INTRODUCTION — Pregnancy is accompanied by insulin resistance, mediated primarily by placental secretion of diabetogenic hormones including growth hormone, corticotropin-releasing hormone, placental lactogen (chorionic somatomammotropin), prolactin, and progesterone. These and other metabolic changes, which are generally most prominent in the third trimester, ensure that the fetus has an ample supply of nutrients.
Gestational diabetes mellitus (GDM) develops in pregnant people whose pancreatic function is insufficient to overcome the insulin resistance associated with the pregnant state. Among the main consequences of GDM are increased risks of preeclampsia, large for gestational age (LGA) newborns, and cesarean birth, and their associated morbidities. Patients with GDM are at high risk of developing type 2 diabetes later in life, which is not surprising since the pathophysiologic path of inadequate insulin secretion in the setting of insulin resistance during pregnancy also underlies type 2 diabetes outside of pregnancy.
There is no universally accepted standard regarding screening for or diagnosis of GDM. Practitioners tend to follow the guidance of their local medical organizations. Approaches to screening and diagnosis will be reviewed here. Management and prognosis are discussed separately:
●(See "Gestational diabetes mellitus: Glucose management and maternal prognosis".)
●(See "Gestational diabetes mellitus: Obstetric issues and management".)
TERMINOLOGY
●Preexisting or pregestational diabetes refers to type 1 or 2 diabetes diagnosed before conception.
●Gestational diabetes traditionally referred to any pregnant person in whom abnormal glucose tolerance was first recognized at any time during pregnancy [1]. A more contemporary definition is diabetes diagnosed in the second or third trimester of pregnancy that was not clearly present prior to gestation [2]. This more contemporary definition does not account for patients diagnosed in the first trimester, who likely have previously undiagnosed type 2 diabetes. The term "overt diabetes" is sometimes used to describe the diabetes status of these individuals during pregnancy. When evaluated again in the nonpregnant state, a formal diagnosis of type 2 diabetes can be made.
BACKGROUND
Prevalence — The prevalence of GDM by the traditional Carpenter and Coustan criteria (table 3B) [3] was 7.8 percent of births in the United States (in 2020) [4] and 2 to 38 percent of pregnant people in other countries [5]. Worldwide prevalence varies because of differences in population characteristics (eg, average maternal age and body mass index [BMI]). Using the 2010 International Association of Diabetes and Pregnancy Study Groups (IADPSG) screening and diagnostic criteria for diabetes in pregnancy (table 3C) [6], the global prevalence of GDM has been estimated to be 17 percent, with regional estimates of 10 percent in North America and 25 percent in Southeast Asia [7].
Prevalence has been increasing over time, likely due to increases in mean maternal age and BMI, particularly increasing obesity [8-14].
Significance — GDM has been associated with increased risks of several adverse outcomes. Although treatment of GDM can reduce the risk of some short-term outcomes (eg, preeclampsia, macrosomia), a favorable effect on long-term offspring outcomes is unclear. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Rationale for treatment'.)
●Short-term – GDM has been associated with increased risks of [15,16]
•Hypertensive disorders of pregnancy (eg, preeclampsia, gestational hypertension and the potential consequence of preterm birth)
•Large for gestational age (LGA) or macrosomic newborn (and the potential consequences: shoulder dystocia, maternal and/or newborn birth trauma)
•Operative delivery (cesarean, assisted vaginal)
•Perinatal mortality
•Fetal/neonatal hypertrophic cardiomyopathy
•Neonatal respiratory problems and metabolic complications (eg, hypoglycemia, hyperbilirubinemia, hypocalcemia, polycythemia)
•Polyhydramnios
Importantly, the risks of these outcomes increase as maternal fasting plasma glucose levels increase above 75 mg/dL (4.2 mmol/L) and as the one- and two-hour oral glucose tolerance test (GTT) values increase from the lowest septile to the highest. This is a continuous effect; there is no clear threshold that defines patients at increased risk of adverse obstetric outcome [17-19].
A small increase in congenital anomalies has been observed in population-based studies of GDM (OR 1.18 [20]) and is likely related to undiagnosed preexisting type 2 diabetes mellitus or maternal obesity [21,22].
●Long-term – GDM has also been associated with increased risks of:
•In mothers: future development of diabetes mellitus (type 2 primarily, also type 1), metabolic syndrome, and cardiovascular disease. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Long-term risk'.)
•In adolescent and adult offspring: obesity, abnormal glucose tolerance, hypertension, and metabolic syndrome. An increased risk of autism and other adverse neurodevelopmental outcomes has also been reported, and may be related to shared environmental and genetic factors [23-28]. However, evidence is circumstantial and of poor quality. (See "Infants of women with diabetes", section on 'Long-term outcome' and "Autism spectrum disorder: Terminology, epidemiology, and pathogenesis", section on 'Environmental and perinatal factors'.)
Risk factors
Individuals at high risk — Pregnant people with any of the following characteristics appear to be at increased risk of developing GDM. The risk varies across these characteristics and is additive when multiple risk factors are present [29-36]:
●Personal history of impaired glucose tolerance, A1C ≥5.7 percent, impaired fasting glucose, or GDM in a previous pregnancy (GDM in previous pregnancy has 40 percent risk of recurrence [37]).
●Family history of diabetes, especially in a first-degree relative.
●Prepregnancy BMI ≥30 kg/m2, significant weight gain in early adulthood or between pregnancies, or excessive gestational weight gain during the first 18 to 24 weeks of pregnancy.
