Master Class

Alarming gaps in gestational diabetes care



Much attention has been given in the media to the incidence of prediabetes in the general population. The Centers for Disease Control and Prevention estimates that approximately 86 million adults have prediabetes, and that the incidence of this condition is similar across racial and ethnic groups. Indeed, the seriousness of this public health concern prompted the Centers for Medicare & Medicaid Services to expand Medicare coverage for interventions for people with prediabetes, a move that was finalized in November 2016.

Despite a widespread focus on the need to prevent prediabetes from becoming type 2 diabetes, women diagnosed with gestational diabetes mellitus (GDM), which accounts for about 9% of women in the United States, may not be receiving critical advice and care.

Dr. E. Albert Reece

Dr. E. Albert Reece

In September 2016, researchers at the University of Illinois at Chicago published a study indicating that women with a history of GDM did not receive adequate diabetes screening postpartum (Prev Chronic Dis. 2016;13:160106. doi:

The investigators analyzed data collected via the National Health and Nutrition Examination Survey from 2007-2012, and identified 284 women with a history of GDM. Only 67% of these women received diabetes screening, and approximately one-third of women included in the study had undiagnosed prediabetes and diabetes. The authors concluded that prediabetes in women who have had GDM may be underdiagnosed. They argued that women with GDM should be encouraged to have additional health visits and screenings to prevent the development of prediabetes or diabetes. Considering the fact that a number of studies have shown that GDM predisposes a woman to developing type 2 diabetes, the University of Illinois findings are alarming.

As ob.gyns., we have increasingly become a woman’s only health care practitioner. Although individuals may skip annual exams with a primary care physician, during which blood work is typically drawn, many women will see their ob.gyn. for regular check-ups. Therefore, we have a unique role to play in our patients’ lifelong health. This is especially important during pregnancy, when it may be easy to focus only on the mother’s health as it pertains to the health of the baby, rather than her health in pregnancy as it may affect her long-term well-being.

We have invited Robert Ratner, MD, the chief scientific and medical officer at the American Diabetes Association, to discuss the need to carefully follow up with patients who have had GDM and to educate them about their risk for developing type 2 diabetes later in life.

Dr. Reece, who specializes in maternal-fetal medicine, is vice president for medical affairs at the University of Maryland, Baltimore, as well as the John Z. and Akiko K. Bowers Distinguished Professor and dean of the school of medicine. Dr. Reece said he had no relevant financial disclosures. He is the medical editor of this column. Contact him at

Why postpartum GDM follow-up is so important


Much of the attention paid to diagnosing gestational diabetes has focused on the fetus and on babies being born very large. However, it is important to appreciate that the original definitions of the condition were based entirely on the long-term outcomes of the mother.

John O’Sullivan, MD, and statistician Claire Mahan published diagnostic criteria in 1964 after performing 3-hour oral glucose tolerance tests (OGTTs) in more than 500 unselected women during their pregnancies, and then following these women and babies out as far as 23 years. Retrospectively, Dr. O’Sullivan and Ms. Mahan defined gestational diabetes mellitus (GDM) as glucose values exceeding two standard deviations above the mean on two out of four OGTT values.

Dr. Robert E. Ratner

Dr. Robert E. Ratner

They came to their conclusions after tracking the later development of diabetes outside of pregnancy. More than 20 years later, 70% of women with the higher OGTT values had developed type 2 diabetes, compared with approximately 10% of women who did not have higher values during pregnancy. The O’Sullivan criteria were established, essentially, based on their association with the development of diabetes after pregnancy. In addition to being a significant predictor of subsequent diabetes, a history of GDM also conferred a three- to fourfold increase in maternal mortality.

Fifty-some years later, these findings have been affirmed through additional research and are the crux of what drives the current recommendations for postpartum follow-up of women with a history of GDM.

Long-term maternal risks

Postpartum, the current recommendation from both the American Diabetes Association and the American College of Obstetricians and Gynecologists is that women with GDM be tested at 6-12 weeks after delivery to ensure that the diabetes has resolved.

This recommendation for initial postpartum testing carries with it a stipulation that’s different from subsequent postpartum testing. It says that postpartum testing at 6-12 weeks should be performed with either a fasting glucose test or a 2-hour OGTT. Since hemoglobin A1c may still be impacted by the rapid red blood cell turnover in pregnancy or blood loss at delivery, A1c testing lacks sensitivity for identifying diabetes during this window of time.

Initial postpartum testing also serves as a way to identify whether the diabetes during pregnancy was preexisting or purely secondary to the hormonal changes associated with the pregnancy.

If this first postpartum test shows diabetes, the patient most likely had preexisting diabetes, and therapy must be initiated immediately. In the case of a normal result, the patient remains at higher risk for the development of type 2 diabetes essentially for the rest of her life and should be tested at least every 3 years for the occurrence of the disease.

Much of the increased risk for different ethnic groups occurs within 5 years of the index pregnancy. This was shown in a systematic review led by Catherine Kim, MD; the review examined more than two dozen studies with follow-up of up to 28 years postpartum. The cumulative incidence of type 2 diabetes increased markedly in the first 5 years and then appeared to plateau after 10 years (Diabetes Care. 2002 Oct;25[10]:1862-8).

The best data on late-occurring diabetes following GDM comes from the multicenter National Institutes of Health–sponsored Diabetes Prevention Program (DPP) trial, which randomized more than 3,000 individuals with baseline impaired glucose tolerance – or prediabetes – to one of two interventions: metformin therapy or intensive lifestyle intervention, or to placebo.

Within this population, there were more than 1,700 women who had a previous live birth. Of these women, 350 reported a history of GDM at a mean of 12 years since the delivery of their first GDM pregnancy. The DPP gave us the opportunity, therefore, to look at a large group of women about 12 years away from their GDM pregnancy who had abnormal glucose levels but had not reached the level of type 2 diabetes, and compare them with women with similarly impaired glucose tolerance who did not have a history of GDM.

There were interesting similarities and differences. Women with a GDM history were on average 8 years younger than women without a GDM history, but they had comparable BMIs. In addition, within the placebo arm, we could observe the natural history of glucose intolerance in women with and without a history of GDM. Despite both groups entering the study with equivalent degrees of impaired glucose tolerance and similar BMI, women with a history of GDM had a 71% higher risk of developing diabetes during the 3-year intervention period than that of parous women without a history of GDM (J Clin Endocrinol Metab. 2008 Dec;93[12]:4774-9).

Clearly, there was something about the history of GDM that puts these women at greater risk for diabetes than women who had the same impaired glucose tolerance, but no GDM. The study demonstrated that GDM is an exceptionally strong predictor of the development of type 2 diabetes, even for those who manage to escape diabetes for the first 10 years.

Next Article: