ORIGINAL
ARTICLE
Weight Gain and Progression to Type 2 Diabetes in Women With a History of Gestational Diabetes Mellitus Joon Ho Moon,* Soo Heon Kwak,* Hye Seung Jung, Sung Hee Choi, Soo Lim, Young Min Cho, Kyong Soo Park, Hak C. Jang, and Nam H. Cho Department of Internal Medicine (J.H.M., S.H.K., H.S.J., S.H.C., S.L., Y.M.C., K.S.P., H.C.J.), Seoul National University College of Medicine, Seoul 110-744, Korea; Department of Internal Medicine (J.H.M., S.H.K., H.S.J., Y.M.C., K.S.P.), Seoul National University Hospital, Seoul 110-799, Korea; and Department of Internal Medicine (S.H.C., S.L., H.C.J.), Seoul National University Bundang Hospital, Seongnam 463-707, Korea; and Department of Preventive Medicine (N.H.C.), Ajou University School of Medicine, Suwon 443-721, Korea
Context: The effect of weight gain on the development of type 2 diabetes after gestational diabetes mellitus (GDM) is not fully understood in Asian women who have a relatively low body mass index (BMI). Objective: We investigated the effect of postpartum longitudinal BMI change on the development of diabetes in Korean women with a history of GDM. Design and Setting: The study included a hospital-based, multicenter, prospective cohort with median follow-up of 4.0 years. Participants: A total of 418 women with previous GDM or gestational impaired glucose tolerance were recruited and underwent an oral glucose tolerance test at 6 weeks postpartum and annually thereafter. Main Outcome Measure: The risk of diabetes was analyzed according to the tertiles of BMI change. Changes in BMI were calculated between the initial postpartum visit and the last follow-up or at the onset of diabetes. Results: The BMI change in each tertile was ⫺1.8 ⫾ 1.1, ⫺0.2 ⫾ 0.3, and 1.6 ⫾ 1.2 kg/m2, respectively. We observed an increased risk of incident diabetes as the tertile of BMI change increased (8.6%, 12.6%, and 16.9%, P ⫽ .039). Postpartum BMI change was an independent predictor of diabetes in a multivariate Cox analysis (hazard ratio 1.27, 95% confidence interval 1.04 –1.56, P ⫽ .021), even after adjusting for BMI at the last follow-up. In the highest tertile group, there was a significant deterioration in cardiovascular risk factors including blood pressure, lipid profile, and insulin sensitivity. Conclusions: Postpartum increase in BMI is significantly associated with a risk of diabetes and deterioration of metabolic phenotypes in Korean GDM women. (J Clin Endocrinol Metab 100: 3548 –3555, 2015)
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in USA Copyright © 2015 by the Endocrine Society Received January 13, 2015. Accepted July 9, 2015. First Published Online July 14, 2015
3548
press.endocrine.org/journal/jcem
* J.H.M. and S.H.K. contributed equally to this work. Abbreviations: BMI, body mass index; BP, blood pressure; CI, confidence interval; DPP, Diabetes Prevention Program; GDM, gestational diabetes mellitus; GIGT, gestational impaired glucose tolerance; HDL, high-density lipoprotein; HOMA-B, homeostasis model assessment of -cell function; HOMA-IR, homeostasis model assessment of insulin resistance; HR, hazard ratio; IGR, impaired glucose regulation; LDL, low-density lipoprotein; OGTT, oral glucose tolerance test.
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
doi: 10.1210/JC.2015-1113
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
he worldwide increase in the prevalence of type 2 diabetes is paralleled by the epidemic of increasing obesity (1). There is a large body of evidence that weight change by itself, in addition to increased body mass index (BMI), has a significant impact on the risk of diabetes and cardiovascular risk factors. In the Nurses’ Health Study, it has been reported that gaining more than 5 kg of body weight increased the risk of developing diabetes by 1.9 times (2). In this regard, interventional trials were conducted to demonstrate the beneficial effect of weight loss on the development of diabetes. The Diabetes Prevention Program (DPP) was a multicenter randomized clinical trial, which aimed to demonstrate that weight loss through lifestyle intervention or metformin could prevent or delay the onset of type 2 diabetes in a high-risk population of impaired glucose tolerance. In the DPP, weight loss was associated with reduced risk of diabetes by 16% for every kilogram of weight loss (3). Women with a previous history of gestational diabetes mellitus (GDM) have an increased risk of developing type 2 diabetes (4, 5). The prevalence of postpartum type 2 diabetes after GDM ranges from 2.6% to greater than 70% (5), and it varies widely by ethnicity and diagnostic methods that are used (6, 7). Various risk factors for the progression to type 2 diabetes include age of the mother (8); prepregnancy (9, 10) and postpartum (11) BMI; a family history of diabetes (12); breast-feeding (13); insulin treatment during pregnancy (5, 14); and fasting glucose during pregnancy (15). Among these, BMI is a modifiable risk factor that could be improved by lifestyle intervention. Several studies have investigated the effect of longitudinal weight gain on the development of type 2 diabetes in women with a previous history of GDM (16 –18). Most of the studies were conducted in Europeans or Hispanics. Although the effect of postpartum weight gain on type 2 diabetes is assumed to be similar for different ethnicities, it has not been thoroughly investigated in a prospective manner in Asian women with a previous history of GDM. Asian women have lower BMI thresholds for increased risk of GDM than Caucasians (19), and the progression rate to postpartum diabetes in Asian immigrant women is reported to be higher compared with Europeans (20). In this multicenter prospective cohort study, we evaluated the impact of the change in BMI on the development of type 2 diabetes and related metabolic phenotypes in Korean women with a previous history of GDM.
T
Materials and Methods Study design We conducted a multicenter prospective cohort study among subjects with a first-time diagnosis of GDM and gestational im-
press.endocrine.org/journal/jcem
3549
paired glucose tolerance (GIGT). Subjects were recruited from August 1995 to May 1997 from four centers in Korea. The initial postpartum follow-up visit was performed at 6 weeks postpartum, and annual follow-up visits were made thereafter.
Subjects All pregnant women performed a screening for GDM with a 50-g oral glucose challenge test at 24 –28 weeks of gestation, and a positive screen was defined as a 1-hour glucose value of 130 mg/dL or greater. Women with a positive screen were given a 3-hour, 100-g oral glucose tolerance test (OGTT). The diagnoses of GDM and GIGT were made using the recommendation of the Third International Workshop-Conference on Gestational Diabetes Mellitus (21). Two or more of the following criteria for GDM and one for GIGT were required for the diagnosis: fasting plasma glucose of 105 mg/dL or greater, 1-hour glucose of 190 mg/dL or greater, 2-hour glucose of 165 mg/dL or greater, or 3-hour glucose of 145 mg/dL or greater. Women with GDM or GIGT who visited the initial postpartum evaluation were enrolled for this study (n ⫽ 1050). Subjects who had persistent diabetes at the initial visit (n ⫽ 113) were excluded. Among the remaining 937 eligible subjects, a total of 418 subjects completed follow-up examinations more than once and had data on BMI and OGTT to diagnose postpartum type 2 diabetes. Study participants (n ⫽ 418) were generally representing potentially eligible subjects (n ⫽ 937), and there was no significant difference in clinical characteristics except for the decreased systolic blood pressure (BP) in the study participants compared with those who did not make follow-up visits (Supplemental Table 1). All subjects participated voluntarily, and informed consent was obtained from each subject. This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (Institutional Review Board number B-1007– 105-007) and was conducted according to the Declaration of Helsinki (22).
Postpartum follow-up examinations Obstetric history including prepregnancy weight, weight at the diagnosis of GDM or GIGT, gestational weight gain, and pregnancy outcome was retrieved from medical records. Faceto-face interviews were conducted at the first postpartum examination using a standardized questionnaire that included past medical and reproductive history and family history of diabetes. For the postpartum follow-up assessment, a standard 75-g OGTT was performed at each follow-up visit. The diagnosis of diabetes was based on the criteria of the American Diabetes Association (23). Impaired glucose regulation (IGR) was defined as a combination of impaired fasting glucose and impaired glucose tolerance. A recent clinical history of diagnosis of diabetes or antidiabetic medication was reviewed at each visit. Physical activity data were collected during the postpartum follow-up visits and were classified as mild, moderate, and vigorous intensity (24). Anthropometric measures including body weight, height, BP, and body composition were measured at each visit. Body weight was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm. BP was measured three times on the left arm by research nurse with mercury sphygmomanometer (Baumanometer; Baum) to the nearest 2 mm Hg. Averaged values were used for analyses. Bioelectrical impedance analysis
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3550
Moon et al
Weight Gain and Future Diabetes in Women With GDM
(body composition analyzer; Giru Co) was performed to evaluate body composition.