●Older maternal age (≥35 years of age).
●Member of one of the following groups, which have a high prevalence of type 2 diabetes: Hispanic American; Native American, Alaska native, or Native Hawaiian; South or East Asian, Pacific Islander. The risk is lower among non-Hispanic White and non-Hispanic Black people [4].
●Medical condition/setting associated with development of diabetes, such as polycystic ovary syndrome (PCOS).
●Previous birth of an infant ≥4000 g (approximately 9 pounds).
Individuals at low risk — By comparison, the risk of developing GDM is low in younger (<25 years of age) non-Hispanic White people, with normal BMI (<25 kg/m2 [<23 kg/m2 in Asian people]), no history of previous glucose intolerance or adverse pregnancy outcomes associated with GDM, and no first-degree relative with diabetes [38]. Only 10 percent of the general obstetric population in the United States meets all of these criteria for low risk of developing GDM, which is the basis for universal rather than selective screening [39]. (See 'Candidates' below.)
Lifestyle interventions for risk reduction — Lifestyle interventions (eg, diet modification, exercise) are targeted at individuals at high risk for developing GDM, particularly individuals who are overweight or obese and those with a history of GDM in a prior pregnancy. The goal of the lifestyle intervention is weight loss leading to improvement in insulin resistance and impaired insulin secretion. Smoking cessation should be encouraged in all patients as well. Although it is often associated with weight gain, it may reduce GDM risk and has multiple maternal and fetal benefits. (See "Tobacco and nicotine use in pregnancy: Cessation strategies and treatment options".)
A meta-analysis of randomized trials has shown that efforts to reduce the risk of GDM can be beneficial. Specifically, in targeted individuals with a higher baseline risk of GDM (eg, 20 percent incidence of GDM), lifestyle interventions could reduce the risk of GDM by approximately 20 percent compared with standard care [40]. Interventions included diet modification, increased exercise, or both, ideally initiated before pregnancy and continuing across gestation; exercising at the proper intensity and frequency (eg, moderate intensity for 50 to 60 minutes at least twice a week throughout pregnancy); and managing gestational weight gain so as to not exceed standard recommendations (table 1). Patients who are overweight or obese appear to benefit from weight loss before conception [41,42], whereas beginning a modest exercise program in the second trimester appears to be inadequate to impact the risk of GDM [43-46]. (See "Exercise during pregnancy and the postpartum period".)
The optimum diet is unclear. There is limited evidence (none from randomized trials) that a diet favoring fruit, vegetables, whole grains, and fish and low in red and processed meat, refined grains, and high-fat dairy reduces the risk of developing GDM [47,48]. There is low-quality evidence from randomized trials that advising patients about a low-glycemic-index diet is not effective for reducing this risk [49].
These interventions are the same as those utilized for prevention of type 2 diabetes and are discussed in detail separately. (See "Prevention of type 2 diabetes mellitus".)
Investigational risk reduction measures
●Metformin has been proposed as a medication to "prevent" GDM. The largest randomized trial of metformin in preventing diabetes in high-risk individuals, the Diabetes Prevention Program, demonstrated a large benefit in patients with prior GDM [50], but whether this finding applies in patients planning pregnancy or during pregnancy is less certain. Although observational studies of metformin use in patients with PCOS supported metformin to prevent GDM [51,52], a randomized trial did not [53]. In the randomized trial, 273 pregnancies among 257 patients with PCOS were randomly assigned to receive metformin (2000 mg/day) or placebo from the first trimester until delivery. There was no significant difference in prevalence of GDM between groups (metformin 17.6 percent versus placebo 16.9 percent). In this trial, and most other metformin trials, patients in the metformin group gained less weight during pregnancy than those in the placebo group (see "Metformin for treatment of the polycystic ovary syndrome", section on 'Prevention of pregnancy complications'). In addition, two randomized trials comparing metformin with placebo in pregnant people with obesity also showed no significant reduction in the rate of GDM [54,55].
●Probiotics do not appear to be useful and may be harmful. In a meta-analysis of randomized trials, probiotics did not clearly reduce the risk of GDM compared with placebo (mean risk ratio [RR] 0.80, 95% CI 0.54-1.20; 6 trials, 1440 participants) but increased the risk of preeclampsia (RR 1.85, 95% CI 1.04-3.29; 4 studies, 955 participants) [56].
●Myoinositol is a naturally occurring sugar in fruits, beans, grains, and nuts that can improve insulin resistance. A meta-analysis pooling data from seven trials (total of 1250 patients) reported a reduction in GDM with supplementation (OR 0.32, 95% CI 0.15-0.72), but there was significant heterogeneity among the trials, which may have been related to different doses and combinations of myoinositol supplementation [57]. Before this intervention can be recommended, large multicenter, blinded, randomized trials are needed to confirm safety and demonstrate improvement in clinically important maternal and/or neonatal outcomes.
BENEFITS AND HARMS OF SCREENING — The commonly used screening and diagnostic tests for diabetes involve drinking a glucose-containing beverage followed by blood glucose measurement. (See 'One- and two-step approaches' below.)
●Potential benefits – Screening for diabetes is performed in pregnancy because identifying patients with diabetes followed by appropriate therapy can reduce some fetal and maternal morbidities: macrosomia, shoulder dystocia, and preeclampsia may be reduced by 40 percent or more [58]. These data are described separately. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Rationale for treatment'.)
