INTRODUCTION — The key elements in management of pregnancies complicated by pregestational (also called preexisting) diabetes are:

●Achieving and maintaining excellent glycemic control, with avoidance of hypoglycemia and diabetic ketoacidosis (DKA).

●Screening, monitoring, and intervention for nonglycemic maternal medical complications (eg, retinopathy, nephropathy, hypertension, cardiovascular disease, thyroid disease).

●Monitoring for fetal and obstetric complications (eg, congenital anomalies, preeclampsia, macrosomia, compromise of fetal well-being), with timely intervention to minimize adverse outcomes, when possible.

Most issues related to the obstetric management of a pregnant woman with diabetes (type 1 or type 2) will be reviewed here. The obstetric management of these pregnancies is largely based on clinical experience, data from observational studies, and expert opinion. There is very limited evidence from randomized trials.

Four important additional issues are discussed in detail separately: prepregnancy evaluation/counseling, glycemic control, maternal medical complications, and neonatal issues:

●(See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management".)

●(See "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control".)

●(See "Infants of women with diabetes".)

Gestational diabetes is also discussed separately:

●(See "Gestational diabetes mellitus: Screening, diagnosis, and prevention".)

●(See "Gestational diabetes mellitus: Glycemic control and maternal prognosis".)

●(See "Gestational diabetes mellitus: Obstetric issues and management".)

FIRST TRIMESTER

First prenatal visit — Ideally, women with pregestational diabetes have received preconception counseling to address maternal and fetal risks during pregnancy, worked with their providers to optimize glycemic control, and undergone screening for complications of diabetes (retinopathy, nephropathy, hypertension, neuropathy, cardiovascular disease, thyroid dysfunction), with management of any complications that were identified. (See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management".)

However, many pregnancies are unplanned and many women do not receive comprehensive management of their disease; thus, a prenatal visit may be the clinician's first opportunity to assess the patient's baseline medical status and counsel them about the management and potential complications of diabetes in pregnancy, as well as routine aspects of pregnancy care. (See "Prenatal care: Initial assessment".)

Classification of diabetes in pregnancy — The following classification system for diabetes in pregnancy, based on the mechanism of disease, has been proposed [1]:

●Type 1 diabetes (autoimmune beta cell destruction, usually leading to absolute insulin deficiency):

a. Without vascular complications.

b. With vascular complications (specify nephropathy, retinopathy, hypertension, arteriosclerotic heart disease, transplant, etc).

●Type 2 diabetes (progressive loss of insulin secretion, often in the setting of insulin resistance):

a. Without vascular complications.

b. With vascular complications (specify nephropathy, retinopathy, hypertension, arteriosclerotic heart disease, transplant, etc).

●Gestational diabetes (diabetes diagnosed during pregnancy and not clearly overt [eg, type 1 or type 2 diabetes]).

●Other diabetes (eg, genetic origin, drug or chemical induced).

Traditionally, the severity of pregestational diabetes was categorized according to the White classification (table 1) [2], which attempts to provide a standardized definition for describing pregnant women with diabetes and has some correlation with pregnancy outcome [3,4]. However, the White classes are not mutually exclusive; thus, some authorities have argued that the classification of diabetes should be reassessed [1]. We believe the presence/absence of vascular complications, as described above, is a better predictor of adverse outcome than the specific White class [5].

Evaluation

Routine — Routine prenatal laboratory evaluations are performed (table 2). Assessment for and treatment of asymptomatic bacteriuria is particularly important because there is a three- to fivefold greater propensity for asymptomatic bacteriuria in diabetic women. Management of pregnant patients with asymptomatic bacteriuria is reviewed separately. (See "Urinary tract infections and asymptomatic bacteriuria in pregnancy", section on 'Asymptomatic bacteriuria'.)

Rescreening at the start of the third trimester is reasonable in women who did not have bacteriuria on the initial test because they remain at high risk for developing bacteriuria.

Glycated hemoglobin — In addition to the standard prenatal laboratory panel, a glycated hemoglobin concentration (hemoglobin A1C) is obtained. In the absence of significant red blood cell abnormalities, A1C reflects the woman's average level of glycemic control over the prior few weeks to months and thus assists in counseling her regarding the risks of miscarriage, congenital malformations, preeclampsia, and other complications. There is usually a glycemia-independent decline of A1C during pregnancy [6], so the reliability of the A1C level to reflect early-pregnancy glycemia may depend on the gestational age of measurement. (See 'Risk of congenital anomalies/miscarriage' below and "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management", section on 'Miscarriage' and "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management", section on 'Preeclampsia and gestational hypertension'.)

Assessment of comorbidities — Additional tests that should be obtained early in pregnancy in women with pregestational diabetes, if not assessed preconceptionally, include [7]:

●Baseline renal function. A serum creatinine is obtained along with quantification of urinary proteinuria. This can be performed on a random urine sample using the urinary protein-to-creatinine ratio, which is both reproducible and more convenient for the patient than a 24-hour collection. However, many providers prefer to obtain a 24-hour collection for direct comparison to collections that may be obtained later in pregnancy for preeclampsia evaluation. (See "Evaluation of proteinuria in pregnancy and management of nephrotic syndrome".)

●Thyroid-stimulating hormone in patients with type 1 diabetes, as the incidence of thyroid dysfunction in type 1 diabetes is increased. If thyroid-stimulating hormone levels are elevated, thyroid peroxidase status should also be checked. (See "Associated autoimmune diseases in children and adolescents with type 1 diabetes mellitus", section on 'Thyroid screening'.)

●Electrocardiogram (ECG) in women with symptoms consistent with cardiovascular disease. Referral to a cardiologist should also be considered for symptomatic patients, regardless of the ECG result [8]. An ECG can be considered in other women with diabetes mellitus, particularly if not previously performed, though its value in women under age 40 without symptoms or other cardiovascular risk factors is unclear. (See "Screening for coronary heart disease in patients with diabetes mellitus".)

●Dilated, comprehensive eye examination by an ophthalmologist to detect retinopathy [8,9]. Close follow-up is indicated during pregnancy, with the frequency determined by baseline findings. The American Diabetes Association (ADA) suggests eye examinations every trimester and for one year postpartum, as indicated by degree of retinopathy and as recommended by the eye care provider [8]. (See "Diabetic retinopathy: Prevention and treatment", section on 'Pregnancy'.)

Ultrasound — Ultrasound examination is obtained for the usual obstetric indications (table 3).

First-trimester ultrasound examination is often obtained to document viability, as the rate of miscarriage is higher in women with diabetes, especially those with poor glycemic control, and to assist in estimation of gestational age. Accurate estimation of gestational age is critical since many of these pregnancies undergo scheduled delivery or have accelerated or, less commonly, restricted fetal growth in the late second and the third trimester. (See "Prenatal assessment of gestational age, date of delivery, and fetal weight".)