Metabolic assessment Metabolic assessment including a 75-g OGTT and fasting serum lipid concentration (total cholesterol, high density lipoprotein [HDL] cholesterol, and triglycerides) was performed at each visit. The homeostasis model assessment of -cell function (HOMA-B) and insulin resistance (HOMA-IR) were calculated (25). Insulin sensitivity was assessed by the Matsuda index (26) as follows: 10 000/⻫([fasting glucose] ⫻ [fasting insulin] ⫻ [mean glucose] ⫻ [mean insulin]). The insulinogenic index (27) was used to estimate insulin secretion as follows: (insulin [30 min] ⫺ insulin [0 min])/(glucose [30 min] ⫺ glucose [0 min]). The disposition index (28) was used to evaluate the composite of insulin secretion considering the degree of insulin sensitivity, as follows: (Matsuda index) ⫻ (insulinogenic index). A 75-g OGTT was performed after a more than 8-hour overnight fast. Plasma glucose was enzymatically measured by an automated analyzer (YSI 2300-STAT; Yellow Springs Instrument Co) using the glucose oxidase method. Plasma insulin was measured by a RIA (Linco Research Inc). Total cholesterol, triglycerides, and HDL (Beckman analyzer; Beckman Instruments) were measured by an enzymatic assay. Low-density lipoprotein (LDL) was calculated using the Friedewald equation (29).
Statistical analysis Postpartum BMI change was defined as the difference between the initial postpartum follow-up and the last visit of examination or at the onset of diabetes. Subjects were divided into tertiles for analysis. Differences among groups were evaluated using an ANOVA for continuous variables and the 2 test and linear-by-linear association for categorical variables. We used the Cox proportional hazard model to assess the difference in incident type 2 diabetes by postpartum BMI change and to adjust for differences in follow-up duration. Subjects were censored at the onset of diabetes or at the last visit of examination (or when lost to follow-up). For the primary analysis, age and initial postpartum BMI were used as covariates in the Cox model. In the secondary analysis, variables that showed a significant association in the univariate analysis and age were used as covariates. Data are presented as the mean ⫾ SD for continuous variables in Gaussian distribution or as the median (interquartile range). Hazard ratios (HRs) are presented with their corresponding 95% confidence intervals (CIs) and P values. Differences were considered significant when P ⬍ .05. Statistical analyses were conducted using SPSS version 20.0 (IBM Co).
Results Prepregnancy BMI and development of diabetes A total of 418 subjects (283 of GDM and 135 of GIGT) were included in the analysis. Because prepregnancy BMI in Korean women with GDM was lower than that in European or Hispanic women with GDM, we first investigated the association between prepregnancy BMI and risk of diabetes. Subjects were divided into three groups by the tertiles of prepregnancy BMI (cutoff 20.6 and 23.2 kg/m2).
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
A total of 53 subjects (12.7%) were found to have type 2 diabetes on follow-up. Within each tertile 12 (8.8%), 15 (10.9%), and 26 (19.0%) subjects developed type 2 diabetes, and the increasing trend was statistically significant (P for trend ⫽ .012). Postpartum BMI change and development of diabetes Next, we investigated whether postpartum BMI change is associated with the development of type 2 diabetes. Subjects were followed up for a median of 4.0 (interquartile range 2.6 –5.0) years. The overall mean postpartum BMI change was ⫺0.1 ⫾ 1.7 kg/m2. The mean age of subjects at the last visit (or at the onset of diabetes) was 36.0 ⫾ 4.6 years. Subjects were categorized into tertiles of postpartum BMI change (cutoff ⫺0.7 and ⫹0.4 kg/m2). Subjects in the first tertile had a decreased BMI of mean ⫺1.8 kg/ m2, whereas subjects in the second and third tertiles had maintained or increased BMIs by a mean ⫺0.2 and ⫹1.6 kg/m2. The baseline characteristics of the subjects at the initial postpartum follow-up according to the tertile are described in Table 1. There were no significant differences among the three groups except for the increased initial postpartum BMI in the first tertile group. In each tertile of postpartum BMI change, the number of subjects who developed type 2 diabetes was 12 (8.6%), 18 (12.6%), and 23 (16.9%), respectively, and the increasing trend was statistically significant (P for trend ⫽ .039). The incidence of type 2 diabetes among the three groups was evaluated by the Cox proportional hazard model. Subjects in the third tertile showed an increased risk of progression to type 2 diabetes compared with the subjects in the first tertile (HR 2.23, 95% CI 1.10 – 4.53, P ⫽ .026) after adjusting for age and initial postpartum BMI (Figure 1). We additionally performed stratified analysis according to last follow-up BMI, considering its significant correlation with postpartum BMI change (r ⫽ 0.364, P ⬍ .001). In each tertile of the last follow-up BMI, there was an increasing trend of incident diabetes as postpartum BMI change increased (Supplemental Table 2 and Supplemental Figure 1). Postpartum BMI change as an independent risk factor for incident type 2 diabetes We further explored the association between various risk factors and incident type 2 diabetes using a multivariate Cox proportional hazard model. In the model, factors that showed a significant association with postpartum type 2 diabetes in the univariate Cox analysis and age were included (Table 2). Postpartum BMI change was an independent risk factor for type 2 diabetes (HR 1.27, 95% CI
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
Table 1.
press.endocrine.org/journal/jcem
3551
Baseline Clinical Characteristics and Biochemical Parameters at Initial Postpartum Follow-Up Postpartum BMI Change
Age, y BMI, kg/m2 Weight, kg Duration of follow-up, y GIGT, % Family history of diabetes, % Breast-feeding, % Postpartum exercisec Insulin treatment during pregnancy, % Systolic BP, mm Hg Fasting glucose, mg/dL Two-hour glucose, mg/dL Fasting insulin, mU/mL HOMA-IR HOMA-B Total cholesterol, mg/dL Triglycerides, mg/dL HDL, mg/dL LDL, mg/dL Body fat percentage by BIA, %
First Tertile (n ⴝ 139)
Second Tertile (n ⴝ 143)
Third Tertile (n ⴝ 136)
P Value
32.1 ⫾ 4.6 24.1 ⫾ 3.4 59.9 ⫾ 9.1 3.9 ⫾ 1.8 33.1 46.8 51.8 55.4 21.2 112.5 ⫾ 11.4 93.3 ⫾ 10.1 129.1 ⫾ 27.4 9.7 ⫾ 3.9 2.2 ⫾ 1.0 127.0 ⫾ 59.9 199.4 ⫾ 34.9 118.9 ⫾ 68.3 53.7 ⫾ 13.7 122.0 ⫾ 30.1 31.4 ⫾ 5.5
32.2 ⫾ 4.4 22.8 ⫾ 2.8 56.3 ⫾ 6.8 3.8 ⫾ 2.0 30.8 52.4 41.3 55.2 21.9 110.3 ⫾ 12.2 92.3 ⫾ 10.1 124.8 ⫾ 30.8 9.3 ⫾ 4.4 2.1 ⫾ 1.1 133.2 ⫾ 130.5 198.9 ⫾ 33.6 127.7 ⫾ 76.1 53.1 ⫾ 14.7 120.2 ⫾ 27.8 30.2 ⫾ 5.5
32.0 ⫾ 4.0 23.1 ⫾ 3.1 57.2 ⫾ 8.4 4.2 ⫾ 1.7 33.1 41.9 52.2 51.5 23.1 109.8 ⫾ 12.0 92.3 ⫾ 10.3 123.3 ⫾ 27.2 9.5 ⫾ 5.0 2.2 ⫾ 1.3 133.5 ⫾ 88.7 197.4 ⫾ 39.3 116.7 ⫾ 62.1 54.0 ⫾ 14.4 119.9 ⫾ 33.9 29.9 ⫾ 4.7
.959 .001a,b .001a,b .171 .982 .211 .957 .516 .706 .133 .632 .211 .693 .711 .821 .883 .373 .855 .836 .058
Abbreviation: BIA, bioelectrical impedance analysis. Subjects were divided into tertiles of postpartum BMI change. P values are for ANOVA (continuous) or linear-by-linear association (dichotomous). Post hoc analysis was performed for variables with P ⬍ .05. a
P ⬍ .05 for first vs second tertile.
b
P ⬍ .05 for first vs third tertile.
c
Exercise was defined as moderate- or vigorous-intensity physical activities.
1.04 –1.56, P ⫽ .021), even after adjusting for other risk factors including prepregnancy BMI and last follow-up BMI. The prepregnancy BMI and fasting glucose during pregnancy were also independent risk factors for the development of type 2 diabetes.
mental Table 4). Improvement of glucose tolerance status, ie, from IGR to normal glucose tolerance, was most frequent in the first tertile group, and deterioration was most frequent in the third tertile group, (P ⬍ .001 by 2 test).
Subgroup analysis and overall glucose tolerance As there might be a residual confounding effect of difference in follow-up duration among the three groups, we performed a sensitivity analysis by dividing our subjects according to follow-up duration of 2 years or more. We observed that incidence of diabetes increased as BMI change increase in the both subgroups (Supplemental Table 3). Among the 135 GIGT subjects, there were significantly fewer subjects who developed diabetes (n ⫽ 5, 3.7%) during the postpartum follow-up. After excluding GIGT subjects, the postpartum BMI change remained associated with the development of type 2 diabetes. Among 283 women with GDM, the number of subjects who progressed to type 2 diabetes in each tertile was 10 (10.8%), 18 (18.2%), and 20 (22.0%), respectively, and the increasing trend was significant (P for trend ⫽ .043). The change in overall glucose tolerance status during follow-up varied by postpartum BMI change (Supple-
Postpartum BMI change and related metabolic phenotypes We examined the change in metabolic phenotypes according to the tertiles of BMI change and observed that increased BMI was associated with deteriorating metabolic phenotypes (Table 3). Subjects in the third tertile had a significantly higher increase in 2-hour glucose and fasting insulin compared with those in the first tertile. BP and lipid profile deteriorated during the follow-up in the third tertile subjects compared with the first tertile subjects. Body fat percentage change also differed by postpartum BMI change, even in subjects of the first or second tertile of last follow-up BMI (Supplemental Table 5). We observed that insulin sensitivity, assessed by the Matsuda index, and composite insulin secretion, assessed by the disposition index, were significantly deteriorated in third tertile subjects who gained weight (Supplemental Figure 2). These differences in metabolic phenotypes by BMI change were maintained after adjusting for last follow-up BMI (data not shown).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3552
Moon et al
Weight Gain and Future Diabetes in Women With GDM
Figure 1. Cumulative incidence of type 2 diabetes mellitus by postpartum BMI change. Subjects were divided into tertiles of postpartum BMI change. A Cox proportional hazard model was used for the analysis (adjusted for age and BMI on initial follow-up). The incidence of type 2 diabetes was higher in the third tertile compared with the first tertile (HR 2.23, 95% CI 1.10 – 4.53, P ⫽ .026).
