Potential Biological Mechanisms Underlying Susceptibility to Nausea and Vomiting of Pregnancy in Women with Polycystic Ovary Syndrome: A Narrative Review

Abstract

Polycystic ovary syndrome (PCOS) is a common endocrine disorder characterised by hormonal and metabolic disturbances that are often present before pregnancy. Emerging evidence suggests that women with PCOS may be more susceptible to nausea and vomiting of pregnancy (NVP), particularly moderate-to-severe symptoms, raising interest in the potential biological links between these conditions. This narrative review examines the endocrine and metabolic characteristics of PCOS during early pregnancy and synthesises the pathways potentially involved in NVP across five domains: neuroendocrine regulation, placental-derived factors, appetite and energy balance, gut microbiota–endocrine interactions, and androgen-related metabolic pathways. Overall, pre-existing endocrine and metabolic disturbances in PCOS may modify maternal adaptation to the physiological changes of early pregnancy, allowing these pathways to collectively influence susceptibility to and severity of NVP. Within this multidomain context, altered serotonergic regulation and growth differentiation factor 15 (GDF15)-related placental signalling may act as two parallel pathways involved in sensitisation and signal amplification and, together with the other pathways, influence the manifestations of NVP. Because the available evidence is largely indirect, these potential mechanisms require validation in translational and prospective clinical studies.

Introduction

Polycystic ovary syndrome (PCOS) is a common endocrine and metabolic disorder in women and was recently renamed polyendocrine metabolic ovarian syndrome (PMOS) through a multistep global consensus process.¹ Given that the term PCOS is widely used in the existing literature and in the studies included in this review, PCOS is retained throughout this manuscript. According to the Rotterdam criteria, its global prevalence is approximately 10–13%. It is primarily characterised by hyperandrogenism, ovulatory dysfunction, and polycystic ovarian morphology and is frequently accompanied by insulin resistance and other metabolic abnormalities.² These abnormalities are generally present before pregnancy and may continue to influence endocrine and metabolic adaptations during gestation. A recent systematic review and meta-analysis encompassing 104 studies and 106,690 pregnancies showed that women with PCOS had increased odds of miscarriage, gestational diabetes, gestational hypertension, pre-eclampsia, and caesarean delivery.³ Therefore, pregnancy in women with PCOS does not merely represent a normal gestational state but rather reflects pregnancy-related physiological changes superimposed on heterogeneous pre-existing endocrine and metabolic abnormalities.⁴ Turning to nausea and vomiting of pregnancy (NVP), this common early-pregnancy condition affects approximately 70% of pregnant women, with symptoms ranging from mild to severe.⁵ Hyperemesis gravidarum (HG) represents the most severe clinical form of NVP and is characterised by severe nausea and vomiting in early pregnancy, an inability to eat or drink normally, and marked limitations in daily activities, potentially accompanied by dehydration, electrolyte disturbances, weight loss, and malnutrition.^5,6

From a biological perspective, pre-existing endocrine and metabolic abnormalities in women with PCOS may affect their adaptation to the rapid physiological changes of early pregnancy and potentially intersect with mechanisms related to NVP. However, direct epidemiological evidence regarding the association between PCOS and NVP remains limited. A conference abstract including 793 women with HG and 571 controls reported a positive association between HG and self-reported PCOS diagnosis (OR=2.0, 95% CI: 1.1–3.5, P=0.02).⁷ Beyond this evidence of association, the available mechanistic evidence has largely been derived separately from PCOS or NVP populations, and the mechanisms underlying NVP in pregnant women with PCOS have not been directly investigated. Therefore, these proposed mechanistic intersections should be regarded as hypothesis-generating explanations rather than established causal mechanisms. On this basis, the present review aims to synthesise the available theoretical and empirical evidence, characterise the endocrine and metabolic features of women with PCOS during early pregnancy, and systematically compare them with NVP-related pathways, thereby providing a theoretical basis for future mechanistic research, risk assessment, and clinical management during pregnancy.