Some major congenital abnormalities (eg, anencephaly) can be detected in the late first trimester by detailed fetal anatomic survey using a transvaginal transducer. Sensitivity is lower earlier in gestation because of difficulty in visualizing small structures and because some abnormalities of organs, such as the gastrointestinal tract, brain, and kidney, can be visualized better in the more physiologically advanced fetus. In particular, the fetal heart, which is a common site of diabetic embryopathy, is optimally visualized in the second trimester at 18 to 22 weeks. (See "Congenital heart disease: Prenatal screening, diagnosis, and management".)

Screening for aneuploidy — Maternal diabetes mellitus is not a risk factor for aneuploidy. Women who have diabetes are offered prenatal screening and diagnosis for aneuploidy and other genetic conditions according to practices in use for the general obstetric population. First-trimester serum and ultrasound markers of aneuploidy are not affected by maternal diabetes. It is presumed that noninvasive screening using cell-free DNA is not affected by maternal diabetes, but this has not been specifically studied. (See "Down syndrome: Overview of prenatal screening".)

If second-trimester testing is performed (eg, quadruple test), then adjustments need to be made since serum alpha fetoprotein and unconjugated estriol (uE3) levels are reduced in women with diabetes. (See 'Screening for aneuploidy' below.)

Counseling and management

General principles — The clinician should emphasize the importance of meticulous glucose self-monitoring and attention to glycemic control, and the need for frequent prenatal visits and intensive fetal surveillance later in pregnancy. Information on diet, insulin therapy, and glucose monitoring should be provided by clinicians with experience in management of diabetes during pregnancy. A team approach is efficient and is usually required to provide the necessary expertise. In addition to the obstetric providers, the team may include an endocrinologist, certified diabetes educator, nutritionist, and the patient's primary care provider. (See "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control".)

Even in early pregnancy, women with pregestational diabetes are often seen more frequently than women with uncomplicated pregnancies. These extra visits can be used to review self-monitored glucose values and results from the ophthalmologic and laboratory examination (eg, renal function, glycated hemoglobin [hemoglobin A1C], thyroid function). In addition, virtual methods of communication can be used to exchange information about glucose values and adjustments in insulin therapy more frequently [10].

The first trimester is a high-risk time for hypoglycemia [11], particularly for women with type 1 diabetes. The contributing factors likely include efforts to obtain tight glycemic control, nausea and vomiting, and an increase in insulin sensitivity [12,13]. The ADA suggests loosening glycemic targets for patients experiencing significant hypoglycemia [8].

Management of comorbidities

Hypertension — Women receiving angiotensin converting enzyme inhibitors or receptor blockers for hypertension or nephropathy should have been taken off these medications prior to pregnancy because of their teratogenic potential. If not discontinued prior to pregnancy, use of these drugs should be suspended during pregnancy. (See "Adverse effects of angiotensin converting enzyme inhibitors and receptor blockers in pregnancy".)

In pregnant patients with diabetes and chronic hypertension, the ADA suggests treatment, with a blood pressure target of 110 to 135/85 mmHg to reduce the risk for accelerated hypertension during pregnancy and minimize the risk of impaired fetal growth [8], though evidence supporting specific blood pressure targets in pregnant women with diabetes is lacking.

The American College of Obstetricians and Gynecologists' (ACOG) guideline for management of chronic hypertension in pregnancy emphasizes the importance of antihypertensive therapy for women with blood pressures ≥160/105 mmHg and no evidence of end-organ damage and suggests maintaining blood pressure at a higher target (120 to 160 mmHg systolic and 80 to 105 mmHg diastolic) in women on medication; for women with end-organ damage, a group which would include many who have long-standing diabetes, they suggest maintaining blood pressure <140/90 mmHg to avoid progression of disease during pregnancy [14]. A detailed discussion of antihypertensive therapy for pregnant women with chronic hypertension can be found separately. (See "Chronic hypertension in pregnancy: Preconception, pregnancy, and postpartum issues and management".)

Other comorbidities

●Chronic kidney disease – Chronic kidney disease increases the risk of adverse maternal and fetal outcomes (table 4). The risk increases as the glomerular filtration rate declines and in the settings of proteinuria and hypertension. Ongoing evaluation and management of these patients are discussed separately. (See "Pregnancy in women with nondialysis chronic kidney disease".)

●Retinopathy – Pregnant women with diabetes should be examined by an ophthalmologist in the first trimester and closely followed until one year postpartum. Pregnancy, as well as the rapid improvement in glycemic control often associated with pregnancy, increases the risk for progression of retinopathy. The risk increases with the severity of retinopathy at baseline [9,15]. (See "Diabetic retinopathy: Prevention and treatment", section on 'Pregnancy'.)

●Obesity – Obesity is common in women with diabetes, particularly type 2 diabetes. Obesity is independently associated with many pregnancy risks and other health consequences. Excessive gestational weight gain amplifies these risks, contributes to postpartum weight retention, and should be avoided (table 5). (See "Obesity in pregnancy: Complications and maternal management" and "Overweight and obesity in adults: Health consequences" and "Gestational weight gain".)

A 2013 study of New York City maternal mortality data from 1995 to 2003 found that maternal obesity and pregestational diabetes were each associated with an increased risk of death during a delivery hospitalization (adjusted odds ratios [ORs] 2.9 and 3.3, respectively), but how concurrence of these two conditions influenced mortality risk was not reported [16]. A second study using New York City birth data argued that diabetes and obesity were independent risk factors for the outcomes of cesarean and preterm delivery [17], in agreement with other studies that showed obesity adds to the risk of pregnancy-associated maternal and neonatal morbidity in women with diabetes [18,19]. Clinicians caring for women with both obesity and diabetes should be mindful of the potential morbidities of both conditions.

●Neuropathy, gastroparesis, and cardiovascular disease – Treatment of these potential comorbidities, if present, is reviewed separately. (See "Management of diabetic neuropathy" and "Treatment of gastroparesis" and "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management", section on 'Maternal medical risks'.)

Prevention of preeclampsia — Pregestational diabetes is a risk factor for preeclampsia (pooled rate 11 percent, 95% CI 8.4-13.8 percent; pooled relative risk [RR] 3.7, 95% CI 3.1-4.3 [20]). The United States Preventive Services Task Force, ADA, and ACOG recommend that women at high risk for preeclampsia, including all those with type 1 and type 2 diabetes, begin low-dose aspirin after 12 weeks of gestation [8,21,22]. It is our practice to ensure that treatment has been initiated by 12 weeks of gestation [23]. Many providers use 81 mg daily as is recommended by ACOG and the Society for Maternal-Fetal Medicine. Doses of 100 to 162 mg daily are recommended by the ADA [8] based on the results of a meta-analysis suggesting that lower doses are not effective [8,24]. Of note, trials of aspirin for preeclampsia prevention have included few women with diabetes, so the effectiveness of this intervention is unclear in this population [25]. (See "Preeclampsia: Prevention", section on 'Low-dose aspirin'.)

