Ferric citrate is a safe and digestible source of iron in broilers and piglets

Background Iron (Fe) is traditionally supplemented in poultry and swine diets using inorganic forms (e.g. sulfates, oxides). However, research suggests that organic sources are more beneficial due to greater bioavailability. In this paper, we present results from four studies aimed at assessing ferric citrate (CI-FER™, Akeso Biomedical Inc., Burlington, MA, USA) as a safe and effective source of Fe for broilers and piglets. Methods A total of four studies were performed in Germany following standard farming practices for each species. One study in day-old broiler chicks and one study in weaned piglets were designed as target animal safety studies where animals were randomly allocated to one of three treatment groups: a negative control group, the proposed dose group and a multifold dose group. Broilers and pigs were fed the experimental diets for 35 and 42 days, respectively. In each study, average daily feed intake, average daily weight gain and feed conversion ratio were measured, and blood samples were taken at study end for routine biochemistry and haematology. The other two studies were designed to evaluate different sources of dietary Fe for weaned piglets bred and managed under standard farm conditions. All piglets received routine Fe injections (200 mg Fe dextran, intramuscular) on day 3 of age, as well as the experimental diets for 42 days. In both studies, performance parameters were measured. In one study, Fe digestibility and serum Fe, superoxide dismutase and haptoglobin were also measured. In all studies, the general health status of the animals was monitored daily and all culls and mortality recorded. Each study followed a complete randomised block design. Results In broilers, ferric citrate was well tolerated up to 2,000 mg/kg feed (×10 the recommended inclusion rate) and no adverse effects on growth, blood parameters or mortality were observed. In piglets, ferric citrate was well tolerated up to 5,000 mg/kg feed (×10 the recommended inclusion rate) with no adverse effects on growth, blood parameters or mortality. In addition, piglets fed ferric citrate performed significantly better than animals fed the negative control diet (containing only endogenous Fe) and those fed inorganic forms of Fe. Moreover, piglets fed ferric citrate demonstrated improved Fe digestibility and improved oxidative status. Altogether, these findings show that ferric citrate is a safe and easily digestible source of dietary Fe for broilers and piglets.


INTRODUCTION
Intensive farming production systems require farmers to pay careful attention to the main factors that determine profitability, such as genetics, health, management and nutrition. In relation to the latter, iron (Fe) as a trace mineral is particularly important since it is involved in oxygen and electron transport, deoxyribonucleic acid (DNA) synthesis, energy metabolism and gene regulation, among others (Beard, 2001). In poultry, the fast-growing modern broiler is sensitive to dietary deficiencies of Fe which can cause severe anaemia with reduction of packed cell volume (PCV), haemoglobin and body weight (Tako, Rutzke & Glahn, 2010;Marks & Kendall, 2020). In suckling piglets, it is well established that insufficient Fe intake results in Fe deficiency, anaemia and increased early morbidity and mortality (Kernkamp, Clawson & Ferneyhough, 1962). Young pigs are commonly at risk of Fe deficiency as the piglet is born with a limited Fe store of approximately 50 mg of Fe, most of which is incorporated in haemoglobin (Venn, McCance & Widdowson, 1947). Sow milk is a poor source of Fe and only provides sucking piglets with approximately 1 mg of Fe per day (Venn, McCance & Widdowson, 1947). Moreover, the modern pig is selected for rapid growth and piglets double their weight in the first seven days of life, requiring increased Fe as a larger volume of blood is needed to support and maintain tissue growth (Peters & Mahan, 2008). The daily Fe requirement for piglets is ca. 7 mg and so the limited store of Fe at birth, coupled with the limited Fe obtained from sow milk, means that, without exogenous Fe supplementation, fast-growing piglets become Fe deficient within the first week of life (Venn, McCance & Widdowson, 1947;Murphy, Friendship & Dewey, 1997;Starzynski et al., 2013). Intramuscular injection with Fe dextran (200 mg) in the first three days is used routinely to prevent Fe deficiency in piglets (Svoboda & Drabek, 2005). However, the introduction of new genetic lines and management practices have raised questions about the efficacy and safety of this method with some authors reporting that additional Fe supplementation is required depending on Fe requirements and the availability of Fe sources (Svoboda, Vanhara & Berlinska, 2017).
Traditionally, trace minerals such as Fe are supplemented using inorganic forms (e.g. sulfates, oxides) but research suggests that the use of organic sources is beneficial due to their greater bioavailability, given that organic compounds can reduce interference from agents that form insoluble complexes with ionic trace elements (Abdallah, El-Husseiny & Abdel-Latif, 2009;Kumar-Mohanta & Kumar-Garg, 2014). This hypothesis is supported by studies such as those of Shinde et al. (2011) and Arnaudova-Matey et al. (2013), who showed that, compared with inorganic forms, broiler diets containing organic sources of Fe increase Fe concentration in different tissues and reduce Fe excretion. Earlier studies in pigs also demonstrated organic forms of Fe to be safe and efficacious (Furugouri & Kawabata, 1975).
In this paper, we present the results from four experiments aimed at assessing the effects of ferric citrate (CI-FER TM ; Akeso Biomedical Inc., Burlington, MA, USA), an organic iron chelate, as a safe and effective source of Fe for broilers and weaned piglets bred and managed under standard farm conditions.

