Introduction

One of the primary economic drivers in pig production is the reproductive efficiency, which is typically defined as the number of piglets produced per sow per year. Optimizing this factor is crucial for the sustainability and productivity of the global pork industry. In recent decades, continuous genetic advancements have focused on increasing sow prolificacy, aiming to raise the number of slaughtered pigs produced per sow per year1. However, this increase in prolificacy has led to a significant reduction in average birth body weight (BW), resulting in a higher number of lightweight piglets at birth2,3. Piglets born small may experience varying degrees of intrauterine growth restriction (IUGR)4,5,6,7 and are likely to struggle to catch up with their heavier counterparts throughout the production cycle2,3,8.

It is estimated that approximately 10–15% of pigs in any given batch exhibit slow growth9. Recent studies have highlighted the identification of this subset of pigs at different production stages by applying different machine learning algorithms to robust and powerful databases10,11. These pigs are at a higher risk of mortality throughout the production cycle9,12. Those that survive present significant management challenges in modern swine production systems, which aim to group contemporary pigs within the same batch to ensure optimal animal health and performance13. In addition, it is also reported that as pigs near the end of the production cycle, the likelihood of lighter pigs being sent to the slaughterhouse at a later stage increase14. Therefore, early identification of slow-growing pigs and the implementation of effective preventive measures are essential to enhance their performance and reduce variability. This is further supported by studies emphasizing that changes in pigs’ BW category are more frequent during the lactation period15,16.

Potentially slow-growing pigs can be easily identified at birth if they exhibit severe IUGR and display a distinctive head morphology previously described17. However, identifying piglets that lack these traits at birth but experience poor subsequent growth performance, persistently remaining in the lower quartile of the birth weight population distribution, becomes a greater challenge. Birth BW is widely recognized as one of the most critical determinants of lifetime growth performance18 and survivability19 in pigs. However, extensive scientific literature research has also explored other individual traits influencing pre-weaning mortality and growth. These include morphometric attributes6,7,20 physiological characteristics21,22,23 and behavioral features24. In addition, risk factors related to the sow, the environment, and BW variability14,25,26 among others, have also been examined highlighting their collective impact on piglet growth and survival.

The identification of slow-growing pigs at birth would facilitate the implementation of targeted management and feeding strategies during early life, allowing farmers to optimize herd management and enhance overall farm production efficiency15. Within these strategies, particular attention should be given to those focusing on piglet care during their first days of life. A common practice in pig production, particularly for supporting low-birth weight piglets, is colostrum supplementation. For example27, reported that orally supplementing weaker piglets with freshly milked colostrum from sows within the same herd helped stabilize their body temperature and enhanced litter performance within the first 24 h. Another strategy is split suckling prior to cross-fostering, which has been shown to improve daily weight gain and litter uniformity at weaning28. Equally important is the feeding management of piglets during the early stages. One widely used strategy is creep feeding, which stimulates feed intake and promotes post-weaning growth. This practice has been associated with increased total weight gain in eater pigs29,30 reduced weight variability at weaning31 and improved growth during the early post-weaning period29,32,33. Although less common than creep feeding, milk replacer supplementation has also demonstrated significant benefits, including higher weaning weights, increased total litter weight34,35 and enhanced weight gain after weaning16.

Among the performance criteria described in the literature for identifying these growth-restricted piglets, birth BW, along with other indicators of body conformation, such as body mass index (BMI)6 and crown-to-rump length (CRL; the piglet’s supine length measured from the crown of the head to the base of the tail), stand out as the most widely studied. Despite extensive investigation, little is known about their correlation with internal measurements, such as those related to the physiological status of organs and bones, which are key factors in postnatal growth. The interest in assessing bone characteristics in piglets at birth lies in finding a more specific indicator of skeletal development than birth weight alone. This could help identify neonates at greater risk of compromised postnatal growth trajectories. In humans, several studies have demonstrated that neonates with IUGR exhibit altered biochemical markers of bone formation, such as lower circulating osteocalcin and 1.25-dihydroxyvitamin D concentrations, relative to appropriate-for-gestational-age infants36. These alterations arise in part from impaired endochondral ossification under conditions of placental insufficiency and maternal metabolic dysfunction, which limit the availability of substrates necessary for normal skeletogenesis36,37. Importantly, bone physiological status has been linked to growth and bone mass accretion during infancy. For example38, showed that IUGR-associated reductions in endochondral ossification correlate with impaired postnatal growth velocity and lower bone strength in early childhood.