Methods

This narrative review employed a structured literature search. Google Scholar, PubMed, Web of Science, and Embase were searched for relevant studies published between January 2015 and June 2026, supplemented by manual screening of the reference lists of included articles. To provide a comprehensive account of the foundational mechanistic evidence underlying the relevant pathways, a limited number of studies published before 2015 were additionally identified through reference screening and included. Search terms included, but were not limited to, “polycystic ovary syndrome”, “PCOS”, “nausea and vomiting of pregnancy”, “NVP”, “hyperemesis gravidarum”, “HG”, “GDF15”, “IGFBP7”, “5-hydroxytryptamine”, “serotonin”, “human chorionic gonadotropin”, “HCG”, “thyroid hormone”, “oestrogen”, “anti-Müllerian hormone”, “AMH”, “prolactin”, “gut microbiota”, “ghrelin”, “leptin”, “androgen”, and “insulin resistance”, combined as appropriate using the Boolean operators “AND” and “OR”.

Studies were eligible if they met at least one of the following criteria: directly evaluated the association between PCOS and NVP or HG; examined endocrine or metabolic characteristics associated with PCOS, healthy pregnancy, NVP, or HG; or provided foundational or mechanistic evidence relevant to the pathways discussed in this review. These pathways included neuroendocrine signalling, placental-derived factors, appetite and energy balance, gut microbiota–endocrine interactions, and androgen-related pathways. Eligible publication types included basic and experimental studies, clinical and observational studies, cohort studies, case–control studies, systematic reviews, meta-analyses, relevant narrative reviews, and international guidelines and consensus statements. Conference abstracts were included only when they provided sufficient methodological and outcome information relevant to the review question. Duplicate records, preprints, conference abstracts lacking adequate methodological or outcome information, and publications for which complete bibliographic information or relevant outcome data were unavailable were excluded. Findings from different sources of evidence were narratively compared and integrated to explore potential, rather than established, mechanistic links between PCOS and NVP.

Hormonal and Metabolic Features of PCOS During Pregnancy

Normal pregnancy requires coordinated endocrine, metabolic, cardiovascular, and immunological adaptations to meet the increasing demands of fetal growth and development. The placenta functions as a major endocrine organ in this process, releasing a variety of hormones into the maternal circulation to regulate maternal physiology and nutrient allocation.⁸ In particular, dynamic changes in insulin sensitivity and the physiological increase in maternal androgen levels are accompanied by compensatory and protective mechanisms that maintain glucose homeostasis and limit the adverse effects of pregnancy-related insulin resistance and androgen elevation on the maternal–fetal unit.⁴ In women with PCOS, these pregnancy-related physiological changes are superimposed on pre-existing hyperandrogenism, elevated anti-Müllerian hormone (AMH) levels, insulin resistance, and associated metabolic abnormalities, and may be accompanied by altered prolactin (PRL) regulation, resulting in an endocrine adaptation that differs from that of healthy pregnancy.

Healthy pregnancy itself is accompanied by dynamic changes in maternal androgen levels. A longitudinal study involving 600 healthy pregnant women demonstrated that total testosterone (T) and androstenedione (A4) concentrations progressively increased with advancing gestation, indicating that a certain degree of androgen elevation represents a physiological endocrine adaptation to pregnancy.⁹ Meanwhile, placental aromatase, encoded by the CYP19A1 gene, converts A4 and T into oestrone and oestradiol, respectively, thereby regulating the androgen–oestrogen balance within the maternal–fetal unit and limiting excessive androgen exposure.¹⁰ In comparison, hyperandrogenism is a core endocrine feature of PCOS and is primarily related to dysregulated ovarian steroidogenesis together with enhanced luteinising hormone (LH) and insulin signalling. Longitudinal evidence indicates that women with PCOS have higher total T and A4 concentrations before conception, whereas during pregnancy these differences become apparent mainly from mid-gestation and persist into the third trimester.¹¹ Another study found that, among women carrying female fetuses, maternal T and A4 concentrations were higher in the PCOS group at approximately 20 weeks of gestation and at delivery than in the corresponding controls.¹² Some studies have also reported a higher free androgen index (FAI) in pregnant women with PCOS than in healthy pregnant controls.¹³ Limited analyses of placental tissue obtained at delivery further suggest reduced placental aromatase activity in PCOS, which may weaken the placental conversion of androgens into oestrogens; however, the timing and precise contribution of this alteration to gestational hyperandrogenism remain to be established.⁴

In healthy women, circulating total AMH levels progressively decline with advancing gestation, with an average reduction of approximately 62% between early pregnancy and 36–39 weeks of gestation.¹⁴ This decline cannot be fully explained by pregnancy-associated haemodilution and may also reflect suppression of ovarian follicular activity and altered hormonal regulation of AMH production. In comparison, elevated AMH is another core endocrine feature of PCOS during early pregnancy. Although AMH levels may also decline with advancing gestation in women with PCOS, they remain relatively high during early pregnancy and are significantly higher than those in non-PCOS controls. Longitudinal evidence further demonstrates that total and cleaved AMH levels remain higher in women with PCOS than in controls from preconception through the third trimester.¹¹

