Embedded Scribd iPaper - Requires Javascript and Flash Player
Concentrations of Free (Ionized) and Total Calcium and Magnesium
in Healthy Captive Asian Elephants (Elephas maximus) and
Effects of Sample Type and pH on measured Free Calcium
and Magnesium Concentrations
Aroch, I.,1* Larian, N.1 and Avni-Magen, N.2
1
2
Koret School of Veterinary Medicine, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot, 761001, Israel.
The Tisch Family Zoological Gardens in Jerusalem (The Biblical Zoo) P.O Box 898, Manhat, Jerusalem 91008, Israel.
* Corresponding author: Itamar Aroch DVM, DECVIM-CA (Internal Medicine), Koret School of Veterinary Medicine, The Hebrew University of
Jerusalem. P.O. Box 12, Rehovot, 761001, Israel. Tel: 972-3-9688556; Fax: 972-3-9604079. Email: itamar.aroch@mail.huji.ac.il
AB ST RAC T
The Asian elephant (Elephas maximus) is an endangered species, with an overall low reproduction rate in
captivity, and a long, 22-month gestation, mostly with a single calve. Calcium and magnesium are important
for the normal progression of gestation and parturition. This study measured blood total and ionized calcium
(tCa and iCa, respectively) and total and ionized magnesium (tMg and iMg, respectively) in four healthy,
captive Asian elephant cows in the Tisch Family Zoological Gardens, Jerusalem, every alternative month,
over a 1-year period, to establish their reference intervals and examine sample pH and sample type effects on
measured iCa and iMg concentrations. iCa and iMg were measured using an ion-selective electrode electrolyte
analyzer. Calcium and magnesium levels in diet samples were measured. The iCa:tCa and iMg:tMg ratios
were 0.44 and 0.73, respectively. Mean iMg concentrations in whole-blood, heparinized plasma and serum
were 0.58, 0.65 and 0.66 mmol/L, respectively. iCa and iMg concentrations in the three sample types were
highly correlated, with no sample type effect on measured iCa concentration, but significant effect on iMg
concentration, with significantly lower whole-blood levels vs. serum and plasma. Serum albumin and both
tCa and tMg concentrations positively correlated. Sample pH had no effect on measured iCa or iMg levels.
This study is the first to measure iMg in Asian elephants, and assess the effects of sample type and pH on the
results. It was concluded that different iMg reference intervals should be established for each sample type.
Keywords: Asian Elephant; Ionized Magnesium; Serum; Plasma; Whole Blood.
INTRODUCTION
The Asian elephant (Elephas maximus) is an endangered
species, with estimated worldwide population size in 2003
of 41,400 to 52,000 elephants, including 15,000 domesticated elephants in Asia, and 1,000 in zoos worldwide (1). In
Israel, there are 12 captive Asian elephants, and those held
in the Tisch Family Zoological Gardens Jerusalem (TFZGJ;
formerly the Jerusalem Biblical Zoo) are the only trained
elephants in Israel, and therefore, the only ones from which
routine blood samples can be obtained. Being an endangered
species, with a long gestation period (22 months), mostly
24
Aroch, I.
JUNE 2016.indb 24
with a single calf, successful gestation and parturition are
essential. Zoos worldwide encounter difficulties in mating
elephants, resulting in low overall reproduction rates (2).
The aggressive (musth) period of elephant bulls makes
their handling extremely dangerous for the staff, limiting
the number of adult elephant bulls in zoos (2-4). Other
reasons for low reproduction rates include lack of breeding
programs, old, non-cyclic females, high infant mortality rate
and diseases (e.g., endotheliotropic elephant herpes virus
infection, a common cause of fetal death and stillbirth in captive Asian elephants), which even in the relatively uncommon
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
31/05/2016 13:35:51
Research Articles
events of successful mating, both gestation and parturition
are frequently complicated (5, 6). Thus, understanding the
pathogenesis of dystocia in the elephant is important in its
prevention.
Calcium and magnesium are important divalent cations,
with significant roles in the normal progression of gestation and parturition (8, 9). Total serum magnesium (tMg)
and calcium (tCa) are distributed among three major fractions: protein-bound, complexed with small anions (e.g.,
phosphates, bicarbonate and lactate) and free-ionized (iMg
and iCa, respectively), which are the active forms (10-12).
Magnesium is mostly an intracellular cation, less tightly regulated compared to calcium, with a wider physiological range.
Its roles include monitoring of calcium pumps, ATPases,
neural and cardiac muscle function, and as a cofactor in
several enzymatic reactions (12-14). It also has important
roles in fetal development (15). During normal pregnancy in
women, serum magnesium levels decline, while severe hypomagnesaemia is associated with abortion, and dietary magnesium deficiency plays an important role in the pathogenesis
of eclampsia and hypertension (15, 16). Hypomagnesaemia
in ruminants and horses might lead to recumbency, ataxia
and other neurological signs (10). Total magnesium (tMg)
concentrations in Asian elephants were previously reported
(8), but their normal iMg levels are unknown.
Calcium, mostly an extracellular cation, is kept within a
narrow physiological range and has important roles in muscle
contractility, skeletal structure and neural function. It is an
intracellular messenger and cofactor in numerous enzymatic
and coagulation reactions (17). Hypocalcaemia leads to post
parturient paresis in lactating dairy cows, to flaccid paralysis
in pregnant ewes, and is one of the major disorders leading to
dystocia in ruminants (10). Pregnant mares and women have
lower iCa than non-pregnant counterparts, probably due to
the growing fetus calcium demands (10, 13, 17, 18). It is suggested, but not confirmed, that captive Asian elephants have
low calcium levels, which might play a role in the pathogenesis
of dystocia, by causing initial progression failure of the second
stage of labor (19). Several, partially documented reports of
dystocia in captive Asian elephants worldwide are available,
some of which responded to intravenous calcium therapy (8).