We hypothesized that piglets with low body metrics at birth exhibit impaired organ and bone physiological status compared to their heavier counterparts. Identifying strong correlations between external and internal measures is essential for reliably identifying potentially slow-growing piglets at an early stage. This early identification would enable the timely implementation of effective intervention strategies for this subset of pigs. In this context, the objectives of this study were to (1) characterize piglets at the level of organ and bone physiological status at birth, and (2) investigate whether these characteristics correlate with individual external physical metrics currently used as indicators of survivability and growth performance in suckling piglets.

Results

Descriptive results

Average birth BW of severe intrauterine growth restricted (sIUGR), low birth weight (LBW), moderate birth weight (MBW) and high birth weight (HBW) piglets were 0.49 ± 0.037 kg, 0.97 ± 0.042 kg, 1.18 ± 0.027 kg, and 1.69 ± 0.121 kg, respectively. Significant differences in birth BW between these four categories of piglets were also reflected in BMI and CRL measurements, likely due to the strong correlation among these variables. Pearson’s correlation coefficient analysis showed that birth BW was significantly positively correlated with BMI (R = 0.870; P ≤ 0.001) and CRL (R = 0.935; P ≤ 0.001). The HBW piglets exhibited the highest BMI and CRL values, while the MBW and LBW piglets had intermediate values that did not differ significantly from each other. The sIUGR piglets showed the lowest values for both BMI and CRL.

Organ weights

Relative organ weights and the different organ ratios calculated are shown in Table 1. The relative brain weight was higher in sIUGR piglets (P < 0.001) and progressively decreased across categories, reaching the lowest values in HBW piglets. The brain: heart, brain: liver, and brain: lungs ratios were also significantly higher in sIUGR piglets compared to all other categories (P ≤ 0.009), which showed no significant differences among themselves. These differences revealed that sIUGR piglets had a significantly larger relative brain size compared to the other categories. The spleen was significantly larger (P = 0.027) in the lowest weight categories (sIUGR and LBW) compared to MBW and HBW piglets, while the relative stomach weight showed only a tendency (P = 0.068) to be greater in the lowest weight categories (sIUGR and LBW).

Table 1 Effect of piglet type (sIUGR, LBW, MBW, or HBW) on organ weights at birth.
Full size table

Computed tomography (CT) measurements

Results of the CT measurements are presented in Table 2. As expected, there was a significant difference in the total bone volume at birth among the four categories of piglets (P < 0.001). The HBW piglets exhibited the largest bone volume, with values ranging from 173.25 to 193.34 cm3while the sIUGR piglets had approximately 70% lower volumes. No significant differences were detected between the LBW and MBW piglets, which had intermediate bone volumes. No differences were observed in the relative total bone volume across the four piglet categories. Furthermore, the proportion of bones with different densities also differed significantly depending on the piglet category (P < 0.001). Specifically, piglets with lower bone volume had a higher percentage of low dense bones (low Hounsfield unit (HU) values), whereas piglets with higher bone volume had a higher percentage of high dense bones (high HU values).

Table 2 Effect of piglet type (sIUGR, LBW, MBW, or HBW) on computed tomography measurements at birth.
Full size table

Specific measurements of the left tibiae revealed that HBW piglets had a higher cortical area (P < 0.001) compared to the other categories, with an average value of 35.67 mm2. The cortical area of the left tibia was reduced by approximately 60% in sIUGR piglets (P < 0.001), while LBW and MBW piglets exhibited intermediate values, showing a tendency to differ from each other (P = 0.069). HBW piglets also showed greater tibial thickness (P = 0.01) compared to the lower weight categories (sIUGR and LBW), with an average value of 2.66 mm. Both sIUGR and LBW piglets had the lowest left tibial thickness, whereas MBW piglets exhibited values not different from other piglets. Similarly, the mean HU values differed significantly among categories (P = 0.041), although no differences were observed between individual categories. The mean HU values for sIUGR and LBW piglets, as well as for MBW and HBW piglets, were numerically similar.

