Dual Metabolic Burden of Polyendocrine Metabolic Ovarian Syndrome (PMOS) and Gestational Diabetes in Pregnancy: Impact on Neonatal Anthropometry: Insights From the Born in Bradford Cohort Authors: Harshal A. Deshmukh, Amal Rauf, Eunice Waife, Aatif Syed, Emmanuel Ssemmondo, Najeeb Shah, Thozhukat Sathyapalan Published: 2026 ABSTRACT ABSTRACTBackgroundPolyendocrine metabolic ovarian syndrome (PMOS) and gestational diabetes mellitus (GDM) independently influence foetal growth, but their combined impact remains poorly characterised. This study examined the individual and joint associations of PMOS and GDM with neonatal anthropometry in a large, multi‐ethnic population.MethodsWe performed a secondary analysis of 9454 singleton pregnancies from the Born in Bradford (BiB) cohort (UK, 2007–2011). Women were classified into four groups: (1) normoglycemic non‐PMOS controls, (2) PMOS only, (3) GDM only and (4) PMOS + GDM. GDM was defined using NICE criteria. The primary outcome was birth weight; secondary outcomes included head circumference, mid‐arm circumference and subscapular skinfold thickness. Multivariable linear regression models were used to assess associations, adjusting for maternal age, pre‐pregnancy BMI and ethnicity.ResultsWomen with PMOS were younger (25.5 ± 4.6 vs. 27.3 ± 5.6 years, p < 0.001) and had higher BMI (27.9 vs. 25.9 kg/m2, p < 0.0001) than controls. Offspring of women with PMOS or GDM had lower mean birth weights than controls (3171 and 3162 g vs. 3242 g, p = 0.0005), with the greatest reduction among PMOS + GDM pregnancies (3118 g). In adjusted models, maternal BMI was strongly associated with birth weight (β = 19.3 g/kg/m2, p < 1 × 10−62), whereas PMOS (β = −123.9 g, p = 0.007), GDM (β = −136.5 g, p = 1.8 × 10−9) and PMOS + GDM (β = −241.8 g, p = 0.046) were associated with lower birth weights. Infants of mothers with combined PMOS + GDM also had the lowest mid‐arm circumference and subscapular skinfold thickness.ConclusionsBoth PMOS and GDM are independently and additively associated with lower foetal birth weight, with coexisting PMOS + GDM conferring the greatest reduction in neonatal size and adiposity. These findings highlight the compounded intrauterine metabolic stress in pregnancies in women with coexisting PMOS and GDM and underscore the importance of targeted antenatal care to mitigate intergenerational metabolic risk. Introduction Polyendocrine metabolic ovarian syndrome (PMOS) is one of the most common endocrine disorders in women of reproductive age, with prevalence estimates around 6%–10% globally, varying by diagnostic criteria and population. As per the Rotterdam consensus, a diagnosis requires two of three features: chronic oligo‐anovulation, clinical or biochemical hyperandrogenism and polycystic ovarian morphology on ultrasound [1, 2, 3] PMOS extends beyond reproduction into metabolism, with insulin resistance, central obesity, dyslipidaemia and elevated type 2 diabetes risk frequently observed, and these disturbances carry into pregnancy where risks of miscarriage, gestational diabetes mellitus (GDM), hypertensive disorders, preterm delivery and abnormal foetal growth are increased [4]. PMOS and GDM have overlapping pathophysiology characterised by insulin resistance, hyperinsulinaemia and metabolic dysregulation. Hyperandrogenism in PMOS exacerbates insulin resistance, while maternal hyperglycaemia in GDM enhances placental nutrient transport and influences foetal growth [5]. PMOS pregnancies may also exhibit placental dysfunction, endothelial dysfunction and heightened inflammation, potentially impairing uteroplacental perfusion; the result can be heterogeneous neonatal size (both SGA and LGA) depending on background risk and comorbidities [6]. If GDM is superimposed on PMOS, it can add further glycaemic burden, making women with co‐existing PMOS and GDM biologically plausible high‐risk candidates—yet direct comparative data remain limited [7]. Large epidemiologic studies report higher risks of preeclampsia, preterm birth and neonatal complications in PMOS independent of age and body mass index (BMI) [4]. Contemporary literature shows that maternal hyperglycaemia—even below traditional diabetes thresholds—is continuously related to adverse outcomes [8]. However, few studies have examined PMOS and GDM together, leaving a gap in understanding their combined impact on foetal growth and adiposity. The Born in Bradford (BiB) cohort [9, 10] enables this question to be addressed at scale. BiB recruited > 12,000 pregnant women in Bradford, UK, a multi‐ethnic, socioeconomically diverse city with high metabolic risk and a large South Asian population—groups with increased susceptibility to insulin resistance, PMOS and GDM and adverse perinatal outcomes at lower BMI thresholds. The objective of this study was