E. Albert Reece
Carol J. Homko
INTRODUCTION
The discovery of insulin in 1921 brought about a significant improvement in the overall outlook for women with diabetes and their reproductive potential. The incidence of both maternal and perinatal mortality has decreased markedly over the past 80-plus years as a result of this discovery as well as a multitude of other scientific advances, including fetal heart rate and blood glucose monitoring and neonatal intensive care. However, despite these advances, women with diabetes and their offspring remain at increased risk for a number of complications. The increased morbidity is directly related to the severity of maternal hyperglycemia. Therefore, the management goal of these pregnancies is strict glycemic control prior to conception and throughout gestation. This is best accomplished through the provision of multidisciplinary team care and targeted self-management education.
FUEL METABOLISM
Pregnancy is recognized as having a profound effect on maternal fuel metabolism. These pregnancy-related alterations in maternal metabolism are necessary to meet the demands of the developing fetus. Studies in lean healthy pregnant women have demonstrated a greater than normal sensitivity to the blood glucose–lowering effect of exogenously administered insulin during early pregnancy as compared with late gestation. In addition, insulin responses to an oral glucose load have been demonstrated to be increased in early pregnancy as compared with the nonpregnant state in the same glucose-tolerant women.
These increases in serum insulin levels and insulin sensitivity produce a milieu during early gestation that favors maternal fat accumulation in preparation for the rise in energy requirements associated with the rapid growth of the fetal–placental unit during late pregnancy. Moreover, the increases in plasma concentrations of estrogen, progestins, and cortisol observed during early pregnancy are also likely to stimulate fat accumulation.
Late gestation is characterized by accelerated growth of the fetoplacental unit, rising plasma concentrations of several diabetogenic hormones including human placental lactogen and estrogens, and increasing insulin resistance. Several investigators have demonstrated increased first- and second-phase insulin release during late gestation, as well as increased plasma insulin-to-glucose ratios. Studies using the euglycemic-hyperinsulinemic clamp and minimal model techniques have reported that peripheral insulin sensitivity is reduced by 33% to 50% during late gestation. During the third trimester of pregnancy, insulin-stimulated carbohydrate oxidation is reduced out of proportion to the decrease observed in insulin-stimulated glucose uptake. Endogenous glucose production is also significantly less inhibited during the third trimester when compared with either the second trimester or the nonpregnant state. Thus, there appears to be general agreement that the second half of pregnancy is associated with increasing insulin resistance, both in the periphery (muscle) and at the hepatic level.
The cause or causes of the increased insulin resistance during late pregnancy are not entirely clear. The parallel development of insulin resistance and increases in blood levels of human placental lactogen, a hormone with strong lipolytic and anti-insulin action, suggests that human placental lactogen and perhaps other diabetogenic hormones, including cortisol, progesterone, and estrogens, may be responsible for much of the observed insulin resistance. In addition, there is also evidence to support a role for plasma free fatty acids in the development of insulin resistance during late pregnancy.
The development of peripheral and hepatic insulin resistance after midpregnancy can be seen as an effort of the mother to adapt to the fuel needs of the rapidly growing fetus. During the third trimester of pregnancy, glucose uptake by the fetus has been estimated to be approximately 33 µmol/kg per minute. To satisfy this additional need, peripheral insulin resistance increases in pregnant women, thus reducing maternal glucose utilization. In addition, hepatic insulin resistance increases, which increases hepatic glucose production. Moreover, by decreasing carbohydrate oxidation, much of the glucose entering muscle is shunted into alanine or lactate, which can be recycled into glucose.
CLASSIFYING DIABETES
Diabetes mellitus generally is classified into the following categories: type 1 or insulin-dependent diabetes mellitus, type 2 or non–insulin-dependent diabetes mellitus, and gestational diabetes mellitus (GDM). Approximately 10% of all individuals with diabetes mellitus have type 1 diabetes. Beta-cell destruction, with resulting insulin deficiency, is the hallmark of this disorder. Onset is generally before the age of 30 and, as a result, this type of diabetes frequently is encountered in women of childbearing age. It is estimated that type 1 diabetes complicates approximately 0.2% of all pregnancies in the United States annually.
Type 2 diabetes is the most common form of the disease, affecting nearly 90% of all individuals with diabetes. Type 2 diabetes is characterized by defects in both insulin action and secretion. It typically is seen in individuals over the age of 40 and, therefore, in the past was felt to be uncommon in women of childbearing age. However, the incidence of type 2 diabetes has been increasing steadily among younger individuals, and data from the National Maternal and Infant Health Survey indicate that type 2 complicates 0.3% of all pregnancies in the United States. GDM is defined as carbohydrate intolerance of variable severity, with onset or first recognition during the index pregnancy.
If the abnormality in glucose tolerance persists after pregnancy, the patient's diagnosis is revised to type 1, type 2, or impaired glucose tolerance.
GESTATIONAL DIABETES MELLITUS
Gestational diabetes is a common problem, complicating approximately 2% to 5% of all pregnancies in the United States. The likelihood of developing gestational diabetes is increased significantly among certain subgroups, and these include individuals with a positive family history of type 2 diabetes, advancing maternal age, obesity, and nonwhite ethnicity. Excess risks for both gestational diabetes and impaired glucose tolerance have been demonstrated in African American, Hispanic, and Native American women, as well as in women from the Indian subcontinent and the Middle East.
