27.2: Serum copper (24b.1)

There are two main forms of copper in serum; one is firmly bound to cerulo­plasmin, the other is reversibly bound to serum albumin. and to amino acids and other, poorly defined components (Linder 2016). Serum also contains the two copper enzymes cyto­chrome c oxi­dase and mono­amine oxidase.

Serum copper con­cen­trations are a reliable bio­marker in cases of severe copper deficiency, when they are almost always low. However, they are not sensitive enough to be used as a measure of marginal copper status, as noted earlier, and do not reflect dietary intake, except when it is low. Normally, sup­plemen­tation with copper does not increase serum copper because con­cent­rations are under strict homeo­static control. Levels are only reduced after copper stores are signif­icantly depleted. In healthy adults consuming diets adequate in copper, the lower end of the normal range for serum copper is 10µmol/L.

Experimental studies have also confirmed that serum copper is not a sensitive index of marginal copper status in adults (Reiser et al., 1985; Turnlund et al., 1989; Milne and Nielsen, 1996). In a depletion–repletion study of post­meno­pausal women fed a diet with only 0.57mg/d Cu for 105d, serum copper levels did not change significantly during the copper-depletion period (Figure 24b.2), despite the presence of other bio­chem­ical changes (Milne and Nielsen, 1996).

Figure 24b.2 Changes in copper biomarkers during copper deprivation. Each data point is the mean of 3wk surrounding date. Data from Milne and Nielsen, American Journal of Clinical Nutrition 63: 358–364, 1996

Eventually serum copper concen­trations do fall. In patients receiving prolonged TPN unsup­ple­mented with copper, falls of about 1.7µmol/L per week may occur (Solomons et al., 1976). Values < 9.5–11µmol/L have been reported in infants with copper deficiency (Cordano et al., 1964; Castillo-Duran and Uauy, 1988).Low serum copper concen­trations also occur with the two genetic disorders of copper meta­bolism: Menkes kinky hair syndrome and Wilson's disease. In Menkes kinky hair syndrome, low serum copper concen­trations arise from an intra­cellular defect of copper utilization, as discussed earlier (Danks et al., 1972). In Wilson's disease, the low serum copper levels are induced by defects in the hepatic storage of copper and in cerulo­plasmin meta­bolism.

Plasma copper concen­trations appear to be consistently lower than corres­ponding values in serum (Table 24b.2) (Smith et al., 1985). In contrast, concentrations of copper in serum and erythrocytes are comparable, so slight hemolysis does not affect serum copper concentrations.

Table 24b.2 Copper concentrations of serum or citrated plasma prepared from single venipuncture samples of eight subjects. Plasma was prepared by adding 0.15mL of 30% sodium citrate per 15mL of whole blood. Percentage difference = [(serum − plasma)/serum] × 100. Data from Smith et al., Journal of the American College of Nutrition 4:627–638, 1985

Subject no.	Serum (μmol/L)	Plasma (μmol/L)	Difference
1	20.0	19.4	3%
2	18.7	17.8	5%
3	19.0	17.3	9%
4	19.2	16.7	13%
5	19.5	16.7	15%
6	19.4	16.5	15%
7	18.1	16.5	9%
8	16.5	15.4	7%
Mean	18.9±1.1	17.0±1.1	9±4

More recently, examination of the non-cerulo­plasmin pool of copper, the exchange­able fraction, has suggested that this might be a good indicator of copper overload, as in diseases such as Wilson's disease (Woimant et al., 2019).

• Factors affecting serum copper

It is important to remember that most assays do not differentiate between the different forms of copper in serum and usually, changes in “serum copper" are generally caused by alterations in the concen­tration of cerulo­plasmin. The vast majority of coper in serum is bound to cerulo­plasmin. Estimates vary from 60% to 95% with the remainder being bound to a poorly characterised series of chelators (Linder 2016). As well as being a ferr­oxidase, involved in iron metabolism, cerulo­plasmin is an acute phase protein, and concen­trations can be altered by other factors than copper status. This makes interpretation of the changes described below complex.

Age-related changes in serum copper concen­trations are well documen­ted. Newborn infants have low values, which rise to adult levels by 6 to 12mo of age (Salmenperä et al., 1986). In men, and in women not taking estrogen-containing prep­arations, serum copper levels increase with age until55–64y, after which they decline (Table 24b.3 (Fischer et al., 1990; Milne and Johnson, 1993).

Table 24b.3 Effect of age in men and women who were not taking estrogen-containing prepar­ations on serum copper concentration and ceruloplasmin activity. Data from Fischer et al., Nutrition Research 10: 1081–1090, 1990.

Age group (years)	Men	Women
	Serum copper (μmol/L)
25–34	14.7 ± 0.6	14.8 ± 0.5
35–44	15.0 ± 0.3	16.7 ± 0.4
45–54	16.2 ± 0.4	17.8 ± 0.6
55–64	16.3 ± 0.4	18.8 ± 0.7
65+	15.3 ± 1.0	16.8 ± 0.5
	Ceruloplasmin (U/dL)
25–34	9.9 ± 0.7	11.0 ± 0.6
35–44	10.8 ± 0.3	12.8 ± 0.4
45–54	11.9 ± 0.4	15.2 ± 1.1
55–64	11.8 ± 0.5	14.1 ± 0.8
65+	11.3 ± 1.2	12.3 ± 1.1

Gender influences serum copper. Levels in adult females tend to be higher than in males (Table 24b.3).

Pregnancy influences serum copper concen­trations. Higher serum copper concen­trations are evident after the third month of pregnancy (Halsted et al., 1968; Hambidge and Droegemueller, 1974). Whether this reflects changes in cerulo­plasmin, which is part of the acute phase response, or an actual change in serum copper concen­trations, is not always clear.