to determine the effect of coexisting PMOS and GDM with one neonatal anthropometry (birthweight, head circumference, mid‐arm circumference, subscapular skinfold thickness), to inform antenatal care and long‐term prevention strategies. Methods Study Design and Setting The BiB study [9, 10] is an ongoing longitudinal birth cohort designed to investigate how environmental, psychological and genetic influences shape the health and development of mothers and their children. Conducted in Bradford, a city in Northern England characterised by high ethnic diversity and socioeconomic disadvantage, the study recruited women attending antenatal clinics at Bradford Royal Infirmary. Recruitment procedures included interviewer‐administered baseline questionnaires carried out in semi‐private interview areas to ensure confidentiality. Participants could choose to complete the questionnaire in English, Urdu or Mirpuri Punjabi, reflecting the linguistic diversity of the local population. Women who were unable to communicate in any of these languages were eligible to join the cohort but did not complete the baseline survey and were therefore excluded from the present analysis. The BiB cohort comprises pregnant women residing in Bradford who consented to participate during routine glucose tolerance testing at approximately 26–28 weeks' gestation, providing a sample broadly representative of the city's multi‐ethnic maternal population. Women were invited to participate if they were attending a routine 26–28 week oral glucose tolerance test (OGTT), were planning to give birth at the Bradford Royal Infirmary, and were able to give informed consent. For the current analysis, we included 9454 women with singleton pregnancies and complete data on PMOS status, OGTT results and neonatal anthropometric outcomes. Exclusion criteria included multiple pregnancies, missing OGTT data or missing birthweight and anthropometric measurements. Maternal BMI in the BiB cohort was derived from weight and height recorded at the booking antenatal visit, which are routinely measured in clinical practice rather than self‐reported. BMI was calculated as weight (kg) divided by height squared (m2). PMOS diagnosis was based on clinical documentation in the electronic health records. All participants underwent a standard 75 g OGTT at 26–28 weeks' gestation. Plasma glucose was measured fasting and at 2 h post‐load. GDM was defined in accordance with NICE 2015 criteria (fasting BG ≥ 5.6 mmol/L and/or a 2 h BG ≥ 7.8 mmol/L). Participants without GDM and with a 2 h glucose value < 7.8 mmol/L were classified as having normal glucose tolerance. Participants were stratified into four mutually exclusive groups: Control group: No PMOS and normal glucose tolerance; PMOS only, GDM only and PMOS + GDM. PMOS diagnosis was ascertained from linked primary care electronic health records using clinician‐recorded diagnostic codes (Read/CTV3 coding systems). The specific diagnostic criteria used (e.g., Rotterdam or NIH) were not available within the data set. While the specific diagnostic criteria (e.g., Rotterdam, NIH) used by individual clinicians are not explicitly captured within the data set, UK clinical practice during the study period predominantly aligns with Rotterdam criteria, although this cannot be verified for all cases. The primary outcome was birthweight measured in grams within 24 h of delivery using calibrated hospital scales. Secondary outcomes included head circumference (cm), mid‐arm circumference (cm) and subscapular skinfold thickness (mm)—anthropometric measures reflecting neonatal growth and adiposity. These measurements were obtained by trained research midwives using standardised protocols and equipment (non‐stretchable measuring tapes and Harpenden skinfold callipers). Maternal characteristics considered as potential confounders included age at delivery, pre‐pregnancy BMI, ethnicity (White European, South Asian, Other), smoking status and socioeconomic status (Index of Multiple Deprivation). These variables were collected from maternal questionnaires and hospital records at recruitment. Statistical Analysis Baseline characteristics were compared across the four exposure groups using one‐way analysis of variance (ANOVA) for continuous variables and chi2 tests for categorical variables. Neonatal anthropometric outcomes were compared across groups using ANOVA with post hoc Tukey tests. Multivariable linear regression models were constructed to assess the independent association between PMOS, GDM and neonatal anthropometry. Adjusted models included maternal age, pre‐pregnancy BMI and ethnicity. Interaction terms were included to test for synergistic effects of PMOS and GDM on birth outcomes. Regression coefficients (β) and 95% confidence intervals (CIs) were reported. All analyses were