Secondary Logo

Journal Logo

Contents: Original Research

Pregnancies After the Diagnosis of Mild Gestational Diabetes Mellitus and Risk of Cardiometabolic Disorders

Varner, Michael W. MD; Rice, Madeline Murguia PhD; Landon, Mark B. MD; Casey, Brian M. MD; Reddy, Uma M. MD, MPH; Wapner, Ronald J. MD; Rouse, Dwight J. MD; Tita, Alan T. N. MD, PhD; Thorp, John M. MD; Chien, Edward K. MD, MBA; Saade, George R. MD; Peaceman, Alan M. MD; Blackwell, Sean C. MD; Vandorsten, J. Peter MD; for the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network

Author Information
doi: 10.1097/AOG.0000000000001863
  • Free

Hormonally mediated physiologic adaptations that occur during pregnancy, including increased insulin resistance, weight gain, atherogenic dyslipidemia, and inflammation, unmask a woman's predisposition to subsequent chronic diseases.1–3 Women who develop gestational diabetes mellitus (GDM), for example, are known to be at increased risk of developing type II diabetes and cardiometabolic disorders, including metabolic syndrome.4 Because it can often be diagnosed in a relatively short follow-up interval after pregnancy, and because it is associated with increased cardiovascular and metabolic disease risks, the diagnosis of metabolic syndrome in women who have experienced the aforementioned pregnancy complications has also been considered as a proxy for subsequent cardiometabolic disease.5,6

Pregnancy complications and their associated future diseases share common pathophysiologies of inflammation and endothelial dysfunction.7,8 If the pregnancy-associated changes play an etiologic role in subsequent cardiometabolic disease, we hypothesized that additional pregnancies occurring after a prior diagnosis of GDM might increase the likelihood of subsequent metabolic syndrome.

The objective of this analysis was to assess the association of subsequent pregnancy with maternal cardiometabolic disorders at 5–10 years after an index pregnancy complicated by mild GDM.


Women enrolled from October 2002 through mid-November 2007 in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network's Mild GDM Trial were followed up 5–10 years later from February 2012 through September 2013. Details regarding the parent study have been previously described.9 Mild GDM was defined with a 3-hour 100-g oral glucose tolerance test (OGTT) with a fasting glucose of less than 95 mg/dL but two of three timed measurements that exceeded established thresholds: 1 hour, 180 mg/dL; 2 hours, 155 mg/dL; and 3 hours, 140 mg/dL. Women did not have a history of pregestational or GDM, renal or cardiovascular disease, or chronic hypertension nor did they have preeclampsia at the time of enrollment in the original study. The mild GDM Trial was approved by the institutional review board of all participating centers.

The prospective observational follow-up study,10 which was not planned at the time of the original study, was approved by the institutional review boards of all participating centers. Eligibility included enrollment in the original GDM study at a center still participating in the Maternal-Fetal Medicine Units Network at the time of the follow-up study (12/16 centers throughout the United States; 94% of the original trial patients), and the woman's index child participated in the follow-up study. After informed consent, women underwent anthropometric and blood pressure measurements and were queried about breastfeeding during the index pregnancy, pregnancies since their index pregnancy, medication use, physical activity, and diet. After an 8-hour fast, participants had their blood drawn followed by a 2-hour 75-g OGTT. Collection, processing, and storage of specimens were conducted per a standardized approach before being shipped on dry ice to the Northwest Lipid Metabolism and Diabetes Research Laboratories for lipid panel and glucose laboratory measurements. Weight and height were measured using a hospital-grade scale and a stationary stadiometer, respectively. Waist circumference was measured by trained personnel who received standardized anthropometric training using an AcuFitness MyoTape just above the uppermost lateral border of the right ilium of the pelvis,11 and an average of three measurements was used in the analysis. Blood pressure was measured by trained personnel after the participant had not smoked or consumed coffee or alcohol for at least 30 minutes and was sitting quietly for at least 10 minutes by auscultation using aneroid sphygmomanometer instruments or a hospital-grade blood pressure–pulse machine, and an average of two measurements was used in the analysis. To minimize bias, research staff involved in data collection and laboratory analyses were masked to the participants' exposures during the original study.

