Which condition is a fetus at risk for if the mother has poor glycemic control?

Hyperbilirubinemia

IDMs also appear to be at greater risk for hyperbilirubinemia than infants born to nondiabetic mothers. In studies that defined hyperbilirubinemia as a serum bilirubin level greater than 12 mgdL or any bilirubin level requiring phototherapy, the prevalence of hyperbilirubinemia in IDMs was 25% (Cordero et al., 1998); other series report a hyperbilirubinemia prevalence of 10%–13% in IDMs. This increased risk might be attributable to polycythemia (larger source of bilirubin to be conjugated by the liver prior to excretion), ineffective erythropoiesis with an increased red blood cell turnover, as well as to immaturity of hepatic bilirubin conjugation and excretion. Macrosomic IDMs appear to have the greatest risk of hyperbilirubinemia; this probably reflects the role of poor maternal glycemic control during pregnancy as IDMs of these women are most likely to be macrosomic and polycythemic.

Abstract

Maternal diabetes, either preexisting or gestational in onset, is well known to contribute to perinatal morbidity. In all cases of maternal diabetes, pancreatic β-cell dysfunction leads to an inability to produce adequate insulin to maintain appropriate glucose levels; glucose freely crosses the placenta and gains access to the fetal circulation leading to hyperglycemia and other metabolic abnormalities in the developing fetus. This can lead to physical developmental problems in the form of congenital anomalies (particularly cardiac anomalies) and macrosomia. Further, metabolic derangements – including severe hypoglycemia, hypocalcemia, and hypomagnesemia – in the neonatal period are frequently present and more severe in the offspring of diabetic mothers.

Growth Dynamics

IDMs with macrosomia follow a unique pattern of in utero growth compared with fetuses in euglycemic pregnancies. During the first and second trimesters, differences in size between fetuses born to diabetic and nondiabetic mothers are usually undetectable with ultrasound measurements. After 24 weeks, however, the growth velocity of the IDM fetus’ abdominal circumference typically begins to rise above normal (Ogata et al, 1980). Reece et al (1990) demonstrated that the IDM fetus has normal head growth, despite marked degrees of hyperglycemia. Landon et al (1989) have reported that although head growth and femur growth of IDM fetuses were similar to those of normal fetuses, abdominal circumference growth significantly exceeded that of controls beginning at 32 weeks’ gestation (abdominal circumference growth in IDM fetuses is 1.36 cm/week, versus 0.901 cm/week in normal subjects).

Morphometric studies of the IDM newborn indicate that the greater growth of the abdominal circumference is caused by deposits of fat in the abdominal and interscapular areas. This central depositing of fat is a key characteristic of diabetic macrosomia and underlies the pathology associated with vaginal delivery in these pregnancies. Acker et al (1986) showed that although the incidence of shoulder dystocia is 3% among infants weighing more than 4000 g, the incidence in infants from diabetic pregnancies who weigh more than 4000 g is 16%. Finally, despite our emphasis on birthweight, this alone may not be a sensitive measure of fetal growth. Catalano et al (2003) conducted body composition studies on infants born to mothers with diabetes and found that even when appropriate for gestational age, these infants have increased fat mass and percent body fat compared with a normoglycemic control group.

44.12.9 Maternal Diabetes Causing Fetal and Neonatal Hypoparathyroidism

Poorly controlled maternal diabetes during pregnancy is known to increase the risk of neonatal hypocalcemia, seizures, and tetany within the first 24–72 h after birth. Hyperphosphatemia may be present, which suggests that some degree of neonatal hypoparathyroidism is present. However, the mechanism by which maternal diabetes leads to fetal and neonatal hypoparathyroidism is unknown. In one case series, the cord blood ionized and total serum calcium levels were increased in infants of diabetic mothers,126 which could explain suppression of the fetal parathyroids. But it is unclear how maternal diabetes would in turn cause fetal hypercalcemia. The glucosuria of uncontrolled diabetes may cause renal wasting of magnesium, and if maternal stores are sufficiently depleted, this could conceivably lead to fetal hypomagnesemia and parathyroid suppression. However, magnesium supplementation had no effect on preventing neonatal hypocalcemia in infants of diabetic mothers.127 Additional risks for hypocalcemia in neonates born of diabetic mothers include preterm birth, lung immaturity, and asphyxia.