Mechanical and geometric magnitudes of piglet tibiae

The results of the mechanical and geometric properties of piglet tibiae are presented in Table 3. The maximum stress values (σC, max and σT, max) were significantly higher (P = 0.002 and P = 0.001, respectively) in the tibiae of sIUGR piglets compared to those of the other piglet categories. The Young’s modulus values also differed significantly (P = 0.006) among the piglet categories, with sIUGR piglets exhibiting higher values than both MBW and HBW piglets (P < 0.05) (differences reaching up to 7281.40 MPa). In contrast, significant differences (P < 0.001) in the area moment of inertia (Iz) and cross-section (A) of the tibiae were observed in the opposite direction, with HBW piglets showing the highest values for both variables.

Table 3 Effect of piglet type (sIUGR, LBW, MBW, or HBW) on mechanical and geometric properties of piglet tibiae at birth.
Full size table

Mineral content in piglet tibiae

The mineral content of piglet tibiae is shown in Table 4. The phosphorus (P), calcium (Ca), and copper (Cu) content in the tibiae of piglets did not differ among the four categories studied. However, the manganese (Mn) content varied depending on the piglet category (P = 0.028), with sIUGR piglets exhibiting the highest levels and HBW piglets the lowest, showing differences of up to 15.4 mg/g (P = 0.031). The Mn content in LBW and MBW piglets was intermediate, not different from that of either sIUGR or HBW piglets. Additionally, there was a tendency toward decreasing zinc (Zn) content (P = 0.067) from sIUGR piglets to HBW piglets, although the differences among the piglet categories were not statistically significant.

Table 4 Effect of piglet type (sIUGR, LBW, MBW, or HBW) on the mineral content of piglet tibiae at birth.
Full size table

Correlations between external physical characteristics (birth BW, BMI, CRL) and internal parameters

The correlations between the three external physical characteristics used to classify piglets at birth (birth BW, BMI, and CRL) and the internal parameters analyzed post-mortem are presented in Table 5. Four correlations involving various organ weight measures and external parameters had strong negative correlation coefficients below − 0.63 (P < 0.01). Among these, the relative brain weight exhibited the strongest negative correlations with external parameters (–0.93 for birth BW, − 0.88 for BMI, and − 0.93 for CRL), followed by the organ weight ratios of brain: lungs, brain: liver, and brain: heart. Similarly, the mechanical properties of maximum compression stress, maximum tensile stress, and Young’s modulus also showed negative correlations with external parameters, with coefficients ranging from − 0.62 to − 0.47 (P < 0.05). The Mn and Zn content in the piglet tibiae also demonstrated negative correlations with external parameters, with coefficients ranging from − 0.56 to − 0.48 (P < 0.05).

Table 5 Correlation coefficients between birth body weight, crown-to-rump length, body mass index, and internal body variables.
Full size table

Conversely, the parameters mainly related to skeletal development were positively correlated with the external parameters. Specifically, the entire bone volume (EBV) showed the strongest positive correlation with birth BW (0.98, P < 0.001), CRL (0.92, P < 0.001), and BMI (0.84, P < 0.001). Bone density was also correlated with external parameters, with the proportion of high and very high-density bones exhibiting positive correlation coefficients ranging from 0.50 to 0.77 (P < 0.05). Furthermore, regarding piglet tibiae measures and geometric magnitudes, the cortical area (0.74 for BMI to 0.91 for birth BW), area (0.74 for BMI to 0.88 for birth BW), and inertia of the Sect. (0.64 for CRL to 0.74 for BMI) were also positively correlated with external parameters (P < 0.001).

Discussion

Birth BW is a critical indicator of postnatal performance, as piglets born small often remain stunted and are unable to catch up to their larger counterparts throughout the entire production cycle8,39,40. Some studies have defined pigs with a birth BW of less than 1.0–1.25 kg and without organ growth restriction as low birth BW pigs15,18,41,42. Accordingly, in the current study, piglets were selected within the lower and upper thresholds of these established cut-off points to identify differences among those considered small at birth. Piglets within the lower threshold had a mean birth BW of 0.97 ± 0.037 kg and were considered LBW, while those within the upper threshold had a mean birth BW of 1.18 ± 0.027 kg and were considered MBW. As some studies suggest that pigs born small may have the potential to compensate during suckling2,15,16 and subsequent growth stages25,43,44 it is important to evaluate and differentiate their characteristics at birth. To encompass the entire birth BW distribution of the population, piglets with the characteristic head morphology of severe IUGR17 along with those with a birth BW exceeding 1.25 kg, were also evaluated.

