Alterations in serum calcium and magnesium concentrations are common in horses that are affected by a variety of disorders (eg, exercise-induced electrolyte loss, chronic renal failure, and sepsis).1–4 In addition, because of their influence on gastrointestinal motility, there is a growing interest on changes in these minerals in horses with gastrointestinal tract disease (mainly horses with colic).4–6
Both tCa and tMg are present in serum in the following 3 forms: ionized (iCa and iMg); bound to proteins (pCa and pMg); and complexed with weak acids (cCa and cMg), including citrate, phosphate, and bicarbonate.7,8 The ionized fractions, which are able to interact with the calcium-sensing receptor and other cationic receptors, are thought to be the only physiologically active forms.9,10 This theoretic concept is reinforced by studies6,11 that have demonstrated that quantification of serum iCa and iMg is superior to measurements of the total electrolyte concentration in diagnosing hypocalcemia and hypomagnesemia in horses. However, it has been suggested that the complexed forms, as a result of the loose bounds between cations and weak acids, may also have a biological role.12
The use of micropartition systems based on the filtration method allows separation of the calcium and magnesium fractions. Micropartition systems contain a membrane that is highly retentive for the serum protein contents and through which serum is filtered. After centrifugation, the protein-bound fractions are retained and the ultrafiltrable fractions (ionized and complexed) pass through the membrane.13,14 Measurement of the ionized fractions by use of selective electrodes permits a distinction between the ionized and complexed minerals in the ultrafiltrate.8,15
Fractionation of serum calcium and magnesium has been reported for humans and dogs.8,15-18 In horses, Holley and Evans19 studied total and μCa and μMg in serum from clinically normal horses. However, these authors did not differentiate the ionized and complexed fractions, which are more clinically relevant.
With the development of ion-sensitive electrodes, iCa and iMg can be measured with relative ease in serum. However, calcium and magnesium electrodes are not readily available to many equine practitioners who may need to submit serum samples for analysis. Therefore, it is important to know the stability of iCa and iMg in serum samples of horses. Purposes of the study reported here were to describe the fractionation of calcium and magnesium in equine serum and study the stability of iCa and iMg in refrigerated and frozen serum samples.
Materials and Methods
Animals—Blood samples were collected from 28 healthy horses (19 mares, 4 stallions, and 5 geldings). Horses were considered healthy on the basis of normal findings on physical examinations, CBCs, and blood biochemical analyses results within reference range. In addition to routine blood biochemical analysis, mineral metabolism was studied in all horses by measuring serum calcium, phosphorus, magnesium, PTH, and calcitriol concentrations. Horses were maintained in paddocks and were fed the same diet (hay, oats, and a vitamin-mineral supplement) for at least 2 months before blood sample collection. Experimental protocols were reviewed and approved by the Ethics Committee for Animal Research of the University of Cordoba (Spain).
Blood sample collection and measurements—Blood samples were obtained anaerobically from the jugular vein and transferred to 10-mL vacuum tubes for serum separation.a Samples were spun at 1,000 × g for 5 minutes to separate serum. The serum was anaerobically transferred from the collection tube to a syringe. Serum iCa and iMg concentrations and pH were measured immediately by use of selective electrodes.b The tCa and tMg were then measured by spectrophotometry.c,d
Fractionation of serum tCa and tMg was performed on the fresh samples of equine serum by use of a micropartition system as previously described for other species.13–18 Briefly, a serum aliquot (2 mL) was gently placed into a disposable ultrafiltration membrane with a conical shape.e The serum was subsequently spun at 4,000 × g for 15 minutes at 4°C to separate the protein-bound fractions, which were retained by the membrane, from the μCa and μMg, which were filtered. Ultrafiltrable fractions were quantified by spectrophotometry.c,d From these measurements, serum concentrations of complexed (cCa and cMg) and protein-bound (pCa and pMg) fractions were calculated as follows: cCa = μCa – iCa; cMg = μMg – iMg; pCa = tCa – μCa; and pMg = tMg – μMg, respectively.
