ARTICLE IN PRESS

Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123

www.elsevier.de/jtemb

EPIDEMIOLOGY
Epidemiology of trace elements deficiencies in Belgian beef and
dairy cattle herds
Hugues Guyota,, Claude Saegermanb, Pascal Lebretonc,
Charlotte Sandersend, Frédéric Rollina
a
University of Liège, Faculty of Veterinary Medicine, Clinic for Ruminants, Boulevard of Colonster,
20 Bât. B42, 4000-Liège, Belgium
b
University of Liège, Faculty of Veterinary Medicine, Epidemiology and Risk Analysis Applied to Veterinary Sciences,
Boulevard of Colonster, 20 Bât. B42, 4000-Liège, Belgium
c
N.B.V.C. Early Health Indicators, Chemin des Joncs, 12, 69570-Dardilly, France
d
University of Liège, Faculty of Veterinary Medicine, Equine Clinic, Boulevard of Colonster, 20 Bât. B41, 4000-Liège, Belgium

Received 4 June 2008; accepted 12 February 2009

Abstract
Selenium (Se), iodine (I), zinc (Zn) and copper (Cu) deficiencies in cattle have been reported in Europe. These
deficiencies are often associated with diseases. The aim of the study was to assess trace element status in Belgian cattle
herds showing pathologies and to compare them to healthy cattle herds. Eighty-two beef herds with pathologies, 11
healthy beef herds, 65 dairy herds with pathologies and 20 healthy dairy herds were studied during barn period. Blood
and/or milk samples were taken in healthy animals. Plasma Zn, Cu, inorganic I (PII) and activity of glutathione
peroxidase in erythrocytes (GPX) were assayed. In milk, I concentration was measured. Data about pathologies and
nutrition in the herds were collected. According to defined thresholds, it appeared that a large proportion of deficient
herds belonged to ‘‘sick’’ group of herds. This conclusion was supported by the mean value of trace elements and by the
fact that a majority of individual values of trace elements was below the threshold. Dairy herds had mean values of trace
elements higher than beef herds. More concentrates and minerals were used in healthy herds versus ‘‘sick’’ herds. These
feed supplements were also used more often in dairy herds, compared to beef herds. Trace elements deficiencies are
present in cattle herds in Belgium and are linked to diseases. Nutrition plays a major role in the trace elements status.
r 2009 Elsevier GmbH. All rights reserved.

Keywords: Trace elements; Cows; Health; Blood; Milk

Introduction in cattle are passed on to humans consuming deficient

Trace elements are essential in human and
animal nutrition. For some trace elements (iodine and
selenium), when food-producing animals suffer from
deficiencies, they are likely to produce milk and meat
that are deficient as well. Thus, the deficiencies observed
Corresponding author. Tel.: +32 4 366 4020; fax: +32 4 366 4024.
E-mail address: Hugues.Guyot@ulg.ac.be (H. Guyot).
animal products. In Britain, iodine (I) deficiency in
humans has regressed due to the I supplementation of
dairy cows, because I passes into milk [1]. Furthermore,
some trace elements (e.g. selenium) may protect animal
products against oxidation [2,3] and improve their shelf
life.
Clinical signs linked to trace element deficiencies are
well described in cattle [4] and are related to problems
in, e.g. reproduction, milk production and neonatology.

0946-672X/$ - see front matter r 2009 Elsevier GmbH. All rights reserved.
doi:10.1016/j.jtemb.2009.02.006
 ARTICLE IN PRESS
H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123 117