Screening and Diagnosis for Gestational Diabetes
Screening of pregnant women for gestational diabetes remains a topic of great debate both in this country and throughout the world. In 1998, the Fourth International Workshop-Conference on GDM acknowledged that there were certain populations of low-risk women in whom it may not be cost effective to screen routinely for GDM. This low-risk group includes women who were not members of ethnic minorities, were less than 25 years of age, had no first-degree relatives with diabetes, and are of normal body weight. Although selective screening may be appropriate in certain populations with a low prevalence of type 2 diabetes, universal screening is still advocated in most centers. The American College of Obstetricians and Gynecologists recommends that all pregnant patients be screened for GDM whether by patient's history, clinical risk factors, or a laboratory screening test. However, the American, Canadian, and British diabetes associations recommend biochemical screening.
Screening should be performed between weeks 24 and 28 of gestation, although women with significant risk factors may benefit from being screened earlier in pregnancy. A 50-gram glucose challenge test is performed without regard to the time of day or interval since the last meal. A venous plasma glucose level is measured 1 hour later, with a value of ≥140 mg/dL indicating the need for a 3-hour, 100-gram oral glucose tolerance test (OGTT). However, lowering the cut-off value to 130 to 135 mg/dL increases the yield of cases of gestational diabetes by 10%. The use of either threshold is acceptable.
The OGTT is performed after an overnight fast and 3 days of an unrestricted carbohydrate diet. A fasting blood glucose level is drawn, and a 100-gram glucose load is then administered. Plasma glucose levels are drawn 1, 2, and 3 hours following ingestion of the glucose solution. Diagnosis requires that at least two of the four glucose levels of the OGTT meet or exceed the upper limits of normal. Two sets of diagnostic criteria are in use in the United States: those of the National Diabetes Data Group and those of Carpenter and Coustan. The Fourth International Workshop-Conference participants believed there was sufficient evidence to suggest that the Carpenter-Coustan criteria, with its lower cut-off values, more accurately predict neonatal risks and have recommended the use of these criteria. American College of Obstetricians and Gynecologists guidelines support the use of either set of criteria.
In addition, the Fourth International Workshop-Conference acknowledged an alterative to the two-step screening-diagnostic procedure described above. The evaluation of glucose intolerance during pregnancy may be made using a one-step approach or 2-hour 75-gram OGTT. This approach is considered most applicable in high-risk populations. Studies are underway to establish the relationships among the various diagnostic criteria and perinatal outcomes in these pregnancies.
Women with gestational diabetes should be evaluated at the first postpartum visit by a 2-hour OGTT using a 75-gram load. More than 90% of women will convert to normal glucose tolerance following delivery. However, studies indicate that the risk for overt diabetes may be as high as 20% to 50% in this population. Long-term annual follow-up is therefore indicated. GDM is greatly influenced by body weight, with the highest rate occurring in obese patients. Women with a history of GDM should be counseled regarding the use of lifestyle changes such as weight loss and regular exercise to reduce this risk.
PREGNANCY COMPLICATIONS
Despite improvements in pregnancy outcomes, women with both gestational and pregestational diabetes are at greater risk for a number of pregnancy-related complications (Table 15.1). These include preterm labor, infectious morbidities, hydramnios, and hypertensive disorders. In addition, multiple investigators have reported a significant association between poor glycemic control and hypertensive disorders, preterm labor, and infections. Furthermore, women with pregestational diabetes are at risk for the acute complications of diabetes because of the metabolic alterations associated with pregnancy, as well as the effects of strict glycemic control. The vascular alterations associated with long-term diabetes also contribute to the higher morbidity rates observed in women with diabetes. Both diabetic nephropathy and retinopathy may progress during pregnancy and should be monitored closely.
|
|
|
TABLE 15.1. Pregnancy complications in diabetes |
Current data would seem to indicate that pregnancy is an independent risk factor for diabetic retinopathy. Hypertension, poor control early in pregnancy, and rapid normalization all appear to be associated with the potential to accelerate retinal deterioration. Furthermore, women with more advanced forms of retinopathy and a longer duration of diabetes are at highest risk for progression. All women who have had type 1 diabetes for 5 years or more or type 2 diabetes at diagnosis require a thorough dilated ophthalmologic evaluation. This may necessitate preconception fluorescein angiography, because dye studies generally are contraindicated during pregnancy. These evaluations ideally should be completed prior to attempting conception. Laser therapy, if indicated, also needs to be completed prior to conception.
In contrast, most studies of the effects of pregnancy on diabetic nephropathy have suggested that pregnancy is not associated with either the development or progression of preexisting nephropathy in women with mild to moderate disease. However, diabetic nephropathy is the complication of diabetes most likely to affect pregnancy outcomes. There are increased risks for pregnancy-induced hypertension or a progression of already existing hypertension, intrauterine growth retardation resulting in small-for-gestational-age infants, preterm deliveries secondary to fetal distress, and a ten-fold increase in the incidence of stillbirth over women with diabetes but without nephropathy. Preeclampsia is the most frequent, serious complication of maternal nephropathy, with implications for both mother and fetus. Close monitoring of blood pressure, with addition or adjustment of antihypertensive agents as needed, is recommended. The drugs of choice are methyldopa, calcium channel blockers, and β blockers. Select calcium channel inhibitors, such as diltiazem, induce mild reductions in blood pressure but have a potent effect on decreasing excess protein excretion.