An augmented insulin secretory response is already evident during early normal pregnancy and precedes the marked reduction in insulin sensitivity that typically occurs later in gestation. As pregnancy progresses, increasing peripheral insulin resistance is normally compensated for by enhanced pancreatic β-cell function, thereby maintaining maternal glucose homeostasis and ensuring nutrient availability for the fetus.¹⁵ In women with PCOS, this pregnancy-related physiological metabolic adaptation is superimposed on pre-existing insulin resistance, potentially contributing to more pronounced metabolic abnormalities during early pregnancy. Women with PCOS have higher fasting insulin levels and higher homeostasis model assessment of insulin resistance (HOMA-IR) values than pregnant women without PCOS, even after adjustment for body mass index (BMI).⁴ Hyperinsulinaemia can further amplify hyperandrogenism by increasing ovarian androgen production and reducing hepatic synthesis of sex hormone–binding globulin (SHBG), thereby increasing circulating levels of bioavailable androgens.¹⁶

During healthy pregnancy, serum PRL levels progressively increase with advancing gestation, with trimester-specific reference intervals substantially exceeding those of non-pregnant women.¹⁷ Against this physiological increase, PRL regulation may be further altered during early pregnancy in women with PCOS. A prospective cohort study showed a positive trend between PCOS diagnosis and higher PRL levels during early pregnancy, although the association did not reach the conventional threshold for statistical significance and its clinical relevance remains to be established.¹⁸

Overall, pregnancy in women with PCOS can be viewed as a state in which the physiological endocrine and metabolic changes of normal pregnancy are superimposed on a heterogeneous pre-existing pathological background. Rather than representing a complete failure of normal gestational adaptation, PCOS pregnancy is more likely characterised by altered trajectories or persistently elevated levels of selected endocrine parameters, including androgens, AMH, insulin-related indices, and possibly PRL. However, the magnitude of these differences may be influenced by obesity, PCOS phenotype, fetal sex, mode of conception, and gestational stage.

Hormonal and Metabolic Features of NVP

The development of NVP has been associated with changes in several hormonal systems, particularly human chorionic gonadotropin (HCG), thyroid hormones, and gut-derived appetite regulators, which may collectively influence gastrointestinal function, central emetic signalling, and energy balance.

HCG is a key hormone in early pregnancy, and its temporal pattern broadly parallels the onset, progression, and resolution of NVP symptoms. Higher HCG levels have been associated with greater symptom severity, and women with HG generally have higher HCG concentrations than those without symptoms.¹⁹ A large cohort study similarly showed that higher HCG levels at 12–14 weeks of gestation were associated with an increased risk of severe NVP (OR 1.47, 95% CI 1.11–1.95).²⁰ Mechanistically, the structural similarity between HCG and thyroid-stimulating hormone (TSH) may allow HCG to stimulate thyroid function through receptor cross-reactivity.²¹ Clinical observations in conditions characterised by markedly elevated HCG levels, including multiple and molar pregnancies, further support its potential contribution to NVP.²²

Changes in thyroid function are also common in NVP, with women with HG frequently exhibiting biochemical hyperthyroidism characterised by increased free triiodothyronine (FT3) and free thyroxine (FT4) levels together with suppressed TSH.²³ Meta-analyses indicate that these abnormalities are more pronounced in HG than in uncomplicated pregnancies and generally resolve as the symptoms improve.²² This transient thyroid dysfunction is largely attributed to HCG-mediated stimulation of the TSH receptor, which increases thyroid hormone production.²¹ However, whether thyroid dysfunction itself is independently associated with NVP severity remains uncertain. A large cohort study found that (sub)clinical hyperthyroidism was not independently associated with severe NVP.²⁰