Nevertheless, many of the clinical signs are not Table 1. Thresholds used to classified ‘‘sick herds’’ and
pathognomonic of trace element deficiencies and their ‘‘healthy herds’’ [31,32].
diagnosis is most commonly confirmed by blood
Pathologies Threshold (tolerance)
analysis. Soil analysis is less informative because
assimilation of trace elements remains complex. The Adult cattle
trace element content of roughages is not always Retention of fetal membranes o5%
representative of the trace element levels in blood Chronic metritis o20%
because the presence of antagonists may reduce the Uterine involution delay o10%
absorption of trace elements from feed and water (e.g. Ovarian cyst o10%
high levels of manganese, molybdenum, iron, calcium or Embryonic mortality o10%
goitrogenic substances) [5]. But logically, insufficient Abortions o3%
Infertility (culling rate due to) o10%
contents of trace elements in feedstuff will definitively
depress respective blood levels.
Trace element deficiencies are frequently reported in Clinical mastitis o25%/year
Europe [6–11]. In Belgium, the trace element content of Sub-clinical mastitis 80% cows o300,000 cells/
soil and roughage has decreased over the last decade, ml
due to the exhaustion of soils and new agricultural
practices (fertilizers containing no or few trace ele- Clinical hypocalcaemia o5%
ments). Furthermore, Belgian legislation limited the Left abomasal displacement o5%
maximal content of trace elements in mineral complexes. Ketosis o10%
For example, the maximal content of selenium (Se) in a
mineral complex was until recently only 20 ppm. Like- Lameness o20%
wise, some Belgian farmers in beef speculation do not Lameness (culling rate due to) o10%
commonly use mineral supplements in the ration of their
herds. Finally, the hyper-muscled Belgian Blue (BB)
Parasitism (external: mange, o5%
breed has increased its trace elements needs in parallel fleas)
with its performances. Mortality (all aetiology) o3%
The aim of the study was to determine zinc (Zn), Digestive troubles o5%
copper (Cu), Se and I status in dairy and beef Belgian Respiratory troubles o5%
cattle herds and to assess the relationship between trace
elements and health status in these herds.
Calves
Mortality rate 0–24 h o6%
Mortality rate 1–30 days o2%
Mortality rate 31–60 days o1%
Material and method Mortality rate 3–24 months o1%