NEONATAL COMPLICATIONS
The offspring of women with diabetes remain at increased risk for a number of complications, which include congenital anomalies, fetal macrosomia, respiratory distress syndrome (RDS), metabolic abnormalities, as well as long-term sequelae (see Table 15.1).
Perinatal Mortality
The two major causes of perinatal mortality are unexplained fetal death and congenital malformations. The causes of unexpected death are not well understood. In animal models sustained hyperglycemia has been associated with increased insulin secretion, elevated fetal oxygen consumption, acidosis, and death. It has been postulated that fetal polycythemia and increased platelet aggregation could explain the increased incidence of intravascular thrombosis in infants of diabetic mothers and that thrombotic episodes could be the underlying cause for late unexplained intrauterine deaths.
Approximately 40% of perinatal deaths that occur among infants of women with diabetes can be attributed to malformations. Diabetes mellitus is one of the most common maternal conditions that results in anomalous offspring. The frequency of major congenital anomalies is increased two-fold to three-fold over that of the general population. Great diversity is seen in the types of malformations associated with insulin-dependent diabetes mellitus. The most frequent types of malformations involve the central nervous system, cardiovascular, gastrointestinal, genitourinary, and skeletal systems, with cardiac malformations being the most common.
The defects most often associated with diabetes occur during organogenesis before 7 weeks gestation. Clinical series in humans have shown an association between malformations and glucose control early in pregnancy. In addition, other investigators have demonstrated that tight glucose control either prior to conception or very early during pregnancy can effectively reduce the rate of major malformations. As a result of these and other studies, the management goal for diabetic pregnancies has become the establishment and maintenance of near euglycemia beginning with the preconceptual period and continuing throughout gestation.
Altered Fetal Growth
Macrosomia is a classic hallmark of the pregnancy complicated by diabetes and is reported to occur in 20% to 25% of pregnancies complicated by diabetes. Macrosomia is defined as excessive birth weight (>90%) for gestational age or as a birth weight over 4,000 grams. Increased adiposity is the primary cause of the increased birth weight seen in the offspring of diabetic women. Numerous studies have established a relationship between the level of maternal glucose control and macrosomia. Mothers of macrosomic infants usually have significantly elevated plasma glucose levels at term, indicating increased glucose availability to the fetus, with hyperinsulinemia a likely intermediate step, during the third trimester. Other factors associated with an increased risk for fetal macrosomia include increased maternal weight, increased parity, previous delivery of a macrosomic infant, and insulin requirements greater than 80 units/day.
Macrosomic fetuses have higher perinatal and neonatal mortality and morbidity rates. Approximately 10% of infants weighing over 4,500 grams at birth will require admission to a neonatal intensive care nursery. In addition, the reported perinatal mortality is 2 to 5 times higher in this group of children than in average-sized children. Delivery of a macrosomic infant is dangerous because of the risk for birth trauma to the head and neck. Fetal asphyxia and meconium aspiration may occur as a result of prolonged labor secondary to unrecognized cephalopelvic disproportion and shoulder dystocia.
At the other extreme, women with type 1 diabetes are also at increased risk for delivering a small-for-gestational-age infant. In general, the risk of growth retardation increases with the severity of the mother's clinical diabetes. Vascular complications, such as retinopathy and nephropathy, are believed to be associated with uteroplacental insufficiency in pregnant women with diabetes. Poor maternal renal function, hypertension, and placental lesions all have been associated with intrauterine growth retardation in the offspring of diabetic mothers. However, more recent evidence suggests that the growth retardation may be related to disturbances in maternal fuels during organogenesis.
Metabolic Abnormalities
Hypoglycemia occurs when plasma glucose levels fall below 35 mg/dL in the term infant and 25 mg/dL in the preterm infant. Infants of diabetic mothers can develop hypoglycemia during the first few hours of life, particularly in cases of poor glycemic control. Macrosomic infants and infants with elevated cord blood C-peptide or immunoreactive insulin levels are also at increased risk. The incidence of hypoglycemia is reported to range from 25% to 40% in infants of mothers with diabetes. Both poor glycemic control during pregnancy and high maternal plasma glucose levels at the time of delivery increase the risk of its occurrence.
The incidence of hypocalcemia also is increased significantly in the infants of women with diabetes. Hypocalcemia generally occurs in association with hyperphosphatemia and occasionally with hypomagnesemia. Neonatal hypocalcemia is defined as a calcium level at or below 7 mg/dL. Serum calcium levels are usually lowest on the second or third day of life. Polycythemia is defined as a venous hematocrit that exceeds 65% and is reported to occur in one third of neonates born to diabetic women. The polycythemia is believed to occur as a result of chronic intrauterine hypoxia, which leads to an increase in erythropoietin and results in an increase in red blood cell production. Neonates born to women with diabetes also have a higher incidence of hyperbilirubinemia as compared with nondiabetic controls. Neonatal hyperbilirubinemia develops in approximately 20% to 25% of cases. Animal studies have demonstrated an association between fetal insulin levels and elevated levels of erythropoietin, which is the major regulator of erythropoiesis in the neonate.
