Noninvasive Ultrasound Monitoring of Embryonic and Fetal Development in Chinchilla lanigera to Predict Gestational Age: Preliminary Evaluation of This Species as a Novel Animal Model of Human Pregnancy.

Ultrasound is a noninvasive routine method that allows real-time monitoring of fetal development in utero to determine gestational age and to detect congenital anomalies and multiple pregnancies. To date, the developmental biology of Chinchilla lanigera has not yet been characterized. This species has been found to undergo placentation, long gestation, and fetal dimensions similar to those in humans. The aim of this study was to assess the use of high-frequency ultrasound (HFUS) and clinical ultrasound (US) to predict gestational age in chinchillas and evaluate the possibility of this species as a new animal model for the study of human pregnancy. In this study, 35 pregnant females and a total of 74 embryos and fetuses were monitored. Ultrasound examination was feasible in almost all chinchilla subjects. It was possible to monitor the chinchilla embryo with HFUS from embryonic day (E) 15 to 60 and with US from E15 to E115 due to fetus dimensions. The placenta could be visualized and measured with HFUS from E15, but not with US until E30. From E30, the heartbeat became detectable and it was possible to measure fetal biometrics. In the late stages of pregnancy, stomach, eyes, and lenses became visible. Our study demonstrated the importance of employing both techniques while monitoring embryonic and fetal development to obtain an overall and detailed view of all structures and to recognize any malformation at an early stage. Pregnancy in chinchillas can be confirmed as early as the 15th day postmating, and sonographic changes and gestational age are well correlated. The quantitative measurements of fetal and placental growth performed in this study could be useful in setting up a database for comparison with human fetal ultrasounds. We speculate that, in the future, the chinchilla could be used as an animal model for the study of US in human pregnancy.


Introduction
e developmental biology of the domesticated long-tailedChinchilla lanigera, a South American species, is not well characterized. Recently, it has been shown that placental and fetal metabolism and the placental vessels of chinchillas are very similar to those of humans [1,2]. e chinchilla placenta is of the haemomonochorial labyrinthine type and therefore resembles the human villous haemomonochorial placenta, demonstrating that this species is suitable for human obstetric research and for the study of placental and fetal functions [3][4][5]. Females typically twin (range: 1-6 pups) after a gestation of ∼112 days (range: 105-115 days) and give birth to 2-3 litters per year, resulting in relatively fewer offspring than those of other rodent species [6]. e reproductive physiology of the hystricomorph chinchilla is different from that of myomorphic rodents typically used in biomedical research. At the same time, the chinchilla has a long gestation period and a long estrus cycle compared with other rodents; for example, the gestational period of guinea pigs ranges from 59 to 72 days, that of rats from 21 to 23 days, and that of mice from 19 to 21 days [7][8][9][10]. e long gestation and the reduced number of fetuses per gestation allow for better visualization of the embryo, longer longitudinal monitoring, and a more detailed analysis of the placenta and fetal organs. In addition, the weight of the neonatal chinchilla is relatively high (50-70 g) compared to the weight of the adult chinchilla and or of other rodent models, which facilitate instrumental ultrasonographic examination [8].
Ultrasonography is the imaging technique of choice for analyzing embryonic and fetal formation in utero [11][12][13][14][15]. Until a few decades ago, structural phenotyping was based on macroscopic examinations and histological techniques that only allowed the postmortem analysis of static structures. Subsequently, different imaging approaches have become available for the study of small animals, such as ultrasonography, magnetic resonance imaging (MRI), optical imaging, and confocal biomicroscopy [2,[16][17][18]. e ability of ultrasonography to acquire longitudinal data in real time noninvasively, the ability to analyze the morphology of nearly all the organs, and the ability to perform quantitative measurements on most of the structures have made this imaging modality particularly advantageous in the field of medical imaging. Conventional ultrasonography uses a range of frequencies, from 2 to 15 MHz, with a spatial resolution of 200-500 μm; the limitations of these systems have become evident in several studies on embryonic development in the mouse, particularly during the morphogenesis phase [11]. Technological progress has also led to the development of high-frequency ultrasonography (HFUS), which achieves microscopic resolution and is thus the best imaging technique to monitor the embryonic development of small animals. HFUS systems use higher frequencies, between 40-100 MHz, and have a spatial resolution of approximately 30 μm; therefore, they are useful for imaging developmental processes occurring in small organisms [11,13].
Previously, our research group used HFUS to monitor embryonic and fetal development in mice. We assessed changes in phenotypic parameters during pregnancy and evaluated physiological fetal parameters of the principal organ development to build a database of normal structural and functional parameters of mouse development [12,13]. Due to the chinchilla embryos' dimensions, it was not possible to monitor the entire pregnancy with HFUS (with the exception of some specific parameters), but it was necessary to use conventional ultrasound in the advanced stage of pregnancy. Up until today, there are no published reports on the use of ultrasound to monitor the fetal development of chinchillas. e aim of our research was to acquire new knowledge of the reproductive physiology of chinchillas based on the gestational similarities with humans and to assess whether this species could potentially be used in obstetric biomedical research.