Gut-derived appetite hormones provide an additional level of regulation by maintaining energy balance through the gut–brain axis, although this balance appears to be disrupted in HG. Circulating ghrelin levels are reduced in women with HG even after adjustment for BMI; as ghrelin is an orexigenic hormone, this reduction may suppress appetite, exacerbate negative energy balance, and contribute to insufficient nutritional intake.²⁴ Conversely, satiety hormones, particularly pancreatic polypeptide (PP) and peptide YY (PYY), are increased. PP has been reported to identify up to 91.1% of HG cases, suggesting its potential value as a biomarker of HG. By enhancing central satiety signalling and slowing gastric emptying, these hormonal changes may further reduce food intake. Together with alterations in HCG and thyroid function, dysregulation of gut-derived appetite hormones may therefore contribute to NVP through multiple related pathways.²⁵

Oestrogen is another pregnancy-related factor that may contribute to NVP through its effects on gastrointestinal motility. During early pregnancy, increasing oestrogen levels can alter the contraction and relaxation of gastrointestinal smooth muscle, potentially contributing to delayed gastric emptying and associated symptoms such as nausea and vomiting.²⁶

Taken together, NVP is associated with alterations in several hormonal systems, including HCG, thyroid hormones, gut-derived peptides, and oestrogen. These pathways converge on the regulation of appetite, gastrointestinal function, and neurotransmitter responsiveness, and their overlap with endocrine and metabolic features of PCOS provides a basis for examining the potential biological links between the two conditions.

Mechanistic Links Between PCOS and NVP Susceptibility

Although PCOS and NVP have distinct endocrine features, they may involve several potentially overlapping pathways related to appetite regulation, metabolic hormone responsiveness, neuroendocrine signalling, and gastrointestinal function. Pre-existing hyperandrogenism, insulin resistance, and metabolic disturbances in women with PCOS may alter maternal adaptation to early pregnancy and modify responses to pregnancy-related signals, including HCG and growth differentiation factor 15 (GDF15), thereby potentially increasing susceptibility to and severity of NVP. This section synthesises the available evidence on possible links between PCOS and NVP across five domains: neuroendocrine signalling, placental-derived factors, appetite and energy balance, gut microbiota–endocrine interactions, and androgen signalling. These proposed multidomain mechanisms and their potential contribution to NVP susceptibility in women with PCOS are summarised schematically in Figure 1.

Figure 1.

Proposed mechanisms linking PCOS to NVP.

Abbreviations: PCOS, polycystic ovary syndrome; AMH, anti-Müllerian hormone; 5-HT, 5-hydroxytryptamine; GI, gastrointestinal; 5-HT₃, 5-hydroxytryptamine type 3 receptor; NTS, Nucleus Tractus Solitarius; GDF15, growth differentiation factor 15; IGFBP7, insulin-like growth factor-binding protein 7; CTZ, chemoreceptor trigger zone; GFRAL, glial cell line-derived neurotrophic factor family receptor alpha-like; RET, rearranged during transfection receptor tyrosine kinase; NK₁, neurokinin-1; PYY, peptide YY; PP, pancreatic polypeptide; LPS, lipopolysaccharide; TLR4, Toll-like receptor 4; SCFA, short-chain fatty acids; T, testosterone; A4, androstenedione; DHEAS, dehydroepiandrosterone sulfate; NVP, nausea and vomiting of pregnancy; HG, hyperemesis gravidarum.

Neuroendocrine Signalling: Serotonin as a Sensitisation Axis

Serotonin (5-hydroxytryptamine, 5-HT) is an important neurotransmitter involved in appetite regulation, reproductive endocrine function, gastrointestinal motility, and emotional regulation. Most of the body’s 5-HT is produced by enterochromaffin cells in the gut, whereas a smaller proportion is synthesised in the central nervous system through tryptophan hydroxylase 2 (TPH2).²⁷ Given its broad physiological roles, alterations in 5-HT signalling have been reported in both PCOS and NVP.

Altered 5-HT signalling has been reported in women with PCOS and may be associated with reproductive and metabolic dysfunction. Clinical studies have demonstrated reduced serum levels of 5-HT and its metabolite, 5-hydroxyindoleacetic acid, in infertile women with PCOS, with these alterations being associated with changes in LH secretion.²⁸ Experimental studies suggest that 5-HT may exert bidirectional effects on gonadotropin-releasing hormone (GnRH) neurons through different receptor subtypes: activation of 5-HT₂A receptors may promote GnRH release, whereas activation of 5-HT₁A receptors may inhibit it. Therefore, reduced 5-HT signalling may affect GnRH pulsatility and downstream LH and follicle-stimulating hormone (FSH) secretion, although the specific relevance of this mechanism to PCOS remains incompletely understood.²⁹ Based on this indirect evidence, we propose a theoretical 5-HT sensitisation model. This model hypothesises that PCOS-associated alterations in serotonergic signalling may lower the threshold at which pregnancy-related serotonergic signals trigger nausea and vomiting, thereby increasing susceptibility to NVP. However, this model remains hypothetical and should not be regarded as an established mechanism.

