Miroslava Gojnic

Background: Fetal programming, conceptualized within the Developmental Origins of Health and Disease framework, describes how intrauterine exposures may shape long-term offspring physiology and disease susceptibility. Metformin is increasingly used during pregnancy for gestational diabetes mellitus, type 2 diabetes mellitus, and selected cases of polycystic ovary syndrome. Its developmental interpretation is complex because the drug improves the maternal metabolic environment but also crosses the placenta and directly exposes the fetus. Methods: This focused narrative review used iterative, targeted searches of PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar to identify pharmacological, placental, mechanistic, clinical, guideline-based, and offspring follow-up evidence. The literature search covered publications available through 31 May 2026. The manuscript was subsequently revised during the editorial submission and peer-review process, and the evidence synthesis was re-evaluated against this final literature cut-off date. Earlier landmark studies were retained where necessary for historical, pharmacological, or methodological context. Results: Metformin reduces hepatic glucose production and maternal insulin resistance and may limit gestational weight gain, insulin requirements, neonatal hypoglycemia, and excessive fetal growth in selected pregnancies. Placental transfer creates biological plausibility for direct effects on AMPK, mitochondrial, mTOR, nutrient-sensing, and epigenetic pathways. Long-term studies are broadly reassuring regarding neurodevelopment and major metabolic disease, but findings on growth trajectory, adiposity, and small-for-gestational-age risk remain heterogeneous and indication-dependent. Conclusions: Metformin should be interpreted neither as uniformly beneficial nor as developmentally neutral. Its use should be individualized according to maternal phenotype, indication, glycemic benefit, placental function, fetal growth, dose, timing, and the limitations of long-term offspring evidence.

Background: Hypertensive disorders of pregnancy (HDP), particularly gestational hypertension and preeclampsia, are major causes of maternal and perinatal morbidity. Metabolic disorders such as gestational diabetes mellitus (GDM), type 2 diabetes mellitus (T2DM), obesity, and polycystic ovary syndrome (PCOS) share mechanisms with HDP, including insulin resistance, inflammation, oxidative stress, endothelial dysfunction, and placental maladaptation. Metformin may influence these pathways, but its clinical effect on hypertensive outcomes remains uncertain. Methods: A focused narrative review was conducted using structured, targeted searches of PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar for literature published primarily from January 2015 through March 2026, supplemented by landmark studies, clinical guidelines, randomized trials, large observational cohorts, systematic reviews, meta-analyses, and mechanistic and placental studies. Evidence was selected for its relevance to the predefined phenotype-based framework and synthesized qualitatively across mechanistic, clinical, and translational domains. Results: Metformin improves insulin sensitivity and may reduce gestational weight gain, metabolic stress, inflammatory signaling, and oxidative stress. Some randomized trials and meta-analyses in GDM, obesity, and PCOS suggest lower rates of pregnancy-induced hypertension or preeclampsia, whereas recent randomized and population-based studies report neutral effects. Findings are limited by heterogeneous populations, late treatment initiation, variable comparators, supplemental insulin use, inconsistent HDP definitions, and insufficient statistical power. Conclusions: Metformin may act as a metabolic and vascular modifier in selected pregnancies, but current evidence does not support its routine use specifically for HDP prevention. Its public health value lies primarily in accessible metabolic treatment and integration into comprehensive antenatal and postpartum risk reduction pathways, rather than replacement of aspirin, blood pressure surveillance, or maternal–fetal monitoring.