Clinical management should consist of increased awareness and monitoring for hypocalcemia in neonates born of diabetic mothers, especially those with poorly controlled diabetes.

50.6.2 Maternal Glucose Control

Maternal diabetes is a common predisposing factor for having infants who are born large for gestational age and subsequently develop an increased risk of obesity and diabetes in adulthood. The direct relationship between birthweight and maternal blood glucose level is noted, even if the maternal glucose level is within the normal range.8 Interestingly, infants born to mothers after a maternal diagnosis of diabetes have a higher risk of raised BMI and diabetes in adulthood than their siblings born before the diagnosis.78

Maternal diabetes during gestation is associated with fetal hyperglycemia, as maternal glucose is transported by the placenta into the fetal circulation. This stimulates the β cell of the fetal pancreas, causing fetal hyperinsulinemia.79 As insulin is a growth promoting factor, the infant is often born large for gestational age.29 During the postpartum period, the infant is no longer exposed to the high glucose environment experienced prenatally due to maternal diabetes. Despite this, the hyperinsulinemia persists. As a result, the infant experiences hypoglycemia and subsequent poor carbohydrate tolerance.80,81

Animal models of maternal diabetes in rodents using an injection of pancreatic islet toxin streptozotocin early in the pregnancy also confirmed these findings and indicate some potential biological mechanisms.35 The offspring of diabetic maternal rodents have higher body weight and appetite with hyperglycemia.36 Structural changes in the hypothalamus are noted, with an altered density of AgRP and POMC expressing neurons in the hypothalamus as well as altered leptin sensitivity.82 This adverse programming of the appetite regulatory pathway in the hypothalamus predisposes the offspring to obesity and metabolic syndrome in adult life. It can potentially be prevented by normalizing maternal glucose during pregnancy.36

Diabetes

Maternal diabetes was first mentioned in 1824, when Heinrich Gottlieb Bennewitz defended his thesis “Diabetes Mellitus: A Symptom of Pregnancy” at the University of Berlin [74]. His thesis was a simple case report and literature review on the causes and treatment of diabetes at that time. Matthew Duncan published the initial study reporting severe maternal and fetal outcomes in maternal gestational diabetes in 1884. He reviewed the pregnancies of 15 diabetic mothers and reported death in 13 out of 19 fetuses. Furthermore, 9 out of these 15 diabetic mothers died of diabetes within a year after they gave birth. Later in 1957, the term gestational diabetes mellitus (GDM) was defined by Carrington describing carbohydrate intolerance during pregnancy that resolves after birth [75].

Estrogen, prolactin, and lactogen [76] are placental hormones that transport nutrients from the mother to the fetus, thereby protecting the fetus from undernutrition and low blood glucose levels. During the course of pregnancy, insulin resistance commonly develops. If the maternal pancreas is unable to increase insulin production accordingly, insulin resistance results with elevated blood glucose levels. This pathophysiology in GDM is coherent to type 2 diabetes. In GDM, increased maternal glucose concentrations are transferred to the fetus. Within the fetal organism, elevated blood glucose in turn stimulates the insulin production of the fetal pancreatic beta cells.

