Archives of Animal Nutrition

ISSN: 1745-039X (Print) 1477-2817 (Online) Journal homepage: http://www.tandfonline.com/loi/gaan20

Effects of different iron supply to pregnant

sows (Sus scrofa domestica L.) on reproductive
performance as well as iron status of new-born
piglets

Marzell Buffler, Christiane Becker & Wilhelm M. Windisch

To cite this article: Marzell Buffler, Christiane Becker & Wilhelm M. Windisch (2017):
Effects of different iron supply to pregnant sows (Sus scrofa domestica L.) on reproductive
performance as well as iron status of new-born piglets, Archives of Animal Nutrition, DOI:
10.1080/1745039X.2017.1301059

To link to this article: http://dx.doi.org/10.1080/1745039X.2017.1301059

Published online: 15 Mar 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=gaan20

Download by: [University of Newcastle, Australia] Date: 17 March 2017, At: 18:48
ARCHIVES OF ANIMAL NUTRITION, 2017
http://dx.doi.org/10.1080/1745039X.2017.1301059

Effects of different iron supply to pregnant sows (Sus scrofa

domestica L.) on reproductive performance as well as iron
status of new-born piglets
Marzell Buffler, Christiane Becker and Wilhelm M. Windisch
TUM School of Life Sciences Weihenstephan, Technical University of Munich, Freising, Bavaria, Germany

ABSTRACT ARTICLE HISTORY

The present study aimed to investigate the effects of different iron Received 27 October 2017
(Fe) supply to sows during gestation on their reproductive perfor- Accepted 23 February 2017
mance and placental Fe load. Additionally, the Fe status of the KEYWORDS
corresponding offspring was assessed. Twenty multiparous sows Iron; nutrient requirements;
were fed from insemination to farrowing with isoenergetic and piglet production;
isonitrogenic balanced diets differing in Fe content. The diet low reproductive performance;
in Fe (Group −Fe) was mainly composed of soybean meal and sow feeding
maize meal and had a Fe content of 114 mg/kg DM. For the diet
high in Fe (Group +Fe), the diet was supplemented with Fe(II)
SO4 · 7H2O to a total Fe content of 256 mg/kg. Blood characteristics
(haemoglobin, haematocrit, mean corpuscular haem concentration,
total Fe-binding capacity, transferrin saturation) of all sows were
measured at the beginning and at the end of gestation. Daily Fe
retention was calculated at the day of farrowing. After birth, repro-
ductive performance (litter size, piglet weight, litter weight), placen-
tal Fe content and Fe blood characteristics of the piglets were
determined. Apparent daily Fe retention tended to be greater in
Group +Fe (p < 0.1). Blood parameters of the sows did not show any
variations between feeding groups, neither at the beginning nor at
the end of pregnancy, whereas placental Fe content was lower in
Group −Fe (p < 0.05). In addition, Fe supply during gestation
improved litter size (p < 0.01) and litter weight (p < 0.05).
Although all sows were supplied according to the current Fe recom-
mendations, a significant decline in reproductive performance of
Group −Fe was recognised. Therefore, it was concluded that the re-
evaluation of the gross Fe requirements of pregnant sows is inevi-
table to accommodate the current feeding recommendations.

1. Introduction
In the organism, the essential trace element iron (Fe) is needed for a multitude of functions
and plays a key role in oxygen transport and as cofactor of many enzymes (Hentze et al.
2010). Therefore, an adequate Fe supply is fundamental for health and adequate perfor-
mance of livestock animals. The current recommendations of Fe supply for pregnant and
lactating sows are 80–100 mg/kg dry matter (DM) (NRC 2012). However, these values
have not been re-evaluated for more than 35 years. During this period, a massive increase
in the reproductive performance of sows was achieved, which was due to quantitative

CONTACT Marzell Buffler marzell.buffler@wzw.tum.de

© 2017 Informa UK Limited, trading as Taylor & Francis Group
2 M. BUFFLER ET AL.

genetics, accompanied by greater nutrient demands of sows (Kim et al. 2005; Ball et al.
2008). Furthermore, in global pig production, Fe contents of diets for gestating sows vary
widely. In the US swine industry, the average dietary Fe content is 1.3 times of the
estimated requirement (Flohr et al. 2016). In German farms, this content is exceeded on
average by 430 mg/kg diet (Humann-Ziehank 2014). As maternal Fe deficiency bears the
risk of reduced fertility and stillbirth (Normand et al. 2012), it is important to evaluate
whether the current data on Fe requirement of pregnant sows are still valid. Although
there is a significant amount of literature on the interconnections between alimentary Fe
supply of sows and the Fe status of the offspring, the current knowledge on Fe demands of
modern sow genotypes is scarce. Therefore, the present study investigated the effects of
different alimentary Fe supply to pregnant sows on their reproductive performance as well
as the Fe status of their offspring.