The cost implications of screening versus not screening have been modeled. In populations with a high prevalence of GDM and type 2 diabetes, screening appears to be cost-effective for prevention of type 2 diabetes, provided that lifestyle interventions are applied subsequent to pregnancy [59].
●Potential harms – None of the screening or diagnostic tests are associated with serious harmful maternal or fetal effects, although some individuals find the hyperosmolar drinks difficult to tolerate.
If GDM is diagnosed, management involves medical nutritional therapy, an increased frequency of prenatal visits, blood glucose monitoring, possible pharmacologic therapy, antenatal fetal monitoring, and possibly an increased risk for induction. Lifestyle interventions are not associated with serious harm, whereas antihyperglycemic drugs can be costly and can have side effects, which vary by drug and dose. Screening and diagnosis of GDM may lead to over- or undertreatment (ie, treatment without a favorable effect on outcome) and heighten maternal anxiety, but there is no evidence of long-term psychological harm.
SCREENING FOR UNDIAGNOSED TYPE 2 DIABETES IN EARLY PREGNANCY
Rationale — An increasing proportion of pregnant patients have unrecognized type 2 diabetes due to the increasing prevalence of obesity and lack of routine glucose screening/testing in reproductive-age individuals [60]. It has been estimated that approximately 30 percent of females aged 18 to 44 years in the United States have diagnosed or undiagnosed abnormal glucose metabolism (26 percent impaired glucose tolerance and/or impaired fasting glucose, 4 percent type 2 diabetes) [61].
Unrecognized, untreated hyperglycemia in early pregnancy is important because it is associated with an increased risk of miscarriage and having a child with a congenital anomaly, and mothers may have unrecognized complications (eg, nephropathy, retinopathy) from diabetes that place them at increased risk during pregnancy [62-64]. If these individuals are identified early in pregnancy, they could benefit from receiving the diagnostic and therapeutic interventions routinely provided to patients with preexisting (pregestational) diabetes mellitus. (See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management".)
Candidates for early pregnancy screening
Our approach — Given the increasing frequency of type 2 diabetes and evidence of a reduced risk of large for gestational age newborns with universal early maternal screening and treatment [65], the author checks an A1C level to screen for diabetes as part of the routine prenatal laboratory studies at the initial prenatal visit.
●A1C ≥6.5 percent (≥48 mmol/mol) – In nonpregnant people, an A1C ≥6.5 percent (≥48 mmol/mol) is one of the criteria used to diagnose diabetes (table 2). Therefore, an A1C ≥6.5 percent early in pregnancy, when A1C levels are generally slightly lower than in the nonpregnant state [66], strongly suggests previously undiagnosed type 2 diabetes. These patients are managed similar to those with preexisting diabetes mellitus. (See "Clinical presentation, diagnosis, and initial evaluation of diabetes mellitus in adults" and "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control" and "Pregestational (preexisting) diabetes mellitus: Obstetric issues and management".)
●A1C <6.5 percent (<48 mmol/mol) – This is a negative test for diabetes. The author screens these patients for GDM with the two-step test at 24 to 28 weeks of gestation. Approximately one-quarter of pregnant people with A1C 5.7 to 6.4 percent ([39 to 46 mmol/mol] suggestive of impaired glucose intolerance) in early pregnancy are diagnosed with GDM when screened with a two-step test later in pregnancy compared with <10 percent of those with A1C <5.7 percent (39 mmol/mol) [67]. (See 'One- and two-step approaches' below.)
The author acknowledges that an A1C <6.5 percent (<48 mmol/mol) is not sufficiently sensitive to detect mildly impaired glucose tolerance, especially if the A1C is ≥5.7 (which is above the upper limit of normal in nonpregnant individuals), but the overall value of detecting and treating mildly impaired glucose tolerance in early pregnancy has not been established.
Other approaches
●Targeted screening – The International Association of Diabetes and Pregnancy Study Groups (IADPSG) [6], the American Diabetes Association (ADA) [2], and the American College of Obstetricians and Gynecologists (ACOG) [29] suggest targeting early pregnancy screening for individuals at increased risk of undiagnosed type 2 diabetes, whereas a United States Preventive Services Task Force (USPSTF) guideline concluded that available evidence was insufficient to assess the balance of benefits and harms of screening asymptomatic pregnant people for glucose intolerance before 24 weeks of gestation [68,69].
The ADA and ACOG define patients at increased risk of type 2 diabetes based on: body mass index (BMI) ≥25 kg/m2 (≥23 kg/m2 in Asian Americans) plus one or more of the following [2,29]:
•GDM in a previous pregnancy.
•Glycated hemoglobin ≥5.7 percent (39 mmol/mol), impaired glucose tolerance, or impaired fasting glucose on previous testing.
•First-degree relative with diabetes.
•High-risk race/ethnicity (eg, African American, Latino, Native American, Asian American, Pacific Islander).
•History of cardiovascular disease.
•Hypertension (≥140/90 mmHg) or on therapy for hypertension.
•High-density lipoprotein cholesterol level <35 mg/dL (0.90 mmol/L) and/or a triglyceride level >250 mg/dL (2.82 mmol/L).
•Polycystic ovary syndrome (PCOS).
•Physical inactivity.
•Other clinical condition associated with insulin resistance (eg, severe obesity, acanthosis nigricans).
In addition, we would include older age and use age ≥35 years as the threshold.
