PDF
Citation
Print
INTRODUCTION
Pregestational diabetes mellitus (PGDM) is a complex metabolic disorder that presents unique challenges to pregnant women. Diabetes mellitus (DM) affects women of reproductive age, with a prevalence of 3.1–6.8%, and PGDM is observed in approximately 1–2% of all pregnancies.123 PGDM, which is characterized by elevated blood glucose levels before conception, can have a profound impact on both maternal and fetal health throughout the gestational period. Managing glucose levels in women with PGDM is of paramount importance to mitigate the associated risks and complications.
The impact of glucose levels in women with PGDM extends beyond the individual and influences the course of pregnancy, fetal development, and long-term health of both the mother and child. Elevated glucose levels in these women can lead to a range of adverse outcomes such as an increased risk of birth defects, macrosomia, preterm birth, preeclampsia, and neonatal hypoglycemia.4567891011 Moreover, uncontrolled diabetes during pregnancy can also contribute to maternal complications, including a higher likelihood of hospital readmission, kidney disease, and the progression of diabetic complications.1213
Therefore, blood glucose control during pregnancy is important in women with PGDM. However, some studies have shown that blood glucose control during pregnancy alone may not fully explain the occurrence of adverse pregnancy outcomes and neonatal morbidity in women with PGDM. With the increasing incidence of type 2 DM, there is a growing emphasis on diagnosing undiagnosed type 2 DM early in pregnancy,14 highlighting the importance of early glucose control. Women who develop gestational diabetes mellitus (GDM) during early pregnancy appear to be more vulnerable to changes in metabolism. They are often insulin-resistant and exhibit characteristics such as greater waist circumference, higher blood pressure, and elevated triglyceride levels.15 Furthermore, pregnancy outcomes are worse in women who develop GDM during early pregnancy.16
In some studies, controlling blood glucose levels after pregnancy may not significantly improve maternal and offspring outcomes, which could be attributed to the intervention starting too late. Catalano17 showed that women with normal prepregnancy glucose tolerance who developed GDM in late pregnancy exhibited subclinical metabolic dysfunction prior to conception compared to women with normal glucose tolerance without GDM. Their research also suggested that maternal pre and early pregnancy metabolic conditions associated with obesity, such as increased insulin resistance and inflammation, affect early placental function and gene expression. These alterations in placental function occur in the first trimester of pregnancy before most intervention trials are initiated.
Therefore, research on maternal blood glucose levels before conception and during early pregnancy is important. Recent studies have demonstrated the potential of early pregnancy glucose screening using specific cutoff values, such as random blood glucose levels of 100 mg/dL and HbA1c levels of 5.2%, to identify high-risk cases and enable early detection of glucose intolerance.18 However, few studies have focused on the preconception maternal glucose levels, and there is a lack of evidence regarding whether specific fasting blood glucose levels in women with diabetes impact pregnancy outcomes. To address this issue, we aimed to evaluate the impact of prepregnancy maternal fasting plasma glucose (FPG) levels on pregnancy and neonatal morbidity in women with PGDM.
METHODS
Study population
Information on the study population was gathered from the Korean Health Insurance Review and Assessment (HIRA) service database, which is linked to the national, single-payer healthcare system. In this system, nearly the entire Korean population (approximately 97%) is enrolled in the National Health Insurance Service (NHIS), and HIRA collects comprehensive claims data for all beneficiaries. Furthermore, the study utilized data from two additional sources: the bi-annual National Health Screening Examination (NHSE) provided by the NHIS and the National Health Screening Program for Infants and Children (NHSP-IC). The NHSE offers insights into pregestational glucose levels, whereas the NHSP-IC, initiated in 2007, provides valuable information on physical examinations, anthropometric measurements, and developmental screening results for infants and children.
Study design
The study population consisted of pregnant women who met the following criteria: 1) singleton pregnancy; 2) delivery between 2015 and 2021; 3) participation in the NHSE within 1 year prior to pregnancy; and 4) diagnosis of any type of DM before pregnancy. Patients lacking detailed clinical information were excluded. Women were excluded from the analysis of neonatal outcomes if their children had not participated in at least one of the seven consecutive NHSP-IC health examinations.
