Abstract
Objectives
This study examined whether early postpartum insulin response patterns (< 6 months) following a glucose load predict the emergence of abnormal glucose metabolism (AGM) in women who were previously diagnosed with gestational diabetes mellitus (GDM).
Methods
We conducted a prospective cohort study involving 308 women who had prior GDM. A 75-gram oral glucose tolerance test was administered to each participant within six months after delivery, and the participant’s insulin levels were measured at baseline, 30 min, and 2 h following glucose administration. Participants were categorized as having delayed insulin secretion (In2h ≥ In30min; n = 196) or a normal insulin response (In2h < In30min; n = 112). Over an average follow-up period of 32.7 months, Cox proportional hazards models were employed to assess adjusted hazard ratios for AGM development.
Results
Among the 308 women, 153 (49.7%) developed AGM. The delayed insulin secretion group faced a significantly elevated risk (adjusted HR 1.739 [95% CI 1.141–2.649], P = 0.010). Other significant risk factors included elevated fasting glucose levels during pregnancy (adjusted HR 1.565 [95% CI 1.152–2.125], P = 0.004) and early postpartum hypertriglyceridemia (adjusted HR 1.448 [95% CI 1.029–2.039], P = 0.034).
Conclusions
A sluggish insulin response during the early postpartum phase is a robust predictor of future AGM in women with prior GDM.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12884-026-08682-5.
Keywords: Gestational diabetes mellitus, Β-cell function, Delayed insulin secretion, Postpartum abnormal glucose metabolism
Introduction
Gestational diabetes mellitus (GDM) is a significant metabolic disorder characterized by glucose intolerance that first manifests during pregnancy [1]. Affecting approximately one in seven pregnancies worldwide, GDM presents not only immediate risks but also lasting health implications for both mothers and offsprings [2]. Women who have had GDM experience an approximately ten times higher risk of transitioning to type 2 diabetes (T2DM) in the postpartum period than their counterparts without such a history [3]. A landmark multinational observational study involving over 10,000 participants revealed striking longitudinal data. The GDM population had diabetes and prediabetes at rates of nearly 29 cases and approximately 150 cases per 1,000 people per year, respectively. In the control group, the rates were a mere 1.73 cases per 1,000 people for diabetes and 49.3 cases per 1,000 people for prediabetes [4]. These findings underscore the urgent need to implement effective postpartum surveillance and prevention strategies for this vulnerable population.
The pathophysiological overlap between GDM and T2DM is well established, with both conditions arising from a combination of β-cell dysfunction—manifested as reduced insulin production—and the resistance to insulin in peripheral tissues, reflecting an impaired biological response to insulin. These dual defects contribute to a progressive decline in glucose metabolism, although the relative impact of each factor may vary depending on the stage of the disease [5, 6]. Emerging evidence from longitudinal metabolic studies suggests that the early postpartum period represents a critical window for diabetes risk stratification because persistent β-cell impairment during this time strongly predicts future metabolic disorders in women with prior GDM [7–9]. Among the various metrics used to assess β-cell function, indices such as the homeostasis model assessment of β-cell function (HOMA-β), the insulinogenic index (IGI), and the insulin secretion-sensitivity index (ISSI) are commonly employed in both large-scale clinical studies and the postpartum monitoring of women with GDM [9–11]. Among these indices, early-phase insulin secretion has been widely recognized as a sensitive and physiologically relevant marker of β-cell integrity. Impairment of first-phase insulin secretion represents a well-established pathological event that occurs in both T1DM and T2DM [12, 13]. However, researches that specifically investigate the dynamics and magnitude of insulin secretion in the early postpartum period among women with prior GDM remain scarce. We designed the present prospective cohort study to systematically investigate the relationship between early postpartum insulin secretion dynamics—specifically the timing and magnitude of insulin responses—and the development of subsequent glucose metabolism disorders in women with prior GDM. By integrating detailed assessments of insulin dynamics with a broad array of gestational and postpartum risk factors, this study aimed to achieve a more comprehensive characterization of postpartum β-cell function, and to assess its predictive value for future metabolic outcomes.
Methods and materials
Study design and participants
In this study, we conducted a prospective cohort investigation involving women diagnosed with GDM who received postpartum care at the First Affiliated Hospital of Sun Yat-sen University between January 1, 2017, and December 31, 2021. Eligible participants were required to complete a 75-gram oral glucose tolerance test (OGTT) within 6 months postpartum, with blood samples collected for insulin analyses at fasting, 30 min, and 2 h after glucose administration. Additionally, participants were required to complete at least one follow-up OGTT after 6 months postpartum, with samples collected at all time points (fasting, 30 min, and 2 h). Follow-up OGTTs were performed during routine postpartum follow-up visits beyond 6 months postpartum and were not scheduled at a fixed interval for all participants. The exclusion criteria were as follows: (1) preexisting diabetes mellitus before pregnancy, (2) multiple pregnancies, (3) a history of organ transplantation requiring long-term hormonal therapy or immunosuppressants, (4) incomplete clinical data or absence of postpartum follow-up, (5) active or untreated malignancies, and (6) a history of psychiatric disorders that prevented patients from complying with follow-up requirements. Ultimately, our analysis incorporated data from 308 women (Fig. 1). The primary endpoint was the occurrence of AGM, which was assessed on the basis of follow-up OGTT results beyond 6 months postpartum.