Cattle herds
A total of 178 cattle herds have been investigated in
Wallonia (Belgium) during the autumn–winter (in-barn)
periods from 2002 to 2006. The herds have been
assigned into 4 groups: (1) sick beef herds (n ¼ 82), (2)
healthy beef herds (n ¼ 11), (3) sick dairy herds (n ¼ 65)
and (4) healthy dairy herds (n ¼ 20). Sick herds were
defined as herds with levels of pathologies above
thresholds as defined in the literature (see Table 1). In
contrast, healthy herds in this study were herds that
showed levels of pathologies below the aforementioned
thresholds. In each farm, the morbidity and mortality
rates of pathologies were objectively measured by the
authors. All sick herds were referred to the Faculty of
Veterinary Medicine of the University of Liège (Clinic
for Ruminants) by veterinary practitioners. It was
assumed that referred cases were representative of
herds with pathologies in the field. Healthy herds
were randomly chosen. The mean size of these herds
Morbidity rate: diarrhoea o20%
Morbidity rate: respiratory o6%
trouble
Morbidity rate: lameness o5%
(as number of calving per year) was 90 (30–300) (median
(range)), 80 (40–150), 45 (20–165) and 50 (25–100),
respectively for herds belonging to groups (1), (2), (3)
and (4). Mean milk production (305 days) in dairy herds
was 712871442 L (mean7SD) and 783871898 L for
groups (3) and (4), respectively. Milk and/or blood
samples were taken in these 4 groups.
Data regarding nutrition and pathologies have been
collected in each farm. Feedstuffs have been assigned
into 7 categories: grass (pasture, hay and silage), corn
(silage), beet pulp (fresh or silage), concentrates (energy
and/or proteins), beet and derivates, cereals (cereals and
derivates) and minerals (mineral, trace elements and
vitamins complexes). Straw in the rations was not taken
 ARTICLE IN PRESS
118 H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123
Table 2. Classification of pathologies observed in the studied
herds.
Animals Code Pathologies
Calves C1 Neonatal mortality, weak calf syndrome,
respiratory distress syndrome
C2 Septicaemia, omphalitis, arthritis, meningitis,
diarrhoea, failure of colostral immunity
transfer
C3 Myopathy, cardiomyopathy
C4 Poor growth, broncho-pneumonia, ear
infection, laryngitis, mortality
Adults RE REproduction: infertility, retention of foetal
membranes, metritis, abortions, embryo
mortality, anoestrus, uterine hypoplasia
UH Udder Health: subclinical and clinical mastitis
MP Not optimal milk production
AM Adult mortality
DE DErmatology: mange, ringworm, poor coat
NE NEurology: cortico-cerebral necrosis,
listeriosis
ME MEtabolism: milk fever, ketosis, left abomasal
displacement, downer cow syndrome
LA LAmeness (all aetiologies)
OT OTher: parasitism, weight loss, diarrhoea
into account. Pathologies in calves and adult cattle have
been classified as described in Table 2 based on the
authors experiences.
Sampling and assays
Within the herds, female healthy animals (females
were 4 years old [median] and ranged from 2 to 9 years
of age in all groups) have been randomly chosen and
matched according to breed, sex, age, and level of
production among the groups. Sick animals recently
treated for pathologies and those close to calving (less
than 2 weeks before or after calving) were excluded. A
total of 669, 94, 484 and 156 animals have been studied,
respectively, in groups (1–3) and (4). In each herd, 872
healthy cows have been investigated, which corresponds
to the standard sampling procedure for the evaluation of
the nutritional status in cattle herds [12–14].
Blood sampling has been performed in the jugular
vein, using 16G needles and heparin tubes (lithium-
heparin, 10 mL, Sarstedts). Blood was sent to the
laboratory within 24 h. Whole blood activity of glu-
tathion peroxidase in erythrocytes (GPX) was measured.
Zn, Cu and inorganic I (PII) were determined in plasma.
During analyses, samples presenting haemolysis were
culled as they can lead to falsely elevated trace element
levels in blood [12].
Regarding milk samples, beef cows were individually
sampled and milk from at least 10 cows was pooled for
Table 3. Thresholds for zinc (Zn), copper (Cu), activity of
glutathione peroxidase in erythrocytes (GPX) and plasmatic
inorganic iodine (PII) in blood and iodine in milk in cattle
[7,11,13,18,21].
Herd Zn Cu GPX PII Milk I
status (mmol/L) (mmol/L) (IU/gHb) (mg/L) (mg/L)
Adequate 14–21 13–18 220–600 45–650 80
Marginal 8–14 8–13 75–220 15–45 30–80
Deficient o8 o8 o75 o15 o30
analysis. Milk was sampled after elimination of the first
sprays. Bulk tank milk was analysed for dairy cows.
Milk was frozen (20 1C) until analysis. Milk samples
were thawed at room temperature and, after homo-
genisation, a 500 mL sample of whole milk was taken for
analysis using alkaline ashing (KOH 4 M) at 105 1C and
Sandell & Kolthoff reaction [15].
GPX activity in erythrocytes was determined as
previously described [16] using a commercial kit
(Ransels, Randox laboratories, Crumlin, UK). Plasma
Zn and Cu were assayed by inductivity coupled plasma-
mass spectrophotometry (ICP-MS) [17]. Finally, PII was
measured by ion-exchange chromatography and the
Sandell & Kolthoff reaction [15]. All analyses were
conducted in the same laboratory (N.B.V.C., 69570
Dardilly, France).
Deficiency thresholds of Zn, Cu, PII and GPX were
established according to previous works [7,11,13,18]
(Table 3). If the mean values were below adequate levels
for Zn, Cu, GPX, PII in the blood or I in milk, the herd
was considered as deficient for the given trace element.
Statistical analysis
All statistical analyses were carried out with Stata/SE
[19]. The mean of several trace elements measurements
in function of different groups of animals were
compared with Welch’s test because of an unequal
variance [20]. Chi-square test (w2) or Fisher’s test were
used in order to compare herds regarding diseases and
nutrition. The limit of statistical significance of the
conducted tests was defined as Pp0.05.
Results
More than 90% of beef breeds in beef cattle herds was
Belgian Blue. Other beef breeds, representing less than
10%, were Limousin, Charolais, Blonde d’Aquitaine,
Aberdeen Angus and BB crossbred. In dairy herds,
Holstein (red or black) breed was the most commonly
represented breed. Other breeds, also representing less
than 10%, were Montbéliarde, Normande and Holstein
crossbred.
 ARTICLE IN PRESS
H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123 119