While the link between birth BW and survivability is well established19 additional measures of body conformation, such as BMI and CRL, also serve as important indicators of survivability6,45,46. Furthermore, body shape at birth has been identified as a reliable predictor of postnatal growth in piglets born with low birth BW25. In this context, it is suggested that not all light pigs are similar and that their performance is more strongly influenced by body shape than by birth BW20. In this study, we found high correlation coefficients between BMI and CRL and birth BW (0.870 and 0.935, respectively). However, no significant differences in these body shape metrics were observed between LBW and MBW piglets. These findings suggest that differentiating light piglets at birth based just on morphometric characteristics is challenging, and that internal physiological status likely may play a critical role in their subsequent growth performance.

Despite the strong interest in extensively studying the internal physiological status of newborn piglets, these traits are difficult to estimate non-invasively, often requiring euthanasia and expensive analytical techniques. The main limitation of this study is the relatively small sample size, with only 6 piglets per category. Consequently, the findings should be regarded as a preliminary step toward understanding the physiological status of organs and bones in newborn piglets based on their birth characteristics, which may significantly impact their future growth performance. Further research involving a larger sample size is crucial to improve the reliability and generalizability of these results. Assessing organ and bone characteristics at birth could aid in identifying neonates at greater risk of compromised postnatal growth trajectories. Similarly, other traits, such as the number and size of muscle fibers, have been extensively studied as critical factors influencing postnatal growth potential. Low birth weight piglets tend to have fewer muscle fibers due to a reduced ratio of secondary to primary fibers47,48. Notably, since muscle fiber hyperplasia is completed by day 90 of gestation, the alterations in fiber number observed at birth are permanent49. These effects directly affect the concept of fetal programming, which may significantly influence neonatal survival, as well as subsequent growth and efficiency in pigs50.

The increased litter size in pigs over the past decades has led to uterine crowding. This, in turn, limits nutrient availability to the embryos during critical periods of development, compromising placental development in surviving fetuses and increasing the likelihood of having runt fetuses451,52 demonstrated that intrauterine crowding becomes a limiting factor when the number of embryos exceeds 14. In this way, piglets affected by IUGR experience asymmetrical organ growth during gestation as part of a fetal adaptative response to placental insufficiency5,53,54. The brain and heart are identified as the organs most protected from growth restriction compared to others. Consistent with these studies, the relative brain weight was significantly greater in sIUGR piglets at birth, showing a significant linear decrease as the piglets exhibited better external physical metrics. However, these differences among piglet categories were not reflected in organ ratios, where significant differences were observed only between the sIUGR group and the other categories23,55. This was most likely due to what has been described as the “brain-sparing effect”54 which primarily affects piglets with the most severe degree of IUGR. Another study revealed that the preferential growth of the nervous and cardiovascular systems in small fetuses can be detected as early as day 45 of gestation56. Interestingly, most relative organ weights were comparable between sIUGR and LBW piglets but differed numerically from those of MBW and HBW piglets, which were also similar to each other.

Differences in bone growth and development were also observed among the 4 categories of piglets. The growth in bone length is stated that depends on body size and, consequently, on nutritional status57. Restricted and undernourished fetuses are smaller than normal and possess bones that are anatomically less developed compared to well-nourished fetuses of the same age57. In agreement with that, HBW piglets exhibited the highest EBV at birth, which was approximately 70% greater than that of the sIUGR piglets. In humans was described that skeletal ossification is more advanced than normal in large fetuses and delayed in those small for their gestational age58. This observation could be related to differences in the proportions of bones with higher or lower densities identified in our study. In this context, sIUGR piglets exhibited a higher percentage of low-density bones and a lower percentage of high-density bones compared to their heavier counterparts. Regarding LBW and MBW piglets, differences in bone density were observed between these two categories, indicating that not all light piglets at birth (< 1.25 kg) exhibit the same physiological status. The proportions of bones with different densities were determined using the HU value cut-off ranges originally described for adult pigs59,60 where higher HU values indicate greater bone density61. Despite this, these ranges successfully differentiated between bones of varying densities in newborn piglets, demonstrating the sensitivity of this method and its applicability to small piglets.