Discussion In this observational cohort study, postpartum BMI change during 4 years of follow-up in women with previous GDM was associated with a significant risk of type 2 diabetes. Postpartum BMI change was an independent risk factor for incident diabetes in Cox analysis, even after adjusting for multiple risk factors including prepregnancy BMI and last follow-up BMI. We demonstrated an association between postpartum BMI change and metabolic phenotypes including BP, insulin sensitivity, and lipid proTable 2.
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
files. To the best of our knowledge, this is one of the first prospective studies that comprehensively assessed the impact of postpartum BMI change on the development of type 2 diabetes and its related metabolic phenotypes in Asian women. Obesity is a well-known modifiable risk factor for type 2 diabetes. Several pivotal studies have reported the association between long-term weight change and the risk of diabetes in both general population and women with history of GDM. In the Nurses’ Health Study, which involved middle-aged women without a history of GDM, those who had weight gain of 5.0 –7.9 kg had an increased relative risk of developing type 2 diabetes by 1.9 (95% CI 1.5–2.3) compared with those who had less than 5 kg of weight gain during the 14 years of follow-up (2). Regarding women with a history of GDM, a prospective observational study by Peters et al (17) that included 666 Hispanics explicitly showed that weight gain increased the risk of type 2 diabetes with an HR of 1.54 (95% CI 1.33– 1.79) per 4.5 kg (10 lb) during 2.2 years of follow-up. In the subgroup analysis of the DPP trial, intensive lifestyle modification or metformin treatment led to postpartum weight loss and 50% reduction in type 2 diabetes risk in women with a history of GDM (18). To have a better comparison with the study results by Peters et al, we reanalyzed our data and observed that the HR for diabetes was 2.00 (95% CI 1.05–3.81) per 4.5 kg (10 lb) weight gain. Although the follow-up duration was longer in our study, it is noteworthy that the risk of diabetes was even higher in Asian women compared with Hispanics for the same degree of weight gain. Whereas the magnitude of the effect of increased weight gain is different according to study design, ethnicity, demographic characteristics, and follow-up duration, it is consistent that weight change is significantly associated with a risk of type 2 diabetes. Our
HRs of Postpartum Type 2 Diabetes by Postpartum BMI Change and Associated Variables by Cox Analysis
Postpartum BMI change Last follow-up BMI Prepregnancy BMI Age Breast-feeding Postpartum exercised Family history of diabetes Parity Fasting glucose during pregnancy
Univariatea
P Value
Multivariateb
P Value
Multivariatec
P Value
1.17 (1.01–1.34) 1.12 (1.04 –1.20) 1.14 (1.05–1.23) 0.95 (0.89 –1.02) 0.50 (0.28 – 0.89) 0.59 (0.35–1.00) 1.28 (0.74 –2.19) 1.00 (0.64 –1.57) 1.03 (1.02–1.04)
.033 .003 .001 .129 .020 .050 .377 .988 ⬍.001
1.27 (1.04 –1.56) 0.90 (0.75–1.07) 1.25 (1.06 –1.47) 0.94 (0.87–1.00) 0.61 (0.33–1.13) 0.49 (0.20 –1.23)
.021 .228 .007 .056 .114 .130
1.17 (1.02 ⫺ 1.35)
.025
1.13 (1.04 ⫺ 1.22)
.002
1.04 (1.02–1.05)
⬍.001
1.32 (1.02–1.04)
⬍.001
Values presented are HR (95% CI). a
HR of postpartum type 2 diabetes for each variable was evaluated with univariate Cox analysis.
b
Factors associated with postpartum type 2 diabetes in univariate Cox analysis and age were included in the multivariate Cox model.
c
Multivariate Cox model only including the significant variables.
d
Exercise was defined as moderate- or vigorous-intensity physical activities.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
Table 3. Change
press.endocrine.org/journal/jcem
3553
Cumulative Incidence of Type 2 Diabetes and Changes of Metabolic Phenotypes by Postpartum BMI Postpartum BMI Change
Type 2 diabetes BMI, kg/m2 Weight, kg Systolic BP, mm Hg Fasting glucose, mg/dL Two-hour glucose, mg/dL Fasting insulin, U/mL HOMA-IR HOMA-B Matsuda index Insulinogenic index Disposition index Total cholesterol, mg/dL Triglycerides, mg/dL HDL, mg/dL LDL, mg/dL Body fat percentage by BIA, %
First Tertile (n ⴝ 139)
Second Tertile (n ⴝ 143)
Third Tertile (n ⴝ 136)
P Value
12 (8.6%) ⫺1.8 ⫾ 1.1 ⫺4.6 ⫾ 5.6 ⫺2.1 ⫾ 10.0 5.4 ⫾ 24.5 5.1 ⫾ 59.8 ⫺1.2 ⫾ 4.8 0.3 ⫾ 1.3 ⫺22.8 ⫾ 66.7 0.5 ⫾ 2.9 0.10 ⫾ 0.52 1.0 ⫾ 3.7 ⫺25.8 ⫾ 36.0 ⫺24.9 ⫾ 55.6 2.2 ⫾ 12.7 ⫺23.0 ⫾ 32.6 ⫺2.7 ⫾ 5.3
18 (12.6%) ⫺0.2 ⫾ 0.3 ⫺0.9 ⫾ 6.0 ⫺0.1 ⫾ 10.0 5.4 ⫾ 15.3 9.0 ⫾ 46.9 0.0 ⫾ 5.1 0.6 ⫾ 2.3 ⫺20.7 ⫾ 134.5 0.0 ⫾ 3.4 0.01 ⫾ 0.38 0.0 ⫾ 2.6 ⫺18.4 ⫾ 34.1 ⫺22.8 ⫾ 84.9 1.0 ⫾ 12.6 ⫺14.7 ⫾ 31.8 ⫺1.0 ⫾ 6.0
23 (16.9%) 1.6 ⫾ 1.1 3.4 ⫾ 5.7 2.0 ⫾ 9.3 10.6 ⫾ 23.9 26.2 ⫾ 59.6 1.0 ⫾ 5.4 0.4 ⫾ 2.2 ⫺22.6 ⫾ 91.0 ⫺1.4 ⫾ 3.2 0.04 ⫾ 0.50 ⫺0.6 ⫾ 6.1 ⫺5.5 ⫾ 34.0 1.4 ⫾ 71.7 ⫺0.8 ⫾ 11.3 ⫺5.0 ⫾ 31.9 1.4 ⫾ 5.0
.039 ⬍.001a,b,c ⬍.001a,b,c .003b .073 .004b,c .002b .531 .983 ⬍.001b,c .441 .013b ⬍.001b,c .004b,c .131 ⬍.001a,b,c ⬍.001a,b,c
Abbreviation: BIA, bioelectrical impedance analysis. Changes of each variable between initial and last postpartum follow-up are presented. P values are for ANOVA (continuous) or linear-by-linear association (dichotomous). Post hoc analysis was performed for variables with P ⬍ .05. a
P ⬍ .05 for first vs second tertile.
b
P ⬍ .05 for first vs third tertile.
c
P ⬍ .05 second vs third tertile.