2. Materials and methods

2.1. Experimental design
This study was conducted at the research facility for livestock Thalhausen (TU
Munich). The experiment was registered and approved by the District Government of
Upper Bavaria of the Federal State of Bavaria (AZ 55.2-1-54-2532-84-13).
The experiment was carried out in two trial periods. In each period, 12 sows
(German Landrace) on heat were allocated to two feeding groups according to an
equal average gestation number (n = 6 sows per group and period). Subsequently, all
sows were inseminated. The animals were housed in gestation crates for 4 weeks until
gestation could be confirmed by ultrasound diagnosis. Because in each group one sow
of was not successfully inseminated, the effective treatment group size amounted to
n = 5 for each of the two experimental periods. Next, pregnant sows were stabled alone
or pair-wise in fattening pens to allow individual feeding. Coprophagy was avoided by
permanent observation of the sows and immediate removal of faeces in the pens. One
week before farrowing, the sows were moved into separate farrowing pens.
The experimental feeding period lasted from insemination to farrowing. During this
time, sows were fed isoenergetic and isonitrogenic diets which differed in Fe content
(low Fe diet vs. high Fe diet). The group low in Fe (Group −Fe) received the basal diet
composed mainly of corn and soybean meal (Table 1) with a native Fe content of
114 mg/kg. The basal diet was designed to meet all nutrient demands including a Fe
supply of 90 mg/kg DM according to GfE (2006) and NRC (2012). In the group with
high Fe content (Group +Fe), the basal diet was supplemented with 637 mg Fe(II)
SO4 · 7H2O per kg diet to reach a total Fe content of 256 mg/kg DM. From insemina-
tion to day 84, the daily feed intake was restricted to 2.5 kg and from day 85 to
farrowing to 3.2 kg. Animals had free access to drinking water.

2.2. Reproductive performance parameters

After farrowing, the litter sizes of each sow as well as number of piglets born alive were
determined. All piglets were weighted immediately after birth and data was used to
calculate the respective litter weights.
 ARCHIVES OF ANIMAL NUTRITION 3

Table 1. Ingredients and metabolisable energy content of the basal diet.

Ingredient Contents (% in dry matter)
Corn 77.36
Cellulose 6.50
Soybean meal (40% CP) 12.00
L-Lysine 0.15
DL-Methionine 0.12
L-Tryptophan 0.02
Soybean oil 1.00
NaCl 0.55
CaCO3 0.75
Ca(H2PO4) 0.80
Premix† 0.75
Metabolisable energy‡ [MJ/kg] 12.86
†
Provided per kg dietary DM: 829 mg MgO; 24 mg CuSO4 · 5H2O; 62 mg MnSO4 · H2O;
0.4 mg Na2SeO3 · 5H2O; 0.7 mg KI; 244 mg ZnSO4 · H2O; 16 mg vitamin A; 0.8 mg
vitamin D3; 60 mg vitamin E; 0.6 mg vitamin K3; 3.4 mg vitamin B1; 10.5 mg vitamin
B2; 22 mg niacin; 26 mg pantothenic acid; 3 mg vitamin B6; 34 µg vitamin B12;
22.0 mg biotin; 2.9 mg folic acid; 2.4 g choline chloride (contents of all vitamins and
trace elements met the requirements according to NRC 2012); ‡Estimated from http://
datenbank.futtermittel.net/.

2.3. Sampling conditions

Two aliquots of each diet (500 g) were sampled in polyethylene bottles and stored
at −20°C. Faecal samples of sows were collected on the day of farrowing and
stored in plastic bags at −20°C. Placenta tissue was packed into plastic bags
directly after birth and frozen at −20°C. At the day of insemination and the day
of farrowing, from sows blood samples (2 × 9 ml) were taken from vena jugularis.
Blood was collected in serum tubes (S-Monovette Z-Gel, Sarstedt AG Co,
Nürnbrecht, Germany) and lithium heparin tubes (S-Monovette Lithium
Heparin, Sarstedt AG &Co; 16 IU heparin/ml of blood), which were free of trace
element contaminants. Serum tubes were centrifuged at 3.000 g for 15 min and
serum was stored in 1.5 ml Eppendorf tubes at −20°C.
Immediately after birth, from each piglet 1 ml of blood was collected from vena
cava cranialis with heparinised syringes and stored at 4°C. One piglet per litter was
euthanised at day 1 postpartum by anaesthesia (azaperone, ketamine) and bleeding.
Blood samples (2 × 9 ml) were collected in serum and lithium heparin tubes as
described earlier. Liver tissue of these piglets was collected in plastic bags, evacuated
and stored at −20°C.

2.4. Crude nutrient analysis and apparent retention of Fe

DM and crude nutrient contents in diets were determined by proximate analysis (crude
protein [Kjeldahl]: method 954.01; ether extract: method 920.39; crude ash: method 942.05;
crude fibre: method 930.10) (AOAC 2005). The contents of metabolisable energy of both
diets were calculated on the basis of the DLG database (http://datenbank.futtermittel.net/)
using listed metabolisable energy (ME) contents of the single feedstuffs. DM digestibility as
well as daily apparent Fe retention was determined according to the indicator dilution
method using lignin as native marker. Lignin contents in feed and faeces were determined
by fibre detergent analysis (method 973.18) according to AOAC (2005).
4 M. BUFFLER ET AL.