In a randomized trial, early pregnancy diabetes screening because of obesity alone (BMI ≥30 kg/m2) did not demonstrate a reduction in a composite of adverse outcomes (eg, macrosomia, primary cesarean delivery, hypertensive disease of pregnancy, shoulder dystocia, neonatal hyperbilirubinemia, or hypoglycemia) [70]. However, the trial findings were limited by the small number of participants diagnosed with early GDM (n = 29) and the timing of the early screening (between 14 and 20 weeks).
Targeting screening to symptomatic patients is not useful because almost all patients with GDM are asymptomatic. However, some patients, usually those with preexisting diabetes first diagnosed during pregnancy, may have symptoms of hyperglycemia (eg, increased thirst, polyuria, weight loss, blurry vision) that are associated with random blood glucose values of ≥200 mg/dL (11.1 mmol/L). (See "Clinical presentation, diagnosis, and initial evaluation of diabetes mellitus in adults".)
●Choice of screening test – Obstetricians in the United States tend to use the two-step approach for screening and diagnosis of diabetes in pregnancy as recommended by ACOG [29], regardless of gestational age, while others tend to use a one-step approach. Neither approach has been validated for diagnosis of diabetes in the first or early second trimester. (See 'One- and two-step approaches' below.)
•If a two-step test is used, criteria for a positive test are shown in the tables (table 3A-B). These criteria are the same as those used for diagnosis of GDM later in pregnancy.
•If a one-step test is used, ADA criteria for a positive test are shown in the table (table 2). These thresholds are the same as those used by the ADA for diagnosis of diabetes in nonpregnant people and were chosen because they correlate with development of adverse vascular events, such as retinopathy and coronary artery disease, in these individuals over time. IADPSG criteria for overt diabetes is fasting blood glucose ≥126 mg/dL (7.0 mmol/L) or A1C ≥6.5 percent or random plasma glucose ≥200 mg/dL (11.1 mmol/L) confirmed by fasting blood glucose or A1C at these levels. These thresholds for diagnosis of type 2 diabetes in early pregnancy are different from those used by the ADA and IADPSG for diagnosis of GDM later in pregnancy (table 3C).
Management of patients with an abnormal or normal early pregnancy glucose tolerance test (GTT)
●Abnormal GTT – If a patient in early pregnancy (before significant insulin resistance) meets criteria for diagnosis of diabetes, they are assumed to have had the disorder prior to the pregnancy and their management is similar to those with documented preexisting diabetes mellitus in pregnancy. (See "Clinical presentation, diagnosis, and initial evaluation of diabetes mellitus in adults" and "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control" and "Pregestational (preexisting) diabetes mellitus: Obstetric issues and management".)
●Normal GTT – Patients with normal 75 or 100 gram oral GTT in early pregnancy are screened for GDM at 24 to 28 weeks of gestation. Those who were screened early in pregnancy by the one-step approach are rescreened using the 75 gram oral GTT. Those who were screened early in pregnancy by the two-step approach can omit the first step (50 gram oral glucose solution with glucose testing after one hour) at 24 to 28 weeks as it is likely to be abnormal; these patients can be rescreened using the three-hour 100 gram oral GTT [23]. (See '100 gram oral glucose tolerance test' below and '75 gram oral glucose tolerance test' below.)
SCREENING FOR GDM
Candidates — In the United States, universal screening appears to be the most practical approach. If a risk-factor-based screening approach is used, 90 percent of pregnant people in the United States would have at least one risk factor for glucose impairment during pregnancy [39] (see 'Risk factors' above). Moreover, consideration of risk factors is a poor method for identifying patients who will have an abnormal glucose tolerance test (GTT) because as many as 20 percent of pregnant people diagnosed with GDM have no risk factors [71-73].
A systematic review by the United States Preventive Services Task Force (USPSTF) concluded that there is a moderate net benefit to screening for GDM at 24 weeks of gestation or after to improve maternal and fetal outcomes [68,69].
One- and two-step approaches — Screening is performed at 24 to 28 weeks of gestation since 24 weeks is the gestational age when insulin resistance is significantly increasing, leading to hyperglycemia in those with insufficient insulin secretory capacity to maintain euglycemia [2,29]. It can be performed as a two- or one-step process. There is no consensus among national and international organizations regarding the optimal approach, and the choice generally depends on local customs. Screening can be performed after 24 to 28 weeks in patients who initially present for prenatal care at >28 weeks.
●Two-step test – The two-step test is the most widely used approach for identifying pregnant people with GDM in the United States. The first step is a one-hour 50 gram oral GTT administered without regard to time of day/previous meals. Screen-positive individuals are at increased risk for GDM so they go on to have the more cumbersome second step, a fasting 100 gram oral GTT, which is a diagnostic test for GDM. Criteria for diagnosis of GDM endorsed by the American College of Obstetricians and Gynecologists (ACOG) are shown in the tables (table 3A-B) [29].
We prefer the two-step approach because fewer patients are diagnosed and managed as GDM, without incurring an increase in adverse outcome (see 'Evidence' below). The first step has the practical advantages that fasting is unnecessary and only one blood sample is required. A minority of patients need to undergo the second step [74]. For the one-hour 50 gram oral GTT, we use ≥135 mg/dL (7.5 mmol/L) as the cutoff for a positive test because our patient population is at high risk for development of GDM. Although there are limited data regarding the clinical benefit of this cutoff, we believe that it offers the optimum combination of sensitivity and specificity for our population. We use Carpenter and Coustan criteria for an abnormal three-hour 100 gram oral GTT as the lower, more stringent criteria for diagnosis identify patients with glucose intolerance at the most risk of adverse perinatal outcomes.