Nevertheless, information of calcium metabolism in
Asian elephants is limited. The mean absorption rate of
dietary calcium is 60%, irrespective of its dietary level. The
digestive physiology of elephants resembles that of horses in
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
JUNE 2016.indb 25
the facts that dietary crude fiber concentration influences dry
matter digestibility, the presence of pelleted feeds influences
calcium absorption, and concentrate feeds lead to increased
fecal volatile fatty acids. In both species, being hindgut fermenters, intestinal calcium absorption is higher compared
to ruminants and in both, excess calcium is excreted in the
urine (6, 8, 20). Unlike ruminants, they are almost completely
dependent on calcium absorption rather than on its skeletal
mobilization (6). A clinical study, performed in Asian elephants in the Rotterdam Zoo, showed that following a high
calcium diet, serum tCa levels increased, while its urinary
excretion remained unchanged, suggesting prior sub-clinical
hypocalcaemia. The authors concluded that captive Asian
elephants should be fed a calcium-rich diet, especially around
parturition. That study has suggested that the reference interval (RI) concentrations of plasma tCa and iCa in captive
Asian elephants should be around 3.6 mmol/L, and 1.25
mmol/L, respectively (8, 21). Another study investigated the
effects of dietary calcium or cholecalciferol supplementation
on serum calcium concentration in elephants. It was concluded, that in Western Europe, captive Asian elephants possibly
suffer from summer-associated subclinical hypocalcemia, and
that the advisable dietary calcium and cholecalciferol levels
should be higher than the currently used guidelines (22).
According to the International Species Information System
(ISIS), mean serum tCa and tMg concentrations in captive
Asian elephants are 2.64 mmol/L (10.6 mg/dL) and 0.893
mmol/L (2.17 mg/dL), respectively, based on measurements
in 216 and 105 healthy Asian elephants, respectively (23). The
RIs or mean concentrations of iMg are not available.
During parturition, calcium levels should be closely
monitored, and if serum tCa decreases <2.5 mmol/L, or if
iCa falls < 1.2 mmol/L, IV or oral calcium should be administered (6). The uterine effect of such calcium therapy
should be confirmed by rectal palpation (i.e., increase in
uterine contractibility is considered a desired response) and
serum calcium measurement (24). Due to the limited data
on the physiological and periparturient serum calcium and
magnesium levels in wild Asian elephants, it is currently unclear whether the serum levels measured in captive elephants
are a dominant factor in the pathogenesis of dystocia. It is
of great importance to establish RIs for both cations, and
to routinely monitor zoo elephants, especially cows, before
and during parturition, and correct abnormalities in order to
improve their reproduction rate.
Calcium and Magnesium in Asian Elephants
25
31/05/2016 13:35:51
Research Articles
The Asian elephant’s nutrition in the wild is based on
plants and trees, while in captivity it mainly includes hay and
pellets, with major differences in elephant diets between zoos
(25). No commercial food designed for elephants is available, except for mineral additives; therefore their nutrition in
captivity, especially regarding calcium and magnesium levels,
might be incomplete and imbalanced, especially during stressful periods, such as gestation, parturition and early lactation.
Calcium absorption is significantly decreased by pelleted feed
addition, in both equids and elephants (26). Most zoos apply
the equine nutritional requirements for elephants. However,
experiments have shown that elephants have much higher
gastrointestinal passage rates compared to horses, thus leading
to a lower absolute digestibility coefficient (20). According
to the Nutrient Requirements Council (NRC) guidelines
for horses, currently serving as elephant nutrition guidelines,
the minimum calcium feed recommended concentrations for
maintenance and late pregnancy, respectively, are 0.3% and
0.5%, and the minimum one for magnesium is 0.1% (26, 27).
Ionized Mg and Ca (iCa and iMg, respectively) can be
measured in whole blood, serum or heparinized plasma, using ion selective electrode analyzers. Their concentrations
decrease with increase in sample pH, in vivo or in vitro,
and vice versa (28-31). Therefore, anaerobic blood sample
handling, using vacuum-sealed tubes is recommended, to
prevent gas-exchange (16, 18).
MATERIALS AND METHODS
Animals and diet
The study included four Asian elephant cows (aged 21 to 28
years) held at the same facility and fed the same diet, in the
TFZGJ, the only zoo in Israel that holds trained elephants,
from which blood samples were obtained every alterative
month over one year. At the beginning of this study, one cow
was in mid-gestation, and was expected to give birth during
the study period. Another, had been lactating over the preceding 3.5 years, and the other two had never been pregnant.
Diet samples were obtained during the trial period for the
measurement of calcium and magnesium concentrations.
Each cow was fed 3 meals daily consisting of wheat
hay (35 kg), dairy cow pellets, (5 kg; 16% protein), pruned
branches, mainly carobs (Ceratonia silique) (10-15 kg), vegetables (20 kg) and one tablet (138 g) of equine vitamin and
mineral supplement (Salvana Tiernahrung GmbH, Klein
26
Aroch, I.
JUNE 2016.indb 26
Offenseth-Sparrieshoop, Germany) containing 9.5% calcium and 1.5% magnesium. The wheat hay contained 0.58%
calcium and 0.1% magnesium in dry matter. The pellets
contained 1.3% calcium and 0.24% magnesium in dry matter.