Fig. 1.
Flowchart of participants inclusion and follow-up in the study
Definitions
GDM was diagnosed on the basis of a 75-g oral glucose tolerance test (OGTT) using the criteria proposed by the International Association of the Diabetes and Pregnancy Study Groups: fasting plasma glucose (FPG) level ≥ 5.1 mmol/L, 1-h plasma glucose (1hPG) level ≥ 10.0 mmol/L, and 2-h plasma glucose (2hPG) level ≥ 8.5 mmol/L [14].
AGM was defined in accordance with the World Health Organization (WHO) standards: (1) diabetes, indicated by a FPG level ≥ 7.0 mmol/L or a 2hPG level ≥ 11.1 mmol/L; (2) impaired fasting glucose (IFG), defined as a FPG level between 6.1 mmol/L and 7.0 mmol/L with 2hPG level < 7.8 mmol/L; and (3) impaired glucose tolerance (IGT), characterized by a FPG level < 7.0 mmol/L with 2hPG level between 7.8 mmol/L and 11.1 mmol/L. β-cell function was assessed using the homeostasis model assessment of β-cell function (HOMA-β) and the insulinogenic index (IGI). HOMA-β was calculated as fasting plasma insulin multiplied by 20 and divided by the difference between fasting plasma glucose (FPG) and 3.5, whereas the IGI was determined as the ratio of the increment in insulin concentration to the increment in glucose concentration during the first 30 min of the OGTT. Insulin resistance was measured using the HOMA-IR, computed by multiplying fasting plasma insulin level (mU/L) by FPG (mmol/L) and dividing by 22.5 [15, 16]. Insulin secretory function was evaluated by comparing the serum insulin concentrations at 30 min (In30min) and 2 h (In2h) during the 75-g OGTT. Based on these measurements within 6 months postpartum, participants were classified into two groups: (1) the delayed insulin secretion group (In2h ≥ In30min, n = 196) and (2) the normal insulin secretion group (In2h < In30min, n = 112) [12, 17, 18].
Data collection
Demographic and clinical information at baseline—including age, BMI, gestational weight gain, delivery mode, family history of diabetes, and prior GDM—was extracted from medical records. Laboratory test results during pregnancy, including glucose values at GDM diagnosis, hemoglobin A1c (HbA1c) levels, and postpartum test results, including lipid profiles, HbA1c levels, and OGTT-derived glucose and insulin measurements, were also collected. All blood samples were collected in the early morning following an overnight fast of at least 8 h and were immediately transferred to the hospital laboratory for processing. Plasma glucose and insulin concentrations were measured by standard enzymatic methods and chemiluminescent immunoassay, respectively. HbA1c was measured using high-performance liquid chromatography. The samples were handled, stored, and analyzed according to standardized laboratory protocols to ensure data accuracy and reliability.
Statistical analysis
IBM SPSS Statistics (Version 25.0) was used for data analysis. A two-sided p value below 0.05 will be determined statistically significant. Continuous variables following a normal distribution are reported as the means ± standard deviations (SDs) and were compared via independent t tests. Nonparametric data, presented as medians and interquartile ranges (IQRs), were analyzed with the Mann‒Whitney U test. Categorical variables are summarized as percentages and were assessed via chi-square tests. The maximum proportion of missing data was 8.7% (for HbA1c level within 6 months postpartum). Under the assumption that data were missing at random, multiple imputation techniques were applied to address missing values. Although missing at random is considered a reasonable assumption, it cannot be statistically verified [19]. Ten estimation datasets, each generated with 50 iterations, were created using predefined predictive variables and combined according to Rubin rules [20]. The risk of AGM beyond 6 months postpartum was evaluated using Kaplan‒Meier curves. The log-rank test was employed to compare survival distributions across groups. Associations between insulin secretion patterns and AGM were assessed using Cox proportional hazards models with adjustment for potential confounding factors, with hazard ratios (HRs) and 95% confidence intervals (CIs) reported for both imputed and complete-case datasets.
Results
Baseline characteristics
The average and median follow-up durations were 32.7 months and 36.0 months postpartum, respectively. Women in the In2h < In30min group had a mean follow-up of 33.8 ± 1.78 months, whereas those in the In2h ≥ In30min group had a mean follow-up of 32.4 ± 1.87 months (p = 0.545). Table 1 shows that women in the In2h ≥ In30min group presented higher 1hPG levels during pregnancy, higher 2hPG levels, and lower IGIs within 6 months postpartum than did those in the In2h < In30min group. Among the 308 women, 29.9% maintained prediabetes at the first OGTT, and the prevalence of prediabetes was 27.7% in In2h < In30min group and 31.1% in In2h ≥ In30min group (p = 0.525). No statistically significant differences were found between groups for other variables, such as age, BMI, gestational weight gain, family history of diabetes, delivery mode, FPG level, 2hPG level, and hemoglobin A1c level during pregnancy, or postpartum metabolic markers, including HOMA-β, HOMA-IR, IGI, HBA1C level, or lipid profiles, within 6 months postpartum.
