The fetus is exposed to high plasma concentrations of unbound estrogens and progesterone throughout pregnancy. However, secretory or decidual changes in the fetal uterus occur relatively infrequently before birth, suggesting a variable endometrial progesterone response at the time of birth. Arguably, partial progesterone resistance that persists into adolescent years may compromise the physiological transformation of the spiral arteries and predispose for defective placentation in the case of pregnancy. Decidualization of the endometrial stromal compartment and junctional zone myometrium precedes trophoblast invasion. It represents the first step in the process of spiral artery remodeling needed to establish effective uteroplacental blood flow by midpregnancy. The major obstetric syndromes caused by impaired placental bed spiral artery remodeling are prevalent in teenage pregnancies, including preeclampsia, fetal growth restriction, and spontaneous preterm labor. Preconditioning of the uterus in response to cyclic menstruation during adolescence may be critical to achieve full uterine responsiveness to hormonal cues. Understanding the mechanisms of functional maturation of the uterus during the early reproductive years may yield novel insights into the major obstetric syndromes.
The core of our knowledge on early uterine development dates back to the middle part of the 20th century and is centered on the neonatal uterus. A major autopsy study entitled “Observations on the endometrium and ovary in the newborn” by the Harvard pathologists Ober and Bernstein was published in 1955. In their analysis of 169 neonatal uteri, the authors found inactive or proliferative endometrium in two-thirds of cases, whereas in the remaining infants, there was evidence of secretory activity in the glandular epithelium and even decidualization of the stromal compartment or menstrual changes.
An important feature of the physiology of the newborn uterus is transient uterine bleeding, described in the French literature as crise génitale du nouveau-né and in German literature as neugeborenenblutung. Neonatal uterine bleeding (NUB) is classified as either overt or occult. Although overt NUB is relatively rare, affecting between 3.3% and 5.3% of newborns, biochemical evidence of vaginal bleeding can be found in 25–61% of the neonates.
In an intriguing study, Berić et al registered the incidence of overt uterine bleeding in all female babies born throughout 1979 in Novi Sad, then the former Yugoslavia. The study included 2477 female newborns and reported an incidence of uterine bleeding in 0.8% in premature neonates, 4.4% in term babies, and 9.1% in infants born postterm. The overall frequency of NUB in this study was 3.9%.
On the basis of a critical reinterpretation of the available literature, we have argued that NUB is akin to menstruation. Furthermore, anatomical aspects of the neonate uterus render it likely that retrograde menstruation be more pronounced than antegrade flow, thus potentially contributing to the pathogenesis of early-onset and premenarcheal endometriosis. On the other hand, there is no evidence of progesterone response in the endometrial stromal compartment in most newborns, raising the possibility that the endometrium, particularly in preterm born girls, is still immature in terms of progesterone responsiveness at the onset of adolescence.
Pregnancy in the human is characterized by spontaneous perivascular decidualization and deep placentation with dramatic remodeling of the spiral arteries in the decidua and inner myometrium. Spiral artery remodeling defects are most pronounced in early-onset preeclampsia and fetal growth restriction. Although some studies have reported an association between impaired spiral artery remodeling and other obstetrical disorders, such as preterm labor or preterm premature rupture of the membranes, the severity and prevalence of the lesions are less pronounced when compared with those found in severe preeclampsia.
How the uterus is prepared for the intense tissue remodeling associated with deep hemochorial placentation is unknown. Arguably this process starts in the perinatal period, although a critical interpretation of uterine development in terms of its subsequent function has been overlooked.
In this communication, we attempt to interpret the available literature on hormonal responses in the uterus of the neonate in the context of subsequent pregnancy disorders, especially in adolescent girls. We argue that the neonatal uterus exhibits intrinsic progesterone resistance and postulate that persistence of hormonal refractoriness into the early reproductive years may restrict spiral artery remodeling and deep placentation, leading to an increased risk of major pregnancy disorders in the young adolescent primigravida.
Strategy of search
As detailed previously, we attempted to retrieve every available scientific article on the presence and characteristics of neonatal uterine bleeding as well as the characteristics of the fetal and neonatal endometrium. In addition, we searched the international scientific literature for the following key words: fetal uterus, fetal endometrium, neonatal uterine bleeding, endometriosis, preeclampsia, fetal growth restriction and small for gestational age, and adolescent pregnancy.
Fetal and perinatal uterine maturation
Developmental anatomy of the cervix and corpus uteri
Based on intravaginal injection of rapidly settling liquid silicon, Terruhn found that the fetal urethra, vagina, uterus, and Fallopian tubes have a defined lumen by the 14th week of gestation. However, after the 26th week of gestation, the cervical canal is no longer patent, presumably because of plugging by secretions of the cervical epithelium that lines the canal.