NHSE before pregnancy
Prepregnancy factors were obtained from the NHSE database, which consists of two main components: health interviews and health examinations. During the preconception health examination, the participants provided overnight fasting blood samples, from which their blood glucose concentrations were measured in the clinic. FPG levels were analyzed using automatic analyzers approved by the Ministry of Food and Drug Safety of Korea and selected by the local laboratories.
Diagnosis of DM and categorization
Women diagnosed with diabetes within 1 year before conception were defined using the International Classification of Diseases-10 (ICD-10) code (E10–14). Incident cases of type 2 DM were defined in accordance with the recommendations of the Korean Diabetes Association.19 The pregestational FPG categories were divided into 6 groups as follows: 1) FPG ≥ 126 mg/dL (diabetic range), 2) 100 mg/dL ≤ FPG < 126 mg/dL (prediabetic range), and the euglycemic range was divided into four groups based on the interquartile range: 3) 95 mg/dL ≤ FPG < 100 mg/dL, 4) 90 mg/dL ≤ FPG < 95 mg/dL, 5) 85 mg/dL ≤ FPG < 90 mg/dL, 6) FPG < 85 mg/dL.
Pregnancy and neonatal outcomes
We utilized the HIRA database to identify women who experienced hypertensive disorders during pregnancy (HDP), postpartum hemorrhage, and chronic hypertension and underwent cesarean section during pregnancy based on ICD-10 diagnostic codes. Information regarding neonatal outcomes such as preterm birth and birth weight was obtained from the NHSP-IC database. The primary outcomes of our study were neonatal outcomes diagnosed within 30 days of delivery. These outcomes included transient tachypnea of newborn, respiratory distress syndrome, necrotizing enterocolitis, intraventricular hemorrhage, bronchopulmonary dysplasia, and hypoglycemia. Additionally, we considered neonatal sex and birth weight as neonatal outcomes under investigation.
Statistical analysis
Continuous variables are described using mean and standard deviation and compared using Student's t-test or ANOVA for multiple group comparisons. Categorical variables are presented as numbers and percentages, and χ2 test was used for comparisons. The study participants were stratified into six groups based on their pregestational fasting glucose levels. Multivariate logistic regression analysis was conducted to estimate the adjusted odds ratios (aORs) and the corresponding 95% confidence intervals (CIs). We conducted a subgroup analysis based on body mass index (BMI) to examine these hypotheses, as blood glucose regulation and insulin resistance can vary depending on BMI, and adiposity may also affect blood glucose control. Participants were categorized into three groups based on the BMI cutoffs recommended by the World Health Organization for the Asia-Pacific region.20 Additionally, to explore the association between FPG and obstetric and neonatal outcomes, we examined these relationships using restricted cubic splines. Analyses were performed using SAS software version 9.4 for Windows (SAS Inc., Cary, NC, USA), and statistical significance was set at a P value < 0.05.
RESULTS
Baseline characteristics
In total, 31,739 pregnant women diagnosed with DM within 1 year prior to pregnancy were screened. Among the participants screened, those who had not undergone NHSE within one year before pregnancy (n = 23,376), had multifetal pregnancy (n = 686), or missing data (n = 135) were excluded. The final cohort comprised 7,542 participants, as shown in Fig. 1.
Fig. 1
Configuration of the cohort. (A) Schematic of the study design. (B) Flow chart for study design and the number of subjects.
FPG = fasting plasma glucose.
Table 1 shows the baseline characteristics of the study population. When comparing groups based on FPG categories, individuals with higher FPG levels were older, had higher prepregnancy BMIs, a higher prevalence of chronic hypertension, and elevated systolic and diastolic blood pressures (Supplementary Table 1). The distribution of prepregnancy FPG levels showed that the highest proportion of participants (22.3%) had glucose levels between 100-126 mg/dL, followed by 18.5% with levels below 85 mg/dL (Supplementary Fig. 1).
Table 1
Characteristics of study participants
Pregnancy outcomes and neonatal outcomes
The incidence rates of pregnancy and neonatal outcomes were assessed according to pregestational FPG categories, as shown in Table 2. Higher FPG levels were associated with increased rates of deliveries at an earlier gestational age and higher offspring birth weights. There was also a higher incidence of delivering large-for-gestational-age (LGA) or macrosomic infants as well as an increased likelihood of undergoing cesarean section. Additionally, the occurrence of HDP and preterm birth increased with higher FPG levels. However, there was no significant correlation between FPG levels and postpartum hemorrhage. Regarding neonatal morbidity, significant increases in transient tachypnea of newborns, respiratory distress syndrome, and hypoglycemia were observed in offspring born to mothers with higher FPG levels.