Additional biochemical measurements included total protein concentrations that were quantified by spectrophotometry in the serum (by use of the biuret technique)f and ultrafiltrate.g Inorganic phosphorus was measured by spectrophotometry,h PTH was quantified by use of an immuno radiometric assay that has been validated for quantification of equine PTH,i and calcitriol was measured by use of a radioimmunoassay.j
To study the effect of storage time and conditions on iCa and iMg measurements, iCa and iMg were measured in 14 serum samples at various times as follows: just after collection; after the samples had been refrigerated at 4°C for 6, 12, 24, 48, and 120 hours; and after the samples had been frozen at −20°C for 15 and 30 days. These serum samples were obtained from the collection tubes as already described and were maintained in anaerobiosis during the storage time.
Statistical analysis—Mean ± SE values were calculated for each calcium and magnesium fraction. Results were expressed in absolute values and as a percentage of tCa and tMg. A Pearson correlation test was used to study the correlation between different calcium and magnesium fractions. Changes in iCa and iMg during storage were studied by paired t tests. For all statistical comparisons, values of P < 0.05 were considered significant.
Results
Horses of this study had a normal mineral metabolism on the basis of the following serum biochemical parameters: calcium, 3.26 ± 0.06 mmol/L; inorganic phosphorus, 0.84 ± 0.06 mmol/L; magnesium, 0.99 ± 0.04 mmol/L; PTH, 42.9 ± 4.1 pg/mL; and calcitriol, 8.9 ± 1.2 pg/mL. The serum tCa concentration (3.26 ± 0.06 mmol/L) was composed of 47% pCa (1.55 ± 0.03 mmol/L) and 53% μCa (1.71 ± 0.04 mmol/L; Table 1). When standardized for the serum total protein concentration (64.3 ± 0.9 g/L), the amount of calcium bound to proteins was 0.024 mmol Ca/g of protein. Greater than 90% of μCa was in the ionized form (iCa, 1.58 ± 0.03 mmol/L), and approximately 9% was complexed (cCa, 0.13 ± 0.02 mmol/L). A good correlation between tCa and pCa (r = 0.602; P = 0.006) was found. However, iCa was poorly correlated with tCa (r = 0.023; P = 0.929). Interestingly, μCa was better correlated with cCa (r = 0.951; P < 0.001) than with iCa (r = 0.264; P = 0.290).
Mean ± SE values obtained for partitioning of calcium in samples of serum from 28 clinically normal horses.
| Variable | Absolute values (mmol/L) | % of tCa |
|---|---|---|
| tCa | 3.26 ± 0.06 | 100 |
| pCa | 1.55 ± 0.03 | 47.4 ± 0.9 |
| μCa | 1.71 ± 0.04 | 52.6 ± 1.1 |
| iCa | 1.58 ± 0.03 | 48.5 ± 0.7 |
| cCa | 0.13 ± 0.02 | 4.1 ± 0.9 |
Serum tMg concentration was 0.99 ± 0.04 mmol/L, and values for the magnesium fractions were as follows: iMg, 0.57 ± 0.02 mmol/L (58%); pMg, 0.33 ± 0.04 mmol/L (33%); and cMg, 0.09 ± 0.02 mmol/L (9%; Table 2). Based on the total protein concentration, the binding of magnesium to serum protein was 0.005 mmol of Mg/g of protein. As with calcium, tMg was well correlated with pMg (r = 0.823; P < 0.001) but not with iMg (r = −0.160; P = 0.527). Again, the ultrafiltrable fraction was better correlated with cMg (r = 0.959; P < 0.001) than with iMg (r = 0.470; P < 0.001).
Mean ± SE values obtained for partitioning of magnesium in samples of serum from 28 clinically normal horses.
| Variable | Absolute values (mmol/L) | % of tMg |
|---|---|---|
| tMg | 0.99 ± 0.04 | 100 |
| pMg | 0.33 ± 0.04 | 33.3 ± 4.2 |
| μMg | 0.66 ± 0.02 | 66.7 ± 2.7 |
| iMg | 0.57 ± 0.02 | 57.6 ± 1.7 |
| cMg | 0.09 ± 0.02 | 9.1 ± 1.9 |
Refrigeration of serum samples for up to 48 hours did not affect iCa values (Figure 1). However, samples that had been refrigerated for 120 hours had a small decline in iCa (from 1.58 ± 0.02 mmol/L to 1.53 ± 0.01 mmol/L; P = 0.04). The iMg was more affected by refrigeration, resulting in a time-dependent decrease in iMg that was significant after only 6 hours of storage. Although the decrease in magnesium values was significant, its magnitude was moderate (8% of baseline values at 120 hours). Freezing the samples for up to 1 month had no significant effect on either the iCa or iMg concentration.