study confirms that this is also true for Asian women with GDM. The initial postpartum mean BMI was in the range of 22.8 –24.1 kg/m2 for the three groups, and this indicates that our study participants were slightly overweight and mostly not obese according to the widely accepted Asian BMI criteria (30). The weight change during follow-up was also modest: 7.7% decrease (⫺4.6 kg) in the first tertile group and a 5.9% increase (3.4 kg) in the third tertile group. Our finding shows that a modest weight change has a significant impact on the development of type 2 diabetes in Korean women with GDM. We observed a significant negative correlation between initial postpartum BMI and postpartum BMI change (r ⫽ ⫺0.145, P ⫽ .003). It is possible that women with a higher initial postpartum BMI are more likely to lose weight during follow-up. Regarding the last follow-up BMI, it was also associated with a risk of diabetes, and it had a significant positive correlation with postpartum BMI change (r ⫽ 0.364, P ⬍ .001). It should be noted that postpartum BMI change was an independent predictor of incident diabetes, even after adjusting for multiple risk factors including the last follow-up BMI. However, it is possible that other risk factors such as follow-up duration, breastfeeding, and physical activity might have confounded our results. It would be of interest to further investigate the
relative contribution of BMI change and last follow-up BMI on incident diabetes. According to the Korean National Health and Nutrition Examination Survey (31), BMI in women aged 20 –39 years remained stable over time since the 1990s (22.1 kg/m2 in 1998 and 21.7 kg/m2 in 2007, P for trend ⫽ .103). The frequency of obesity (defined as a BMI higher than 25 kg/m2) was also similar (16.5% in 1998 and 13.0% in 2007, P for trend ⫽ .212). It is notable that BMI in Korean young women was stable over time, in contrast to the global epidemic increase in obesity (32). This is expected to be the consequence of weight consciousness among young women in Korea (33). Our study is applicable to Korean women who are mostly not obese currently and can also be extended to Asian women with a relatively low BMI. In the study, the third tertile group of postpartum BMI change had a significantly increased risk of diabetes compared with the first tertile group in Cox analysis. Regarding the second tertile group, all the participants had a stable weight with a change between ⫺0.7 and 0.4 kg. However, the risk for diabetes was also increased by an HR of 2.01, although statistical significance was not met (95% CI 0.95– 4.24, P ⫽ .067). It should also be noted that body fat percentage decreased in subjects who reduced BMI postpartum, even in subjects with low to middle BMI
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3554
Moon et al
Weight Gain and Future Diabetes in Women With GDM
on last follow-up. Taken together, our study implies that women with GDM should try at least not to gain weight to prevent the development of diabetes. Our study included participants who had GIGT but who did not fulfill the criteria of GDM. Because the diagnosis of GIGT was based on one abnormal value of a 100-g OGTT, these women might reflect, in part, those who would be newly diagnosed by the International Association of the Diabetes and Pregnancy Study Groups criteria (34). Among the 135 GIGT women, only five (3.7%) of them progressed to type 2 diabetes, but 51 (37.8%) progressed to IGR. It is not known whether these women with IGR would eventually progress to type 2 diabetes. Furthermore, long-term follow-up studies are required to determine the clinical significance of GIGT in terms of risk of type 2 diabetes. Our study has some limitations. First, the follow-up duration and interval were not uniform in all the study subjects. Nevertheless, the mean duration of follow-up in each BMI change tertile was not significantly different, and we applied the Cox proportional hazard model to overcome this issue. Because we analyzed data from a hospital-based prospective cohort, it is more likely to reflect settings of real clinical practice. Second, due to the observational nature of this study, reverse causation and various confounding factors might have contributed to the results. In our study, changes in the insulin level during postpartum follow-up were higher as the BMI change tertile increased. As suggested by Corkey (35), there is a potential that hyperinsulinemia itself might have resulted in weight gain and aggravation of other metabolic phenotypes, rather than changes in BMI causing detrimental metabolic effects including hypersecretion of insulin. In addition, although we had data on physical activity, we did not have dietary data, which were a major part of lifestyle intervention. In conclusion, this study demonstrated that postpartum BMI change was significantly associated with a risk of type 2 diabetes and related metabolic phenotypes in Korean women with a history of GDM. We found that BMI change during follow-up was an independent risk factor for postpartum diabetes. Our study indicates that women with a history of GDM should try not to gain weight and closely monitor their weight to maintain proper body weight to prevent the development of type 2 diabetes.
Acknowledgments Address all correspondence and requests for reprints to: Hak C. Jang, PhD, Department of Internal Medicine, Seoul National University Bundang Hospital, 82, Gumi-ro, 173 beon-gil, Bundang-gu, Seongnam, Gyeonggi-do, 463-707, Korea. E-mail:
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
[email protected]; or Nam H. Cho, PhD, Department of Preventive Medicine, Ajou University School of Medicine, 164, World Cup-ro, Yeongtong-gu, Suwon, Gyeonggi-do, 443-721, Korea. E-mail: [email protected]. J.H.M. and S.H.K. contributed to researching the data and writing the manuscript. H.S.J., S.H.C, S.L., Y.M.C., K.S.P., H.C.J., and N.H.C. contributed to the discussion and reviewed and edited the manuscript. H.C.J. and N.H.C. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. This work was supported by the Korea Healthcare Technology Research and Development Project, Ministry of Health and Welfare Grant A111362 and the Seoul National University Bundang Hospital Research Fund Grant 02-2010-013. Disclosure Summary: The authors have nothing to disclose.
References 1. Geiss LS, Wang J, Cheng YJ, et al. Prevalence and incidence trends for diagnosed diabetes among adults aged 20 to 79 years, United States, 1980 –2012. JAMA. 2014;312:1218 –1226. 2. Colditz GA, Willett WC, Rotnitzky A, Manson JE. Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med. 1995;122:481– 486. 3. Hamman RF, Wing RR, Edelstein SL, et al. Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes Care. 2006;29: 2102–2107. 4. Jang HC. Gestational diabetes in Korea: incidence and risk factors of diabetes in women with previous gestational diabetes. Diabetes Metab J. 2011;35:1–7. 5. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care. 2002; 25:1862–1868. 6. Ferrara A. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care. 2007;30(suppl 2):S141– S146. 7. Fujimoto W, Samoa R, Wotring A. Gestational diabetes in high-risk populations. Clin Diabetes. 2013;31:90 –94. 8. Grant PT, Oats JN, Beischer NA. The long-term follow-up of women with gestational diabetes. Aust N Z J Obstet Gynaecol. 1986;26:17–22. 9. Catalano PM, Vargo KM, Bernstein IM, Amini SB. Incidence and risk factors associated with abnormal postpartum glucose tolerance in women with gestational diabetes. Am J Obstet Gynecol. 1991; 165:914 –919. 10. Kwak SH, Choi SH, Jung HS, et al. Clinical and genetic risk factors for type 2 diabetes at early or late post partum after gestational diabetes mellitus. J Clin Endocrinol Metab. 2013;98:E744 –E752. 11. Steinhart JR, Sugarman JR, Connell FA. Gestational diabetes is a herald of NIDDM in Navajo women. High rate of abnormal glucose tolerance after GDM. Diabetes Care. 1997;20:943–947. 12. Henry OA, Beischer NA. Long-term implications of gestational diabetes for the mother. Baillieres Clin Obstet Gynaecol. 1991;5:461– 483. 13. Taylor JS, Kacmar JE, Nothnagle M, Lawrence RA. A systematic review of the literature associating breastfeeding with type 2 diabetes and gestational diabetes. J Am Coll Nutr. 2005;24:320 –326. 14. Lobner K, Knopff A, Baumgarten A, et al. Predictors of postpartum diabetes in women with gestational diabetes mellitus. Diabetes. 2006;55:792–797. 15. Coustan DR, Carpenter MW, O’Sullivan PS, Carr SR. Gestational diabetes: predictors of subsequent disordered glucose metabolism. Am J Obstet Gynecol. 1993;168:1139 –1145.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
16. Ehrlich SF, Hedderson MM, Quesenberry CP Jr, et al. Post-partum weight loss and glucose metabolism in women with gestational diabetes: the DEBI Study. Diabet Med. 2014;31:862– 867. 17. Peters RK, Kjos SL, Xiang A, Buchanan TA. Long-term diabetogenic effect of single pregnancy in women with previous gestational diabetes mellitus. Lancet. 1996;347:227–230. 18. Ratner RE, Christophi CA, Metzger BE, et al. Prevention of diabetes in women with a history of gestational diabetes: effects of metformin and lifestyle interventions. J Clin Endocrinol Metab. 2008;93: 4774 – 4779. 19. Hedderson M, Ehrlich S, Sridhar S, Darbinian J, Moore S, Ferrara A. Racial/ethnic disparities in the prevalence of gestational diabetes mellitus by BMI. Diabetes Care. 2012;35:1492–1498. 20. Ignell C, Shaat N, Ekelund M, Berntorp K. The impact of ethnicity on glucose homeostasis after gestational diabetes mellitus. Acta Diabetol. 2013;50:927–934. 21. Metzger BE. Summary and recommendations of the Third International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes. 1991;40:197–201. 22. World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Bull World Health Organ. 2001;79(4):373–374. 23. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(suppl 1):S62–S69. 24. Sallis JF, Haskell WL, Wood PD, et al. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol. 1985;121: 91–106. 25. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and -cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412– 419. 26. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from
press.endocrine.org/journal/jcem
27.
28.
29.
30.
31.
32.
33.
34.
35.
