Introduction
Gestational diabetes mellitus (GDM) is one of the most common metabolic complications of pregnancy and has a high prevalence globally,1,2 3,4 5–7 8,9 10,11 subsequent pregnancy is uncertain.12 subsequent pregnancy . Because few studies have addressed these knowledge gaps to date, we performed a multicenter retrospective cohort study to investigate the incidence of GDM among women without a GDM history and to identify the potential risk factors of GDM in subsequent pregnancy , with the goal of providing some references for the prediction and prevention of GDM.
Methods
Study design and subjects
This retrospective cohort study was performed at 18 research centers selected from the Chinese GDM Collaboration Group (Peking University First Hospital, Tianjin Central Hospital of Obstetrics and Gynecology, Hubei Maternal and Child Health Hospital, Fujian Maternal and Child Health Hospital, Jinan Maternal and Child Health Hospital, First affiliated Hospital of Zhengzhou University, Dalian Maternity Hospital, Taiyuan Maternal and Child Health Hospital Shanxi Province, Affiliated Hospital of Inner Mongolia Medical University, Peking University Shenzhen Hospital, Suzhou Kowloon Hospital, Shanghai Jiao Tong University School of Medicine, Huazhong University of Science and Technology Tongji Hospital, Qilu Hospital of Shandong University, Third affiliated Hospital of Zhengzhou University, Beijing Shunyi District Hospital, Hainan General Hospital, Harbin Medical University Fourth Hospital, Harbin Medical University First Hospital). These hospitals came from 14 provinces, 15 cities, three specialist hospitals, and 15 general hospitals, all of which were tertiary hospitals. The study was reviewed and approved by the Institutional Review Board of the Peking University First Hospital in 2013 (Reference number: 2013(578)). Because this is a retrospective study, the informed consent form is exempted from signing.
The inclusion criteria were all participants who had received prenatal care at the 18 hospitals and had two deliveries in same hospital, who delivered first time after 2011, and the second delivery was from January 1, 2018 to December 31, 2018. All participants have done oral glucose tolerance test (OGTT) in subsequent pregnancy . The exclusion criteria were as follows: delivery before 28 weeks of gestation, known diabetes mellitus (DM) before index pregnancy, multiple pregnancy, and missing data on the major items. We enrolled 6204 women in the initial sample, 5180 women without GDM in the index pregnancy remained in the final analysis. Detailed information on the study population is shown in Figure 1 .
Figure 1: Flow diagram of participant selection. DM: Diabetes mellitus; GDM: Gestational diabetes mellitus.
Quantitative variables for each pregnancy included age, height, prepregnancy body weight, weight before birth, 75-g OGTT results, fetal birth weight, and gestational age at delivery. We used 1 or 2 as the suffix of each variable: 1 for the index pregnancy and 2 for the subsequent pregnancy . Through calculations, we obtained data on the prepregnancy body mass index (BMI), gestational weight gain (GWG), and weight retention during the interval between the two pregnancies (prepregnancy body weight 2 minus prepregnancy bodyweight 1) and the length of the interdelivery interval (date of birth 2 minus date of birth 1).
Definitions
The GDM diagnostic criteria followed the new guidelines established from 2011 in China,13
According to the standard of the Working Group on Obesity in China,14 2 ; normal weight, 18.5 ≤ BMI < 24.0 kg/m2 ; overweight, 24.0 ≤ BMI < 28.0 kg/m2 ; and obese, BMI ≥28.0 kg/m2 .
According to the Institute of Medicine,15 2 ) should gain 12.5 to 18.0 kg; normal-weight women (18.5 ≤ BMI < 25.0 kg/m2 ) should gain 11.5 to 16.0 kg; overweight women (25.0 ≤ BMI < 30.0 kg/m2 ) should gain 7.0 to 11.5 kg; and obese women (BMI ≥30.0 kg/m2 ) should gain 5.0 to 9.0 kg. According to each BMI category, GWG below the lower limit was defined as inadequate weight gain, GWG above the upper limit was defined as excessive weight gain, and GWG within the limits was defined as adequate weight gain.
