Epidemiological characteristics of DIP
The present study showed that the incidence of DIP is higher in China than in other countries, while the incidence of GDM is similar [3, 5, 6]. The high incidence of DIP may be related to the establishment of the diabetes innovation center in our hospital, which receives many patients who are referred from other hospitals.
Poor glycemic control in DIP results in poor maternal and fetal prognoses. The incidence rates of abortion and stillbirth in DIP are 10–17% [19] and 5.9% [20], respectively, while the incidence rates of premature delivery in PGDM and GDM are 14.24% and 8.98%, respectively [21]. The aORs of PGDM and GDM for fetal malformations were 2.44 and 1.28, respectively [22]. The incidence rates of macrosomia in T1DM and T2DM were 51% and 38%, respectively [23]. The results of our study suggested that DIP has a higher incidence and that it is associated with adverse pregnancy outcomes, and thus, this condition warrants greater clinician attention.
High-risk factors for GDM
Studies have suggested that the high-risk factors for GDM include race, advanced age, overweight or obesity, history of diabetes among first-degree relatives, and previous history of macrosomia or GDM, HDPs, and polycystic ovary syndrome [24]. Asians are a high-risk group for GDM. According to the previous reports, among the high-risk factors for GDM, age ≥ 32 years and overweight accounted for 27.9% and 28.3%, respectively [25]. The risk of GDM increases by 1.883 times when the maternal age is ≥36 years [26]. The present study showed that advanced age is the most common risk factor for GDM, accounting for 40% of cases, followed by overweight, which accounted for 24.37% of cases. The incidence of GDM increases with age. Owing to changes in our lifestyle, the number of overweight or obese women has been increasing yearly, which may be one of the reasons for the increase in the incidence of GDM. Giving birth at a suitable age, if possible, can reduce the risk of GDM. Women who are overweight or obese before pregnancy can undertake lifestyle interventions for proper weight control or weight loss to help reduce their risk of GDM and adverse pregnancy complications. Zhang et al. [27] found that the incidence of GDM among women undergoing assisted reproductive technology (ART) treatment was 1.9 times that among women who conceived naturally. The use of progesterone in early pregnancy may have an impact on glucose metabolism and increase the incidence of GDM. The present study found that IVF-ET accounted for 22.18% of GDM cases, and is very likely to be a high-risk factor for GDM. In addition, women undergoing IVF-ET have an increased rate of multiple pregnancies, which is also a high-risk factor for GDM. In this study, 65.20% of women diagnosed with GDM had at least one high-risk factor. Consistent with this, O’Sullivan et al. [28] found that if only women with high-risk factors for GDM were screened using OGTTs, the diagnosis rate of GDM would be 62%. The mean gestational age at the time of GDM diagnosis was lower in women with high-risk factors than in women without high-risk factors. Thus, early screening of those with high-risk factors for GDM can not only increase the diagnosis rate of GDM but also lead to the early detection of GDM. This is useful because early intervention can effectively reduce the occurrence of adverse pregnancy outcomes in such patients. Nevertheless, if only those with high-risk factors for GDM are screened, the missed diagnosis rate will be 34.8%. Therefore, while early screening of women with high-risk factors is recommended, women without high-risk factors should be routinely screened during the period of rapid fetal growth period at 24–28 weeks.
Comorbidities and complications of DIP
HDPs are one of the most common complications of DIP. Women with DIP are 2–4 times more likely to develop HDPs than women without DIP [16, 29]. When diabetes is accompanied by microvascular disease, particularly renal microvascular disease, the incidence of HDPs and preeclampsia can be as high as above 50% [16]. Women with HDPs often have abnormal glucose metabolism, and these two conditions affect each other. In this study, HDPs were the main comorbidity of DIP, accounting for 6.75% of cases. PROM was the main complication, accounting for 18.81% of cases. This finding may be explained as follows: abnormal glucose metabolism can easily alter the vaginal flora, which reduces the local tension of the fetal membranes and results in PROM. Therefore, glycemic control in pregnancy may help reduce the occurrence of PROM. Polyhydramnios may be related to fetal hyperglycemia, hyperosmolar diuresis, and increased fetal urine excretion. Bicocca et al. [30] found that the incidence of polyhydramnios in DIP was 10.5%. However, the incidence of polyhydramnios in this study was low (1.05%), and the incidence of puerperal infection (6.05%) was consistent with that in the normal population (6%) [16]. These findings may be attributable to the establishment of the innovative O2O office for GDM by our team, which helped to achieve good glycemic control in DIP, and thereby reduce the incidence of polyhydramnios and puerperal infections. For women who had a normal OGTT in the second trimester but are found to have polyhydramnios in the third trimester, the OGTT may be repeated to detect latent GDM in a timely manner. For women with third-trimester polyhydramnios who have already been diagnosed with DIP, the diabetes diet and exercise treatment plan should be adjusted as soon as possible.