Risk of congenital anomalies/miscarriage — Data from multiple studies have consistently shown a higher risk of major congenital malformations and miscarriage associated with increasing first-trimester A1C values (figure 1) [26-29]. Although A1C values from different laboratories may not have been comparable at the times these studies were conducted because of differences in methodology and a lack of standardization among laboratories, a value >1 percent above the upper limit of the normal range is associated with an increased risk of congenital anomalies. The relationship between A1C and congenital anomalies/miscarriage is discussed in detail separately. (See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management" and "Measurements of glycemic control in diabetes mellitus".)

We typically inform patients of the increased risk of congenital anomalies, including neural tube and cardiac defects [30], particularly if hemoglobin A1C is markedly elevated. We tell them that more information about fetal development will be obtained from first- and second-trimester sonographic examinations and maternal serum alpha-fetoprotein (MSAFP) results, if performed.

Women with diabetes should take at least 400 mcg of folic acid daily to reduce the risk of neural tube defects (NTDs). Higher doses up to 5 mg daily have sometimes been recommended. Ideally, folic acid supplementation is begun prior to conception. (See "Folic acid supplementation in pregnancy", section on 'Folic acid supplementation for prevention of neural tube defects'.)

Risk of macrosomia — The risk of macrosomia is related to the degree of glycemic control [31]. Pre-pregnancy obesity and gestational weight gain likely play a role as well [32,33]. All pregnant women should be counseled regarding the Institute of Medicine (IOM, now the National Academy of Medicine) weight gain recommendations (table 5), which are stratified and inversely related to starting body mass index. Appropriate gestational weight gain is especially important for those who have obesity (see 'Other comorbidities' above). Gestational weight gain above IOM recommendations increases the chances of large for gestational age (LGA) and macrosomic infants [34-38]. (See "Gestational weight gain", section on 'Recommendations for gestational weight gain' and "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control", section on 'Calorie requirements'.)

Pneumococcal vaccination — Adults with diabetes are at risk for pneumococcal disease and should receive the PPSV23 vaccine (see "Pneumococcal vaccination in adults"). If not previously done, some providers defer vaccination to the second trimester since more data about its safety are available. (See "Immunizations during pregnancy".)

SECOND TRIMESTER

General principles — Women are seen by the obstetric provider every two to four weeks through the second trimester, but more frequently if complications arise or glycemic control is suboptimal. This schedule of visits should be individualized based upon the severity of the diabetes, the degree of glycemic control, and the presence of other pregnancy complications. Some programs review blood glucose values remotely via phone or electronic communication and will make needed insulin adjustments outside of in-person visits.

Close follow-up for glycemic control is especially necessary after 18 weeks of gestation, as insulin resistance resulting from changes in placenta-produced hormones may increase rapidly and, consequently, insulin requirements can also increase quickly after this point [12].

Screening for aneuploidy — As discussed above (see 'Screening for aneuploidy' above), screening for aneuploidy is offered, according to routine obstetric practice. (See "Down syndrome: Overview of prenatal screening".)

Diabetes does not increase the risk of fetal aneuploidy. However, levels of maternal serum alpha-fetoprotein (MSAFP), unconjugated estriol (uE3), and inhibin A, which are components of some second-trimester Down syndrome screening tests, are significantly reduced in women with diabetes, thereby mimicking the pattern suggestive of Down syndrome. Therefore, multiples of the median (MoM) values should be adjusted in women with diabetes. (See "Laboratory issues related to maternal serum screening for Down syndrome", section on 'Previous false-positive result'.)

Screening for neural tube defects — The prevalence of neural tube defects (NTDs) is higher in women with pregestational diabetes mellitus. As an example, in a study from 1982 (before recommendations for folic acid supplementation and food fortification), NTDs occurred in 2 percent of pregnancies complicated by diabetes versus 0.1 to 0.2 percent of the general population [39]. In a study from 2004, NTDs occurred in 0.19 percent of pregnancies complicated by diabetes versus 0.07 percent of pregnancies in women without diabetes [40]. The lower prevalence in 2004 likely reflects trends in better periconceptional glucose control as well as increased periconceptional folic acid exposure.

We use ultrasound alone to screen for NTDs, but it may be used in combination with measurement of MSAFP. Since the median MSAFP level is 15 percent lower and the prevalence of NTDs is higher in women with diabetes than in those without diabetes, a lower threshold MSAFP value (eg, approximately 1.5 MoM) has typically been used in women with diabetes to obtain the same negative predictive value for NTDs as in women without diabetes. Laboratory requisitions for MSAFP typically ask providers to indicate if the patient has diabetes; however, the need for correction for diabetes independent of maternal weight has been challenged [41]. (See "Open neural tube defects: Risk factors, prenatal screening and diagnosis, and pregnancy management".)

Screening for other congenital anomalies — A detailed ultrasound examination of fetal anatomy is performed at approximately 18 weeks of gestation and by 22 weeks in pregnancies complicated by pregestational diabetes because of the increased prevalence of congenital anomalies in this group. If the sonologist performing the ultrasound is aware of the diagnosis of diabetes, they can be particularly mindful of evaluating for anomalies common to such pregnancies (table 6). Early detection of congenital anomalies allows parents and clinicians to prepare for the birth of an infant who may require specialized care. Alternatively, some parents may choose pregnancy termination; such procedures are more easily and safely undertaken at earlier gestational ages.

As noted above, fetuses of women with diabetes are at risk for NTDs, and evidence of such anomalies is often apparent on ultrasound. (See "Neural tube defects: Prenatal sonographic diagnosis".)

Another focus of this examination is cardiac anatomy, including a four-chamber view of the heart and visualization of the outflow tracts. Detailed examination of the fetal heart is important because congenital heart disease occurs more frequently in the offspring of women with diabetes than in the general population and accounts for approximately one-half of diabetes-related major congenital anomalies [30,42]. As an example, in a series of 535 pregnant women with preexisting diabetes, 30 (5.6 percent) delivered an infant with confirmed congenital heart disease; the risk was 8.3 percent in women with A1C ≥8.5 percent versus 3.9 percent of those with an A1C below this level [43]. Some centers refer all women with diabetes for fetal echocardiograms, while others restrict fetal echocardiography to fetuses with abnormalities on imaging the four chambers and outflow tracts or women with elevated periconception or first trimester A1C. A selective approach is acceptable because routine echocardiography has a low additional yield in centers with high-volume, skilled, comprehensive ultrasound services [44,45]. (See "Congenital heart disease: Prenatal screening, diagnosis, and management".)