Geographical distribution of the herds is presented in
Fig. 1. Of the 82 sick beef herds, 36 were located in the
province of Luxembourg, 20 in Liège, 12 in Hainaut, 12
in Namur and 2 in Wallonian Brabant. Of the 65 sick
dairy herds, 30 were located in the province of Liège, 14
in Luxembourg, 12 in Hainaut, 4 in Namur, 4 in
Wallonian Brabant and 1 in Limbourg.
The distribution of pathologies in the sick herds is
given in Table 4. Many herds (34/82 beef herds and
31/65 dairy herds) suffered from more than one class of
pathologies (median: 2 pathologies for both groups),
Fig. 1. Geographical distribution of the herds (healthy,
unhealthy, dairy and beef) studied in Belgium.
ranging from 1 to 5 per herd. Beef herds had more C1-2-
3-4 and AM pathologies (w2 ¼ 30; Po0.001) but no
UH, MP, ME and less LA pathologies, compared to
dairy herds (w2 ¼ 55; Po0.001).
The use of the different classes of feedstuffs in the
studied farms is given in Table 5. Concentrates were
more often used in healthy beef herds compared to sick
beef herds (w2 ¼ 7.2; P ¼ 0.007). Concentrates were
more often used in sick dairy herds than in sick beef
herds (w2 ¼ 22.9; Po0.001). Minerals were used in a
lower proportion in sick beef herds compared to sick
dairy herds (w2 ¼ 7.7; P ¼ 0.005). Furthermore, it also
appeared that minerals were more often used in healthy
herds (beef and dairy) compared to sick herds (beef and
dairy) (Fisher’s exact; Po0.001).
Table 6 presents the mean and percentile (25, 50, 75)
values of Zn, Cu, GPX, PII in blood and I in milk in the
4 groups of herds, as well as their statistical significance.
For all studied trace elements (Zn, Cu, GPX, PII, I
milk), the mean value was higher in healthy herds
compared to sick herds (Pp0.0001). Considering GPX
(Pp0.0001) and I in plasma and milk (0.01XPp0.03),
mean value of healthy dairy herds was higher than beef
herds. Table 7 shows the number of deficient herds. The
term ‘‘deficient herd’’ refers to herds with mean trace
element values below the threshold values for adequate
status (see Table 3). It is supposed that if the mean value
of the sampled healthy animals for a given trace element
is below the defined threshold for adequate status, then
there is a trace element deficiency in the herd.

Table 4. Distribution of pathologies in the studied sick herds.

C1 C2 C3 C4 RE UH MP AM DE NE ME LA OT X2a

Beef 33b 35b 10b 30b 17 0b 0b 4b 8 2 0b 6b 6 34

(n ¼ 82) 40% 43% 12% 37% 21% 0% 0% 5% 10% 2% 0% 7% 7% 41%
Dairy 12 16 1 14 24 22 5 0 3 1 11 15 2 31
(n ¼ 65) 18% 25% 2% 22% 37% 34% 8% 0% 5% 2% 17% 23% 3% 48%

Absolute numbers+percentage (in italic character). C1: calves pathologies, C2: calves pathologies, C3: calves pathologies, C4: calves pathologies,
RE: reproduction, UH: udder health, MP: milk production, AM: adult mortality, DE: dermatology, NE: neurology, ME: metabolism, LA: lameness,
OT: others.
a
X2: number of herds with X2 classes of pathologies.
b
Significant difference between Beef and Dairy herds (Po0.01).

Table 5. Use of the different classes of feedstuffs in the studied farms.

Grass Corn Beet pulp Concentrates Cereals Beet and Minerals

(%) (%) (%) (%) (%) derivates (%) (%)

Beef/sick n ¼ 82 94 87 38 39a,b 35 18 46a,b

Beef/healthy n ¼ 11 100 73 45 82 36 27 100
Dairy/sick n ¼ 65 97 68 48 78 46 29 69a
Dairy/healthy n ¼ 20 100 60 45 95 50 30 100
a
Significant difference between ‘‘healthy’’ herds and ‘‘sick’’ herds (Po0.05).
b
Significant difference between beef herds and dairy herds (Po0.05).
 ARTICLE IN PRESS
120 H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123

Table 6. Values of plasmatic zinc (Zn), copper (Cu), activity of glutathione peroxidase in erythrocytes (GPX), plasmatic inorganic
iodine (PII) and iodine in milk (Milk I) in 4 groups of herds.