The area under the curve (AUC) of the time-blood glucose curve of the OGTT can be used as a measurement method of the severity of maternal hyperglycemia16
Statistical analysis
SPSS 16.0 (SPSS, Inc., Chicago, IL, USA) was used for the statistical analysis. Continuous variables were presented as mean ± standard deviation (SD ), and categorical variables were presented as number and percentage. Quantitative data conform to normal distribution. A univariate analysis of potential risk factors was performed using the Chi-squared test and/or t -test for qualitative or quantitative variables, respectively. Of these variables, associations with P values <0.1 were chosen to be included in the multivariate binary logistic regression model. The cut-off value of the AUC of the OGTT and predictive ability was studied by means of a receiver operating characteristic (ROC) curve. P values <0.05 were considered statistically significant.
Results
Primary analysis
In this study, 6204 women without DM before the index pregnancy enrolled in 18 research centers were included in the primary analysis. there were 1002 (1002/6204, 16.2%) women diagnosed with GDM and 5202 (5202/6204, 83.8%) women without GDM in index pregnancy. In subsequent pregnancy , there were 490 women diagnosed with GDM in women with GDM history, the incidence of GDM was 48.9% (490/1002), 835 women diagnosed with GDM in women without GDM history, the incidence of GDM was 16.1% (835/5202). The incidence of GDM (48.9% vs . 16.1%, odds ratio (OR )=5.01, 95% confidence interval (CI )=4.33–5.78, P <0.001) is significantly higher in women with GDM history than that in women without GDM history.
Demographic factors related to GDM in subsequent pregnancy
We focused on the 5180 women who had no GDM history in index pregnancy and no DM in subsequent pregnancy . Table 1 shows the participant characteristics assessed during the two pregnancies classified by non-GDM or GDM, based on the blood glucose values of the subsequent pregnancy . We found that the GDM group was older than the non-GDM group at both the index pregnancy (28.9±3.2 years vs . 28.1±3.2 years, P <0.001) and subsequent pregnancy (32.6±3.3 years vs . 31.6±3.4 years, P <0.001) and that the percentage of women over 35 years old was higher in the GDM group. There were more overweight and obese women in the GDM group than the non-GDM group at subsequent pregnancy (33.3% vs . 20.8%, P <0.001), and this difference was also evident and significant at the index pregnancy (23.0% vs . 16.4%, P <0.001). The GWG at index pregnancy was not different between the GDM and non-GDM groups, but the mean GWG was lower in the GDM group than in the non-GDM group at subsequent pregnancy (12.6±4.8 kg vs . 13.9±4.9 kg, P <0.001).
Table 1 -
Characteristics of the women with and without GDM at
subsequent pregnancy .
Variables
n
Non-GDM
n
GDM
t /χ
2
P
Maternal age 1 (years)
4344
28.1±3.2
835
28.9±3.2
–6.374*
<0.001
≥35
109 (2.5)
36 (4.3)
8.368
0.004
Maternal age 2 (years)
4344
31.6±3.4
835
32.6±3.3
–7.731*
<0.001
≥35
757 (17.4)
225 (26.9)
40.691
<0.001
Prepregnancy BMI 1 (kg/m2 )
3582
21.3±3.0
718
21.9±3.1
–4.912*
<0.001
≥24
586 (16.4)
165 (23.0)
18.043
<0.001
Prepregnancy BMI 2 (kg/m2 )
3538
21.8±3.1
715
23.1±3.4
–9.245*
<0.001
≥24
736 (20.8)
238 (33.3)
52.016
<0.001
GWG 1 (kg)
3516
15.3±5.5
710
15.3±5.0
0.003*
0.998
Inadequate
631 (17.9)
115 (16.2)
Adequate
1294 (36.8)
243 (34.2)
4.509
0.105
Excessive
1591 (45.3)
352 (49.6)
GWG 2 (kg)
3196
13.9±4.9
584
12.6±4.8
6.098*
<0.001
Inadequate
732 (22.9)
172 (29.5)
Adequate
1293 (40.5)
205 (35.1)
12.619
0.002
Excessive
1171 (36.6)
207 (35.4)
Gestational age at delivery 1 (weeks)
4261
39.4±1.5
824
39.3±1.6
2.094*
0.036
Gestational age at delivery 2 (weeks)
4335
39.0±1.4
832
38.9±1.2
1.464*
0.143
Infant birth weight 1 (g)
4261
3324.0±448.0
824
3392.8±508.0
–3.626*
<0.001
Macrosomia 1
265 (6.2)
88 (10.7)
21.265
<0.001
Infant birth weight 2 (g)
4342
3359.2±444.3
835
3389.0±449.6
–1.772*
0.076
Macrosomia 2
304 (7.0)
69 (8.3)
1.668
0.196
Weight retention (kg)
3410
1.5±5.9
690
3.1±5.9
–6.588*
<0.001
IDI (year)
4345
3.5±1.4
835
3.7±1.3
–4.148*
<0.001
Data are presented as the mean ± SD for parametric variables and as n (%) for categorical variables.