Table 2
Pregnancy and neonatal outcomes
The risk of adverse pregnancy outcomes and neonatal morbidity
Pregnancy and neonatal outcomes were assessed according to the pregestational FPG categories after adjusting for age, parity, hypertensive disorder during pregnancy, chronic hypertension, BMI, systolic and diastolic blood pressures, and neonatal sex, women with poorly controlled FPG levels, which remained at or above 126 mg/dL during prepregnancy, had an increased risk of preterm birth compared to women with FPG levels below 85 mg/dL (aOR, 1.66; 95% CI, 1.18–2.34). Similarly, the risk of delivering a macrosomic baby was significantly higher for women with prediabetes (100 ≤ FPG < 126) with an aOR of 2.97 (95% CI, 2.04–4.33), and even higher for women with diabetes (FPG ≥ 126) with an aOR of 4.77 (95% CI, 3.25–6.99), after adjusting for other variables.
Regarding neonatal morbidity, infants born to women with diabetes-stage FPG levels had a higher risk of transient tachypnea of newborn compared to infants born to women with FPG < 85 (aOR, 1.96; 95% CI, 1.27–3.03). The risk of respiratory distress syndrome was also increased in this group (aOR, 1.44; 95% CI, 1.03–2.04). The risk of neonatal hypoglycemia escalated from the prediabetes stage (100 ≤ FPG < 126) (aOR, 3.21; 95% CI, 2.13–4.83) and reached its highest in the diabetes stage (FPG ≥ 126) (aOR, 7.72; 95% CI, 5.14–11.59) (Table 3). A trend of increased risk with higher levels of FPG for preterm birth, macrosomia, and LGA baby delivery was observed. Conversely, the risk of small for gestational age baby delivery decreased with increasing FPG levels. Regarding neonatal outcomes, as FPG levels increased, the risk of transient tachypnea and hypoglycemia in the newborn also increased. The relationships between the risk of adverse outcomes or neonatal morbidity and FPG values are visualized in Fig. 2 using restricted cubic splines.
Table 3
Pregnancy and neonatal outcomes
CI = confidence interval, FPG = fasting plasma glucose.
aAdjusted for maternal age, parity, chronic hypertension, hypertensive disorder during pregnancy, body mass index before pregnancy, systolic and diastolic blood pressure, fetal sex, smoking and exercise.
bAdjusted for maternal age, parity, chronic hypertension, body mass index before pregnancy, systolic and diastolic blood pressure, fetal sex, smoking and exercise.
cAdjusted for maternal age, parity, chronic hypertension, hypertensive disorder during pregnancy, body mass index before pregnancy, systolic and diastolic blood pressure, fetal sex, preterm delivery, smoking and exercise.
Subgroup analysis
We conducted a subgroup analysis based on the prepregnancy BMI categories. The participants were divided into three groups: those who were underweight (< 18.5 kg/m2), those with normal weight (18.5–22.9 kg/m2), and those classified as overweight/obese (≥ 23 kg/m2). In the case of preterm delivery, in the underweight group, the risk of preterm delivery increased steeply beyond a certain high FPG cutoff, indicating that the risk escalated as FPG levels increased. Similarly, the OR for delivering a LGA baby also demonstrated a tendency to increase most rapidly in the underweight group. In the case of HDP, there was an observed tendency for an increased risk of HDP as FPG levels increased only in the obese group. Regarding the risk of neonatal hypoglycemia, women with a normal BMI experienced the most rapid increase in risk, followed by those who were underweight and obese, in terms of the steepness of risk escalation (Supplementary Fig. 2).
DISCUSSION
The main findings of the current study are as follows: 1) Elevated FPG levels were associated with an increased likelihood of delivering infants at an earlier gestational age and with higher birth weights; 2) A higher incidence of delivering infants with large for gestational age or macrosomia, an increased likelihood of undergoing a cesarean section, of developing hypertensive disorder during pregnancy, and a preterm birth was noted as FPG levels increased; 3) Regarding neonatal morbidity, infants born to women with diabetes-stage FPG levels had a higher risk of transient tachypnea of newborn, respiratory distress syndrome, and hypoglycemia compared to infants born to women with euglycemic FPG levels.