The total daily calcium and magnesium intake, excluding the
pruned branches and vegetables, were 280 g (0.7%) and 47
g (0.12%), respectively. The diet analysis was carried out by
the Forage and Feed Laboratory, E.H. Smoler Consulting
and Research for Agricultural Science Ltd., Yoav Business
Center, Ree’m Junction, Israel.
Blood samples and laboratory methods
Blood was obtained from the saphenous or cephalic veins.
In order to minimize gas exchange and resultant in vitro
sample pH changes, samples were collected using vacuumsealed tubes (Vacutainer, BD Diagnostics and Preanalytical
Systems, Franklin Lakes, NJ, USA), and were analyzed within
four hours from collection. Blood from each elephant was
collected in three sealed lithium-heparin tubes and two plain
serum tubes with separators. The heparin tubes were used for
whole-blood and plasma iCa and iMg measurements. One
was centrifuged within 30 minutes from collection, and the
plasma was immediately transferred to sealed Eppendorf
tubes, filled to their full capacity to prevent gas exchange,
and analyzed within four hours from collection. The plain
tubes were allowed to clot, centrifuged, and harvested serum
was handled as described above. One plain separator tube
was used for tCa and tMg concentration measurement, and
the other was used for iCa and iMg measurement. All tubes
were kept chilled on ice pending analysis.
Electrolytes were measured using an ion selective electrode (ISE) electrolyte analyzer (Nova 8, Nova Biomedical,
Waltham, MA, USA; at 37°C). Serum biochemistry analysis,
including tCa and tMg, phosphorus and albumin, was performed by wet chemistry autoanalyzers (Vitros 5,1 FS, Ortho
Clinical Diagnostics, NJ; Cobas Integra 400 Plus, Roche,
Mannheim, Germany; at 37°C).
Statistical analysis
The distribution pattern of continuous variables was assessed
using the Kolmogorov-Smirnov test. Pearson’s correlation
coefficient was used to evaluate the correlation between two
continuous variables (e.g., pH and iCa). Intra-class correlation (both definitions of consistency and absolute agreement)
was calculated in order to evaluate the association between
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
31/05/2016 13:35:51
Research Articles
Table 1: Concentrations of ionized calcium and magnesium in whole-blood, plasma and serum, total serum calcium and magnesium in four
captive Asian elephant cows in the Jerusalem Biblical Zoo over one year
 
Sample
Mean
SD
Minimum
Maximum
Ionized calcium(mmol/L)
Whole
blood
1.12
0.18
0.65
1.34
tCa
iCa:tCa
(mmol/L)
Plasma
Serum
Serum
Serum
Reporteda
tCa
(mmol/L)
1.11
0.23
0.45
1.38
1.16
0.20
0.68
1.39
2.70
0.39
1.49
3.17
0.44
0.12
0.4
0.56
2.65
0.2
NA
NA
Ionized magnesium (mmol/L)
Whole
blood
0.58
0.10
0.31
0.72
tMg
Plasma
Serum
Serum
Reported
tMga
(mmol/L)
0.65
0.09
0.47
0.82
0.66
0.08
0.46
0.81
0.91
0.12
0.54
1.1
0.85
0.2
NA
NA
tCa = total calcium; iCa = ionized calcium; tMg = total magnesium; iMg = ionized magnesium; a = ISIS physiological reference values (23); NA = not available.
variable measurements in plasma, serum, and whole blood.
Assessment of differences between analyte concentrations
in the different sample types was carried out using repeated
measures ANOVA, applied twice, with, and without elephant
effect. The Greenhouse-Geisser test was applied to test the
effect within subjects. If differences between sample types
were statistically significant in the ANOVA, Student’s paired
t-tests, with Bonferroni’s correction of the significance level
were used to compare sample type pairs at each time-point.
All tests applied were two-tailed, and a P value ≤0.05 was
considered statistically significant.
RESULTS
All measured analytes and calculated variables were normally
distributed. The mean iCa, tCa, iMg and tMg concentrations,
the calculated iCa:tCa and iMg:tMg in whole-blood, plasma
and serum in the four elephant cows over the study period
are presented in Table 1. The mean sample pH of wholeblood, plasma and serum over the study period were 7.38 (SD
0.05; range 7.29-7.51), 7.49 (SD 0.03; range 7.43-7.57) and
7.51 (SD 0.04; range 7.45-7.62), respectively. Mean serum
albumin, total protein (TP) and phosphorus concentrations
over the study period were 3.04 (SD 0.4) g/dL, 8.1 (SD 0.52)
g/dL and 4.41 (SD 0.83) mg/dL, respectively.
The intra-class correlation coefficients using consistency
definition (assuming absence of interaction, because it was
not estimable otherwise) for iMg, iCa and pH in whole
blood, serum and plasma samples were r=0.891, r=0.790 and
r=0.632, respectively. The respective intra-class correlation
coefficients using absolute agreement definition (assuming
absence of interaction, because it is not estimable otherwise)
were r=0.815, r=0.791, and r=0.230, respectively.
Whole blood and plasma iMg were strongly correlated
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
JUNE 2016.indb 27
(r=0.826, P<0.005; n=20), with a weaker correlation between
whole blood and serum iMg (r=0.690, P=0.001; n=20) and
between plasma and serum iMg (r=0.678, P=0.001; n=20).
