Abstract
The iron status of sows has a great influence on reproductive performance. Iron deficiency reduces reproductive performance and newborn piglet survival rate of sow. The hemoglobin is a potential predictor for the iron status of sows and is convenient for rapid detection in pig farms. However, the relationship between iron status, hemoglobin, placental trace elements, and reproductive performance remains unclear. In this study, the hemoglobin and reproductive performance of more than 500 sows with first to sixth parities at different gestation stages (25, 55, 75, 95, and 110 d of gestation) in two large-scale sow farms were collected, and the content of placental Fe, Zn, Mn, and Cu was analyzed. The results show that hemoglobin levels of sows during pregnancy (days 75, 95, and 110) decreased significantly (P < 0.001). As the parity increases, the hemoglobin levels of sows at days 25 and 55 of gestation and placental mineral element contents including Fe, Zn, Mn, and Cu at delivery decreased (P < 0.05), while the litter size, birth alive, and litter weights increased gradually (P < 0.001). Furthermore, hemoglobin during pregnancy had a negative linear correlation with litter weight and average weight (P < 0.05), and higher hemoglobin at day 25 of gestation may reduce the number of stillbirths (P = 0.05), but higher hemoglobin at day 110 of gestation may tend to be a benefit for the birth (P = 0.01). And there was a significant positive linear correlation between hemoglobin at day 110 of gestation and placental Fe and Mn levels (P = 0.002, P = 0.013). There was also a significant positive linear correlation among Fe, Zn, Mn, and Cu in the placenta (P < 0.001). The levels of Fe, Zn, and Mn in the placental at delivery were positively related to the average weight of the fetus (P = 0.048, P = 0.027, P = 0.047), and placental Cu was linearly correlated with litter size (P = 0.029). Our research revealed that the requirements for iron during gestation were varied in different gestation periods and parities. The feeds should be adjusted according to the gestation periods, parities, or iron status to meet the iron requirements of sows and fetal pigs.
Keywords: gestation, hemoglobin, iron status, pregnant sows, reproductive performance
Lay Summary
Iron deficiency and iron excess may cause adverse outcomes during pregnancy. In sows’ feed, iron is added as ferrous sulfate, ferrous glycine, or other forms to improve their reproductive performance and prevent iron-deficiency anemia in their offspring. However, it is always ineffective and iron-deficiency anemia often occurs in piglets. To explore the iron requirements in pregnant sows, we conducted a large-scale farm study to track the hemoglobin levels, placental trace element content, and reproductive performances of hundreds of sows. The correlation between the hemoglobin levels, placental trace element content, and reproductive performance indicators of sows during pregnancy at different parities was analyzed. We found that pregnancy hemoglobin level of sows decreases during the gestation and varies at different parities. The hemoglobin level of sows during pregnancy was linearly negatively correlated with reproductive performance. The content of iron, zinc, manganese, and copper in the placenta was linearly positively correlated. Our results revealed that iron deficiency or excess in sows’ feed may not be conducive to the improvement of reproductive performance, and the optimal iron supplementation dose during pregnancy may depend on the iron status and number of fetuses of sow.
Optimal iron levels in sows’ diet during pregnancy are critically important for its outcomes.
Introduction
Trace elements, such as iron, zinc, copper, and manganese, are essential during pregnancy to perform important functions (Lewicka et al., 2017). The reproductive performance indexes of sows mainly include litter size, litter weight, and adverse pregnancy outcomes such as mummies, stillbirth, and weak-born piglets (Salak-Johnson, 2017). Unsatisfying reproductive performance and low piglet survival rates caused due to iron deficiency are difficult challenges in the sows breeding industry (Wang et al., 2014). Fetal growth and development during pregnancy lead to a rapid increase in the demand for iron, thereby resulting in iron deficiency in sow. The inability of sows to obtain iron from their diets for the demand of gestation seriously affects reproductive performance and growth of piglets, and iron-deficiency anemia can lead to adverse pregnancy outcomes such as abortion and stillbirth (Means, 2020). In addition, trace elements such as iron, copper, zinc, and manganese have common transporters. They interact closely with each other during absorption and metabolism (Gulec and Collins, 2014). For example, deficiency of iron could inhibit other trace elements absorption and lead to copper (Solomons, 1986) and zinc deficiency (Kelkitli et al., 2016). Therefore, iron supplementation and iron injection for sows and neonatal piglets are used to prevent iron deficiency in sows and piglets (Peters and Mahan, 2008; Buffler et al., 2017). Improving iron bioavailability of sows during pregnancy could significantly increase their litter weight, placental iron concentration, and piglet iron status (Wan et al., 2018). However, few studies also revealed that iron supplements did not improve the iron status and reproductive performance of sow without iron deficiency (Bhattarai et al., 2019c). Excessive iron supplementation could not increase the absorption and utilization of iron but impacts the health and reproductive performance (Georgieff et al., 2019). Sows may generally face the risk of iron deficiency and iron excess during the first, middle, and late trimesters of pregnancy (Bhattarai et al., 2019b). Evaluating the accurate iron requirements during different gestation periods may help to improve their productive performance. But the studies on recommended iron intake for sows during different gestation periods are still limited.
Iron is found in animals mainly in the form of hemoglobin and myoglobin, which has a function to carry oxygen (Gell, 2018). Hemoglobin and liver iron concentration are recognized as common indicators of iron status in the body, and the most intuitive change in anemia is the decrease in hemoglobin (Padjen et al., 2017). In addition, the detection method of hemoglobin is convenient and less invasive enough to facilitate multiple data collection of sows at different gestation periods. Studies have reported the correlation between hemoglobin during pregnancy and reproductive performance of sows. A negative correlation between stillbirth rate and hemoglobin in late-gestation sows was found. Hemoglobin levels in newborn piglets were found to increase with sow hemoglobin (Bhattarai et al., 2019a). A large-scale study suggested that higher hemoglobin levels of sows may be related to piglet survival, litter weight, and average birth weight (Hollema et al., 2020). A study in humans had shown a U-shaped curve relationship between the risk of adverse birth outcomes and hemoglobin concentrations during pregnancy—both high and low hemoglobin levels increase the risk of abortion and stillbirth—and this impact is most pronounced during early pregnancy (Dewey and Oaks, 2017). Therefore, the pregnancy hemoglobin level of sows is a potential connection for iron requirements and reproductive performance. The correlation between pregnancy hemoglobin and the reproductive performance of sows should be clarified during sow production.