Other Infant Morbidities
Offspring born to women with diabetes mellitus are also at increased risk of developing various hypertrophic types of cardiomyopathy and congestive heart failure. The exact incidence is not known, but one study reported that 10% of infants born to women with diabetes may have evidence of myocardial and septal hypertrophy. Thickening of the interventricular septum and of the left or right ventricular wall can occur, and it is felt to be a result of the fetal hyperinsulinemic state. In the majority of cases the infants are asymptomatic, and the myocardial changes are detectable only by electrocardiogram or echocardiogram. In severe cases, left ventricular outflow obstruction can result and may lead to reduced cardiac output and congestive heart failure during the first few days of life. The abnormalities do not appear to damage the myocardium permanently and in most cases regress by 6 months of age.
RDS is another common complication associated with diabetes. In the past, offspring born to women with diabetes had a four-fold to six-fold greater incidence of RDS at every week of gestation. This incidence was decreased dramatically with the initiation of strict metabolic control. More recent studies seem to indicate that stringent metabolic control may reduce the incidence of RDS in neonates of women with diabetes to near background level in the population.
Lastly, there are long-term consequences of diabetic pregnancies. These include childhood obesity, neuropsychological deficits, and an increased tendency to develop overt diabetes.
PRECONCEPTION CARE
Care of women with type 1 or type 2 diabetes ideally begins before conception. Although numerous clinical trials have demonstrated that strict glycemic control prior to and during early gestation can reduce the rate of structural defects, the vast majority of women with diabetes still seek medical care only after they learn they are pregnant. Consequently, the rate of congenital malformations in infants of mothers with diabetes has continued to remain significantly higher than that of the general population.
A prepregnancy assessment should be undertaken to document a woman's overall fitness for pregnancy. This includes a careful history and thorough physical examination to assess the patient's vascular status. Baseline creatinine clearance and protein excretion levels should be evaluated and an electrocardiogram performed. These women also should be referred for an ophthalmologic consultation. Optimization of blood glucose control should be achieved before the woman is advised to become pregnant. Women should receive appropriate contraceptive therapy while they are preparing for pregnancy. For patients who are not already following an intensive regimen, an extensive period of education and the institution of self–blood glucose monitoring is also necessary.
Prepregnancy counseling for women with gestational diabetes should begin immediately after delivery. These women need to be advised that they are at significant risk for developing GDM in subsequent pregnancies and that they are at increased risk for developing type 2 diabetes as they age.
DIABETES MANAGEMENT
The main goal of management for pregnancies complicated by either gestational or pregestational diabetes is to achieve and maintain euglycemia throughout gestation. The treatment approach requires a combination of diet, exercise, intensive insulin regimens, and daily multiple blood glucose determinations.
Diet
Diet therapy is the cornerstone of diabetes management in pregnancy, just as it is in the nonpregnant state. The goal of diet therapy is to meet the additional nutrition requirements of pregnancy while at the same time maintaining good glycemic control.
All pregnant women with diabetes should be seen by a nutritionist for individualized diet counseling. There are no universal recommendations for determining the number of calories required during pregnancy. Most experts advocate an additional intake of 300 to 400 kcal/day to meet the needs of pregnancy. Recommendations regarding total caloric intake often are based on the mother's pregravid weight: 30 kcal/kg of body weight per day for normal-weight women with diabetes, 40 kcal/kg per day for underweight women, and 24 kcal/kg per day for overweight women (Table 15.2). Although it is generally agreed that pregnancy is not the time for weight reduction, obese women with type 2 or GDM should be encouraged to achieve a weight gain of no more than 15 pounds. Modest caloric restrictions have not been associated with either ketonuria or elevated plasma ketone concentrations.
|
|
|
TABLE 15.2. Preparation of a diet for a pregnant diabetic woman |
The most current guidelines suggest that the composition of the diabetic diet be based on an individualized nutrition assessment. Because the postprandial blood glucose level has been shown to be a major factor in the development of neonatal macrosomia, a goal of nutrition therapy is to blunt postprandial hyperglycemia. The postprandial blood glucose level is most influenced by the carbohydrate content of the meal (see Table 15.2). Carbohydrate levels of 40% to 45% of total calories are considered appropriate during pregnancy. To achieve euglycemia, dietary fat can be liberalized up to 40% of calories. Saturated fats should be limited to one third of fat calories or less. The remaining calories should come from either monounsaturated or polyunsaturated fats. The recommended dietary allowance for protein in pregnancy is 60 grams per day, and this typically ranges from 12% to 20% of total calories.
A number of different approaches to meal planning can be used and range from general guidelines for good nutrition to structured meal planning strategies. The most appropriate approach will depend on the patient's type of diabetes, educational level, lifestyle, habits, motivation, and economic restraints. Although the nutrition needs and distribution of calories for women with gestational diabetes and pregestational diabetes are similar, the number and timing of snacks will vary. Two or more snacks generally are recommended for women with type 1 diabetes, while women with type 2 diabetes or gestational diabetes require only a bedtime snack. The bedtime snack should include a complex carbohydrate and protein to prevent late-night hypoglycemia and early morning starvation ketosis. Artificial sweeteners are considered safe to use during pregnancy, although no specific recommendations concerning intake levels exist.
Exercise
Although exercise has been demonstrated to be beneficial in nonpregnant individuals, evidence regarding the risk and benefits of either periodic or regular exercise in pregnant women with diabetes is limited. Little has been published regarding exercise during pregnancy complicated by preexisting diabetes, either type 1 or type 2. However, walking is possible for most women and has been reported to improve lipid profiles and blood glucose control.