Trace element Group Number Mean P25 P50 P75

a
Zn (mmol/L) Sick beef herds 57 13.1 12.3 13.0 14.2
Healthy beef herds 10 16.6 15.8 16.6 17.6
Sick dairy herds 27 13.4b 12.6 13.3 14.3
Healthy dairy herds 18 16.3 15.4 15.9 17.2

Cu (mmol/L) Sick beef herds 57 11.8a 9.8 12.0 13.1

Healthy beef herds 10 14.7 14.1 14.9 15.4
Sick dairy herds 27 12.4b 11.5 13.0 13.6
Healthy dairy herds 18 14.5 13.3 14.7 15.2

GPX (IU/gHb) Sick beef herds 82 111a 58 102 150

Healthy beef herds 11 290c 273 295 314
Sick dairy herds 65 158b,d 76 144 207
Healthy dairy herds 20 379 315 389 430

PII (mg/L) Sick beef herds 45 31a 15 19 37

Healthy beef herds 6 79e 76 81 82
Sick dairy herds 51 41b 19 28 44
Healthy dairy herds 11 107 86 95 131

Milk I (mg/L) Sick beef herds 44 44a 28 40 61

Healthy beef herds 9 103e 84 95 118
Sick dairy herds 39 49b 31 38 50
Healthy dairy herds 12 145 95 154 188

P25, P50 and P75: percentiles 25, 50 and 75, respectively.

a
Significant difference between )sick* and )healthy* beef herds (Pp0.0001).
b
Significant difference between )sick* and )healthy* dairy herds (Pp0.0001).
c
Significant difference between )healthy* beef herds and )healthy* dairy herds (Pp0.0001).
d
Significant difference between )sick* beef herds and )sick* dairy herds (Pp0.0001).
e
Significant difference between )healthy* beef herds and )healthy* dairy herds (0.01XPp0.03).