For iCa, the strongest correlation was between plasma and serum iCa (r=0.808, P<0.005; n=21), followed by that between
whole blood and serum iCa (r=0.454, P=0.039; n=21) and between whole blood and plasma iCa (r=0.395, P=0.077; n=21).
There were significant moderate correlations between whole
blood and plasma pH (r=0.584, P=0.005; n=21) and between
plasma and serum pH (r=0.584, P=0.005; n=21). There was
no significant correlation between blood and serum pH.
There were significant strong correlations between albumin and tCa (r=0.877, P<0.001) and between albumin
and tMg (r=0.869, P<0.001) (Figure 1) and a moderate
significant correlation between albumin and phosphorus
(r=0.536, P=0.012). There were no other significant correlations between measured analytes.
When analyzing the sample type and elephant effects on
the results using the Greenhouse-Geisser’s test, there was a
significant sample type effect on pH (P<0.001) and on iMg
concentration (P=0.00001), but no significant sample type
effect on iCa (P=0.367).There were no elephant effects on
pH or iMg and iCa concentrations (P=0.413, P=0.840 and
P=0.675, respectively). When examining the elephant effect
on the concentrations of total protein, albumin and phosphorus, the only significant difference was in mean serum total
protein concentration (P=0.001).
Based on paired Student’s t-test, significant differences
were found between whole-blood and heparinized plasma
pH (P<0.001) and between whole-blood and serum pH (P
<0.001), with no significant difference between plasma and
serum pH (P=0.038). Sample pH was consistently lower in
whole-blood compared to plasma and serum.
There were significant iMg concentration differences beCalcium and Magnesium in Asian Elephants
27
31/05/2016 13:35:51
Research Articles
study. It did not proceed beyond the second stage
of labor, while the calf was still within the birth
canal and did not respond to oral and intravenous
calcium treatment (Table 2). A vaginal vestibulotomy was performed, and attempts were made to
pull the calf out, but these were unsuccessful, as it
was too large. This was followed by a fetotomy procedure, a procedure performed only twice before
in elephants, and was successful only once (19).
Only one third of the calf was evacuated, while
the rest remained in the birth canal. Three days
later, despite intensive supportive care, the elephant
suddenly died. Necropsy was not conducted, so the
cause of death was undetermined. Blood samples
obtained during parturition were analyzed only
for total calcium (Table 2) due to the urgency
of events, and were not included in the analyses.
DISCUSSION
This study followed over a period of one year iCa,
iMg concentrations and pH in three different
sample types every alternative month in four capFigure 1: Changes in mean total calcium (A) and total magnesium (B) vs. serum
tive, healthy adult Asian elephant cows in different
albumin over the study period in four captive Asian elephants in the Jerusalem
physiological states. To the best of our knowledge,
Biblical Zoo. Note the parallel changes in both total calcium and total magnesium
this is the first study to report iMg concentrations
with changes in serum albumin. There were significant (P<0.001), positive strong
in elephants in general, and particularly in Asian
correlations between changes in total calcium and total magnesium vs. serum
albumin (r=0.877 and r=0.869, respectively).
elephants. Because these results are novel, a comparison with previously published data regarding
tween whole-blood and heparinized plasma (P<0.0001) and
iMg could not be performed. These results may serve as
between whole-blood and serum (P<0.0001), with no sigreference intervals for future studies, although, these results
nificant difference between plasma and serum iMg (P=0.69).
should be applied cautiously, due to the limited number of
iMg concentrations were consistently lower in whole-blood
elephants included herein. Additionally, the present study
compared to heparinized plasma or serum. There were no
includes novel findings regarding the influence of the sample
significant differences between iCa concentrations in differtype (e.g., whole-blood, heparinized plasma and serum) on
ent sample types.
iCa and iMg measurements, and their associations of sample
There was a significant moderate correlation between
pH in Asian elephants.
whole blood iCa and iMg (r=0.699, P=0.001), and no correlaThe mean serum tCa concentration measured presently
tions between whole blood pH and iMg or iCa. There were
(2.7±0.37 mmol/L) is similar to findings in 216 healthy
significant moderate correlations between plasma iCa and
Asian elephants (2.64±0.22 mmol/L) (23), while mean
iMg (r=0.575, P=0.006), and between serum iCa and iMg
plasma iCa concentration herein (1.11±0.23 mmol/L) is
(r=0.567, P=0.007), and a weak one between plasma pH and
somewhat higher compared to previously published data
(0.93±0.11 mmol/L) (8, 22, 32). This difference appears
iMg (r=-0.459, P = 0.036). There was no correlation between
clinically insignificant. Nevertheless, both iCa concentraplasma pH and iCa, and between serum pH and iCa or iMg.
tion herein and in the previously published study are lower
The pregnant elephant cow included in the present study
than its recommended target concentration of 1.25 mmol/L
had started labor five months from the beginning of the
28
Aroch, I.