performed using R 4.4.2. A two‐sided p < 0.05 was considered statistically significant. Model assumptions were verified by inspection of residual plots and variance inflation factors. The BiB study received ethical approval from the Bradford Research Ethics Committee (Ref: 07/H1302/112). Results Table 1 shows the demographic characteristics of the study population. Mothers with PMOS were significantly younger at conception compared with controls (mean age 25.5 ± 4.6 vs. 27.3 ± 5.6 years, p < 0.001). The distribution of ethnicity differed between groups, with a higher proportion of South Asians in the PMOS group (67% vs. 52%) and fewer Caucasians (28% vs. 43%, p < 0.001). Women with PMOS had higher mean booking weight (78.1 vs. 73.7 kg) and BMI (27.9 vs. 25.9 kg/m2, both p < 0.0001). The age at menarche was similar between groups (p = 0.8). A lower proportion of women with PMOS reported ever smoking regularly (19% vs. 29%, p < 0.0001) or consuming alcohol during pregnancy or the 3 months preceding conception (14% vs. 31%, p < 0.001). Employment status and IMD scores were comparable between the two groups. Table 1Demographic characteristics of the study population.CharacteristicPMOS (n = 279)Control (n = 9175) p valueMean maternal age, years25.5 (4.6)27.3 (5.6)< 0.001Ethnicity< 0.001Caucasians79 (28%)4408 (43%)South Asians187 (67%)5314 (52%)Others5 (2%)551 (5%)Mother's weight, kg78.0673.7< 0.0001Mother's booking BMI27.8525.9< 0.0001Age at first period, years13.2113.240.8Mother ever regularly smoked55 (19%)2987 (29%)< 0.0001Mother smoked during pregnancy29 (10%)1736 (17%)0.07Mother drank alcohol during pregnancy or within 3 months before41 (14%)3245 (31%)< 0.001Mother currently employed93 (33%)4637 (45%)0.1Index of multiple deprivation (IMD) score41.2 (17.8)42.1 (17.0)0.41John Wiley & Sons, Ltd. Demographic characteristics of the study population. Gestational age at the time of the OGTT was comparable across all four groups, ranging from 26.7 to 27.0 weeks. Women with GDM were generally older than their normoglycaemic counterparts, both in the non‐PMOS group (30.5 vs. 27.2 years) and in those with PMOS (27.6 vs. 25.4 years). BMI was highest among women with PMOS and GDM (31.3 kg/m2), followed by those with non‐PMOS GDM (28.7 kg/m2). As expected, fasting and 2 h plasma glucose levels (Figure 1) were significantly higher in women with GDM compared with normoglycaemic women, irrespective of PMOS status (non‐PMOS: 5.33 vs. 4.43 mmol/L and 8.69 vs. 5.37 mmol/L; PMOS: 5.25 vs. 4.45 mmol/L and 8.84 vs. 5.52 mmol/L, respectively). Fasting and 2 h glucose levels in the study subgroups. Table 2 summarises neonatal anthropometric measures at birth according to maternal PMOS and GDM status. Overall, mean head circumference was comparable across groups (p = 0.066), with no significant difference between infants of mothers with or without PMOS or GDM. Mid‐arm circumference differed modestly between groups (p = 0.017), with slightly lower mean values observed in infants born to mothers with both PMOS and GDM (10.4 ± 1.17 cm) compared to those without either condition (10.7 ± 1.17 cm). Subscapular skinfold thickness, an indicator of neonatal adiposity, also varied significantly (p < 0.001), being highest among offspring of non‐PMOS women with GDM (4.93 ± 1.21 mm) and lowest among those of women with PMOS and GDM (4.64 ± 0.87 mm). Birth weight differed significantly across groups (p = 0.0005). Infants of non‐PMOS normoglycaemic mothers had the highest mean birth weight (3242 ± 561 g), while those of PMOS mothers with GDM had the lowest (3118 ± 565 g). Table 2Effect of PMOS and GDM on neonatal birthweight and anthropometric parameters.CharacteristicNon‐PMOS normoglycaemicNon‐PMOS GDMPMOS normoglycaemiaPMOS GDM p value (ANOVA)Head circumference (cm)34.3 (1.63)34.1 (1.47)34.2 (1.81)34.3 (2.16)0.066Mid‐arm circumference (cm)10.7 (1.17)10.6 (1.10)10.5 (1.20)10.4 (1.17)0.017Subscapular skinfold (mm)4.70 (1.09)4.93 (1.21)4.68 (0.98)4.64 (0.87)< 0.001Birth weight (g)3242 (561)3162 (573)3171 (626)3118 (565)0.0005John Wiley & Sons, Ltd. Effect of PMOS and GDM on neonatal birthweight and anthropometric parameters. Figure 2 shows the regression analysis evaluating predictors of birth weight demonstrated that higher maternal BMI was a strong positive determinant (β = 19.33 g per kg/m2, SE = 1.15, p < 1 × 10−62). Gestational age showed a positive but nonsignificant association (β = 2.82 g per week, SE = 3.50, p = 0.42). In contrast, both GDM and PMOS were associated with lower birth weight. Compared with normoglycaemic non‐PMOS pregnancies, infants of mothers with GDM had markedly lower birth weights (β = −136.54 g, SE = 22.66, p = 1.77 × 10−9), while those born to normoglycaemic women with PMOS were also lighter (β = −123.91 g, SE = 46.13, p = 0.007). The combination of