We use 1 or 2 as the suffix of each variable; 1 represents the index pregnancy, and 2 represents the subsequent pregnancy .
*t value.
BMI: Body mass index; GDM: Gestational diabetes mellitus; GWG: Gestational weight gain; IDI: Interdelivery interval; SD : Standard deviation.
There was little difference in the gestational age at delivery between the groups in the index pregnancy (39.4±1.5 weeks vs . 39.3±1.6 weeks, P =0.036). The neonate birth weight was higher in the GDM group than in the non-GDM group at index pregnancy (3392.8±508.0 g vs . 3324.0±448.0 g, P <0.001), and a difference in the rate of macrosomia was evident only in the index pregnancy (10.7% vs . 6.2%, P <0.001) but not in the subsequent pregnancy (8.3% vs . 7.0%, P =0.196). The GDM group had higher weight retention than the non-GDM group (3.1±5.9 kg vs . 1.5±5.9 kg, P <0.001). The interdelivery interval was slightly longer in the GDM group than that in the non-GDM group.
OGTT and GDM in subsequent pregnancy
The results of our comparisons of the blood glucose values at each time point of the OGTT showed that the GDM group had higher values at each point than the non-GDM group in both index pregnancy (fasting 4.53±0.35 mmol/L vs . 4.39±0.36 mmol/L, 1 h 7.82±1.25 mmol/L vs . 7.28±1.32 mmol/L, 2 h 6.65±1.01 mmol/L vs . 6.22±1.01 mmol/L, P <0.001) and subsequent pregnancy (fasting 5.03±0.50mmol/L vs . 4.40±0.37 mmol/L, 1 h 9.16±1.67 mmol/L vs . 7.29±1.33 mmol/L, 2 h 7.78±1.42 mmol/L vs . 6.29±1.00mmol/L, P <0.001). The AUC of the OGTT showed similar trends (13.41±1.58 mmol·h/L vs . 12.58±1.68 mmol·h/L, P <0.001 in index pregnancy, 15.56±2.02 mmol·h/L vs . 12.64±1.71 mmol·h/L, P <0.001 in subsequent pregnancy ) (Table 2 ). The ROC curve was drawn between the AUC of the OGTT in the index pregnancy and GDM in the subsequent pregnancy , and the cut-off value of 12.6 mmol·h/L ensured the best sensitivity and specificity of 70.1% and 50.6%, respectively (area under the ROC curve 0.638, 95% CI : 0.616–0.660, P <0.001).
Table 2 -
Comparison of metabolic variables between the non-GDM and GDM groups.
Variables
n
Non-GDM
n
GDM
t
P
OGTT 1 (mmol/L)
Fasting
3783
4.39±0.36
718
4.53±0.35
–10.093
<0.001
1 h
3770
7.28±1.32
715
7.82±1.25
–10.415
<0.001
2 h
3765
6.22±1.01
715
6.65±1.01
–10.497
<0.001
OGTT 2 (mmol/L)
Fasting
4345
4.40±0.37
835
5.03±0.50
–34.502
<0.001
1 h
4345
7.29±1.33
835
9.16±1.67
–30.596
<0.001
2 h
4345
6.29±1.00
835
7.78±1.42
–28.833
<0.001
AUC of OGTT 1 (mmol·h/L)
3759
12.58±1.68
713
13.41±1.58
–12.671
<0.001
AUC of OGTT 2 (mmol·h/L)
4345
12.64±1.71
835
15.56±2.02
–39.166
<0.001
Data are presented as the mean±SD for parametric variables.