Several studies have explored the relationship between PGDM and pregnancy outcomes.9212223 Recently, research has not only focused on the association with underlying medical conditions but also on managing blood glucose levels during pregnancy and the improvement in pregnancy outcomes and impact on women's future health.242526 This perspective aligns with the findings of the current study, which indicate that in women with diabetes, particularly those with FPG levels above 126 mg/dL (diabetic range) or in the prediabetic range, an increased risk of conditions such as preterm birth, delivery of LGA baby and macrosomia is observed.
In PGDM, particularly when FPG levels are elevated, the increased risks of preterm birth, delivery of LGA baby and macrosomia can be attributed to complex and multifactorial causes. First, insulin function is compromised in the presence of DM.27 Prolonged exposure to high blood glucose further impairs insulin secretion28 leading to an increased risk of both microvascular and macrovascular complications.29 This, in turn, can increase the risk of developing hypertensive disorders during pregnancy and preterm birth. Second, continual impaired fasting glucose or hyperglycemia in the placenta can result in increased glucose transfer to the fetus, triggering the expansion of islet beta cells in the fetal pancreas.30 This, in turn, can lead to elevated levels of insulin in the fetus, a well-established factor in the development of large-for-gestational age or macrosomic infants. Third, women with hyperglycemia frequently exhibit elevated levels of insulin-like growth factor I (IGF-I). This hormone stimulates the transfer of nutrients across the placenta, thereby boosting fetal growth. Elevated IGF-I levels are associated with an increased likelihood of producing largely for gestational-age infants.31 Furthermore, this excessive fetal growth typically results in disproportionate abdominal enlargement, with head-to-abdominal circumference ratios < 0.95 being associated with neonatal hypoglycemia in PGDM pregnancies.32
According to a previous meta-analysis, the risk of neonatal respiratory distress syndrome is increased in infants born to women with PGDM. The pooled OR for the association between PGDM and the risk of neonatal respiratory distress syndrome was 2.66 (95% CI, 2.06–3.44; I2 = 73.7%; heterogeneity = 0.010).33 In our study, similar to these results, infants born to women with FPG levels in the diabetic range showed an increased risk of transient tachypnea of newborn when compared to infants born to women with FPG levels < 85 (aOR for transient tachypnea of newborn, 1.96; 95% CI, 1.27–3.03). Additionally, the risk of respiratory distress syndrome was also elevated, with an aOR of 1.44 (95% CI, 1.02–2.04) for infants born to women with diabetes-range FPG levels. A possible mechanism explaining the connection between maternal DM and the heightened risk of neonatal respiratory distress syndrome is linked to the integrity and composition of the fetal pulmonary surfactant. DM is associated with the delayed secretion of phosphatidylglycerol, which is a crucial lipid component of surfactants.34 Furthermore, insulin inhibits the gene expression of surfactant proteins A and B in lung epithelial cells. These proteins are typically elevated in newborns exposed to hyperglycemia during pregnancy.35
Another important consideration is the potential genetic effects of maternal DM. However, the genetic factors associated with diabetes are not fully understood. Genome-wide association studies have identified more than 40 loci that increase an individual's risk of developing type 2 diabetes36 and over 60 loci that contribute to susceptibility to type 1 diabetes.37 However, the combined effects of these genetic factors account for only approximately 5–10% of the heritability of diabetes. This suggests that the inheritance of diabetes susceptibility is complex, polygenic, and involves epigenetic modifications as well as nutritional and environmental factors. Diabetes during pregnancy can affect the developing fetus, neonates, and children. The mechanisms underlying the effect of maternal hyperglycemia on fetal development include heightened oxidative stress, hypoxia, apoptosis, and epigenetic alterations. Evidence supports these epigenetic changes, including the fact that not all offspring are equally affected, and the degree of impact can vary. The maternal diet can also play a role in influencing pregnancy outcomes. Additionally, maternal diabetes can lead to changes in embryonic transcriptional profiles, resulting in increased variability between transcriptomic profiles owing to alterations in gene regulation.3839
Considering the aforementioned mechanisms, it is conceivable that poor blood glucose control in diabetes could have implications for pregnancy outcomes and offspring health. However, previous research has often focused on euglycemia and hyperglycemia or examined the relationship between blood glucose levels during pregnancy and pregnancy outcomes. In this study, we categorized pregestational blood glucose levels in women with diabetes into various groups and assessed the associated risks for pregnancy outcomes and neonatal morbidity. This study provides clinical evidence for the benefits of strict blood glucose control before pregnancy, thereby offering valuable insights.