In the Western world, the incidence of GDM has continuously increased over the last decades and is the most common maternal complication during pregnancy [77]. Despite new medical strategies and improved preventive care, gestational diabetes mellitus remains a serious medical condition for the developing fetus and is associated with metabolic and physiologic complications. Studies in type 1 diabetic mothers report a correlation between unstable glucose levels during the first trimester and increased risk for neurological and cardiac defects in the offspring, while unstable diabetes during the third trimester is associated with intrauterine growth restriction. Infants born to mothers that suffered from severe type 2 diabetes are more commonly macrosomic and are subsequently obese later in life, and they are also more likely to develop cardiovascular and metabolic diseases as adults [78]. These well-documented relationships led to the preventive evaluation for gestational diabetes and therapeutic strategies for most pregnant women. The association between gestational diabetes and long-term fetal consequences is well established. Furthermore, various studies document the association between preterm birth and the development of diabetes later in life. However, studies investigating the impact of maternal diabetes on preterm birth are elusive. By analyzing registries from Norway, Eidem and coworkers identified 1307 births among women with pregestational type 1 diabetes [79]. Twenty-six of these 1307 women that were diagnosed with type 2 diabetes delivered preterm of which 42% delivered spontaneously preterm. Compared to the 7% preterm birth rate in the control population, the rate of preterm deliveries was much higher in the type 1 diabetes group. Based on these data and the emphasis that preterm delivery is the most important single factor in perinatal mortality, Eidem and coworkers speculate that an excess risk of perinatal infant death in women with diabetes could be due to preterm births [79]. Further studies are needed to verify this hypothesis. However, improving maternal nutrition and optimizing maternal metabolic conditions may be the key to preventing the detrimental effects of maternal diabetes on fetal development and preterm birth.

4.9.6 Maternal Diabetes

Maternal diabetes has been shown to increase the risk for abnormal glucose metabolism in youth [52, 53]. In a study of 88 offspring of women with diabetes (including T1D, T2D, and gestational diabetes or GDM), the prevalence of IGT during adolescence was associated with amniotic fluid insulin (AFI) at 32–38 weeks gestation. The prevalence of IGT at 10–16 years was 19.5% and was not associated with the etiology of the mother’s diabetes. Seventeen subjects > 10 years of age had IGT with one having T2D. Of adolescents that had normal AFI during gestation, only 3.7% had IGT, whereas 33.3% of adolescents with elevated AFI had IGT. This study suggests that not only the history of maternal diabetes but the adequacy of glucose control during the pregnancy can be deemed important in determining risk. A study of 150 overweight or obese Hispanic youth aged 8–13 years with a family history (first- or second-degree relative) of diabetes showed 28% to have IGT [52]. None of the subjects had IFG. Of subjects whose mother had gestational diabetes (GDM), 41% had IGT.

Large-for-Gestational-Age Infants

Maternal diabetes is the most common cause of large-for-gestational-age (LGA) infants; however, post-term pregnancies (>42 weeks' gestational age), excessive maternal weight gain, multigravidity, and familial genetic factors (large parental size) can contribute to the incidence of LGA newborns. Fetal hyperinsulinism and excessive intrauterine growth is the fetal response to an abundance of glucose in the intrauterine environment. Hypoglycemia has been estimated to occur in 16% of term LGA infants within the first 24 hours of life.5 Hypoglycemia in LGA infants is generally transient and usually responds well to oral feedings (see Table 54-3 for treatment recommendations).

12.09.1.2 Maternal Diabetes As a Teratogen during Pregnancy

Maternal diabetes is a well-known teratogenic condition (Freinkel 1988; Goto and Goldman 1994; Kousseff 1999). Abnormalities most prominently associated with diabetic pregnancies include heart defects, neural tube defects (NTDs), and aberrant growth regulation, including agenesis/dysgenesis of caudal tissues (Martinez-Frias 1994). It has been shown that, at least if the readout for exposure is incidence of NTDs, high maternal blood glucose levels, achieved by periodic injection of glucose, produce the same outcomes as maternal diabetes (Horal et al. 2004). It has been suggested that excess and aberrant glucosylation may alter or impair protein function (Ahmed et al. 2005; Brownlee 2001; Sheetz and King 2002; Wautier and Schmidt 2004), but direct demonstrations of this phenomenon as a causative mechanism in developmental defects (Mills 1982) are still lacking. Similarly, it has been speculated that high-glucose levels provoke excessive insulin production (Buchanan and Kitzmiller 1994; Hornberger 2006), but maternal insulin is unlikely to directly affect the developing embryo in early gestational stages because it does not readily cross the placenta (Metzger 1991).

