23.8: Biochemical indices of calcium status (23a.8)

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 (between1.1–1.4mmol/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:

\[\begin{gathered}
\text { Adjusted calcium }(\mathrm{mmol} / \mathrm{L})= \\
\text { Totalcalcium }(\mathrm{mmol} / \mathrm{L})+0.02(40-\text { serum albumin }(\mathrm{g} / \mathrm{L}))
\end{gathered}\nonumber\]

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 pro­vides a summary of these assumptions.

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

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) 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).

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)

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).