Objectives: To compare placental inflammatory lesions and basal plate vascularization between pregnancies complicated by maternal vascular–metabolic disorders and normoglycemic, normotensive pregnancies, and to determine whether basal plate vascularization differs among metabolic, hypertensive, thrombophilic, and combined disorder phenotypes. Methods: This retrospective observational study included 680 singleton pregnancies with placental histopathology. Pregnancies were classified as complicated by maternal vascular–metabolic disorders or as belonging to a normoglycemic, normotensive comparison group without documented maternal vascular–metabolic disease. Placental inflammatory lesions (villitis and decidualitis) were recorded as present or absent. Basal plate vascularization was assessed using a three-level ordinal scoring system and compared between study groups and among vascular–metabolic disorder phenotypes. Results: Maternal vascular–metabolic disorders were documented in 205 pregnancies, whereas 475 pregnancies constituted the comparison group. Villitis was more frequent in the vascular–metabolic disorder group than in the comparison group (13.2 % vs. 3.2 %, p<0.001), as was decidualitis (71.2 % vs. 16.8 %, p<0.001). Basal plate vascularization scores were higher in the vascular–metabolic disorder group (2.76 ± 0.45 vs. 2.59 ± 0.52, p<0.001). However, the absolute difference was modest, and the median score was 3.0 in both groups. Basal plate vascularization did not differ significantly among isolated metabolic, hypertensive, thrombophilic, and combined disorder phenotypes (p=0.216). Conclusions: Maternal vascular–metabolic disorders were associated with a higher frequency of placental inflammatory lesions and slightly higher basal plate vascularization scores. However, the restricted scoring range, ceiling effect, unadjusted analyses, and retrospective observational design limit the clinical interpretation of these findings and preclude causal inference.

Background: Gestational diabetes mellitus (GDM) and polycystic ovary syndrome (PCOS) are clinically distinct disorders that share important metabolic features. This Perspective proposes a non-causal intergenerational framework linking maternal PCOS-related metabolic vulnerability, susceptibility to GDM, intrauterine metabolic exposure, and later reproductive-metabolic risk in female offspring. Methods: A narrative synthesis of evidence from reproductive endocrinology, obstetrics, diabetology, developmental biology, and public health was undertaken. The strength of evidence supporting individual components of the proposed continuum was qualitatively appraised. Separately, aggregated surveillance data from Belgrade for 2015–2024 were used as a registry-based illustration of ICD-10 O24.4-coded GDM diagnoses. No formal time-series analysis, causal modeling, or forecasting was performed. Results: Evidence was strongest for the increased risk of GDM among women with PCOS and for the association between intrauterine exposure to maternal diabetes and later offspring metabolic susceptibility. Evidence linking maternal PCOS with offspring metabolic vulnerability was less definitive, while direct evidence that GDM exposure leads to clinically diagnosed PCOS in female offspring remained weak and indirect. Registry-recorded GDM rates varied across years, but this variability may reflect differences in screening, diagnostic criteria, coding, reporting completeness, and healthcare access rather than true changes in prevalence. Conclusions: The proposed GDM–PCOS continuum is a hypothesis-generating and probabilistic framework, not an established causal pathway. Its clinical value lies in connecting prevention opportunities before conception, during pregnancy, after delivery, and across the life course. Longitudinal maternal–offspring studies are required to test the proposed relationships.

Background: Progesterone is an endogenous pregnancy hormone widely used in reproductive medicine and obstetrics. Its developmental relevance should be interpreted by distinguishing direct molecular plausibility from indirect obstetric effects, including stabilization of the maternal–decidual–placental maternal–decidual–placental environment, cervical stability, and prolongation of gestation. Methods: A focused qualitative narrative review was performed using targeted literature searches and thematic synthesis. This approach was chosen because the available evidence is heterogeneous with respect to indication, formulation, route, dose, timing, comparator group, and offspring follow-up, making quantitative synthesis inappropriate for the present objective. Results: Direct progesterone-related pathways include receptor-mediated signaling, neuroactive metabolites, and biologically plausible epigenetic regulation; however, human evidence for persistent therapy-induced molecular programming remains limited. Indirect pathways are better supported and include decidualization, immune tolerance, placental stabilization, cervical integrity, uterine quiescence, and prevention of prematurity in selected pregnancies. Clinical evidence should be interpreted separately for placebo-controlled trials, untreated pathological cohorts, and observational follow-up studies because underlying maternal conditions may themselves influence offspring outcomes. Potential risk signals also require formulation-specific interpretation, particularly for synthetic progestogens and 17-alpha hydroxyprogesterone caproate, for which long-term observational data and regulatory reassessments have raised concerns regarding efficacy and possible offspring safety signals. Inadvertent exposure to contraceptive progestins during unrecognized pregnancy, including progestin-only preparations and levonorgestrel-releasing intrauterine systems, should be considered separately from therapeutic obstetric progesterone use. Available evidence has not shown a consistent early adverse signal for natural progesterone when used for recognized indications, although long-term offspring data remain limited and cannot be generalized to synthetic progestogens or 17-alpha hydroxyprogesterone caproate. Conclusions: Progesterone-related therapies should be considered indication-specific and formulation-specific rather than interchangeable. Current evidence supports a cautious and comparator-aware interpretation: natural progesterone appears acceptable for recognized clinical indications based on available short-term and early childhood data, but evidence beyond early childhood remains insufficient for broad safety generalizations.