Study B
The objective of this study was to evaluate the safety of ferric citrate as a source of dietary Fe for weaned piglets bred and managed under standard farm conditions. A total of 144 healthy weaned piglets (25 days of age) from a single weaning batch of ca. 310 piglets (Danbred x Piètrain) were used. Piglets received routine Fe injections (200 mg Fe dextran, intramuscular) on day 3 of age (as per standard farming practice) and creep feed (without addition of Fe) from day 8 to 24 of age, respectively. Details of the diet composition are presented in Table 3. The analysed Fe content in the creep feed was 59.4 mg/kg (Table 4). The piglets were allocated to 3 treatment groups (T1 -negative control; T2 -500 mg ferric citrate/kg feed; T3 -5,000 mg ferric citrate/kg feed) balanced as far as possible for body weight (BW), litter origin, and gender in 24 pens (12 male and 12 female pens, 6 piglets per pen, 8 pens per treatment) from 25 to 66 d of age (42-d feeding period). Starter mash was fed for 14 days, and grower mash was fed subsequently, until study end (42 days post-weaning). The basal diets (T1 Control) contained 69.2 and 76.2 mg endogenous Fe/kg feed for starter and grower diets, respectively, and were Fe deficient compared to the recommended 100 mg Fe/kg for growing piglets fed an ad libitum diet (based on 90% dry matter) between 5-25 kg of weight (National Research Council (NRC), 2012, Table 4). The ferric citrate used contained microtracer (MicroTracers, San Francisco, CA USA, at an inclusion rate of 10%) as a marker. Microtraced ferric citrate was added to the basal diets instead of Tixosil Ò (>97% colloidal silica). Feed samples were analysed for dry matter, crude protein, crude fat, ash and Fe using VDLUFA (1993) methods. Details of analysed nutritional values and microtracer recovery are shown in Table 4. Post-weaning piglets had ad libitum access to feed (mash form) and water (drinking bowls). The pens were equipped with solid partitions and slotted floors in a climate-controlled post-weaning house, with 2 rows of pens separated by a central aisle. Pen measured 3.2 × 2.0 m, providing 1.1 m 2 per animal.