Conotruncal and ventricular septal defects are the most common cardiac defects found in infants of women with diabetes. Significant augmentation of interventricular septal thickness may be noted in midtrimester fetuses of these women and often progresses during the course of pregnancy [46]. The hypertrophy primarily occurs in pregnancies with poor glycemic control. Although this condition is usually mild and asymptomatic in the neonate, congestive cardiomyopathy, which is a more diffuse process of hypertrophy and hyperplasia of the myocardial cells, can also occur. Both disorders are transient and managed with supportive care. (See "Infants of women with diabetes", section on 'Cardiomyopathy'.)

The performance of ultrasound for prenatal detection of congenital anomalies was illustrated in a retrospective study of pregnant women with pregestational diabetes who received detailed sonography with fetal echocardiography at 18 to 22 weeks in the authors' hospital system and subsequently delivered a liveborn or stillborn neonate at their hospital between 2011 and 2017 [47]. Major anomalies were present in 7 percent of newborns, of which 76 percent were detected prenatally; 68 percent of the prenatally detected anomalies were identified during the initial detailed fetal sonogram with echocardiography, and another 8 percent were identified on follow-up sonography. Prenatal detection by organ system was over 85 percent for central nervous system, genitourinary, and musculoskeletal anomalies; 65 percent for cardiac anomalies; and 43 percent for craniofacial anomalies.

THIRD TRIMESTER

General principles — In the third trimester, women with diabetes are seen frequently, as often as every one to two weeks until 36 weeks of gestation, and then weekly until delivery. The major concerns of the third trimester are:

●Continued close monitoring of maternal blood glucose levels

●Fetal testing and monitoring to minimize the risk of intrauterine fetal demise

●Monitoring for obstetric or medical complications necessitating premature delivery

●Evaluation for excessive or insufficient growth

Obstetric management consists of reinforcement of good glycemic control, electronic and sonographic fetal monitoring, estimation of fetal size, surveillance for pregnancy complications such as preeclampsia or polyhydramnios, and, in some cases, determination of fetal pulmonary maturity.

As in the late second trimester, frequent escalation of insulin doses is commonly required to maintain euglycemia. Insulin requirements generally plateau around 37 weeks of gestation. (See "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control".)

Assessment of fetal well-being — Intrauterine fetal demise is now a rare complication in pregnancies complicated by diabetes, primarily because the ability to achieve glycemic control has improved. The fetus of the mother with diabetes is at risk for hypoxia primarily from two mechanisms: (1) fetal hyperglycemia and hyperinsulinemia increase fetal oxygen consumption, which may induce fetal hypoxemia and acidosis if the oxygen needs of the fetus are not met [48-52], and (2) maternal vasculopathy and hyperglycemia can lead to reduced uteroplacental perfusion, which may be associated with reduced fetal growth [53].

The American College of Obstetricians and Gynecologists (ACOG) recommends antepartum fetal testing for pregnancies complicated by pregestational diabetes [22]. There are no data from large or randomized trials on which to make an evidence-based recommendation as to which pregnancies complicated by diabetes should undergo fetal surveillance, when to start, what test to order, or how often to perform it [54]. As a result, management is largely based on clinical experience and expert opinion and varies widely [55]. The most common techniques for antepartum monitoring are the biophysical profile, modified biophysical profile, or nonstress test, which should be performed at appropriate intervals (once or twice weekly). Testing is generally begun at 32 weeks of gestation, unless there is an indication for earlier testing [22]. A 2009 National Institutes of Health workshop recognized the limitations of available data and concluded that, in managing pregnancies in women with diabetes, it was "not clear which method [of antenatal testing], if any, is superior" [56].

Our approach — We begin antepartum surveillance at approximately 32 weeks of gestation, increasing the frequency of testing to two times per week from 36 weeks until delivery [57]. In complicated patients with fetal growth restriction, oligohydramnios, preeclampsia, or poorly controlled blood glucose concentrations, testing may start as early as 26 weeks of gestation and is performed more frequently. Any significant deterioration in maternal status or decline in insulin requirements necessitates reevaluation of the fetus. The frequency of fetal death (excluding fetuses with congenital malformations) with such protocols is approximately 3 per 1000 pregnancies in women with type 1 diabetes [58].

If nonreassuring fetal testing is related to a potentially reversible problem such as hyperglycemia or ketoacidosis, the fetus can often be resuscitated in utero by prompt correction of the mother's metabolic derangement. Pathologic fetal heart rate patterns will often revert to normal.

If nonreassuring fetal testing appears to be related to a nonreversible problem, the gestational age and the associated risk for sequelae of prematurity strongly influence management. In preterm fetuses, we try to delay delivery, at least long enough to treat the mother with betamethasone to accelerate fetal lung maturation (see 'Antenatal glucocorticoids' below). This can be done while monitoring the fetus more intensively or even continuously. The degree of compromise indicated by fetal surveillance is used to weigh the relative risks and benefits of delaying delivery.

Assessment of fetal growth — Evaluation of fetal growth is a particularly important component of third trimester obstetric care, as pregnancies complicated by maternal diabetes are commonly associated with accelerated growth [59], but are also at increased risk of impaired fetal growth.

We obtain an ultrasound examination at 28 to 32 weeks of gestation to assess fetal growth. If fetal growth is normal, we typically perform another sonogram at approximately 38 weeks of gestation to estimate fetal weight and assist with delivery plans. Alternatively, some obstetricians limit the 28- to 32-week ultrasound examination to diabetic patients with hypertension or nephropathy and wait until 38 weeks of gestation for evaluation of others.

Pregnancies with accelerated or insufficient fetal growth are discussed below.

Accelerated fetal growth — Accelerated growth is most common among women whose diabetes is marked by insulin resistance; high insulin requirements are associated with accelerated fetal growth even in euglycemic pregnancies [60]. If present, accelerated fetal growth in fetuses of pregnancies of women with diabetes often becomes apparent at 26 to 28 weeks of gestation, which is the rationale behind an early third-trimester ultrasound examination [42,61,62]. A possible explanation for this observation is that insulin is the major regulatory hormone of fetal growth and fetal insulin receptors are maximally expressed at 19 to 25 weeks of gestation. In spite of this physiology, ultrasound examination at 29 to 34 weeks is not highly predictive of high birth weight in these pregnancies [63].