Table 1.
Baseline characteristics of participants stratified by insulin levels within 6 months postpartum
| Characteristic | In2h < In30min (n = 112) | In2h ≥ In30min (n = 196) | P value |
|---|---|---|---|
| Maternal age (year) | 35 (32–38) | 33 (30–38) | 0.075 |
| ≥ 35 years, n (%) | 60 (53.6) | 85 (43.4) | 0.084 |
| Height (cm) | 158 (156–161) | 159 (156–162) | 0.157 |
| Pre-pregnancy weight (kg) | 54.5 (50.3–60.0) | 54.0 (48.0–60.0) | 0.186 |
| Pre-pregnancy BMI (kg/m2) | 21.56 (20.25–24.05) | 21.41 (19.23–23.33) | 0.081 |
| Weight at delivery (kg) | 65.2 (59.1–71.3) | 64.9 (58.5–68.7) | 0.105 |
| Gestational weight gain (kg) | 10.0 (7.5–12.1) | 10.0 (7.4–12.5) | 0.786 |
| Family history of diabetes, n (%) | 31 (27.7) | 64 (32.7) | 0.363 |
| Previous GDM, n (%) | 14 (12.5) | 30 (15.3) | 0.498 |
| Gestational age (week) | 38 (38–39) | 38 (38–39) | 0.793 |
| Mode of delivery, n (%) | 0.801 | ||
| Cesarean delivery | 53 (47.3) | 97 (49.5) | |
| Forceps delivery | 6 (5.4) | 13 (6.6) | |
| Spontaneous labor | 53 (47.3) | 86 (43.9) | |
| During pregnancy | |||
| OGTT: FPG (mmol/L) | 4.5 (4.3–4.8) | 4.5 (4.2–4.8) | 0.955 |
| OGTT: 1hPG (mmol/L) | 10.0 (9.1–10.6) | 10.2 (9.3–10.9) | 0.047 |
| OGTT: 2hPG (mmol/L) | 9.0 (8.6–9.7) | 9.2 (8.6–10.0) | 0.112 |
| HbA1c (%) | 5.1 (4.9–5.3) | 5.1 (4.9–5.4) | 0.079 |
| Within 6 months postpartum | |||
| OGTT: FPG (mmol/L) | 4.8 (4.3–5.1) | 4.6 (4.4-5.0) | 0.697 |
| OGTT: 30minPG (mmol/L) | 9.3 (8.3–10.7) | 9.2 (8.3–10.2) | 0.430 |
| OGTT: 2hPG (mmol/L) | 7.3 (6.4-8.0) | 7.4 (6.8–8.3) | 0.048 |
| OGTT: fasting insulin (µU/mL) | 4.77 (3.48–7.56) | 4.73 (3.15–6.74) | 0.500 |
| OGTT: Insulin at 30 min (µU/mL) | 49.72 (35.39–75.63) | 29.62 (21.62–43.21) | < 0.001 |
| OGTT: Insulin at 2 h (µU/mL) | 32.36 (22.68–43.13) | 56.77 (35.07–73.61) | < 0.001 |
| HOMA-β | 89.1 (62.0-122.0) | 80.5 (53.5-127.3) | 0.315 |
| HOMA-IR | 0.99 (0.70–1.72) | 0.99 (0.63–1.50) | 0.540 |
| IGI | 10.17 (6.75–15.10) | 5.97 (3.73–9.10) | < 0.001 |
| HbA1c (mmol/L) | 5.5 (5.1–5.7) | 5.4 (5.2–5.6) | 0.594 |
| TC (mmol/L) | 5.5 (4.8–6.3) | 5.3 (4.7-6.0) | 0.192 |
| TG (mmol/L) | 0.82 (0.66–1.21) | 0.83 (0.62–1.18) | 0.526 |
| HDL (mmol/L) | 1.63 (1.37–1.83) | 1.56 (1.32–1.84) | 0.424 |
| LDL (mmol/L) | 3.26 (2.83–3.73) | 3.14 (2.73–3.65) | 0.300 |
BMI body mass index, OGTT oral glucose tolerance test, FPG fasting plasma glucose, PG postprandial glucose, HbA1c hemoglobin A1c, TC total cholesterol, TG triglycerides, HDL high-density lipoprotein, LDL low-density lipoprotein
Values are presented as mean ± standard deviation (SD), median (interquartile range [IQR]) or number (percentage), as appropriate
Associations between insulin secretion patterns and the risk of future AGM
Among the 308 women, 153 (49.7%) developed AGM beyond six months postpartum. Among these women, 3 women (2.68%) in the In2h < In30min group and 12 women (6.12%) in the In2h ≥ In30min group were diagnosed with T2DM. Additionally, 45 women (40.16%) in the In2h < In30min group and 93 women (47.45%) in the In2h ≥ In30min group were diagnosed with prediabetes. The In2h ≥ In30min group had a greater incidence of AGM (53.6%) than the In2h < In30 min group (42.8%) (Fig. 2). Kaplan‒Meier analysis demonstrated a significantly elevated cumulative AGM risk in In2h ≥ In30min group (Fig. 3).
Fig. 2.
Postpartum glucose metabolism status beyond 6 months in women stratified by insulin secretion timing. NGT. normal glucose tolerance
Fig. 3.
Kaplan–Meier curves illustrating the cumulative incidence of postpartum abnormal glucose metabolism
Risk factors for the development of AGM beyond 6 months postpartum
To determine risk factors associated with the development of AGM beyond 6 months postpartum, we conducted Cox proportional hazards regression analyses on both the imputed and complete-case datasets (Table 2). Univariate analyses revealed several significant associations: In2h ≥ In30min status; FPG, 1hPG, and HbA1c levels during pregnancy; and triglyceride levels within 6 months postpartum were significantly associated with a higher risk of developing postpartum AGM. Notably, women in the In2h ≥ In30min group had a significantly greater hazard ratio (HR = 1.690 [95% CI 1.198–2.385], P = 0.003) than did those in the In2h < In30min group. In contrast, the HOMA-β, HOMA-IR, and IGI showed no significant association with postpartum AGM. Comparable results were observed in the analysis of the available datasets.
Table 2.
Univariate analysis of risk factors for postpartum abnormal glucose metabolism in imputed and available datasets
| Variables | Available dataset (n = 234) | Imputation dataset (n = 308) | ||
|---|---|---|---|---|
| HR [95% CI] | P value | HR [95% CI] | P value | |
| Maternal age (year) | 0.983 (0.944–1.024) | 0.404 | 0.983 (0.948–1.019) | 0.348 |
| ≥ 35 years | 0.857 (0.597–1.231) | 0.403 | 0.840 (0.611–1.155) | 0.283 |
| Height (cm) | 0.999 (0.960–1.039) | 0.952 | 1.002 (0.969–1.036) | 0.906 |
| Pre-pregnancy weight (kg) | 0.998 (0.977–1.019) | 0.861 | 1.002 (0.983–1.022) | 0.829 |
| Pre-pregnancy BMI (kg/m2) | 0.995 (0.935–1.058) | 0.872 | 1.005 (0.951–1.062) | 0.865 |
| Weight delivery (kg) | 0.996 (0.975–1.018) | 0.716 | 0.999 (0.979–1.018) | 0.883 |
| Gestational weight gain (kg) | 0.993 (0.954–1.033) | 0.721 | 0.988 (0.953–1.024) | 0.953 |
| Family history of diabetes | 1.119 (0.761–1.643) | 0.568 | 1.315 (0.942–1.836) | 0.107 |
| Previous GDM | 1.004 (0.600–1.680) | 0.988 | 1.103 (0.718–1.694) | 0.653 |
| Gestational age (weeks) | 1.031 (0.886-1.200) | 0.694 | 1.060 (0.946–1.188) | 0.314 |
| Cesarean delivery | 0.765 (0.532-1.100) | 0.149 | 0.863 (0.628–1.187) | 0.365 |
| During pregnancy | ||||
| OGTT: FPG (mmol/L) | 1.554 (1.234–1.956) | < 0.001 | 1.643 (1.337–2.020) | < 0.001 |
| OGTT: 1hPG (mmol/L) | 1.157 (1.011–1.324) | 0.035 | 1.176 (1.041–1.329) | 0.009 |
| OGTT: 2hPG (mmol/L) | 1.078 (0.950–1.223) | 0.243 | 1.125 (0.999–1.265) | 0.051 |
| HbA1c (%) | 1.850 (1.250–2.738) | 0.002 | 1.917 (1.346–2.731) | < 0.001 |
| Within 6 months postpartum | ||||
| Insulin secretion | ||||
| In2h < In30min | Reference | Reference | ||
| In2h ≥ In30min | 1.643 (1.124–2.402) | 0.010 | 1.690 (1.198–2.385) | 0.003 |
| HOMA-β | 0.999 (0.996–1.002) | 0.591 | 1.000 (0.998–1.002) | 0.990 |
| IGI | 0.988 (0.970–1.006) | 0.180 | 0.986 (0.968–1.004) | 0.132 |
| HOMA-IR | 1.029 (0.883–1.199) | 0.711 | 1.080 (0.946–1.232) | 0.255 |
| HbA1c (%) | 1.495 (0.921–2.426) | 0.104 | 1.451 (0.927–2.272) | 0.103 |
| TC (mmol/L) | 1.091 (0.914–1.304) | 0.335 | 1.033 (0.878–1.215) | 0.669 |
| TG (mmol/L) | 1.327 (1.012–1.741) | 0.041 | 1.390 (1.081–1.787) | 0.010 |
| HDL (mmol/L) | 0.921 (0.541–1.565) | 0.760 | 0.896 (0.554–1.450) | 0.655 |
| LDL (mmol/L) | 1.125 (0.879–1.439) | 0.350 | 1.064 (0.854–1.325) | 0.581 |
HR hazard ratio, CI confidential interval, BMI body mass index, OGTT oral glucose tolerance test, FPG fasting plasma glucose, PG postprandial glucose, HbA1c glycated hemoglobin, TC total cholesterol, TG triglycerides, HDL high-density lipoprotein, LDL low-density lipoprotein
Multivariate Cox regression analyses based on the imputed dataset demonstrated that In2h ≥ In30min status (HR = 1.771 [95% CI 1.221–2.570], P = 0.003), FPG levels during pregnancy (HR = 1.524 [95% CI 1.159–2.005], P = 0.003), and TG levels within 6 months postpartum (HR = 1.333 [95% CI 1.016–1.749], P = 0.038) were independent risk factors. After adjusting for confounding factors, including maternal age, pregestational BMI, family history of T2DM, prior history of GDM, gestational weight gain, mode of delivery, postpartum IGI, HOMA-IR, HOMA-β and lipid levels, these associations remained statistically significant: In2h ≥ In30min status (adjusted HR 1.739 [95% CI 1.141–2.649], P = 0.010), FPG levels during pregnancy (adjusted HR 1.565 [95% CI 1.152–2.125], P = 0.004), and TG levels within 6 months postpartum (adjusted HR 1.448 [95% CI 1.029–2.039], P = 0.034). In the available dataset, triglyceride levels within six months postpartum were no longer a significant factor associated with long-term AGM risk (Table 3).