The placenta is for the transport of nutrients from mothers to the fetus (Cao and Fleming, 2016). In humans and mice, iron from the blood is delivered to the fetus through transferrin (Tf), transferrin receptor 1 (TfR1), divalent metal-ion transporter-1(DMT1), or Zrt and Irt-like protein 8 (ZIP8) in trophoblasts of the placental villi (Sangkhae and Nemeth, 2019). However, the accumulated amounts of trace elements in the placenta may result in iron overload, which could affect the iron transportation from the mother blood to the fetus and lead to iron deficiency anemia in newborn piglets (Dong et al., 2020).
In this research, the hemoglobin and reproductive performance of more than 500 sows with first to sixth parities at different gestation stages (25, 55, 75, 95, and 110 d of gestation) in two large-scale sow farms were collected; placenta was sampled after delivery; and the contents of Fe, Zn, Mn, and Cu were analyzed. We analyzed the hemoglobin at different gestational periods, reproductive performance, and placental trace element content of sows between parities and conducted a linear regression analysis to explore the correlation between the three factors. Our results would help to establish the optimal iron levels in sows’ diet during pregnancy and improve its outcomes.
Materials and Methods
The care and handling of the sows used in this study followed the standards of the Animal Care and Use Committee of the Subtropical Agriculture, Chinese Academy of Science (No. ISA-2017-012). This study was conducted at Xinwufeng and Jinzhunongmu farms in the Hunan province. The experimental period was from August 2018 to August 2019.
Animals, diets, and experimental design
Long-term gestational hemoglobin and reproductive performance data were collected from more than 500 sows. During the entire experiment, the estrus and mating time, litter size, healthy litter size, litter weight, average weight, weak litter size, mummy fetus, and stillbirth of sows with different parities were recorded; the hemoglobin levels of sows at days 25, 55, 75, 95, and 110 of gestation were measured; and placenta samples collected after delivery were used to detect the content of Fe, Zn, Mn, and Cu in tissues. Detailed information about each pig was recorded during the experiment, including mating date, estrus interval, dystocia, abortion, estrus, and health records. The experimental diets used for pig farm production were from the company named Xinqidian in Hunan province according to the recommendations NRC (2012). The composition and nutrient levels of basal diet for sows in the experimental period are as follows (Table 1).
Table 1.
Composition and nutrient levels of gestation and lactation diets (air-dry basis, 90%)
| Gestation (<95 d) diet | Gestation (>95 d) diet | Lactation diet | |||
|---|---|---|---|---|---|
| Composition | Content, % | Composition | Content, % | Composition | Content, % |
| Corn | 56.60 | Corn | 57.07 | Corn | 59.07 |
| Soybean meal | 19.27 | Soybean meal | 20.00 | Soybean meal | 19.50 |
| Bran | 4.80 | Bran | 2.50 | Bran | 2.50 |
| Wheat bran | 12.00 | Wheat bran | 12.00 | Wheat bran | 10.00 |
| Stone powder | 1.20 | Stone powder | 1.20 | Stone powder | 1.20 |
| Soybean oil | 2.70 | Fish meal | 1.00 | Fish meal | 1.50 |
| L-Lysine HCl (70%) | 0.40 | Soybean oil | 2.70 | Soybean oil | 2.70 |
| CaHPO4 | 1.00 | L-Lysine HCl (70%) | 0.50 | L-Lysine HCl (70%) | 0.50 |
| NaCl | 0.50 | CaHPO4 | 1.00 | CaHPO4 | 1.00 |
| L-Threonine (98.5%) | 0.07 | NaCl | 0.50 | NaCl | 0.50 |
| DL-Methionine (98.5%) | 0.08 | L-Threonine (98.5%) | 0.07 | L-Threonine (98.5%) | 0.07 |
| Phytase | 0.01 | DL-Methionine (98.5%) | 0.08 | DL-Methionine (98.5%) | 0.08 |
| Choline chloride(60%) | 0.17 | Phytase | 0.01 | Phytase | 0.01 |
| Chinese herbal extract | 0.20 | Choline chloride (60%) | 0.17 | Choline chloride (60%) | 0.17 |
| Premix1 | 1.00 | Chinese herbal extract | 0.20 | Chinese herbal extract | 0.20 |
| Premix1 | 1.00 | Premix1 | 1.00 | ||
| Total | 100.00 | Total | 100.00 | Total | 100.00 |
| Nutrient levels2 | Content, % | Nutrient levels | Content, % | Nutrient levels | Content, % |
| Dry matter (DM) | 86.75 | DM | 86.82 | Dry matter (DM) | 86.82 |
| DE, kMcal/kg | 3.29 | DE, kMcal/kg | 3.30 | DE, kMcal/kg | 3.32 |
| CP | 16.65 | CP | 17.42 | CP | 17.37 |
| EE | 6.23 | EE | 5.97 | EE | 5.99 |
| CF | 2.60 | CF | 2.50 | CF | 2.38 |
| Ash | 4.40 | Ash | 4.43 | Ash | 4.42 |
| Ca | 0.82 | Ca | 0.87 | Ca | 0.89 |
| Total P | 0.68 | Total P | 0.69 | Total P | 0.69 |
The vitamin and mineral premix contains the following per kilogram of diets: VA 15,000 IU, VD3 2,500 IU, VE 20 IU, VB1 2 mg, VB2 5 mg, VB6 2 mg, VB12 18 mg, VC 200 mg, VK3 3 mg, folic acid 0.6 mg, d-pantothenic acid 10 mg, nicotinic acid 22 mg, d-biotin 0.5 mg, Cu (CuSO4·5H2O) 150 mg, Fe (FeSO4·H2O) 100 mg, Zn (ZnSO4·H2O) 200 mg, Mn (MnSO4·H2O) 50 mg, I (CaI2O6) 1 mg, and Se (Na2SeO3) 0.25 mg.