2.5. Fe content analysis

Fe contents of drinking water, feed, faeces, placenta tissue, and liver tissue were deter-
mined by atomic absorption spectrometry (AAS; novAA 350, Analytik Jena AG, Jena,
Germany). Before analysis, faeces, placenta, and liver were freeze-dried for 72 h and the
DM contents were assessed. Dried samples were pulverised using a pestle in a ceramic
mortar. Extraction was performed in duplicate using 1 g pulverised material mixed with
5 ml double-distilled H2O, 6.25 ml HNO3 (65%), and 3 ml H2O2 (30%). This mixture was
wet extracted using a microwave system (Ethos 1, MLS GmbH, Leutkirch, Germany). The
extracts were made up to a total volume of 50 ml with double-distilled H2O and
measured directly. For calibration, a certified AAS standard material was used (Merck,
119781). Fe content in drinking water was measured directly without extraction process.

2.6. Fe content and total Fe-binding capacity in serum

Serum Fe content was assessed photometrically using the commercial Iron Ferene Kit
(Bioanalytic GmbH, Umkirch, Germany) according to the manufacturer’s guidelines. In
brief, per sample 100 µl serum was mixed with reducing acetic acetate buffer to dissolve
Fe3+ from transferrin and to reduce it to Fe2+. A chromogenic reagent was added and
the extinction was measured at 590 nm. A Fe standard (140 µg Fe3+/dl) included in the
kit served as calibration solution. The serum Fe content was calculated by means of Fe
standard extinction.
Total Fe-binding capacity (TIBC) was determined by using the Fe-binding capacity kit
(Bioanalytic GmbH). In brief, serum transferrin was saturated with Fe3+ solution and
excess Fe was precipitated with magnesium hydroxide carbonate. Subsequently, the Fe
content within the supernatant was measured photometrically using the Iron Ferene Kit
as described earlier. For calculation, the Fe standard (140 µg Fe3+/dl) was measured and a
dilution factor of 2.7 due to the precipitation step was taken into account. Transferrin
saturation was calculated as the ratio from serum Fe content to TIBC.

2.7. Haemoglobin and haematocrit in blood

Haemoglobin (Hb) contents were determined by the cyanohaemoglobin method mod-
ified from DIN 58931 using a commercial reagent (Haemoglobin, Bioanalytic GmbH).
A volume of 5 µl of whole blood from lithium heparin tubes was added to 1.25 ml
haemoglobin reagent and mixed gently. After 3 min incubation at room temperature,
extinction (Es) was measured at 540 nm with respect to a blank with pure Hb reagent.
Hb content was calculated as follows:
Haemoglobin ½g=dl ¼ E s  368

Determination of haematocrit (Hct) was performed in Haematocrit 150 Centrifuge

(Hettich, Lab Technology, Tuttlingen, Germany). Lithium heparin blood was soaked in
micro capillaries, closed with sealing compound, and centrifuged for 3 min at 14,000 g.
Haematocrit content was estimated by evaluation desk.
Mean corpuscular haem concentration (MCHC) was calculated according to the
following formula:
 ARCHIVES OF ANIMAL NUTRITION 5

Haemoglobin ½g=dl  100%

MCHC ½g=dl ¼
Haematocrit ½%

2.8. Statistical analysis

Data were analysed with two-way analysis of variance (ANOVA) [experimental period
(1, 2), diet (−Fe, +Fe)], using the GLM procedure in the SAS 9.3 software package (SAS
Institute Inc., Cary, NC, USA). Significantly different means were tested with Student–
Newman–Keuls method at a 5% significance level.

3. Results
3.1. Fe supply and digestibility
The sows of the two feeding groups (n = 10) had approximately identical average gestation
numbers of 4.9 and 5.4, respectively. During the entire study, the feed portions were
consumed completely by the experimental animals. Daily Fe intake in early pregnancy
was 258 mg in Group −Fe and 578 mg in Group +Fe. In late pregnancy, daily Fe intake
increased in Groups −Fe and + Fe to 329 mg and to 738 mg, respectively (Table 2). In the
whole period of gestation, the sows in Groups −Fe and +Fe ingested 32 and 71 g Fe,
respectively. Fe content of drinking water was virtually zero (0.047 ± 0.0047 µg/l).
Faeces’ Fe contents differed significantly between groups (p < 0.001). Group −Fe
excreted 814 mg Fe/kg faeces, whereas faeces Fe content of Group +Fe was 2.1-fold higher
(1.675 g/kg). DM digestibility of the feed was on average 85% with no differences between
groups (85.2% vs. 85.3%). Based on this data, daily faecal Fe excretion was calculated. The
sows of Group −Fe excreted on average 342 mg Fe/d, whereas Group +Fe excreted 708 mg
Fe/d (p < 0.001; Table 3). Calculated apparent Fe retention in Group −Fe tended to be lower
than in Group +Fe (−13 mg/d vs. 30 mg/d; p < 0.1).