Table 3.
Multivariate analysis of risk factors for postpartum abnormal glucose metabolism in imputed and available datasets
| Variables | Available dataset (n = 234) | Imputation dataset (n = 308) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR [95% CI] | P value | Adjusted HR [95% CI] a |
P value | HR [95% CI] | P value | Adjusted HR [95% CI] a |
P value | |||||||||
| Insulin secretion | ||||||||||||||||
| In2h < In30min | Reference | |||||||||||||||
| In2h ≥ In30min | 1.831 (1.232–2.722) | 0.003 | 1.713(1.110–2.645) | 0.015 | 1.771 (1.221–2.570) | 0.003 | 1.739 (1.141–2.649) | 0.010 | ||||||||
| FPG during pregnancy (mmol/L) | 1.407 (1.051–1.883) | 0.022 | 1.501 (1.106–2.035) | 0.009 | 1.524 (1.159–2.005) | 0.003 | 1.565 (1.152–2.125) | 0.004 | ||||||||
| 1hPG during pregnancy (mmol/L) | 1.084 (0.916–1.283) | 0.345 | 1.074 (0.899–1.282) | 0.434 | 1.027 (0.878–1.201) | 0.738 | 1.067 (0.905–1.258) | 0.441 | ||||||||
| 2hPG during pregnancy (mmol/L) | 1.016 (0.890–1.159) | 0.818 | 1.018 (0.882–1.174) | 0.807 | 1.050 (0.922–1.196) | 0.461 | 1.039 (0.901–1.198) | 0.596 | ||||||||
| HbA1c during pregnancy (%) | 1.406 (0.840–2.355) | 0.195 | 1.598 (0.910–2.808) | 0.103 | 1.306 (0.823–2.071) | 0.257 | 1.419 (0.847–2.379) | 0.184 | ||||||||
| TG within 6 months postpartum (mmol/L) | 1.319 (0.991–1.757) | 0.058 | 1.373 (0.965–1.954) | 0.078 | 1.333 (1.016–1.749) | 0.038 | 1.448 (1.029–2.039) | 0.034 | ||||||||
a, Adjustments were made for maternal age, pregestational BMI, family history of T2DM, prior history of GDM, gestational weight gain, mode of delivery, postpartum IGI, HOMA-IR, HOMA-β, and lipid levels within 6 months postpartum. FPG fasting plasma glucose, PG postprandial glucose, HbA1c glycated hemoglobin, TG triglycerides
Discussion
This study provides valuable insights into the intricate association between early postpartum insulin secretion dynamics and long-term metabolic outcomes in women with prior GDM. Our data demonstrate that women with delayed insulin responses postpartum face a markedly greater likelihood of progressing to diabetes or prediabetes than women with normal insulin kinetics. In addition, elevated fasting plasma glucose (FPG) levels during pregnancy and increased triglyceride (TG) concentrations within six months postpartum were also identified as significant determinants of long-term postpartum AGM. Impairment of first-phase insulin secretion is a preliminary signal of pancreatic dysfunction—a key player in the transition from GDM to T2DM [21] that reflects a decrease in the ability of β-cells to release insulin promptly when faced with rising glucose levels. This observation is particularly relevant in Asian populations, where β-cell dysfunction is often the predominant driver of dysglycemia progression, overshadowing the role of obesity-induced insulin resistance [7, 22, 23]. Previous studies have shown that in nondiabetic populations, a delayed insulin peak after glucose challenge is linked to increased diabetes risk compared with insulin concentrations peaking at 30–60 min [17, 18]. In this study, the predictive performances of traditional measures of β-cell function, such as the IGI and HOMA-β, were inferior to that of the In2h ≥ In30min status marker. This difference may be attributed to the inherent limitation of steady-state models, which rely solely on fasting measures, in capturing β-cell function under stimulated conditions [24]. Thus, these models may underestimate the extent of dysfunction in people with impaired glucose tolerance [21]. Notably, β-cell defects in the early stages of glucose metabolism disorders are often only detectable in response to glucose loading [25]. Most prior investigations have focused predominantly on fasting indicators, such as fasting plasma glucose and fasting insulin levels, to evaluate insulin sensitivity and β-cell function [9, 10]. However, such measures may fail to distinguish variations in insulin secretory capacity under stimulated conditions. Our findings underscore the limitations of assessing islet function using fasting-state models alone in postpartum women with prior GDM. Although In2h ≥ In30min status has not yet been universally accepted as a definitive marker and may not precisely capture the actual peak time of insulin secretion following glucose intake, it offers a simple, practical, and novel metric for assessing delayed insulin response in the early postpartum period. As anticipated, this index also demonstrated substantial value for its association with future AGM risk.