3555
oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care. 1999;22:1462–1470. Tura A, Kautzky-Willer A, Pacini G. Insulinogenic indices from insulin and C-peptide: comparison of -cell function from OGTT and IVGTT. Diabetes Res Clin Pract. 2006;72:298 –301. Retnakaran R, Qi Y, Goran MI, Hamilton JK. Evaluation of proposed oral disposition index measures in relation to the actual disposition index. Diabet Med. 2009;26:1198 –1203. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499 – 502. World Health Organization Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157–163. Khang YH, Yun SC. Trends in general and abdominal obesity among Korean adults: findings from 1998, 2001, 2005, and 2007 Korea National Health and Nutrition Examination Surveys. J Kor Med Sci. 2010;25:1582–1588. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980 –2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384:766 –781. Wardle J, Haase AM, Steptoe A. Body image and weight control in young adults: international comparisons in university students from 22 countries. Int J Obes (Lond). 2006;30:644 – 651. International Association of Diabetes Pregnancy Study Groups Consensus Panel, Metzger BE, Gabbe SG, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33:676 – 682. Corkey BE. Banting Lecture 2011: hyperinsulinemia: cause or consequence? Diabetes. 2012;61:4 –13.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
ARTICLE
Weight Gain and Progression to Type 2 Diabetes in Women With a History of Gestational Diabetes Mellitus Joon Ho Moon,* Soo Heon Kwak,* Hye Seung Jung, Sung Hee Choi, Soo Lim, Young Min Cho, Kyong Soo Park, Hak C. Jang, and Nam H. Cho Department of Internal Medicine (J.H.M., S.H.K., H.S.J., S.H.C., S.L., Y.M.C., K.S.P., H.C.J.), Seoul National University College of Medicine, Seoul 110-744, Korea; Department of Internal Medicine (J.H.M., S.H.K., H.S.J., Y.M.C., K.S.P.), Seoul National University Hospital, Seoul 110-799, Korea; and Department of Internal Medicine (S.H.C., S.L., H.C.J.), Seoul National University Bundang Hospital, Seongnam 463-707, Korea; and Department of Preventive Medicine (N.H.C.), Ajou University School of Medicine, Suwon 443-721, Korea
Context: The effect of weight gain on the development of type 2 diabetes after gestational diabetes mellitus (GDM) is not fully understood in Asian women who have a relatively low body mass index (BMI). Objective: We investigated the effect of postpartum longitudinal BMI change on the development of diabetes in Korean women with a history of GDM. Design and Setting: The study included a hospital-based, multicenter, prospective cohort with median follow-up of 4.0 years. Participants: A total of 418 women with previous GDM or gestational impaired glucose tolerance were recruited and underwent an oral glucose tolerance test at 6 weeks postpartum and annually thereafter. Main Outcome Measure: The risk of diabetes was analyzed according to the tertiles of BMI change. Changes in BMI were calculated between the initial postpartum visit and the last follow-up or at the onset of diabetes. Results: The BMI change in each tertile was ⫺1.8 ⫾ 1.1, ⫺0.2 ⫾ 0.3, and 1.6 ⫾ 1.2 kg/m2, respectively. We observed an increased risk of incident diabetes as the tertile of BMI change increased (8.6%, 12.6%, and 16.9%, P ⫽ .039). Postpartum BMI change was an independent predictor of diabetes in a multivariate Cox analysis (hazard ratio 1.27, 95% confidence interval 1.04 –1.56, P ⫽ .021), even after adjusting for BMI at the last follow-up. In the highest tertile group, there was a significant deterioration in cardiovascular risk factors including blood pressure, lipid profile, and insulin sensitivity. Conclusions: Postpartum increase in BMI is significantly associated with a risk of diabetes and deterioration of metabolic phenotypes in Korean GDM women. (J Clin Endocrinol Metab 100: 3548 –3555, 2015)
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in USA Copyright © 2015 by the Endocrine Society Received January 13, 2015. Accepted July 9, 2015. First Published Online July 14, 2015
3548
press.endocrine.org/journal/jcem
* J.H.M. and S.H.K. contributed equally to this work. Abbreviations: BMI, body mass index; BP, blood pressure; CI, confidence interval; DPP, Diabetes Prevention Program; GDM, gestational diabetes mellitus; GIGT, gestational impaired glucose tolerance; HDL, high-density lipoprotein; HOMA-B, homeostasis model assessment of -cell function; HOMA-IR, homeostasis model assessment of insulin resistance; HR, hazard ratio; IGR, impaired glucose regulation; LDL, low-density lipoprotein; OGTT, oral glucose tolerance test.
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
doi: 10.1210/JC.2015-1113
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
he worldwide increase in the prevalence of type 2 diabetes is paralleled by the epidemic of increasing obesity (1). There is a large body of evidence that weight change by itself, in addition to increased body mass index (BMI), has a significant impact on the risk of diabetes and cardiovascular risk factors. In the Nurses’ Health Study, it has been reported that gaining more than 5 kg of body weight increased the risk of developing diabetes by 1.9 times (2). In this regard, interventional trials were conducted to demonstrate the beneficial effect of weight loss on the development of diabetes. The Diabetes Prevention Program (DPP) was a multicenter randomized clinical trial, which aimed to demonstrate that weight loss through lifestyle intervention or metformin could prevent or delay the onset of type 2 diabetes in a high-risk population of impaired glucose tolerance. In the DPP, weight loss was associated with reduced risk of diabetes by 16% for every kilogram of weight loss (3). Women with a previous history of gestational diabetes mellitus (GDM) have an increased risk of developing type 2 diabetes (4, 5). The prevalence of postpartum type 2 diabetes after GDM ranges from 2.6% to greater than 70% (5), and it varies widely by ethnicity and diagnostic methods that are used (6, 7). Various risk factors for the progression to type 2 diabetes include age of the mother (8); prepregnancy (9, 10) and postpartum (11) BMI; a family history of diabetes (12); breast-feeding (13); insulin treatment during pregnancy (5, 14); and fasting glucose during pregnancy (15). Among these, BMI is a modifiable risk factor that could be improved by lifestyle intervention. Several studies have investigated the effect of longitudinal weight gain on the development of type 2 diabetes in women with a previous history of GDM (16 –18). Most of the studies were conducted in Europeans or Hispanics. Although the effect of postpartum weight gain on type 2 diabetes is assumed to be similar for different ethnicities, it has not been thoroughly investigated in a prospective manner in Asian women with a previous history of GDM. Asian women have lower BMI thresholds for increased risk of GDM than Caucasians (19), and the progression rate to postpartum diabetes in Asian immigrant women is reported to be higher compared with Europeans (20). In this multicenter prospective cohort study, we evaluated the impact of the change in BMI on the development of type 2 diabetes and related metabolic phenotypes in Korean women with a previous history of GDM.
T
Materials and Methods Study design We conducted a multicenter prospective cohort study among subjects with a first-time diagnosis of GDM and gestational im-
press.endocrine.org/journal/jcem
3549
paired glucose tolerance (GIGT). Subjects were recruited from August 1995 to May 1997 from four centers in Korea. The initial postpartum follow-up visit was performed at 6 weeks postpartum, and annual follow-up visits were made thereafter.
Subjects All pregnant women performed a screening for GDM with a 50-g oral glucose challenge test at 24 –28 weeks of gestation, and a positive screen was defined as a 1-hour glucose value of 130 mg/dL or greater. Women with a positive screen were given a 3-hour, 100-g oral glucose tolerance test (OGTT). The diagnoses of GDM and GIGT were made using the recommendation of the Third International Workshop-Conference on Gestational Diabetes Mellitus (21). Two or more of the following criteria for GDM and one for GIGT were required for the diagnosis: fasting plasma glucose of 105 mg/dL or greater, 1-hour glucose of 190 mg/dL or greater, 2-hour glucose of 165 mg/dL or greater, or 3-hour glucose of 145 mg/dL or greater. Women with GDM or GIGT who visited the initial postpartum evaluation were enrolled for this study (n ⫽ 1050). Subjects who had persistent diabetes at the initial visit (n ⫽ 113) were excluded. Among the remaining 937 eligible subjects, a total of 418 subjects completed follow-up examinations more than once and had data on BMI and OGTT to diagnose postpartum type 2 diabetes. Study participants (n ⫽ 418) were generally representing potentially eligible subjects (n ⫽ 937), and there was no significant difference in clinical characteristics except for the decreased systolic blood pressure (BP) in the study participants compared with those who did not make follow-up visits (Supplemental Table 1). All subjects participated voluntarily, and informed consent was obtained from each subject. This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (Institutional Review Board number B-1007– 105-007) and was conducted according to the Declaration of Helsinki (22).
Postpartum follow-up examinations Obstetric history including prepregnancy weight, weight at the diagnosis of GDM or GIGT, gestational weight gain, and pregnancy outcome was retrieved from medical records. Faceto-face interviews were conducted at the first postpartum examination using a standardized questionnaire that included past medical and reproductive history and family history of diabetes. For the postpartum follow-up assessment, a standard 75-g OGTT was performed at each follow-up visit. The diagnosis of diabetes was based on the criteria of the American Diabetes Association (23). Impaired glucose regulation (IGR) was defined as a combination of impaired fasting glucose and impaired glucose tolerance. A recent clinical history of diagnosis of diabetes or antidiabetic medication was reviewed at each visit. Physical activity data were collected during the postpartum follow-up visits and were classified as mild, moderate, and vigorous intensity (24). Anthropometric measures including body weight, height, BP, and body composition were measured at each visit. Body weight was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm. BP was measured three times on the left arm by research nurse with mercury sphygmomanometer (Baumanometer; Baum) to the nearest 2 mm Hg. Averaged values were used for analyses. Bioelectrical impedance analysis
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3550
Moon et al
Weight Gain and Future Diabetes in Women With GDM
(body composition analyzer; Giru Co) was performed to evaluate body composition.
Metabolic assessment Metabolic assessment including a 75-g OGTT and fasting serum lipid concentration (total cholesterol, high density lipoprotein [HDL] cholesterol, and triglycerides) was performed at each visit. The homeostasis model assessment of -cell function (HOMA-B) and insulin resistance (HOMA-IR) were calculated (25). Insulin sensitivity was assessed by the Matsuda index (26) as follows: 10 000/⻫([fasting glucose] ⫻ [fasting insulin] ⫻ [mean glucose] ⫻ [mean insulin]). The insulinogenic index (27) was used to estimate insulin secretion as follows: (insulin [30 min] ⫺ insulin [0 min])/(glucose [30 min] ⫺ glucose [0 min]). The disposition index (28) was used to evaluate the composite of insulin secretion considering the degree of insulin sensitivity, as follows: (Matsuda index) ⫻ (insulinogenic index). A 75-g OGTT was performed after a more than 8-hour overnight fast. Plasma glucose was enzymatically measured by an automated analyzer (YSI 2300-STAT; Yellow Springs Instrument Co) using the glucose oxidase method. Plasma insulin was measured by a RIA (Linco Research Inc). Total cholesterol, triglycerides, and HDL (Beckman analyzer; Beckman Instruments) were measured by an enzymatic assay. Low-density lipoprotein (LDL) was calculated using the Friedewald equation (29).
