have been developed and refined to measure calcium in hard tissues, and thus estimate of total body mineral con­tent, which in turn can be used to calculate total body calcium. These technologies are also used to determine bone integrity for the diagnosis of osteoporosis and other bone diseases. They can also be used during treatment to evaluate efficacy as well as in research studies for the development of new drugs, exercise regimens, or nutrition interventions (IOF, 2020).

The measure­ment of bone mineral using DXA is usually expressed as the bone mineral density, which corrects for pro­jected bone area. However, the measure­ment is not a true density measure­ment, but instead, a measure of areal density (g/cm²) derived by dividing bone mineral con­tent (g) by the scanned bone area. Bone mineral density is calculated in this way to reduce the bio­logical variation observed in values for bone mineral con­tent at all ages and thus to increase the statistical power of detecting abnormal values. Nevertheless, bone mineral density does not adequately correct for bone area and body size and thus must be interpreted carefully (Prentice et al., 1994). In con­trast, QCT uses multiple slices to con­struct three-dimensional images, thus pro­viding a true volumetric measure for BMD (Adams, 2009).

Diagnosis for osteoporosis is based on the difference between the measured bone mineral density (BMD) of the participant and the mean BMD of healthy young adults, matched for sex and ethnicity. The difference is expressed relative to the young adult population standard deviation, and reported as a T‑score. WHO, 1994 has defined both osteoporosis and osteopenia based on T‑score values. An individual with a T‑score less than −2.5 at the spine, hip, or forearm is classified as having osteoporosis; a T‑score between −2.5 and −1 is classified as osteopenia; and a T‑score greater than −1 is healthy. These WHO definitions should not be applied to other bone mineral density measure­ment sites.

The International Society for Clinical Densitometry (ISCD) has established guidelines for bone density testing using DXA. In addition, they pro­vide specific guidelines for diagnosing osteoporosis based on T‑score values for postmenopausal women, men (20y and older), pre-menopausal women (20y to menopause), and children (males or females < 20y). Persons using DXA technology should be certified, and facilities should be accredited; con­sult the ISCD for advice on certification and accreditation.

23a.9.1 Single-photon absorptiometry

Single-photon absorptiometry was the first practical noninvasive method that was developed to examine the peripheral skeleton (Wahner et al., 1983). although now it has mostly been replaced by DXA (IOF, 2020). The original method was based on the assumption that the bone mineral con­tent is inversely pro­portional to the amount of photon energy transmitted through the bone under study. A mono-energetic photon source is used, usually ¹²⁵I or ²⁴¹Am. The technique is fast, taking approximately 5min. With this method, bone mineral density is measured with a precision (%CV) of less than 2% and with an effective dose to the participant of only 0.1µSv. The site most frequently selected for measure­ments by single-photon absorptiometry is the lower radius, at approximately one-third of the distance from the styloid pro­cess to the olecranon. Single-photon absorptiometry is not suitable for measure­ments on the axial skeleton because the technique requires that a uniform thickness of soft tissue surrounds the bone (Neer, 1992). Newer technologies have replaced SPA for human studies as SPA cannot measure those sites (hip or spine) needed for clinical diagnoses. In the past four decades SPA machines have been adapted and used for animal studies of bone health (Sequeira et al, 2020).

23a.9.2 Dual X‑ray absorptiometry (DXA)

The first com­mercial DXA scanner was marketed in 1987. There are now several com­mercial versions, each using similar measure­ment pro­cedures The original DXA method used an isotopic source, typically ¹⁵³Gd , that emitted two low-energy gamma rays at 44Kev and 100KeV. The short half-life of the isotopic source ¹⁵³Gd, however, limited the precision with which long-term changes in bone mineral con­tent in the same participant could be made using this technique. This limitation led to the radioactive ¹⁵³Gd source being replaced with an X‑ray tube behind a K‑edge filter. The filter con­verts the polychromatic X‑ray beam into one with two main energy peaks at 40Kev and 70KeV. These two con­gruent X‑ray beams are passed through the body, and the ratio of beam attenuation at the lower energy relative to that at the higher energy differentiates between bone mineral, bone-free mass, and the fat mass (Kim et al., 2018). As a result, this modified instrument is termed a dual energy X‑ray absorptiometer, abbreviated as DXA (or DEXA).