Fluhmann documented a tremendous growth of the cervix during the last 3 months of fetal life, a phenomenon also apparent in the vagina but not shared by the uterine corpus. Consequently, the length of the cervix is approximately 2- to 2.5-fold that of the corpus in the newborn. In fact, the length of the corpus increases proportionally with the fetal weight during the last 3 months of pregnancy, whereas the length of the cervix increases at an accelerated rate. Immediately after birth, there is a marked decrease in the length of the cervix and, much more modestly, in the length of the corpus. The difference is likely explained by the entodermal (urogenital sinus) origin of the cervix with enhanced cellular activity, which is absent in the mesodermal Müllerian duct.
Using real-time ultrasound, Hata et al studied the morphology of the normal uterus and cervix in neonates delivered at term. The uterus was visualized in 41 of the 46 neonates (89.1%), and the cervical volume was found to be approximately 3 times the corporeal volume (3.65 ± 1.36 cm 3 vs 1.18 ± 0.42 cm 3 , respectively; P < .001). Arguably, the anatomical structure of the cervix and uterus at the time of birth could promote the retrograde flux of menstrual effluent, thus obscuring the true incidence of NUB.
Plasma estrogens and progesterone in maternal and fetal circulation
Throughout pregnancy, the fetus is exposed to high concentrations of unbound estrogens and progesterone. The mean concentrations of unbound estrone, estradiol, and estriol measured by radioimmunoassay in the umbilical vein are significantly greater than those found in the maternal circulation at term. Furthermore, the mean concentrations of total and unbound plasma progesterone at term are, respectively, 5- and 7-fold higher in the umbilical vein than in the peripheral maternal circulation ( Table ).
| Variable | Total | Unbound | ||
|---|---|---|---|---|
| n | Measure, ng/mL | % | Measure, ng/mL | |
| Luteal phase | 5 | 17 ± 3 | 4.83 ± 0.1 | 0.82 ± 0.14 |
| End of first trimester | 10 | 41 ± 5 | 4.93 ± 0.1 | 2.03 ± 0.24 |
| Term pregnancy | 10 | 160 ± 10 | 5.02 ± 0.1 | 8.03 ± 0.50 |
| Umbilical vein at term | 10 | 788 ± 30 | 7.01 ± 0.2 | 55.20 ± 2.10 |
Hill et al suggested that placental distribution of enzymes involved in steroidogenesis and metabolism account for the higher circulating progesterone levels in the fetus compared with the mother. For example, type 2 17β-hydroxysteroid dehydrogenase, which oxidizes estradiol to estrone and 20α-dihydroprogesterone to progesterone, is highly expressed in placental endothelial cells lining fetal vessels.
Conversely, syncytium, which is directly in contact with maternal blood, produces higher amounts of estradiol and 20α-dihydroprogesterone because of the expression of other steroid dehydrogenases, including 17β-hydroxysteroid dehydrogenase types 1, 5, and 7. Circulating progesterone levels in neonates drop rapidly after birth. Ferris and Green found that the excretion of pregnanediol, the major inactive progesterone metabolite, in neonatal urine disappeared after the fifth day of life.
Fetal and perinatal uterine maturation
Developmental anatomy of the cervix and corpus uteri
Based on intravaginal injection of rapidly settling liquid silicon, Terruhn found that the fetal urethra, vagina, uterus, and Fallopian tubes have a defined lumen by the 14th week of gestation. However, after the 26th week of gestation, the cervical canal is no longer patent, presumably because of plugging by secretions of the cervical epithelium that lines the canal.
Fluhmann documented a tremendous growth of the cervix during the last 3 months of fetal life, a phenomenon also apparent in the vagina but not shared by the uterine corpus. Consequently, the length of the cervix is approximately 2- to 2.5-fold that of the corpus in the newborn. In fact, the length of the corpus increases proportionally with the fetal weight during the last 3 months of pregnancy, whereas the length of the cervix increases at an accelerated rate. Immediately after birth, there is a marked decrease in the length of the cervix and, much more modestly, in the length of the corpus. The difference is likely explained by the entodermal (urogenital sinus) origin of the cervix with enhanced cellular activity, which is absent in the mesodermal Müllerian duct.
Using real-time ultrasound, Hata et al studied the morphology of the normal uterus and cervix in neonates delivered at term. The uterus was visualized in 41 of the 46 neonates (89.1%), and the cervical volume was found to be approximately 3 times the corporeal volume (3.65 ± 1.36 cm 3 vs 1.18 ± 0.42 cm 3 , respectively; P < .001). Arguably, the anatomical structure of the cervix and uterus at the time of birth could promote the retrograde flux of menstrual effluent, thus obscuring the true incidence of NUB.