We use 1 or 2 as the suffix of each variable; 1 represents the index pregnancy, and 2 represents the subsequent pregnancy .
AUC: Area under the curve; GDM: Gestational diabetes mellitus; OGTT: Oral glucose tolerance test; SD : Standard deviation.
Independent risk factors of GDM in subsequent pregnancy
As shown in Table 3 , four variables became independent risk factors for GDM in the multivariate analysis: age at subsequent pregnancy , blood glucose value during index pregnancy, macrosomia history, and weight retention. We used two models to describe the blood glucose value during index pregnancy, as the blood glucose value in OGTT in model 1, and the AUC of OGTT 1 ≥12.6mmol·h/L in model 2. In model 1, women who were more than 35 years old had a 1.5-fold increased risk of GDM compared with women <35 years old in subsequent pregnancy (OR =1.540, 95% CI =1.257–1.886, P <0.001). Although the blood glucose level during the index pregnancy did not meet the diagnostic criteria for GDM, each value for OGTT 1 had an independent and significant effect on the GDM diagnosis at subsequent pregnancy ; among them, fasting blood glucose had the most significant effect (OR =2.487, 95% CI =1.883–3.285, P <0.001). Women who gave birth to infants with macrosomia in the index pregnancy had a 1.7-fold increased risk of GDM in the subsequent pregnancy compared with women who gave birth to infants without macrosomia in the index pregnancy (OR =1.749, 95% CI =1.277–2.395, P =0.001). As weight retention increased by 1 kg, the incidence of GDM in the subsequent pregnancy increased by 5% (OR =1.052, 95% CI =1.035–1.068, P <0.001). In model 2, when we used the AUC of the OGTT to represent the blood glucose level of each point, women whose AUCs of OGTT 1 were more than 12.6 mmol·h/L had a 2.0-fold increased risk of GDM compared with women whose AUCs of OGTT 1 were less than 12.6 mmol·h/L (OR =2.040, 95% CI =1.687–2.467, P <0.001).
Table 3 -
Multivariate analysis of the risk factors for GDM.
Variables
β
P
OR
95% CI
Model 1
Maternal age 2 ≥35 years
0.432
<0.001
1.540
1.257–1.886
OGTT 1 (mmol/L)
Fasting
0.911
<0.001
2.487
1.883–3.285
1 h
0.133
0.002
1.142
1.051–1.241
2 h
0.255
<0.001
1.290
1.162–1.432
Macrosomia 1
0.559
0.001
1.749
1.277–2.395
Weight retention (kg)
0.050
<0.001
1.052
1.035–1.068
Constant
–8.580
<0.001
0.000
Model 2
Maternal age 2 ≥35 years
0.500
<0.001
1.649
1.350–2.014
AUC of OGTT 1 ≥12.6 mmol×h/L
0.713
<0.001
2.040
1.687–2.467
Macrosomia 1
0.590
<0.001
1.805
1.321–2.464
Weight retention (kg)
0.051
<0.001
1.052
1.036–1.068
Constant
–2.310
<0.001
0.099
AUC: Area under the curve; CI : Confidence interval; GDM: Gestational diabetes mellitus; OGTT: Oral glucose tolerance test; OR: Odds ratio.
Discussion
This hospital-based retrospective study demonstrated that the frequency of GDM in subsequent pregnancy was 16.1% among women without GDM history. Independent risk factors for GDM were older age at subsequent pregnancy , macrosomia history, higher OGTT value during index pregnancy, and higher weight retention.
One of the key features of this study was that, to our knowledge, this is the first to report the frequency and risk factors of GDM in subsequent pregnancy among women without a GDM history according to the IADPSG 2011 GDM diagnostic criteria. We chose 2011 as starting point for our study because the National Health and Family Planning Commission of China adopted testing and diagnostic criteria based on the IADPSG guidelines in 2011.13 et al .12
Another advantage of our study is that the large sample size of subjects (5180 women) comprised patients from 18 hospitals in 14 provinces and 15 cities in China, allowing a representative group of multiparous women in China. After the implementation of the two-child policy in October 2015 and three-child policy in 2021, an increasing number of women had a second or third child. Therefore, researchers have begun to pay more attention to multiparous women since 2015 because this group is older, has a higher BMI, and has more pregnancy complications.