Materials and Methods
e study was performed with the consent of chinchilla owners and breeders. Ethical clearance (50380-2018) from the Ethical Committee for Animal Experimentation of the University of Naples Federico II was obtained.
Healthy pregnant chinchillas were analyzed at the Interdepartmental Radiology Center of the University of Naples Federico II and at the La Plata breeding center in Cupello (CH), Italy.
We divided the pregnancy into four stages: very early stage (T1) from embryonic day (E) E15 to E30, early stage (T2) from E31 to E46, intermediate stage (T3) from E47 to E70, and advanced stage (T4) from E71 to E115. e pregnancy stage of the monitored females was determined by the date of their last birth. e chinchilla, as well as the guinea pig, begins a new estrous 12-48 h after birth [6]. From this date, we assumed a period of approximately 115 days of gestation; at parturition, both the effective time of conception and the stages of gestation were retrospectively confirmed.
2.1. Ultrasound Imaging. All ultrasound (US) exams were performed by an experienced veterinary ultrasonographer on awake subjects gently restrained to avoid any possible stress to the pregnant females. We monitored 35 pregnant females with a mean age of 3.5 years. Of those, 15 were in their second coupling, 10 in their first, and 10 in their third. e animals used for this study underwent a trichotomy of the pelvic and abdominal area for which no sedation was necessary since the chinchilla is a docile animal. All measurements were made by the same ultrasonographer to ensure consistency during the investigation. In each pregnant female, 2-3 embryos were imaged. e number of fetuses ultrasonographically assessed was compared to the actual number of fetuses at parturition, as communicated to us by the breeders and owners. After obtaining the US results of the entire study, the pregnancy was divided into four stages in order to longitudinally analyze the same morphometric parameter. Each chinchilla was monitored with US in a period ranging from E15 to E115, for a total of four US examinations for each chinchilla. US was performed every two weeks at the beginning of the pregnancy, at T1 and T2, and about every 40 days during T3 and T4, to avoid excessive stress on the animals. e exact day of pregnancy, in which US was performed, was retrospectively determined only after the delivery date, which was communicated to us by the breeders and the owners.
During stage T1, we used the high-frequency ultrasound system, Vevo 770 (Visualsonics, Canada), equipped with a 40 MHz high-resolution linear transducer (focal length 6 mm, depth of penetration 5-15 mm, resolution 30-40 mm axial, and 70-90 mm lateral). During the other stages, we used a US device (MyLab 30, Esaote, Firenze, IT) equipped with a 12 MHz linear probe. A series of measurements were obtained: the longitudinal and transversal diameters of the gestational sacs and the diameter and thickness of the placenta at T1. At T2, T3, and T4, we measured the size of the gestational sacs, the diameter and thickness of the placenta, and the crown-rump length (CRL). Furthermore, we measured the head diameter (HD), the body diameter (BD), the occipital-snout length (OSL), and the heart rate (HR); the femur length (FL) and the stomach were measured from T3, and eye and lens diameter were measured at T3.

Statistical Analysis.
Continuous variables were reported as the mean ± standard deviation. Univariate analysis between each variable and embryonic day was performed using Spearman's rank correlation (Rs). Linear regression analysis was used to assess the relationships between embryonic variables and gestational time. A p value < 0.05 was considered statistically significant. Statistical analyses were performed using MedCalc 18.2.1 software (Ostend, Belgium).