The DXA instrument is used not only for bone mineral measures but also to assess body com­position, as described in Chapter 14. Only the measure­ments of bone mineral mass (g), bone mineral con­tent (g/cm), and “areal” bone mineral density (g/cm²) are discussed here. Using DXA, measure­ments of areal bone mass density (aBMD) can be made at the lumbar spine, femur, and forearm, as well as total body. Measurements of body mineral density at the lumbar vertebrae (L2 to L4) are often favored because they are sensitive to the changes associated with aging, disease, and therapy. Care must be taken when positioning the participant for the scan to ensure that precise and accurate data are obtained. Under some circumstances, it may be appro­priate to make a whole-body scan. This is quick, taking from 3min to 35min, depending on the instrument used and the age of the participant, and requires very little cooperation from the participant. As a result, scans can be readily performed on young children, the elderly, and even persons who are sick, although pregnant women should be excluded (Kim et al., 2018).

Dual-energy X‑ray absorptiometry is now the dominant method for measuring bone loss for clinical diagnosis of osteopenia and osteoporosis. In intervention studies care must be taken to ensure that both the length of the intervention and the size of the study population are sufficient to allow the detection of small changes in bone mineral density. A meta-analysis of trials reported that increasing calcium intake from dietary sources increased BMD by 0.6–1.0% at the total hip and total body at one year, and by 0.7–1.8% at these sites and the lumbar spine and femoral neck at two years (Tai et al., 2015). Larger decreases in bone loss have been observed after treatment with estrogen, calcitonin, or biphosphates.

When com­paring measure­ments using DXA based on different instruments and technologies, it is important to acknowledge the potential for systematic differences arising from methodological factors which could mask those associated with bio­logical variation. Methodological factors may include differences in instrument calibration methods, whereas variations in body thickness, distribution of body fat, and the fat con­tent of the bone marrow are major sources of bio­logical variation. Of the latter, DXA is especially sensitive to discrepancies in body thickness, which can lead to systematic differences in the bone mineral estimates on thin and obese subjects. In a global survey of fracture liaison services, although access to DXA was reported to meet the needs (Clynes et al., 2020), only around 50% of institutions con­firmed adherence to basic DXA quality and reporting pro­cedures, and many required educating the operators/interpreters. Indeed, significant variability worldwide in both the quality and access to DXA services was reported in this survey.

Calcium balance using DXA may be determined by obtaining a total body (TB) DXA scan that pro­vides bone mineral con­tent (BMC). Scans of total body bone mineral con­tent (TBBMC) can be used to calculate calcium requirements of specific life cycle groups, on the assumption that bone mineral is 32·2% calcium (Vatanparast et al., 2010). Use of this value is a better approximation of calcium in bone than simply dividing the atomic weight of calcium (40g/mol) by the molecular weight (i.e., 1005g/mol) of hydroxyapatite Ca₁₀(PO₄)₆(OH)₂ because the mineral in bone is hydrated so water molecules con­tribute to the weight of bone mineral con­tent.

To derive the Estimated Average Requirement (EAR) for calcium for adolescents 9–18y, the U.S Food and Nutrition Board applied the factorial approach to estimate the physiological calcium requirements (IOM, 2011). Estimates of calcium losses via urine, feces, and sweat were derived from metabolic balance studies to which was added the calcium accrued as bone mineral for growth (derived from TBBMC). The latter was measured in a cohort of Canadian males and females in two age groups, 9–13y and 14–19y as shown in Table 23a.7

Table 23.7. Determining calcium requirements during adolescence (age 9–18y) using total body bone mineral content (TBBMC). Source: Vatanparast et al., (2009). British Journal of Nutrition 103: 575–580.