Macrosomia is considered a high-risk factor for GDM. Lee et al .17 et al .18 16 subsequent pregnancy , and it would be acceptable to state that the blood glucose level during the index pregnancy is related to GDM in subsequent pregnancy . We found that to predict the risk of GDM in subsequent pregnancy , the cut-off point of AUC needed to be 12.6mmol·h/L in the index pregnancy; thus, this method would a useful tool for predicting GDM risks in clinical work.
Two results in this study seemed to be inconsistent with our expectations. The first was that in subsequent pregnancy , GWG in the GDM group was lower than that in the non-GDM group (12.6±4.8 kg vs. 13.9±4.9 kg). The second was that the rates of macrosomia in the index pregnancy were significantly different between the GDM group and the non-GDM group (10.7% vs. 6.2%), but the difference became nonsignificant in the subsequent pregnancy (8.3% vs. 7.0%). These findings may be related to the systematic management of women with GDM in these hospitals. The IPD meta-analysis19 20
In the future clinical work, we propose that a multipara without GDM history should be screened for high risk factors of GDM in early pregnancy. Women who are more than 35 years old, with macrosomia history, AUC of OGTT ≥12.6 mmol·h/L in index pregnancy, weight retention ≥1kg, should be provided early diet and exercise guidance to reduce the risk of GDM. Some hospitals have antenatal clinics, which conduct a comprehensive assessment for women of childbearing age. At this time, guidance can also be given to multiparas with high-risk factors of GDM.
Some potential limitations are worth mentioning. This was a retrospective study, and some of the information was not available in the medical records; thus, there may be an inevitable risk of selection bias. Data on other risk factors, such as family history of diabetes mellitus, were not available, and these factors were not evaluated in our study.
Conclusion
In summary, women without a GDM history are still at risk of GDM in subsequent pregnancy . Risk factors for GDM, including age, macrosomia history, OGTT value, AUC of OGTT in index pregnancy, and weight retention, can be evaluated before or at the early stage of a subsequent pregnancy . Early warning and intervention are needed for high-risk women.
Acknowledgements
The authors would like to thank all the research staff of the 18 hospitals for their excellent contribution to the completion of the study protocol.
Funding
The study was supported by grants from the National Program on Basic Research Project of China (2019FY101005 to Geng Song), the World Diabetes Foundation (No. WDF 10-517 and No. WDF 14-908 to Huixia Yang) and Scientific Research Seed Fund of Peking University First Hospital (2018SF046 to Geng Song).
Author Contributions
G.S. analyzed the data and wrote the manuscript. J.Y.Y., M.X., X.L.Z., M.H.Z., Y.Y.M., H.W.L., J.X.S., and K.J.H. contributed to the data collection. H.X.Y., G.S., R.N.S., and Y.M.W. contributed to the study design and implementation. G.S. and J.J. contributed to the interpretation of the data analyses. H.X.Y. revised the manuscript. All authors contributed to the critical interpretation of the results, reviewed the manuscript and approved the final version for publishing. G.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Conflicts of Interest
None.
Editor Note
Huixia Yang is the Editor-in-Chief of Maternal-Fetal Medicine . The article was subject to the journal’s standard procedures, with peer review handled independently of this editor and her research groups.
Data Availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
1. Hod M, Kapur A, Sacks DA, et al. The International Federation of Gynecology and Obstetrics (FIGO) Initiative on gestational diabetes mellitus: A pragmatic guide for diagnosis, management, and care. Int J Gynaecol Obstet 2015;131(Suppl 3):S173–S211. doi:10.1016/S0020-7292(15)30033-3.
2. Billionnet C, Mitanchez D, Weill A, et al. Gestational diabetes and adverse perinatal outcomes from 716,152 births in France in 2012. Diabetologia 2017;60(4):636–644. doi:10.1007/s00125-017-4206-6.
3. Zhu WW, Yang HX, Wang C, et al. High prevalence of gestational diabetes mellitus in Beijing: effect of maternal birth weight and other risk factors. Chin Med J (Engl) 2017;130(9):1019–1025. doi:10.4103/0366-6999.204930.