Study C
The objective of this study was to evaluate three different ferric compounds as sources of dietary Fe for weaned piglets bred and managed under standard farm conditions. At weaning (25 days of age), 220 healthy piglets from a single weaning batch of ca. 340 piglets (Danbred × Piètrain) were used. Piglets were allocated to four treatment groups in a randomised block design: T1 -negative control; T2 -500 mg ferric lactate (ca. 17% Fe)/kg feed; T3 -500 mg ferric citrate (ca. 17% Fe)/kg feed; and T4 -500 mg ferric tartrate (ca. 20% Fe)/kg feed. Treatments were balanced as far as possible for body weight (BW), litter origin, and gender in 44 mixed-sex pens (5 piglets/pen, males and females in each pen) giving 11 replicate pens per treatment. Basal prestarter and basal starter batches were produced and then split into 4 aliquots prior to adding the Fe sources to T2, T3 and T4 aliquots. Prestarter mash was fed for 14 days, then starter mash was fed for 28 days, until study end at 42 days post-weaning (66 d of age). Piglets were housed in post-weaning units, each unit with pens on either side of a central aisle with ad libitum access to feed (mash form) and water (drinking bowls). The basal diets (T1 Control) contained 59.1 and 75.4 mg endogenous Fe/kg feed for starter and grower diets, respectively, and were Fe deficient compared to the recommended Fe inclusion rate for growing pigs (National Research Council (NRC), 2012). The ferric sources contained microtracers (MicroTracers, San Francisco, CA USA, at inclusion rate of 10%) as markers. Microtraced ferric sources were added to the basal diets instead of Tixosil Ò (>97% colloidal silica). Feed samples were analysed for dry matter, crude protein, crude fat, ash and Fe using VDLUFA (1993) methods. Details of diets composition, analysed nutritional values and microtracer recovery are shown in Tables 5 and 6.

Study D
The objective of this study was to evaluate organic and inorganic sources of Fe as sources of dietary Fe for weaned piglets bred and managed under standard farm conditions. In this study, faecal samples were collected to determine Fe digestibility. At weaning (25 days of age), 175 healthy piglets from a single weaning batch of ca. 280 piglets (Danbred × Piètrain) were selected. Piglets had received routine Fe injections (Fe dextran, 200 mg intramuscular) on day 3 of age and creep feed (without addition of ferrous (II) sulphate) from day 8 to 24 of age, respectively. Weaned piglets were allocated to 5 dietary treatments: T1 -negative control; T2 -85 mg Fe/kg from ferrous (II) sulphate; T3 -85 Fe mg/kg from ferric (III) oxide; T4 -500 mg ferric citrate/kg, supplying 85 mg Fe/kg feed; T5 -500 mg ferric citrate/kg supplying 85 mg Fe/kg feed plus 85 mg Fe/kg feed from ferrous sulphate, (Table 7). Treatments were balanced as far as possible for body weight (BW), litter origin, and gender, and piglets were housed in 35 mixed-sex pens (5 piglets/pen, males and females in each pen, 7 pens per treatment). Fe supplements in treatments T2, T3 and T4 matched normal commercial practice (85 mg Fe/kg feed). Since some pig producers combine both inorganic and organic trace minerals, and to consider the effects of double supplementation, T5 contained ferric citrate and ferrous (II) sulphate, both supplying 85 mg Fe/kg feed. None of the diets exceeded the maximum EU permitted Fe in weaned piglet feeds. Starter mash was fed for   14 days, and grower mash was fed thereafter, until study end (42 days post-weaning). The creep feed contained 56.8 mg endogenous Fe/kg feed and the basal diets (T1 Control) contained 89.6 and 75.4 mg endogenous Fe/kg feed for starter and grower diets, respectively and were Fe deficient compared to the recommended Fe inclusion rate for growing pigs (National Research Council (NRC), 2012, Table 8). The ferric citrate used contained microtracer (MicroTracers, San Francisco, CA USA, at inclusion rate of 10%) as a marker. Microtraced ferric citrate was added to the basal diets instead of Tixosil Ò (>97% colloidal silica). Titanium dioxide was included in the grower feed for faecal digestibility analysis. Feed samples were analysed for dry matter, crude protein, crude fat, ash and Fe using VDLUFA (1993) methods. Details of diets composition, analysed nutritional values and microtracer recovery are shown in Tables 7 and 8. Post-weaning piglets had ad libitum access to feed (mash form) and water (drinking bowls).
For all studies, temperature, relative humidity, lighting and forced ventilation were controlled and followed the breeder recommendations. All study personnel involved in taking measurements were blinded to treatments, while the study director had full knowledge of the study design.
In all studies, the general health status of the animals was monitored daily and culls and mortality recorded. During the daily health checks, if any animal was identified as showing signs of ill health or suffering, they were either culled immediately whereas if the animal was exhibiting symptoms that were potentially treatable, they were given appropriate medication. Any animal that was found dead during the daily rounds, was removed from the pen and examined. The cause of death was established and attributed to either sudden death or non-response to treatment. At study end, all surviving animals were examined by a veterinarian, fed the control diet for a week and, afterwards, returned to the farm of origin. All left-over feeds, test items and animal carcasses were disposed of according to local regulations at study end.     In all studies, samples were collected from the anterior vena cava into different tubes: plain, heparinised for clinical chemistry, and EDTA tubes for blood counts.
For Study D, blood serum was additionally analysed for Fe, superoxide dismutase (SOD) and haptoglobin. SOD is an enzyme that eliminates superoxide radicals and is an important indicator of antioxidation and oxidation levels caused by free radicals created by unbound Fe. Haptoglobin is an acute phase protein that binds to free haem and reduces oxidative activity. Blood cells were measured by flow cytometry, blood concentrations in plasma were determined by using photometry, while sodium, potassium, and chlorides were measured by ionic liquid-polyacrylamide gel electrophoresis.