Pregnancy is not a time to initiate a new exercise program, although women who have been exercising regularly prior to pregnancy can be counseled to continue. Appropriate exercises are those that use upper body muscles and place little mechanical stress on the trunk region. Women should be taught to palpate their uterus during exercise and stop the exercise if they detect contractions. Until additional studies are completed, however, medical supervision is recommended for all pregnancies complicated by diabetes. Women with type 1 diabetes are vulnerable to exercise-related hypoglycemia and, therefore, should monitor their blood glucose levels closely and always have a readily available source of glucose. Women should be cautioned against becoming dehydrated, overheated, tachycardic, or dyspneic. Exercise should not be prescribed for patients with hypertension or an autonomic dysfunction with a diminished counter-regulatory response.
In regard to gestational diabetes, exercise has been recommended as an adjunct to nutrition therapy. Regular aerobic exercise has been shown to lower fasting and postprandial glucose concentrations. Several randomized controlled trials have demonstrated that upper extremity exercise for 20 minutes 3 times a week can lower blood glucose levels significantly in women with gestational diabetes. In addition, these trials found no significant increase in either maternal or neonatal complications related to exercise.
Insulin Therapy
Other than diet, insulin is the only therapy recommended to treat diabetes during pregnancy. The safety of all available oral agents has not been established in pregnancy and could lead to prolonged neonatal hypoglycemia. Therefore, they not recommended. However, a study in women with GDM has found glyburide to be a safe and clinically effective alternative to insulin.
The goal of insulin therapy is to achieve blood glucose levels that are nearly identical to those observed in healthy pregnant women. Therefore, multiple injections of insulin usually are required. Human insulin is the least immunogenic of all insulins and is recommended exclusively in pregnancy. Insulin requirements may change dramatically throughout the various stages of gestation. In the first trimester, the maternal insulin requirement is approximately 0.7 U/kg of body weight per day. This is increased in the third trimester to 1.0 U/kg per day.
Several different approaches to insulin administration can be used during pregnancy. The superiority of one regimen over another never has been fully demonstrated. The new rapid-acting insulin analogs, with peak hypoglycemic action 1 to 2 hours after injection, offer the potential for improved postprandial glucose control. However, their safety during pregnancy and their ability to improve perinatal outcomes has yet to be established. Studies in women with gestational diabetes have demonstrated that these agents do not cross the placenta.
The continuous subcutaneous insulin infusion pump is another treatment option which has been used successfully during pregnancy. Insulin therapy delivered by this mode more closely resembles that of physiologic insulin release. The pump delivers a continuous basal rate of insulin infusion, with pulse-dose increments before meals. In published studies, researchers have demonstrated that comparable glucose control and pregnancy outcomes can be achieved by both conventional insulin therapy and pump therapy. However, it has been reported that hemoglobin A1C (HbA1C) levels were significantly lower 1 year after delivery in women who elected to remain on pump therapy after delivery compared with women who had continued to use conventional insulin treatment. Other advantages of insulin pump therapy during pregnancy include more rapid and predictable insulin absorption, enhanced lifestyle flexibility, and simplified morning sickness management.
Insulin therapy should be initiated in all women with GDM who fail to maintain euglycemia with diet. Women typically are started on a dosage of 20 U of NPH and 10 U of regular insulin daily. Insulin dosage is adjusted according to blood glucose levels, and an evening injection is added if fasting hyperglycemia persists. Some investigators have advocated the use of prophylactic insulin in GDM to reduce the risk of macrosomia. However, the advantages of this therapy must be weighted against the disadvantages.
MONITORING METABOLIC STATUS
Self-monitoring of Blood Glucose
Self-blood glucose monitoring has become the mainstay of management for pregnancies complicated by diabetes mellitus. A variety of small, battery-powered blood glucose reflectance meters are available for home use. Accurate readings depend on performing the test correctly; however, most of the newest models are less technique dependent. Some models are extremely sophisticated, having memories that permit the storage of results with the date and time they were collected, whereas others can even be downloaded onto a personal computer. Each woman should be seen by a qualified nurse educator to ensure that her technique is accurate. Ongoing education also is important to help the woman make necessary changes in her treatment plan to maintain euglycemia throughout gestation.
Although it has been shown that, in general, self-monitored blood glucose values correlate very well with those measured in automated laboratories, reports have shown that sometimes patients report blood glucose levels falsely both during and outside of pregnancy. These findings are worrisome, because accurate information is essential for optimal management of the pregnancy complicated by diabetes. Therefore, verified blood glucose determinations (i.e., the use of blood glucose meters with memory) are recommended to enhance the reliability and accuracy of self-monitored blood glucose results.