These results revealed that more deficient herds are
present among the ‘‘sick’’ herds, compared to the
healthy herds. These conclusions are stated on the fact
that confidence intervals do not cross over. Sick herds
have a higher incidence of multiple deficiencies (72/82
beef herds and 51/65 dairy herds with more than one
deficiency; 0/82 beef herds and 2/65 dairy herds with no
deficiency) compared to healthy herds (0/11 beef herds
and 1/20 dairy herds with more than one deficiency; 8/11
beef herds and 15/20 dairy herds with no deficiency).
Discussion and conclusion
To determine the trace element status in cattle,
different tools have been described in the literature
[4,21]. The most common and pertinent ones were
investigated in this study. To assess I intake in cattle, the
dosage of PII [7] or I content in milk [22] is frequently
performed. These markers are short-term markers of I
nutritional status. After interruption of I supplementa-
tion, PII decreases within a week, while milk I decreases
more slowly [22]. This indicates that PII may be a less
reliable marker of I status if analysed shortly after
discontinuation of I supplementation. Thyroid hor-
mones may be used as a long-term markers for I status,
although authors argued that changes in the concentra-
tion of thyroid hormones are related to thyroid function
and thus cannot be used for assessing a nutritional
factor such as I [22,23]. For assessing Se status, GPX
activity was used because this parameter is correlated
with blood Se concentration [24]. GPX is linked to the
life duration of the erythrocytes and is a long-term
marker of Se status [25]. Bulk milk Se is also a good
mean for measuring Se intake in dairy herds, as
suggested in a recent review on Se and I deficiencies in
ruminants [21]. For assessment of the Cu and Zn status,
the plasma levels were determined because this method
is the one most commonly used [26]. Liver biopsy is
described to be more reliable to assess the Zn or Cu
 ARTICLE IN PRESS
H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123
Table 7. Number of deficient herds (and confidence interval) according to the threshold of deficiency in the 4 groups for plasmatic
zinc (Zn), copper (Cu), activity of glutathion peroxidase in erythrocytes (GPX), plasmatic inorganic iodine (PII) and iodine in milk.
‘‘Sick’’ beef ‘‘Healthy’’ beef
a
Zn 42/57 (74%) 0/10 (0%)
60–84%b 0–26%
Cu 40/57 (70%) 1/10 (10%)
57–82% 0–45%
GPX 78/82 (95%) 0/11 (0%)
88–99% 0–36%
PII 36/45 (80%) 0/6 (0%)
65–90% 0–39%
Milk I 42/44 (95%) 2/9 (22%)
85–99% 0–60%
a
Number of deficient herds on total herds (and %).
b
The 95% confidence interval (binomial exact).
status but such sampling method is technically demand-
ing in a large population of animals on the field.
Independent of the analysis of the sample, there are
also standard procedures to be applied when sampling
cattle for assessing their trace elements status [12–14].
Only healthy animals have to be sampled within sick or
healthy herds. When diagnosing clinical disease, animals
with clinical signs are sampled. However, when assessing
nutritional status, only healthy animals are selected [12].
Furthermore, some factors may affect trace elements
status in blood. For example, inflammation [27], stress
[14] and infections [28] cause an increase of plasma Cu
and a decrease in plasma Zn content.
Even in healthy animals, other factors, considered as
‘‘within-herd’’ factors, can also influence trace mineral
status. To minimize these factors, the animals sampled
were selected within the same age group, level of
production, gestation state and lactation state. If this
was not possible, more samples were taken to achieve
acceptable statistical strength. The smallest number of
animals to be sampled depends on the expected
variability of the population and the selection of animals
within the herd. Previous works [12,13] considered that
7 animals is the minimum for the evaluation of blood
components, although rarely smaller numbers of ani-
mals are sufficient for nutritional assessment. In the
present study, a mean of 8 animals were sampled in each
farm. Since the cows were matched according to their
physiological state and age, ‘‘within-herd’’ variability
was limited.
To assess trace elements status on animals or herds,
thresholds need to be determined. However, a wide
variety of thresholds are published by different authors.
121
‘‘Sick’’ dairy ‘‘Healthy’’ dairy
18/27 (67%) 1/18 (6%)
46–83% 0–27%
13/27 (48%) 3/18 (17%)
51–90% 4–41%
50/65 (77%) 1/20 (5%)
65–86% 0–25%
38/51 (75%) 0/11 (0%)
60–86% 0–24%
34/39 (87%) 1/12 (8%)
73–96% 0–39%
When insufficient intake is present, the literature usually
defines 4 progressive steps: (1) depletion; (2) deficiency;
(3) dysfunction and (4) disease [29]. The thresholds used
in this study were chosen according to the second step