Zootechnical parameters
Standard zootechnical parameters (body weight, BW; feed intake, FI; weight gain, WG; mortality-corrected feed conversation rate, MFCR) were recorded in all studies for the different study periods.

Fe digestibility (Study D)
On day 42, the apparent total digestibility of crude ash and Fe was determined using titanium (IV) dioxide added at a rate of 3 g/kg feed as an inert indigestible marker. A total of seven replicates from each treatment (pooled faeces collected rectally from 5 piglets per pen, giving a total of seven replicates) were collected.
To calculate the apparent total Fe digestibility, the following formula was used: Total digestibilityð%Þ ¼ 100 À %Marker in feed %Nutrient in faeces Â %Nutrient in faeces %Nutrient in feed Â 100

Statistical analyses
Study A and Study B were based on demonstrating non-inferiority, where the tolerance group (T3) does not perform worse than the untreated control (T1). The null hypothesis being that the animals fed the tolerance dose perform worse that the control animals.
The alternative hypothesis was that the animals fed the tolerance dose perform the same or not worse than the animals fed the control diet. Study C and Study D were based on demonstrating superiority, where the animals fed diets supplemented with ferric citrate or another source of Fe, perform better than the untreated control. The null hypothesis being that the animals fed diets containing ferric citrate (or another source of Fe) perform the same as the control animals. The alternative hypothesis being that the animals fed ferric citrate (or another source of Fe) perform better than the animals fed the control diet.
For the different studies, sample size was calculated according to the hypothesis (Sealed Envelope Ltd, 2012; primary outcome measure = BW), and were designed to be optimal or near optimal in the trade-off between power and resources.
For all studies, the basic study design was a random complete block design, with pen as the experimental unit for statistical purposes for zootechnical and faecal parameters. The individual animal was used as the experimental unit for blood parameters. All parameters were analysed using a general lineal model as follows: Y ij = m + treatment i + block j + ε ij , where Y ij was the dependent variable, m was the overall mean, treatment i was the effect of treatment (Study A = 3 levels; Study B = 3 levels; Study C = 4 levels and Study D = 5 levels), block j was the effect of the block, and ε ij was the residual error. All experimental units were used for statistical analysis.
All parameters were reported as group least squares mean. Standard error of the mean were also reported. Multiple comparisons between treatment groups were made by Tukey test and significant differences were declared at P ≤ 0.05, while near significant trends were considered for 0.05 < P ≤ 0.10.
The analyses were performed using the software packages SPSS (IBM SPSS, Version 21, Study A, Study B, Study D) and Unistat (Unistat Ltd., version 10.05, Study D).

RESULTS
General health in all studies was good. Overall mortality rates were as follows: study A = 2.5%; study B = none; study C = 0.9%; study D = 1.14%. Details of the veterinary treatments administered are given below.

