Doppler Studies in Pregnancies with diabetes mellitus

Maternal diabetes mellitus is associated with a high risk of fetal death. In the past, before the introduction of insulin, the main cause of death was in association with maternal keto-acidosis, but now most fetal deaths are non-keto-acidotic and occur in association with fetal macrosomia.

The major source of fetal glucose is the mother and there is a good correlation between maternal and fetal blood glucose concentrations 1. In pregnancies complicated by diabetes mellitus, the maternal hyperglycemia causes fetal hyperglycemia and hyperinsulinemia 2,3. Furthermore, the fetal insulin to glucose ratio is increased because hyperglycemia and/or the other metabolic derangements associated with maternal diabetes mellitus act on the fetal pancreas to cause b -cell hyperplasia and precocious pancreatic maturation 2. Fetal hyperinsulinemia causes macrosomia, either directly through its anabolic effect on nutrient uptake and utilization, or indirectly through related peptides such as insulin-like growth factors 4.

Although good diabetic control in the third trimester of pregnancy reduces the incidence of macrosomia, the latter is not always preventable. In pregnant women with diabetes mellitus, despite stringent maternal glycemic control, the fluctuation in maternal glucose concentration is greater than in non-diabetics and it is possible that, during short-lived episodes of hyperglycemia, an already hyperplastic fetal pancreas will respond with a disproportionately high release of insulin.

In diabetic pregnancies, analysis of blood samples obtained by cordocentesis has demonstrated significant acidemia and hyperlacticemia in the absence of hypoxemia 5-7. Fetal acidemia, which may offer an explanation for the unexplained stillbirths of diabetic pregnancies, is likely to be the consequence of increased metabolic rate. Salvesen et al. performed cordocentesis in diabetic pregnancies and reported a significant association between fetal plasma insulin concentration and the degree of fetal acidemia 2. In pregnant sheep, chronic hyperglycemia results in increased aerobic and anaerobic glucose metabolism, with consequent increased oxygen consumption, lactate production and fall in pH and pO 28-13 . Glucose oxidation and oxygen consumption are also increased by hyperinsulinemia, and this effect is independent of that caused by hyperglycemia 12. Hyperlacticemia occurs because the fetus has a reduced capacity for oxidative metabolism and low pyruvate dehydrogenase activity. Severe hyperglycemia is characterized by acidemia and hypoxemia, but minor degrees of hyperglycemia are associated with acidemia in the absence of hypoxemia 8 . However, in the presence of mild fetal hypoxemia, minor degrees of fetal hyperglycemia do result in severe acidosis and even fetal death 13.

The alternative explanation for fetal acidemia in maternal diabetes mellitus is impaired placental perfusion. Histological studies have reported decreased villous surface area, villous edema and thickening of the basement membrane 14 . However, the finding that acidemia is not accompanied by hypoxemia suggests that the acidemia is unlikely to be due to impaired placental function; in pregnancies complicated by intrauterine growth restriction due to uteroplacental insufficiency, acidemia is accompanied by hypoxemia (see Chapter 4 ).

Infants of diabetic mothers are at increased risk of hypertrophic cardiomyopathy. This disease is characterized by a thickening of the interventricular septum and ventricular walls and by systolic and diastolic dysfunction, which may result in congestive heart failure.

Rizzo et al. examined 40 well-controlled insulin-dependent diabetic pregnancies at 20–38 weeks of gestation and reported a significant increase in fetal interventricular septal thickness (Figure 1) and an associated decrease in the ratio between the peak velocities during early passive ventricular filling and active atrial filling at the level of the atrioventricular valves 34. These findings, which were unrelated to maternal glycosylated hemoglobin levels, suggest that, even in well-controlled maternal diabetes mellitus, there is fetal interventricular septal hypertrophy that affects cardiac diastolic function. A longitudinal study of 14 well-controlled, insulin-dependent diabetic pregnancies at 20–36 weeks of gestation confirmed the presence of hypertrophy in the interventricular septum and the right and left ventricular walls, as well as abnormal development of cardiac function – decrease in the ratio between early and active ventricular filling at the level of both the mitral and tricuspid valves (Figure 2) 35. The cardiomegaly and cardiac dysfuncion increased with gestation but they were evident from as early as 20 weeks. Since the diabetic control in these pregnancies was good, it was suggested that fetal cardiomegaly may be the consequence of increased insulin sensitivity of the fetal myocardium. This hypothesis is supported by the data of Thorsson and Hintz, showing a reduction from fetus to adult in the number and affinity of insulin receptors 36.

Figure 1: Real-time and M-mode tracing of a fetus of an insulindependent diabetic mother at 36 weeks of gestation. The interventricular wall septal thickness is increased (10 mm compared to the expected mean of 5 mm for this gestation)

Figure 2: Flow velocity waveforms across the tricuspid valve in a fetus of an insulin-dependent diabetic mother at 32 weeks of gestation. The E/A ratio is decreased (0.48 compared to expected mean of 0.77 for this gestation)

The lower ratio between early and active ventricular filling at the level of the atrioventricular valves in fetuses of diabetic mothers may be due to impaired development of ventricular compliance, possibly secondary to cardiac wall thickening. In addition, the ratio may be influenced by reduced preload, as a consequence of the polycythemia, and therefore increased blood viscosity in fetuses of diabetic mothers. Thus, in a Doppler study of 37 fetuses of insulin-dependent diabetic mothers, immediately before an elective Cesarean section, the ratio between early and active ventricular filling was significantly and independently affected by both the interventricular wall thickness and fetal hematocrit 37.