(deficiency) and are higher than those commonly used.
The goal in trace element nutrition is to avoid clinical
diseases but also to optimize herd health, welfare,
productivity and food quality. Sub-clinical diseases
may appear while trace elements are given according
to the National Research Council (NRC) requirements
if concomitant negative factors, e.g. stress, exist. Never-
theless, recent literature also proposed higher thresholds
[7,11,13,18]. Another threshold, much lower than
deficient, is also defined for animals with clinical signs
of deficiency (steps 3 and 4). This clinically deficient
threshold was never reached in this study, as only
clinically healthy animals were sampled.
Diseases observed in the farms generally belong to the
group of multifactorial diseases in which environment,
infectious agents, genetics and nutrition (not only trace
elements) play an important role as predisposing or
determining factors for the apparition of the disease.
Several authors have described clinical signs directly
linked to a specific trace element deficiency [4]. Some
have also attributed to these deficiencies a risk factor for
the development of pathologies (odds ratio) [11]. In this
study, as pathologies were investigated at a herd level
and only clinically healthy animals were sampled, it was
not possible to relate clinical signs to a specific trace
element deficiency. Indeed, most of the time, multiple
pathologies, as well as multiple trace elements deficien-
cies, were recorded in the farms. Moreover, other critical
points have to be taken into consideration in these farms
 ARTICLE IN PRESS
122 H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123
as well (e.g. nutrition, management, infectious diseases).
In these herds, the next step will be to develop more
detailed and accurate information on relationships
between individual herd trace mineral contents in blood
or milk and specific herd pathologies.
The interpretation of results was made using the herd
level, although, the same pattern was found when
comparing groups at an individual level (data not
shown). According to the results, we could use the
mean of the herd for diagnosing trace element nutri-
tional status. Globally, if the herd mean value was below
the threshold, most of the sampled animals were below
the threshold, as well. Accordingly, pooled samples
within a herd could be used in order to reduce costs of
analyses. Nevertheless, pooling of the samples is only
advised if the variability of results is low, which occurred
in the present study. An appropriate method of selection
of the samples ensures that variability remains low.
Trace element status found in the herds of the present
study was compared to status in Ireland and France and
to results previously obtained in Belgium. PII and Cu
values in sick herds were similar to values in diseased
cattle herds from Ireland [7] while GPX activity was
higher in the present study. Values of GPX activity, and
plasma Zn and Cu content in healthy and sick herds
from France [11] were globally similar to the present
ones, as well as to the values obtained in the previous
Belgian study [10]. Indeed, the Se [9] and I [8,30] status
in Europe are often described as deficient.
Lower trace element levels were found in sick herds, in
comparison with healthy herds. The origin of the
difference might be the more frequent absence of
minerals and concentrates in herds with a higher number
of pathologies. In Belgium, dairy farmers generally buy
more concentrates and minerals than beef farmers, as
observed in this study.
Conclusions
As described in other European countries, deficiencies
of the trace elements Zn, Cu, Se and I are present in
Belgium. The presence of trace element deficiencies
increases the risk of pathologies in cattle herds. Trace
element deficiencies in Belgium are more frequent when
cows are not complemented with concentrates and/or
minerals. Measuring trace elements in blood or milk is
expensive. Further studies are needed to determine the
accuracy of pooled samples in the frame of diagnosing
trace element deficiencies. Assaying I and Se in bulk
milk is the first step to evaluate nutritional status in a
single representative sample.
This study has consequently shown the interest to
supplement cattle with trace elements. The investment in
trace element supplementation is profitable for animal
health and will reduce the cost of diseases and loss of
production (mortalities, antibiotics, decreased weight
gain or milk production and eventually extra-labour).
Acknowledgement
This research was funded by the University of Liège
(Special Research Founds C-06/68).
References
[1] Phillips DI. Iodine, milk, and the elimination of endemic
goitre in Britain: the story of an accidental public health
triumph. J Epidemiol Community Health 1997;51:391–3.
[2] Lindmark-Mansson H, Akesson B. Antioxidative factors