PMOS and GDM was associated with the largest reduction in birth weight (β = −241.76 g, SE = 121.37, p = 0.046). We tested for interaction between PMOS and GDM by including an interaction term in the multivariable model. The interaction term was not statistically significant (β = 18.7 g, p = 0.887), indicating no evidence of a multiplicative interaction between PMOS and GDM on birth weight. Effect of PMOS and GDM on birth weight. *Non‐PMOS‐normoglycaemic patient as reference group. Discussion In this large, multi‐ethnic pregnancy cohort, we observed that women with coexisting PMOS and GDM were associated with higher maternal adiposity and glycaemic levels during gestation, suggesting an additive metabolic burden. Although both PMOS and GDM individually influence foetal growth, our findings indicate that their coexistence may further perturb intrauterine metabolic environments potentially influencing neonatal growth patterns. While such differences in early‐life anthropometry have been associated with later metabolic outcomes in previous studies, our data are limited to birth measurements and do not allow direct inference regarding long‐term metabolic risk. PMOS and GDM share overlapping pathophysiological mechanisms rooted in insulin resistance, hyperinsulinaemia and β‐cell dysfunction [4]. In PMOS, hyperandrogenism amplifies peripheral insulin resistance and disrupts glucose homoeostasis, while pregnancy imposes additional insulin demands that may exceed β‐cell compensatory capacity, precipitating GDM. The resultant hyperglycaemia enhances transplacental glucose and lipid transport, promoting foetal adiposity through increased foetal insulin secretion [8]. Beyond glucose metabolism, inflammation, endothelial dysfunction and altered placental vascularity—hallmarks of both PMOS and GDM—may compromise uteroplacental perfusion and nutrient partitioning, contributing to heterogeneous neonatal anthropometry [6]. A recent systematic review [11] including 45 studies and more than 8000 women with PMOS reported that offspring of women with PMOS have significantly lower mean birth weight compared to those born to women without PMOS. This association persisted after sensitivity analyses excluding pregnancies exposed to metformin or following bariatric surgery. The review also noted that mothers with PMOS had a higher pre‐pregnancy BMI and greater gestational weight gain, despite being slightly younger than the comparison group. Our findings are consistent with these observations—women with PMOS in our cohort were younger and had a higher BMI than non‐PMOS controls. Similarly, consistent with our study findings, a Norwegian cohort study [12] of 390 women with PMOS found that their offspring had lower birth weight, birth length and head circumference than those of mothers without PMOS. In the present study, we extend these observations by showing that when PMOS coexists with GDM, the effects on birth weight and neonatal anthropometry are even more pronounced. To our knowledge, this is only the second study to directly compare birth weight and detailed anthropometric characteristics between offspring of women with PMOS with and without GDM. A recent Swedish nationwide register‐based historical cohort study [13] examined the combined impact of PMOS and GDM on birth weight and neonatal outcomes but did not demonstrate any additive effects. However, it is important to note that this study reported a very low prevalence of GDM, likely due to the use of higher diagnostic thresholds than those recommended by the WHO in 2013. Consequently, some cases of GDM may have been missed, potentially diluting the observed associations between PMOS–GDM cooccurrence and adverse perinatal outcomes [13]. In this cohort, women with PMOS and GDM were associated with lower birth weight, a finding that contrasts with classical observations of increased foetal growth in untreated hyperglycaemia, such as those reported in the HAPO Study [8]. However, our findings are consistent with previous analyses from the BiB cohort [14], where lower birth weight has also been observed among women diagnosed with GDM. A likely explanation is that the observed associations reflect the combined effect of GDM diagnosis and subsequent clinical management, rather than the natural history of untreated hyperglycaemia. In routine clinical practice, women diagnosed with GDM undergo dietary intervention, glucose monitoring and pharmacological treatment where required, which may mitigate excessive foetal growth. Therefore, the lower birth weight observed in this study may represent treated GDM and improved glycaemic control, rather than an intrinsic growth‐restricting effect of GDM itself. In addition, the BiB cohort includes a large proportion of South Asian women, a population characterised by lower