Introduction

Nutritional imbalances in developed societies are traditionally linked to excessive food intake, but currently the incidence of eating disorders and voluntary caloric restriction for aesthetical reasons are increasing.Reference Brett, Ferraro, Yockell-Lelievre, Gruslin and Adamo ¹ These circumstances, besides the traditional high prevalence of maternal undernutrition in developing countries,Reference Ramakrishnan, Imhoff-Kunsch and Martorell ² make research on the effects of undernutrition in the reproductive outputs of the exposed individuals necessary. Maternal undernutrition is considered a major cause of intrauterine growth restriction (IUGR).Reference Murthy, Agarwal and Khanna ³ Other maternal causes of IUGR described in the bibliography are hypertensive disorders of pregnancy, diabetic vasculopathy, chronic renal disease, collagen vascular disease and thrombophilia.Reference Chia and Huang ⁴

IUGR is defined as the failure of the fetus to reach its genetically established growth rate. It is annually related to 800,000 neonatal deaths worldwideReference Bhutta, Das and Rizvi ⁵ and can be classified as ‘symmetrical’ or ‘asymmetrical’, with different degrees of severity. Symmetrical IUGR is characterized by a uniform reduction in size of the fetus and its organs and is associated with genetic and infectious factors. Asymmetrical IUGR, however, is characterized by a reduction in size of some organs, while other organs remain normal. It is mainly related with insufficient nutritional delivery to the fetus by maternal undernutrition or placental insufficiency. Predisposing factors that alter the fetal growth trajectory are a detrimental maternal nutritional status and inadequate diet during pregnancy.Reference Cetin, Mando and Calabrese ^{6 ,} Reference Nardozza, Araujo Junior and Barbosa ⁷ Moreover, maternal undernutrition may affect placental growth and function, thus decreasing nutrient availability to the fetus and as a result affecting neonatal size.Reference Ergaz, Avgil and Ornoy ^{8 ,} Reference Redmer, Wallace and Reynolds ⁹

Furthermore, a clear relationship between IUGR infants and metabolic, cardiovascular and neurological pathologies in adulthood has been documented.Reference Gonzalez-Bulnes, Ovilo and Lopez-Bote ^{10 ,} Reference Ross and Desai ¹¹ The association between inadequate intrauterine nutrition with the occurrence of IUGR and disease risk at postnatal life gave basis to the concept of the Developmental Origins of Health and Disease, formerly known as fetal programming or the Barker hypothesis.Reference Barker ¹² This concept predicts that specific situations experienced in early life may increase the risks of suffering late-onset diseases.Reference Saffery ¹³ However, it is unclear how the fetus responds to this prenatal environment to increase its chances of survival.

There is a serious necessity to address the potential long-term effects of undernutrition on the offspring via preventative procedures and focused treatments. Seeing as experimental studies on humans are limited due to ethical constraints, translational research based on the use of animal models have become imperative. This is verified by the fact that around three quarters of IUGR experiments have been performed in rats and mice.Reference Schroder ¹⁴ However, in the recent years, the rabbit has been established as one of the most amenable animal model for reproductive studies.Reference Puschel, Daniel and Bitzer ¹⁵ In brief, ovulation in rabbits is induced by coitus, resulting in a precise pregnancy and embryonic age. Implantation starts on day 6–7, completing the chorioallantoid placentation around day 9.Reference Fischer, Chavatte-Palmer, Viebahn, Santos and Duranthon ¹⁶ After 31 days of pregnancy, females are able to deliver more than eight fetuses, hence enabling a high number of samples per female when compared with other mammal models, reinforcing the 3 R concept on Animal Experimentation.Reference Russell, Burch and Hume ¹⁷ The placental structure of the rabbit is hemodichorial (two cellular layers of chorion between the maternal and the fetal blood) and so it is closer to the human structure than that of other laboratory animals, like rodents (hemotrichorial). Moreover, the similar development of the brain in rabbits and humans (white matter maturation process starts during the intrauterine periodReference Derrick, Drobyshevsky and Ji ¹⁸ ) favours the use of rabbits as an useful model to study perinatal processes of brain development.Reference van Vliet, Eixarch and Illa ¹⁹ Furthermore, the fetal size in rabbits, significantly larger than in rodents, allows accurate fetal measurements performed with common veterinary ultrasound equipments.Reference Chavatte-Palmer, Laigre and Simonoff ²⁰