JUNE 2016.indb 28
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
31/05/2016 13:35:52
Research Articles
Table 2: Serum total calcium concentration, treatment and events in an Asian Elephant cow with dystocia at the Jerusalem Biblical Zoo
14.7.09
16.7.09
19.7.09
Serum calcium
(mmol/L) [mg/dL]
2.4 mmol/L [9.6 mg/dL]
2.5 mmol/L [10 mg/dL]
2.4 mmol/L [9.6 mg/dL]
20.7.09
2.6 mmol/L [10.4 mg/dL]
21.7/09
22.7.09
23.7.09
24.7.09
25.7.09 (16:00)
2.6 mmol/L [10.4 mg/dL]
2.7 mmol/L [10.8 mg/dL]
2.6 mmol/L [10.4 mg/dL]
26.7.09 (05:20)
2.42 mmol/L [9.7 mg/dL]
26.7.09 (22:30)
27.7.09
28.7.09 (14:30)
2.28 mmol/L [9.1 mg/dL]
Date*
2.63 mmol/L [10.5 mg/dL]
2.25 mmol/L [9.0 mg/dL]
Calcium treatment and comments
4 Salvana horse pills in the food
4 Salvana horse pills in the food
4 Salvana horse pills in the food
4 Salvana horse pills in the food; Ca-borogluconate (2L, 23% solution) PO.
Serum progesterone concentration decreased.
4 Salvana horse pills in the food;Ca-borogluconate (2L, 23% solution) PO
4 Salvana horse pills in the food;Ca-borogluconate (2L, 23% solution) PO
4 Salvana horse pills in the food;Ca-borogluconate (2L, 23% solution) PO
4 Salvana horse pills in the food;Ca-borogluconate (2L, 23% solution) PO
4 Salvana horse pills in the food; Ca-borogluconate (5L, 23% solution) PO
At 06:00: Ca-borogluconate IV (800 mL; 23% solution)
At 12:30: Ca-borogluconate (2L, 23% solution) PO;4 Salvana horse pills in the food
At 23:00: Ca-borogluconate IV (400 mL; 23% solution)
At 07:10: Ca-borogluconate IV (400 mL; 23% solution)
Vestibulotomy**
* when the time is mentioned, it refers to the time in which the blood sample was obtained;
** no further measurements of serum calcium were made after the vestibulotomy.
in Asian elephants, fed a calcium-rich diet (8). It has been
recommended that when iCa level decrease to <1.2 mmol/L
during parturition, IV calcium should be administered in
order to maintain these iCa levels, a calcium-rich diet should
be fed at all times, particularly around parturition. Avoiding
pellets in the diet might improve calcium absorption in
elephants (8, 26). Since most zoos (including the TFZGJ)
include pellets in elephant diets, and since blood calcium
concentrations recorded herein and previously are below its
target concentration, possibly, such elephant diets in zoos fail
to meet physiologic calcium demands. A summer-associated
hypocalcemia has been suggested in captive Asian elephants
in Western Europe as well (22), and may play a role in the
high incidence of dystocia among captive Asian elephants.
If so, increasing dietary calcium content in captive Asian
elephant diets should be considered to maintain health and
prevent the potential effects of subclinical hypocalcemia on
reproduction. iCa was unfortunately not measured in the
elephant cow that had dystocia during this study. However,
total serum calcium concentration range, measured during
parturition in this cow, was 2.275 to 2.7 mmol/L, despite
oral and intravenous calcium administration throughout
parturition (Table 2). Given a calculated average iCa:tCa
ratio of 0.44, an iCa concentration range of 1.001 to 1.188
mmol/L during parturition may be assumed, which is below
the desired concentration of 1.25 mmol/L (8). This supports
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
JUNE 2016.indb 29
the presence of hypocalcemia in this cow during dystocia,
but cannot prove a direct cause and effect between this hypocalcemia and dystocia. Therefore, the present study, as in
previous ones (8, 22), did not conclusively prove a direct association between low blood Ca concentrations and dystocia
in elephants.
The serum iCa:tCa ratio in this study (0.44) is higher
compared to previous findings (0.32) (8), despite the higher
iCa concentration in the present study, and obviously higher
results compared to the tCa concentration in the previous
one. Since both ratios are lower than the ratio reported in
horses (0.52) (18), this questions whether the current nutrient
calcium demands for elephants, that are based on recommendations in equines, are adequate for elephants, and whether
such species differences warrant devising elephant-specific
dietary values. If the requirement in horses is to be applied in
elephants, it seems that iCa concentration, measured herein
and previously (8), are below optimal values.
Our results show significant strong correlations between
iCa concentration among all three sample types, for both
consistency and absolute agreement, with no significant
sample type effect on measured iCa, suggesting that practically, any sample type of the three can be used to reliably
measure iCa, accurately and interchangeably. These findings
are not consistent with previous results which showed higher
iCa concentrations in serum (20 samples; three elephants)
Calcium and Magnesium in Asian Elephants
29
31/05/2016 13:35:53
Research Articles
compared to heparinized plasma (25 samples; three elephants) (8). In horses, iCa concentration was also higher in
serum compared to heparinized plasma and whole-blood,
attributed to significantly higher sample pH in heparinized
plasma compared to serum (18). The reasons for this discrepancy are unclear, and it might have resulted from sample pH
differences between this study and the previous ones.
The present study’s mean serum tMg concentration is
similar to previous findings in 105 healthy Asian elephants
(0.893±0.08 mmol/L) (23). In the present study, the correlations between iMg concentrations in the three sample
types were strong and significant, and stronger than those
concerning iCa concentration. The iMg oncentration range
herein is narrower than that of iCa. However, in contrast
with iCa, significant sample type effects were noted on iMg
concentrations as demonstrated by pair-wise analyses, and
these were consistently lower in heparinized whole-blood,
in agreement with previous findings reported in horses (18).
There are several potential explanations for these results.