In NVP, clinical studies have shown that plasma 5-HT levels are higher in women with HG than in those with mild symptoms or uncomplicated pregnancies. These levels are also positively correlated with Pregnancy-Unique Quantification of Emesis (PUQE) scores.³⁰ 5-HT can trigger vomiting through 5-HT₃ receptors located in both the gastrointestinal tract and the central nervous system. In the peripheral pathway, activation of gastrointestinal 5-HT₃ receptors transmits emetic signals through vagal afferents to the nucleus tractus solitarius (NTS). These peripheral inputs are subsequently integrated with signals from central emetic regions, including the area postrema or chemoreceptor trigger zone, and may interact with the neurokinin-1 (NK₁) receptor pathway. 5-HT₃ receptor antagonists can block these signals. Their effects include receptor binding, receptor internalisation, and reduced interaction between the 5-HT₃ and NK₁ pathways.^31,32 Although most of this evidence comes from studies on chemotherapy-induced nausea and vomiting, it still helps explain how similar pathways may operate in NVP. Although women with PCOS tend to have lower baseline 5-HT levels, whereas NVP is associated with increased 5-HT activity, these findings are not necessarily contradictory. Prolonged exposure to low 5-HT levels could theoretically alter receptor expression or sensitivity, causing women with PCOS to respond more strongly to pregnancy-related serotonergic changes. Such heightened responsiveness could enhance vagal afferent signalling and facilitate the activation of central emetic pathways, thereby increasing susceptibility to and severity of nausea and vomiting. This theoretical process of sensitisation followed by signal amplification remains hypothetical and requires direct translational and clinical validation.

Placental-Derived Factors: GDF15 and IGFBP7 as Potential Stress-Response Mediators

Placental-derived factors have been linked to NVP. Among them, GDF15 and insulin-like growth factor-binding protein 7 (IGFBP7) are often discussed. Both are elevated in early pregnancy and are associated with symptom severity.³³

GDF15

GDF15, a member of the transforming growth factor-β superfamily, is involved in appetite control, body weight regulation, and glucose metabolism, and may also participate in inflammatory processes, although the underlying mechanisms remain incompletely understood.³⁴

Experimental studies indicate that oxidative stress may increase GDF15 expression in ovarian granulosa cells; for example, arachidonic acid has been shown to induce this response, which may represent a protective mechanism.³⁵ However, clinical findings are not entirely consistent. Relatively low GDF15 levels have been reported in adolescents with PCOS and may be associated with impaired regulation of adiposity, whereas treatment with spironolactone, pioglitazone, and metformin has been reported to increase circulating GDF15, accompanied by improvements in insulin resistance, inflammation, and ectopic fat deposition.³⁶ Furthermore, GDF15 levels are positively associated with age and indices of insulin resistance but negatively associated with osteocalcin, suggesting a potential role in the interplay between energy and bone metabolism.³⁷ Metformin treatment may further increase circulating GDF15 in women with PCOS, and the observed associations of GDF15 with insulin signalling and adipogenesis-related pathways suggest that it may contribute to the metabolic effects of this treatment.³⁸ Metformin is primarily used to improve PCOS-related insulin resistance and associated reproductive and metabolic abnormalities. A recent overview of reviews indicated that metformin use during pregnancy may reduce the risks of preterm birth and pregnancy-induced hypertension in women with PCOS, whereas its preventive effects on miscarriage and gestational diabetes remain unclear. Although metformin is generally well tolerated during pregnancy, it may cause gastrointestinal adverse effects, and its long-term metabolic safety for offspring has not been fully established. Its use should therefore be individualised according to maternal metabolic characteristics and pregnancy risks.³⁹