These findings on organ development, ossification, and bone growth in newborn piglets support the statement that the lower preweaning growth observed in runt pigs cannot be fully explained by their lower birth BW. This suggests that IUGR exerts a more complex influence on development potential62. This influence likely originates during gestation, where growth-retarded pig fetuses are associated with smaller placentas characterized by slower blood flow63 and fewer, less dense areolae45 compared to their littermates. The placenta plays a pivotal role in fetal growth and development as it facilitates the transport of nutrients and respiratory gases between maternal and fetal circulation64. Consequently, fetuses associated with smaller placentas exhibit alterations in fetal organ development, changes in the number and type of muscle fibers4 and impaired endochondral ossification36,37 thereby compromising their future developmental potential. Within-litter variation in fetal weight is reported to be detectable as early as day 28 of gestation and increases as gestation progresses and fetuses grow56. This finding suggests that small fetuses identified early in gestation are likely to maintain a reduced growth rate, may experience further compromise in later gestation, and are ultimately born as low-weight or IUGR piglets. The differences we observed, primarily in brain weight, EBV, and bone density among the various categories of piglets, may help identify varying degrees of restriction likely established during gestation. These intrauterine growth restrictions seem to have a direct impact on the piglets’ postnatal growth and development.

To gain a deeper understanding of the differences among these growth-restricted piglets at birth, we also analyzed various parameters related to their tibiae. The tibia was selected as the optimal long bone for analysis for several reasons. First, it allows for the most precise measurements using CT, facilitating the calculation of specific metrics previously described in the literature60. Additionally, the tibia is well-known as a significant mineral reservoir, particularly for Ca and P. Its mineral content and mechanical properties have been studied in pigs in previous research65,66. Proper bone growth, along with normal joint structure and function, is essential for the overall development of the organism67. Therefore, piglets with a larger cortical area and greater tibial thickness are likely to exhibit better development at birth compared to their littermates, as demonstrated in our study. High birth BW piglets had the largest cortical area, approximately 60% greater than that of sIUGR piglets. Similarly, HBW piglets exhibited greater tibial thickness compared to the lower weight categories (sIUGR and LBW). These two measures are influenced by the process of bone growth, which is regulated by a complex system of endocrine signals. It is described that the growth of long bones is influenced by growth factors, dietary factors, and other cellular signals that are essential for chondrocyte differentiation68,69,70. Among those factors that can damage the growth plate, IUGR piglets are being associated with impaired plasma IGF-1 levels71. Since insulin/IGF-1 signaling plays a pivotal role in early development and fetal growth, IUGR pigs are more prone to skeletal disorders caused by disruptions in the endocrine regulation of the growth plate67. Indeed, IGF-1 not only stimulates longitudinal bone growth but also has demonstrated direct effects on bone modeling, supported by both in vivo72 and in vitro73 evidence.

Although our results suggest that piglets born light are more likely to experience impaired ossification, the mechanical properties of the tibiae, including maximum stress in compression and tension, as well as rigidity, were highest in the sIUGR group compared to the other categories. This could be supported by the following: (1) these mechanical properties are intrinsic to the bone itself and, therefore, remain independent of both its size and the piglet’s birth BW, and (2) as part of a fetal adaptive response to placental insufficiency and limited access to nutrients from the sow, similar to the “brain-sparing effect”, piglets affected by IUGR may prioritize intrinsic bone quality over longitudinal bone growth. Although this specific physiological phenomenon has not been extensively studied in the literature57, described that full-term runts had bones and epiphyses with a chemical composition more aligned with their age than their size. This implies that, regardless of weight, piglets can achieve chemical maturation of their bones as a function of age. Moreover, this study identified differences in the mineral content of the tibiae, which correlate with the observed variations in mechanical properties. Greater bone mineralization is associated with improved intrinsic bone quality. Ca and P, both essential for optimal bone formation and mineralization66 displayed similar levels in both sIUGR and HBW piglets. Hydroxyapatite (Ca5(PO4)3(OH)), a mineral form of Ca and P that provides bones with their rigidity and strength, is the primary and essential component of bones in animals74,75. Differences were observed in the Mn and Zn content, both of which play crucial roles in bone and cartilage development76,77,78. The highest Mn and Zn levels were found in sIUGR piglets, with a linear decrease observed toward the HBW piglets. Given that Ca and P levels were consistent across piglet categories, and sIUGR exhibited higher Mn and Zn content in their bones, the impaired ossification identified in these sIUGR piglets did not lead to changes in intrinsic bone properties, such as rigidity and strength. Mn plays a role in bone and cartilage matrix formation through its involvement with proteoglycans76 while Zn influences bone mass and is critical for matrix quality77,78. Regarding the mechanical properties and mineralization of the bone, it is important to note that no differences were found between LBW and MBW piglets. Since piglets were euthanized immediately after birth without having access to colostrum, this suggests the existence of an anticipated fetal adaptive response in bone quality that appears to be evident only in piglets with a severe degree of IUGR.