Weber et al. performed serial evaluations of cardiac growth and ventricular diastolic filling using M-mode and Doppler echocardiography at 20–40 weeks of gestation and at 48–72 hours after birth in 11 fetuses of non-diabetic mothers and nine fetuses of well-controlled insulin-dependent diabetic mothers. Cardiac growth and birth weight in the two groups were similar 38. Ventricular diastolic filling increased with gestation in both groups but the increase was delayed in the diabetic group. Tsyvian et al. examined 15 third-trimester pregnancies complicated by insulin-dependent maternal diabetes mellitus and reported a delay in fetal left ventricular filling, which may reflect changes in myocardial relaxation 39. Weiner et al. examined 31 well-controlled insulin-dependent diabetic pregnancies at 22–40 weeks of gestation and reported that, after 34 weeks, in the fetuses of diabetic pregnancies, compared to normal pregnancies, there was a lower ratio between the peak velocities during early passive ventricular filling and active atrial filling at the level of the atrioventricular valves 40.

Macklon et al. examined well-controlled diabetic pregnancies at 18–20 weeks of gestation and reported that the fetal intraventricular septal thickness was increased, but transvalvular peak flow velocities and the duration of ventricular ejection in the fetal heart were not significantly different from normal 41. It was concluded that, in fetuses of well-controlled diabetic pregnancies, altered cardiac morphology is evident early in pregnancy, before any obvious alteration in cardiac function. Rizzo et al. examined 27 insulin-dependent diabetic pregnancies from 12 weeks of gestation 42. In the fetuses of diabetic mothers, compared to normal pregnancies, there was a lower ratio between early and active ventricular filling at the level of the atrioventricular valves, a higher percentage of reverse flow during atrial contraction in the inferior vena cava, and a higher proportion of cases with pulsations in the umbilical vein. These findings, demonstrating impaired development of cardiac and venous blood flow patterns from as early as at 12 weeks of gestation, were more evident in pregnancies with poorer glycemic control but they were also found in the presence of good metabolic control.

Peak velocities at the level of the aortic and pulmonary outflow tracts were significantly higher in fetuses of diabetic mothers than in normal fetuses 34. The most likely explanations for the increased peak velocities are increased cardiac contractility (also found in postnatal studies in infants of diabetic mothers) and increased intracardiac flow volume due to the relatively large size of such fetuses, since cardiac output is a function of fetal weight. These abnormalities in cardiac hemodynamics also impair the venous circulation. Rizzo et al. carried out a cross-sectional study of 62 third-trimester fetuses in insulin-dependent diabetic pregnancies and reported that increased preload index in the inferior vena cava (in the absence of alterations in fetal peripheral vessels) was associated with a lower umbilical arterial blood pH and higher hematocrit at birth, as well as increased neonatal morbidity 43. These findings suggest that the mechanisms inducing fetal distress in diabetic pregnancies (where the development of hypertrophic cardiomyopathy plays a pivotal role in the genesis of fetal distress) are different from those in fetuses with intrauterine growth restriction, where the change in cardiac function is secondary to the alteration in peripheral resistance.

In neonates from normal pregnancies, the ratio between early and active ventricular filling at the level of the atrioventricular valves increases during the first few days of postnatal life; the early wave is usually higher than the active one, resulting in a ratio between early and active ventricular filling that is higher than one. In newborns of diabetic mothers, there are no changes in this ratio during the first 5 days of life and its value remains lower than one 44. These anomalies might explain the relatively high incidence of transitory tachypnea and pulmonary edema in neonates from diabetic pregnancies. The cardiac hypertrophy of fetuses of diabetic mothers resolves during the first year of postnatal life. However, it is possible that the cardiac hypertrophy and dysfunction observed in intrauterine life may affect cardiac function in adult life.

In maternal diabetes mellitus, impedance to flow in the uterine and umbilical arteries is not related to either short-term or long-term maternal glycemic control.

In maternal diabetes mellitus, impedance to flow in the uterine arteries is normal, even in patients with nephropathy and vasculopathy. However, increased impedance, as in non-diabetic pregnancies, identifies a group at high risk for subsequent development of pre-eclampsia and/or intrauterine growth restriction.

In maternal diabetes mellitus, increased impedance to flow in the umbilical artery is associated with the development of pre-eclampsia and/or intrauterine growth restriction. There is contradictory evidence concerning a possible increase in impedance in pregnancies with maternal vasculopathy.

In maternal diabetes mellitus, there is no evidence of redistribution in the fetal circulation with decreased PI in the middle cerebral artery and increased PI in the descending thoracic aorta. This is presumably because, in diabetes, there may be acute fluctuations in fetal blood pH, since the latter is associated with the maternal glucose concentration. Furthermore, unlike intrauterine growth restriction, in diabetes metabolic derangements in the fetus may lead to acidemia without hypoxemia. Therefore, the classic redistribution seen in fetal hypoxemia due to uteroplacental insufficiency may not occur even in severely compromised fetuses, and it is therefore important not to be misled by apparently normal fetal Doppler results.

In maternal diabetes mellitus, the fetus is at increased risk of hypertrophic cardiomyopathy. This disease is characterized by a thickening of the interventricular septum and cardiac dysfunction, which may be evident from as early as 12 weeks of gestation.

produced at Centrus ®