Blood glucose measurements should be obtained at least 4 times a day (fasting and 1–2 hours after meals) in women with gestational diabetes and 5 to 7 times a day in women with preexisting diabetes (Table 15.3). In addition to this regular monitoring, patients also should test whenever they feel symptoms of either hyperglycemia or hypoglycemia. Detailed record keeping helps to identify glucose patterns. Daily urine ketone testing should be performed to ensure early identification of starvation ketosis or ketoacidosis. Ketone testing also should be performed anytime the blood glucose level exceeds 200 mg/dL, during illness, or when the patient is unable to eat.
|
|
|
TABLE 15.3. Monitoring the diabetic during pregnancy |
Blood glucose levels are measured in both the fasting and postprandial states. A randomized controlled trial compared the efficacy of preprandial and postprandial glucose determinations in reducing the incidence of neonatal macrosomia and other complications in women with gestational diabetes. Women requiring insulin treatment were randomly assigned to have their diabetes managed according to the results of preprandial self–blood glucose monitoring or postprandial monitoring performed 1 hour after meals. Both groups had similar success in achieving blood glucose level targets and demonstrated similar degrees of adherence with the monitoring schedule. Nevertheless, the women in the postprandial monitoring group received significantly more insulin and achieved a greater decrease in glycosylated hemoglobin values during treatment than those in the preprandial monitoring group. In addition, there was significantly less macrosomia and neonatal hypoglycemia among the offspring of the mothers in the postprandial monitoring group. These data suggest that adjustment of insulin therapy according to the results of postprandial blood glucose values improves glycemic control and pregnancy outcomes.
Previous studies in pregnant women with pregestational diabetes also have found that postprandial blood glucose levels are better predictors of fetal macrosomia than are fasting blood glucose levels. The Diabetes in Early Pregnancy Study demonstrated that third-trimester nonfasting glucose levels to be the strongest predictors of percentile birth weight in infants of diabetic mothers. Similar results have been reported by Combs and colleagues, who found that the incidence of macrosomia rose progressively with increasing postprandial glucose levels. Postprandial glucose levels of less than 130 mg/dL reduced the incidence of macrosomia, but levels of less than 120 mg/dL totally eliminated this complication. They recommended 130 mg/dL as a reasonable target for 1-hour postprandial glucose levels.
Although it is widely accepted that the level of metabolic control achieved in the pregnancy complicated by diabetes significantly affects perinatal outcome, what constitutes optimal control has not been established. Emerging evidence suggests that a continuum of risk exists between carbohydrate intolerance and both perinatal and neonatal morbidity. In theory, then, the target ranges for blood glucose values during pregnancy should be based on maternal plasma glucose levels in normal pregnancy. The logical approach is to achieve as near-to-normal glucose levels as possible without undue severe hypoglycemia. Current recommendations are that whole blood glucose levels should not exceed 95 mg/dL in the fasting state and 120 mg/dL after meals.
Although the current data demonstrate a relationship between metabolic control and neonatal complications, maternal glycemia may not be the sole parameter of optimal control. Buchanan and colleagues have suggested the use of fetal ultrasonography in women with gestational diabetes to identify pregnancies at risk for fetal macrosomia and related morbidity. They have found that during the late second trimester or early third trimester, a fetal abdominal circumference under or over the 75th percentile for gestational age can distinguish pregnancies at low risk from those at high risk, respectively, for producing large-for-gestational-age infants. Their data suggest that maternal glucose concentrations alone may not accurately predict which fetuses are at high risk for excessive fetal growth and support the use of fetal criteria to direct metabolic therapies in GDM.
Glycosylated Hemoglobin
The glycosylated hemoglobin assay is an accurate, objective measure of long-term glycemic control in diabetes. It can be measured either as total glycohemoglobin or as the particular configuration known as HbA1C. Proteins undergo a nonenzymatic, postsynthetic modification that results in the attachment of glucose to various amino acids. This process, which is known as glycosylation, occurs slowly and is irreversible. The amount of hemoglobin that becomes glycosylated depends on the concentration of glucose over the time of the exposure and, therefore, is not influenced significantly by recent or transient blood glucose excursions. Because erythrocytes in the peripheral circulation have a half-life of 100 days, the glycosylated hemoglobin level reflects glycemic control over the preceding 2 to 3 months. Most studies have found positive correlation between HbA1C values and other parameters of glucose control during gestation, similar to findings in the nonpregnant state.
Several investigators have reported an association between malformations and glucose control early in pregnancy. The higher the woman's glycosylated hemoglobin level, the greater is her risk of having a severely affected infant. The risk of miscarriage also has been shown to be increased with marked elevations in first-trimester glycosylated hemoglobin. Ideally, glycosylated hemoglobin levels should be measured before conception, and pregnancy should be delayed until normal levels are reached. Unfortunately, the majority of pregnancies in women with and without diabetes are not planned. Therefore, the HbA1Clevel can be obtained at the first prenatal visit and reassessed every 4 to 6 weeks to document the degree of glycemic control maintained throughout the pregnancy. Although elevations in first-trimester glycosylated hemoglobin levels indicate an increased risk for congenital malformations, an elevation during the second half of pregnancy appears to identify infants at risk for perinatal morbidity and mortality. Several investigators have demonstrated an association between HbA1C and neonatal hypoglycemia, neonatal hyperbilirubinemia, perinatal death, and macrosomia. In women with gestational diabetes, however, HbA1C levels have not been found useful as either a screening or diagnostic tool.
FETAL ASSESSMENT
All pregnancies complicated by diabetes require additional fetal evaluation and assessment (see Table 15.3). Ultrasonography provides the clinician with essential information about the fetus and its development. A first-trimester scan should be performed to date the pregnancy and establish viability. All women with pregestational diabetes should also be evaluated for possible fetal anomalies. Although the risk for delivering an anomalous infant cannot be fully determined by a glycosylated hemoglobin level, an elevated HbA1C should alert the practitioner to an increased risk for structural defects. Evaluation includes targeted ultrasonography to survey general fetal anatomy and fetal echocardiography at approximately 20 to 22 weeks gestation. In addition, the maternal serum α-fetoprotein screening test should be performed at 16 to 18 weeks gestation because of the increased risk of neural tube defects.