First, whole-blood and plasma samples were obtained with
heparin as the anticoagulant, which binds Mg ions, thereby
lowering their free fraction (33, 34). Second, erythrocytes
and platelets, present in whole blood, might interfere with
iMg analysis, through their influence on the electric potential
of the ISE analyzer (33, 34). Third, platelets contain high
intracellular magnesium, which is released during clot formation ands ample centrifugation in serum harvesting. These
findings are not consistent with those reported in humans,
in which no sample type effect on measured iMg concentrations was noted (35). It is unknown if platelet magnesium
levels in humans are lower compared to those in horses and
elephants, thus warranting further studies. Nevertheless, the
iMg concentration differences recorded, albeit statistically
significant, are probably clinically insignificant.
The pH levels were significantly and consistently correlated, albeit only moderately, among the three sample
types. The sample pH was consistently lower in whole-blood
samples, similar to iMg and iCa concentrations, while pH
was highest in serum samples. This is surprising, because
previous studies have shown that a rise in pH leads to decreased iMg and iCa concentrations, and vice versa, since
in alkaline solution calcium and magnesium ion binding to
plasma protein increases, thereby lowering their free (ionized)
form concentration (28-31). The higher pH levels in serum
samples, compared to whole blood and plasma likely resulted
30
Aroch, I.
JUNE 2016.indb 30
from some CO2 evaporation during harvesting the serum,
despite the strict protocol used to minimize it, which may
have resulted in an increase in sample pH (36).
A previous study has suggested a theory to explain the
lower whole-blood pH levels compared to serum and heparinized samples pH. According to this theory, in whole-blood
samples, a fraction of CO2 is associated with erythrocytes,
and therefore evenly distributes within the collection tube
similarly to physiological conditions. In contrast, when
whole-blood is centrifuged, erythrocytes are packed, and
CO2 is trapped with the erythrocytes, and thus the supernatant plasma (or serum) CO2 level is relatively lower (18). In
addition, in vitro metabolism of glucose by leukocytes and
platelets in whole-blood sample might have contributed to
the formation of CO2, thereby lowering the sample pH (35).
Another possible explanation is post sampling in vitro changes, mainly CO2 evaporation (from plasma and serum samples)
during separation, despite the precautions taken to avoid
such changes. Possibly, some evaporation is inevitable (36).
There were significant correlations between concentrations of iCa and iMg in the three different sample types,
and mostly in whole-blood samples. In contrast, sample pH
did not correlate with iCa or iMg concentrations in all three
sample types. This may be explained by other more dominant
factors, mentioned above, influencing the ions concentration (e.g., effects of heparin, platelets and erythrocytes).
Conversely, significant correlations between sample pH and
iCa concentration in serum and heparinized plasma were
noted previously in Asian elephants (8, 31), with a much
stronger correlation in serum samples compared to heparinized plasma samples (r=0.819 vs. r=0.496, respectively)
(8). A significant, albeit weak negative correlation between
sample pH and iCa concentration was noted in horses as well
between sample pH and iMg levels (18). Our findings are in
agreement with the latter.
Mean serum albumin and total protein and phosphorus
concentrations in this study are similar to previous findings
(32±5 g/L, n=206; 81±8 g/L, n=209 and 1.55±0.39 mmol/L
[4.8±1.2 mg/dL]; n=201) (23). We did find a significant
correlation between serum tCa and serum albumin, consistent with previous findings in horses (18), although in the
present study the correlation is stronger (r=0.877 vs. r=0.44,
respectively). This is probably because calcium strongly binds
to serum proteins (about 45%), mainly albumin (11). In this
study however, there was no correlation between serum tCa
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
31/05/2016 13:35:53
Research Articles
and serum total protein, in contrast to a weak correlation
described in horses (18). In this study there was also a significant correlation between serum tMg and albumin, in contrast
to horses (18), likely due to strong magnesium binding to serum proteins (about 30%), mainly albumin (11). Surprisingly,
there were no significant correlations between phosphorus
and both tCa and tMg, although physiologically, phosphorus
and calcium are closely regulated, and as dietary phosphorus
influences intestinal calcium absorption, and serum phosphorus concentrations influence calcium resorption from bones
and absorption from the intestine. Hyperphosphatemia can
induce hypocalcaemia, and hypophosphatemia can induce
hypercalcemia. Phosphorus also associates with magnesium;
high dietary phosphorus inhibits magnesium absorption in
the small intestine of monogastric animals (11).
There were no elephant effects on pH levels and iCa or
iMg concentrations, supporting that iMg concentration differences resulted only from sample type differences and that
the analytical method is not affected by individual animals.
Although this study included only four elephants, these were
in different physiological states (pregnancy, lactation and
non-pregnancy), suggesting that the physiological status has
no effect on measured pH, iCa and iMg.
The major limitation of this study is the small number
of elephants included, that has been reduced further in the
mid-study period, due to death of one elephant cow during
parturition. However, this limitation is common worldwide
in studies of captive elephants (8, 22, 32), and is inevitable,
as in Israel, with only 12 Asian elephants in the country, of
which only the four studied here were trained, allowing blood
collection. Nevertheless, in order to partially overcome this
limited number of elephants included, these were followed
over a 12-months period. Thus, the total number of samples
analyzed in the study is actually relatively high. Second,
although most of the diet was unchanged during the study
period, in some, dietary components changes were made over
the study period, and this probably resulted in changes in
their ingredients, precluding exact calculation of dietary calcium, magnesium and phosphorus at all times. Thus, potential
changes in their dietary levels might have affected their blood
concentrations, which could not be assessed in this study,
thereby limiting our interpretation and recommendations.