3.2. Fe status parameters in blood of sows

All sows exhibited comparable levels of blood Fe status parameters at the beginning of
gestation (Table 3). Furthermore, despite different feeding regimes, no differences
between Groups −Fe and +Fe were evident on day 115 of pregnancy.

Table 2. Analysed crude nutrient and iron (Fe) contents in experimental diets.
Experimental diets
−Fe† +Fe‡
Dry matter (DM) [%] 89.5 89.8
Crude protein [% in DM] 12.1 12.3
Ether extract [% in DM] 4.2 4.2
Crude ash [% in DM] 3.7 3.9
Crude fibre [% in DM] 6.9 6.8
Lignin [% in DM] 2.3 2.3
Fe content [mg/kg as-fed] 103 231
Fe content [mg/kg DM] 114 256
† ‡
Basal diet containing low Fe content; Basal diet supplemented with 637 mg
Fe(II)SO4 · 7H2O/kg diet.
6 M. BUFFLER ET AL.

Table 3. Effects of different iron (Fe) supply on Fe retention and blood parameters of sows during
gestation.
Experimental groups
−Fe +Fe SEM# p-Value
†
Fe intake [mg/d] 329 738 – –
Faecal Fe excretion [mg/d]‡ 342 708 27 <0.001
Apparent Fe retention [mg/d] −13 30 27 0.090
Haemoglobin [g/dl]
Day 1 of gestation 133.5 129.8 4.2 0.072
Day 115 of gestation 119.3 123.9 7.4 0.405
Alteration days 1 to 115 −14.9 −5.9 7.5 0.413
Haematocrit [%]
Day 1 of gestation 40.8 40.2 1.3 0.647
Day 115 of gestation 35.4 35.7 1.9 0.734
Alteration days 1 to 115 −6.3 −4.4 2.1 0.359
MCHC◊ [g/dl]
Day 1 of gestation 32.9 32.3 1.2 0.803
Day 115 of gestation 33.7 34.8 1.3 0.432
Alteration days 1 to 115 1.1 2.5 2.2 0.734
Serum Fe content [µg/dl]
Day 1 of gestation 145.3 138.5 12.7 0.175
Day 115 of gestation 128.3 130.4 10.5 0.719
Alteration days 1 to 115 −6.7 −6.2 11.0 0.167
TIBC¶ [µg/dl]
Day 1 of gestation 641.4 641.2 43.4 0.943
Day 115 of gestation 540.1 553.6 21.0 0.612
Alteration days 1 to 115 −101.3 −80 37.1 0.773
Transferrin saturation [%]
Day 1 of gestation 23.2 21.7 2.9 0.257
Day 115 of gestation 23.2 23.4 1.5 0.770
Alteration days 1 to 115 2.8 1.8 1.9 0.414
†
Days 85–115 of pregnancy at a feeding level of 3.2 kg/d; ‡Calculation based on dry matter digestibility and analysed Fe
content in faeces at day 115 of pregnancy; #SEM, standard error of means representing the pooled standard error of
the respective general linear model; ◊MCHC, mean corpuscular haem concentration; ¶TIBC, total iron-binding capacity.

Hb and haematocrit decreased numerically throughout gestation (−14.9 vs. −5.9 g/dl
and −6.3 vs. −4.4%, respectively) in both groups; however, these trends missed the
threshold of statistical significance. MCHC rose numerically in both groups by 3.3%
and 7.6%, respectively.
Serum Fe contents declined almost identically by 6.7 and 6.2 µg/dl in Groups −Fe and +Fe,
respectively (Table 3). The main shift was observed in TIBC, with a numerically reduction by
a factor of 0.83 (Group −Fe) and 0.86 (Group +Fe) at the end of pregnancy compared with
insemination date, respectively.

3.3. Placental Fe content, reproductive performance and Fe status parameters in

piglets
Fe contents in placental tissue differed significantly between the two feeding groups. In
Group −Fe, the average Fe content was at 276.8 µg/g DM (Table 4), whereas content in
Group +Fe was 1.7-fold higher (p < 0.05; Table 4).
With an average litter size of 9.8 piglets, Group −Fe showed significantly lower
offspring numbers than in Group +Fe (12.1 piglets; Table 4). This difference between
groups was identical with the number of piglets born alive. Average piglet weight at
 ARCHIVES OF ANIMAL NUTRITION 7