We further observed that high triglyceride levels significantly predict AGM, independent of other factors, which aligns with the findings of previous studies [26–28]. The connection between triglyceride levels and diabetes risk appears particularly strong in women [29]. Similar findings were reported in an Australian study that included 150 GDM survivors and 72 women with normal glucose tolerance [30]. Elevated TG levels may promote insulin resistance and impair insulin sensitivity. A potential underlying mechanism is that excess TG accumulation increases the free fatty acids in circulation, which in turn may promote the generation of lipotoxic metabolites that impair insulin signaling and β-cell function [31–33].
Following adjustments for potential confounding factors, FPG levels during pregnancy remained a key predictor, irrespective of data imputation. Several high-quality investigations have demonstrated that, compared with the 2-hour plasma glucose level following an OGTT, the FPG level during pregnancy serves as a more robust indicator for predicting postpartum disturbances in glucose metabolism [30, 34–37]. A prior study reported that the risk of postpartum diabetes was elevated 1.61 times for each 1 mmol/L increase in the FPG level [38]. Other researchers have similarly identified the FPG level as a key predictor of T2DM, particularly among women with FPG levels of 5.6 mmol/L at diagnosis, who presented a markedly greater risk than those with baseline values of 4.7 mmol/L [39]. Emerging evidence also indicates that women with impaired fasting glycemia exhibit more pronounced β-cell dysfunction and a chronic reduction in β-cell mass [40, 41]. These underlying pathophysiological changes may contribute to their increased susceptibility to the development of T2DM.
Despite enhancing risk stratification in this population, several limitations should be acknowledged. Notably, the single-center design and relatively small sample size may have introduced selection bias, which limits the generalizability of the findings. Additionally, participants who returned for follow-up may systematically differ from those lost to follow-up. Second, we lacked insulin measurements at 1 and 3 h during the postpartum OGTT. This limitation hindered our ability to accurately characterize the full trajectory of insulin secretion following glucose intake. Third, the duration of follow-up was confined to the first 3 years postpartum, which may not fully capture the true long-term incidence of AGM after GDM gestation. In addition, the predictive accuracy of HbA1c is limited during pregnancy as well as in the early postpartum period, largely due to physiological changes affecting hemoglobin metabolism. Future studies involving larger, multicenter cohorts and more comprehensive metabolic profiling are warranted to confirm the predictive value of delayed insulin secretion and to refine risk stratification strategies in this population.
In summary, this study revealed that delayed insulin secretion upon glucose stimulation in the early postpartum period, as well as the FPG levels during pregnancy and TG levels within early postpartum period, are significantly associated with subsequent metabolic dysregulation in women with prior GDM.
Supplementary Information
Authors’ contributions
Shuhui Liang and Chenxue Wang contributed to the investigation, data analysis and interpretation, and the preparation of the original draft. Fenghua Lai was responsible for critical review and revision of the manuscript. Changliu Xu, Nan Chen, Xuejie Wang were responsible for investigation, data collection and validation. Haipeng Xiao and Yanbing Li provided critical revision of the manuscript. Xiaopei Cao contributed to conceptualization, funding acquisition and critical revision of the manuscript.
Funding
This study was supported by grants from the 5010 Project Foundation of Sun Yat-sen University (Grant number: 2017001) and the Noncommunicable Chronic Diseases-National Science and Technology Major Project (2024ZD0532100).
Data availability
The datasets analyzed in this study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was approved by the Institutional Review Board of The First Affiliated Hospital of Sun Yat-sen University (No.[2015]101) with informed consent waiver and was conducted in compliance with the Declaration of Helsinki.
Consent for publication
This section is not applicable for this paper.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Shuhui Liang and Chenxue Wang contributed equally to this work.