Study B
Results summarising the safety of ferric citrate in weaned piglets are presented in Table 10. For the overall study period (0-42 days) there were no significant differences between treatment groups for BW, ADG or ADFI. However, FCR was significantly improved in  piglets fed T2 (ferric citrate, 500 mg/kg) and T3 (ferric citrate, 5,000 mg/kg) diets compared to those fed the T1 Control diet (T2 = −4.02%; T3 = −5.31%). There were no significant differences between treatment groups for any of the blood parameters measured.

Study C
Results summarising the effects of ferric citrate and other ferric Fe sources fed to weaned piglets are presented in Table 11. For the overall study period, significant improvements were observed in groups fed the ferric Fe sources compared to those fed the control diets. On day 42, animals in T3 (ferric citrate) and T4 (ferric tartrate) were 1.41 kg and 1.45 kg heavier, respectively compared to the T1 Control animals. Furthermore, animals fed T3 and T4 diets grew significantly better than T1 Controls (ADG: T3 = +7.09%; T4 = +7.30%). FCR was also significantly improved in animals fed T3 and T4 compared to  Tables 13 and 14. On day 0 there were no significant differences in blood parameters between treatment groups. On day 42, no significant differences were observed for any of the blood parameters except for total bilirubin and SOD. Total bilirubin was significantly higher in T2 compared to the other treatment groups (T1 = +74.73%; T3 = +63.54%; T4 = +78.49%; T5 = +95.29%), while SOD was significantly higher in the animals fed ferric citrate (T4 and T5) compared to the piglets fed the T1

DISCUSSION
The objective of the four studies presented here was to assess the safety and efficacy of ferric citrate as a safe and available source of Fe for broilers and weaned piglets bred and managed under standard farming conditions.