Serum iCa and iMg concentrations as a percentage ofbaseline values (iCa [mean SE], 1.58 0.02 mmol/L; iMg, 0.58 ± 0.02 mmol/L) versus time in storage under refrigeration (4°C) or freezing (−20 °C) conditions (n = 14 samples). *Significantly (P < 0.05) different from baseline value (0 hours).
Citation: American Journal of Veterinary Research 67, 3; 10.2460/ajvr.67.3.463
Discussion
Fractionation of serum calcium and magnesium has been previously reported for horses. However, in their classical study, Holley and Evans19 did not differentiate fractions that compose the ultrafiltrate (ie, ionized and complexed). Thus, to our knowledge, prior to our study, reference values for either complexed calcium or magnesium in equine had not been determined. In addition, the percentage of ionized versus complexed fractions was unknown.
Results of our study allow establishment of reference range values for serum tCa concentrations of clinically normal horses.5 The percentage of pCa in equine serum obtained in our study (47.4 ± 0.9%) is slightly higher than values reported by Holley and Evans.19 However, our percentage of pCa is in good agreement with most human values obtained by ultrafiltration and ultracentrifugation.20–23 It is noteworthy that Holley and Evans19 also analyzed some samples of human serum to validate the ultrafiltration procedure and the values that they reported for pCa were also slightly lower than those found in other human studies.20–23 The slightly higher percentage of pCa found in our study may be the result of a greater ability of the ultrafiltration membrane to retain proteins. This contention is supported by the minute concentration of total proteins measured in the ultrafiltrate, < 0.4% of total protein content in the serum. Another factor that may influence the amount of calcium bound to proteins is the pH of the sample, which determines the equilibrium between pCa and iCa. To avoid exposure to air during centrifugation, a covering of mineral oil, which prevents the air exposure of the serum, was used by Holley and Evans.19 However, this system has been criticized because the head pressure of the mineral oil may cause removal of calcium from protein, therefore resulting in a lower pCa.24
The iCa values obtained in our study are also in good agreement with previous reports1,5 on iCa values in healthy horses. The iCa in equine serum represented slightly < 50% of tCa and > 90% of the μCa, whereas the cCa was 4% of the tCa and < 10% of the μCa. To our knowledge, no reports have been published on partitioning of μCa in equine serum. Data from other species indicate that complexed calcium is slightly higher in dogs, > 15% of the μCa,16 and > 20% of the μCa in humans.15 Thus, compared with other species, cCa seems to represent a smaller fraction in equine serum. However, it should be taken into account that because tCa values in horses are higher than in carnivores and humans, the absolute amount of serum cCa in horses is similar (0.13 ± 0.02 mmol/L in our study) to that of dogs (0.13 mmol/L)16 and humans (0.15 mmol/L).15 Measurement of the cCa fraction is of clinical interest because cCa has been shown to increase in certain diseases (eg, renal disorders).25,26
Information available regarding serum magnesium concentration in horses is more limited than that of calcium. The tMg values were within the reference range for horses.5,27 As with calcium, pMg was slightly higher than previously reported values.19 In agreement with a prior study,19 our data indicate that horses have less pMg than pCa, not only in absolute values (0.005 vs 0.024 mmol/g of protein) but also in percentage (33.3 ± 4.2% vs 47.4 ± 0.9%). Compared with other species, humans have lower (19%) pMg than horses.8 However, the relationship between pMg and μMg is similar in horses (33% and 67%, respectively) and in dogs (37% and 63%, respectively).18
Measurement of serum calcium and magnesium has important clinical applications in horses.1–6 Results of several studies indicate that for diagnostic purposes, measurement of the ionized fractions (iCa and iMg) is superior to quantification of serum tCa and tMg.6,11 In addition, as demonstrated by the lack of correlation between the total and ionized fractions, the total serum concentration is a poor estimator of the ionized value. Thus, it becomes clinically important to quantify iCa and iMg. However, most equine practitioners do not have ready access to ion-selective electrodes and may need to submit samples for analysis. Therefore, it is important to know the stability of iCa and iMg in samples of equine serum. Our results confirm previous findings indicating that iCa is fairly stable and that samples can be safely refrigerated without interfering with iCa values.28,29 To our knowledge, no previous data on iMg stability in equine serum are available. Information from other species indicates that iMg is stable in samples of human30 and canine31 serum. Our data reveal that although the changes are not prominent, iMg content significantly declines in refrigerated samples of equine serum. By contrast, no changes in either iCa or iMg values were found in samples frozen for up to 30 days. Thus, bearing in mind that in a clinical setting the storage conditions may be less perfect in comparison to our study (ie, complete anaerobiosis), we recommend that serum samples in which iMg cannot be promptly measured should be frozen.