In recent years, DIP associated with cervical insufficiency has garnered increasing attention from obstetricians. The incidence of cervical insufficiency in this study was 3.60%, which is higher than the rates of 0.1–1.0% reported in the literature [31]. This may be related to the fact that our hospital is a treatment center for severely ill pregnant women in Guangdong Province, the establishment of an innovative office for the management of cervical insufficiency, and the referral of many patients with cervical insufficiency from other hospitals. Cervical insufficiency is an important factor for abortion and premature delivery in the second trimester. Relative bed rest is one of the methods used to manage this condition.
DIP treatment and insulin usage characteristics
After standardized diet and exercise treatment, if the blood glucose still does not reach the target, medication should be promptly added. At present, insulin is the first-line treatment for hyperglycemia due to DIP. It has been reported that 24.1–33.7% of GDM patients require insulin [32, 33]. Among PGDM patients, the maximum insulin dose to treat T1DM during pregnancy is at least twice as high as that required before pregnancy, while T2DM often necessitates additional insulin treatment or a rapid increase in insulin dose during 28–32 gestational weeks, which is a period of rapid fetal development [15]. Padmanabhan et al. [34] found that in the third trimester, the required insulin dose increased by 22.9% in T1DM and by 44% in T2DM; furthermore, patients with T1DM had a slight decrease in insulin dose before delivery, while those with T2DM did not. Roeder et al. [35] found that the insulin dose immediately after delivery in T1DM patients should be 30–35% less than the pre-pregnancy dose. If the pre-pregnancy insulin dose is unknown, the insulin dose should be reduced to 50% of the pre-delivery dose and adjusted according to the blood glucose level.
The present study found that the incidence rates of insulin use were 14.06% in DIP patients, 55% in PGDM patients, and 8.53% in GDM patients, which are lower than the rates reported in the literature [32, 33]. This finding may be related to the implementation of the O2O management mode and the standardized implementation of diet and exercise in our hospital. The mean gestational age at which insulin treatment was initiated was lower in the PGDM group than in the GDM group, and the mean initial and maximum insulin doses were higher in PGDM group than in the GDM group. These findings may be related to the presence of islet dysfunction in women with PGDM and the aggravation of insulin resistance during pregnancy. The mean gestational age at the time of the peak insulin dose in women with DIP was 32 weeks, and this parameter did not significantly differ between the PGDM and GDM groups, indicating that in both groups, insulin resistance was most obvious during the period of rapid fetal development, and it was necessary to adjust the dose of insulin in a timely manner to maintain a stable blood glucose level.
In the third trimester, the required insulin dose decreased in 50.80% of women with PGDM and 47.44% of women with GDM, with no significant between-group difference. This is inconsistent with the results reported by Padmanabhan et al. [34], and may be related to the failure to detect the type of PGDM in this study and to distinguish between the doses used in the third trimester of pregnancy and the pre-delivery period. Our results suggested that GDM is also associated with the phenomenon of physiological insulin reduction in the third trimester.
In the present study, postpartum insulin use was required in 21.14% of women with DIP, 39.04% of women with PGDM, and 5.58% of women with GDM. The mean postpartum insulin dose was 32.07% and 26.20% of the antepartum dose in the PGDM and GDM groups, respectively, which is consistent with the literature [16, 35]. With the delivery of the placenta, insulin resistance is significantly reduced, and the insulin concentration is significantly increased. Therefore, the insulin dose should be carefully adjusted in the postpartum period, particularly for breastfeeding patients. It is recommended that the initial postpartum dose be one-third of the pre-delivery or pre-pregnancy dose to avoid the occurrence of hypoglycemia.
Risk analysis of adverse maternal and fetal pregnancy outcomes in DIP
Abortion
The incidence of abortion in the DIP group was 3.62%, which is lower than that reported in the literature [16]. This may be due to the fact that our hospital is a specialized treatment center for severely ill pregnant women, and few patients who require abortion seek admission into our hospital. To reduce the incidence of abortion in DIP, we recommend the following measures: (1) in the case of women with PGDM, the blood glucose level should be controlled within the normal range before pregnancy, (2) the HbA1c level should be <6.5%, and (3) multidisciplinary assessment should be used to determine whether these women can safely become pregnant.
Premature delivery
The incidence of premature delivery in DIP was 10–25%. The risk of polyhydramnios in women with DIP was 10 times that in women without DIP. The higher the blood glucose level, the more common is polyhydramnios, which can lead to premature delivery [16]. Lin et al. [36] found that among women with GDM, the incidence of premature delivery was significantly higher in those with poor glycemic control than in those with good glycemic control.
The present study found that the incidence of premature delivery in DIP was 23.33%, which is consistent with that reported in the literature [16]. The incidence of premature delivery in the PGDM group was significantly higher than that in the GDM A1 and A2 groups, and the incidence in the GDM A2 group was significantly higher than that in the GDM A1 group. Compared with GDM, PGDM was associated with 1.533 times the risk of premature delivery, which is consistent with the risk reported in the literature [13].
During pregnancy, women with DIP require more frequent blood glucose monitoring, good control of the blood glucose level within the standard range, and timely insulin treatment when necessary to reduce the incidence of premature delivery.
