To the Editor: Gestational diabetes mellitus (GDM) is a pathologic condition in which glucose intolerance and insulin resistance develop to different degrees during pregnancy. Women with GDM have higher odds of adverse maternal and fetal outcomes. Patients with a history of GDM were at a significantly increased risk for recurrence of GDM in subsequent pregnancies. Our previous and some other studies found an increased incidence of adverse perinatal outcomes in pregnant with recurrent GDM.[1] So far, women with a prior history of GDM are not recommended to refrain from subsequent pregnancy in light of the limited available data. What about those women with a history of recurrent GDM? Should women with recurrent GDM be recommended to avoid further pregnancy? Currently, studies on perinatal outcomes of women with re-recurrent GDM and their prognostic factors are very rare. Therefore, this retrospective cohort study was performed to explore perinatal outcomes of pregnant woman with re-recurrent GDM and their prognostic factors.

This cohort study was approved by the Ethics Committee of Fujian Maternity and Child Health Hospital (No. 2020–2049). Written informed consent was obtained from all participants. Women with GDM were diagnosed according to the diagnostic criteria established by the International Association of Diabetes and Pregnancy Study Groups (IADPSG),[2] which are also the diagnostic criteria for GDM in China. All recruited women were checked if they had a subsequent pregnancy and followed-up until termination before May 31, 2022. Maternal and fetal data were collected from the medical records of all participants.

Sample size estimation was performed using PASS software (version 15; Beijing Tianyan Rongzhi Software Co., Ltd., Beijing, China). All continuous values were given as mean ± standard deviation (SD) or median interquartile range (IQR). Student's t-test or Mann–Whitney test was used for continuous variables. Chi-squared test was used for categorical variables. Analyses were conducted using SPSS Statistics (version 26.0; IBM, Armonk, NY, USA), and a two-sided P-value <0.05 was taken as the level of significance. Risk factor values were extracted for a random subset of ΔNPV pixels using the R package "raster".

During 10 years from 2010 to 2022, 1547 women were diagnosed with recurrent GDM in Fujian Maternity and Child Hospital, a regional hospital in southeast of China. Of those, 170 women were excluded due to lack of oral glucose tolerance test after six weeks of postpartum or other data, 5 women were diagnosed with diabetes after delivery and 1372 women were eventually recruited. Among all recruited women, 22 women were lost in the follow-up and 182 women got pregnancy again. One hundred and eleven of them delivered after 24 weeks while the other 71 women terminated pregnancy before 24 weeks.

Among these 111 women, 65 (64%) complicated with GDM (re-recurrent group), 37 (37%) uncomplicated with GDM (recurrent group), and the other 9 (9%) women complicated with pre-gestational diabetes mellitus (PGDM) in the third pregnancy. One of woman complicated with PGDM was diagnosed by fasting glucose (7.29 mmol/L) at 18 weeks and stillbirth occurred finally at 28 weeks due to a hyperosmolar hyperglycemic state. The other three were diagnosed by oral glucose tolerance test (OGTT) at 8–15 weeks and delivered by vagina smoothly.

Between recurrent group (n = 37) and re-recurrent group (n = 65), there was no statistically significant difference in all baseline characteristics, including maternal age (years) (33.5 ± 3.4 vs. 33.7 ± 2.9), qualifications (upgraduate, 61.5% [40/65] vs. 78.4% [29/37]), mode of conception (natural conception, 96.9% [63/65] vs. 97.3% [36/27]), family history of diabetes (12.3% [8/65] vs. 8.1% [3/37]), gravidity (4.0 [3.0–5.0] vs. 4.0 [3.0–5.0]), macrosomia history (10.8% [7/65] vs. 8.1% [3/37]), preterm history (7.7% [5/65] vs. 13.5% [5/37]), and gestational weight gain (kg) (13.6 ± 3.6 vs. 15.0 ± 4.6) except pre-pregnancy body mass index (BMI) (kg/m2) (23.5 [22.0–24.9]) vs. (21.8 [19.8–23.7]) and interpregnancy interval (IPI) (all P >0.05). IPI months to first pregnancy (55.3 ± 5.8 vs. 68.9 ± 3.0, P <0.001) and second pregnancy (months) (26.9 ± 7.3 vs. 41.0 ± 5.7, P <0.001) were significantly shorter in re-recurrent group than recurrent group.