Table 4. Effects of different iron (Fe) supply during pregnancy on placental Fe content, reproductive
performance, and blood parameters of the offspring.
Experimental groups
−Fe +Fe SEM# p-Value
Placental Fe content [µg/g DM] 276.8 462.0 42.2 0.026
Litter size 9.8 12.1 0.9 0.016
Piglets born alive 9.0 11.4 0.9 0.017
Piglets born alive [%] 92.8 91.1 7.3 0.778
Piglet weight [kg] 1.51 1.57 0.06 0.050
Litter weight [kg] 15.5 19.8 1.6 0.035
Haemoglobin [g/dl] 80.2 89.3 5.0 0.513
Haematocrit [%] 27.0 31.3 1.3 0.632
MCHC† [g/dl] 30.1 28.6 1.1 0.718
Serum Fe content [µg/dl] 39.2 40.7 12.1 0.930
TIBC‡ [µg/dl] 147 117 16 0.466
Transferrin saturation [%] 29.6 39.9 14.8 0.610
Liver Fe content [µg/g DM] 500 809 183.7 0.152
SEM, standard error of means (representing the pooled standard error of the respective general linear model); †MCHC,
#

mean corpuscular haem concentration; ‡TIBC, total iron-binding capacity.

birth was higher in Group +Fe (1.51 vs. 1.57 kg). Due to different litter sizes and piglet
weights, the total litter weight in Group +Fe was 4.3 kg greater than in Group −Fe
(p < 0.05).
Although blood analyses revealed numerically differences, the levels of haemoglobin,
haematocrit, MCHC, serum Fe, TIBC, as well as transferrin saturation of the piglets
were not significantly influenced by the experimental feeding of sows (Table 4). Also, Fe
stores in the liver of piglets were not significantly different between both groups.
Nevertheless, numerically it could be shown that liver Fe content in Group −Fe was
nearly 40% lower than in Group +Fe (500 vs. 809 µg/g DM).

4. Discussion
The present study investigated the effects of varying Fe supply strategies during the
gestation of sows on their reproductive performance and Fe status as well as the Fe
status of their progeny. For this purpose, two groups of multiparous sows were fed with
a basal diet based on corn and soybean meal. These components are commonly used in
practical feeding and are known to be rich in phytate which binds dietary Fe as
insoluble complexes and thereby decreases Fe absorption in the duodenum (De
Boland et al. 1975). No phytase was supplemented and, in addition, native phytase
activity was inhibited by heating in the pelleting process to keep Fe bioavailability on a
low level. The analysed average native Fe content of the basal diet (114 mg Fe/kg DM)
was still above the recommended Fe supply of 80 mg/kg diet (at 90% DM; NRC 2012)
or 90 mg/kg DM (GfE 2006). In practical feeding, sow diets exhibit Fe contents far
above recommendations ranging between 200 mg and 600 mg/kg DM (Humann-
Ziehank 2014; Flohr et al. 2016). Therefore, Group +Fe was supplemented with
120 mg Fe/kg diet to supply sows with Fe contents common in practice.
Previous studies in pregnant sows investigated the influence of either excess
(Kirchgessner and Pallauf 1973) or deficient (Roth-Maier et al. 1985) Fe supply on
reproductive performance. None of the models are appropriate to examine the actual
8 M. BUFFLER ET AL.

demand, because the sows are exposed to extreme physiological challenges with high
risks to the offspring (Normand et al. 2012). In contrast, the experimental approach
applied in the present study avoids drastic metabolic dysbalances in response to severe
alimentary Fe deficiency or overload. Therefore, it is possible to investigate the effects of
differential Fe supply strategies to sows under physiological conditions.
In both groups of sows, haemoglobin and haematocrit values were within the
published reference ranges (Friendship and Henry 1992) and did not differ significantly,
neither at the beginning nor at the end of pregnancy. The numerical decrease in
haemoglobin and haematocrit values during pregnancy is not mandatory affected by
nutritional factors, especially Fe supply levels, but also by an increase of the total plasma
volume of up to 25% during pregnancy (Matte and Girard 1996). In this study, serum
Fe contents of the sows also did not differ between both sampling times (Calvo et al.
1989) and are within the physiological reference range during the whole pregnancy
(Kraft et al. 1999). This parameter is influenced by nutritional and environmental
factors such as feed intake and daytime and shows diurnal variations of about 50%
(Chikazawa et al. 2013). Taken together, the serum data suggest that despite different Fe
supply to pregnant sows, no significant effects on blood parameters became evident.
This strongly questions the usability of such measures to serve as diagnostic parameters
to detect alimentary Fe deficiency in pregnant sows.
No mechanisms of active Fe excretion are known and endogenous losses play a
minor role (Conrad et al. 1964; Hentze et al. 2010). Hence, apparent Fe retention may
potentially indicate Fe demand. The reference values for Fe retention, however, are not
consistent within literature. While some authors reported very low retention of 26 mg/d
(Roth-Maier et al. 1985), other authors showed Fe retention rates in the range of
130 mg/d (Cao and Chavez 1995). Furthermore, the absorption of trace elements
decreases during the final week of pregnancy along with the maternal preparation for
the upcoming farrowing (Kirchgessner et al. 1980). This may cause an underestimation
of Fe retention when based on faecal samples collected during the final stage of
pregnancy. As in the present study, the faecal samples were derived at the day of
farrowing, the negative Fe retention observed in −Fe sows seems to reflect such an
underestimation rather than an actual mobilisation of maternal Fe stores. Nevertheless,
sows of Group +Fe tended to exhibit a higher Fe retention (+45 mg/d compared with
Group −Fe; p < 0.1). This difference seems to reflect a higher metabolic Fe requirement
of Group +Fe sows due to the larger litter size and/or an insufficient dietary Fe supply
in −Fe sows. The latter might have resulted from the low bioavailability of native Fe in
plant-based diets (around 5–12%; Hurrell and Egli 2010), particularly in presence of
phytate-rich feedstuffs and absence of relevant phytase activities as it was the case in our
study. Indeed, +Fe sows were supplemented with FeSO4, which is known for its high
bioavailability (Allen et al. 2006; Ettle et al. 2008). But since Fe homoeostasis should
have channelled any dietary Fe quantities exceeding the requirements towards the faecal
route of excretion, the different Fe retentions observed in the present study seem to
directly indicate respective differences in Fe status levels between sows of Groups −Fe
and +Fe.
Taken together, these results indicate that Fe requirements in pregnant sows prob-
ably have gradually increased during the last 30 years. The numerically increased
retention in Group +Fe confirms the hypothesis that current recommendations may
 ARCHIVES OF ANIMAL NUTRITION 9