Most methods used to induce IUGR are based on nutrient restriction or surgical modification.Reference Huizinga, Engelbregt and Rekers-Mombarg ^{21 ,} Reference Eixarch, Hernandez-Andrade and Crispi ²² Some authors propose the ligature of uteroplacental vessels as the most precise method, since restrict both nutrient and oxygen supply; however, that procedure is performed on the last few days of pregnancy and is associated with high mortality.Reference Eixarch, Hernandez-Andrade and Crispi ²² As an alternative, underfeeding regimes are an easier way to induce IUGR in fetuses and are also useful for understanding the pathology of IUGR and its connection to inadequate dietary habits. The availability of a reliable animal model to study fetoplacental hemodynamics in IUGR pregnancies would be highly beneficial in order to develop diagnostic, preventive and therapeutic strategies in humans and animals. Our hypothesis is that the rabbit is a suitable model for the study of the IUGR syndrome caused by maternal undernutrition.

Therefore, the present work aimed to evaluate the suitability of the rabbit as a maternal undernutrition IUGR model. Suitability will be determined by ultrasonography and direct measurements of effects of such perturbations on the development of the offspring and characterizing fetal hemodynamic features at 70% pregnancy.

Methods

Conceptuses study

Ultrasound evaluations were performed in all rabbits on the 21st day of pregnancy (≈70% of the total pregnancy length). At this stage, fetal viability is critical since a significant change in the uteroplacental blood flow takes place.Reference Hafez and Tsutsumi ²⁵ This is, when the irrigation of the placenta becomes more abundant than that of the uterus, in order to cover the needs of the developing conceptuses. Moreover, at this moment, organogenesis is assumed to be achieved.Reference Eixarch, Hernandez-Andrade and Crispi ²² Rabbits were shaved in the abdominal area and manually restrained in dorsal recumbence, without anesthesia to avoid any effect on the heart rate or blood flow during the observations. All the females had been handled and restrained before the experimental phase, which diminished animal stress. The same blind operator scanned all rabbits using a Vivid-I ultrasound machine (General Electric, Fairfield, CT, USA) equipped with a multifrequency (8–12 MHz) lineal array probe. Scans were recorded using the ‘cine-loop’ option; this allowed animals to be released quickly, minimizing the restraining time. A complete scan exploration did not last >20 min/female.

To visualize the uterine horns and fetuses on the transverse, frontal and sagittal planes, rabbits were scanned by placing the transducer on one flank of the rabbits and moving it to the opposite flank. Four fetuses were selected at random in each female to minimize individual effects. Hence, 16 fetuses were analyzed in the control group and 20 in the underfed group, but the four fetuses measured per mother were averaged to result in one data point. Measurements were obtained with built-in electronic calipers on the cine-loop once the complete examination was recorded. Since the size of the fetus was too large for viewing the entire body-length at this pregnancy stage, measurements included the thoracic diameter, the occipito-nasal length and the biparietal diameter (Fig. 1).

Fig. 1

Ultrasound image of the head (a) and the thorax (b) of a rabbit fetus at day 21 of pregnancy. ONL, occipito-nasal length; BPD, biparietal diameter; TD, thoracic diameter.

Immediately after birth, all the kits were classified as newborns or stillborns. Only newborn were measured. Using a slide caliper, we also determined biparietal diameter (from one parietal eminence to the other), thoracic diameter (adjusted between underarms) and crown-rump length (maximum distance from crown to tail basis; see Fig. 2).

Fig. 2

Measurements obtained at birth: (a) biparietal diameter, (b) transversal thoracic diameter and (c) crown-rump length.