The value of 2 h-plasma glucose (PG, mmol/L) (9.0 [8.6–9.7] vs. 8.1 [6.7–8.6], P <0.001), number of OGTT abnormal items (P = 0.001), level of cholesterol (TC, mmol/L) (6.9 ± 1.0 vs. 6.1 ± 0.7, P <0.001) in third trimester and level of glycosylated hemoglobin (HbA1c, %) in second (5.6 [5.3–5.9] vs. 5.4 [5.1–5.5], P <0.001) and third trimester (5.6 [5.4–5.9] vs. 5.0 [4.9–5.4], P <0.001) during second pregnancy was significantly higher in re-recurrent group than recurrent group. While the level of fasting plasma glucose (FPG, mmol/L) in first trimester (4.9 [4.5–5.0] vs. 4.8 [4.6–5.0]), 0 h (4.7 ± 0.6 vs. 4.6 ± 0.5), and 1 h-PG (mmol/L) (10.5 [10.2–11.0] vs. 10.8 [10.20–11.07]), total cholesterol (TG) in third trimester (mmol/L) (4.1 [3.1–5.1] vs. 3.2 [3.0–4.6]), and insulin therapy (7.7% [5/65] vs. 0% ) between two groups showed no significant difference.

The level of TG (mmol/L) (2.3 ± 0.5 vs. 1.8 ± 0.4, P = 0), TC (4.7 [4.4–5.1] vs. 3.8 [3.8–4.3], P <0.001), FPG in first trimester (mmol/L) (4.9 ± 0.4 vs. 4.7 ± 0.3, P = 0.03) and FPG before delivery (5.09 ± 0.74 vs. 4.16 ± 0.34, P <0.001) were significantly higher in re-recurrent group than recurrent group.

The rate of neonatal intensive care unit (NICU) admission in re-recurrent group was significantly higher than recurrent group (26.2% [17/65] vs. 8.1% [3/37], P = 0.03). There was no significant difference in other perinatal outcomes including hospitalization days (days) (4.0 [4.0–6.0] vs. 5.0 [3.0–6.5]), hospitalization expenses (yuan) (6086.69 [4862.38–9960.77] vs. 7496.34 [4709.96–9172.01]), gestational week of delivery (weeks) (38.6 ± 2.0 vs. 39.1 ± 0.9), insulin therapy (9.2% [6/65] vs. 10.8% [4/37]), mode of delivery (cesarean-section, 15.4% [10/65] vs. 13.5% [5/37]), gestational hypertension disease (GHD) (6.2% [4/65] vs. 13.5% [5/37]), thyroid dysfunction (16.9% [11/65] vs. 5.4% [2/37]), intrahepatic cholestasis of pregnancy (ICP) (1.5% [1/65] vs. 2.7% [1/37]), premature rupture of membranes (PROM) (15.4% [10/65] vs. 13.5% [5/37]), precipitate labor (9.2% [6/65] vs. 10.8% [4/37]), postpartum hemorrhage (PPH) (4.6% [3/65] vs. 5.4% [2/37]), small for gestational age (SGA) (6.2% [4/65] vs. 0%), large for gestational age (LGA) (15.4% [10/65] vs. 10.8% [4/37]), macrosomia (12.3% [8/65] vs. 10.8% [4/37]), Apgar score at 1 min (score) (10 [10–10] vs. 10 [10–10]) and weight of fetus (grams) (3460.00 [3445.00–3750.00] vs. 3320.00 [3170.00–3450.00]) between two groups (all P >0.05).

Random forest algorithm was used to analyze the risk factors affecting perinatal outcomes. The main influencing factors included leading factors on GHD, thyroid dysfunction, PROM, delivery way, PPH, rate of NICU admission, LGA, and SGA were respectively pre-pregnancy BMI, the value of HbA1c in third trimester during second pregnancy, number of OGTT abnormal items during second pregnancy, level of TC in third trimester during second pregnancy, IPI months to second pregnancy, the value of HbA1c in third trimester during second pregnancy, level of TC in third trimester during second pregnancy and level of TC in first trimester during third pregnancy [Supplementary Figure 1, https://links.lww.com/CM9/B736].

This study demonstrated that the frequency of re-recurrent GDM for women with recurrent GDM in the third pregnancy was nearly 65%. Many studies evaluated and reported the rates of recurrence of GDM that range widely from 30% to 84%.[3] The finding of this study is reasonable. On the one hand, the reported average frequency of GDM recurrence was 48%. On the other hand, primiparous women experienced a lower rate of GDM recurrence than multiparous women, the frequency will be increased along with parity. Besides, women with recurrent GDM have cumulative damage in every additional pregnancy, such as their β-cell reserves are diminished, and therefore are at higher increased risk for GDM recurrence in third pregnancy.

This study found that the rate of common obstetric complications showed no significant difference in re-recurrent GDM women except for hypothyroidism. In fact, the relationship between hypothyroidism and GDM is still controversial. Some studies conclude that hypothyroidism during pregnancy is mostly caused by Hashimoto's thyroiditis and inflammatory factors in these patients may lead to oxidative stress injury and increase insulin resistance and cause GDM.