Statistical analysis Postpartum BMI change was defined as the difference between the initial postpartum follow-up and the last visit of examination or at the onset of diabetes. Subjects were divided into tertiles for analysis. Differences among groups were evaluated using an ANOVA for continuous variables and the 2 test and linear-by-linear association for categorical variables. We used the Cox proportional hazard model to assess the difference in incident type 2 diabetes by postpartum BMI change and to adjust for differences in follow-up duration. Subjects were censored at the onset of diabetes or at the last visit of examination (or when lost to follow-up). For the primary analysis, age and initial postpartum BMI were used as covariates in the Cox model. In the secondary analysis, variables that showed a significant association in the univariate analysis and age were used as covariates. Data are presented as the mean ⫾ SD for continuous variables in Gaussian distribution or as the median (interquartile range). Hazard ratios (HRs) are presented with their corresponding 95% confidence intervals (CIs) and P values. Differences were considered significant when P ⬍ .05. Statistical analyses were conducted using SPSS version 20.0 (IBM Co).
Results Prepregnancy BMI and development of diabetes A total of 418 subjects (283 of GDM and 135 of GIGT) were included in the analysis. Because prepregnancy BMI in Korean women with GDM was lower than that in European or Hispanic women with GDM, we first investigated the association between prepregnancy BMI and risk of diabetes. Subjects were divided into three groups by the tertiles of prepregnancy BMI (cutoff 20.6 and 23.2 kg/m2).
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
A total of 53 subjects (12.7%) were found to have type 2 diabetes on follow-up. Within each tertile 12 (8.8%), 15 (10.9%), and 26 (19.0%) subjects developed type 2 diabetes, and the increasing trend was statistically significant (P for trend ⫽ .012). Postpartum BMI change and development of diabetes Next, we investigated whether postpartum BMI change is associated with the development of type 2 diabetes. Subjects were followed up for a median of 4.0 (interquartile range 2.6 –5.0) years. The overall mean postpartum BMI change was ⫺0.1 ⫾ 1.7 kg/m2. The mean age of subjects at the last visit (or at the onset of diabetes) was 36.0 ⫾ 4.6 years. Subjects were categorized into tertiles of postpartum BMI change (cutoff ⫺0.7 and ⫹0.4 kg/m2). Subjects in the first tertile had a decreased BMI of mean ⫺1.8 kg/ m2, whereas subjects in the second and third tertiles had maintained or increased BMIs by a mean ⫺0.2 and ⫹1.6 kg/m2. The baseline characteristics of the subjects at the initial postpartum follow-up according to the tertile are described in Table 1. There were no significant differences among the three groups except for the increased initial postpartum BMI in the first tertile group. In each tertile of postpartum BMI change, the number of subjects who developed type 2 diabetes was 12 (8.6%), 18 (12.6%), and 23 (16.9%), respectively, and the increasing trend was statistically significant (P for trend ⫽ .039). The incidence of type 2 diabetes among the three groups was evaluated by the Cox proportional hazard model. Subjects in the third tertile showed an increased risk of progression to type 2 diabetes compared with the subjects in the first tertile (HR 2.23, 95% CI 1.10 – 4.53, P ⫽ .026) after adjusting for age and initial postpartum BMI (Figure 1). We additionally performed stratified analysis according to last follow-up BMI, considering its significant correlation with postpartum BMI change (r ⫽ 0.364, P ⬍ .001). In each tertile of the last follow-up BMI, there was an increasing trend of incident diabetes as postpartum BMI change increased (Supplemental Table 2 and Supplemental Figure 1). Postpartum BMI change as an independent risk factor for incident type 2 diabetes We further explored the association between various risk factors and incident type 2 diabetes using a multivariate Cox proportional hazard model. In the model, factors that showed a significant association with postpartum type 2 diabetes in the univariate Cox analysis and age were included (Table 2). Postpartum BMI change was an independent risk factor for type 2 diabetes (HR 1.27, 95% CI
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
Table 1.
press.endocrine.org/journal/jcem
3551
Baseline Clinical Characteristics and Biochemical Parameters at Initial Postpartum Follow-Up Postpartum BMI Change
Age, y BMI, kg/m2 Weight, kg Duration of follow-up, y GIGT, % Family history of diabetes, % Breast-feeding, % Postpartum exercisec Insulin treatment during pregnancy, % Systolic BP, mm Hg Fasting glucose, mg/dL Two-hour glucose, mg/dL Fasting insulin, mU/mL HOMA-IR HOMA-B Total cholesterol, mg/dL Triglycerides, mg/dL HDL, mg/dL LDL, mg/dL Body fat percentage by BIA, %
First Tertile (n ⴝ 139)
Second Tertile (n ⴝ 143)
Third Tertile (n ⴝ 136)
P Value
32.1 ⫾ 4.6 24.1 ⫾ 3.4 59.9 ⫾ 9.1 3.9 ⫾ 1.8 33.1 46.8 51.8 55.4 21.2 112.5 ⫾ 11.4 93.3 ⫾ 10.1 129.1 ⫾ 27.4 9.7 ⫾ 3.9 2.2 ⫾ 1.0 127.0 ⫾ 59.9 199.4 ⫾ 34.9 118.9 ⫾ 68.3 53.7 ⫾ 13.7 122.0 ⫾ 30.1 31.4 ⫾ 5.5
32.2 ⫾ 4.4 22.8 ⫾ 2.8 56.3 ⫾ 6.8 3.8 ⫾ 2.0 30.8 52.4 41.3 55.2 21.9 110.3 ⫾ 12.2 92.3 ⫾ 10.1 124.8 ⫾ 30.8 9.3 ⫾ 4.4 2.1 ⫾ 1.1 133.2 ⫾ 130.5 198.9 ⫾ 33.6 127.7 ⫾ 76.1 53.1 ⫾ 14.7 120.2 ⫾ 27.8 30.2 ⫾ 5.5
32.0 ⫾ 4.0 23.1 ⫾ 3.1 57.2 ⫾ 8.4 4.2 ⫾ 1.7 33.1 41.9 52.2 51.5 23.1 109.8 ⫾ 12.0 92.3 ⫾ 10.3 123.3 ⫾ 27.2 9.5 ⫾ 5.0 2.2 ⫾ 1.3 133.5 ⫾ 88.7 197.4 ⫾ 39.3 116.7 ⫾ 62.1 54.0 ⫾ 14.4 119.9 ⫾ 33.9 29.9 ⫾ 4.7
.959 .001a,b .001a,b .171 .982 .211 .957 .516 .706 .133 .632 .211 .693 .711 .821 .883 .373 .855 .836 .058
Abbreviation: BIA, bioelectrical impedance analysis. Subjects were divided into tertiles of postpartum BMI change. P values are for ANOVA (continuous) or linear-by-linear association (dichotomous). Post hoc analysis was performed for variables with P ⬍ .05. a
P ⬍ .05 for first vs second tertile.
b
P ⬍ .05 for first vs third tertile.
c
Exercise was defined as moderate- or vigorous-intensity physical activities.
1.04 –1.56, P ⫽ .021), even after adjusting for other risk factors including prepregnancy BMI and last follow-up BMI. The prepregnancy BMI and fasting glucose during pregnancy were also independent risk factors for the development of type 2 diabetes.
mental Table 4). Improvement of glucose tolerance status, ie, from IGR to normal glucose tolerance, was most frequent in the first tertile group, and deterioration was most frequent in the third tertile group, (P ⬍ .001 by 2 test).
Subgroup analysis and overall glucose tolerance As there might be a residual confounding effect of difference in follow-up duration among the three groups, we performed a sensitivity analysis by dividing our subjects according to follow-up duration of 2 years or more. We observed that incidence of diabetes increased as BMI change increase in the both subgroups (Supplemental Table 3). Among the 135 GIGT subjects, there were significantly fewer subjects who developed diabetes (n ⫽ 5, 3.7%) during the postpartum follow-up. After excluding GIGT subjects, the postpartum BMI change remained associated with the development of type 2 diabetes. Among 283 women with GDM, the number of subjects who progressed to type 2 diabetes in each tertile was 10 (10.8%), 18 (18.2%), and 20 (22.0%), respectively, and the increasing trend was significant (P for trend ⫽ .043). The change in overall glucose tolerance status during follow-up varied by postpartum BMI change (Supple-
Postpartum BMI change and related metabolic phenotypes We examined the change in metabolic phenotypes according to the tertiles of BMI change and observed that increased BMI was associated with deteriorating metabolic phenotypes (Table 3). Subjects in the third tertile had a significantly higher increase in 2-hour glucose and fasting insulin compared with those in the first tertile. BP and lipid profile deteriorated during the follow-up in the third tertile subjects compared with the first tertile subjects. Body fat percentage change also differed by postpartum BMI change, even in subjects of the first or second tertile of last follow-up BMI (Supplemental Table 5). We observed that insulin sensitivity, assessed by the Matsuda index, and composite insulin secretion, assessed by the disposition index, were significantly deteriorated in third tertile subjects who gained weight (Supplemental Figure 2). These differences in metabolic phenotypes by BMI change were maintained after adjusting for last follow-up BMI (data not shown).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3552
Moon et al
Weight Gain and Future Diabetes in Women With GDM
Figure 1. Cumulative incidence of type 2 diabetes mellitus by postpartum BMI change. Subjects were divided into tertiles of postpartum BMI change. A Cox proportional hazard model was used for the analysis (adjusted for age and BMI on initial follow-up). The incidence of type 2 diabetes was higher in the third tertile compared with the first tertile (HR 2.23, 95% CI 1.10 – 4.53, P ⫽ .026).