Ultrasonographic Findings.
e ultrasonographic examination was feasible in all chinchilla subjects. Every ultrasonographic examination lasted approximately 10 min. Both techniques, HFUS and US, were able to diagnose pregnancy. However, not all fetal structures could be visualized with both techniques. In the first stage (T1), we found 74 embryos, but at the delivery, this number decreased as the 35 chinchillas effectively delivered 67 cubs. Considering the 35 pregnant females, the pregnancy loss was 11%.
Since it was not possible to obtain the measurements of each parameter at every ultrasonographic examination, we reported the exact number of fetuses, for which each parameter was measured (Table 1).
It was possible to monitor the chinchilla embryos with HFUS from E15 to E60 and with US from E15 to E115. is was due to the increasing fetal dimensions that do not permit accurate measurement of several fetal structures with HFUS. e assessability of the analyzed fetal structures depended on their dimensions and thickness: the placenta was 100%, the HD was 82%, the CRL was 57%, and the femur was 26%. We summarized the assessed structures per gestational stage and the employed ultrasound technique in Table 2.
At T1, from E15 to E30, it was possible to visualize the gestational sac with both HFUS and US. However, the placental diameter and thickness were visualized and measured only with HFUS because the placenta at this stage is not yet discoidal in shape. In the very early stage, it was also possible to identify and measure the gestational sac and to distinguish the primitive node (Figure 1).
At T2, from E30 to E46, the placental diameter and the placental thickness were visualized with US. From E30, it was possible to measure the CRL with both HFUS and US (Figure 2), and from E38, the BD was also measurable. All these parameters were better visualized with HFUS at T2. e OSL, in particular, could be visualized and measured from E40 because, at this gestational age, the head becomes well defined.

Contrast Media & Molecular Imaging
At E30, the heartbeat was detectable for the first time with HFUS, and it was possible to visualize the completed neural tube at E46 (Figure 3). From E55, it was possible to detect the stomach, the eyes, and the lenses (Figure 4).
At E66, the femur became clearly visible and measurable because of increased mineralization (Figure 5(a)). In the advanced pregnancy stage T4 (E75-E115), it was no longer possible to measure the CRL due to fetal dimensions, bone mineralization was pronounced, and the dimensions of eyes and lenses were approximately the same as those in newborns ( Figure 5).
From the beginning of the intermediate stage T3 (E47-E70), it was possible to monitor the chinchillas only with US due to the growing dimensions of the fetal structures. However, we were able to use HFUS to measure the anteroposterior and laterolateral lens diameters until E115. HFUS in this stage of gestation (E60) was also useful to visualize the interventricular septum of the heart and to quantify the HR until the end of pregnancy.

Statistical Analysis.
Summarizing the most important results of the first two stages (T1 and T2), the diameter of the placenta increases from 0.8 cm to 1.6 cm (about 2-fold) and   Figure 6. e longitudinal size of the gestational sacs increased from 1.17 to 2.39 cm (2-fold) until E49. Also, the CRL, HD, and BD gradually increased (CRL: 0.7 to 2.7, about 4-fold; HD: 0.25 to 0.65, about 5-fold; BD: 0.85 to 1.9 cm, about 2fold) till the end of the pregnancy (Figures 2 and 3). e precise measurements of the parameters from the first day of pregnancy to the last are reported in Table 3.
All structural measurements, except for femoral length, were significantly correlated with gestational age. e results of correlation analysis are reported in Table 4.
Based on the period during which it was possible to measure the aforementioned variables simultaneously, they embryos measured were placed in three groups for the regression analysis: group 1 included variables measurable between time T1 to T3 (i.e.,gestational sac diameter and placental diameter and thickness); group 2 included variables measurable between time T2 to T3 (i.e.,placental diameter and thickness, CRL, HD, BD, OSL, and HR); group 3 included variables measurable between time T2 to T4 (i.e.,placental diameter and thickness, BD, OSL, HD, and HR). Using this analysis, three models for predicting gestational age were obtained (Table 5).