Because women with diabetes are at risk for fetal growth aberrations, frequent ultrasonographic scans are recommended to identify states of altered growth. Ultrasonographic examinations should be performed at 4- to 6-week intervals during the second and third trimesters of pregnancy to assess not only fetal growth, but amniotic fluid volume as well.
Fetal death is more common in pregnancies complicated by diabetes than in the general population. The goal of antepartum surveillance is avoidance of intrauterine death by early detection of fetal compromise. The nonstress test (NST) has become the preferred antepartum fetal heart rate test for screening the fetal condition in pregnancies complicated by diabetes. The NST evaluates the presence of accelerations from the baseline fetal heart rate. An alternative fetal test is the fetal biophysical profile (BPP), which also is used to evaluate the significance of a nonreactive NST. Because the BPP employs ultrasonography, it permits evaluation of amniotic fluid volume and may detect major fetal malformations in patients who have not been studied earlier in pregnancy. Maternal evaluation of fetal movement counts should be integrated into the surveillance program. Women with diabetes are instructed to count fetal movements beginning as early as 28 weeks of gestation. Although the false-positive rate is high, the technique is inexpensive and simple and augments the total antepartum surveillance program.
Most perinatal centers institute a program of weekly fetal monitoring beginning at 32 to 34 weeks gestation. Because fetal death is more common in women with poor glycemia control, hydramnios, fetal macrosomia, hypertension, or vasculopathy, these women receive twice weekly NST testing. Patients with diet-controlled gestational diabetes who maintain normal fasting and postprandial glucose levels are probably at low risk for an intrauterine fetal death and are not tested as early or as frequently. However, women with gestational diabetes and chronic hypertension, a previous stillbirth, preeclampsia, and who require insulin therapy receive more intense surveillance.
Timing and Mode of Delivery
It is generally accepted that if a pregnant diabetic patient is in good metabolic control and is receiving fetal surveillance on a regular basis, delivery may be delayed safely until term or the onset of spontaneous labor. Women with poor metabolic control, worsening hypertensive disorders, fetal macrosomia, growth retardation, or polyhydramnios may be delivered electively after fetal lung maturity has been confirmed. If an elective delivery is planned before 38 weeks gestation, amniocentesis should be performed for confirmation of fetal lung maturity. Fetal lung maturation is better predicted by the amniotic fluid phosphatidylglycerol content than by the lecithin/sphingomyelin ratio.
Cesarean delivery should be performed on most patients with an estimated fetal weight of greater than 4,500 grams to prevent shoulder dystocia and birth trauma. Management should be individualized for patients with an estimated fetal weight between 4,000 and 4,500 grams. The decision is based on the size of the pelvis and progress of labor, as well as the patient's obstetric history. A history of shoulder dystocia is often an indication for repeat caesarean section.
During labor and delivery, good blood glucose control should be maintained to prevent neonatal hypoglycemia. Blood glucose levels should be maintained below 100 mg/dL with the use of an insulin infusion. After delivery, insulin requirements tend to fall dramatically as a result of the significant decrease in levels of placental hormones. Once the patient is able to eat regular meals, she should receive subcutaneous insulin at approximately one half of her prepregnancy dose.
CONTRACEPTION AND FAMILY PLANNING
The use of contraception in all women with diabetes or a history of GDM cannot be emphasized strongly enough. This is the only way to ensure that preconception care can be provided. A variety of contraceptive methods are available. Barrier methods of contraception create mechanical or chemical barriers, or both, to fertilization and include diaphragms, male condoms, spermicidal foam or jelly, cervical caps, and female condoms.
Although these methods pose no health risks to women with diabetes, they are user dependent, require correct application or insertion before intercourse, and have a high failure rate of 12% to 28% during the first year because of improper use. With experience and motivation, these failure rates may be reduced to levels of 2% to 6%.
Oral contraceptives remain the most popular form of birth control despite controversy over potential side effects. The main reasons for their popularity are their failure rate of generally less than 1% and ease of use. Low-dose formulations are preferred and are recommended only for patients without vascular complications or additional risk factors, such as smoking or a strong family history of myocardial disease. Progestin-only (“mini-pill”) oral contraceptives are an option for women with contraindications to the estrogen component, such as hypertension or thrombosis.
An intrauterine device is the most effective nonhormonal device. It should be offered only to women with diabetes who have a low risk of sexually transmitted diseases, because any infection might place the patient with diabetes at risk for sepsis and ketoacidosis. Patient education should include the early signs of sexually transmitted diseases, such as increased and abnormal vaginal discharge, dyspareunia, heavy or painful menses, lower abdominal pain, and fever.
Depo-Provera is a long-acting progestin that provides highly effective pregnancy prevention. Depo-Provera is administered intramuscularly every 3 months and works by inhibiting ovulation. The high efficacy and long period of action of Depo-Provera makes it an attractive option for women with diabetes, especially women with a history of poor medication compliance. Unfortunately, this long-acting progestin has not been studied in women with diabetes.