Doppler evaluation of fetal hemodynamics

At the 21 day of pregnancy, blood flow parameters from the umbilical cord arteries (UCAs; Fig. 3a) and from the middle cerebral artery (MCA; Fig. 3b) were determined in the same fetus in which the body size was ultrasonically evaluated. Briefly, after identifying the vessels with color Doppler (UCAs were found at the free-floating umbilical cord proximal to the placental insertion whereas MCA were located after the Circle of Willis identification), the sample pulsed Doppler gate was placed over the vessels. Then, the waveforms of three consecutive cardiac cycles in each vessel were recorded, disregarding views with insonation angles between 20–50°. Measurements were obtained once the entire examination was recorded and included resistance index (RI), pulsatility index (PI), systolic peak velocity (SPV), end diastolic velocity (EDV) and time-averaged mean velocity (MV), measured at UCA and MCA (Fig. 3). Finally, the cerebroplacental ratios (i.e. the ratios between MCA and UCA values) for RI, PI, SPV, EDV and MV were also calculated.

Fig. 3

Doppler image of UCA (a); and MCA (b) of a rabbit fetus at day 21 of pregnancy. UCA, umbilical cord artery; MCA, middle cerebral artery.

Results

Morphometry of fetuses and offspring

Table 1 depicts the morphometric data of fetuses and kits. On day 21 of pregnancy maternal underfeeding was related to a reduction in fetal size, specifically in a shorter occipito-nasal length (P<0.05), while biparietal and thoracic diameters were similar in control and underfed fetuses. At birth, a total of 43 newborns from control group and 53 in the underfed group were included in the study. Kits from the underfed group showed the lowest size (in terms of crown-rump length, biparietal and thoracic diameters; P<0.05) and a decreased weight when compared with that of the control group (P<0.05). Time of delivery was not affected by the feed restriction (all females delivered on day 31 of pregnancy). No difference was obtained between groups in the mortality rate at birth (1 out of 44 and 3 out of 56 in control and underfed groups, respectively; P>0.05).

Table 1

Offspring development of rabbit does in the control group and underfed group at day 21 of pregnancy and at birth

^a,bDifferent superscripts within a row indicate significant differences among groups (P<0.05).

Doppler evaluation of fetoplacental hemodynamics

At the UCAs, significantly higher SPVs and time-averaged mean velocities (P=0.02 and P=0.01, respectively) were observed in the underfed fetuses when compared with the control group (Table 2). Whereas, fetuses from the underfeed group showed a decreased EDV (P=0.05). Indexes of pulsatility (P=0.07) and resistance (P=0.06) tended to be higher in the underfed fetuses, without being statistically significant.

Table 2

Fetal hemodynamic parameters obtained from rabbit does in the control group and underfed group at day 21 of pregnancy

UCA, umbilical cord artery; MCA, middle cerebral artery.

^a,bDifferent superscripts within a row indicate significant differences among groups (P<0.05).

There were no significant differences in the hemodynamic parameters at the MCA between both groups at any analyzed variable. Finally, the cerebroplacental ratio for MV was significantly lower in the underfed fetuses (P=0.02).

Discussion

The results of the present study show that underfeeding pregnant rabbits, at a level that does not affect maternal body weight, can be considered a useful model for IUGR studies. The level of restriction affected fetal development by reducing the prenatal growth of the offspring and therefore their size and weight at birth, without being associated to a higher mortality rate. Our data also indicate that IUGR fetuses develop compensative ‘brain-sparing effect’ for preserving adequate blood supply to the brain, with this effect being compromised in fetuses with severe IUGR. Thus, these data are similar to that previously reported in humans at 70% pregnancy.Reference Vik, Vatten, Jacobsen and Bakketeig ²⁶ Our study also confirms the adequacy of the use of ultrasound scanning for the assessment of fetal characteristics, reinforcing previous studies in rabbits.Reference Chavatte-Palmer, Laigre and Simonoff ²⁰ This finding gives a way for future translational studies in this model since, in humans, ultrasound technology is widely used in prenatal care to estimate gestational age, to assess fetal growth and to determine physical abnormalities.Reference Smarr, Vadillo-Ortega, Castillo-Castrejon and O’Neill ²⁷ Moreover, fetal biometry during pregnancy is a confirmed method to predict probable later adverse perinatal outcomes.Reference Smith, Smith, McNay and Fleming ^{28 ,} Reference Pedersen, Wojdemann, Scheike and Tabor ²⁹