not be appropriate for modern sow genotypes. In addition, the differences in Fe

retention between groups highlight the daily Fe retention to be a more sensitive
indicator of Fe supply than blood parameters.
Within the homoeorhetic hierarchy of Fe delivery during pregnancy, requirements
of the foetuses rank first followed by maternal red blood cells and Fe stores of the
sow (Cetin et al. 2011). Therefore, foetal Fe level and, hence, foetal hepcidin
expression are key regulators for Fe transfer through the placenta by different
molecular mechanisms (Lipinski et al. 2013). Under the assumption that the current
recommendations including an additional safety margin are appropriate to cover the
demand of pregnant sows, the regulatory system should maintain uterine Fe transfer
in both groups on a similar level. However, the reduced Fe content in placenta of
Group −Fe compared with Group +Fe suggests that the daily alimentary Fe is
insufficient, although placental Fe transporter are up-regulated in maternal Fe defi-
ciency (Gambling et al. 2001). Within the hierarchy of Fe distribution, sows mobilise
endogenous Fe to meet the offspring’s demand (Mahan 1990) and maintain maternal
red blood cells. This is also in line with the results of blood parameters which are
only marginally affected in Group −Fe. Nevertheless, it can be hypothesised that the
mobilisation of maternal endogenous Fe stores of the Group −Fe was not able to
fully compensate the insufficient alimentary Fe intake and, hence, the Fe demand of
the foetuses was not satisfied (Gambling et al. 2009). For further understanding, the
maternal liver Fe contents have to be analysed.
Roth-Maier et al. (1985) showed that severe Fe deficiency impaired reproductive
performance, when sows were fed with dietary Fe contents far below recommended
levels (40 mg/kg). As the sows in this study were repleted with Fe pre-experimentally,
the reduced litter size and fewer piglets born alive suggests that Fe status of the sow in
early pregnancy is less critical than the influence of Fe-deficient nutrition during this
time. Similar results are indicated in the current study. Although the average numbers
of pregnancy and previous litter sizes were equal in both groups (data not shown) and
dietary Fe contents are above the current recommendations, Group −Fe had fewer
piglets per litter and reduced piglet weights. This leads to the assumption of an under-
estimation of the sows Fe demand when suggesting levels of 80–90 mg/kg DM for
modern pig lines. Different Fe supply of the sow at the beginning of pregnancy (first
trimester) had a clear influence on the number of developing foetuses but minor effects
on the prenatal weight development of piglets itself, which has been already described
by Venn et al. (1947). Insufficient essential trace element supply in early pregnancy
increases the risk of embryonic death (Hostetler et al. 2003), which is consistent with
the lower litter sizes in the −Fe.
Several other authors demonstrated the absence of effects after different dietary Fe
supply on litter size and number of live born piglets, but the majority of these
experimental models varied Fe supplementation to the sows not from the beginning
but only from the second trimester of gestation (Kirchgessner and Pallauf 1973; Roth-
Maier et al. 1985) or just a few weeks before farrowing (O’Connor et al. 1989; Egeli et al.
1998; Petrichev and Bambova 2005). Even the results of similar experimental designs
are ambivalent. Whereas Guise and Penny (1990) demonstrated significant more living
piglets after Fe injection 3 weeks ante partum, according to Egeli et al. (1998),
parenteral Fe supplementation at the same period of pregnancy had no effects on
10 M. BUFFLER ET AL.