CP, crude protein; EE, ether extract; CF, crude fiber.
Sample and reproductive performance data collection
After delivery, reproductive performance data of sows were recorded, including litter size, mummy fetus, stillborn, healthy litter size, weak litter size, and litter weight at birth, and the average weight is litter weight divided by litter size. Placenta samples were collected and stored at −20 °C until analysis.
Hemoglobin detection
The sow hemoglobin detection was performed on days 25, 55, 75, 95, and 110 of gestation. Two hours after feeding, disposable steel needles (Aikang, Hangzhou, China) were used to puncture the ear epiderms of sows, and 10 to 50 μL of blood samples were collected from the ear vein. The content of hemoglobin was detected on-site with a hemoglobin test strip (Aikang, Hangzhou, China) and a portable hemoglobin meter (CYHB, Hangzhou, China), and records were made in time. Each test was repeated more than twice, and the mean of the duplicates was used for data. Sows collected on different dates were measured at the same time of day.
Tissue trace element analysis
The contents of Fe, Zn, Mn, and Cu in placenta samples were determined using an inductively coupled plasma emission spectrometer (ICP). Nitric acid (5 mL) was added to a suitable dry tissue sample and digested in a microwave digestion system (Milestone, Italy), then cool to below 50 °C and volume with water to 50 mL. The solution was submitted to the ICP-MS analyses (iCAPQ, Thermo). All of the laboratory analyses procedures followed the manufacturer’s instructions and included three duplicates in the same sample. The mean of the duplicates was used for analysis.
Statistical analysis
All the original records of the experiment were recorded in Microsoft Office Excel; statistical analysis was performed using the SPSS Statistics 25.0 software; and all figures were finished in GraphPad Prism 9.0. Comparison between the two groups was done using independent-samples t-test, and comparison between more than three groups was analyzed by one-way analysis of variane (ANOVA). Results are given as mean values. P-value < 0.05 was used to indicate statistical significance. Regression analyses were done using (both linear regression and quadratic regression) SPSS. The results of linear regression analysis are given as β, R, and P-value. β means the linear regression coefficient, R is the coefficient of determination and reflects the fitting degree of the regression equation, P-value means the significance of linear regression, and P < 0.05 was used to indicate statistical significance.
Results
Effect of gestational age on hemoglobin content of sows
Hemoglobin levels of sows from first to sixth parities at days 25, 55, 75, 95, and 110 of gestation and reproductive performance data were collected in each sow at different periods. Totally, hemoglobin levels of the sows in the first (n = 304), second (n = 204), third(n = 23), fourth (n = 29), fifth (n = 19), and sixth (n = 18) parities, respectively, were collected for statistical analysis (Table 2). In first to sixth parities, hemoglobin level of sows increased from day 25 to 55, then decreased from day 55 to 110, and peaked at 10.045 g/L during the gestational age. Specifically, hemoglobin level had a significant reduction in the second trimester and reached the lowest point of 9.458 g/L at day 110 (P < 0.001). Similarly, the hemoglobin level of the sows increased first and then decreased obviously with gestational age in the first and second parities (P < 0.001, P = 0.022). The hemoglobin levels tend to increase in the early stage of the pregnancy and rapidly decreased in the middle and late pregnancy in the first, second, and third parities (P < 0.001). No significant differences were observed due to limited sample size in larger parities (P = 0.078).
Table 2.
Hemoglobin content of sows at different gestation days
| Parity | Gestational age | SEM | P-value∗ | ||||
|---|---|---|---|---|---|---|---|
| Day 251 | Day 55 | Day 75 | Day 95 | Day 110 | |||
| 1 to 6 | 9.619bc | 10.045a | 10.001a | 9.690b | 9.458c | 0.030 | <0.001 |
| 1 | 9.992ab | 10.202a | 10.082ab | 9.729bc | 9.460c | 0.035 | <0.001 |
| 2 | 9.259bc | 9.778a | 9.694ab | 9.403ab | 9.122b | 0.072 | 0.022 |
| 3 to 6 | 9.090b | 9.630a | 9.571ab | 9.721a | 9.642a | 0.083 | 0.078 |
Day 25, gestation 25 d; day 55, gestation 55 d; day 75, gestation 75 d; day 95, gestation 95 d; and day 110, gestation 110 d.
P-value reflects differences between sow hemoglobin at different gestational ages.
Values mean thesignificant differences (P ≤ 0.05) between different groups.
Effect of parities on hemoglobin content in pregnant sows
To explore the variation of hemoglobin level in different parities during pregnancy, we compared the hemoglobin level of sows with different parities at days 25, 55, 75, 95, and 110 of gestation (Table 3). The results show that hemoglobin at days 25 and 55 decreased significantly with the increase of parities (P < 0.001, P = 0.003). However, there was no significant difference in hemoglobin from first to sixth parity at days 75, 95, and 10 (P = 0.196, P = 0.286, P = 0.276), and hemoglobin after day 75 of gestation was generally lower than that at early stage of the pregnancy.
Table 3.
Hemoglobin content of sows at different parities
| Gestational age | Parity | SEM | P-value∗ | |||||
|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |||
| Day 251 | 9.992a | 9.259b | 9.388ab | 8.906b | 9.181b | 9.370ab | 0.062 | <0.001 |
| Day 55 | 10.202a | 9.778b | 9.733ab | 9.483b | ——2 | —— | 0.060 | 0.003 |
| Day 75 | 10.082 | 9.694 | 9.700 | 9.380 | 9.867 | 9.500 | 0.066 | 0.196 |
| Day 95 | 9.729 | 9.403 | 9.896 | —— | —— | —— | 0.072 | 0.286 |
| Day 110 | 9.460 | 9.122 | 9.808 | —— | —— | —— | 0.080 | 0.276 |
Day 25, gestation 25 d; day 55, gestation 55 d; day 75, gestation 75 d; day 95, gestation 95 d; day 110, gestation 110 d.
“——” represents uncollected data.
P-value reflects differences between sow hemoglobin at different parities.
Values means significant differences (P ≤ 0.05) between different groups.