The rate of NICU admission in re-recurrent GDM was much higher in this study. The high value of HbA1c in third trimester during second pregnancy and FPG in first trimester during third pregnancy may explain this result and also indicate the importance of blood-glucose management for pregnant women with re-recurrent GDM.

The literature on risk factors for outcomes of recurrent GDM is unclear. In Israel, Schwartz et al[4] performed a cross-sectional cohort study that included 788 women with GDM in ten years and had consecutive deliveries. The study revealed that shorter inter-pregnancy interval and fewer weight gain were preferable. In order to further demonstrate the role of inter-pregnancy interval and BMI gain between the pregnancies, the authors calculated two scenarios with fixed information and different combinations of the main risk factors. They got that by losing weight between the pregnancies (a reduction of 0.5 BMI units) instead of gaining weight (increase of 1.5 BMI units), and waiting one year between the pregnancies (instead of two years), the woman will have a decrease of 15–19% in the probability of GDM recurrence. A possible explanation is that among women who suffered from GDM, the β-cell reserves are already diminished and therefore longer inter-pregnancy interval suggests a longer time in which the reserve is decreased even further. However, some studies found that a short interval between pregnancies was risk factor for recurrence of GDM. There are two possible explanations. One is that the damage of β-cell reserves caused by GDM can recover to a certain extent in some interval of inter-pregnancy, and the other is that there is no enough time to recover or lose weight, which is a protective factor for recurrence of GDM. In this study, the mean intervals to first pregnancy and second pregnancy were respectively 4 years and 25 months for women with re-recurrent GDM and which were shorter than women without re-recurrent. Appropriate interval between pregnancies, however, needs further studies in the future.

This study found that the level of plasma lipid and glucose during the previous pregnancy and first trimester of this pregnancy may be risk factors for the outcome of re-recurrent GDM. The results are consistent with some other studies.[5] Higher levels of lipid and FPG are associated with insulin resistance and β-cell dysfunction. With the development of pregnancy, insulin resistance aggravates, β-cells secrete unenough insulin, insulin sensitivity decrease, and GDM recurrent at last for women with higher levels of lipid and FPG in early pregnancy.

Studies with greater numbers of subjects are required to clarify the exact re-recurrence rate of GDM. Prospective studies regarding the reduction of the subsequent GDM by advice on the interval with the next pregnancy are required. Although a prospective study is required, our study demonstrates that there is a possibility that the adequate advice concerning IPI and the level of plasm of lipid and glucose could help prevent the recurrence of GDM and improve the prognosis of perinatal outcomes.

Acknowledgments

The authors thank the statistical guidance and assistance of Prof. Yingying Su from the School of Public Health, Xiamen University, China.

Funding

This study was supported by grants from the Guide Fund for the Development of Local Science and Technology from the Central Government (No. 2020L3019), the Natural Science Foundation of Fujian Province of China (No. 2020J01314), and Joint Funds for the Innovation of Science and Technology, Fujian province (No. 2020Y9161).

Conflicts of interest

None.

References 1. Guo Y, Xu X, Xu W, Liao T, Liang J, Yan J. Subsequent perinatal outcomes of pregnancy with two consecutive pregnancies with gestational diabetes mellitus: A population-based cohort study. J Diabetes 2022;14: 282–290. doi: 10.1111/1753-0407.13263. 2. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, Dyer AR, Leiva Ad, Hod M, Kitzmiler JL, Lowe LP, McIntyre HD, Oats JJ, Omori Y, Schmidt MIInternational Association of Diabetes and Pregnancy Study Groups Consensus Panel. International Association of Diabetes and Pregnancy Study Groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy: Comment to the International Association of Diabetes and Pregnancy Study Groups Consensus Panel. Diabetes Care 2010;33: 676–682. doi: 10.2337/dc09-1848. 3. Filardi T, Tavaglione F, Di Stasio M, Fazio V, Lenzi A, Morano S. Impact of risk factors for gestational diabetes (GDM) on pregnancy outcomes in women with GDM. J Endocrinol Invest 2018;41: 671–676. doi: 10.1007/s40618-017-0791-y. 4. Schwartz N, Green MS, Yefet E, Nachum Z. Modifiable risk factors for gestational diabetes recurrence. Endocrine 2016;54: 714–722. doi: 10.1007/s12020-016-1087-2. 5. Bao W, Dar S, Zhu Y, Wu J, Rawal S, Li S, et al. Plasma concentrations of lipids during pregnancy and the risk of gestational diabetes mellitus: A longitudinal study. J Diabetes 2018;10: 487–495. doi: 10.1111/1753-0407.12563.