reproductive performance. The nutrient demand of sow and offspring increases during
the exponential embryonic growth in the second half of pregnancy (McPherson et al.
2004). A lack in nutrient supply during this period induces foetal growth retardation
and lower birth weights (Campos et al. 2012). Therefore, Fe application at mid-
pregnancy and during the period of strongest foetal growth in the last trimester
improves piglet weights (Buchanan and Bolin 1949; Petrichev and Bambova 2005).
Although group size (n = 10) of this study is below the required number per treatment
(<20) to generate reliable data for interpretation (Hays and Aaron 2000), significant
differences in the litter size and also in total litter weight were evident. Nevertheless,
further experiments have to be done to evaluate these results even if there are no
available data based on large-scale experimental designs.
However, not only reproductive performance of the sow but also the Fe status of the
offspring may be affected by maternal Fe supply. While maternal Fe dowry to the
foetuses cannot be improved by excessive Fe supplementation due to homoeostatic
counter-regulations, dietary Fe deficiency during pregnancy depresses liver Fe stores of
new-born piglets (Chaney and Barnhart 1963; Kirchgessner and Pallauf 1973; Egeli et al.
1998). Also in the present study, liver Fe was numerically by 38% lower in −Fe
compared with the offspring of Group +Fe. Although statistically not evident from
ANOVA, lower liver Fe corresponded to lower haemoglobin, haematocrit, and placen-
tal Fe contents, as well as to higher TIBC levels. Respective Pearson correlation
coefficients were statistically significant for haematocrit (r = 0.67, p < 0.008) and
placental Fe content (r = 0.53; p < 0.048), supporting the hypothesis of an impaired
Fe dowry from the −Fe sows to their offspring due to an insufficient dietary Fe supply
(Spruill et al. 1971; O’Connor et al. 1989).

5. Conclusions
In this study, the effects of varying alimentary Fe supply of sows on their offspring
could be demonstrated, although both diets had Fe contents slightly above current
recommendations. Sows that fed a low Fe diet showed lower uterine Fe transfer, which
was reflected in different placental Fe contents compared with sows that fed a high Fe
diet. The clearest effects of the varying alimentary Fe supply were decreased litter size
and reduced birth weight of new-born piglets in Group −Fe. This indicates an urgent
need to re-evaluate Fe requirements and feeding recommendations of modern sows,
like it is already in progress for other nutrients. Moreover, it was evident that the
maternal blood parameters are not suitable to evaluate foetal Fe supply; therefore,
analysis of Fe stores of the sows would be necessary.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Bayerische Arbeitsgemeinschaft Tierernährung (BAT e.V.) and
the Hildegard Grunow Foundation.
 ARCHIVES OF ANIMAL NUTRITION 11

References
Allen L, De Benoist B, Dary O, Hurrell R, editors. 2006. Guidelines on food fortification with
micronutrients. Geneva: World Health Organization, Food and Agricultural Organization of
the United Nations.
[AOAC] Association of Analytical Communities. 2005. Official methods of analysis of AOAC
International. 18th ed. Arlington, VA: Assoc Off Anal Chem.
Ball RO, Samuel R, Moehn S. 2008. Nutrient requirements of prolific sows. Adv Pork Prod.
19:223–236.
Buchanan M, Bolin D. 1949. Anemia in suckling pigs. Bi-Monthly Bulletin, Dakota Agric Exp
Stn. 11:106–107.
Calvo JJ, Allue JR, Escudero A, Garcia LJ. 1989. Plasma ferritin of sows during pregnancy and
lactation. Cornell Vet. 79:273–282.
Campos P, Silva B, Donzele J, Oliveira R, Knol E. 2012. Effects of sow nutrition during gestation
on within-litter birth weight variation: a review. Animal. 6:797–806.
Cao J, Chavez ER. 1995. Comparative trace mineral nutritional balance of first-litter gilts under
two dietary levels of copper intake. J Trace Elem Med Biol. 9:102–111.
Cetin I, Berti C, Mandò C, Parisi F. 2011. Placental iron transport and maternal absorption. Ann
Nutr Metabol. 59:55–58.
Chaney C, Barnhart C. 1963. Effect of iron supplementation of sow rations on the prevention of
baby pig anemia. J Nutr. 81:187–192.
Chikazawa S, Hori Y, Kanai K, Ito N, Hoshi F, Orino K, Watanabe K, Higuchi S. 2013. Factors
influencing measurement of serum iron concentration in dogs: diurnal. J Vet Med Sci.
Dec;75:1615–1618.
Conrad ME, Weintraub LR, Crosby WH. 1964. The role of the intestine in iron kinetics. J Clin
Invest. 43:963.
De Boland AR, Garner GB, O’Dell BL. 1975. Identification and properties of phytate in cereal
grains and oilseed products. J Agric Food Chem. 23:1186–1189.
Egeli AK, Framstad T, Gronningen D. 1998. The effect of peroral administration of amino acid-
chelated iron to pregnant sows in preventing sow and piglet anaemia. Acta Vet Scand. 39:77–
87.
Ettle T, Schlegel P, Roth F. 2008. Investigations on iron bioavailability of different sources and
supply levels in piglets. J Anim Physiol Anim Nutr. 92:35–43.
Flohr JR, Derouchey JM, Woodworth JC. 2016. A survey of current feeding regimens for
vitamins and trace minerals in the US swine industry. J Swine Health Prod. 24:290–303.
Friendship RM, Henry SC. 1992. Cardiovascular System, hematology and clinical chemistry. 7th
ed. Ames, Iowa: Iowa State University Press.
Gambling L, Czopek A, Andersen HS, Holtrop G, Srai SK, Krejpcio Z, Mcardle HJ. 2009. Fetal
iron status regulates maternal iron metabolism during pregnancy in the rat. Am J Physiol
Regul Integr Comp Physiol. 296:R1063–1070.
Gambling L, Danzeisen R, Gair S, Charania Z, Solanky N, Joory KD, Srai SKS, Mcardle HJ. 2001.
Effect of iron deficiency on placental transfer of iron and expression of iron transport proteins
in vivo and in vitro. Biochem J. 356:883–889.
[GfE] Gesellschaft für Ernährungsphysiologie. 2006. Empfehlungen zur Energie- und
Nährstoffversorgung bei Schweinen [Recommendations for the supply of energy and nutrients
to pigs]. Frankfurt am Main: DLG-Verlag.
Guise HJ, Penny RH. 1990. Influence of supplementary iron in late pregnancy on the perfor-
mance of sows and litters. Vet Rec. 127:403–405.
Hays VW, Aaron DK. 2000. Statistical techniques for the design and analysis of swine nutrition
experiments. Swine nutrition, second edition. Boca Raton (FL): CRC Press.
Hentze MW, Muckenthaler MU, Galy B, Camaschella C. 2010. Two to tango: regulation of
mammalian iron metabolism. Cell. 142:24–38.
Hostetler CE, Kincaid RL, Mirando MA. 2003. The role of essential trace elements in embryonic
and fetal development in livestock. Vet J. 166:125–139.
12 M. BUFFLER ET AL.