In conclusion, our study reports the complete fractionation of serum calcium and magnesium in horses. In addition, our results indicate that serum iMg is less stable than serum; thus, when samples are not going to be analyzed promptly, freezing is preferable to refrigeration for sample storage.
| tCa | Total calcium |
| tMg | Total magnesium |
| iCa | Ionized calcium |
| iMg | Ionized magnesium |
| pCa | Protein-bound calcium |
| pMg | Protein-bound magnesium |
| cCa | Calcium complexed with weak acids |
| cMg | Magnesium complexed with weak acids |
| μCa | Ultrafiltrable calcium |
| μMg | Ultrafiltrable magnesium |
| PTH | Parathyroid hormone |
Vacutainer, Becton-Dickinson, Plymouth, England.
Nova Stat Profile Critical Care Xpress, Nova Biomedical, Waltham, Mass.
Calcium, Sigma Diagnostics, St Louis, Mo.
Magnesium, Sigma Diagnostics, St Louis, Mo.
Amicon Ultra-4, Millipore, Carrigtwohill, Ireland.
BioSystems Reagents & Instruments, Barcelona, Spain.
Bio-Rad protein assay, Bio-Rad Laboratories GmbH, Munchen, Germany.
Phosphorus inorganic, Sigma Diagnostics, St Louis, Mo.
Nichols Institute Diagnostics, San Juan Capistrano, Calif.
Immunodiagnostic Systems, Bolton, England.
References
- 1
Aguilera-Tejero E, Estepa JC, Lopez I, et al.Plasma ionized calcium and parathyroid hormone concentrations in horses after endurance rides. J Am Vet Med Assoc 2001;219:488–490.
- 2
Tennant B, Bettleheim P, Kaneko JJ. Paradoxic hypercalcemia and hypophosphatemia associated with chronic renal failure in horses. J Am Vet Med Assoc 1982;180:630–634.
- 3
Sevinga M, Barkema HW, Hesselink JW. Serum calcium and magnesium concentrations and the use of a calcium-magnesium-borogluconate solution in the treatment of Friesian mares with retained placenta. Theriogenology 2002;15:941–947.
- 4
Johansson AM, Gardner SY, Jones SL, et al.Hypomagnesemia in hospitalized horses. J Vet Intern Med 2003;17:860–867.
- 5↑
Toribio RE, Kohn CW, Chew DJ, et al.Comparison of serum parathyroid hormone and ionized calcium and magnesium concentrations and fractional urinary clearance of calcium and phosphorus in healthy horses and horses with enterocolitis. Am J Vet Res 2001;62:938–947.
- 6
Garcia-Lopez JM, Provost PJ, Rush JE, et al.Prevalence andprognostic importance of hypomagnesemia and hypocalcemia in horses that have colic surgery. Am J Vet Res 2001;62:7–12.
- 7
Pedersen KO. Determination of calcium fractions of serum. I. The separation of protein-bound and protein-free fractions bymeans of a simplified ultrafiltration technique. Scand J Clin Lab Invest 1969;24:69–81.
- 8↑
Altura BT, Altura BM. A method for distinguishing ionized, complexed and protein-bound Mg in normal and diseased subjects. Scand J Clin Lab Invest 1994;54(suppl 217):83–87.
