Physiological acromegaly of pregnancy

Pregnancy resets many endocrine systems and may cause endocrine disturbance. For example, when the insulin resistance of pregnancy is not met by an adequate insulin secretory response, gestational diabetes ensues; also, gestational hyperthyroidism develops as a result of an abnormally high output of human chorionic gonadotropin – for example, in multiple or molar pregnancy – or a genetic hypersensitivity to human chorionic gonadotropin. While the growth hormone (GH)–IGF axis is often overlooked in reviews on the endocrinology of pregnancy, evidence has been accumulating since the late 1980s that pregnancy is a state of mild physiological acromegaly.

Many pregnant women complain of swollen limbs and face. It has become clear that the traditional explanations of leg varicose veins and hypoproteinemia are insufficient. The facial edema and coarsening of facial features that gravidas may experience is reminiscent of acromegaly. The same is true for the edema of the forearms and hands, which can lead to morning stiffness of the fingers and mild-to-severe carpal tunnel syndrome with unremitting paresthesia. Textbooks mention the ‘meaty’ hands of acromegalics, and clinical Phase I or II trials examining the use of recombinant GH or IGF-I in adults consistently report forearm edema and carpal tunnel symptoms as side effects.

Pregnancy is accompanied by hypersecretion of GH and IGF-I, because the placenta produces a variant of GH known as placental growth hormone (PGH). PGH is encoded by the GH-V (for variant) gene, unlike pituitary GH, which is encoded by the GH-N (normal) gene; yet, both genes are part of the somatogenic cluster of genes located on chromosome 17q22–24. The molecular weight of PGH (22.3 kDa) is slightly higher than that of pituitary GH (20 kDa); there is also a minor glycosylated PGH isoform (25 kDa). PGH contains 191 amino acids, as does GH, but 13 amino acids are different [1].

Placental GH is expressed exclusively in the multinucleate syncytiotrophoblast. Hence, PGH behaves as a typical placental product: its secretion is tonic rather than pulsatile as for pituitary GH, it is detectable in maternal plasma as early as 5 weeks gestational age (GA), rises exponentially until a GA of 37–38 weeks and is cleared very rapidly after delivery (65–89% within 30 min) [2,3]. Of note, PGH is not detectable in the fetal circulation. Similar to GH, PGH is partly bound to the GH-binding protein (BP) in the maternal circulation. Yet, individual PGH concentrations vary greatly, with peak levels ranging between 4.6 and 69.2 ng/ml in one study [2]. The bioavailable GH-BP-unbound fraction and free-PGH concentrations are even more variable [3]. Overweight gravidas typically have lower circulating PGH levels [2–4], although the underlying mechanism is uncertain.

Placental GH is a genuine somatogenic hormone. It binds with the same potency to the hepatic GH receptor as pituitary GH does. Preclinical models show a comparable efficacy of GH and PGH in stimulating growth. Transgenic mice have been generated that express PGH levels relevant to human pregnancy; these animals weigh 85% more than normal mice do, primarily by an increase in fat-free mass. They also display insulin resistance and hyperinsulinemia, possibly by perturbed postreceptor insulin signaling [5].

Together with insulin and nutritional factors, GH is a pivotal regulator of IGF-I secretion in nonpregnant adults. Longitudinal studies demonstrate that circulating IGF-I is not augmented from early pregnancy onward; in fact, a decline of approximately 30% was observed in some studies [6,7]. The estrogen surge may be reponsible for this early reduction, but this hypothesis needs to be verified. However, circulating IGF-I increases unequivocally in the second half of pregnancy, reaching maximum levels at 37–38 weeks. The difference between trough and peak values in pregnancy is two- to threefold [2,6,7]. Circulating IGF-I drops by approximately 40% in the first 48 h after delivery.

Importantly, many studies have shown significant correlations between circulating PGH and IGF-I at any stage of pregnancy, as well as between the calculated slopes of PGH and IGF-I [2,4,8]. However, the correlation appears strongest at the end of pregnancy or at delivery. These findings have led to the tenet that PGH is the principal regulator of IGF-I secretion during pregnancy.