In the current study, ultrasonography showed fetal morphometric changes that were induced by maternal nutritional restriction at 70% pregnancy. Underfed fetuses showed a smaller occipito-nasal length than fetuses in the control group, without any other major changes. The occipito-nasal length is a more accurate indirect measure for fetal development than other parameters since the skull remains in a good examinable positionReference Kelly and Newnham ³⁰ and the hyperechogenic limit of the bones enables an easier measurement. From the morphometric changes, the limited impact of maternal restriction on embryo growth, together with the fact that rabbit and other species including human fetuses show a faster body growth toward the end of the pregnancy can be explained.Reference Beaudoin, Barbet and Bargy ³¹ In the early pregnancy, embryo–fetal needs are scarce, with maternal metabolism being anabolic in this periodReference Vernon, Clegg and Flint ³² to ensure enough energetic reserves. Moreover, the does in our study were at optimal body weight at the beginning of the pregnancy and could have partially satisfied the low fetal growth requirements during the two first thirds of pregnancy without needing to mobilize their own body reserves.

The last third of pregnancy is characterized by maximum fetal growth rate and, therefore, due to the need for a higher amount of nutrients; maternal metabolism becomes catabolic,Reference Symonds and Clarke ³³ and nutritional restriction during this phase of pregnancy normally has a severe impact on fetal growth, which caused the differences found at birth. In our study, feed intake was not significantly altered during the pregnancy in the control group, neither the maternal weight; this was consistent with other studies in rabbits with different levels of restriction showing no differences in maternal weight at the end of the trials.Reference Nafeaa, Ahmed and Fat Hallah ^{34 ,} Reference Menchetti, Brecchia, Cardinali, Polisca and Boiti ³⁵ Hence, we can hypothesize that despite the fetuses requiring a higher level of resources near pregnancy term, the mother may have invested her limited resources for her own metabolism, subsequent lactation and in her future reproduction instead of her litter growth or placental development.Reference Fowden and Moore ³⁶ There was not a higher mortality rate among restricted newborns, demonstrating that the maternal ‘decision’ of not dedicating more energy to their litter was right, in the view of survival rate. Probably a more severe restriction would induce changes in weight of the mothers, but further research is necessary to explore this issue and to determine if it would be a limiting factor in the translational value of the rabbit model.

The results obtained after the evaluation of the fetal blood flow by Doppler ultrasound in the present study reinforce the use of rabbits as a model for IUGR studies. Fetuses showed a high flow velocity at the umbilical arteries, resembling values in humans at the second trimester of pregnancy.Reference Ichizuka, Ando and Ichihara ^{37 –} Reference Gudmundsson and Marsal ⁴⁰ We focused our research on the umbilical artery since previous studies have shown that the measurement of insonation at umbilical arteries is easy to obtain and offers reliable data on blood exchange with the placenta than insonation of other vessels.Reference Acharya, Wilsgaard, Berntsen, Maltau and Kiserud ⁴¹ Hence, a high blood flow induces high PI and RI values, like in humans.Reference Polisca, Scotti, Orlandi, Brecchia and Boiti ³⁹