Background: Type 2 diabetes mellitus (T2DM) in pregnancy is increasingly recognized as a high-risk metabolic condition, frequently accompanied by overweight, obesity, insulin resistance, and adverse maternal and neonatal outcomes. This systematic review and meta-analysis aimed to evaluate body weight-related parameters in pregnancies complicated by T2DM compared with those in non-T2DM control groups. Methods: A systematic search of PubMed, Scopus, and Web of Science was conducted up to 19 August 2025. Original studies reporting body weight, body mass index (BMI), or gestational weight gain (GWG) in pregnant women with T2DM and different control groups were included. Data were synthesized using standardized mean differences (SMDs) with fixed or random-effects models. Maternal, metabolic, delivery, and neonatal outcomes were summarized descriptively. Results: Eighty-seven studies were included in the systematic review and seventy-two were included in the meta-analysis. Pregnant women with T2DM had significantly higher pre-pregnancy body weight and first-trimester body weight than normoglycemic and T1DM controls. Pre-pregnancy BMI was also significantly higher in T2DM pregnancies compared with normoglycemic, T1DM, and gestational diabetes controls. In contrast, GWG did not differ significantly between T2DM and normoglycemic or gestational diabetes pregnancies, while it was significantly lower in T2DM than in type 1 diabetes pregnancies. Adverse maternal and neonatal outcomes, including hypertensive disorders, preterm delivery, fetal growth abnormalities, macrosomia, congenital anomalies, and fetal/neonatal loss, were frequently reported across the included studies. Conclusions: Pregnancies complicated by T2DM are characterized by an unfavorable preconception anthropometric profile. The contrasting patterns across diabetes types suggest different periods for weight-related care: preconception weight optimization appears particularly relevant in T2DM, whereas the higher gestational weight gain observed in T1DM relative to T2DM supports individualized monitoring of gestational weight trajectories during pregnancy. These strategies should be incorporated into comprehensive preconception and antenatal care alongside glycemic optimization and assessment of diabetes-related complications.

Background/Objectives: Gestational diabetes (GD) is a well known risk factor for future metabolic diseases. However, the long-term time-dependent risk of non-alcoholic fatty liver disease (NAFLD) remains unexplored. The aim of this meta-analysis was to quantify the long-term risk of NAFLD in women with previous GD and evaluate if the risk persists beyond the postpartum period. Methods: A systematic search was performed in PubMed using appropriate medical subject headings to identify trials evaluating the incidence of NAFLD among women with previous GD compared to those with normal glucose tolerance (NGT). Studies reporting adjusted risk estimates with a follow-up duration beyond pregnancy were included. Data were extracted and analyzed using relevant statistical methods, with the level of significance at p < 0.05. Results: A total of four studies (N = 2873) were included in the meta-analysis. Women with previous GD had a 2.46-fold higher odds of NAFLD compared to those with NGT (95% CI 1.88–3.23, p < 0.001). The slope for years since delivery was not significant (β = 0.001 per year, 95% CI −0.037 to 0.040, p = 0.901), indicating that the likelihood of NAFLD in women with prior GD did not change over time. Conclusions: GD is associated with a substantially increased and sustained risk of NAFLD, persisting decades beyond pregnancy. These findings identified GD as a potential early risk marker of future liver outcomes and highlight the need for long-term metabolic screening and preventive strategies for this high-risk population.

