Get Adobe Flash player


In all women with preexisting diabetes mellitus, there is a 9-14% rate of miscarriage. Current data suggest a strong association between the degree of glycemic control before pregnancy and the miscarriage rate. Suboptimal glycemic control has been shown to double the miscarriage rate in women with diabetes. A correlation also exists between more advanced diabetes and miscarriage rates. Patients with long-standing (>10 y) and poorly controlled diabetes (glycohemoglobin exceeding 11%) have been shown to have a miscarriage rate of up to 44%. Conversely, excellent glycemic control normalizes the miscarriage rate.


Birth defects

Among the general population, major birth defects occur in 1-2% of the population. In women with overt diabetes and suboptimal glycemic control before conception, the likelihood of a structural anomaly is increased 4- to 8-fold.

Although initial reports demonstrated anomaly rates as high as 18% in women with preexisting diabetes mellitus,[21] more recent studies, in patients who received more aggressive preconception and first trimester management, report anomaly rates between 5.1 and 9.8%.[22, 23]

Two-thirds of birth anomalies involve the cardiovascular and central nervous systems. Neural tube defects occur 13-20 times more frequently in diabetic pregnancies, and genitourinary, gastrointestinal, and skeletal anomalies are also more common.

It is notable that no increase in birth defects occurs among the offspring of fathers who have diabetes or the offspring of women who develop gestational diabetes after the first trimester. This suggests that periconceptional glycemic control is the main determinant of abnormal fetal development in diabetic women.

When the frequency of congenital anomalies in patients with normal or high first-trimester maternal glycohemoglobin values was compared to the frequency in healthy patients, the rate of anomalies was only 3.4% with glycosylated hemoglobin values (HbA1C) of less than 8.5%, versus 22.4% with poorer glycemic control in the periconceptional period (HbA1C >8.5%). An overall malformation rate of 13.3% was reported in 105 patients with diabetes, but the risk of delivering a malformed infant was comparable to a normal population when the HbA1C was less than 7%.[24] More recently, in a review of 7 cohort studies, researchers found that patients with a normal glycohemoglobin (0 SD above normal), the absolute risk of an anomaly was 2%. At 2 SD above normal, this risk was 3%, with an odds ratio of 1.2 (1.1- 1.4). As the glycohemoglobin increased so did the risk for malformation, in a direct relationship.[25]

Clinical trials of intensive metabolic care have demonstrated that malformation rates similar to those in the nondiabetic population can be achieved with meticulous preconceptional glycemic control.[23] Subsequent trials comparing a preconceptional intensive metabolic program to standard treatment have demonstrated lowered rates of perinatal mortality (0% vs 7%) and congenital anomalies (2% vs 14%). In addition, when the preconceptional counseling program was discontinued, the congenital anomaly rate increased by over 50%.[26]

Growth restriction

Although most fetuses of diabetic mothers exhibit growth acceleration, growth restriction occurs with significant frequency in pregnancies in women with preexisting type 1 diabetes. The most important predictor of fetal growth restriction is underlying maternal vascular disease. Specifically, pregnant patients with diabetes-associated retinal or renal vasculopathies and/or chronic hypertension are most at risk for growth restriction.


Excessive body fat stores, stimulated by excessive glucose delivery during diabetic pregnancy, often extends into childhood and adult life. Maternal obesity, common in type 2 diabetes, appears to significantly accelerate the risk of infants being LGA. Approximately 30% of fetuses of women with diabetes mellitus in pregnancy are large for gestational age (LGA). In preexisting diabetes mellitus, this incidence appears to be slightly higher (38%).[13]

In women with gestational diabetes, weight gain during pregnancy that exceeds Institute of Medicine (IOM) weight-gain guidelines increases the risk of preterm delivery, of having a newborn who is LGA, and of requiring a cesarean delivery.[27]The chance that a newborn would be small for gestational age (SGA) was greater among women with gestational diabetes whose weight gain was below the IOM guidelines.

Plotting of serial ultrasonographic examination findings from diabetic fetuses shows that the growth velocity of the abdominal circumference is often well above the growth percentiles seen in nondiabetic fetuses, and it is higher than the fetal head and femur percentiles. The growth of the abdominal circumference begins to rise significantly above normal after 24 weeks.


Macrosomia is typically defined as a birth weight above the 90th percentile for gestational age or greater than 4000 g. Macrosomia occurs in 15-45% of babies born to diabetic women, a 3-fold increase from normoglycemic controls.