Although FPG provides valuable information and is an important indicator of its correlation with outcomes, there may be cases in which only FPG levels are low, unlike overall blood glucose control status. For instance, patients with type 1 diabetes often have low FPG levels, depending on their medication regimen. In addition, nonobese patients tend to have lower FPG levels. Considering these factors, additional research utilizing data on glycated hemoglobin or postprandial blood glucose levels is needed.
Our analysis revealed an unexpected U-shaped relationship between maternal FPG levels and certain neonatal outcomes, particularly for transient tachypnea and respiratory complications at FPG < 85 mg/dL. While the adverse effects of hyperglycemia are well-established, the interpretation of outcomes associated with lower FPG levels requires careful consideration. Our methodological approach in selecting FPG < 85 mg/dL as a reference group was inspired by the landmark HAPO study, which demonstrated the importance of examining glycemic effects across a continuous spectrum. Similar to the HAPO study, where the lowest glycemic category served as a reference point to demonstrate the progressive relationship between increasing glucose levels and adverse pregnancy outcomes, we aimed to examine this continuous relationship and identify potential threshold effects in women with PGDM. However, it is crucial to emphasize that this reference point was selected purely for statistical comparison purposes and should not be interpreted as suggesting that maintaining FPG below 85 mg/dL would lead to better pregnancy outcomes in women with diabetes. The observed U-shaped pattern might reflect the complex interplay between maternal glucose regulation and fetal development, where both extremely low and high glucose levels could potentially impact outcomes. However, several factors warrant consideration in interpreting these findings: the presence of underlying maternal conditions requiring strict glucose control, potential unmeasured confounders, and the relatively wider CIs observed in the lower FPG range due to smaller sample sizes. The optimal glycemic targets for pregnant women with PGDM should be determined based on individual clinical assessment, taking into account the risks of both hyperglycemia and hypoglycemia, and following established clinical guidelines for diabetes management during pregnancy. These observations underscore the importance of balanced glucose management during pregnancy and highlight the need for further research to better understand the implications of lower maternal glucose levels on neonatal outcomes.
This study represents the first investigation of the impact of pregestational FPG levels on pregnancy outcomes and neonatal morbidity by categorizing FPG levels into various ranges. Considering that women with diabetes constitute approximately 1–2% of the total pregnant population, studying the effects of FPG levels as designed in this study requires a large-scale population, which was made possible by utilizing national databases. By grouping FPG into different ranges and examining the relative ORs, this study provides valuable clinical insights into the effects of glycemic control.
However, this study has several limitations. First, owing to limitations in the available information within the national database, data on factors such as socioeconomic status, alcohol use, and glycated hemoglobin levels and specific types of treatments and degree of glycemic control during pregnancy which can affect adverse pregnancy outcomes and neonatal morbidity, were not available. In addition, information on intrapartum events that may affect the fetus during pregnancy is unavailable. Third, incorrect or insufficient diagnostic code input may have led to the underestimation of events. Furthermore, the difficulty in classifying subtypes of DM made it challenging to distinguish between Type 1 and Type 2 diabetes, and we were unable to obtain information regarding the use of medication, which represents a limitation. Fourth, owing to the retrospective nature of this study, well-designed prospective studies are needed to address these limitations.
In this study, elevated prepregnancy FPG levels were associated with higher frequencies of adverse pregnancy outcomes, including preterm birth, macrosomia, and delivery of LGA babies. We also observed that neonates born to women with diabetes-range FPG had higher rates of respiratory complications and neonatal hypoglycemia. While these associations suggest the potential importance of preconception glucose control, prospective longitudinal studies incorporating both prepregnancy and gestational glycemic status are needed to establish the causal relationship between glycemic control and pregnancy outcomes in women with diabetes.
XML Download