●One-step test – The one-step approach simplifies screening by performing only a diagnostic test, typically a fasting 75 gram oral GTT, in all patients. The American Diabetes Association (ADA) and International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria for diagnosis of GDM are shown in the table (table 3C).
Evidence — In a 2022 meta-analysis of randomized trials comparing outcomes of the one- versus the two-step approach:
●The one-step approach diagnosed twice as many patients with GDM (16.3 versus 8.3 percent; RR 2.13, 95% CI 1.61-2.82) and resulted in twice as many patients receiving antihyperglycemic pharmacotherapy (7.1 versus 3.8 percent; RR 2.24, 95% CI 1.21-4.15)
●Neonatal hypoglycemia was more common in the one-step group (9.3 versus 7.6 percent; RR 1.23, 95% CI 1.13-1.34)
●Other maternal and neonatal outcomes were similar for both groups, for example:
•LGA (8.8 versus 9.2 percent; RR 0.95, 95% CI 0.88-1.04)
•Hypertensive disorder of pregnancy (13.6 percent versus 13.6 percent; RR 1.00, 95% CI 0.94-1.07)
•Primary cesarean birth (24.0 versus 24.7 percent; RR 0.98, 95% CI 0.93-1.02)
•Macrosomia (>4000 grams, 11.2 versus 11.3 percent; RR 0.99, 95% CI 0.92-1.07)
•Shoulder dystocia (2.1 versus 2.0 percent; RR 1.06, 95% CI 0.89-1.28)
•SGA (8.6 versus 8.3 percent; RR 1.04, 95% CI 0.95-1.13)
•Stillbirth (0.5 versus 0.6 percent; RR 0.90, 95% CI 0.64-1.28)
Results were similar in high-quality observational studies, except for LGA (RR 0.97, 95% CI 0.95-0.98) where the confidence interval was consistent with a modest reduction.
Based on these findings, it appears that using the one-step method increases the number of patients who receive a diagnosis of GDM and thus has the potential for increased patient and medical system burden (eg, more prenatal visits, fetal and maternal surveillance, lifestyle changes, and intervention) with economic, personal, and psychological consequences [75-77], but without a clear direct benefit over the two-step approach in maternal and newborn outcomes. Therefore, we believe it is prudent to use the two-step approach to diagnose GDM followed by treatment of affected patients according to standard guidelines. Nevertheless, because of limitations of the available trials, the best approach remains controversial [78]. (See 'One- and two-step approaches' above.)
An additional issue is whether increased identification of glucose intolerance in pregnancy would lead to improved long-term metabolic outcomes. Theoretically, if patients with GDM make postpartum lifestyle changes (with or without metformin), development of diabetes may be prevented or delayed [50]. Data are lacking due to the low percentage of patients with GDM who complete the recommended postpartum follow-up. Counseling regarding long-term risk, adherence to a risk-reduction program, and long-term glucose monitoring for diabetes would be essential to assess this issue.
Procedure and laboratory issues
Step one of the two-step approach: 50 gram glucose GTT — A 50 gram oral glucose solution is given without regard to the time elapsed since the last meal and venous plasma glucose is measured one hour later (sometimes called a glucose challenge test [GCT] or glucose loading test). Glucose concentration should be measured using an accurate and precise enzymatic method. The following thresholds have been proposed to define a positive screen: ≥130 mg/dL, ≥135 mg/dL, or ≥140 mg/dL (7.2 mmol/L, 7.5 mmol/L, or 7.8 mmol/L).
The original threshold for an elevated test (equivalent to 143 mg/dL [7.9 mmol/L] with current methodology) was arbitrary, used whole blood and a nonspecific glucose assay, and was validated by its ability to predict an abnormal 100 gram oral GTT [79]. Use of a lower threshold (≥130 mg/dL [7.2 mmol/L] with current methodology) provides greater sensitivity, but results in more false positives and would require administering a three-hour 100 gram oral GTT to more patients [3,80]. In a systematic review of cohort studies of screening tests for GDM by the USPSTF, at the 130 mg/dL (7.2 mmol/L) threshold, sensitivity and specificity were 88 to 99 percent and 66 to 77 percent, respectively [81]. At the 140 mg/dL (7.8 mmol/L) threshold, sensitivity was lower (70 to 88 percent), but specificity was higher (69 to 89 percent).
Management of patients with markedly elevated results — Patients who have a very high glucose level on their one-hour 50 gram oral GTT have a high likelihood of an abnormal three-hour 100 gram oral GTT; the positive predictive value (PPV) varies depending on the prevalence of GDM in the population tested and the GTT criteria used for diagnosis. For example, Carpenter and Coustan found that patients whose one-hour 50 gram oral GTT plasma glucose concentration was >182 mg/dL (10.1 mmol/L) had >95 percent probability of an abnormal three-hour 100 gram oral GTT [3]. At glucose levels ≥200 mg/dL (11.1 mmol/L), others have reported PPVs of 47 to 80 percent for an abnormal three-hour 100 gram oral GTT [82-84].
Based on these findings, for patients with one-hour 50 gram glucose concentration ≥200 mg/dL (11.1 mmol/L), the author makes a presumptive diagnosis of GDM, unless the patient prefers to undergo a three-hour 100 gram oral GTT for definitive diagnosis. The GTT can be performed safely, as the 100 gram glucose load would rarely, if ever, lead to a hyperglycemic emergency or other serious adverse events in patients with GDM or unrecognized type 2 diabetes. It has been performed in thousands of patients with no reports of serious adverse events.