This characterization of piglets at organ and skeletal development levels will enable a more precise assessment of their susceptibility and subsequent risk of mortality due to their IUGR status at birth. Previous studies5,79 noted that some small piglets can still achieve their genetic growth potential, exhibiting normal growth, whereas piglets affected by IUGR do not. These differences in genetic growth potential and survivability among low birth BW may be attributed to the asymmetrical organ and skeletal growth experienced by IUGR piglets during gestation. In the present study, the most pronounced differences were observed between HBW and sIUGR piglets, while less significant differences were noted between LBW and MBW piglets. LBW piglets exhibited greater similarity to sIUGR piglets, whereas MBW piglets were more comparable to HBW piglets. This suggests that although light piglets at birth (LBW and MBW piglets) may be affected to varying degrees by a “mild-moderate” form of IUGR, distinguishing them clearly across different categories remains challenging. These two categories of light piglets studied align with the findings of15 which identified two distinct populations among piglets born with low BW based on their future growth performance. The authors further propose that incorporating additional variables measured at birth, especially in population with high birth BW variability, could allow for the identification and study of additional piglet categories. Other easily measurable on-farm indicators, such as the piglet’s relative birth BW within its litter6 the difference between the piglet’s birth BW and the average litter weight at birth or after cross-fostering10,11 and colostrum intake5 among others, should be used as complementary indicators alongside those employed in this study. Incorporating these measures would enable a more precise and robust classification of light piglets at birth.

Identifying piglets with some degree of IUGR at birth using easily measurable on-farm indicators offers farmers a quick and practical method to determine which piglets require additional support to optimize neonatal survival and growth performance. Our study demonstrates that birth BW, BMI, and CRL are strongly correlated with most of the internal characteristics analyzed, confirming their reliability as indicators of survivability and growth potential. Specifically, these morphological traits showed strong negative correlations with relative brain weight and organ ratios, and strong positive correlations with total bone volume and the longitudinal growth of the tibia. The results of the present study indicate that these external indicators can explain the organ and skeletal physiological status of piglets at birth but are not precise or accurate enough to distinguish between light piglets at birth (LBW and MBW piglets). The interest in this pig population lies in their varying growth performance potential and their ability to transition between categories of weight during the early stages of life. Some authors have highlighted the potential benefits of prediction methods using large data sets to identify pigs at risk of growth retardation10,11. Further research is needed to improve the early identification of piglets affected by varying degrees of IUGR.

This study provides a deeper understanding of organ and bone physiological status in newborn piglets classified by birth weight, body mass index, and crown-to-rump length. Piglets experiencing some degree of IUGR exhibit asymmetrical organ growth favoring brain development, along with impaired ossification and longitudinal bone growth. Furthermore, these piglets show enhanced mechanical properties of the tibia, attributed to increased manganese and zinc content. The findings of this study reveal that the strong correlation between morphological characteristics at birth and organ and skeletal physiological characteristics is insufficient to accurately identify piglets classified as LBW or MBW, which are small at birth and with a “mild-moderate” degree of IUGR. Improving this classification criterion by incorporating complementary indicators at birth could enable pig farmers to design specific management and nutritional strategies targeted at this vulnerable population, ultimately enhancing their performance and survivability.

Methods

All animal experimentation procedures were approved by the Ethics Committee of the Universitat Autònoma de Barcelona (CEEAH2788M2) in accordance with the European Union guidelines for the care and use of animals in research80. This study is reported in accordance with ARRIVE guidelines.