Plasma estrogens and progesterone in maternal and fetal circulation
Throughout pregnancy, the fetus is exposed to high concentrations of unbound estrogens and progesterone. The mean concentrations of unbound estrone, estradiol, and estriol measured by radioimmunoassay in the umbilical vein are significantly greater than those found in the maternal circulation at term. Furthermore, the mean concentrations of total and unbound plasma progesterone at term are, respectively, 5- and 7-fold higher in the umbilical vein than in the peripheral maternal circulation ( Table ).
| Variable | Total | Unbound | ||
|---|---|---|---|---|
| n | Measure, ng/mL | % | Measure, ng/mL | |
| Luteal phase | 5 | 17 ± 3 | 4.83 ± 0.1 | 0.82 ± 0.14 |
| End of first trimester | 10 | 41 ± 5 | 4.93 ± 0.1 | 2.03 ± 0.24 |
| Term pregnancy | 10 | 160 ± 10 | 5.02 ± 0.1 | 8.03 ± 0.50 |
| Umbilical vein at term | 10 | 788 ± 30 | 7.01 ± 0.2 | 55.20 ± 2.10 |
Hill et al suggested that placental distribution of enzymes involved in steroidogenesis and metabolism account for the higher circulating progesterone levels in the fetus compared with the mother. For example, type 2 17β-hydroxysteroid dehydrogenase, which oxidizes estradiol to estrone and 20α-dihydroprogesterone to progesterone, is highly expressed in placental endothelial cells lining fetal vessels.
Conversely, syncytium, which is directly in contact with maternal blood, produces higher amounts of estradiol and 20α-dihydroprogesterone because of the expression of other steroid dehydrogenases, including 17β-hydroxysteroid dehydrogenase types 1, 5, and 7. Circulating progesterone levels in neonates drop rapidly after birth. Ferris and Green found that the excretion of pregnanediol, the major inactive progesterone metabolite, in neonatal urine disappeared after the fifth day of life.
The hormonal responsiveness of the endometrium in the fetus, newborn, and infant
As early as 1934, it was recognized that proliferative activity in the fetal endometrium during the third trimester of pregnancy could not reflect ovarian activity but must be caused by stimulation from placental hormones. As aforementioned, Ober and Bernstein described in detail the changes in the endometrium and ovaries in a series of 169 autopsies, which included 68 infants born at term (38-42 weeks’ gestation). Among these infants, 57 died within the first 3 days of life and the remaining 11 died before 14 days. The endometrium was classified as proliferative in 116 (68%), secretory in 45 (27%), and progestational in 8 (5%) cases. It should be noted that 60 of 116 cases classified as proliferative showed a basal phenotype (resting phase), and 37 of 45 classified as secretory showed only subnuclear vacuolation.
The group with progestational changes consisted of the 8 cases with histological evidence of decidualization or menstrual tissue breakdown. Importantly, evidence of ovarian follicular development to the antral stage was observed only in a single case, and none of the ovaries had a Graafian follicle or corpus luteum. From these observations, the authors correctly concluded that the secretory and progestational changes in the endometrium could not be accounted for by fetal or neonatal ovarian activity.
The aforementioned study by Huber et al involved 82 uteri of fetuses, infants, and children. This study showed that glandular formation is absent from the fetal endometrium before the 20th week of gestation. Gradually proliferation becomes apparent in both the epithelial and stromal compartments, and secretory changes in the glands are increasingly found after 33 weeks, with activity peaking at birth. Immediately after birth, tissue involution and regression are associated with desquamation of the epithelial layer and discrete interstitial bleeding. Immediately after birth regressional changes of the endometrium set in, and after the eighth postnatal day, a transitional type of endometrium without glandular activity appears. The epithelium is comparatively low, and no evidence of glycogen is found.
After the transitional phase, the endometrium consists of a very thin, nearly flat epithelial layer and remains in an atrophic, inactive state. This seems a characteristic pattern during infancy and early childhood until the child shows the first visible signs of ovarian activity at about the age of 7 or 8 years. Huber et al concluded that we have to assume that functional changes in the fetal and infantile endometrium take place independently of any ovarian activity of the child itself and are brought about by the effect of maternal (placental) hormones. At the same time, the fetal and infantile tissues have not reached their full maturity and can therefore not react like that of a mature woman. These observations were confirmed by Hiersche and Meinen and Pryse-Davies and Dewhurst.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
