By Ahizechukwu C. Eke, MD, MPH
Assistant Professor in Maternal Fetal Medicine, Division of Maternal Fetal Medicine, Department of Gynecology & Obstetrics, Johns Hopkins University School of Medicine, Baltimore
Dr. Eke reports no financial relationships relevant to this study.
SYNOPSIS: In this retrospective cohort study of 243 pregnant women who had combined hemoglobin A1c (HbA1c) and a two-step oral glucose tolerance testing at less than 21 weeks of gestation, median values of HbA1c were higher in women with gestational diabetes compared to nondiabetics (5.8% compared to 5.3%; P < 0.001). The predictive probability of using HbA1c in diagnosing diabetes in early pregnancy was high compared to two-step testing (area under the curve, 0.8), with an optimal diagnostic threshold of 5.6%. Although a HbA1c level of > 6.5% is diagnostic of early gestational diabetes, a lower diagnostic threshold might be justified during pregnancy.
SOURCE: Battarbee AN, Grant JH, Vladutiu CJ, et al. Hemoglobin A1c and early gestational diabetes. J Womens Health (Larchmt) 2020; July 15. [Online ahead of print].
Gestational diabetes is a common medical complication during pregnancy, and a major cause of neonatal morbidity and mortality.1,2 Although the 50-g, one-hour oral glucose tolerance testing (if
> 140 mg/dL, but < 200 mg/dL), followed by a 100-g, three-hour oral glucose testing (two-step testing) is the preferred algorithm for diagnosing gestational diabetes in the United States, the utility of the two-step diabetic test is greater during the second trimester of pregnancy (24 to 28 weeks) because of worsening insulin resistance from placental hormones with advancing gestation.3 Hence, investigators have considered other alternatives to the two-step testing for earlier diagnosis of gestational diabetes.4
Increasingly, hemoglobin A1c (HbA1c) is being used to diagnose diabetes in early pregnancy.5,6 The ease of obtaining a single blood draw and not subjecting pregnant women to glucose loads (50 grams, then 100 grams) makes HbA1c an attractive option to the two-step oral glucose tolerance testing in diagnosing diabetes in pregnant women. Although a HbA1c concentration of ≥ 6.5% is diagnostic of diabetes outside of pregnancy, its utility in diagnosing diabetes during pregnancy is still a subject of debate.
In this retrospective cohort study in Chapel Hill, NC, Battarbee and colleagues reported their findings in pregnant women at high risk for gestational diabetes who were screened by either using oral glucose tolerance testing or HbA1c at < 21 weeks of gestation between July 2016 and July 2017. Women were eligible for inclusion if they were considered high risk for gestational diabetes, including women with a prior history of gestational diabetes or obese women (body mass index of ≥ 30 kg/m2) at < 21 weeks of gestation. Pregnant women were excluded if they were pregestational diabetics, on metformin therapy, anemic (hematocrit of < 27 mg/dL), and homozygous for hemoglobinopathy.7
To estimate the predictive probability of using HbA1c or the two-step test in the diagnosis of early gestational diabetes at < 21 weeks of gestation, the area under the receiver operative curve (ROC) was used, and the concordance probability method that maximized sensitivity and specificity for defining the optimal HbA1c cut-off value in the ROC area under the curve was used. Other HbA1c diagnostic cut-offs were assessed using alternate sensitivity, specificity, and diagnostic accuracy.7
Among 426 high-risk women initially identified for inclusion, 243 women were selected after excluding all ineligible women. Seventy-seven (32%) of the women were non-Hispanic Black, 89 (36%) were non-Hispanic white, 55 (23%) were Hispanic, and 22 (9%) were from other ethnicities.7 Women diagnosed with gestational diabetes at < 21 weeks of gestation were more likely to be obese and/or had a history of gestational diabetes in a previous pregnancy. Median values of HbA1c were higher in women with gestational diabetes compared to nondiabetics (5.8% compared to 5.3%; P < 0.001). The ROC area under the curve for HbA1c when compared to two-step oral glucose tolerance testing was 0.8 (95% confidence interval, 0.69-0.91), with an optimal threshold of 5.6% (64% sensitivity and 84% specificity).7 Fourteen (5.8%) of the 243 were diagnosed with diabetes based on the two-step oral glucose tolerance testing, while nine women were diagnosed with gestational diabetes using an optimal threshold of 5.6%.