Animals and experimental design

This study was conducted with 24 piglets (Pietrain x [Danish Landrace x Danish Yorkshire]) from 6 multiparous sows of the same batch, sourced from a commercial farm in Lleida, Spain. The average total litter size was 20.2 ± 1.20 piglets, and the average total litter weight was 26.8 ± 2.00 kg. On the day of parturition (day 0), before litter equalization, piglets were individually weighed, and the BMI and CRL were measured. The degree of IUGR was determined based on the head morphology criteria described by17. Four piglets per sow were selected according to the previously described measures to generate four different groups (n = 6 piglets/group): sIUGR, LBW, MBW, and HBW. Piglets classified as sIUGR exhibited a steep, dolphin-like forehead, bulging eyes, and wrinkles perpendicular to the mouth. Piglets with a birth BW between 0.75 and 1.0 kg, without morphological signs of sIUGR, were categorized as LBW, while those with a birth BW between 1.0 and 1.25 kg were considered MBW. Piglets with a birth BW exceeding 1.25 kg were classified as HBW. All descriptive statistics by group are provided in Table 6. After classifying the piglets, they were euthanized with an intramuscular injection of azaperone (Sediron 40 mg/ml; Livisto Int’l, S.L.) and ketamine (Ketamidor 100 mg/ml; VetViva Richter GmbH), followed by an intravenous injection of sodium pentobarbital (Euthasol 400 mg/ml; Dechra Pharmaceuticals PLC).

Table 6 Descriptive statistics for birth body weight (BW), body mass index (BMI), and crown-to-rump length (CRL) [Mean ± Standard deviation (SD)] for piglets (n = 6 piglets per group) classified as severe intrauterine growth restricted (sIUGR), low birth weight (LBW), moderate birth weight (MBW), and high birth weight (HBW).
Full size table

Computed tomography, post-mortem examination, and organ and tibiae sampling

Piglets were fully scanned post-mortem after birth with a Philips Brilliance 16 scanner (Philips Healthcare, Madrid, ES). The CT scanner undergoes daily internal calibration prior to use. The acquisition conditions were as follows: 120 kV, 200 mA, field of view (FOV) 150 mm, 512 × 512 matrix, collimation 16 × 1.5, helical 3 mm-thick images every 3 mm (pitch 0.938). The reconstruction algorithm used was the standard.

Image analysis was performed with software Matlab 2008b (The MathWorks Inc., Natick, MA, USA) using a customized script. From all the images of each piglet, frequency of voxels (3D pixels) associated with each HU value, were obtained and transformed into volume using the image thickness, matrix size and FOV values81. The volume corresponding to HU values between + 140 and + 1,700 was classified as bone tissue82 and this range was used to calculate the total volume of bones. Moreover, four bone density ranges were defined in order to obtain volume measurements of bones with different densities: low dense bones between HU values + 140 and + 499, medium dense bones between HU values + 500 and + 999, high dense bones between HU values + 1,000 and + 1,499, and very high dense bones between HU values + 1,500 and + 1,700, following60 but including an additional range. Based on the volume of each range and the total bone volume, the proportion corresponding to each density range was determined.

Measurements were taken axially in the central part of the left tibia using VisualPork software83. External and internal cortical bone areas were manually selected as regions-of-interest (ROI), and the area of the cortical bone was obtained as a difference between the two areas. Moreover, the average HU value of the cortical area and the thickness of the cortical bone were also determined (Fig. 1).

Fig. 1
Fig. 1
Full size image

The area of the cortical bone (left) and thickness of the cortical bone (right) measured in tibia bone.

Afterwards, the brain, heart, liver, lungs, kidneys, and spleen were harvested and blotted dry before these were weighed on a precision scale (Sartorius AG, Model ED224S, Germany). The intestine and stomach were carefully removed intact, rinsed thoroughly and weighed. The hind legs of each piglet were separated from the rest of the body and individually frozen for subsequent tibiae analyses (mechanical and geometric properties and mineral determination).