Humann-Ziehank E. 2014. Zur Regulation des Eisenstoffwechsels durch Hepcidin und seiner
Bedeutung für Tierernährung und Tiermedizin [Regulation of iron metabolism by hepcidin
and its importance in animal nutrition and veterinary medicine]. Übers Tierernährg. 42. DLG
Verlag, Frankfurt am Main, Germany.
Hurrell R, Egli I. 2010. Iron bioavailability and dietary reference values. Am J Clin Nutr. 91:3.
Kim S, Wu G, Baker D 2005. Amino acid nutrition of breeding sows during gestation and
lactation. Pig News Info. CABI. 26:89N–99N.
Kirchgessner M, Pallauf J. 1973. Zum Einfluß zusätzlicher Eisengaben an Muttersauen ante
partum als Anämieprophylaxe bei Saugferkeln. [Influence of iron supplementation to gestating
sows as anemia prophylaxis for suckling piglets.]. Züchtungskunde. 45:245–248.
Kirchgessner M, Sporl R, Roth-Maier DA. 1980. Fecal excretion and apparent absorption of
copper, zinc, nickel and manganese of nongravid and gravid sows with different dietary supply
of trace elements. J Anim Physiol Anim Nutr. 44:98–111.
Kraft W, Dürr UM, Fürll M, Bodstedt H, Heinritzi K. 1999. Hematology. In: Kraft W, Dürr UM,
editors. Clinical laboratory diagnostics in veterinary medicine. 6th ed. Stuttgart: Schattauer
Verlag; p. 43–77.
Lipinski P, Stys A, Starzynski RR. 2013. Molecular insights into the regulation of iron metabo-
lism during the prenatal and early postnatal periods. Cell Mol Life Sci. 70:23–38.
Mahan DC. 1990. Mineral nutrition of the sow: a review. J Anim Sci. 68:573–582.
Matte J, Girard C. 1996. Changes of serum and blood volumes during gestation and lactation in
multiparous sows. Can J Anim Sci. 76:263–266.
McPherson RL, Ji F, Wu G, Blanton jr JR, Kim SW. 2004. Growth and compositional changes of
fetal tissues in pigs. J Anim Sci. 82:2534–2540.
[NRC] National Research Council. 2012. Nutrient requirements of swine: 11th revised edition.
Washington, DC: The National Academies Press.
Normand V, Perrin H, Auvigne V, Robert N, Laval A. 2012. Anaemia in the sow: a cohort study
to assess factors with an impact on haemoglobin concentration, and the influence of haemo-
globin concentration on the reproductive performance. Vet Rec. 171:8.
O’Connor DL, Picciano MF, Roos MA, Easter RA. 1989. Iron and folate utilization in reprodu-
cing swine and their progeny. J Nutr. 119:1984–1991.
Petrichev MH, Bambova M. 2005. The effects of oral administration of iron methionate to
pregnant sows and their litters. Folia Veter. 49:125–128.
Roth-Maier DA, Kirchgessner M, Spoerl R. 1985. Fe balances of pregnant and lactating breeding
sows by different alimentary iron administration. Zentralbl Veterinärmed A. 32:739–751.
Spruill DG, Hays VW, Cromwell GL. 1971. Effects of dietary protein and iron on reproduction
and iron-related blood constituents in swine. J Anim Sci. 33:376–384.
Venn JA, McCance RA, Widdowson EM. 1947. Iron metabolism in piglet anaemia. J Comp
Pathol Ther. 57:314–325.