Diagnostic testing: 75 or 100 gram glucose GTT
●The diagnosis of GDM is based on results of a 75 or 100 gram glucose oral GTT. However, it should be noted that the GTT is an imprecise test with poor reproducibility [85]. In a study that performed two 100 gram oral GTTs one to two weeks apart in 64 pregnant people whose one-hour 50 gram oral GTT was ≥135 mg/dL, 48 had normal/normal, 11 had normal/abnormal, 3 had abnormal/normal, and 2 had abnormal/abnormal results [86]. Thus, only 50 of 64 (78 percent) had reproducible test results. Nevertheless, the 75 or 100 gram oral GTT is a practical and widely utilized means of diagnosing GDM, and the 75 gram oral GTT is a widely utilized means of diagnosing diabetes in nonpregnant individuals. Proper handling and processing of blood samples are important for accurate results [87].
●There is some variation regarding the thresholds for a positive test (table 3B-C).
●Although the 100 gram oral GTT is typically performed as the second step of the two-step approach while the 75 gram test is performed as the only test in the one-step approach, this is arbitrary. In fact, the Canadian Diabetes Association clinical guidelines suggest the 75 gram oral GTT as the second step of the two-step approach [88]. Historically, carbohydrate loading for three days before the test was recommended but is often no longer recommended and is not necessary if the patient is not on a very-low-carbohydrate diet [89-92].
Some clinicians obtain a fasting glucose level before administering the GTT. If a 75 gram oral GTT is planned and the fasting glucose level is ≥92 mg/dL (5.1 mmol/L), then the diagnosis of GDM is made and the GTT is cancelled. If a 100 gram oral GTT is planned, no data support a particular fasting cutoff for diagnosing GDM and an abnormal fasting glucose level alone is not diagnostic of GDM. However, a glucose level ≥126 mg/dL (7 mmol/L) is a reasonable threshold for cancelling the GTT as it is diagnostic of diabetes in the general population. This approach requires asking the patient to have blood drawn for their fasting glucose level and then wait for the results before proceeding with the GTT later on the same day (and remain fasting) or on another day (and fast again), which is cumbersome.
As discussed above, it is not necessary to exclude fasting hyperglycemia to safely perform the test. (See 'Management of patients with markedly elevated results' above.)
100 gram oral glucose tolerance test — The three-hour 100 gram oral GTT is diagnostic of GDM when two glucose values are elevated. The most commonly used thresholds for defining elevated values have been proposed by Carpenter and Coustan (table 3B) [3], which is a modification of thresholds proposed by O'Sullivan and Mahan [93], originally based on venous whole blood samples. The Carpenter and Coustan values are based on newer enzymatic assays performed on plasma samples and reflect current laboratory practices.
The requirement for two abnormal values on the GTT was an arbitrary decision made by O'Sullivan and Mahan in 1964 [93]. Treatment of patients who meet these diagnostic criteria for GDM improves some pregnancy outcomes (eg, pregnancy-induced hypertension, macrosomia, shoulder dystocia) [94].
However, a 2016 systematic review including 25 studies noted that pregnant people with one abnormal value on the 100 gram oral GTT generally had increased risks for the same poor outcomes as those with two abnormal values (ie, GDM) [95]. This subset of patients may warrant closer monitoring for fetal overgrowth and other adverse outcomes of GDM. Whether they would benefit from treatment is unknown. The author of this topic obtains a third-trimester fetal growth scan on these patients. If fetal overgrowth or polyhydramnios is present, she discusses the option of home glucose monitoring.
75 gram oral glucose tolerance test — The two-hour 75 gram oral GTT is diagnostic of GDM when one glucose value is elevated. The most commonly used thresholds for defining elevated values have been proposed by the IADPSG (table 3C). The 75 gram oral GTT is generally more convenient, better tolerated and more sensitive for identifying the pregnancy at risk for adverse outcome (eg, gestational hypertension, preeclampsia, LGA) than the 100 gram oral GTT [96]. Increased sensitivity is primarily related to the fact that only one elevated glucose value is needed for a positive test [97], although the cutoffs are also slightly lower. In addition, all patients must fast before the one-step 75 gram oral GTT. With the two-step method, the first-step (50 gram oral glucose solution with glucose testing after one hour) does not require fasting; the majority of patients (70 to 80 percent) screen-negative at this step and only those who screen-positive need to go on to fast for the 100 gram test [3].
The IADPSG-defined thresholds for the 75 gram oral GTT are primarily based on outcome data reported in the HAPO study, a prospective observational study of more than 23,000 pregnancies in nine countries evaluated with a 75 gram oral GTT at 24 to 32 weeks of gestation [6,17]. These thresholds represent the glucose values at which the odds of infant birth weight, cord C-peptide (surrogate for fetal insulin level), and neonatal percent body fat >90th percentile were 1.75 times the estimated odds of these outcomes at mean glucose levels, based on fully adjusted logistic regression models. Compared with participants in the HAPO study with all glucose values below the thresholds, those who exceeded one or more of these thresholds had a twofold higher frequency of LGA infants and preeclampsia and >45 percent increase in preterm delivery and primary cesarean delivery. Using an odds ratio (OR) of 2 for the thresholds defined a population with a higher frequency of these outcomes, but the difference was modest and resulted in failure to identify many pregnant people who were at almost similar risk of these outcomes but had all glucose values below the thresholds. IADPSG-defined thresholds are also predictive of a more than threefold increased long-term risk of diabetes and metabolic syndrome [98,99]. In a post hoc analysis of data from the HAPO Study, 52 percent of pregnant people who would meet IADPSG criteria for GDM developed a disorder of glucose metabolism by 10 to 14 years postpartum, compared with 20 percent of mothers without GDM (OR 3.44, 95% CI 2.85-4.14 [99]). Among those who met criteria for GDM, rates of type 2 diabetes and prediabetes were 10.7 and 41.5 percent, respectively, versus rates of 1.6 and 18.4 percent, respectively, in those without GDM. The children of these individuals were more likely to be "overweight or obese" (39.5 versus 28.6 percent, OR 1.21, 95% CI 1.00-1.46) after adjustment for maternal body mass index (BMI) during pregnancy, and the difference in prevalence of obesity was statistically significant (19 versus 10 percent, OR 1.58, 95% CI 1.24-2.01).