Bending tests for mechanical properties of piglet tibiae determination

The left hind leg of each piglet was thawed and placed in the autoclave at 121 °C for 30 min. Then, the flesh was removed from those hind legs to obtain only the clean left tibiae. The tibia density (ρ) of each specimen was determined from two measurements, the unsubmerged weight (Pair) and the weight of the same sample submerged (Psum, the latter measurement being equal to the weight in air plus the buoyant force of the fluid). The geometric magnitudes, including the cross-sectional area and moment of inertia, were calculated and subsequently used to determine the following properties. The mechanical properties of porcine bone (left piglet tibiae) were determined from three point bending tests using a Universal Test Machine (UTM) Zwick All-around table-top 5 kN (ZwickRoell S.L., Ulm, Germany), equipped with a load cell of 500 N. The real values of A and Iz of the fractured cross-section were computed for each specimen from an image using ImageJ® software. Specifically, the mechanical properties calculated were the following: stiffness (Young’s modulus) (E), maximum stress (σmax), ultimate strain (εu), proof resilience (We), and toughness (Wu). The mechanical properties E, σmax, We, and Wu were determined through analytical approximations from the final stress-strain curves. The εu was calculated from the maximum displacement.

All tibia specimens were tested at the same displacement rate (v = 0.5 mm/min). This rate was controlled by the UTM, which simultaneously recorded both displacement and applied force. Due to differences in specimen size and strength, displacement at failure consistently ranged between 0.4 and 1.0 mm, resulting in total test durations of 75 to 125 s. Both the strain rate (dε/dt) and the loading rate (dF/dt) increased linearly with the imposed displacement rate (v), although they were also influenced by the cross-sectional dimensions and strength of each tibia. The calculated strain rate values ranged from 2.0 × 10^(−4) to 1.0 × 10^(−5) s−1, depending on the specimen geometry. These values fall within the quasi-static regime, ensuring that the mechanical behavior corresponds to a rigid-elastic model with brittle failure. Higher testing speeds would induce viscoelastic effects, which would hinder meaningful comparisons between specimens. The loading rate, in turn, was affected not only by the displacement rate but also by the stiffness and strength of the individual tibiae. In all cases, the loading rate ranged from 1.0 to 4.0 N/s.

Inductively coupled plasma mass spectrometry (ICP-MS) for piglet tibiae mineral determination

For the right hind legs, as with the left, they were thawed and placed in an autoclave at 121 °C for 30 min to obtain clean right tibiae. Tibiae bones were then dried at 103 °C for 12 h, followed by immersion in acetone for 48 h to remove any remaining fat residues. Finally, to obtain the tibiae ash, they were heated again at 103 °C for 12 h and then placed in a muffle furnace at 550 °C for 12 h.

Concentrations of P, Ca, Zn, Cu, and Mn were determined in the ash obtained from the right tibiae samples according to the AOAC 984.27 method, using a mass spectrometer ICP-MS 7900 (Agilent Technologies, Santa Clara, California, USA).

To validate the accuracy of the analytical method, a standard was added to the samples (spiked samples). The recovery percentages for these spiked samples were 98% for P, 97% for Ca, 105% for Mn, 101% for Cu, and 93% for Zn. Furthermore, the minimum concentrations quantifiable in the sample using the described methodology were 8 mg/g for P, 10 mg/g for Ca, 0.2 mg/g for Mn, 2 mg/g for Cu, and 2 mg/g for Zn.

Statistical analysis

All calculations and statistical analyses were performed with open-source software R v.4.4.084. Pig was used as the experimental unit in all data analyses. Residuals were tested for normality using the Shapiro-Wilk test (ols_test_normality function, olsrr package) and assessed with normal probability plots. Birth BW, BMI, and CRL measures were analyzed using a linear mixed-effect model (lmer function, lme4 package), with piglet category as a fixed effect and sow as a random effect. All other examined variables, except those related to the proportion of bones with different densities, were similarly analyzed using a linear mixed-effect model (lmer function, lme4 package), including category of piglet as fixed effect and sow as random effect. The variables related to the proportion of bones with different densities were analyzed using a generalized linear mixed-effect model (glmer function, lme4 package) with a binomial family, also including piglet category as fixed effect and sow as random effect. Multiple means comparisons were done using Tukey-Kramer’s correction. Results for the fixed effects are presented as least squares means with the standard error of the mean (SEM). Results were considered significant at P ≤ 0.05 and a tendency when 0.05 < P ≤ 0.10. Pearson correlations were calculated using the cor.test function.