4. Wei Y, Yang H. Perspectives on diagnostic strategies for hyperglycemia in pregnancy - dealing with the barriers and challenges in China. Diabetes Res Clin Pract 2018;145:84–87. doi:10.1016/j.diabres.2018.04.005.
5. Metzger BE, Lowe LP, et alHAPO Study Cooperative Research Group. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358(19):1991–2002. doi:10.1056/NEJMoa0707943.
6. Lowe WL Jr, Scholtens DM, Kuang A, et al. Hyperglycemia and adverse pregnancy outcome follow-up study (HAPO FUS): maternal gestational diabetes mellitus and childhood glucose metabolism. Diabetes Care 2019;42(3):372–380. doi:10.2337/dc18-1646.
7. Shou C, Wei YM, Wang C, et al. Updates in long-term maternal and fetal adverse effects of gestational diabetes mellitus. Maternal Fetal Med 2019;1(2):91–94. doi:10.1097/FM9.0000000000000019.
8. Kwak SH, Kim HS, Choi SH, et al.
Subsequent pregnancy after gestational diabetes mellitus: frequency and risk factors for recurrence in Korean women. Diabetes Care 2008;31(9):1867–1871. doi:10.2337/dc08-0384.
9. Kim C, Berger DK, Chamany S. Recurrence of gestational diabetes mellitus: a systematic review. Diabetes Care 2007;30(5):1314–1319. doi:10.2337/dc06-2517.
10. England L, Kotelchuck M, Wilson HG, et al. Estimating the recurrence rate of gestational diabetes mellitus (GDM) in Massachusetts 1998–2007: methods and findings. Matern Child Health J 2015;19(10):2303–2313. doi:10.1007/s10995-015-1750-x.
11. Schwartz N, Nachum Z, Green MS. The prevalence of gestational diabetes mellitus recurrence–effect of ethnicity and parity: a metaanalysis. Am J Obstet Gynecol 2015;213(3):310–317. doi:10.1016/j.ajog.2015.03.011.
12. Young C, Kuehl TJ, Sulak PJ, et al. Gestational diabetes screening in subsequent pregnancies of previously healthy patients. Am J Obstet Gynecol 2000;182(5):1024–1026. doi:10.1067/mob.2000.105395.
13. Yang HX. Diagnostic criteria for gestational diabetes mellitus (WS 331-2011). Chin Med J (Engl) 2012;125(7):1212–1213. doi:10.3760/cma.j.issn.0366-6999.2012.07.004.
14. Group of China obesity task force. Guidelines for prevention and control of overweight and obesity in Chinese adults. (In Chinese). Acta Nutrimenta Sinica 2004;26(1):1–4. doi:10.3321/j.issn:0512-7955.2004.01.001.
15. Institute of Medicine Weight Gain during Pregnancy: Reexamining the Guidelines. Washington, DC: National Academies Press, 2009.
16. Zhang C, Wei Y, Sun W, et al. The area under the curve (AUC) of oral glucose tolerance test (OGTT) could be a measure method of hyperglycemia in all pregnant women. Open J Obstet Gynecol 2019;9(2):186–195. doi:10.4236/ojog.2019.92019.
17. Lee KW, Ching SM, Ramachandran V, et al. Prevalence and risk factors of gestational diabetes mellitus in Asia: a systematic review and meta-analysis. BMC Pregnancy Childbirth 2018;18(1):494. doi:10.1186/s12884-018-2131-4.
18. Zhang C, Su S, Liu C, et al. Impact of area under the curve of oral glucose tolerance test on pregnant woman with gestational diabetes mellitus (in Chinese). Zhonghua Fu Chan Ke Za Zhi 2015;50(9):658–663. doi:10.3760/cma.j.issn.0529-567X.2015.09.004.
19. International Weight Management in Pregnancy (i-WIP) Collaborative Group. Effect of diet and physical activity based interventions in pregnancy on gestational weight gain and pregnancy outcomes: meta-analysis of individual participant data from randomised trials. BMJ 2017;358:j3119. doi:10.1136/bmj.j3119.
20. Wei Y, Yang H. Variation of prevalence of macrosomia and cesarean section and its influencing factors (in Chinese). Zhonghua Fu Chan Ke Za Zhi 2015;50(3):170–176. doi:10.3760/cma.j.issn.0529-567x.2015.03.002.