- 9
Chattopadhyay N, Mithal A, Brown EM. The calcium-sensing receptor: a window into the physiology and pathophysiology of mineral ion metabolism. Endocrinol Rev 1996;17:289–307.
- 10
Zholos AV, Bolton TB. Effect of divalent cations on muscarinic receptor cationic current in smooth muscle from guinea-pig small intestine. J Physiol 1995;486:67–82.
- 11
Dart AJ, Snyder JR, Spier SJ, et al.Ionized calcium concentration in horses with surgically managed gastrointestinal disease: 147 cases (1988–1990). J Am Vet Med Assoc 1992;201:1244–1248.
- 12↑
Toffaletti J, Bowers GN Jr. The possible physiologic importance of calcium complexes: an opinion. Clin Chim Acta 1979;97:101–105.
- 13
Farese G, Mager M, Blatt WF. A membrane ultrafiltration procedure for determining diffusible calcium in serum. Clin Chem 1970;16:226–228.
- 14
Farese G, Mager M. Protein-free filtrates obtained by membrane ultrafiltration. Clin Chem 1970;16:280–281.
- 15↑
Moore EW. Ionized calcium in normal serum, ultrafiltrates, and whole blood determined by ion-exchange electrodes. J Clin Invest 1970;49:318–334.
- 16↑
Schenck PA, Chew DJ, Brooks CL. Fractionation of canine serum calcium, using a micropartition system. Am J Vet Res 1996;57:268–271.
- 17
Schenck PA, Chew DJ. Determination of calcium fractionation in dogs with chronic renal failure. Am J Vet Res 2003;64:1181–1184.
- 19↑
Holley DC, Evans JW. Determination of total and ultrafilterable calcium and magnesium in normal equine serum. Am J Vet Res 1977;38:259–262.
- 20
Loken HF, Havel RJ, Gordan GS, et al.Ultracentrifugal analysis of protein-bound and free calcium in human serum. J Biol Chem 1960;235:3654–3658.
- 21
Prasad AS, Flink EB. Effect of carbon dioxide on concentration of calcium in an ultrafiltrate of serum obtained by centrifugation. J Appl Physiol 1957;10:103–107.
- 22
Walser M. Ion association. VI. Interactions between calcium, magnesium, inorganic phosphate, citrate and protein in normal human plasma. J Clin Invest 1961;40:723–730.
- 23
Osmun JR. Total and ultrafiltrable calcium and magnesium in human plasma: improved method and normal values. Am J Med Technol 1967;33:448–458.
- 24↑
Wortsman J, Killam LI, Traycoff RB. A rapid method for the determination of ultrafilterable calcium in serum. J Lab Clin Med 1981;98:226–228.
- 25
Ledermann SE, Johnson A, Dillon MJ, et al.Serum intact parathyroid hormone and ionised calcium concentration in children with renal insufficiency. Pediatr Nephrol 1994;8:561–565.
- 26
Kelly PM, Bending MR, Barron JL. Comparison of serum calcium fractions during haemodialysis and peritoneal dialysis. Ann Clin Biochem 1987;24:585–590.
- 27
Stewart AJ, Hardy J, Kohn CW, et al.Validation of diagnostic tests for determination of magnesium status in horses with reduced magnesium intake. Am J Vet Res 2004;65:422–430.
- 28
Szenci O, Nemeth F, Stollar Z, et al.Effects of storage time and temperature on ionized calcium concentration in equine blood, plasma, and serum. J Am Vet Med Assoc 1994;204:1242–1244.
- 29
Schenck PA, Chew DJ, Brooks CL. Effects of storage onserum ionized calcium and pH from horses with normal and abnormal ionized concentrations. Vet Clin Pathol 1996;25:118–120.
- 30↑
Greenway DC, Hindmarsh JT, Wang J, et al.Reference interval for whole blood ionized magnesium in a healthy population and the stability of ionized magnesium under varied laboratory conditions. Clin Biochem 1996;29:515–520.
- 31↑
Unterer S, Lutz H, Gerber B, et al.Evaluation of an electrolyte analyzer for measurement of ionized calcium and magnesium concentrations in blood, plasma, and serum of dogs. Am J Vet Res 2004;65:183–187.