The IGF-I biovailability might also be increased during pregnancy. In one study, the free-IGF-I fraction was 1.5–2.4% during pregnancy compared with 0.9% in nonpregnant women [9]. The binding of IGF-I to the IGF BPs determines its bioavailability. Six IGF BPs have been characterized, with IGF BP-3 and IGF BP-1 the most important IGF BPs. IGF BP-3 binds 80–90% of circulating IGF-I in a trimeric complex that also includes the acid-labile subunit; IGF-I, IGF BP-3 and acid-labile subunit are all GH-regulated. However, IGF BP-3 (as well as IGF BP-2, -4 and -5) are proteolyzed by a pregnancy-induced circulating endoprotease, rendering these IGF BPs undetectable by western ligand or immunoblots. The IGF BP-3 protease is discernible as early as 6 weeks GA and is trophoblast derived, possibly corresponding to a disintegrin–metalloproteinase type enzyme [10]. IGF BP-3 proteolysis during pregnancy is expected to result in increased bioavailable IGF-I, although full consensus on this issue is lacking.

Consequent to the surge in PGH and IGF-I, pituitary GH secretion is suppressed with very low or undetectable concentrations in the second half of pregnancy [8]. Suppression of GH during pregnancy also appears to occur in acromegalic women, and their IGF-I concentrations are within the normal (non)pregnant range [11]. PGH and estrogens may be involved in the spontaneous amelioration of acromegaly; however, the present data are too scant to be conclusive.

Maternal circulating PGH and IGF-I have been related to ultrasound indices of fetal growth and birth weight. There is a definite correlation close to or at delivery, whereas the clinical value of maternal PGH or IGF-I sampled at some point in the second or in the early third trimester – or calculated from individual slopes when multiple data are available – in the prediction of birth weight remains uncertain [2,4]. PGH and IGF-I reflect placental size [2,6], which may explain the correlation with fetal growth indices. Unsurprisingly, both analytes are lower during the third trimester in pregnancies complicated by severe preeclampsia or fetal growth restriction [12].

It is believed that IGF-II is a less important somatogenic hormone than IGF-I during postnatal life. Nonetheless, there appears to be a modest and reversible rise (20–25%) in circulating IGF-II during pregnancy [7], the role of which needs further study. IGF-II is abundantly expressed in both the placenta and the fetus, and is believed to stimulate fetoplacental growth by a local rather than an endocrine mode of action. Indeed, fetal mice with ablated placental IGF-II expression show smaller placentas and develop growth restriction, stressing the somatogenic role of IGF-II in prenatal life [13].

Although IGF BP-1 binds only approximately 2% of IGF-I, it is metabolically important because IGF BP-1 secretion is upregulated by fasting and hypoinsulinemia (thus explaining the diurnal variation) and also by hypoxemia. Apart from the highly phosphorylated IGF BP-1 produced in the liver, the decidua is an abundant source of variably phosphorylated IGF BP-1 isoforms. Decidual IGF BP-1 diffuses into the amniotic fluid [14], causing a steep rise in IGF BP-1 in this compartment from 10 weeks GA onward; concentrations are attained that are 100–1000-times higher than in maternal plasma. Decidual IGF BP-1 affects placental growth and maturation [14], possibly through interaction with trophoblastic IGF-II. In the maternal circulation, IGF BP-1 is not proteolyzed and increases two- to threefold in the course of pregnancy [6]; there are many reports of lower concentrations in gravidas who later develop preeclampsia, possibly reflecting their relative hyperinsulinemia and/or impaired decidual function [15]. Yet, the value of maternal plasma or amniotic fluid IGF BP-1 in the second or early third trimester to predict growth restriction at birth is inconsistent among studies, similarly to maternal plasma PGH and IGF-I [4,16].

The physiological hypersecretion of PGH, IGF-I, IGF-II and IGF BP-1 should serve a purpose. Studies in animal models and in vitro indicate that PGH stimulates trophoblast invasion [17], IGF-II boosts placental growth and diffusion capacity, and IGF-I and IGF-II enhance placental maturation and nutrient (amino acids and glucose) uptake [18]; hence, they optimize nutrient availability for the fetus, fetal growth and survival. The placental effects appear to be more important than the maternal-metabolic effects engendered by PGH and the IGFs, such as a possible hyperglycemic effect of PGH [5] or a possible lipolytic effect of PGH and IGF-I [18]. However, the interesting data of these preclinical models have yet to be translated into strong evidence that applies to human pregnancy. Further studies are eagerly awaited to disclose the effects of maternal and fetal somatogenic hormones at the deciduoplacental interface.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.