birth weight and distinct foetal growth patterns, even at comparable levels of maternal glycaemia. Ethnic‐specific differences in insulin sensitivity, body composition and placental function may also contribute to these findings. The combination of earlier diagnosis, lower BMI thresholds and proactive management in this population may further influence neonatal outcomes. We also demonstrate that offspring of women with combined PMOS and GDM have the lowest mid‐arm circumference and subscapular skinfold thickness compared with those born to mothers with PMOS or GDM alone. This finding is clinically significant, as a lower mid‐upper arm circumference has been associated with an increased risk of metabolic syndrome and insulin resistance later in life [15]. Likewise, lower subscapular skinfolds at birth or early life indicate undernutrition and impaired foetal growth and have been linked to reduced metabolic flexibility and higher risk of insulin resistance during later catch‐up growth [16]. Our findings highlight the importance of targeted screening and individualised management strategies for women with PMOS during pregnancy. Optimising pre‐pregnancy metabolic health through weight management, insulin sensitisation and lifestyle modification is critical. Furthermore, the observed differences in neonatal adiposity measures underscore the need for postnatal surveillance, as altered foetal body composition is an early determinant of childhood obesity and insulin resistance. Postpartum, women with both PMOS and GDM remain at high risk of developing type 2 diabetes and cardiovascular disease, warranting structured follow‐up and preventive strategies that extend beyond delivery. Integration of obstetric and endocrine care pathways, with postpartum glycaemic re‐evaluation and lifestyle support, could mitigate intergenerational transmission of metabolic risk. The principal strength of this study lies in the utilisation of the BiB cohort—a large, ethnically diverse and well‐characterised data set with detailed metabolic and anthropometric measures. The prospective design, standardised OGTT data and inclusion of multiple neonatal anthropometric indices enable comprehensive evaluation of maternal metabolic influences on foetal growth. Adjustment for confounders such as maternal age, BMI and ethnicity enhances robustness and internal validity. However, several limitations must be acknowledged. PMOS diagnosis was based on CTV3 codes and reported by general physicians rather than uniform biochemical or ultrasound verification, which may introduce misclassification. The OGTT was performed at a single time point in mid‐pregnancy and may not capture earlier glycaemic fluctuations or late‐onset GDM. As an observational analysis, residual confounding from diet, gestational weight gain or insulin treatment cannot be excluded. Moreover, while neonatal anthropometry offers useful surrogate markers of foetal growth, it cannot fully represent later metabolic outcomes. Further, the number of women with both PMOS and GDM was smaller than other subgroups, potentially limiting statistical power to detect subtle synergistic effects. PMOS is a heterogeneous condition encompassing multiple phenotypes with differing metabolic profiles, including hyperandrogenic and non‐hyperandrogenic subtypes. In this study, PMOS status was derived from routinely collected clinical data, and detailed phenotypic information—such as androgen levels, ovulatory status or ovarian morphology—was not available. As a result, we were unable to examine phenotype‐specific associations. Finally, although we adjusted for key maternal confounders, residual confounding from unmeasured or incompletely measured factors cannot be excluded. In particular, data on gestational weight gain, post‐diagnosis glycaemic control, treatment for PMOS (e.g., metformin use) and the timing and intensity of GDM management were not consistently available. These factors may influence foetal growth and could partially account for the observed associations. In summary, this study demonstrates that coexisting PMOS and GDM confer a compounded metabolic burden during pregnancy, with implications for both maternal and neonatal health. These findings emphasise the need for early identification, preconception optimisation and tailored antenatal management for women with PMOS. Future research should explore mechanistic pathways linking dual maternal insulin resistance and placental function, and assess whether interventions targeting glucose and weight management before conception can attenuate the intergenerational transmission of metabolic risk. Ethics Statement The Bradford Research Ethics Committee granted ethical approval for data collection (Ref 07/H1302/112). Conflicts of Interest The authors declare no conflicts of interest.