In the pathogenesis of NVP, fetoplacental GDF15 and maternal sensitivity to its effects may play important roles. Most circulating GDF15 during pregnancy originates from the fetoplacental unit and contributes to appetite suppression and nausea and vomiting through signalling mediated by glial cell line-derived neurotrophic factor family receptor alpha-like (GFRAL) and rearranged during transfection receptor tyrosine kinase (RET) in the brainstem. Clinical studies have shown that circulating GDF15 concentrations are higher in women with vomiting during pregnancy or HG than in women without vomiting or with less severe symptoms.⁴⁰ Whole-exome sequencing further identified an association between the GDF15 rs1058587 locus and HG, with the C allele being associated with increased risk in both European (OR=1.64, 95% CI: 1.37–1.97) and non-European populations (OR=2.13, 95% CI: 1.19–3.80).⁴¹ Consistent with this signalling pathway, the predominant localisation of GFRAL to the area postrema and NTS in the brainstem provides a mechanistic link between circulating GDF15 and NVP, as the area postrema is positioned to detect circulating hormonal signals.⁴²

Taken together, PCOS-related alterations in GDF15 homeostasis and the metabolic milieu may cause the gestational rise in placenta-derived GDF15 to occur against differing levels of prepregnancy GDF15 exposure, thereby influencing maternal sensitivity to GDF15 and the emetic response threshold for NVP. Within this theoretical framework, metformin, which improves insulin resistance and increases circulating GDF15, may further modify this process. However, no studies have directly demonstrated that PCOS-related changes in GDF15 or metformin intervention affect the risk or severity of NVP in this population; therefore, these proposed associations remain theoretical.

IGFBP7

In addition to GDF15, alterations in the IGFBP7 pathway have been proposed as another potential mechanism involved in NVP. IGFBP7 is a widely expressed secreted glycoprotein belonging to the insulin-like growth factor-binding protein family. Although it has relatively low affinity for insulin-like growth factors (IGFs), it exhibits a higher affinity for insulin and acts predominantly through IGF-independent pathways. By binding to insulin receptors and reducing their activation, IGFBP7 may impair downstream insulin signalling and thereby contribute to insulin resistance.⁴³

In PCOS, increased IGFBP7 expression has been observed in ovarian granulosa cells and has been associated with ovarian dysfunction. Elevated IGFBP7 expression may inhibit granulosa-cell proliferation, promote apoptosis, and increase T production, whereas reducing IGFBP7 expression may partially restore granulosa-cell function. Consistent with these findings, IGFBP7 deficiency in dehydroepiandrosterone (DHEA)-induced PCOS mouse models improves oestrous-cycle abnormalities and attenuates hyperandrogenic features, suggesting that IGFBP7 may contribute to PCOS development, potentially through the insulin-like growth factor 1/phosphoinositide 3-kinase/protein kinase B (IGF-1/PI3K/Akt) signalling pathway.⁴⁴ Experimental evidence further suggests that IGFBP7 may participate in immune dysregulation by altering the expression of immune-related genes in the spleen and T-cell receptor β-chain diversity. Accordingly, reduced IGFBP7 expression may attenuate inflammatory signalling and partially improve the chronic low-grade immune imbalance associated with PCOS.⁴⁵

In NVP, serum IGFBP7 levels appear to follow a gestational pattern, with higher concentrations observed at approximately 12 weeks of gestation in women with HG than in those with mild symptoms or uncomplicated pregnancies. By 24 weeks, however, this difference is no longer evident as symptoms improve, suggesting that IGFBP7 levels may be associated with the clinical course of the condition.⁴⁶ Genetic studies provide further support, as genome-wide association studies have identified the 4q12 locus encompassing IGFBP7 as being associated with HG risk, with this finding subsequently replicated across different populations.⁴⁷

Women with PCOS commonly exhibit insulin resistance, low-grade inflammation, and oxidative stress, which may create a metabolically vulnerable background during early pregnancy. In this context, pregnancy-related changes in GDF15 and IGFBP7 could elicit greater physiological responses and thereby contribute to more severe or persistent nausea and vomiting. However, this proposed link has not been directly demonstrated and should therefore be regarded as a hypothetical mechanism through which placental-related factors may contribute to increased NVP susceptibility in women with PCOS.

Appetite and Energy Balance: A Metabolic Vulnerability Axis

Ghrelin and leptin are two key hormones involved in appetite regulation, energy metabolism, and gastrointestinal motility. Their levels change during early pregnancy and may be related to NVP.