The term "large for gestational age" (LGA) usually refers to a fetus or newborn that is greater than the 90^th centile for fetuses or infants of that gestational age (possibly including adjustments for fetal sex and ethnicity). At 40 weeks of gestation, the 90^th percentile for birth weight in the United States is approximately 4060 grams [64]. The term "macrosomia" refers to a fetus or infant that is greater than some defined weight regardless of gestational age, sex, or ethnicity. Various authors and professional organizations have defined macrosomia as greater than 4000, greater than 4250, and greater than 4500 grams. ACOG suggests a threshold of 4500 grams because maternal and infant morbidity increases sharply above this level [65].

Maternal diabetes mellitus may double the incidence of LGA infants; it also changes the anthropometric measurements of infants of diabetic mothers (IDMs) compared with offspring of women without diabetes [66]. Specifically, the chest-to-head and shoulder-to-head ratios are increased in IDMs [67].

LGA fetuses are at increased risk for a prolonged second stage of labor, shoulder dystocia, operative delivery, maternal and infant birth trauma, and perinatal death [68]. Maternal diabetes mellitus increases the likelihood of shoulder dystocia two- to sixfold compared with the population without diabetes [69] and increases the likelihood of dystocia-associated fetal morbidity, such as brachial plexus injury [70]. The correlation between shoulder dystocia and birth weight in gravida with and without diabetes is shown in the table (table 7) [71]. (See "Shoulder dystocia: Risk factors and planning delivery of high-risk pregnancies".)

Accelerated fetal growth is also associated with an increased risk of neonatal metabolic and physiologic disturbances. Continued control of blood glucose concentration during the third trimester is important to minimize the risk of these complications. (See "Infants of women with diabetes", section on 'Neonatal effects'.)

Although neonatal weight is an important predictor of neonatal morbidity and estimation of fetal weight at term is an important variable in delivery planning, there is no highly reliable method for identifying LGA fetuses before delivery [42,72,73]. In a review of studies of ultrasound for predicting estimated fetal weight (EFW) >4000 grams in women with diabetes, sensitivity ranged from 33 to 83 percent and specificity ranged from 77 to 98 percent [42]. (See "Fetal macrosomia", section on 'Patients with diabetes'.)

Given the limitations of fetal weight estimates, some investigators have used other measurements for predicting LGA and shoulder dystocia. LGA is most apparent in the liver and abdomen and occurs in approximately 88 percent of fetuses in whom the abdominal circumference and EFW both exceed the 90^th percentile [74]. Enlarged biparietal diameter and head circumference are less predictive of LGA than enlarged abdominal measurements. Fetal fat thickness or body habitus and a variety of equations (eg, chest minus biparietal diameter ≥1.4 cm) have also been used to predict LGA and risk of shoulder dystocia. These assessments have yielded sensitivities of 83 to 96 percent in pregnancies complicated by diabetes [75]. Although these assessments can be somewhat predictive of LGA and shoulder dystocia, many of the measurements are difficult to obtain and reproduce accurately, and these formulas have not been validated in large studies or at a variety of sites.

Growth restriction — Impaired growth is more common among women with diabetic vasculopathy and/or superimposed preeclampsia. It is associated with increased fetal and neonatal morbidity and mortality, and has long-term health implications. (See "Infants with fetal (intrauterine) growth restriction".)

If there is evidence of fetal growth restriction, which is uncommon but often related to preeclampsia or preexisting maternal vasculopathy, tests of fetal well-being are initiated as in any pregnancy complicated by fetal growth restriction. (See "Fetal growth restriction: Evaluation and management", section on 'Pregnancy management'.)

Preeclampsia — The incidences of hypertension and preeclampsia are increased in pregnant women with diabetes and are related to pregestational hypertension and vascular and renal disease. Poor glycemic control also appears to play a role [76].

●In a series of 462 women with pregestational diabetes, the rate of preeclampsia in women with White classification B, C, D, and F/R (table 1) was 11, 22, 21, and 36 percent, respectively [77].

●In another study, the risk of preeclampsia increased significantly with increasing A1C values above optimal levels [76]. Compared with women with A1C <6.1 percent at 26 weeks of gestation, the odds of preeclampsia for women with A1C 6.1 to 6.9 percent, 7.0 to 7.9 percent, and ≥8.0 percent were 2.1, 3.2, and 3.8, respectively. At 34 weeks of gestation, the odds of preeclampsia with A1C values ≥7.0 percent and ≥8.0 percent were 3.3 and 8.0, respectively.

The increased risk of preeclampsia is concordant with the observation that insulin resistance appears to increase the risk of developing preeclampsia, even in the absence of overt diabetes [78,79]. Impaired endothelium-dependent vasodilation appears to be related to the duration of diabetes [80].

Diagnosis and management of preeclampsia are similar to that in women without diabetes, except among those who enter pregnancy with preexisting nephropathy. In these women, diagnosing preeclampsia can be difficult and requires relying on deterioration of other markers. (See "Hypertensive disorders in pregnancy: Approach to differential diagnosis" and "Preeclampsia: Management and prognosis".)

Polyhydramnios — Maternal diabetes is a common etiology of polyhydramnios, although the mechanism for the increased amniotic fluid volume has not been clearly defined. Possibilities include fetal polyuria secondary to maternal and fetal hyperglycemia, decreased fetal swallowing, or an imbalance in water movement between the maternal and fetal compartments [81]. Polyhydramnios is frequently associated with accelerated fetal growth. (See "Physiology of amniotic fluid volume regulation".)

Fetal outcomes in pregnancies with diabetes-associated polyhydramnios may not be as poor as outcomes in pregnancies in which polyhydramnios is associated with fetal neurologic disease, twin-to-twin transfusion, or other syndromes [82]. Diabetes-associated polyhydramnios rarely requires intervention for severe maternal symptoms. (See "Polyhydramnios: Etiology, diagnosis, and management".)

Preterm labor — Compared with controls without diabetes or hypertension, women with pregestational diabetes have higher rates of both indicated preterm delivery (22 versus 3 percent, odds ratio [OR] 8.1, 95% CI 6.0-10.9) and spontaneous preterm delivery (16 versus 11 percent, OR 1.6, 95% CI 1.2-2.2) [83]. Indicated preterm delivery is primarily initiated because of preeclampsia [83,84], but both gestational and pregestational diabetes have been associated with indicated preterm delivery independent of preeclampsia. The reasons for an increased risk of spontaneous preterm delivery are not clear [85,86].

The indications for inhibition of preterm labor are similar to those in the general obstetric population. Our preferences for tocolytic therapy are nifedipine or indomethacin (for pregnancies less than 32 weeks of gestation). We avoid beta-adrenergic receptor agonist therapy as these drugs can cause severe hyperglycemia in women with diabetes. (See "Inhibition of acute preterm labor".)