Discussion In this observational cohort study, postpartum BMI change during 4 years of follow-up in women with previous GDM was associated with a significant risk of type 2 diabetes. Postpartum BMI change was an independent risk factor for incident diabetes in Cox analysis, even after adjusting for multiple risk factors including prepregnancy BMI and last follow-up BMI. We demonstrated an association between postpartum BMI change and metabolic phenotypes including BP, insulin sensitivity, and lipid proTable 2.
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
files. To the best of our knowledge, this is one of the first prospective studies that comprehensively assessed the impact of postpartum BMI change on the development of type 2 diabetes and its related metabolic phenotypes in Asian women. Obesity is a well-known modifiable risk factor for type 2 diabetes. Several pivotal studies have reported the association between long-term weight change and the risk of diabetes in both general population and women with history of GDM. In the Nurses’ Health Study, which involved middle-aged women without a history of GDM, those who had weight gain of 5.0 –7.9 kg had an increased relative risk of developing type 2 diabetes by 1.9 (95% CI 1.5–2.3) compared with those who had less than 5 kg of weight gain during the 14 years of follow-up (2). Regarding women with a history of GDM, a prospective observational study by Peters et al (17) that included 666 Hispanics explicitly showed that weight gain increased the risk of type 2 diabetes with an HR of 1.54 (95% CI 1.33– 1.79) per 4.5 kg (10 lb) during 2.2 years of follow-up. In the subgroup analysis of the DPP trial, intensive lifestyle modification or metformin treatment led to postpartum weight loss and 50% reduction in type 2 diabetes risk in women with a history of GDM (18). To have a better comparison with the study results by Peters et al, we reanalyzed our data and observed that the HR for diabetes was 2.00 (95% CI 1.05–3.81) per 4.5 kg (10 lb) weight gain. Although the follow-up duration was longer in our study, it is noteworthy that the risk of diabetes was even higher in Asian women compared with Hispanics for the same degree of weight gain. Whereas the magnitude of the effect of increased weight gain is different according to study design, ethnicity, demographic characteristics, and follow-up duration, it is consistent that weight change is significantly associated with a risk of type 2 diabetes. Our
HRs of Postpartum Type 2 Diabetes by Postpartum BMI Change and Associated Variables by Cox Analysis
Postpartum BMI change Last follow-up BMI Prepregnancy BMI Age Breast-feeding Postpartum exercised Family history of diabetes Parity Fasting glucose during pregnancy
Univariatea
P Value
Multivariateb
P Value
Multivariatec
P Value
1.17 (1.01–1.34) 1.12 (1.04 –1.20) 1.14 (1.05–1.23) 0.95 (0.89 –1.02) 0.50 (0.28 – 0.89) 0.59 (0.35–1.00) 1.28 (0.74 –2.19) 1.00 (0.64 –1.57) 1.03 (1.02–1.04)
.033 .003 .001 .129 .020 .050 .377 .988 ⬍.001
1.27 (1.04 –1.56) 0.90 (0.75–1.07) 1.25 (1.06 –1.47) 0.94 (0.87–1.00) 0.61 (0.33–1.13) 0.49 (0.20 –1.23)
.021 .228 .007 .056 .114 .130
1.17 (1.02 ⫺ 1.35)
.025
1.13 (1.04 ⫺ 1.22)
.002
1.04 (1.02–1.05)
⬍.001
1.32 (1.02–1.04)
⬍.001
Values presented are HR (95% CI). a
HR of postpartum type 2 diabetes for each variable was evaluated with univariate Cox analysis.
b
Factors associated with postpartum type 2 diabetes in univariate Cox analysis and age were included in the multivariate Cox model.
c
Multivariate Cox model only including the significant variables.
d
Exercise was defined as moderate- or vigorous-intensity physical activities.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
Table 3. Change
press.endocrine.org/journal/jcem
3553
Cumulative Incidence of Type 2 Diabetes and Changes of Metabolic Phenotypes by Postpartum BMI Postpartum BMI Change
Type 2 diabetes BMI, kg/m2 Weight, kg Systolic BP, mm Hg Fasting glucose, mg/dL Two-hour glucose, mg/dL Fasting insulin, U/mL HOMA-IR HOMA-B Matsuda index Insulinogenic index Disposition index Total cholesterol, mg/dL Triglycerides, mg/dL HDL, mg/dL LDL, mg/dL Body fat percentage by BIA, %
First Tertile (n ⴝ 139)
Second Tertile (n ⴝ 143)
Third Tertile (n ⴝ 136)
P Value
12 (8.6%) ⫺1.8 ⫾ 1.1 ⫺4.6 ⫾ 5.6 ⫺2.1 ⫾ 10.0 5.4 ⫾ 24.5 5.1 ⫾ 59.8 ⫺1.2 ⫾ 4.8 0.3 ⫾ 1.3 ⫺22.8 ⫾ 66.7 0.5 ⫾ 2.9 0.10 ⫾ 0.52 1.0 ⫾ 3.7 ⫺25.8 ⫾ 36.0 ⫺24.9 ⫾ 55.6 2.2 ⫾ 12.7 ⫺23.0 ⫾ 32.6 ⫺2.7 ⫾ 5.3
18 (12.6%) ⫺0.2 ⫾ 0.3 ⫺0.9 ⫾ 6.0 ⫺0.1 ⫾ 10.0 5.4 ⫾ 15.3 9.0 ⫾ 46.9 0.0 ⫾ 5.1 0.6 ⫾ 2.3 ⫺20.7 ⫾ 134.5 0.0 ⫾ 3.4 0.01 ⫾ 0.38 0.0 ⫾ 2.6 ⫺18.4 ⫾ 34.1 ⫺22.8 ⫾ 84.9 1.0 ⫾ 12.6 ⫺14.7 ⫾ 31.8 ⫺1.0 ⫾ 6.0
23 (16.9%) 1.6 ⫾ 1.1 3.4 ⫾ 5.7 2.0 ⫾ 9.3 10.6 ⫾ 23.9 26.2 ⫾ 59.6 1.0 ⫾ 5.4 0.4 ⫾ 2.2 ⫺22.6 ⫾ 91.0 ⫺1.4 ⫾ 3.2 0.04 ⫾ 0.50 ⫺0.6 ⫾ 6.1 ⫺5.5 ⫾ 34.0 1.4 ⫾ 71.7 ⫺0.8 ⫾ 11.3 ⫺5.0 ⫾ 31.9 1.4 ⫾ 5.0
.039 ⬍.001a,b,c ⬍.001a,b,c .003b .073 .004b,c .002b .531 .983 ⬍.001b,c .441 .013b ⬍.001b,c .004b,c .131 ⬍.001a,b,c ⬍.001a,b,c
Abbreviation: BIA, bioelectrical impedance analysis. Changes of each variable between initial and last postpartum follow-up are presented. P values are for ANOVA (continuous) or linear-by-linear association (dichotomous). Post hoc analysis was performed for variables with P ⬍ .05. a
P ⬍ .05 for first vs second tertile.
b
P ⬍ .05 for first vs third tertile.
c
P ⬍ .05 second vs third tertile.