Reproductive performance of sows with different parities
Reproductive performance including litter size, healthy litter size, litter weight, weak litter size, mummy fetus, and stillbirth sows in first (n = 99), second (n = 32), third (n = 11), fourth (n = 19), fifth (n = 11), and sixth (n = 5) parities was recorded for statistical analysis (Table 4). Litter size, healthy litter size, and litter weight increased gradually with parity, suggesting that multiparity sows have better reproductive performance (P < 0.001, P < 0.001, P < 0.001). However, adverse reproductive performance, such as weak litter size, mummy fetus, and stillbirth, had a downtrend in the second and third parities but an uptrend was noticed above the fourth parity without significant difference (P = 0.253, P = 0.055, P = 0.153).
Table 4.
Reproductive performance of sows at different parities
| Reproductive performance | Parity | SEM | P-value∗ | ||||
|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | |||
| Litter size | 9.330d | 10.125cd | 14.000a | 11.053bc | 12.455ab | 0.244 | <0.001 |
| Healthy litter size | 8.229c | 9.219bc | 11.364a | 9.842ab | 10.727ab | 0.234 | <0.001 |
| Litter weight, kg | 11.237b | 13.755a | ——1 | 14.753a | 16.007a | 0.404 | <0.001 |
| Weak litter size | 0.214 | 0.281 | 0.636 | 0.263 | 0.364 | 0.045 | 0.253 |
| Mummy fetus | 0.133a | 0.094ab | 0.000b | 0.105ab | 0.455ab | 0.030 | 0.055 |
| Stillbirth | 0.857ab | 0.531b | 1.364a | 0.947ab | 1.091ab | 0.078 | 0.153 |
“—” represents uncollected data.
P-value reflects differences between reproductive performance at different parities.
Values means significant differences (P ≤ 0.05) between different groups.
Effects of parities on placental trace elements of sows
The content of Fe, Zn, Mn, and Cu that are regarded essential during pregnancy was detected. We detected first (n = 104), second (n = 52), third (n = 20), fourth (n = 12), fifth (n = 3), and sixth (n = 5) placental samples from different parities of sows (Table 5). Obviously, the contents of Zn and Mn in the placenta decreased significantly with the increase of parities (P = 0.002, P < 0.001). And the iron and copper content in the placenta of sows at the first three parities also showed an obvious decreasing trend (P = 0.070, P = 0.088).
Table 5.
Content of trace elements in the placenta of sows at different parities
| Content of trace elements in the placenta, mg/kg | Parity | SEM | P-value∗ | |||||
|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |||
| Iron | 540.042a | 495.231ab | 424.100ab | 277.786b | 198.667ab | 278.400ab | 26.960 | 0.070 |
| Zinc | 98.563b | 74.956ac | 61.400ac | 149.829ab | 64.900abc | 68.080abc | 5.027 | 0.002 |
| Manganese | 27.394a | 14.787b | 12.822b | 7.471bd | 2.283c | 7.750bc | 1.722 | <0.001 |
| Copper | 11.402a | 11.004a | 9.390ab | 7.209b | 9.667ab | 9.915ab | 0.339 | 0.088 |
P-value reflects differences between placental trace elements at different parities.
Values means significant differences (P ≤ 0.05) between different groups.
Correlation of iron, zinc, copper, and manganese contents in the placenta of sows
Considering the close interaction between Fe, Zn, Mn, and Cu, we performed linear and quadratic regression analyses of Fe, Zn, Mn, and Cu contents in the placenta of sow (Table 6). Interestingly, the results show that placental Fe, Zn, Mn, and Cu have a significant positive correlation with each other in linear regression rather than quadratic regression (P < 0.001, P < 0.001, P < 0.001, P < 0.001). And this result of linear regression analysis and curve fitting shown in Figure 1 suggests that Fe, Zn, Mn, and Cu may promote placental transport of the other mineral elements in the four trace elements.
Table 6.
Linear regression of iron, zinc, and manganese content in sow placenta
| Content of trace elements in the placenta, mg/kg | Iron | Zinc | Manganese | Copper | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β1 | R2 | P-value∗ | β | R | P-value | β | R | P-value | β | R | P-value | |
| Iron | —— | 0.067 | 0.475 | <0.001 | 0.046 | 0.687 | <0.001 | 0.006 | 0.407 | <0.001 | ||
| Zinc | 3.373 | 0.475 | <0.001 | —— | 0.295 | 0.619 | <0.001 | 0.044 | 0.433 | <0.001 | ||
| Manganese | 10.238 | 0.687 | <0.001 | 1.298 | 0.619 | <0.001 | —— | 0.087 | 0.453 | <0.001 | ||
| Copper | 27.221 | 0.407 | <0.001 | 4.265 | 0.433 | <0.001 | 2.358 | 0.453 | <0.001 | —— | ||
β is the linear regression coefficient.
R is the coefficient of determination and reflects the fitting degree of the regression equation.
P-value means the significance of linear regression.
Figure 1.
Linear regression analysis and curve fitting of placental trace elements. Linear correlation between iron, zinc, manganese, and copper in placental tissue of sows. R is the coefficient of determination and reflects the fitting degree of the regression equation. P-value means the significance of linear regression. Linear regression analysis between placental iron and zinc (A); Linear regression analysis between placental manganese and iron (B); Linear regression analysis between placental copper and iron (C); Linear regression analysis between placental manganese and zinc (D); Linear regression analysis between placental copper and zinc (E); Linear regression analysis between placental copper and manganese (F).
Linear negative correlation between hemoglobin content and reproductive performance of pregnant sows
The linear and quadratic regression analyses of hemoglobin at different gestation periods with reproductive performance of 177 sows are presented in Table 7 and also shown in Figure 2A–G. The results show no significant quadratic regression between hemoglobin and reproductive performance including litter size, healthy litter size, litter weight, and average weight. However, we observed a generally significant linear negative correlation between litter weight and hemoglobin at day 25 (P = 0.030), day 55 (P = 0.012), day 75 (P = 0.029), day 95 (P = 0.030), and day 110 (P = 0.011). As the same trend in average weight, the hemoglobin content at days 25 and 110 has a significant linear negative correlation with average weight (P < 0.001, P = 0.006). Surprisingly, contrary to what has been reported in previous studies, hemoglobin content in pregnant sows is negatively correlated with several typical reproductive performance, such as litter size and healthy litter size with no significance (P > 0.05).