Factorial Components	Girls 9–18y (mg/d)	Boys 9–18y (mg/d)
Ca accretion from TBBMC	121	175
Urinary Ca Losses	106	127
Endogenous Fecal Ca	112	108
Sweat Ca Losses	55	55
Total Ca physiological need	394	465
Absorption (%)	38	38
Dietary Ca requirement (EAR)	1037	1224

(Vatanparast et al., 2010). To translate the physiological requirements for absorbed calcium (from the factorial approach) into the EAR, it is necessary to take into account the pro­portion of calcium in the habitual diet that is absorbed by the intestine. Assuming an efficiency of absorption of 38% from an omnivorous diet, the equation used for the estimated Ca dietary requirement (i.e., the EAR) using the formula for adolescence which is a period of rapid growth:
EAR = Ca losses + (Ca needs for growth / 38%)
(IOM, 2011).

23a.9.3 Com­puterized tomography (CT) and peripheral quantitative CT (pQCT)

Computerized tomography (CT), often called quantitative CT (QCT) when referring to bone measure­ments, can be used to measure bone mass of both the appendicular and axial skeleton, although the method is not feasible for population studies. The advantage of QCT is that a true three-dimensional (i.e., volumetric) bone density can be obtained, so results can be expressed in g/cm³ rather than the two-dimensional areal density measured by DXA. Bone can also be identified as cortical or trabecular in nature on the basis of bone density measure­ments. This is an advantage because some forms of osteoporosis are predominantly trabecular in character. The equipment is not portable, and the radiation dose required per slice for imaging is relatively high (50–500µSv), although it can be reduced if scans are limited to regions of specific interest. The cost of the equipment and its main­tenance is high.

The use of QCT technology applied to specific bone sites yields information about micro­architecture. Methods that can be used include micro‑CT and peripheral Quantitative CT (pQCT), with the latter having widespread use. The equipment for pQCT is small relative to DXA and CT, and pro­vides imaging of both the arm at the distal radius and the lower leg (tibia) sites to assess bone micro­architecture. High-Resolution pQCT (HR‑pQCT) has been developed to improve the resolution pro­blems of pQCT. Recently, second-generation HrpQCT has been introduced which differs from the first-generation HR‑pQCT in scan region, resolution, and morphological measure­ment techniques (Whittier et al., 2020).

23a.9.4 Quantitative ultrasound

Quantitative ultrasound (QUS) is used for measure­ments of the bone density of the peripheral skeleton such as the proximal arm (radius), lower leg (tibia) or heel. A wide variety of equipment is now available to measure the attenuation of a sonographic pulse as it passes through bone and is scattered and absorbed by the trabeculae. The heel is the most frequently used measure­ment site because it has a large volume of trabecular bone that is accessible for transmission measure­ments. However, there are ultrasound machines on the market that measure multiple skeletal sites. Quantitative Ultrasound has the advantage of being radiation-free and a portable system.

Quantitative ultrasound measures the speed of sound (SoS) in meters per second (m/s). In a study of different age groups using QUS, measures of SoS (m/s) varied according to life stages that reflected bone development (Rivas-Ruiz et al., 2015). For example, bone SoS accretion was found to begin 5y earlier in girls than boys (p < 0.05), a sex-related difference that mirrors results seen in DXA scans (Vatanparast et al., 2010). The maximal (peak bone) SoS was noted at 28y at the radius site and at 22y at the tibia site. Women who were postmenopausal (age 45–50y) showed a significant decrease in SoS measures at both these sites (p < 0.05) compared with men, as would be expected. Thus, while the precision of QUS is poor relative to DXA, the equipment may be suitable for characterizing bone in populations where DXA is not a feasible option. However, QUS has no clinical applications and is not used to define T‑scores for diagnosing osteoporosis or monitoring treatment. However, a systematic review has shown that quantitative ultrasound is an independent predictor of fracture for men and women when QUS values are low (McCloskey et al., 2015).