Maternal obesity has a strong and independent effect on fetal macrosomia. Birth weight is largely determined by maternal factors other than hyperglycemia, with the most significant influences being gestational age at delivery, maternal prepregnancy body mass index (BMI), maternal height, pregnancy weight gain, the presence of hypertension, and cigarette smoking.

When women who are very obese (weight >300 lb) were compared with women of normal weight, the newborns of obese women had more than double the risk of macrosomia compared to those of women of normal weight.[28] This may explain the failure of glycemic control to completely prevent fetal macrosomia in several series.

Excess nutrient delivery to the fetus causes macrosomia and truncal fat deposition, but whether fasting or peak glucose values are more correlated with fetal overgrowth is less clear. Data from the Diabetes in Early Pregnancy project indicate that fetal birth weight correlates best with second- and third-trimester postprandial blood sugar levels and not with fasting or mean glucose levels.[29]When postprandial glucose values average 120 mg/dL or less, approximately 20% of infants can be expected to be macrosomic. When postprandial levels range as high as 160 mg/dL, macrosomia rates can reach 35%.

Macrosomia is associated with excessive rates of neonatal morbidity, as illustrated by a study by Hunter et al in 1993, in which the infants of diabetic mothers had 5-fold higher rates of severe hypoglycemia, a 4-fold increase in macrosomia, and a doubled increase in neonatal jaundice relative to infants of mothers without diabetes.[30]

The macrosomic fetus in diabetic pregnancy develops a unique pattern of overgrowth, involving central deposition of subcutaneous fat in the abdominal and interscapular areas.[31] Skeletal growth is largely unaffected.

Neonates of diabetic mothers have a larger shoulder and extremity circumference, a decreased head-to-shoulder ratio, significantly higher body fat, and thicker upper extremity skin folds compared with nondiabetic control infants of similar weights. Because fetal head size is not increased during poorly controlled diabetic pregnancy, but shoulder and abdominal girth can be markedly augmented, the risk of injury to the fetus after delivery of the head (eg, Erb palsy) is significantly increased. Thus, birth injury, including shoulder dystocia and brachial plexus trauma, are more common among infants of diabetic mothers, and macrosomic fetuses are at the highest risk.

More recent data from the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS) trial demonstrated a positive relationship between severity of maternal fasting hyperglycemia and risk of shoulder dystocia, with a 1 mmol increase in fasting glucose leading to a 2.09 relative risk for shoulder dystocia.[32]

In addition, there appears to be a role for excessive fetal insulin levels in mediating accelerated fetal growth. In the study by Simmons et al which compared umbilical cord sera in infants of diabetic mothers and controls, the heavier, fatter babies from diabetic pregnancies were also hyperinsulinemic.[33]

Metabolic syndrome

The adverse downstream effects of abnormal maternal metabolism on the offspring have been documented well into puberty. Glucose intolerance and higher serum insulin levels are more frequent in children of diabetic mothers than in normal controls. By age 10-16 years, offspring of diabetic pregnancy have a 19.3% rate of impaired glucose intolerance.[34]

A study by Patel et al found that maternal pregnancy glycosuria, gestational diabetes, and existing diabetes demonstrate some association with higher offspring fasting glucose and insulin; however, little evidence suggests an association between maternal diabetes or glycosuria with offspring lipids, blood pressure, and C-reactive protein.[35]

The childhood metabolic syndrome includes childhood obesity, hypertension, dyslipidemia, and glucose intolerance. A growing body of literature supports a relationship between intrauterine exposure to maternal diabetes and risk of a metabolic syndrome later in life.[36, 37] Fetuses of diabetic women that are born large for gestational age appear to be at the greatest risk.[37]

Cardiovascular risk factors

An associational study examined the effect on maternal diabetes in utero and cardiovascular risk factors in offspring.[38] Offspring born to mothers with diabetes exhibited higher levels of biomarkers for endothelial damage and inflammation, as well as higher leptin levels, BMI, waist circumference, and systolic blood pressure and decreased adiponectin levels. The association remained significant when controlling for maternal prepregnancy BMI.

Neurocognitive development

In a study of 212 preschool children, Nomura et al found that maternal GDM and low socioeconomic status were associated with an increased risk for attention-deficit/hyperactivity disorder (ADHD) at age 6 and that children exposed to both GDM and low socioeconomic status were at even greater risk for ADHD and also at increased risk for compromised neurobehavioral functioning.

Add comment

Security code

Hội Nội Tiết- Đái Tháo Đường Miền Trung Việt Nam

Diabetes- Endocrinology- Metabolism Association of central Vietnam


Liên hệ ban biên tập: 0903574457, email: hoặc

Năm xuất bản: 12/2010