The primary outcome, metabolic syndrome, was defined per the American Heart Association and National Heart Lung and Blood Institute Scientific Statement as three or more of the following: high waist circumference (greater than 88 cm), high triglycerides (150 mg/dL or greater), low high-density lipoprotein cholesterol (less than 50 mg/dL), blood pressure 130/85 mm Hg or greater, or fasting glucose 100 mg/dL or greater or relevant treatment for any of these components.12 Diabetes mellitus outside of pregnancy was a secondary outcome and was defined as currently treated for diabetes with an oral agent or insulin or meeting American Diabetes Association diagnostic criteria for an abnormal 75-g glucose tolerance test.13 The primary exposure was the number of subsequent pregnancies since the index pregnancy and a secondary exposure was number of subsequent pregnancies that were, or were not, complicated with GDM.

Descriptive analyses used the χ2 test for categorical variables and the Kruskal-Wallis test for continuous variables. Components of the metabolic syndrome, OGTT results, and C-peptide were evaluated as continuous outcomes by constructing least squares means based on linear multivariable regression. Metabolic syndrome, diabetes mellitus outside of pregnancy, and binomial components of metabolic syndrome were evaluated as binomial outcomes using log-Poisson multivariable regression to estimate relative risks and 95% confidence intervals. The multivariable models were adjusted for race–ethnicity, mild GDM treatment group in the index pregnancy, pregnancy-associated hypertension (gestational hypertension or preeclampsia) in the index pregnancy, maternal age at follow-up, current self-reported daily servings of vegetables other than potatoes or corn, and current log body mass index. Continuous variables were assessed to evaluate whether they were normally distributed and log-transformed when appropriate. SAS software was used for the analyses. All tests were two-tailed and P<.05 was used to define statistical significance. No imputation for missing data was performed.


Of the 905 eligible women with mild GDM from the original trial, 666 (74%) were successfully contacted, of whom 483 women consented and participated in the follow-up study. The maternal baseline characteristics were generally similar between women who did and did not participate in the follow-up study, although a higher percentage of the non-Hispanic white women participated in the follow-up study (Appendix 2, available online at Of the 483 women who were included in the follow-up, 26 were pregnant at the time of follow-up and 31 had missing laboratory data, leaving 426 included in this analysis. Of these, 212 had no subsequent pregnancies since the index pregnancy (group A), 143 women had only one subsequent pregnancy (group B), and 71 women had two or more subsequent pregnancies (group C). There were 309 subsequent pregnancies among the 214 women in groups B and C, eight of which were multiple gestations (twins).

Demographics of the three groups at the time of enrollment in the parent study and at the time of this follow-up evaluation are presented in Tables 1 and 2. Overall, 59% of the participants were Hispanic. Women in group A also had higher body mass indices than did women in groups B and C. There were no differences between groups in treatment group, race–ethnicity, or frequency of pregnancy-associated hypertension during the index pregnancy. When including all pregnancies before and after the index pregnancies, women in group C had higher median parity.

Table 1.
Table 1.:
Demographics of Follow-up Participants at the Time of Enrollment in the Parent Study, October 2002–November 200713
Table 2.
Table 2.:
Demographic and Lifestyle Information For Participants at Follow-up, February 2012–September 201314

At the time of enrollment in the follow-up study, women in group A were significantly older (median 38 years) than women in groups B (median 35 years) and C (median 33 years) (P<.001). The interval from participation in the parent study to participation in the follow-up study was longer for women in group C (median 8 years) than for women in groups A or B (median for both 7 years) (P=.03). There were no other significant differences among the groups in any other self-reported demographic information, smoking status, dietary intake, minutes per week of vigorous physical activity, or breastfeeding in the index pregnancy. Almost one third (69/214 [32%]) of women with subsequent pregnancies (groups B and C) had another pregnancy complicated with GDM.

Results for laboratory markers of vascular risk, by group and adjusted for various potential confounders, are presented in Table 3. Median unadjusted insulin concentrations were not significantly different between groups, although mean adjusted insulin was highest in group A. No other significant differences were observed.

Table 3.
Table 3.:
Adjusted* Results for Laboratory and Physical Examination Markers of Vascular Risk by Number of Pregnancies After the Index Gestational Diabetes Mellitus Pregnancy

The frequency of metabolic syndrome and diabetes mellitus at the time of follow-up evaluation was 32.9% (140/426) and 8.0% (34/426), respectively (Table 4). Among the 34 women diagnosed with diabetes, five were being treated with an oral agent or insulin before the follow-up visit; the remainder (n=29, 6.8% of all participants) were diagnosed with an abnormal 75-g OGTT at the follow-up visit.