References
- 1.Jovanovic L, Pettitt DJ. Gestational diabetes mellitus. JAMA. 2001;286(20):2516–8. [DOI] [PubMed] [Google Scholar]
- 2.Sweeting A, Hannah W, Backman H, Catalano P, Feghali M, Herman WH et al. Epidemiology and management of gestational diabetes. Lancet. 2024;404(10448):175–92. [DOI] [PubMed] [Google Scholar]
- 3.Vounzoulaki E, Khunti K, Abner SC, Tan BK, Davies MJ, Gillies CL. Progression to type 2 diabetes in women with a known history of gestational diabetes: systematic review and meta-analysis. BMJ. 2020;369:m1361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shen Y, Wang P, Wang L, Zhang S, Liu H, Li W, et al. Gestational diabetes with diabetes and prediabetes risks: a large observational study. Eur J Endocrinol. 2018;179(1):51–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Homko C, Sivan E, Chen X, Reece EA, Boden G. Insulin secretion during and after pregnancy in patients with gestational diabetes mellitus. J Clin Endocrinol Metab. 2001;86(2):568–73. [DOI] [PubMed] [Google Scholar]
- 6.McIntyre HD, Catalano P, Zhang C, Desoye G, Mathiesen ER, Damm P. Gestational diabetes mellitus. Nat Rev Dis Primers. 2019;5(1):47. [DOI] [PubMed] [Google Scholar]
- 7.Li W, Zhang S, Liu H, Wang L, Zhang C, Leng J, et al. Different associations of diabetes with β-cell dysfunction and insulin resistance among obese and Nonobese Chinese women with prior gestational diabetes mellitus. Diabetes Care. 2014;37(9):2533–9. [DOI] [PubMed] [Google Scholar]
- 8.Retnakaran R, Qi Y, Sermer M, Connelly PW, Hanley AJG, Zinman B. Beta-cell function declines within the first year postpartum in women with recent glucose intolerance in pregnancy. Diabetes Care. 2010;33(8):1798–804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Retnakaran R, Ye C, Kramer CK, Hanley AJ, Connelly PW, Sermer M, et al. Deteriorating beta cell function is the dominant determinant of progression from normal glucose tolerance to prediabetes/diabetes in young women following pregnancy. Diabetologia. 2023;66(11):2154–63. [DOI] [PubMed] [Google Scholar]
- 10.Miao Z, Wu H, Ren L, Bu N, Jiang L, Yang H, et al. Long-Term postpartum outcomes of insulin resistance and β-cell function in women with previous gestational diabetes mellitus. Int J Endocrinol. 2020;2020:7417356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kramer CK, Swaminathan B, Hanley AJ, Connelly PW, Sermer M, Zinman B, et al. Each degree of glucose intolerance in pregnancy predicts distinct trajectories of β-cell function, insulin sensitivity, and glycemia in the first 3 years postpartum. Diabetes Care. 2014;37(12):3262–9. [DOI] [PubMed] [Google Scholar]
- 12.Del Prato S. Loss of early insulin secretion leads to postprandial hyperglycaemia. Diabetologia. 2003;46(Suppl 1):M2–8. [DOI] [PubMed] [Google Scholar]
- 13.Porte D. Clinical importance of insulin secretion and its interaction with insulin resistance in the treatment of type 2 diabetes mellitus and its complications. Diabetes Metab Res Rev. 2001;17(3):181–8. [DOI] [PubMed] [Google Scholar]
- 14.International Association of Diabetes and Pregnancy Study Groups Consensus Panel, Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33(3):676–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998;21(12):2191–2. [DOI] [PubMed] [Google Scholar]
- 16.Seltzer HS, Allen EW, Herron AL, Brennan MT. Insulin secretion in response to glycemic stimulus: relation of delayed initial release to carbohydrate intolerance in mild diabetes mellitus. J Clin Invest. 1967;46(3):323–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sun Y, Han J, Lin Z, Song L, Wang C, Jia W. Delayed insulin secretion response during an OGTT is associated with an increased risk for incidence of diabetes in NGT subjects. J Diabetes Complications. 2016;30(8):1537–43. [DOI] [PubMed] [Google Scholar]
- 18.Hayashi T, Boyko EJ, Sato KK, McNeely MJ, Leonetti DL, Kahn SE, et al. Patterns of insulin concentration during the OGTT predict the risk of type 2 diabetes in Japanese Americans. Diabetes Care. 2013;36(5):1229–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Little RJA, Rubin DB. Statistical analysis with missing data. Third Edition. Hoboken: Wiley; 2019.