Estrogen-containing preparations, such as oral contra­ceptive agents and estrogen replace­ment therapy, elevate serum copper concen­trations (hyper­cupremia) (Fischer et al., 1990; Nielsen et al., 1992). This effect is due to an estrogen-mediated rise in serum ceruloplasmin concentration No differences in copper absorption or copper balance have been reported between users and non-users of oral contra­ceptive agents (King et al., 1978; Crews et al., 1980). (Table 24b.4).

Table 24b.4 Effect of oral contraceptive agents (OCA) and estrogen replacement therapy on serum copper concentration and ceruloplasmin activity. Data from Fischer et al., Nutrition Research 10: 1081–1090,1990.

	Premenopausal women
	Non-users of OCA	Users of OCA
Serum copper (µmol/L)	16.2±0.3	26.8±1.3
Ceruloplasmin (U/dL)	12.2±0.4	20.4±0.9
	Postmenopausal women
	Non-users of estrogens	Users of estrogens
Serum copper (µmol/L)	18.1±0.4	25.9±1.2
Ceruloplasmin (U/dL)	14.5±0.7	19.7±1.0

Diurnal variation in serum copper concen­trations has been noted (Cartwright, 1950; Guillard et al., 1979) with the highest levels occur in the morning (Lifschitz and Henkin, 1971; Cartwright et al., 1954; Yokoyama et al., 2000).

Regular strenuous exercise may also affect serum copper concen­trations. Lukaski et al. (1983) showed values to be higher in male university athletes relative to controls.

Smoking raises serum copper concen­trations (Kocyigit et al., 2001).

Infection, inflam­mation, and stress all elevate serum copper concen­trations (Honkanen et al., 1991; Brown et al., 1993). This is a result of an increase in cerulo­plasmin (an acute phase protein), effected by leuko­cytic endo­genous mediators (Pekarek et al., 1972).

Malabsorption syndromes such as celiac disease, cystic fibrosis, and ulcerative colitis (Sternlieb and Janowitz, 1964) are often associated with low serum copper concen­trations, arising from pro­longed diarrhea which prevents reabsorp­tion of copper from the bile (Rodriguez et al., 1985).

Certain disease states lead to alterations in serum copper concen­trations. Levels increase in leukemia, Hodgkin's disease, various anemias, collagen disorders, hemo­chrom­atosis, and myo­cardial infarction and in patients with dilated cardio­myopathy (Mason, 1979; Oster, 1993). In other diseases (e.g., the nephrotic syndrome), serum copper concen­trations are low (Cartwright et al., 1954).

• Interpretive criteria

Table 24b.5 Plasma copper concentrations in children, grouped according to age (mean±SD). Data from Rükgauer et al., Journal of Trace Elements in Medicine and Biology 11: 92–98, 1997.

Age group (years)	n	Serum Cu (µmol/L)
1 to < 2	15	21.1±4.6
2 to < 4	23	21.5±3.9
4 to < 6	19	19.4±5.3
6 to < 10	25	23.4±2.5
10 to < 14	21	21.1±3.7
14 to < 18	17	20.5±4.4

Interpretive criteria used for serum copper concentrations for adults are 8.8–17.5µmol/L for men and10.7–26.6µmol/L for women who are not taking oral contraceptive agents. A slightly higher range(15.7–31.5µmol/L) has been proposed for women taking estrogen-containing prepar­ations (Milne and Johnson, 1993). Rükgauer et al. (1997) have published reference values for plasma copper for children ranging from 1mo to 18y, although the number of children studied in some of the age groups was small. These reference values are based on self-selected samples of apparently healthy subjects who were not taking any vitamin or mineral supplements (Table 24b.5). Salmenperä et al. (1986) have compiled reference values for serum copper in exclusively breast­fed infants followed longitudinally for the first year of life. These reference values are shown in (Figure 24b.3). Serum copper levels were measured in subjects from 3 to 74y in the Canada Health Survey (Health and Welfare Canada, 1981). The National Health and Nutrition Examination Survey (NHANES II) also collected data on serum copper concentrations, but were not measured in the more recent U.K. national surveys.

Figure 24b.3 Median serum copper and ceruloplasmin concentrations in exclusively breastfed infants during the first year of life. Data from Salmenperä et al., American Journal of Clinical Nutrition 43: 251–257, 1986.

• Measurement of serum copper

The most widely used method for measuring serum copper is AAS, although inductively coupled mass spectrometry is becoming more common (McMillen et al., 2009). Generally, a direct technique is used, involving sample dilution with deionized water (1 part plasma or serum to 1 part deionized water) (Osheim, 1983; Smith et al., 1985): a signal-enhancing mixture such as butanol and water is also some­times used(Meret and Henkin, 1971). In some instances, use of a “high solids” burner head may be necessary (Boling, 1966).Sometimes, the protein in the blood sample is removed using an acid such as trichloro­acetic acid (Kelson and Shamberger, 1978),but this procedure may introduce volume errors during the deprotein­ization step and adventitious contamin­ation from the acid and is not recommended. Analytical variation for serum copper by AAS is usually small; Gallagher et al. (1989) reported it to be < 2.6%. For very small pediatric samples, graphite furnace AAS, with Zeeman-effect background correction, may be used (Wang and Demshar, 1993). Increasingly inductively coupled plasma spectrometry (ICP) is being used because it is a multi-element technique with a detection limit and sensitivity better than flame AAS (Nixon et al., 1986). Standard reference materials for the analysis of copper in serum are available from NIST, Gaithersburg, MD.