Table 7.
Linear regression between hemoglobin at different gestation ages and reproductive performance of sow
| Hemoglobin, g/L | Litter size | Healthy litter size | Litter weight, kg | Average weight, kg | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β2 | R3 | P-value∗ | β | R | P-value | β | R | P-value | β | R | P-value | |
| Day 251 | −0.471 | 0.185 | 0.086 | −0.371 | 0.141 | 0.194 | −1.058 | 0.269 | 0.030 | −0.267 | 0.555 | <0.001 |
| Day 55 | −0.453 | 0.171 | 0.063 | −0.307 | 0.122 | 0.191 | −0.964 | 0.257 | 0.012 | −0.028 | 0.120 | 0.248 |
| Day 75 | −0.299 | 0.120 | 0.199 | −0.184 | 0.077 | 0.417 | −0.858 | 0.234 | 0.029 | −0.007 | 0.029 | 0.787 |
| Day 95 | −0.266 | 0.100 | 0.254 | −0.255 | 0.102 | 0.244 | -0.841 | 0.231 | 0.030 | -0.036 | 0.178 | 0.100 |
| d 110 | 0.086 | 0.031 | 0.745 | -0.270 | 0.101 | 0.287 | -1.063 | 0.287 | 0.011 | -0.079 | 0.313 | 0.006 |
Day 25, gestation 25 d; day 55, gestation 55 d; day 75, gestation 75 d; day 95, gestation 95 d; day 110, gestation 110 d.
β is the linear regression coefficient.
R is the coefficient of determination and reflects the fitting degree of the regression equation.
P-value means the significance of linear regression.
Figure 2.
Linear regression analysis and curve fitting between hemoglobin and reproductive performance, and between hemoglobin and content of placental trace elements. R is the coefficient of determination and reflects the fitting degree of the regression equation. P-value means the significance of linear regression. Abbreviations: Hb, hemoglobin; 25 d, gestation 25 d; 55 d, gestation 55 d; 75 d, gestation 75 d; 95 d, gestation 95 d; 110 d, gestation 110 d. Linear regression analysis between litter weight and 25 day-Hb (A); Linear regression analysis between litter weight and 55 day-Hb (B); Linear regression analysis between litter weight and 75 day-Hb (C); Linear regression analysis between litter weight and 95 day-Hb (D); Linear regression analysis between litter weight and 110 day-Hb (E); Linear regression analysis between average weight and 25 day-Hb (F); Linear regression analysis between average weight and 110 day-Hb (G); Linear regression analysis between placental zinc and 55 day-Hb (H); Linear regression analysis between placental iron and 110 day-Hb (I); Linear regression analysis between placental manganese and 110 day-Hb (J).
Effects of hemoglobin content on adverse reproductive performance of pregnant sows
The adverse reproductive performance and hemoglobin levels of 177 sows were collected and analyzed (Table 8). Due to the number of weak litter size, mummified fetus and stillborn were mostly distributed between 0 and 4 and less than 2. We divided them into groups based on the number of adverse reproductive performance and compared the hemoglobin level between different groups. The majority of adverse reproductive performance was observed independent of the hemoglobin level during the whole pregnancy, but the hemoglobin level at day 25 was significantly lower when the number of stillbirths is more than one compared with that without stillbirth (P = 0.050). On the contrary, higher hemoglobin at day 110 may lead to a higher number of stillbirths (P = 0.010).
Table 8.
Effects of hemoglobin level at different gestational ages on adverse reproduction of sow
| Hemoglobin (g/L) | Weak litter size | P-value∗ | Mummy fetus | P-value | Stillbirth | P-value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | >1 | 0 | >1 | 0 | >1 or 1 | 2 | 3 | 4 | ||||
| Day 251 | 9.615 | 9.694 | 0.507 | 9.626 | 9.665 | 0.824 | 9.698 | 9.575 | —— | —— | —— | 0.050 |
| Day 55 | 10.262 | 9.760 | 0.856 | 10.160 | 10.164 | 0.430 | 10.206 | 10.122 | —— | —— | —— | 0.926 |
| Day 75 | 10.122 | 10.158 | 0.609 | 10.132 | 10.103 | 0.311 | 10.196 | 10.078 | —— | —— | —— | 0.424 |
| Day 95 | 9.864 | 10.304 | 0.814 | 9.922 | 10.167 | 0.020 | 9.802 | 10.091 | —— | —— | —— | 0.091 |
| Day 110 | 9.407 | 9.691 | 0.702 | 9.461 | 9.491 | 0.877 | 9.166c | 9.555bc | 9.922ab | 10.675a | 8.970bc | 0.010 |
Day 25, gestation 25 d; day 55, gestation 55 d; day 75, gestation 75 d; day 95, gestation 95 d; day 110, gestation 110 d.
P-value reflects the differences of hemoglobin whether adverse reproduction appears or different stillbirth size.
Values means significant differences (P ≤ 0.05) between different groups.
Linear correlation between hemoglobin content of pregnant sows and placental trace elements content
The linear regression analysis between hemoglobin and the content of trace elements in the placenta was performed (Table 9, Figure 2H–J). Consistent with expectations, the majority of hemoglobin level at days 25, 55, 75, and 95 has no linear correlation with placental trace elements content, while pre-delivery hemoglobin level at day 110 has a significant linear positive correlation with placental Fe and Mn content (P = 0.002, P = 0.013). Beside, hemoglobin level at day 55 was significant and negatively related with placental Zn content (P = 0.042). In addition, there was no significant correlation between placental copper and hemoglobin throughout pregnancy (P > 0.05).
Table 9.