Discussion
We examined embryo and fetus development in the C. lanigera using an ultrasonographic technique, and we predicted gestational age with the analysis of ultrasound parameters.
Ultrasonography is a noninvasive method that provides a longitudinal, real time, and detailed morphological evaluation of fetuses in vivo. Furthermore, US is quick and less expensive than other imaging modalities when screening for neonatal defects. We performed a longitudinal, qualitative, and morphometric evaluation of chinchilla embryonic parameters from E15 to the last day of pregnancy using HFUS and US. As in human obstetrics, we analyzed different morphometric parameters useful to predict the gestational age of the fetus (such as the CRL or the HR) and others like the HD, the OSL, or the BD, embryo anatomical size measurements, position of the placenta, and structures, which are useful for predicting anomalies in the unborn child. e ultrasonographic exam was feasible in all subjects, even when subjects were awake. e main difficulty encountered when performing the US exam was related to the anatomy of the chinchilla, which has a highly developed caecum that can sometimes hinder the sight of some structures of interest. Furthermore, the fetuses of chinchillas have a big head that, especially in the last part of pregnancy, when mineralization is advanced, can obstruct the visualization of other anatomical structures of the fetus. Lastly, US is a subjective technique in which the expertise and capabilities of the operator are of paramount importance. In further studies, it would be useful to compare US evaluation performed by two or more operators to assess reliability, sensitivity, and specificity of the method.
Since the animals examined in this study were bred or owned as pets, we had to perform our exams on awake subjects. is circumstance and the abovementioned anatomic characteristics of the chinchilla limited the analysis of some anatomical structures. However, the morphometric data we obtained for the different parameters during pregnancy were consistent and reproducible within the same gestational stage, and no female showed any signs of stress during or following US examinations. All subjects completed their pregnancies although some fetuses were reabsorbed. Studies performed in mice under general anesthesia are more detailed; however, the use of anesthetics often affects the newborn offspring, thus disrupting the possible comparison with human ultrasound findings [19].
Our intent was to protect the well-being of the cubs and fetuses and to demonstrate the noninvasiveness and feasibility of ultrasonographic examination, which can also be performed on awake subjects. In our study, some observation days were missed because we performed only four US examinations per chinchilla (from E15 to E115). is was done to avoid excessive stress on the pregnant females. Also limiting in our study is that, in the 115 days, we performed the first two US analyses in the first 115 days, about every 2 weeks and then about every 40 days thereafter, so we lucked some temporal points of observation. is was due to difficulties reaching the breeding center and also because the chinchillas used for the study were domesticated animals and not experimental animals. us, we adjusted our research to suit the availability of the breeding center and owners.
We used a best-fit regression coefficient to analyze which parameters could be predicted for each gestational age. Based on our results, the placental thickness at T1-T3, the HR, and the CRL at T2-T3, and the HD and HR at T2-T4 could be used to predict the exact gestational age of the chinchilla fetuses. However, some parameters, such as the femoral length, which are used to predict gestational age in human pregnancy, have to be further evaluated to increase casuistry. Furthermore, other parameters will likely be needed to better predict gestational age. For this reason, it is useful to have a precise database of physiological morphometric parameters of embryo and fetal development  during chinchilla pregnancy, even if divided in stages and not in days, in order to diagnose pathological conditions that could occur. Since 1947, the efficiency of animal models for the study of human pathologies of pregnancy has been investigated. Even today an adequate model for the study of human pregnancy has yet to be found [20]. A recent study shows that, in animal perinatology, the use of animal models does not give adequate mathematical correlations with human pregnancy preterm birth [21]. e choice of the right animal model in human obstetrics is challenging because there are several aspects to consider. e duration of gestation, the number of fetuses for each pregnancy, and the placental morphology are all important parameters to evaluate when comparing placentation and fetal morphology between rodents and humans. Other aspects that should not be underestimated are the management costs of the animal, the ease of animal housing, and the legislation (Directive 2010/ 63/EU) that regulates the use of one species instead of another as an animal model for biomedical research. e chinchilla might be more suitable as an animal model for human obstetrics than the guinea pig and other rodents because of the longer duration of pregnancy (105-115 days) [22][23][24]. e female chinchilla's placenta is comparable to the human haemomonochorial placenta, which is formed of only one layer of syncytiotrophoblasts, unlike the other rodents [2]. Furthermore, chinchilla cubs are fully developed, covered with hair, have already developed sight and hearing, and are able to move a few hours after birth. ese characteristics make chinchilla development resemble humans, who have a longer intrauterine development than myomorphic rodents and rabbits. Finally, the reduced number of offspring (1-6, on average 1 or 2 cubs per gestation) facilitates monitoring of the same fetus with US, avoiding errors due to the presence of numerous fetuses in a single pregnancy and allowing a higher spatial resolution and sensitivity using imaging techniques. e discordant data between the number of US examinations and the number of fetuses communicated by the owners and breeders are probably due to fetal reabsorption that often occurs in this species [25].

Conclusions
In conclusion, US lasts only a few minutes and is well tolerated in chinchillas, which have proved to be a suitable animal model for the study of human pregnancy and biomedical research in general [25]. Our research has provided important data, which allow the division of the gestation in four stages. is is particularly relevant in chinchillas since estrus is often silent and does not permit detection of the exact time of mating; with ultrasound, it is possible to establish the exact stage of pregnancy because of the reliability of morphometric measurements. One limitation of the study is the fact that we analyzed each chinchilla only four times during the entire pregnancy, which resulted in some  longitudinal observations being missed. Further studies will be carried out by our group to obtain all morphometric measurements that will cover the entire pregnancy of the chinchilla.

Data Availability
e ultrasound data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.