Ghrelin

Ghrelin is a gut-derived peptide hormone secreted predominantly by the gastric mucosa and is the endogenous ligand of the growth hormone secretagogue receptor (GHSR). In addition to regulating appetite, ghrelin contributes to gastric motility, glucose metabolism, neuroendocrine function, and the physiological response to energy deficiency.⁴⁸

Circulating ghrelin levels are generally lower in women with PCOS than in weight-matched controls, which may reflect the metabolic and endocrine disturbances associated with the condition.^49,50 Reduced ghrelin levels have also been associated with insulin resistance, as indicated by their relationships with the glucose-to-insulin ratio and the HOMA-IR, whereas ghrelin concentrations may increase following metformin treatment alongside improvements in insulin sensitivity.^51,52 Moreover, ghrelin and growth hormone secretagogue receptor type 1a (GHSR-1A) are expressed in the ovary, suggesting potential roles in steroidogenesis and reproductive endocrine regulation. The inverse association between circulating ghrelin and hyperandrogenism further suggests an interaction between ghrelin regulation and androgen excess,⁵³ while its involvement in energy metabolism may provide an additional link with the reproductive axis.⁵⁴ Women with PCOS also exhibit an altered postprandial ghrelin response, characterised by attenuated ghrelin suppression following oral glucose intake, which may indicate impaired metabolic adaptation to nutrient intake.⁵⁵

Reduced ghrelin levels have also been reported in HG, with clinical studies showing lower serum concentrations in affected women than in those with uncomplicated pregnancies, independently of BMI.²⁴ Women with HG frequently experience negative energy balance and substantial weight loss, while satiety hormones such as PYY and PP may remain elevated even during fasting, potentially suppressing food intake and further exacerbating energy deficiency.²⁵ This hormonal pattern may overlap with the reduced ghrelin levels and impaired postprandial regulation observed in PCOS, thereby limiting metabolic adaptability during early pregnancy. Against the background of rapid pregnancy-related endocrine changes, this metabolic vulnerability could contribute to gastric dysmotility, unstable energy balance, and increased susceptibility to nausea and vomiting.

Leptin

Leptin is an adipose-derived hormone that reflects energy stores and contributes to appetite control, body weight regulation, reproductive endocrine activity, and neuroendocrine balance. Acting primarily on the hypothalamus, it activates pro-opiomelanocortin neurons while inhibiting neuropeptide Y/agouti-related peptide pathways.⁵⁶ Leptin levels are often higher in women with PCOS, whereas its regulation appears to be altered in NVP. Although the patterns differ between the two conditions, these changes in leptin signalling may affect energy balance and gastrointestinal function, potentially contributing to increased susceptibility to NVP in women with PCOS.

Leptin levels are often elevated in women with PCOS and are associated with disturbances in energy metabolism and endocrine function. Among women with obesity and PCOS, leptin concentrations are positively associated with BMI, waist-to-hip ratio, and fat mass, suggesting that adiposity contributes to their elevation.⁵⁷ However, leptin levels remain higher in women with PCOS than in healthy controls even after adjustment for age, BMI, and waist-to-hip ratio, with PCOS itself identified as an independent factor associated with elevated leptin.⁵⁸ These findings indicate that the altered leptin profile in PCOS cannot be explained by adiposity alone. Furthermore, the positive associations of leptin with fasting insulin and HOMA-IR suggest a link with insulin resistance and disrupted energy balance,⁵⁹ while its positive association with the FAI and inverse association with sex hormone-binding globulin (SHBG) suggest potential involvement in reproductive endocrine regulation.⁶⁰

In NVP, leptin follows a different pattern, with clinical studies indicating that the increase in leptin from early to mid-pregnancy is attenuated in affected women compared with those without symptoms.⁶¹ Thus, although leptin regulation differs between the two conditions, elevated leptin levels in PCOS may reflect long-standing metabolic imbalance, whereas the blunted gestational leptin response observed in NVP may indicate impaired adaptation of energy homeostasis. These distinct abnormalities could influence maternal physiological adaptation during early pregnancy and potentially increase susceptibility to nausea and vomiting, although this proposed link has not been directly confirmed in pregnant women with PCOS.