Antenatal glucocorticoids — If preterm birth between 22+0 and 33+6 weeks of gestation is anticipated or planned, administration of betamethasone improves neonatal outcome. Administration of betamethasone to reduce neonatal complications associated with preterm birth should be done cautiously. Transient hyperglycemia induced by glucocorticoids can be severe in women with diabetes, even when glucose levels are closely monitored and treated [87,88]. The hyperglycemic effect begins approximately 12 hours after the first steroid dose and lasts for approximately five days [89,90]. This was illustrated in a series in which 16 women with type 1 diabetes requiring insulin therapy were given betamethasone for fetal lung maturation [89]. Their daily insulin dose for the following five days increased by 6, 38, 36, 27, and 17 percent above baseline, respectively. It is our clinical experience that patients with insulin resistance or obesity may experience greater increases in insulin requirement. Although administration of betamethasone also had potential benefits before preterm births between 34+0 and 36+6 weeks of gestation in a randomized trial (Antenatal Betamethasone for Women at Risk for Late Preterm Delivery [91]), women with diabetes were specifically excluded from this trial, and, concordant with recommendations from specialty societies [92], we recommend not administering steroids to improve neonatal outcome in diabetic pregnancies at late preterm gestational ages.

We preemptively increase subcutaneous insulin doses in women receiving betamethasone, starting with the algorithm suggested by Mathiesen, et al [89] and using additional correctional subcutaneous insulin along with rapid titration as needed. Capillary blood glucose concentrations are checked before and after meals and at bedtime. A continuous intravenous insulin infusion is initiated on the labor unit if values continue to rise despite such treatment or if values are above 180 to 200 mg/dL (10.0 to 11.1 mmol/L). (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

DELIVERY — A number of issues arise peripartum, such as assessment of fetal maturity, timing and route of delivery, and risk of birth trauma from macrosomia.

Fetal pulmonary maturity — Planned delivery before 39+0 weeks is increasingly discouraged as the morbidities associated with early-term (37+0 to 38+6 weeks) and late preterm (34+0 to 36+6 weeks) deliveries have become evident [93,94]. An amniocentesis demonstrating fetal lung maturity does not obviate all such morbidity [95], and fetal lung maturity testing is no longer performed.

When a medically/obstetrically indicated early delivery is being considered, it is important to remember that respiratory distress syndrome (RDS) is more likely to develop in infants of women with diabetes delivered early than in infants of women without diabetes delivered early; this risk does not become equivalent in the two groups until after 38.5 weeks of gestation [96]. The endocrine changes associated with maternal diabetes delay fetal lung maturation [97]. Specifically, high fetal insulin levels enhance cellular hypertrophy and hyperplasia at the expense of cellular maturation, thus leading to macrosomia and immature lung function. In the era prior to the availability of fetal pulmonary maturity tests, RDS accounted for 52 percent of neonatal deaths among infants born to women with pregestational diabetes [98].

We evaluate timing of delivery based on clinical circumstances and maternal and fetal condition, mindful of the risk of pulmonary morbidity among preterm and early-term newborns of women with diabetes. Although RDS occurs rarely at or after 39 weeks in patients with poor glycemic control, the risk of in utero death in this setting probably far exceeds the risk of severe neonatal respiratory morbidity or mortality.

Timing of delivery — We and others [99] see little benefit in continuing pregnancy beyond 39 weeks in women with diabetes and thus suggest induction of labor for these pregnancies by 40+0 weeks of gestation. Preterm delivery should be avoided, except when glycemic control is suboptimal or there are other maternal or fetal reasons for concern (eg, maternal vascular disease). In these cases, an acceptable approach is to induce labor at 36+0 to 38+6 weeks (or earlier depending on the specific clinical setting). When a plan for induction is chosen, the risks of a prolonged induction due to an unfavorable cervix must be weighed against the risks associated with continuing the pregnancy.

The American College of Obstetricians and Gynecologists (ACOG) and the Society for Maternal-Fetal Medicine have published similar recommendations: deliver at 39+0 to 39+6 weeks if well-controlled glucose levels and no vascular disease; deliver at 36+0 to 38+6 weeks if poorly controlled glucose levels or vascular disease (even earlier if severity of complications warrants earlier delivery); expectant management beyond 40+0 weeks is not recommended [22,100,101].

Preterm delivery is performed for the usual obstetric indications (eg, preeclampsia, fetal growth restriction, abruption, premature labor with or without premature rupture of membranes, nonreassuring fetal testing) or for worsening maternal renal insufficiency or active proliferative retinopathy. Decisions to undertake a preterm delivery only because of maternal diabetes are balanced against morbidity associated with delivery at an early gestational age. In past eras, elective early delivery of women with pregestational diabetes had been advocated to prevent fetal death in late gestation [102-104]. This was a reasonable approach prior to 1950 since one-half of stillbirths in this population occurred after the 38^th week of gestation [105]. However, fetal mortality has fallen precipitously among both diabetic women and the general obstetric population over the past few decades; thus, for most patients, the morbidity and mortality from prematurity and failed induction should be weighed carefully against contemporary estimates of potential benefit from early delivery [106]. Currently, it is unclear whether women with good glycemic control and reassuring antepartum surveillance are at any increased risk of an intrauterine demise at term. Although there are case reports of fetal deaths occurring within hours of reassuring fetal testing in gravidas with diabetes (and in nondiabetic women, as well), despite maternal euglycemia [107], there are insufficient data from large well-designed studies to convincingly demonstrate that the risk of fetal demise with modern, intensive perinatal care is increased compared with controls without diabetes.

The only randomized trial evaluating the timing of delivery of 200 women with uncomplicated insulin-requiring diabetes (13 pregestational, 187 gestational) and appropriately grown fetuses showed induction during the 38^th week was advantageous compared with expectant management [108]. Women randomly assigned to active induction of labor within five days of reaching 38 weeks of gestation had a lower prevalence of macrosomia compared with those managed expectantly with biweekly nonstress tests and amniotic fluid volume assessment until 42 weeks (10 versus 23 percent) and fewer cases of shoulder dystocia (0 versus 3 percent). The rate of cesarean delivery was similar for the two groups. Moreover, 50 percent of women in the expectant management group ultimately required induction for obstetric indications. This study generally supports our approach, but more data are needed to reliably assess the risks and benefits of induction before 39 weeks of gestation.