Table 4.
Table 4.:
Adjusted* Results for Metabolic Syndrome and Its Components and Diabetes Mellitus by Number of Pregnancies After the Index Gestational Diabetes Mellitus Pregnancy
Table 4-A.
Table 4-A.:
Adjusted* Results for Metabolic Syndrome and Its Components and Diabetes Mellitus by Number of Pregnancies After the Index Gestational Diabetes Mellitus Pregnancy

The specific components of metabolic syndrome in each group are shown in Table 4. In all three groups the diagnosis of metabolic syndrome was primarily based on waist circumference and low high-density lipoprotein cholesterol. There were no differences among the groups in the diagnosis of metabolic syndrome (34%, 33%, 30%, respectively; relative risks [RR] 1.15 [95% confidence interval {CI} 0.79–1.69] and 1.00 [0.60–1.68] for the second and third groups, respectively, compared with the first group); however, there was a difference among groups in diabetes (5.2%, 10.5%, 11.3%, respectively; RRs 2.62 [95% CI 1.16–5.91] and 2.83 [1.06–7.59] for the second and third groups, respectively, compared with the first group). The association with diabetes was driven mostly by having a subsequent pregnancy complicated with GDM, with RRs 1.99 (95% CI 0.82–4.84) and 3.75 (1.60–8.82) for subsequent pregnancies without or with GDM, respectively, compared with women who did not have subsequent pregnancies (Table 5). Body mass index was the one risk factor consistently associated with all of the outcomes evaluated.

Table 5.
Table 5.:
Adjusted* Results for Metabolic Syndrome and Diabetes Mellitus by Whether or Not Pregnancies After the Index Gestational Diabetes Mellitus Pregnancy Were Complicated With Gestational Diabetes Mellitus


We found that pregnancies after a diagnosis of mild GDM did not increase a woman's risk of developing metabolic syndrome but were associated with an increased likelihood of diabetes mellitus outside of pregnancy at 5–10 years after the index pregnancy. The magnitude of the association with postgestational diabetes was strongest in those having a subsequent pregnancy complicated with GDM. Although this could be consistent with either a stronger genetic or an environmental predisposition to type II diabetes, it could also represent confounders that were not measured in this prospective observational follow-up study.

In women with a previous diagnosis of mild GDM, metabolic syndrome was common, being diagnosed in 32.9% of these women at 5–10 years. Subsequent pregnancies were not associated with an increased likelihood of developing metabolic syndrome at 5–10 years follow-up but were associated with an increased incidence of diabetes mellitus outside of pregnancy over the same interval. The association with diabetes was driven mostly by subsequent pregnancy(ies) complicated with GDM.

Women whose pregnancies are complicated by GDM are known to be at increased risk for subsequent metabolic syndrome and other long-term vascular disease. Several previous reports have identified metabolic syndrome rates in parous women with a prior diagnosis of GDM. Verma et al14 evaluated 106 women with a previous diagnosis of GDM at 11 years after delivery and identified metabolic syndrome in 29 (27.2%). A subsequent report from Denmark evaluated 481 women with a previous diagnosis of GDM at a median of 9.8 years (interquartile range 6.4–17.2 years) after delivery and diagnosed metabolic syndrome in 38.4%.15 Most recently, Li et al16 reported on 1,263 Chinese women with a previous diagnosis of GDM with 23.8% of them having a diagnosis of metabolic syndrome at 1–5 years after delivery. These incidence figures (23.8–38.4%) are similar to our findings (32.9%). However, none of these reports provided data on the number of intervening pregnancies from the index pregnancy to the time of metabolic syndrome diagnosis. Although we observed a high overall frequency of metabolic syndrome in our cohort, our data suggest that subsequent pregnancies do not increase a woman's risk of developing metabolic syndrome. Our data are limited, however, by the fact that we do not know how many patients had metabolic syndrome at the time of the index pregnancy.

Gunderson et al17 have demonstrated that waist circumference increases with each birth, whereas body weight increased only after the first birth. However, we did not identify any significant differences in waist circumference among our study groups, perhaps because our group A women older less likely to be nulliparous at entry into the parent study, but our group C women had higher overall parity by the time of the follow-up study.