HbA1c is a glycated hemoglobin resulting from sustained exposure of red blood cells to glucose in the setting of hyperglycemia.8 Although a level of ≥ 6.5% is recommended for diagnosing diabetes in nonpregnant women, no consensus has been reached in the utility of HbA1c for diagnosing diabetes during pregnancy. However, levels of HbA1c ≥ 6.5% have been used in diagnosing diabetes in early pregnancy on the basis that HbA1c levels in early pregnancy are similar to those to in nonpregnant women (4.5% to 5.7% in early pregnancy vs. 4.7% to 6.3% in non-pregnant women).5,6 Therefore, HbA1c levels of ≥ 6.5% in early pregnancy implies undiagnosed pregestational diabetes. The study by Batterbee and colleagues was unique compared to other HbA1c studies in early pregnancy because they studied an ethnically and racially diverse group of pregnant women. In addition, they analyzed their samples with modern technology, and used a single site, thereby increasing precision of HbA1c reported values. Their optimal threshold for diagnosing diabetes was 5.6% and is consistent with the values reported by other studies.5,6 The lower HbA1c thresholds reported in these studies might be related to the physiological changes that occur during normal pregnancy — for example, the red blood cell lifespan shortens to approximately 90 days, with HbA1c declining from rapid red blood cell turnover,9 physiologic anemia of pregnancy, and increased intestinal transit times that occur during pregnancy.4 In addition, there are racial differences in HbA1c normal values.4 Because of these factors, using lower HbA1c levels (< 6.5%) in early pregnancy for making a diagnosis of diabetes, while reasonable, remain a subject of debate.
In conclusion, the HbA1c threshold of 5.6% for diagnosing diabetes in early pregnancy, although modest, is not currently recommended. Until well-designed studies are completed, a HbA1c of ≥ 6.5% could remain useful as a screening test in early pregnancy as an alternative to the two-step oral glucose tolerance testing in women intolerant to large glucose loads (for example, patients with a history of Roux-en-Y gastric bypass surgery). This is important because making the diagnosis of gestational diabetes in early pregnancy and instituting early management improves outcomes and decreases the future risk of cardiovascular disease, obesity, type 2 diabetes, and metabolic syndrome.
- Domanski G, Lange AE, Ittermann T, et al. Evaluation of neonatal and maternal morbidity in mothers with gestational diabetes: A population-based study. BMC Pregnancy Childbirth 2018;18:367.
- Riskin A, Itzchaki O, Bader D, et al. Perinatal outcomes in infants of mothers with diabetes in pregnancy. Isr Med Assoc J 2020;9:503-509.
- Committee on Practice Bulletins – Obstetrics. ACOG Practice Bulletin No. 190: Gestational Diabetes Mellitus. Obstet Gynecol 2018;131:e49-e64.
- Shinar S, Berger H. Early diabetes screening in pregnancy. Int J Gynaecol Obstet 2018;142:1-8.
- Radder JK, van Roosmalen J. HbA1c in healthy, pregnant women. Neth J Med 2005;63:256-259.
- Nielsen LR, Ekbom P, Damm P, et al. HbA1c levels are significantly lower in early and late pregnancy. Diabetes Care 2004;27:1200-1201.
- Battarbee AN, Grant JH, Vladutiu CJ, et al. Hemoglobin A1c and early gestational diabetes. J Womens Health (Larchmt) 2020; July 15. doi: 10.1089/jwh.2019.8203. [Online ahead of print.]
- Cohen RM, Franco RS, Khera PK, et al. Red cell life span heterogeneity in hematologically normal people is sufficient to alter HbA1c. Blood 2008;112:4284-4291.
- Schrader HM, Jovanovic-Peterson L, Bevier WC, Peterson CM. Fasting plasma glucose and glycosylated plasma protein at 24 to 28 weeks of gestation predict macrosomia in the general obstetric population. Am J Perinatol 1995;12:247-251.