In 2022, the IADPSG-defined thresholds for the 75 gram oral GTT were reassessed in a trial that randomly assigned 4061 pregnant patients undergoing GDM screening to the standard glycemic threshold criteria for diagnosis (table 3C) or higher thresholds (ie, fasting plasma glucose ≥99 mg/dL [≥5.5 mmol/L] or two-hour glucose ≥162 mg/dL [≥9.0 mmol/L]) [100]. Patients who met the threshold criteria for GDM for their assigned group received usual care for patients with GDM (per New Zealand Ministry of Health clinical practice guideline on GDM). Major findings were:
●Use of lower glycemic threshold criteria resulted in more patients diagnosed with GDM (15.3 versus 6.1 percent), but this was not accompanied by a significant reduction in birth of large for gestational age (8.8 versus 8.9 percent) or macrosomic newborns (11.8 versus 12.3 percent).
●Use of the lower glycemic threshold criteria resulted in more inductions of labor (33.7 versus 30.2 percent), use of health services, and use of pharmacotherapy (10.9 versus 4.6 percent). The lower glycemic threshold group also had a higher rate of neonatal hypoglycemia warranting treatment (10.7 versus 8.4 percent), which was likely related to hospital protocols requiring screening for hypoglycemia in infants of diabetic mothers.
●Rates of preeclampsia, preterm birth, shoulder dystocia, cesarean birth, and small for gestational newborn were similar in the two trial groups.
Interestingly, about half of the patients with glucose test results that fell between the lower and higher glycemic threshold criteria were considered by their providers to have a mild degree of GDM and were treated for GDM. These patients had maternal and infant health benefits, including fewer large for gestational age newborns, compared with those who did not receive a diagnosis of a mild degree of GDM and therefore not treated.
The results of this trial suggest that the optimum glucose thresholds for diagnosis of GDM by the 75 gram oral GTT still need to be determined.
Options for patients unable to tolerate oral hyperosmolar glucose — The highly concentrated hyperosmolar glucose solution used for the GCT and GTT can cause gastric irritation, delayed emptying, and gastrointestinal osmotic imbalance, leading to nausea and, in a small percentage of patients, vomiting [101-103]. Options in these cases include:
●Serving the hyperosmolar glucose drink on ice may reduce nausea and vomiting.
●If the patient vomited during the oral GTT and is willing to come back another day for repeat testing, premedication with an antiemetic drug may allow the test to be completed.
●The GTT can be performed intravenously instead of orally, although this is rarely done.
●Periodic fasting and one- or two-hour postprandial blood glucose tests can be obtained in pregnant people at high risk for GDM. For example, they can be asked to keep a weekly log of self-monitored glucose values from 24 to 28 weeks (the gestational age when screening would be performed) and at 32 weeks (the gestational age of peak insulin resistance).
This approach is also useful for pregnant people who have dumping syndrome after a Roux-en-Y gastric bypass procedure; these individuals are unlikely to tolerate a hyperosmolar glucose solution [104]. Monitoring glucose values will only identify those cases of GDM that might require intervention for hyperglycemia and not all cases of GDM. (See "Fertility and pregnancy after bariatric surgery".)
Other approaches have been tried in patients who do not tolerate or decline to have an oral GTT, but are not recommended because they have not been validated or they perform poorly.
●Candy, a predefined meal, or commercial soft drinks with 75 or 100 grams of glucose have been used instead of a standard glucose monomer or polymer solution [105-110]. These oral glucose loads are better tolerated but appear to be less sensitive and have not been validated in large studies. None have been endorsed by the ADA or ACOG.
●A1C – In a USPSTF systematic review [81], no threshold for glycated hemoglobin (A1C) in the second and third trimesters had both good sensitivity and specificity as a screening test for GDM.
●Fasting glucose – The USPSTF review [81] also found that a fasting plasma glucose level less than 85 mg/dL (4.7 mmol/L) by 24 weeks of gestation performed well for identifying pregnant people who did not have GDM, whereas a value over 85 mg/dL (4.7 mmol/L) performed less well than the one-hour 50 gram oral GTT for identifying those with GDM.
However, in low-resource settings where universal screening with a GTT is not feasible, use of fasting plasma glucose at 24 to 28 weeks to screen pregnant people may be a practical approach, a reasonable trade-off between cost savings and adverse outcomes in missed cases. In a study from 15 Chinese hospitals, if performance of the GTT was restricted to pregnant people with fasting glucose from 79 mg/dL (4.4 mmol/L) to 90 mg/dL (5.0 mmol/L), then 50 percent of pregnant people could avoid a GTT since 38 percent of this population had fasting glucose <79 mg/dL (4.4 mmol/L) and 12 percent had fasting glucose >90 mg/dL (5.0 mmol/L), diagnostic of GDM in this system; 12 percent of patients with GDM were missed [111]. These findings may not be generalizable to other low-resource populations since Asian females have a higher incidence of type 2 diabetes and GDM than White females, and fasting hyperglycemia among Asian females with GDM is less prominent than in non-Asian populations [97]. In non-Asian populations, 50 to 75 percent of patients with GDM have fasting glucose levels ≥92 mg/dL (5.1 mmol/L) on an oral GTT [97].