The fact that participation in the parent study was limited to women with mild GDM could be seen as a limitation, particularly in light of the relatively low incidence of diabetes at the time of follow-up. Likewise, these findings may not apply to women with GDM diagnosed early in pregnancy. However, our frequency of metabolic syndrome is high and consistent with other reports in the literature.14–16 Still, our results may not be generalizable to the full spectrum of GDM. In addition, this is a relatively small retrospective study that reported on less than half (426/905) of the participants in the parent study. We can thus not exclude the possibility of a selection bias. A higher percentage of the non-Hispanic white women participated in the follow-up study. Despite this, 59% of the women included in this analysis were Hispanic. Another limitation of this report is the lack of metabolic syndrome data from both the index pregnancy and before the index pregnancy. Likewise, given the lack of contact with study participants between the parent and follow-up studies, it is not possible to know when women with metabolic syndrome developed the condition, specifically in relation to any intervening pregnancies. Lastly, our limited significant findings should be interpreted in the context of multiple comparisons. Strengths of this report include the parent study's precise definitions of mild GDM and of metabolic syndrome components and diabetes in the follow-up study.


1. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 2002;25:1862–8.
2. Bellamy L, Casas JP, Hingorani AD, Williams DJ. Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis. BMJ 2007;335:974.
3. McDonald SD, Malinowski A, Zhou Q, Yusuf S, Devereaux PJ. Cardiovascular sequelae of preeclampsia/eclampsia: a systematic review and meta-analyses. Am Heart J 2008;156:918–30.
4. Sattar N, Greer IA. Pregnancy complications and maternal cardiovascular risk: opportunities for intervention and screening? BMJ 2002;325:157–60.
5. Isomaa B, Almgren P, Tuomi T, Forsén B, Kahti K, Nissén M, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care 2001;24:683–9.
6. Wilson PW, D'Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation 2005;112:3066–72.
7. Rodie VA, Freeman DJ, Sattar N, Greer IA. Pre-eclampsia and cardiovascular disease: metabolic syndrome of pregnancy? Atherosclerosis 2004;175:189–202.
8. Craici I, Wagner S, Garovic VD. Preeclampsia and future cardiovascular risk: formal risk factor or failed stress test? Ther Adv Cardiovasc Dis 2008;2:249–59.
9. Landon MB, Spong CY, Thom E, Carpenter MW, Ramin SM, Casey B, et al. A multicenter, randomized trial of treatment for mild gestational diabetes. N Engl J Med 2009;361:1339–48.
10. Landon MB, Rice MM, Varner MW, Casey BM, Reddy UM, Wapner RJ, et al. Mild gestational diabetes mellitus and long-term child health. Diabetes Care 2015;38:445–52.
11. National Health and Nutrition Examination Survey (NHANES) anthropometry procedures manual, January 2009. Available at: Retrieved December 21, 2015.
12. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement [published erratum appears in Circulation 2005;112:e297, e298]. Circulation 2005;112:2735–52.
13. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2011;34(suppl 1):S62–9.
14. Verma A, Boney CM, Tucker R, Vohr BR. Insulin resistance syndrome in women with prior history of gestational diabetes. J Clin Endocrinol Metab 2002;87:3227–35.
15. Lauenborg J, Mathiesen E, Hansen Glümer TC, Jørgensen T, Borch-Johnsen K, Hornnes P, et al. The prevalence of the metabolic syndrome in a Danish population of women with previous gestational diabetes mellitus is three-fold higher than in the general population. J Clin Endocrinol Metab 2005;90:4004–10.
16. Li W, Liu H, Qiao Y, Lv F, Zhang S, Wang L, et al. Metabolic syndrome of weight change from pre-pregnancy to 1–5 years post-partum among Chinese women with prior gestational diabetes. Diabet Med 2015;32:1492–9.
17. Gunderson EP, Murtaugh MA, Lewis CE, Quesenberry CP, West DS, Sidney S. Excess gains in weight and waist circumference associated with childbearing: the Coronary Artery Risk Development in Young Adults Study (CARDIA). Int J Obes Relat Metab Disord 2004;28:525–35.

Supplemental Digital Content

© 2017 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.