Introduction: Hypercholesterolemia during reproductive age increases long-term cardiovascular risk and may affect hormonal balance, fertility, and pregnancy outcomes. Despite its importance, it remains frequently underdiagnosed and undertreated in women. Modifiable lifestyle factors, including diet, physical activity, and smoking, influence lipid profiles, making their association with hypercholesterolemia important for prevention strategies. This study aims to assess the prevalence of hypercholesterolemia and its association with dietary and lifestyle characteristics among women of reproductive age in Serbia. Methods: This secondary analysis used data from the cross-sectional Serbian National Health Survey 2019, which included non-institutionalized Serbian residents aged ≥15 years. The Serbian National Health Survey used the European Health Interview Survey (EHIS). Dietary intake was assessed using a Food Frequency Questionnaire (FFQ), and physical activity was assessed using the European Health Interview Survey Physical Activity Questionnaire (EHIS-PAQ). Results: The prevalence of self-reported hypercholesterolemia among women of reproductive age in Serbia was 3.4%. Smokers had approximately twofold higher odds, and obese women had ninefold higher odds of hypercholesterolemia than non-smokers and non-obese women, respectively. Consumption of non-alcoholic beverages and juices was associated with lower odds of hypercholesterolemia. Conclusion: Hypercholesterolemia among women of reproductive age in Serbia is associated with modifiable lifestyle factors, highlighting the need for interventions promoting healthy weight management, smoking cessation, and sustainable lifestyle changes.

Background and Objectives: Gestational diabetes mellitus (GDM) is a major public health concern associated with adverse maternal and neonatal outcomes. It was found that even physiological pregnancy is followed by a significant shift in serum lipid profile, and even more pronounced in GDM pregnancies. We aimed to comprehensively assess lipid parameters among pregnant women with and without GDM. Materials and Methods: A systematic review, covering PubMed, WoS, and SCOPUS until 23 July 2024, with meta-analysis and meta-regression, was conducted, comprising studies measuring TG, TC, LDL-C, HDL-C, VLDL-C, and TG/HDL ratio in pregnant women diagnosed with GDM, and those with normal glucose tolerance. The overall effect size measure was the SMD. NOS and JADAD scales were used for quality assessment, I² statistics for heterogeneity evaluation, and funnel plots for publication bias inspection. Results: A total of 457 studies were included in the qualitative analysis, and 74, 277, and 122 studies were included in the quantitative analysis for the 1st 2nd, and 3rd trimester, respectively. TG and TG/HDL levels were significantly elevated in all three trimesters (TG: SMD = 0.61, 0.57, and 0.48, p < 0.001 for all, and TG/HDL: SMD = 0.44, 0.66, and 0.49; p < 0.001 for all), while TC and LDL-C levels showed significant increases in the 1st and 2nd trimesters (TC: SMD = 0.38, 0.27, p < 0.001 for both, LDL-C: SMD = 0.33, 0.20, p < 0.001 for both), in pregnant women with GDM compared to those without the condition. Conclusions: GDM is associated with significant lipid abnormalities, particularly elevated TG and decreased HDL-C levels. These lipid changes are most pronounced in the first and second trimesters, highlighting the importance of early detection and management.

Cancer stem cells (CSCs) significantly contribute to heterogeneity, malignancy, and therapy resistance in ovarian cancer. Recent studies emphasize the role of the vitamin D receptor (VDR) in regulating cell differentiation and stemness in various types of cancer. This study aims to determine the expression levels of CD44, CD133, and VDR in epithelial ovarian tumors (EOTs) and to compare these levels across different tumor types, including benign, atypical proliferative tumors, and five types of malignant phenotypes, in order to evaluate their potential as diagnostic tools for malignancy. Tissue samples from 218 patients diagnosed with EOT were analyzed. Clinical and histopathologic parameters were recorded. Quantitative immunohistochemical tissue microarray analysis was used to assess the expression levels of CD44, CD133, and VDR using two different scoring systems. Comparisons were made between benign tumors (n = 45), atypical proliferative tumors (n = 42), and ovarian carcinomas (n = 131), including high-grade serous (HGSC) and non-HGSC subtypes. Ovarian cancer, especially HGSC, showed a significantly higher expression of CD44 and VDR (p < 0.05) compared to atypical proliferative tumors and benign tumors. The expression of CD133 was highest in atypical proliferative tumors (p < 0.05). A moderate positive correlation was found between CD44, CD133, and VDR in all groups, with significant correlations with tumor grade and FIGO stage in ovarian cancer (p < 0.05). The increased expression of CD44 and VDR in aggressive ovarian cancer, along with elevated CD133 levels in atypical proliferative tumors, highlights the complexity of tumor biology. These markers may serve as valuable targets for the diagnosis of ovarian cancer.