●Testing for glucosuria – A positive urine dipstick for glycosuria is not very predictive of GDM, and a negative urine dipstick for glycosuria is not very predictive of absence of GDM, so should not be used for ruling the diagnosis in or out [112-114]. Glycosuria with a normal blood glucose level is common in pregnant people as pregnancy is associated with reductions in fractional reabsorption of glucose, which results in higher rates of urinary excretion.
Should testing be repeated later in pregnancy? — We recommend not repeating a GTT later in pregnancy. Although some studies have shown that repeating the test after an initially normal GTT (defined as the absence of at least two glucose values above threshold) will identify additional cases in 4 to 29 percent of cases depending on the timing and indication for repeat testing [115-118], the value of identifying these individuals versus nondiagnosis of GDM is unproven. Furthermore, patients with a normal GTT and sonographic signs of fetal overgrowth or one abnormal value on the GTT can be given the option of insulin therapy in an attempt to improve pregnancy outcome. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Pharmacotherapy'.)
RECOMMENDATIONS OF NATIONAL AND INTERNATIONAL ORGANIZATIONS — The optimum strategy for diagnosis of GDM to improve maternal and infant health is unclear [119]. Many organizations have published recommendations for screening and diagnosis of diabetes in pregnancy:
●American College of Obstetricians and Gynecologists (ACOG, two-step approach (table 3A and table 3B)) [29]
●International Association of Diabetes and Pregnancy Study Groups (IADPSG, one-step approach (table 3C)) [6]
●American Diabetes Association (ADA, one- or two-step approach) [2]
●World Health Organization (WHO, one-step approach (table 4)) [120]
●Canadian Diabetes Association (CDA, two-step [preferred] or one-step approach) [88]
●Australasian Diabetes in Pregnancy Society (WHO approach) [38]
●National Institute for Health and Care Excellence (NICE, United Kingdom, one-step approach)
●International Federation of Gynecology and Obstetrics (FIGO, one-step approach) [121]
POSTDELIVERY FOLLOW-UP — Individuals with GDM should be screened again postpartum and periodically thereafter because they are at increased risk for developing type 2 diabetes mellitus. Postdelivery follow-up is described separately. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Follow-up'.)
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 5th to 6th 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 10th to 12th 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.)
●Basics topics (see "Patient education: Gestational diabetes (diabetes that starts during pregnancy) (The Basics)")
●Beyond the Basics topics (see "Patient education: Gestational diabetes (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Terminology – Gestational diabetes mellitus (GDM) refers to diagnosis of diabetes at 24 to 28 weeks of gestation. Diagnosis of diabetes in early pregnancy is more consistent with previously undiagnosed type 2 diabetes. The term "overt diabetes" is sometimes used to describe the diabetes status of these individuals until they are re-evaluated in the nonpregnant state and a formal diagnosis of type 2 diabetes can be made. (See 'Terminology' above.)
●Overall approach – An overall approach to screening for diabetes in pregnancy and after delivery is shown in the algorithm (algorithm 1).
●Early pregnancy screening for previously undiagnosed type 2 diabetes – There is no standard approach to early pregnancy screening for undiagnosed type 2 diabetes. The author obtains an A1C level at the initial prenatal visit in all patients as part of the initial prenatal blood work; a value ≥6.5 percent (≥48 mmol/mol) is diagnostic of diabetes and the patient is managed accordingly. If the A1C is <6.5 percent (<48 mmol/mol), the author performs standard screening for GDM at 24 to 28 weeks. Alternatively, several national organizations have suggested a targeted approach that limits screening to patients at high risk. (See 'Screening for undiagnosed type 2 diabetes in early pregnancy' above.)
●Screening for GDM at 24 to 28 weeks – We suggest screening for GDM at 24 to 28 weeks of gestation (Grade 2B). In the United States, universal rather than selective screening appears to be the most practical approach because 90 percent of pregnant people have at least one risk factor for glucose impairment during pregnancy. Identifying pregnant people who have GDM followed by medical nutritional therapy, and then pharmacotherapy if glycemic goals are not achieved, can reduce the rate of some adverse pregnancy outcomes, particularly macrosomia, shoulder dystocia, and preeclampsia. (See 'Candidates' above and 'Significance' above and 'Benefits and harms of screening' above and 'One- and two-step approaches' above.)
●One- and two-step testing approaches – We use a two-step approach for screening/diagnosis of GDM (table 3A-B), but a one-step approach is also acceptable (table 3C). (See 'One- and two-step approaches' above and 'Procedure and laboratory issues' above.)
●Postpartum follow-up – Individuals with GDM should be screened for diabetes postpartum and periodically thereafter because they are at increased risk for developing type 2 diabetes mellitus. (See "Gestational diabetes mellitus: Glucose management and maternal prognosis", section on 'Follow-up'.)
ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Lois Jovanovic, MD, Donald R Coustan, MD, and Michael Greene, MD, who contributed to earlier versions of this topic review.