Linear regression between hemoglobin at different gestation ages and placental trace elements content of sow
| Hemoglobin, g/L | Placental iron, mg/kg | Placental zinc, mg/kg | Placental manganese, mg/kg | ||||||
|---|---|---|---|---|---|---|---|---|---|
| β2 | R3 | P-value∗ | β | R | P-value | β | R | P-value | |
| Day 251 | 66.263 | 0.191 | 0.063 | 2.11 | 0.052 | 0.658 | 2.179 | 0.115 | 0.322 |
| Day 55 | 21.987 | 0.065 | 0.512 | −9.929 | 0.219 | 0.042 | −1.738 | 0.088 | 0.417 |
| Day 75 | 2.671 | 0.011 | 0.916 | 4.658 | 0.138 | 0.223 | 1.608 | 0.092 | 0.416 |
| Day 95 | −6.856 | 0.037 | 0.730 | −3.524 | 0.121 | 0.231 | 0.489 | 0.035 | 0.749 |
| Day 110 | 110.923 | 0.328 | 0.002 | 9.565 | 0.178 | 0.106 | 6.859 | 0.269 | 0.013 |
Day 25, gestation 25 d; day 55, gestation 55 d; day 75, gestation 75 d; day 95, gestation 95 d; day 110, gestation 110 d.
β is the linear regression coefficient.
R is the coefficient of determination and reflects the fitting degree of the regression equation.
P-value means the significance of linear regression.
Linear correlation between placental trace elements content and reproductive performance of pregnant sows
Because of the correlation between the content of trace elements in the placenta and hemoglobin, we also conducted a correlation analysis on the content of trace elements in the placenta and reproductive performance (Table 10). To our surprise, placental Fe, Zn, and Mn have a significant linear positive correlation with average weight (P = 0.048, P = 0.027, P = 0.047) rather than other reproductive performances (P > 0.05). And the increase of placental Cu significantly increased litter size (P = 0.029).
Table 10.
Linear regression between placental trace elements content and reproductive performance of sow
| Content of trace elements in the placenta, mg/kg | Litter size | Healthy litter size | Litter weight, kg | Average weight, kg | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β1 | R2 | P-value∗ | β | R | P-value | β | R | P-value | β | R | P-value | |
| Iron | −0.0004 | 0.071 | 0.442 | 0.0002 | 0.037 | 0.665 | 0.0003 | 0.041 | 0.642 | 0.0001 | 0.182 | 0.048 |
| Zinc | 0.0003 | 0.008 | 0.933 | 0.002 | 0.049 | 0.574 | 0.006 | 0.093 | 0.287 | 0.001 | 0.204 | 0.027 |
| Manganese | −0.012 | 0.135 | 0.144 | −0.003 | 0.027 | 0.754 | 0.001 | 0.007 | 0.937 | 0.002 | 0.183 | 0.047 |
| Copper | 0.241 | 0.364 | 0.029 | 0.281 | 0.298 | 0.078 | 0.310 | 0.294 | 0.082 | 0.009 | 0.108 | 0.531 |
β is the linear regression coefficient.
R is the coefficient of determination and reflects the fitting degree of the regression equation.
P-value means the significance of linear regression.
Discussion
This study is the first to monitor the hemoglobin levels and reproductive performance of sows in different parities and gestation periods with large samples, and then a detailed regression analysis was performed between the hemoglobin, reproductive performance, and trace elements in the placenta. Our study is statistically more repetitive and comprehensive than other studies that only compare low parities (Girard et al., 1996); furthermore, different from other studies, we measured hemoglobin levels of sows throughout pregnancy, rather than only in the middle and late trimesters or early trimester, and found a statistically significant linear correlation between hemoglobin levels and litter weight, and average weight and stillbirth (Normand et al., 2012). We also found a possible relationship between hemoglobin and placental trace elements, and between placental trace elements and reproductive performance, that has not been mentioned in other studies.
In all data collected in the first to sixth parities, the hemoglobin levels of sows increased first (d 25 to d 75) and then decreased (day 75 to 110) during the whole pregnancy, and the decrease was especially obvious after day 75. Iron status of sow increases before day 75, which may provide sufficient iron for the fetus to prepare for its growth and development demand. And this trend is extremely obvious in first and second parities, but the decrease of hemoglobin after day 75 was not significant in the third to sixth parities. This may be due to the limited sample size and the low level of hemoglobin in third to sixth parities compared with first and second parities. The comparison of hemoglobin levels at the same gestational periods with different parities showed that hemoglobin levels at days 25 and 55 decreased significantly with the increase of parities, which was consistent with other reports (Normand et al., 2012). However, there was no significant difference in hemoglobin level after day 75 between different parities. Multiple pregnancies reduce iron reserves in the body of sows leading to this change, and as reported in the previous results, hemoglobin level is lower after day 75 in all parities without a difference.
Our results show that sows with higher parities have better reproductive performance, and this trend was evident in litter size, healthy litter size, and litter weight, contradicting previous studies (Hagan and Etim, 2019). But a few other studies have reported that sows with higher parity have better reproductive performance, which is consistent with us (Lavery et al., 2019; Yang et al., 2019). This divergence may be caused by the variety, feeding, season, and other factors affecting reproductive performance beside parity.
Similar to hemoglobin level, the content of Fe, Zn, Mn, and Cu in the placenta decreased significantly with the increase in parities, especially in the sixth parity compared with the first parity. In addition, there is a significant positive linear correlation between Fe, Zn, Mn, and Cu in the placenta, indicating that Fe, Zn, Mn, and Cu have close interaction with each other; iron supplementation during pregnancy plays an important role in the absorption and transport of other elements, and the deficiency of iron may lead to the deficiency of other trace elements, which is consistent with other reports (Osredkar and Sustar, 2011), and thus impact on reproductive performance is more serious than predicted.