In IUGR pregnancies, the increase in the resistance at the small arteries and arterioles of the villi, decreases the diastolic flow, which raises the MV at the UCAs.Reference Gudmundsson and Marsal ^{40 –} Reference Bamfo and Odibo ⁴² Concomitantly, in the present study we observed that fetuses exposed to maternal underfeeding evidenced a lower EDV, higher systolic peak and time-averaged mean velocities, with a trend for higher indexes of pulsatility and resistance when compared with those of the control group. These alterations, usually detected in IUGR, are related to a lower supply of oxygen and nutrients to the fetus as a consequence of a placental ischemia.Reference Bamfo and Odibo ⁴² Placental ischemia is considered the ultimate cause for IUGR and is associated with high perinatal morbility and mortality.Reference Roberts ⁴³ It could be hypothesized that fetuses in the underfed group were in the first stages of IUGR and demanded more blood flow to try to increase blood exchange rate,Reference Roberts ⁴³ since the nutrient content and oxygen supply were reduced. Consequently, the differences found in the current study could indicate a lower blood perfusion in fetuses from underfed mothers caused by the nutritional restriction.

Fetuses at IUGR risk, under circumstances such as poor nutrition or inadequate placental supply, develop a redistribution of blood circulation to increase the blood supply to the fetal brain. This phenomenon is known as ‘the compensatory brain-sparing effect’Reference Makhseed, Jirous, Ahmed and Viswanathan ⁴⁴ and its purpose is to assure adequate brain development and function, since a failure of this causes a broad spectrum of adverse neurological outcomes, thus compromising postnatal vitality and survival of the neonate.Reference Peeling and Smart ^{45 ,} Reference Salihagic-Kadic, Medic and Jugovic ⁴⁶ In human medicine, the occurrence of brain sparing consecutive to IUGR has been identified by assessing Doppler ratios.Reference Makhseed, Jirous, Ahmed and Viswanathan ⁴⁴ The cerebroplacental ratios quantify the redistribution of the cardiac output by dividing Doppler index from cerebral and fetoplacental vascularity. In this scenario, umbilical arteries represent the right ventricular afterload, whereas, the MCA reflects the left ventricular afterload.Reference Baschat and Gembruch ⁴⁷

The brain-sparing effect was manifest in the underfed fetuses of the current study (significantly lower cerebroplacental ratio and a trend for higher PI); this is evidence of a redistribution of the cardiac output to the cerebral circulation (Table 2). Our study also identified the weakness and even the failure of this compensatory mechanism in those fetuses suffering a more severe IUGR, those with smaller biparietal and thoracic diameters, as evidenced by the Pearson correlations.

Our results indicate the reliability and usefulness of the rabbit model and of the Doppler procedure for the study of IUGR. However, we cannot leave aside some limitations of this study. The first limitation would be a systematic assessment of the changes in IUGRs throughout the entire pregnancy. However, we preferred limited scanning sessions since repeated examinations of the pregnant does could have caused stress-induced abortion, even in trained animals, especially during the last third of pregnancy, when the decidua, the maternal side of the placenta is thinner and less attached.Reference Mossman ⁴⁸ A second limitation would be the small sample size used in this study, which could have reduced the power of the study and masked significant differences. Finally, more information about the UCA diameters could have been relevant, there can be errors when measuring vessel diameter, in particular for vessels with a high pulsating blood flow.Reference Hecher, Campbell, Doyle, Harrington and Nicolaides ⁴⁹ In our case those measurements were more complicated since no anesthetic drugs were used, since they could interference with maternal uteroplacental blood flow.

In conclusion, our data indicate that rabbit fetuses exposed to maternal underfeeding are prone to show a disrupted developmental trajectory and, due to this restriction, to undergo IUGR. IUGR fetuses were able to develop compensative brain-sparing effects in a similar way to that previously described for humans. Thus, this study supports the rabbit as a reliable translational model for studies on IUGR associated with feed restriction and associated diseases. The use of the Doppler technique to explore umbilical arteries, as well as the estimation of the cerebroplacental ratios led to the early detection of fetal blood flow alterations and should be taken in consideration as a primary diagnosis tool. The information herein could be of interest to researchers studying the vascular adaptation and/or deterioration processes and help in exploring possible therapies in experimental models of IUGR.