study confirms that this is also true for Asian women with GDM. The initial postpartum mean BMI was in the range of 22.8 –24.1 kg/m2 for the three groups, and this indicates that our study participants were slightly overweight and mostly not obese according to the widely accepted Asian BMI criteria (30). The weight change during follow-up was also modest: 7.7% decrease (⫺4.6 kg) in the first tertile group and a 5.9% increase (3.4 kg) in the third tertile group. Our finding shows that a modest weight change has a significant impact on the development of type 2 diabetes in Korean women with GDM. We observed a significant negative correlation between initial postpartum BMI and postpartum BMI change (r ⫽ ⫺0.145, P ⫽ .003). It is possible that women with a higher initial postpartum BMI are more likely to lose weight during follow-up. Regarding the last follow-up BMI, it was also associated with a risk of diabetes, and it had a significant positive correlation with postpartum BMI change (r ⫽ 0.364, P ⬍ .001). It should be noted that postpartum BMI change was an independent predictor of incident diabetes, even after adjusting for multiple risk factors including the last follow-up BMI. However, it is possible that other risk factors such as follow-up duration, breastfeeding, and physical activity might have confounded our results. It would be of interest to further investigate the
relative contribution of BMI change and last follow-up BMI on incident diabetes. According to the Korean National Health and Nutrition Examination Survey (31), BMI in women aged 20 –39 years remained stable over time since the 1990s (22.1 kg/m2 in 1998 and 21.7 kg/m2 in 2007, P for trend ⫽ .103). The frequency of obesity (defined as a BMI higher than 25 kg/m2) was also similar (16.5% in 1998 and 13.0% in 2007, P for trend ⫽ .212). It is notable that BMI in Korean young women was stable over time, in contrast to the global epidemic increase in obesity (32). This is expected to be the consequence of weight consciousness among young women in Korea (33). Our study is applicable to Korean women who are mostly not obese currently and can also be extended to Asian women with a relatively low BMI. In the study, the third tertile group of postpartum BMI change had a significantly increased risk of diabetes compared with the first tertile group in Cox analysis. Regarding the second tertile group, all the participants had a stable weight with a change between ⫺0.7 and 0.4 kg. However, the risk for diabetes was also increased by an HR of 2.01, although statistical significance was not met (95% CI 0.95– 4.24, P ⫽ .067). It should also be noted that body fat percentage decreased in subjects who reduced BMI postpartum, even in subjects with low to middle BMI
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
3554
Moon et al
Weight Gain and Future Diabetes in Women With GDM
on last follow-up. Taken together, our study implies that women with GDM should try at least not to gain weight to prevent the development of diabetes. Our study included participants who had GIGT but who did not fulfill the criteria of GDM. Because the diagnosis of GIGT was based on one abnormal value of a 100-g OGTT, these women might reflect, in part, those who would be newly diagnosed by the International Association of the Diabetes and Pregnancy Study Groups criteria (34). Among the 135 GIGT women, only five (3.7%) of them progressed to type 2 diabetes, but 51 (37.8%) progressed to IGR. It is not known whether these women with IGR would eventually progress to type 2 diabetes. Furthermore, long-term follow-up studies are required to determine the clinical significance of GIGT in terms of risk of type 2 diabetes. Our study has some limitations. First, the follow-up duration and interval were not uniform in all the study subjects. Nevertheless, the mean duration of follow-up in each BMI change tertile was not significantly different, and we applied the Cox proportional hazard model to overcome this issue. Because we analyzed data from a hospital-based prospective cohort, it is more likely to reflect settings of real clinical practice. Second, due to the observational nature of this study, reverse causation and various confounding factors might have contributed to the results. In our study, changes in the insulin level during postpartum follow-up were higher as the BMI change tertile increased. As suggested by Corkey (35), there is a potential that hyperinsulinemia itself might have resulted in weight gain and aggravation of other metabolic phenotypes, rather than changes in BMI causing detrimental metabolic effects including hypersecretion of insulin. In addition, although we had data on physical activity, we did not have dietary data, which were a major part of lifestyle intervention. In conclusion, this study demonstrated that postpartum BMI change was significantly associated with a risk of type 2 diabetes and related metabolic phenotypes in Korean women with a history of GDM. We found that BMI change during follow-up was an independent risk factor for postpartum diabetes. Our study indicates that women with a history of GDM should try not to gain weight and closely monitor their weight to maintain proper body weight to prevent the development of type 2 diabetes.
Acknowledgments Address all correspondence and requests for reprints to: Hak C. Jang, PhD, Department of Internal Medicine, Seoul National University Bundang Hospital, 82, Gumi-ro, 173 beon-gil, Bundang-gu, Seongnam, Gyeonggi-do, 463-707, Korea. E-mail:
J Clin Endocrinol Metab, September 2015, 100(9):3548 –3555
[email protected]; or Nam H. Cho, PhD, Department of Preventive Medicine, Ajou University School of Medicine, 164, World Cup-ro, Yeongtong-gu, Suwon, Gyeonggi-do, 443-721, Korea. E-mail: [email protected]. J.H.M. and S.H.K. contributed to researching the data and writing the manuscript. H.S.J., S.H.C, S.L., Y.M.C., K.S.P., H.C.J., and N.H.C. contributed to the discussion and reviewed and edited the manuscript. H.C.J. and N.H.C. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. This work was supported by the Korea Healthcare Technology Research and Development Project, Ministry of Health and Welfare Grant A111362 and the Seoul National University Bundang Hospital Research Fund Grant 02-2010-013. Disclosure Summary: The authors have nothing to disclose.
References 1. Geiss LS, Wang J, Cheng YJ, et al. Prevalence and incidence trends for diagnosed diabetes among adults aged 20 to 79 years, United States, 1980 –2012. JAMA. 2014;312:1218 –1226. 2. Colditz GA, Willett WC, Rotnitzky A, Manson JE. Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med. 1995;122:481– 486. 3. Hamman RF, Wing RR, Edelstein SL, et al. Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes Care. 2006;29: 2102–2107. 4. Jang HC. Gestational diabetes in Korea: incidence and risk factors of diabetes in women with previous gestational diabetes. Diabetes Metab J. 2011;35:1–7. 5. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care. 2002; 25:1862–1868. 6. Ferrara A. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care. 2007;30(suppl 2):S141– S146. 7. Fujimoto W, Samoa R, Wotring A. Gestational diabetes in high-risk populations. Clin Diabetes. 2013;31:90 –94. 8. Grant PT, Oats JN, Beischer NA. The long-term follow-up of women with gestational diabetes. Aust N Z J Obstet Gynaecol. 1986;26:17–22. 9. Catalano PM, Vargo KM, Bernstein IM, Amini SB. Incidence and risk factors associated with abnormal postpartum glucose tolerance in women with gestational diabetes. Am J Obstet Gynecol. 1991; 165:914 –919. 10. Kwak SH, Choi SH, Jung HS, et al. Clinical and genetic risk factors for type 2 diabetes at early or late post partum after gestational diabetes mellitus. J Clin Endocrinol Metab. 2013;98:E744 –E752. 11. Steinhart JR, Sugarman JR, Connell FA. Gestational diabetes is a herald of NIDDM in Navajo women. High rate of abnormal glucose tolerance after GDM. Diabetes Care. 1997;20:943–947. 12. Henry OA, Beischer NA. Long-term implications of gestational diabetes for the mother. Baillieres Clin Obstet Gynaecol. 1991;5:461– 483. 13. Taylor JS, Kacmar JE, Nothnagle M, Lawrence RA. A systematic review of the literature associating breastfeeding with type 2 diabetes and gestational diabetes. J Am Coll Nutr. 2005;24:320 –326. 14. Lobner K, Knopff A, Baumgarten A, et al. Predictors of postpartum diabetes in women with gestational diabetes mellitus. Diabetes. 2006;55:792–797. 15. Coustan DR, Carpenter MW, O’Sullivan PS, Carr SR. Gestational diabetes: predictors of subsequent disordered glucose metabolism. Am J Obstet Gynecol. 1993;168:1139 –1145.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/JC.2015-1113
16. Ehrlich SF, Hedderson MM, Quesenberry CP Jr, et al. Post-partum weight loss and glucose metabolism in women with gestational diabetes: the DEBI Study. Diabet Med. 2014;31:862– 867. 17. Peters RK, Kjos SL, Xiang A, Buchanan TA. Long-term diabetogenic effect of single pregnancy in women with previous gestational diabetes mellitus. Lancet. 1996;347:227–230. 18. Ratner RE, Christophi CA, Metzger BE, et al. Prevention of diabetes in women with a history of gestational diabetes: effects of metformin and lifestyle interventions. J Clin Endocrinol Metab. 2008;93: 4774 – 4779. 19. Hedderson M, Ehrlich S, Sridhar S, Darbinian J, Moore S, Ferrara A. Racial/ethnic disparities in the prevalence of gestational diabetes mellitus by BMI. Diabetes Care. 2012;35:1492–1498. 20. Ignell C, Shaat N, Ekelund M, Berntorp K. The impact of ethnicity on glucose homeostasis after gestational diabetes mellitus. Acta Diabetol. 2013;50:927–934. 21. Metzger BE. Summary and recommendations of the Third International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes. 1991;40:197–201. 22. World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Bull World Health Organ. 2001;79(4):373–374. 23. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(suppl 1):S62–S69. 24. Sallis JF, Haskell WL, Wood PD, et al. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol. 1985;121: 91–106. 25. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and -cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412– 419. 26. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from
press.endocrine.org/journal/jcem
27.
28.
29.
30.
31.
32.
33.
34.
35.
3555
oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care. 1999;22:1462–1470. Tura A, Kautzky-Willer A, Pacini G. Insulinogenic indices from insulin and C-peptide: comparison of -cell function from OGTT and IVGTT. Diabetes Res Clin Pract. 2006;72:298 –301. Retnakaran R, Qi Y, Goran MI, Hamilton JK. Evaluation of proposed oral disposition index measures in relation to the actual disposition index. Diabet Med. 2009;26:1198 –1203. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499 – 502. World Health Organization Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157–163. Khang YH, Yun SC. Trends in general and abdominal obesity among Korean adults: findings from 1998, 2001, 2005, and 2007 Korea National Health and Nutrition Examination Surveys. J Kor Med Sci. 2010;25:1582–1588. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980 –2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384:766 –781. Wardle J, Haase AM, Steptoe A. Body image and weight control in young adults: international comparisons in university students from 22 countries. Int J Obes (Lond). 2006;30:644 – 651. International Association of Diabetes Pregnancy Study Groups Consensus Panel, Metzger BE, Gabbe SG, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33:676 – 682. Corkey BE. Banting Lecture 2011: hyperinsulinemia: cause or consequence? Diabetes. 2012;61:4 –13.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 September 2015. at 06:38 For personal use only. No other uses without permission. . All rights reserved.