To explore the effect of gestational hemoglobin on reproductive performance and explain the above differences, we conducted a detailed correlation analysis of all data. Dewey and Oaks (2017) have reported a U-shaped curve relationship between human gestational hemoglobin and adverse reproductive performance, the risk of stillbirth and preterm birth increases when hemoglobin level is too low as well as too high, and the effect of early pregnancy hemoglobin levels on adverse pregnancy outcome was much greater than that in the middle and late stages. Therefore, we performed linear and quadratic regression analyses on the relationship between hemoglobin levels and reproductive performance at each gestation stage and found that hemoglobin level in all pregnancies has a significant negative linear rather than the parabolic correlation with reproductive performance. The litter weight and average weight were statistically significantly with the increase of hemoglobin, and litter size, healthy litter size also has tend to decreased with higher hemoglobin. Unlike the study on humans, our results are consistent with another sow study, which indicated that higher gestational hemoglobin may be associated with lower litter weight and average weight and better for piglet survival (Hollema et al., 2020). This negative correlation also explained higher-parities sow has better reproductive performance but lower hemoglobin level. Due to the poor efficiency of iron transport from the placenta to fetuses, lower hemoglobin level where dietary iron supplementation is adequate may mean higher efficiency of iron transport in the placenta. It can improve reproductive performance such as litter weight, because of increased iron availability to fetuses. Interestingly, for adverse pregnancy outcomes, we found that higher hemoglobin level at day 25 may reduce the number of stillbirths, while higher hemoglobin level at days 75 and 110 may be associated with a higher risk of mammy fetus and stillbirth, respectively, and the negative trend between stillbirth and hemoglobin in the late pregnancy was consistent with another study (Bhattarai et al., 2019b). Although there is a plenty amount of iron in feedstuff, their insufficient bioavailability always resulted in iron deficiency in target animals. Thus, commercial iron additives are widely used and supplemented, especially to sows’ diet to prevent their iron deficiency. In our present study, the results indicated that very few sows face iron deficiency under the current iron supplementation. What’s more, our results also found the reproductive performance decreases, along with the rise of hemoglobin, indicating the iron in these sows were overload. Moreover, excessive iron supplementation may not improve reproductive performance and even be detrimental to sow production. Consequently, our results revealed that the U-shaped curve relationship between gestational hemoglobin and adverse reproductive performance was also found in pregnant sows. The appropriate iron supplementation dosage recommend for pregnant sows are urgency. And, based on our results, the optimal iron dosage may depend on their own parities, pregnancy periods, iron status, and litter size. Therefore, monitoring the iron status of sows during pregnancy is important to adjust the iron dosage and improve the reproductive performance for clinical application.
As expected, hemoglobin at day 110 is positively related with Fe and Mn in the placenta, suggesting that both hemoglobin levels and placental iron may reflect the iron status of sows. However, placental trace elements have no significant correlation with reproductive performance except average weight, and higher placental iron levels may indicate a lower average weight which is antagonistic to other studies in mice (Kwan et al., 2020; Sangkhae et al., 2020). However, the ability of the placenta to transport iron rather than placental iron is the most important factor which determines the reproductive performance of sows, because the litter weight is affected by litter size and average weight. It is worth further study in relationship between the structure of the placenta and the expression of trace element transporters in the placenta and reproductive performance.
Conclusions
Our study reveals the variation of hemoglobin levels in sows at different parities and gestation periods, which suggests the changes in iron status and requirements during pregnancy. We showed a negative linear correlation between the gestational hemoglobin and reproductive performance, which demonstrates that the pregnancy hemoglobin levels can be regarded as potential predictors of reproductive performance. Meanwhile, the ability of the placenta to transport iron from the mother to the fetus when dietary iron supplementation is adequate may be a determinant of reproductive performance. In addition, higher-parity sows may have better reproductive performance due to higher iron transport capacity. Meanwhile, our study reinforces the importance of reasonable iron supplementation in sows during pregnancy.
Acknowledgments
This work was supported by the National Key R&D Program of China (2016YFD0501201), the Outstanding Youth Fund of Hunan Natural Science Foundation (2021JJ20045), the Young Elite Scientists Sponsorship Program by China Association for Science and Technology (2018QNRC001), the Science and Technology Program of Hunan Province (2020NK2013 and 2020GK4095), and the key R&D Program of Guangxi Province (2021AB200663).
Glossary
Abbreviations
- Hb
hemoglobin
- ICP-MS
inductively coupled plasma mass spectrometry
Conflict of Interest Statement
L.G., D.Z., W.T., Z.D., Y.Y., and D.W. report no conflict of interest. Y.Z. and S.W. are employed by the Changsha Xinjia Bio-Engineering Co., Ltd.
Literature Cited
- Bhattarai, S., Framstad T., and Nielsen J. P.. . 2019a. Association between sow and piglet blood hemoglobin concentrations and stillbirth risk. Acta Vet. Scand. 61:61. doi: 10.1186/s13028-019-0496-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bhattarai, S., Framstad T., and Nielsen J. P.. . 2019b. Hematologic reference intervals of Danish sows at mid-gestation. Acta Vet. Scand. 61:16. doi: 10.1186/s13028-019-0451-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bhattarai, S., Framstad T., and Nielsen J. P.. . 2019c. Iron treatment of pregnant sows in a Danish herd without iron deficiency anemia did not improve sow and piglet hematology or stillbirth rate. Acta Vet. Scand. 61:60. doi: 10.1186/s13028-019-0497-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Buffler, M., Becker C., and Windisch W. M.. . 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. Arch. Anim. Nutr. 71:219–230. doi: 10.1080/1745039x.2017.1301059 [DOI] [PubMed] [Google Scholar]