Safety in broilers -Study A
Haemoglobin is an Fe-rich protein involved in cellular metabolism and the transport of oxygen through animal tissues (Snyder & Sheafor, 1999). Thus, haemoglobin and haematocrit-related parameters are key to detecting any possible negative effects of dietary Fe supplementation in broilers. No significant differences between diets were detected for haemoglobin, haematocrit, MCH or MCHC. In the case of erythrocyte and MCV values, despite the statistically significant differences with the Control diet, all results were in line with values reported in previous studies (Milanovic et al., 2008;Saripinar-Aksu, Aksu & Özsoy, 2010;Tako, Rutzke & Glahn, 2010). Likewise, the statistically significant differences detected between treatments for sodium, potassium, chloride, urea and glucose were considered to have no clinical relevance since there were no indications of a dose-dependent effect and results were in agreement with other reported data for chickens (Simaraks, Chinrasri & Aengwanich, 2004;Café et al., 2012;Odunitan-Wayas, Kolanisi & Chimonyo, 2018). Furthermore, the inflammatory marker serum-amyloid A (O'Reilly, Bailey & Eckersall, 2018) was below the detection level, indicating good health status of all birds. Finally, and with the few exceptions described above, blood cells, enzymes, electrolytes and biochemical blood constituents at the end of the 35 days feeding period did not change significantly across the different doses of ferric citrate. Altogether, the analysis of blood data shows that ferric citrate is a safe source of supplementary Fe for broilers, since no negative effects were detected in either of the ferric citrate treatments, where supplementary Fe from ferric citrate supplied an additional 34 mg (T2) and 340 mg (T3) Fe/kg feed, approximately, in comparison with T1 Control (only endogenous dietary Fe). Additionally, it was confirmed that ferric citrate is well tolerated up to 2,000 mg/kg (×10 the recommended inclusion rate). These results are in line with a recent assessment performed by the European Food Safety Authority (EFSA, 2019), who concluded that ferric citrate is a safe feed additive for weaned piglets up to 5,000 mg/kg of feed, (supplying ca. 850 mg Fe/kg feed). As indicated by the National Research Council (NRC) (1994), some of the response criteria for Fe supplementation in chickens are growth and feed efficiency. The results presented in Study A support this as, after 35 days, animals fed diets supplemented with ferric citrate at either 200 mg/kg or 2,000 mg/kg feed (34 and 340 mg Fe/kg feed, respectively) showed better performance in terms of ADG, MFCR and final weight compared to birds with no Fe supplementation (T1 Control = x 79 mg Fe/kg Fe). The low mortality rate reported (2.5% at 35 days), coupled with the blood analysis and the growth performance, support the safety of ferric citrate in broilers.
The results of Study A are in agreement with an earlier study that evaluated ferric citrate as a source of Fe, using broiler chickens in a specialised model intended as an intermediate test of in vivo Fe bioavailability of human Fe supplements (Tako, Rutzke & Glahn, 2010). Intestinal Fe absorption and haemoglobin maintenance was assessed in broilers using ferric citrate as the Fe source. From one week of age the control group was fed a diet deficient in Fe (51 mg Fe/kg feed), and the other group was fed a diet supplemented with 500 mg ferric citrate/kg (total Fe: 141 mg/kg feed). During the 6-week study period, there was no difference in feed intake between the treatments, but Fe intake was significantly increased in the ferric citrate supplemented group compared to the Fe deficient control group. Blood haemoglobin (Hb) concentrations were also significantly higher in the ferric citrate group compared with the control. The Hb maintenance efficiency values (indicator for dietary Fe availability) were significantly higher in birds fed the Fe-deficient control diet, indicating that the Fe uptake mechanisms in broiler are responding to dietary Fe concentrations. The final total body Hb and the increase in total body haemoglobin Fe during the study period was significantly greater in the ferric citrate supplemented group than in the control group (85 vs. 54 mg, P < 0.05). No mortalities or adverse effects were reported.
Whilst not evaluated in the present study, it has been demonstrated that organic sources of Fe are more bioavailable in broilers than inorganic forms (Pla & Fritz, 1970;Shinde et al., 2011;Arnaudova-Matey et al., 2013). For example, Pla & Fritz (1970) investigated the bioavailability of organic and inorganic sources of Fe in broiler chicks with severe anaemia. Over a two-week period, response to dietary Fe supplementation was measured using blood Hb. The authors categorized the sources of Fe into good, mediocre and poor. They reported that the organic sources of iron, including ferric citrate, were good sources of iron compared to ferrous sulphate. Shinde et al. (2011) looked at the efficiency of organic and inorganic sources of Fe in Fe-depleted broilers and compared ferric sulphate and ferrous methionine with a negative control diet. During 35 days, improved body weight gain, FCR, and greater dry matter, crude protein retention and Fe digestibility in faeces were observed in birds fed on Fe supplemented diets when compared with birds fed the control diet. After 35 days, the red blood cell, Hb, haematocrit, and Fe concentration in plasma, tibia and liver were higher in birds fed on Fe-supplemented diets than birds fed on the control diet. Most importantly, supplementation of Fe in the organic form resulted in greater Fe concentration in the tibia and liver, and less Fe excretion, when compared with birds receiving inorganic Fe.

Safety in weaned piglets
The results obtained in Study B support the safety of ferric citrate. In this study, no adverse effects on growth or blood parameters were noted. Furthermore, no mortality was recorded, and animals were healthy for the duration of the trial. These results support the safety of ferric citrate when added to the diet of weaned piglets at up to 10 times the recommended dose, and are in agreement with the recent assessment performed by the European Food Safety Authority (EFSA, 2019), who concluded that ferric citrate is a safe feed additive for weaned piglets up to 5,000 mg/kg of feed. Results obtained in Studies C and D further support the safety of ferric citrate as a source of dietary Fe. In these studies, no adverse effects were observed in the animals fed ferric citrate at the recommended dose (ferric citrate at 500 mg/kg feed supplying 85 mg Fe/kg feed); animals fed ferric citrate performed better than animals fed the control diet (containing endogenous Fe only), and the blood data were within the normal ranges for healthy piglets.