Stillbirth
Reports on the incidence of stillbirth vary globally, with the reported rates in developed and developing countries being 3.1% and 30%, respectively [37, 38]. In China, the incidence of stillbirth is 8.8% [39]. If DIP is not controlled, hyperglycemia and diabetic ketoacidosis can occur, which can lead to stillbirth. A large study [20] analyzed the data of 10,733,983 newborns in the United States from 1995 to 1997. The results showed that the risk of stillbirth was higher in women with DIP than in women without DIP (5.9% vs. 4.0%). Tennant et al. [40] found that the risk of stillbirth was increased by 4 times in women with PGDM compared to women without DIP. In the third trimester, HbA1c > 5.8% and lack of antenatal folic acid supplementation were the only variables that were significantly associated with stillbirth. HbA1c > 6.6% was independently associated with the risk of stillbirth. Among women with HbA1c levels > 6.6%, every 0.1% increase in the HbA1c level during the perinatal period increased the probability of stillbirth by 2% [40].
The incidence of stillbirth in the present study was 0.56%. The incidence of stillbirth was significantly lower in the good glycemic control group than in the poor glycemic control, while it did not differ among the GDM A1, GDM A2, and PGDM groups. Actively controlling the perinatal blood glucose levels and the maternal-fetal weight gain to remain within the standard target values is an effective way to prevent stillbirth in DIP.
LGA
Landon et al. [41] found that insulin treatment decreased the risk of macrosomia from 14.3–5.9% among women with GDM. Mackin et al. [23] found that the incidence of macrosomia in T1DM and T2DM was 51% and 38%, respectively. In the present study, the incidence of LGA was 3.95%, which is significantly lower than that reported in the literature.
Birth defects
Wu et al. [22] found that the aORs of PGDM and GDM for fetal malformations were 2.44 (95% CI: 2.33–2.55) and 1.28 (95% CI: 1.24–1.31), respectively. Nielsen et al. [9] found that increased HbA1c level was related to pregnancy outcomes. For every 1% increase in the HbA1c level, the risk of adverse pregnancy outcomes increased by 3.8–7.3%.
In our study, the incidence of birth defects was 9.37% in the DIP group, and this incidence did not significantly differ among the GDM A1, GDM A2, and PGDM groups, or between the good and poor glycemic control groups.
Neonatal asphyxia and transfer to NICU
Hyperglycemia and hyperinsulinemia in DIP can affect the biosynthesis of fetal type 2 alveolar cell surface-active substances and the development and maturation of the fetal lungs, which increases the risk of neonatal respiratory distress syndrome, neonatal asphyxia, and transfer to NICU. In the present study, the incidence of neonatal asphyxia was 7.90%. The incidence rates of transfer to the NICU and neonatal asphyxia significantly differed between the GDM A1, GDM A2, and PGDM groups, but did not significantly differ between the good and poor glycemic control groups nor between the insulin and non-insulin groups.
The high incidence of transfer to the NICU in the PGDM group may be related to the high incidence of premature delivery. Clinicians must pay careful attention to prevent premature delivery in women with DIP. For women who are at risk of premature delivery, steroids should be administered in time to promote fetal lung maturation after controlling the blood glucose level to within the standard range. Studies have found that after the administration of steroids to women with DIP, the fasting and postprandial blood glucose levels are increased, and most women need more than twice the previous insulin dose [42]. Therefore, during steroid administration, it is necessary to closely monitor the blood glucose level and to use insulin in a timely manner to reduce the incidence of adverse pregnancy outcomes. Our team found that among women who receive intramuscular injections of dexamethasone 6 mg twice a day for a total of 2 days to promote fetal lung maturation, dexamethasone-induced hyperglycemia can be adequately managed with the concurrent subcutaneous injections of 4–6 U insulin (Detemir) once per day for a total of 2 days.
Gestational age and mode of delivery
The ACOG guidelines [43] recommend that GDM A1 patients with good glycemic control through exercise and diet and no other indications for induction of labor are not usually recommended to undergo delivery before 39 weeks. In such patients, close monitoring until 40+6 weeks is appropriate. In the case of GDM A2, patients who need drugs for glycemic control are recommended to deliver at 39 to 39+6 weeks; for women with poor glycemic control, early delivery is recommended. Delivery at 37–38+6 weeks can be considered, if the blood glucose level is not well controlled after hospitalization; if the prenatal fetal monitoring is abnormal, delivery at 34–36+6 weeks should be considered [45]. According to the guidelines for DIP in China, pregnant women with PGDM on insulin therapy can deliver after 39 weeks of gestation if their blood glucose is well controlled, and there are no maternal and fetal complications. If blood glucose control is not satisfactory, or if maternal or fetal complications occur, these patients should be promptly admitted to a hospital for observation, and the timing of delivery should be determined according to their specific condition [17].
Limitations and future prospects
This was a single-center study, and the number of cases in some subgroups was limited. This study only analyzed the short-term adverse pregnancy outcomes of DIP in mothers and infants. A follow-up study on the long-term effects is ongoing. A multi-center, large-scale, prospective group-controlled study is feasible in the future, as well as long-term follow-up of mothers with DIP and their offspring after delivery.