- Cao, C., and Fleming M. D.. . 2016. The placenta: the forgotten essential organ of iron transport. Nutr. Rev. 74(7):421–431. doi: 10.1093/nutrit/nuw009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dewey, K. G., and Oaks B. M.. . 2017. U-shaped curve for risk associated with maternal hemoglobin, iron status, or iron supplementation. Am. J. Clin. Nutr. 106:1694s–1702s. doi: 10.3945/ajcn.117.156075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dong, Z., Wan D., Yang H., Li G., Zhang Y., Zhou X., Wu X., and Yin Y.. . 2020. Effects of iron deficiency on serum metabolome, hepatic histology, and function in neonatal piglets. Animals (Basel).10(8):1353. doi: 10.3390/ani10081353 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gell, D. A. 2018. Structure and function of haemoglobins. Blood Cells Mol. Dis. 70:13–42. doi: 10.1016/j.bcmd.2017.10.006 [DOI] [PubMed] [Google Scholar]
- Georgieff, M. K., Krebs N. F., and Cusick S. E.. . 2019. The benefits and risks of iron supplementation in pregnancy and childhood. Annu. Rev. Nutr. 39:121–146. doi: 10.1146/annurev-nutr-082018-124213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Girard, C. L., Robert S., Matte J. J., Farmer C., and Martineau G. P.. . 1996. Serum concentrations of micronutrients, packed cell volume, and blood hemoglobin during the first two gestations and lactations of sows. Can. J. Vet. Res. 60:179–185. [PMC free article] [PubMed] [Google Scholar]
- Gulec, S., and Collins J. F.. . 2014. Molecular mediators governing iron-copper interactions. Annu. Rev. Nutr. 34:95–116. doi: 10.1146/annurev-nutr-071812-161215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hagan, J. K., and Etim N. N.. . 2019. The effects of breed, season and parity on the reproductive performance of pigs reared under hot and humid environments. Trop. Anim. Health Prod. 51:411–418. doi: 10.1007/s11250-018-1705-5 [DOI] [PubMed] [Google Scholar]
- Hollema, B. L., Zwiers S., and Hermesch S.. . 2020. Genetic parameters for haemoglobin levels in sows and piglets as well as sow reproductive performance and piglet survival. Animal 14:688–696. doi: 10.1017/s1751731119002532 [DOI] [PubMed] [Google Scholar]
- Kelkitli, E., Ozturk N., Aslan N. A., Kilic-Baygutalp N., Bayraktutan Z., Kurt N., Bakan N., and Bakan E.. . 2016. Serum zinc levels in patients with iron deficiency anemia and its association with symptoms of iron deficiency anemia. Ann. Hematol. 95:751–756. doi: 10.1007/s00277-016-2628-8 [DOI] [PubMed] [Google Scholar]
- Kwan, S. T. C., Kezer C. A., Helfrich K. K., Saini N., Huebner S. M., Flentke G. R., Kling P. J., and Smith S. M.. . 2020. Maternal iron nutriture modulates placental development in a rat model of fetal alcohol spectrum disorder. Alcohol 84:57–66. doi: 10.1016/j.alcohol.2019.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lavery, A., Lawlor P. G., Magowan E., Miller H. M., OʹDriscoll K., and Berry D. P.. . 2019. An association analysis of sow parity, live-weight and back-fat depth as indicators of sow productivity. Animal. 13:622–630. doi: 10.1017/s1751731118001799 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lewicka, I., Kocyłowski R., Grzesiak M., Gaj Z., Oszukowski P., and Suliburska J.. . 2017. Selected trace elements concentrations in pregnancy and their possible role – literature review. Ginekol. Pol. 88:509–514. doi: 10.5603/GP.a2017.0093 [DOI] [PubMed] [Google Scholar]
- Means, R. T. 2020. Iron deficiency and iron deficiency anemia: implications and impact in pregnancy, fetal development, and early childhood parameters. Nutrients 12(2):447. doi: 10.3390/nu12020447 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Normand, V., Perrin H., Auvigne V., Robert N., and Laval A.. . 2012. Anaemia in the sow: a cohort study to assess factors with an impact on haemoglobin concentration, and the influence of haemoglobin concentration on the reproductive performance. Vet. Rec. 171:350. doi: 10.1136/vr.100404 [DOI] [PubMed] [Google Scholar]
- NRC. 2012. Nutrient requirements of swine. 11th rev. ed. Washington (DC): National Academy Press. [Google Scholar]
- Osredkar, J., and Sustar N.. . 2011. Copper and zinc, biological role and significance of copper/zinc imbalance. J. Clin. Toxicol. S. 3:0495. [Google Scholar]
- Padjen, I., Öhler L., Studenic P., Woodworth T., Smolen J., and Aletaha D.. . 2017. Clinical meaning and implications of serum hemoglobin levels in patients with rheumatoid arthritis. Semin. Arthritis Rheum. 47:193–198. doi: 10.1016/j.semarthrit.2017.03.001 [DOI] [PubMed] [Google Scholar]
- Peters, J. C., and Mahan D. C.. . 2008. Effects of neonatal iron status, iron injections at birth, and weaning in young pigs from sows fed either organic or inorganic trace minerals. J. Anim. Sci. 86:2261–2269. doi: 10.2527/jas.2007-0577 [DOI] [PubMed] [Google Scholar]
- Salak-Johnson, J. L. 2017. Social status and housing factors affect reproductive performance of pregnant sows in groups. Mol. Reprod. Dev. 84:905–913. doi: 10.1002/mrd.22846 [DOI] [PubMed] [Google Scholar]
- Sangkhae, V., Fisher A. L., Wong S., Koenig M. D., Tussing-Humphreys L., Chu A., Lelić M., Ganz T., and Nemeth E.. . 2020. Effects of maternal iron status on placental and fetal iron homeostasis. J. Clin. Invest. 130:625–640. doi: 10.1172/jci127341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sangkhae, V., and Nemeth E.. . 2019. Placental iron transport: the mechanism and regulatory circuits. Free Radic. Biol. Med. 133:254–261. doi: 10.1016/j.freeradbiomed.2018.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Solomons, N. W. 1986. Competitive interaction of iron and zinc in the diet: consequences for human nutrition. J. Nutr. 116:927–935. doi: 10.1093/jn/116.6.927 [DOI] [PubMed] [Google Scholar]
- Wan, D., Zhang Y. M., Wu X., Lin X., Shu X. G., Zhou X. H., Du H. T., Xing W. G., Liu H. N., Li L., . et al. 2018. Maternal dietary supplementation with ferrous N-carbamylglycinate chelate affects sow reproductive performance and iron status of neonatal piglets. Animal 12:1372–1379. doi: 10.1017/s1751731117003172 [DOI] [PubMed] [Google Scholar]
- Wang, J., Li D., Che L., Lin Y., Fang Z., Xu S., and Wu D.. . 2014. Influence of organic iron complex on sow reproductive performance and iron status of nursing pigs. Livest. Sci. 160:89–96. doi: 10.1016/j.livsci.2013.11.024 [DOI] [Google Scholar]
- Yang, K. Y., Jeon J. H., Kwon K. S., Choi H. C., Kim J. B., and Lee J. Y.. . 2019. Effect of different parities on reproductive performance, birth intervals, and tail behavior in sows. J. Anim. Sci. Technol. 61:147–153. doi: 10.5187/jast.2019.61.3.147 [DOI] [PMC free article] [PubMed] [Google Scholar]