The embryo itself may exhibit increased insulin production, but since insulin-producing cells that respond to glucose stimulation appear relatively late in development, for example, embryonic day (E)14.5 in mouse and 5th month in human gestation, this mechanism does not account for the elevated risk for birth defects that have their etiology shortly after gastrulation (E9.5 in mouse, 4 weeks in human). Therefore, it has been concluded that diabetes-induced teratogenesis is mediated by indirect processes downstream of high-glucose levels and may include deficiencies in yolk sac function (Reece 1999) compromising nutrient transport to the embryo, or altered gene regulation in the embryo itself (Chang and Loeken 1999). The first possibility (nutrient transport) has been extensively studied, culminating in the demonstration of aberrant arachidonic acid metabolism and the formulation of the so-called yolk sac theory (Reece et al. 1994). A prominent example for the latter mechanism (altered gene expression) is the role of Pax3 downregulation (Chang and Loeken 1999; Loeken 2006) in diabetes-exposed embryos. In extensive studies, a drop in Pax3 expression in embryos was detected in murine diabetic pregnancies by E8.5 (Phelan et al. 1997). This drop was associated with hypoxia and oxidative stress (Horal et al. 2004; Li et al. 2005). Increased apoptosis was also reported (Fine et al. 1999), suggesting that the cell death could be contributing to faulty neural tube closure. Partial genetic deficiency of Pax3 (i.e., heterozygosity for a Pax3 null allele) has been shown to predispose to NTDs under conditions of maternal diabetes (Machado et al. 2001; Pani et al. 2002b). Recently, reduced Pax3 expression in neural crest cells has been linked to heart defects (Morgan et al. 2008a,b), which is a characteristic feature of diabetic embryopathy, thus suggesting Pax3 involvement in a second diabetes-induced phenotype. Nevertheless, the fact that reduced Pax3 levels are not associated with NTDs in a strain considered resistant to NTDs (Pani et al. 2002a) indicates that additional pathways are likely involved. Regulation of cell proliferation/survival by the tumor suppressor gene, p53, has been suggested to play a role (Pani et al. 2002b), but the requirement for apoptosis in NTDs remains controversial (Dunlevy et al. 2006).

Well supported, however, by work from two independent groups is the involvement of Wnt3a signaling as a likely cause of caudal growth defects in diabetic embryopathy (Chan et al. 2002; Pavlinkova et al. 2008; Shum et al. 1999). Although it has not been studied yet whether partial genetic deficiency (i.e., heterozygosity for a Wnt3a null allele) predisposes to caudal defects in diabetic pregnancy, the combination of exposure to maternal diabetes and RA produced embryos with caudal phenotypes similar to those found in human caudal dysgenesis (Chan et al. 2002).

Taken together, these studies implicate two molecular pathways in embryonic development under conditions of maternal diabetes, namely the pathways involving Wnt3a and Pax3. The evidence also establishes precedent for the notion that altered gene expression in the embryo mediates diabetes-induced malformations. This rationale and the findings that both RA (Chan et al. 2002; Shum et al. 1999), a known teratogen, and the genetic background (Pani et al. 2002a) are able to modulate the embryonic response to maternal diabetes prompted us to undertake an unbiased search for genes whose expression is altered in embryos under adverse conditions. We had three major goals in this quest: (1) to identify molecular pathways in embryos altered by exposure to maternal diabetes; (2) to derive a set of functional candidate genes that could be responsible for the phenotypic defects associated with exposure to diabetes in pregnancy; and (3) to begin to understand the molecular basis for the variation in embryonic response to this exposure.

Development of Type II Diabetes in Later Life in Macrosomic Offspring

IDMs, particularly those with macrosomia, have increased risk of developing type II diabetes earlier in life. Mechanisms responsible for this sequence of events include insulin resistance and insufficient insulin secretion (β-cell dysfunction) in response to hyperglycemia. Typically glucose intolerance from obesity and increased insulin resistance progress to fasting hyperglycemia and the inability of β-cells to compensate by increasing their rate of insulin secretion. This form of β-cell failure appears to be reversible over short periods by improved glycemic control, but long-term exposure to hyperglycemia can lead to β-cell exhaustion and specific inhibition of insulin secretion. The insulin resistance also extends to the liver where glucose production increases. This triad of insulin resistance, reduced β-cell insulin secretion, and increased hepatic glucose production produces type II diabetes.