in milk. Br J Nutr 2000;84(Suppl):S103–10.
[3] O’Grady MN, Monahan FJ, Fallon RJ, Allen P. Effects
of dietary supplementation with vitamin E and organic
selenium on the oxidative stability of beef. J Anim Sci
2001;79:2827–34.
[4] Graham TW. Trace elements deficiencies in cattle. Vet
Clin North Am Food Anim Pract 1991;7:153–215.
[5] Puls R. Mineral levels in animal health. Clearbrook,
Canada: Sherpa International; 1994. 356pp.
[6] Lamand M. Les acquisitions récentes sur les carences en
oligo-éléments du sol aux ruminants. Supplément du
bulletin technique CRZV. Theix: INRA; 1975. p. 90–94.
[7] Mee JF, O’Farrell KJ, Rogers PAM. Base-line survey of
blood trace element status of 50 dairy herds in the south
of Ireland in the spring and autumn of 1991. Ir Vet J
1994;47:115–22.
[8] Delange F, Burgi H, Chen ZP, Dunn JT. World status of
monitoring iodine deficiency disorders control programs.
Thyroid 2002;12:915–24.
[9] Oldfield JE. Selenium World Atlas. Grimbergen: Sele-
nium-Tellurium Development Association (STDA); 2002.
[10] Rollin F, Lebreton P, Guyot H. Trace elements deficien-
cies in Belgian beef and dairy herds in 2000–2001. In:
Proceedings of the 22nd World Buiatrics Congress,
Hanover, Germany, 2002. p. 72.
[11] Enjalbert F, Lebreton P, Salat O. Effects of copper, zinc
and selenium status on performance and health in
commercial dairy and beef herds: retrospective study.
J Anim Physiol Anim Nutr 2006;90:459–66.
[12] Herdt HT. Variability characteristics and test selection in
herd-level nutritional and metabolic profile testing. Vet
Clin North Am Food Anim Pract 2000;16:387–403.
[13] Maas J. Diagnostic considerations for evaluating nutri-
tional problems in cattle. Vet Clin North Am Food Anim
Pract 2007;23:527–39.
[14] Herdt HT, Rumbeiha W, Braselton WE. The use of blood
analyses to evaluate mineral status in livestock. Vet Clin
North Am Food Anim Pract 2000;16:423–44.
[15] Aumont G, Tressol J-C. Rapid method for the direct
determination of inorganic iodine in plasma using ion-
exchange chromatography and the Sandell and Kolthoff
reaction. Analyst 1987;112:875–8.
[16] Paglia DE, Valentine WN. Studies on the quantitative
and qualitative characterization of erythrocyte glu-
tathione peroxidase. J Lab Clin Med 1967;70:158–9.
 ARTICLE IN PRESS
H. Guyot et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 116–123
[17] Lumet L, Negriolli J. Dosage des oligo-éléments dans le
sang par torche à plasma couplée à la spectrométrie de
masse. Bull GTV 2007;38:27–30.
[18] Rollin F. Detection of trace minerals deficiency in cattle.
In: Proceedings of the European Meeting organized by
Société Française de Buiatrie (from research to clinic),
Paris, France, 2003.
[19] StataCorp. Stata statistical software: release 8.0. College
Station, TX: Stata Corporation; 2003.
[20] Dagnelie P. Statistique théorique et appliquée. Tome 2.
Inférence statistique à une et à deux dimensions. De Boeck
& Larcier, 1998.
[21] Guyot H, Rollin F. Selenium and Iodine deficiencies
diagnosis in cattle. Ann Med Vet 2007;151:166–91.
[22] Swanson EW, Miller JK, Mueller FJ, Patton CS, Bacon
JA, Ramsey N. Iodine in milk and meat of dairy cows fed
different amounts of potassium iodide or ethylenediamine
dihydroiodide. J Dairy Sci 1990;73:398–405.
[23] Hemingway RG, Fishwick G, Parkins JJ, Ritchie NS.
Plasma inorganic iodine and thyroxine concentrations for
beef cows in late pregnancy and early lactation associated
with different levels of dietary iodine supplementation.
Vet J 2001;162:158–60.
[24] Counotte GHM, Hartmans J. Relation between selenium
content and glutathione-peroxidase activity in blood of
cattle. Vet Q 1989;11:155–60.
123
[25] Whitaker DA. Interpretation of metabolic profiles in
dairy cows. Cattle Pract 1997;50:498–501.
[26] Kincaid RL. Assessment of trace mineral status of
ruminants: a review. J Anim Sci 2000:1–10.
[27] Milanino R, Cassini A, Conforti A, Franco L, Marrella
M, Moretti U, et al. Copper and zinc status during acute
inflammation: studies on blood, liver and kidneys metal
levels in normal and inflamed rats. Inflammat Res
1986;19:215–23.
[28] Orr CL, Hutcheson DP, Grainger RB, Cummins JM,
Mock RE. Serum copper, zinc, calcium and phosphorus
concentrations of calves stressed by bovine respiratory
diseases and infectious rhinotracheitis. J Anim Sci 1990;
68:2893–900.
[29] Suttle NF. Problems in the diagnosis and anticipation of
trace element deficiencies in grazing livestock. Vet Rec
1986;119:148–52.
[30] World Health Organization. Assessment of the iodine
deficiency disorders and monitoring their elimination.
WHO/NHD/01.1. Genova: World Health Organization;
2001.
[31] Brand A, Noordhuizen JPTM, Schukken YH. Herd
health and production management in dairy practice.
The Netherlands: Wageningen Pers; 1996.
[32] Radostits OM. Herd health – food animal production
medicine. New York: W.B. Saunders Company; 2001.