Is early induction recommended when macrosomia is suspected? — Data concerning induction of labor for suspected macrosomia suggest limited benefits. In a large trial that randomly assigned women with singleton fetuses whose estimated weight exceeded the 95^th percentile to receive induction of labor within three days between 37+0 weeks and 38+6 weeks of gestation or expectant management, induction resulted in a slightly higher likelihood of vaginal delivery (relative risk [RR] 1.14, 95% CI 1.01-1.29) and a lower risk of significant shoulder dystocia (RR 0.32, 95% CI 0.12-0.85) [109]. There were two fractures in the induction group compared with eight in the expectant management group, and this difference was not statistically significant; no brachial plexus injuries occurred in either group. The proportions of infants with a high bilirubin concentration and receiving phototherapy were significantly higher in the induction group. Importantly, this trial specifically excluded patients with insulin-requiring diabetes. Moreover, recognizing the risk of respiratory morbidity in early-term neonates, contemporary guidelines urge avoiding elective delivery before 39 weeks of gestation in the absence of concerns about fetal or maternal well-being.

Given such guidance and the absence of high-quality data establishing a clear benefit of induction for imminent or suspected macrosomia, our practice is to use evaluation of glycemic control and standard maternal and fetal concerns to inform timing of delivery <39 weeks rather than estimated fetal weight (EFW) alone, and, as discussed above (see 'Timing of delivery' above), we favor delivery at 39 weeks for women with well-controlled diabetes mellitus.

A detailed discussion of the evidence related to this issue is available separately. (See "Shoulder dystocia: Risk factors and planning delivery of high-risk pregnancies", section on 'Planning delivery in high-risk pregnancies'.)

Route of delivery — Maternal diabetes alone is not an indication for cesarean birth in the absence of the usual obstetric indications. Macrosomia may be considered an indication for cesarean delivery due to the risk of morbidity from shoulder dystocia [110-113]. It has been suggested that neonates with shoulder dystocia have greater shoulder and chest-to-head disproportion than macrosomic infants without this complication [67,114]. In particular, macrosomic infants of mothers with diabetes are more likely to exhibit this disproportion than infants of nondiabetic mothers of comparable weight [112]. (See 'Accelerated fetal growth' above.)

For these reasons, the position of ACOG is that, although the diagnosis of macrosomia is imprecise, prophylactic cesarean delivery is reasonable to prevent brachial plexus injury when the EFW is greater than 4500 grams in a woman with diabetes [22,115]. If the patient has had a previous child with shoulder dystocia, then EFW, gestational age, and the severity of the prior neonatal injury, if any, should also be considered in making the decision about route of delivery [116]. (See "Shoulder dystocia: Risk factors and planning delivery of high-risk pregnancies".)

Since assisted vaginal delivery is associated with an additional risk for shoulder dystocia, a lower weight threshold (eg, 4000 g) may be used when making a decision to perform an operative vaginal delivery in a woman with diabetes [117].

Maternal diabetes is not a contraindication to a trial of labor after a previous cesarean delivery (TOLAC); however, the success rate may be lower than in women without diabetes (64 versus 74 percent in one study [118]). (See "Choosing the route of delivery after cesarean birth".)

Labor and delivery — The woman with diabetes and her fetus are continuously monitored on the labor and delivery unit, as these pregnancies are at increased risk for nonreassuring fetal heart rate patterns [119,120]. Peripartum maintenance of maternal euglycemia is essential and generally requires hourly capillary glucose determinations, intravenous solutions containing glucose, and intravenous insulin infusion if hyperglycemia is present. Management of glucose and insulin during labor, induction, and cesarean delivery is discussed separately. (See "Pregestational (preexisting) and gestational diabetes: Intrapartum and postpartum glycemic control".)

We schedule induction or cesarean delivery early in the morning, when possible, as this facilitates management of glucose and insulin in a fasting patient.

There are no contraindications to natural childbirth, epidural anesthesia, or general anesthesia. However, maternal hypotension is more commonly associated with spinal than epidural anesthesia and may be associated with a lower pH and a greater base deficit in the infant of a diabetic mother [121]. (See "Umbilical cord blood acid-base analysis at delivery".)

POSTPARTUM — Insulin requirements drop sharply after delivery and should be recalculated at this time based on serial blood glucose determinations. Immediate postpartum insulin requirements are substantially lower than requirements prior to pregnancy [122]. Postpartum calorie requirements are approximately 25 kcal/kg per day, but somewhat higher (27 kcal/kg per day) in lactating women. (See "Pregestational (preexisting) and gestational diabetes: Intrapartum and postpartum glycemic control".)

Breastfeeding should be encouraged [22] (see "Infant benefits of breastfeeding"). An additional 500 kcal per day above the prepregnancy baseline is recommended to meet the metabolic demands of breastfeeding. Women with diabetes appear to have more problems, and may benefit from consultation with a lactation specialist (see "Common problems of breastfeeding and weaning"). Women with type 1 diabetes who breastfeed may have a basal insulin requirement that is approximately 15 percent lower than prior to pregnancy through at least six months after delivery [123].

Postpartum depression is more common among women with diabetes (pregestational or gestational) than in nondiabetic women [124], so screening is warranted. (See "Postpartum unipolar major depression: Epidemiology, clinical features, assessment, and diagnosis".)

The United States Medical Eligibility Criteria for Contraceptive Use consider all hormonal methods acceptable for women with diabetes and no vascular disease [125]; thus, selection should be based upon the same factors that apply to women without diabetes [8] (see "Contraception: Counseling and selection"). Although evidence from randomized trials is limited, both progestin-only methods and low-dose combined oral contraceptives appear to have only minor effects on glucose and fat metabolism [126]. Depot medroxyprogesterone acetate and combined estrogen-progestin contraceptives are generally avoided in women with vascular disease [125]. The progestin-releasing intrauterine device (IUD), copper IUD, and etonogestrel implant have lower risk of thromboembolic events than estrogen-progestin contraceptives [127].

Ophthalmologic follow-up during the first year postpartum is advised since retinopathy can be aggravated anytime during pregnancy or postpartum [8].

DIABETIC KETOACIDOSIS — Physiologic changes and pathological conditions related to pregnancy predispose women with diabetes to worsening glycemic control. Despite this, diabetic ketoacidosis (DKA) occurs in only an approximate 0.5 to 3 percent of diabetic pregnant women [128]. (See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management" and "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control".)

DKA results from absolute or relative insulin deficiency combined with counterregulatory hormone excesses (ie, glucagon, glucocorticoids, catecholamines, and growth hormone). Pregnancy is a state of relative insulin resistance, which can be exacerbated by systemic infection and insulin pump technical failure. Counterregulatory hormone excesses can result from betamimetic tocolytic therapy and betamethasone administered to accelerate fetal lung maturity. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis", section on 'Precipitating factors'.)