Permanent sterilization, including tubal ligation and vasectomy, may be considered by the patient or her partner when they desire no more children.
CONCLUSION
The diagnosis of diabetes mellitus during pregnancy has certain implications for the well-being of both the mother and the fetus. Advances in medical and obstetric care have dramatically improved the outlook for women with diabetes and their offspring. However, both mother and child remain at increased risk for a number of complications. Research indicates that the majority of these complications are associated with hyperglycemia. The achievement and maintenance of euglycemia has, therefore, become the major focus of management.
SUMMARY POINTS
· The development of insulin resistance during late pregnancy is a normal physiologic adaptation that shifts maternal energy metabolism from carbohydrate to lipid oxidation and thus spares glucose for the growing fetus.
· All pregnant women should be screened for GDM between 24 and 28 weeks gestation.
· Women with diabetes and their offspring are at greater risk for a number of pregnancy-related complications.
· Strict blood glucose control prior to conception and throughout gestation can reduce and or eliminate the excess risk for both mother and baby.
· All diabetic women of childbearing age should be counseled regarding the importance of preconception care and planning their pregnancies.
RECOMMENDED READINGS
American College of Obstetricians and Gynecologists. Gestational diabetes. ACOG Practice Bulletin No. 30. Washington DC: ACOG, 2001.
Buchanan TA, Kjos SL, Montoro MN, et al. Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Diabetes Care 1994;17:275–283.
Chew EY, Mills JL, Metzger BE, et al. Metabolic control and progression of retinopathy. The diabetes in early pregnancy study. National Institute of Child Health and Human Development Diabetes in Early Pregnancy Study. Diabetes Care 1995;18:631–637.
Combs CA, Gunderson E, Kitzmiller JL, et al. Relationship of fetal macrosomia to maternal postprandial glucose control during pregnancy. Diabetes Care1992;15:1251–1257.
deVeciana M, Major CA, Morgan MA, et al. Postprandial versus preprandial blood glucose monitoring in women with gestational diabetes requiring insulin therapy. N Engl J Med 1995;333:1237–1241.
Engelgau MM, Herman WH, Smith PJ, et al. The epidemiology of diabetes and pregnancy in the U.S., 1988. Diabetes Care 1995;18:1029–1033.
Gabbe SG. New concepts and applications in the use of the insulin pump during pregnancy. J Matern Fetal Med 2000;9:42–45.
Homko CJ, Reece EA. Ambulatory management of the pregnant woman with diabetes. Obstet Gynecol Clin North Am 1998;41:584–596.
Homko CJ, Sivan E, Reece EA, et al. Fuel metabolism during pregnancy. Semin Reprod Endocrinol 1999;17:119–125.
Jovanovic-Peterson L, Peterson CM, Reed GF, et al. Maternal postprandial glucose levels and infant birth weight. The diabetes in early pregnancy study. Am J Obstet Gynecol 1991;164:103–111.
Jovanovic-Peterson L, Abrams RS, Coustan DR, et al., eds. Medical management of pregnancy complicated by diabetes. Alexandria, VA: American Diabetes Association, 1995.
Jovanovic-Peterson L, Peterson CM. Exercise and the nutritional management of diabetes during pregnancy. Obstet Gynecol Clin North Am 1996;23:75–86.
Khandelwal M, Homko C, Reece EA. Gestational diabetes mellitus: controversies and current opinions. Curr Opin Obstet Gynecol 1999;11:157–165.
Kjos SL. Contraception in women with diabetes mellitus. Obstet Gynecol Clin North Am 1996;23:243–258.
Langer O, Conway DL, Berkus M, et al. A comparison of glyburide and insulin in women with GDM. N Engl J Med 2000;343:1134–1138.
Langer O, Rodriguez DA, Xenakis MJ, et al. Intensified versus conventional management of gestational diabetes. Am J Obstet Gynecol 1994;170:1036–1047.
Leguizamon G, Reece EA. Effect of medical therapy on progressive nephropathy: Influence of pregnancy, diabetes and hypertension. J Matern Fetal Med2000;9:70–78.
Metzger BE, Coustan DR. Summary and recommendations of the Fourth International Workshop-Conference on gestational diabetes. Diabetes Care1998;21[Suppl 2]:B161–B167.
Reece EA, Homko CJ, Leguizamon G. Stringent controls in diabetic nephropathy associated with optimization of pregnancy outcomes. J Matern Fetal Med1998;7:213–216.
Reece EA, Coustan DR, eds. Diabetes mellitus in pregnancy, second ed. New York: Churchill-Livingston, 1995.
Reece EA, Homko CJ. Infant of the diabetic mother. Semin Perinatol 1994;18:459–469.
Reece EA, Homko CJ. Why do diabetic women deliver malformed infants? Obstet Gynecol Clin 2000;43:32–45.
Reece EA, Homko CJ. Diabetes mellitus in pregnancy: what are the best treatment options? Drug Safety 1998;18:209–220.
Sermer M, Naylor CD, Gare DJ, et al. Impact of increasing carbohydrate intolerance on maternal-fetal outcomes in 3637 women without gestation diabetes. The Toronto tri-hospital gestational diabetes project. Am J Obstet Gynecol 1995;173:146–156.
Weintrob N, Karp M, Hod M. Short- and long-range complications in offspring of diabetic mothers. J Diab Comp 1996;10:294–301.