Third, the ISE electrolyte analyzer used in this study was
designed for human blood; its performance in elephant blood
has never been described or validated. Species differences
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
JUNE 2016.indb 31
might have affected the results. Nevertheless, the fact that
intra-assay correlations were high and significant, suggested
that this factor is probably unimportant. Since this study is
the first to examine the iMg levels in Asian elephants, our
results could not be compared to an established reference
interval or any other previous results. Lastly, during the unsuccessful parturition of the pregnant elephant, in the middle
of the study period blood samples for iCa, iMg and pH were
not obtained, due to the emergency nature of events, which
would have been important to our understanding changes in
these over this crucial period of dystocia.
In conclusion, this is the first study blood iMg and
iMg:tMg ratio in elephants, and the first to compare the sample type and pH effects on iCa and iMg levels in elephants.
Sample type affected iMg concentration, necessitating establishing different reference intervals for its concentration
in whole-blood, plasma and serum. The sample type did not
affect iCa levels, suggesting that a single reference interval
could be applied for all three sample types. However, due to
the small number of elephants and the fact that they were in
different physiological states, these results should be applied
cautiously. Additionally, this study showed no sample pH effect on iMg or iCa levels, in contrast with results in humans.
ACKNOWLED GEMENTS
We would like to thank Prof. Jacob Lavy, Head, Department of Heart
Transplantations, “Sheba” Hospital, Tel-Hashomer, Israel and Prof.
Shaul Dollberg, Head, Department of Neonatology, Sorasky Medical
Center, Tel-Aviv, Israel, for their help in some of the analyses, as well
as the Jerusalem Biblical Zoo elephants and their keepers.
REFERENCES
1. International Union for Conservation of Nature red list of
threatened species. Available athttp://www.iucnredlist.org/
details/7140/0;Access date: 30.01.16.
2. Taylor, V.J. and Poole, T.B.: Captive breeding and infant mortality
in Asian elephants: A comparison between twenty western zoos
and three eastern elephant centers. Zoo Biol. 17:311-332, 1998.
3. Brown, J.L., Somerville, M., Riddle, H.S., Keele, M., Duer, C.K.
and Freeman, E.W.: Comparative endocrinology of testicular,
adrenal and thyroid function in captive Asian and African elephant
bulls. Gen. Comp.Endocrinol. 151:153-162, 2007.
4. Sanyathitisaeree, P., Yartbantoong, N., Thongthipsiridej, S. and
Theeraphan, W.: Elephant Health Problems: An accumulative
case report from the Kasetsart University Veterinary Teaching
Hospital – Kamphaengsaen: EU-Asia Link Project Symposium
“Managing the Health and Reproduction of Elephant Populations
in Asia”. pp. 41-48, 2007.
Calcium and Magnesium in Asian Elephants
31
31/05/2016 13:35:53
Research Articles
5. Hermes, R., Goritz, F., Streich, W.J. and Hildebrand, T.B.: Assisted
reproduction in female rhinoceros and elephants – current status
and future perspective. Reprod. Domest. Anim. 42:33-44, 2007.
6. Hermes, R., Saragusty, J., Schaftenaar, W., Goritz, F., Schmitt,
D.L. and Hildebrandt, T.B.: Obstetrics in elephants. Theriogenol.
70:131–144, 2008.
7. Almonte, R.A., Heath, D.L., Whitehall, J., Russell, M.J., Patole,
S. and Vink, R.: Gestational magnesium deficiency is deleterious
to fetal outcome. Biol. Neonate. 76:26-32, 1999.
8. Van Der Kolk, J.H., Leeuwen, J.P.T.M., Belt, A.J.M., Schaik,
R.H.N. and Schaftenaar, W.: Subclinical Hypocalcaemia in Captive
Asian Elephants (Elephas maximus). Vet. Rec. 162:475-478, 2008.
9. Elin, R.J.: Assessment of magnesium status. Clin. Chem. 33:
1965-1970, 1987.
10. McCue, P.M. and Troedsson, M.H.T., Alterations in Sexual Function, In: Large Animal Internal Medicine, 3rd edition. Smith, B.P.
ed. St. Louis, Mosby, pp. 217-227, 2002.
11. Stockham, S.L. and Scott, M.A.: Calcium, Phosphorus, Magnesium
and their Regulatory Hormones, In: Fundamentals of Veterinary
Clinical Pathology, 2nd ed.Ames, Blackwell, pp. 594-625, 2008.
12. Elin, R.J.: Assessment of magnesium status. Clin. Chem. 33:19651970, 1987.
13. Altura, B.M.: Introduction: importance of Mg in physiology and
medicine and the need for ion selective electrodes. Scand. J. Clin.
Lab. Invest. 54: (Suppl. 217), 5-9, 1994.
14. Elin, R.J.: Magnesium: the fifth but forgotten electrolyte. Am. J.
Clin. Pathol. 102:616-622, 1994.
15. Almonte, R.A., Heath, D.L., Whitehall, J., Russell, M.J., Patole,
S. and Vink, R.: Gestational magnesium deficiency is deleterious
to fetal outcome. Biol. Neonate. 76:26-32, 1999.
16. Saris, N.E.L., Mervaala, E., Karppanen, H., Khawaja J.A. and
Lewenstam, A.: Magnesium - an update on physiological, clinical
and analytical aspects. Clin. Chim. Acta. 294:1-26, 2000.
17. Rosol, T.J. and Capen, C.C.: Calcium-regulating hormones and
diseases of abnormal mineral (calcium, phosphorus, magnesium)
metabolism, In: Clinical Biochemistry of Domestic Animals, 5th
ed., Kaneko, J.J., Harvey, J.W., Bruss, M.L. eds. San Diego, Academic Press, pp. 619-702, 1997.