Clinical presentation — The presentation of DKA is similar in pregnant women to that in nonpregnant persons, with symptoms of nausea, vomiting, thirst, polyuria, polydipsia, abdominal pain, and, when severe, a change in mental status. Laboratory findings include hyperglycemia (usually >250 mg/dL [13.9 mmol/L]), acidemia (arterial pH <7.30), an elevated anion gap (>12 mEq/L), ketonemia, low serum bicarbonate (<15 mEq/L), elevated base deficit (>4 mEq/L), and renal dysfunction [129]. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis", section on 'Clinical presentation' and "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis", section on 'Laboratory findings'.)

Hyperglycemia is usually severe in nonpregnant persons, however, DKA is well documented to occur at much lower blood glucose levels during pregnancy. In one series, 4 of 11 pregnant women with DKA had blood glucose levels less than 200 mg/dL (11.1 mmol/L) [130]. Severe hyperglycemia can cause an osmotic diuresis resulting in maternal volume depletion. This, in turn, can result in reduced uterine perfusion and, in association with the metabolic abnormalities of DKA, produce life-threatening fetal hypoxemia and acidosis. Maternal mortality is less than 1 percent, but fetal mortality rates of 9 to 36 percent have been reported, as well as increased risks of preterm birth [128]. Thus, DKA is a true obstetric emergency.

During acute DKA, the fetal heart rate often has minimal or absent variability and absent accelerations, as well as repetitive decelerations [128]. These abnormalities usually resolve with resolution of DKA, but it may take several hours before the tracing is normal [131].

Management — Other than close attention to fetal heart rate monitoring, the principles of DKA management are similar in pregnant and nonpregnant patients [22,128]. These include the use of intravenous insulin, appropriate volume replacement, correction of electrolyte abnormalities (including potassium, phosphate, and magnesium), monitoring acidosis, and a search for precipitating causes, such as infection [22]. Glucose targets during DKA treatment are lower in pregnant women (100 to 150 mg/dL) than in nonpregnant adults (150 to 250 mg/dL) [22]. Glucocorticoids and beta-mimetics should be avoided during DKA, as they will worsen hyperglycemia. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Treatment".)

DKA alone is generally not an indication for delivery. Emergency delivery before maternal stabilization should be avoided because it increases the risk of maternal morbidity and mortality and may result in delivery of a hypoxic, acidotic preterm infant for whom in utero resuscitation may have resulted in a better outcome. The timing of delivery needs to be individualized based on multiple factors, including gestational age, maternal condition (whether the mother is responding to aggressive therapy or deteriorating), and fetal condition (whether the fetal heart rate pattern is improving or deteriorating). Fetal heart rate abnormalities resulting from maternal acidosis will often improve as DKA is treated and maternal condition improves [128].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diabetes mellitus in pregnancy".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Care during pregnancy for women with type 1 or 2 diabetes (Beyond the Basics)" and "Patient education: Gestational diabetes (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Ideally, women with pregestational diabetes will have received preconceptional counseling to assess their baseline medical status and educate them about the management and potential complications of diabetes in pregnancy. (See 'First prenatal visit' above.)

●In addition to routine prenatal testing, assessment of the diabetic gravida should include: glycated hemoglobin concentration (hemoglobin A1C), baseline renal function, thyrotropin and free thyroxine, electrocardiogram (ECG), dilated and comprehensive eye examination by an ophthalmologist, and first-trimester ultrasound examination if pregnancy dating is uncertain. (See 'Evaluation' above.)

Laboratory monitoring across the remainder of pregnancy is described in the table (table 8).

●The care of women with diabetes during pregnancy generally requires a team approach to provide the necessary expertise. Information on diet, insulin therapy, exercise and glucose monitoring should be provided by clinicians with experience in management of diabetes during pregnancy. (See 'Counseling and management' above.)

●A markedly elevated glycohemoglobin value in the first trimester is associated with increased risks of both first-trimester miscarriage and congenital malformations. We typically advise patients of these risks and evaluate fetal development via second-trimester sonographic examination and maternal serum multiple marker screening. (See 'Risk of congenital anomalies/miscarriage' above.)

●Screening for Down syndrome and neural tube defects (NTDs) are offered, according to routine obstetric practice. Diabetes does not increase the risk of fetal aneuploidy, but it does affect interpretation of the analyte panels. The risk of NTDs is increased for fetuses of diabetic gravidae. (See 'Screening for aneuploidy' above and 'Screening for neural tube defects' above.)

●We suggest ultrasound examination and fetal echocardiogram at approximately 18 weeks of gestation to evaluate for congenital anomalies, particularly congenital heart disease. (See 'Screening for other congenital anomalies' above.)

●We begin antepartum surveillance with weekly nonstress tests at 32 weeks of gestation, increasing the frequency of testing to two times per week from 36 weeks until delivery. (See 'Assessment of fetal well-being' above.)

●We obtain an ultrasound examination at approximately 38 weeks of gestation to estimate fetal weight and assist with delivery plans. (See 'Assessment of fetal growth' above.)

●We suggest nifedipine or indomethacin for tocolysis of preterm labor instead of a beta-adrenergic receptor agonist (Grade 2C). (See 'Preterm labor' above.)

●If antenatal betamethasone is administered to accelerate fetal lung maturation between 23 and 34 weeks of gestation, we closely monitor capillary blood glucose concentrations (eg, every one to four hours depending on glucose levels and difficulty in obtaining control) beginning 12 hours after the first dose of betamethasone and continuing for 24 hours after the second dose. We administer insulin intravenously as needed to maintain euglycemia. (See 'Antenatal glucocorticoids' above.)

●If the expected fetal weight is over 4500 grams, we suggest cesarean delivery to avoid possible trauma from shoulder dystocia (Grade 2B). (See 'Route of delivery' above.)

●We suggest induction of labor at 39 to 40 weeks of gestation in women with favorable cervices and fetuses less than 4500 grams (Grade 2B). The presence of high-risk factors, such as poor glucose control, worsening nephropathy or retinopathy, preeclampsia, or restricted fetal growth, warrant consideration of earlier delivery. Awaiting the spontaneous onset of labor is reasonable if there is good glycemic control and no pregnancy or additional maternal complications. However, extending pregnancy beyond 40 weeks of gestation is generally not advised unless the patient has gestational diabetes with excellent glucose control with dietary modification alone. If induction of an unfavorable cervix is planned, use of cervical ripening agents is safe and effective. (See 'Timing of delivery' above.)

●We suggest not performing induction <39 weeks for the indication of suspected fetal macrosomia alone (Grade 2C). Glycemic control and standard maternal and fetal indications for induction should inform timing of delivery <39 weeks rather than estimated fetal weight (EFW) alone. (See 'Is early induction recommended when macrosomia is suspected?' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Michael F Greene, MD, who contributed to an earlier version of this topic review.