Nutritional efficacy in weaned piglets
The results from Studies C and D showed that ferric citrate is an effective source of iron in weaned piglets. Study C demonstrated that ferric citrate and other sources of ferric Fe improved growth rates in pigs fed these compounds compared to pigs fed the Fe deficient negative control diet containing only endogenous Fe. The results from Study D showed that pigs fed ferric citrate alone or in combination with ferrous sulphate performed better than the control animals and the animals fed inorganic forms of iron. Moreover, in Study D, Fe digestibility was significantly improved in piglets fed ferric citrate alone or in combination with ferrous sulphate compared to the control animals and the animals fed the inorganic forms of Fe, thus supporting the nutritional efficacy of ferric citrate. In the same study, no significant improvements in blood Fe parameters were observed. At the beginning of the study (25 days of age), all piglets had Hb of 100 g/L to 102 g/L, which indicated normal iron concentrations and that the animals were not iron deficient or anaemic (Wei et al., 2005). At day 42 (66 days of age), piglets fed ferric citrate had numerically higher Hb compared to the negative controls, but all groups were within the expected range. No significant differences were noted for other blood parameters indicative of Fe status (MCV, MCH, MCHC), highlighting that all piglets included in this study were not seriously iron deficient or anaemic. Furthermore, no significant differences in Fe serum concentrations were seen between treatment groups on day 0 or day 42. These results are not surprising given that all piglets received a parenteral iron dextran injection at 3 days old. Administration of iron dextran is standard practice and performed for welfare reasons to reduce severe Fe deficiency and high mortality in pre-and post-weaning piglets (Perri et al., 2016;Svoboda, Vanhara & Berlinska, 2017). As these studies were run for regulatory approval, using Fe-depleted animals was not appropriate as healthy animals must be used to support the safety and efficacy of feed additives (EFSA, 2017;EFSA, 2018). However, earlier studies reported that ferric citrate supplemented at daily use rates of 48 mg and 144 mg/feed increased blood haemoglobin concentration and plasma Fe in pigs (Furugouri & Kawabata, 1975) and interstingly, Furugouri & Kawabata (1975) also cited a study performed by Ullrey et al. (1973), who reported that ferric citrate is a more effective source of supplemental Fe for piglets than ferrous sulfate.
Even though a Fe-depletion model which is often described as the standard for assessing bioavailability of iron (Henry & Miller, 1995) was not used in the present studies, improved growth performance in animals fed ferric citrate compared to the animals fed diets containing endogenous Fe only was observed. This, coupled with the faecal digestibility of Fe, indicates that ferric citrate is a safe and effective source of dietary Fe. Furthermore, piglets receiving ferric citrate had significantly higher levels of SOD compared to the negative control group and the groups fed the inorganic forms of Fe, indicating that oral administration of ferric citrate improved the oxidative status of the animals. These results were in agreement with a recent study that compared oral administration of ferrous glycine and parenteral iron dextrose with a negative control in anaemic pigs (Dong et al., 2020). These authors reported significantly decreased serum SOD levels in the control animals and those that received iron dextran compared to the animals fed the organic form of Fe.
Some animals in studies C and D were treated with antibiotics, which could raise some concern given that there are reports showing an interference of these compounds with iron absorption (Djaldetti et al., 1981). However, the data presented in study D shows that all animals had recommended Fe levels and no significant differences were found in Fe serum concentrations between treatment groups. Additionally, the proportion of animals treated in each study was low (<9% out of 220 and 175 piglets, respectively) and antibiotics were applied roughly equally across all treatment groups.
Iron requirements for piglets are influenced by several factors, the most important being growth intensity during pre-weaning and post-weaning. Modern pig breeds are selected for fast growth and several recent studies have questioned whether current Fe supplementation practices are meeting Fe requirements (Perri et al., 2016;Svoboda, Vanhara & Berlinska, 2017). Hence, dietary supplementation with Fe is required to ensure that adequate stores of Fe are achieved. Inadequate iron leads to health and economic implications as piglets with depleted iron stores are at risk of a suppressed immune system, leading to an impaired ability to resist infection and parasitic diseases, and have slower growth rate and increased morbidity and mortality (Svoboda & Drabek, 2005).

CONCLUSIONS
In conclusion, the results from these four experiments show that ferric citrate is a safe source of dietary Fe for broilers up to 2,000 mg/kg feed and up to 5,000 mg/kg for piglets. Additionally, ferric citrate is an easily digestible source of Fe in piglets and can be used alone or with other Fe sources.