- 20.Donald B. Rubin. Multiple imputation for nonresponse in surveys. New York, NY;Chichester, UK: Wiley; 1987. [Google Scholar]
- 21.Festa A, Williams K, Hanley AJG, Haffner SM. β-Cell dysfunction in subjects with impaired glucose tolerance and early type 2 diabetes: comparison of surrogate markers with First-Phase insulin secretion from an intravenous glucose tolerance test. Diabetes. 2008;57(6):1638–44. [DOI] [PubMed] [Google Scholar]
- 22.Ma Y, Wang N, Gu L, Wei X, Ren Q, Huang Q, et al. Postpartum assessment of the beta cell function and insulin resistance for Chinese women with previous gestational diabetes mellitus. Gynecol Endocrinol. 2019;35(2):174–8. [DOI] [PubMed] [Google Scholar]
- 23.Kahn SE. The relative contributions of insulin resistance and beta-cell dysfunction to the pathophysiology of type 2 diabetes. Diabetologia. 2003;46(1):3–19. [DOI] [PubMed] [Google Scholar]
- 24.Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care. 2004;1(6):1487–95. [DOI] [PubMed] [Google Scholar]
- 25.Mari A, Ahrén B, Pacini G. Assessment of insulin secretion in relation to insulin resistance. Curr Opin Clin Nutr Metab Care. 2005;8(5):529–33. [DOI] [PubMed] [Google Scholar]
- 26.Retnakaran R, Qi Y, Connelly PW, Sermer M, Hanley AJ, Zinman B. Risk of early progression to prediabetes or diabetes in women with recent gestational dysglycaemia but normal glucose tolerance at 3-month postpartum. Clin Endocrinol (Oxf). 2010;73(4):476–83. [DOI] [PubMed] [Google Scholar]
- 27.Dotevall A, Johansson S, Wilhelmsen L, Rosengren A. Increased levels of triglycerides, BMI and blood pressure and low physical activity increase the risk of diabetes in Swedish women. A prospective 18-year follow-up of the BEDA study. Diabet Med. 2004;21(6):615–22. [DOI] [PubMed] [Google Scholar]
- 28.Kim SH, Kim MY, Yang JH, Park SY, Yim CH, Han KO, et al. Nutritional risk factors of early development of postpartum prediabetes and diabetes in women with gestational diabetes mellitus. Nutrition. 2011;27(7–8):782–8. [DOI] [PubMed] [Google Scholar]
- 29.McPhillips JB, Barrett-Connor E, Wingard DL. Cardiovascular disease risk factors prior to the diagnosis of impaired glucose tolerance and non-insulin-dependent diabetes mellitus in a community of older adults. Am J Epidemiol. 1990;131(3):443–53. [DOI] [PubMed] [Google Scholar]
- 30.Barden A, Singh R, Walters B, Phillips M, Beilin LJ. A simple scoring method using cardiometabolic risk measurements in pregnancy to determine 10-year risk of type 2 diabetes in women with gestational diabetes. Nutr Diabetes. 2013;3(6):e72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Oh YS, Bae GD, Baek DJ, Park EY, Jun HS. Fatty Acid-Induced lipotoxicity in pancreatic Beta-Cells during development of type 2 diabetes. Front Endocrinol (Lausanne). 2018;9:384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Lasram MM, Bouzid K, Douib IB, Annabi A, El Elj N, El Fazaa S, et al. Lipid metabolism disturbances contribute to insulin resistance and decrease insulin sensitivity by malathion exposure in Wistar rat. Drug Chem Toxicol. 2015;38(2):227–34. [DOI] [PubMed] [Google Scholar]
- 33.Kotronen A, Velagapudi VR, Yetukuri L, Westerbacka J, Bergholm R, Ekroos K, et al. Serum saturated fatty acids containing triacylglycerols are better markers of insulin resistance than total serum triacylglycerol concentrations. Diabetologia. 2009;52(4):684–90. [DOI] [PubMed] [Google Scholar]
- 34.Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care. 2002;25(10):1862–8. [DOI] [PubMed] [Google Scholar]
- 35.Bardugo A, Bendor CD, Rotem RS, Tsur AM, Derazne E, Gerstein HC, et al. Glucose intolerance in pregnancy and risk of early-onset type 2 diabetes: a population-based cohort study. Lancet Diabetes Endocrinol. 2023;11(5):333–44. [DOI] [PubMed] [Google Scholar]
- 36.Hiersch L, Shah BR, Berger H, Geary M, McDonald SD, Murray-Davis B, et al. Oral glucose tolerance test results in pregnancy can be used to individualize the risk of future maternal type 2 diabetes mellitus in women with gestational diabetes mellitus. Diabetes Care. 2021;44(8):1860–7. [DOI] [PubMed] [Google Scholar]
- 37.Ekelund M, Shaat N, Almgren P, Groop L, Berntorp K. Prediction of postpartum diabetes in women with gestational diabetes mellitus. Diabetologia. 2010;53(3):452–7. [DOI] [PubMed] [Google Scholar]
- 38.Liu H, Zhang S, Wang L, Leng J, Li W, Li N, et al. Fasting and 2-hour plasma glucose, and HbA1c in pregnancy and the postpartum risk of diabetes among Chinese women with gestational diabetes. Diabetes Res Clin Pract. 2016;112:30–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Damm P, Kühl C, Bertelsen A, Mølsted-Pedersen L. Predictive factors for the development of diabetes in women with previous gestational diabetes mellitus. Am J Obstet Gynecol. 1992;167(3):607–16. [DOI] [PubMed] [Google Scholar]
- 40.Faerch K, Borch-Johnsen K, Holst JJ, Vaag A. Pathophysiology and aetiology of impaired fasting glycaemia and impaired glucose tolerance: does it matter for prevention and treatment of type 2 diabetes? Diabetologia. 2009;52(9):1714–23. [DOI] [PubMed] [Google Scholar]
- 41.Anastasiou E, Vasileiou V, Athanasiadou A, Stavrianos C, Saltiki K, Alevizaki M. Phenotypic and metabolic characteristics of women with isolated hyperglycemia in pregnancy-is the time-point important? Diabetes Res Clin Pract. 2010;90(3):333–8. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analyzed in this study are available from the corresponding author on reasonable request.