18. Berlin, D. and Aroch, I.: Concentrations of ionized and total magnesium and calcium in healthy horses: effects of age, pregnancy,
lactation, pH and sample type. Vet. J. 181:305-311, 2008.
19. Thitaram, C., Pongsopawijit, P., Thongtip, N., Angkavanich, T.,
Chansittivej, S., Wongkalasin, W., Somgird, C., Suwankong,
N., Prachsilpchai, W., Suchit, K., Clausen, B., Boonthong, P.,
Nimtrakul, K., Niponkit, C., Siritepsongklod S., Roongsri, R. and
Mahasavankul, S.: Dystocia following prolonged retention of a
dead fetus in an Asian elephant(Elephas maximus). Theriogenol.
66:1284-1291, 2006.
20. Clauss, M., Loehlein, W., Kienzle, E. and Wiesner, H.: Studies on feed digestibilities in captive Asian elephants (Elephas
maximus). J. Anim. Physiol. Anim. Nutrition 87:160-173,
2003.
21. Lihong, W., Liu, L., Qian, H., Jinguo, Z. and Li, Z.: Analysis of
nutrient components of food for Asian elephants in the wild and
in captivity. Front. Biol. China. 2:351-355, 2007.
22. van Sonsbeek, G.R., van der Kolk, J.H., van Leeuwen, J.P.,
32
Aroch, I.
JUNE 2016.indb 32
Everts, H., Marais, J., and Schaftenaar, W.: Effect of calcium and
cholecalciferol supplementation on several parameters of calcium
status in plasma and urine of captive Asian (Elephas maximus)
and African elephants (Loxodonta africana). J. Zoo Wildl. Med.
2013 44:529-540.
23. International Species Information System (ISIS), Physiological Reference Intervals for Captive Wildlife, 2013, available at:
https://zims.isis.org/Main.aspx:. Access date: 08.02.2016.
24. Schaftenaar, W. and Hildebrandt, T.B.: Veterinary guidelines for
reproduction-related management in captive female elephants,
2005. Available at:http://www.elephantcare.org/protodoc_files/
new%2006/Repro_guidelines_2005_09_05.pdf. Access date:
02.02.2016.
25. Ullrey, D.E., Crissey, S.D. and Hintz, H.F.: Elephants: Nutrition
and Dietary Husbandry: Nutrition Advisory Group Handbook,
Michigan State University, East Lansing, Fact Sheet 004, 1997.
26. Dierenfeld, E.S., Nutrition. In: Fowler, M.E. and Mikota, S.K. eds.
Biology, Medicine, and Surgery of Elephants, 1st edition. Ames,
Blackwell, pp. 57-65, 2006.
27. Elephant Nutrition Guide; Zutrition.com; available at: http://
www.zutrition.com/elephant-nutrition-guide/. Access date:
11.02.2016.
28. Moore, E.W.: Ionized calcium in normal serum, ultrafiltrates,
and whole blood determined by ion-exchange electrodes. J. Clin.
Invest. 49:318-334, 1970.
29. Conway, J. C.D., Liparini, A., Oliveira, J.R. and Belchior, J.C.:
Analyses of the temperature and pH effects on the complexation of magnesium and calcium in human blood plasma: an approach using artificial neural networks. Analyt. Bioanalyt. Chem.
389:1585-1594, 2007.
30. Wang, S., McDonnell, E.H., Sedor, F.A. and Toffaletti, J.G.: pH effects on measurements of ionized calcium and ionized magnesium
in blood. Arch. Pathol. Lab. Med. 126:947-950, 2002.
31. Altura, B.T., Burack, J.L., Cracco, R.Q., Galland, L., Handwerker,
S.M., Markell, M.S., Mauskop, A., Memon, Z.S., Resnick, L.M.,
Zisbrod, Z. and Altura, B.M.: Clinical studies with the NOVA
ISE for IMg2+. Scand. J. Clin. Lab. Invest. 54 (Suppl. 217), 53-67,
1994.
32. van Sonsbeek, G.R., van der Kolk, J.H., van Leeuwen, J.P. and
Schaftenaar, W.: Preliminary validation of assays to measure
parameters of calcium metabolism in captive Asian and African
elephants in western Europe. J. Vet. Diag. Invest. 2011 23:504-510.
33. Fogh-Andersen, N. and Siggaard-Andersen, O.: Standardizing
and reporting results from Mg2+ ISEs, with some notes on sample handling. Scand. J.Clin. Lab. Invest. 54 (Suppl. 217):89-96,
1994.
34. Greenway, D.C., Hindmarsh, J.T., Wang, J., Khodadeen, J.A. and
Hebert, P.C.: Reference interval for whole blood ionized magnesium
in a healthy population and the stability of ionized magnesium under varied laboratory conditions. Clin. Biochem. 29:515-520, 1996.
35. Thode, J., Juul-Jorgensen, B., Seibaek, M., Elming, H., Borrensen,
E. and Jordal, R.: Evaluation of an ionized magnesium-pH analyzer - NOVA 8. Scand. J. Clin. Lab. Invest. 58:127-133, 1998.
36. Stockham, S.L. and Scott, M.A.: Blood gases, blood pH, and
strong ion difference, In: Fundamentals of Veterinary Clinical
Pathology, 2nd ed. Blackwell, Ames p. 573; 2008.
Israel Journal of Veterinary Medicine  Vol. 71 (2)  June 2016
31/05/2016 13:35:53