Gut Microbiota–Endocrine Interactions: An Inflammation–Gut–Brain Axis

The gut microbiota comprises a diverse community of microorganisms inhabiting the human gastrointestinal tract, with Firmicutes and Bacteroidetes among the predominant bacterial phyla. Its composition is shaped by host genetics, diet, and environmental exposures, and it contributes to energy absorption, metabolic homeostasis, and immune regulation. Accordingly, alterations in gut microbial composition have been associated with metabolic disorders, including obesity.⁶²

In PCOS, gut dysbiosis is commonly reported and is characterised by reduced α-diversity and altered β-diversity, with these changes appearing more pronounced in women with insulin resistance. At the compositional level, Gram-negative bacteria such as Bacteroides and Escherichia/Shigella are often increased, whereas short-chain fatty acid (SCFA)-producing bacteria, including Lachnospira, Faecalibacterium, and Bifidobacterium, are reduced. Collectively, these alterations may impair intestinal barrier integrity, promote systemic inflammation, and disrupt gut–brain signalling.⁶³ One possible mechanism involves lipopolysaccharide (LPS) derived from Gram-negative bacteria. Following disruption of the intestinal barrier, LPS may enter the circulation and interact with LPS-binding protein, cluster of differentiation 14 (CD14), and Toll-like receptor 4 (TLR4), thereby activating downstream inflammatory signalling and promoting the release of cytokines such as tumour necrosis factor-α and interleukin-6. This process may contribute to chronic low-grade inflammation, impaired insulin signalling, and the metabolic abnormalities associated with PCOS.⁶⁴ Beyond this inflammatory pathway, metabolic and microbial disturbances may also reinforce each other: hyperinsulinaemia can stimulate ovarian androgen production, whereas androgen excess may further alter the gut microbiota, suggesting a potential bidirectional relationship between metabolic dysfunction and microbial imbalance.⁶⁵ The reduced abundance of SCFA-producing bacteria may also decrease SCFA availability. In addition to maintaining intestinal barrier integrity, SCFAs regulate enteroendocrine hormone secretion through G protein-coupled receptors 41 and 43 (GPR41 and GPR43), thereby influencing the release of PYY, glucagon-like peptide 1 (GLP-1), and 5-HT.⁶⁶ These findings show that gut microbiota changes are closely linked to metabolic and inflammatory processes in PCOS.⁶⁷

Notably, gut microbial dysbiosis is not only a prominent feature of PCOS, but alterations in gut microbial composition have also been observed in patients with NVP, particularly those with HG. Studies have shown that patients with NVP may exhibit changes in microbial α-diversity and a relative enrichment of Gram-negative bacteria, including members of the family Bacteroidaceae.⁶⁸ Mendelian randomisation studies have further suggested that the genera Defluviitaleaceae UCG011 and Ruminococcus 1/2 are positively associated with the risk of HG, whereas Coprococcus 2 may have a protective effect, providing genetic support for the potential involvement of specific microbial taxa in susceptibility to HG.⁶⁹ Mechanistically, the gut microbiota may influence neuroendocrine networks through the gut–brain axis. Certain microbial metabolites may stimulate 5-HT release from enterochromaffin cells and affect brainstem pathways involved in emesis through vagal and humoral signalling, thereby potentially enhancing nausea and vomiting.⁷⁰ Meanwhile, pregnancy-related hormonal fluctuations, including changes in progesterone, may also reshape the gut microbial composition and alter the abundance of bacteria such as Bifidobacterium. Interactions between microbial dysbiosis and the hormonal environment may therefore influence the manifestations of NVP.⁷¹ However, most existing observational studies collected faecal samples after the onset of nausea and vomiting and therefore cannot determine whether microbial alterations occurred before or after the development of NVP. Reduced food intake, nutritional deprivation, and changes in the metabolic milieu resulting from severe emesis may themselves induce secondary alterations in the gut microbiota. Therefore, the temporal sequence and causal direction of the relationship between the gut microbiota and NVP remain unclear, and the relationship may be bidirectional.

Both PCOS and NVP have been reported to involve microbial features such as an increased abundance of Gram-negative bacteria, a reduction in SCFA-producing bacteria, and altered gut microbial diversity, suggesting that they may share certain functional patterns of gut microbial dysbiosis. These microbial alterations may affect gastrointestinal motility and neuroendocrine responses through the LPS–TLR4-mediated inflammatory pathway and the SCFA–gut hormone axis, involving mediators such as 5-HT, GLP-1, and PYY, thereby providing biological plausibility for the involvement of the gut microbiota in NVP. Accordingly, pre-existing microbial dysbiosis in women with PCOS may alter their adaptation to pregnancy-related hormonal and metabolic changes and may theoretically increase susceptibility to NVP. However, this proposed association is primarily inferred from indirect evidence, and NVP itself may also induce secondary changes in the gut microbiota. Therefore, the causal direction of this relationship cannot currently be established.

Androgen Signalling: A Metabolic–Neuroendocrine Feedback Loop

Androgens are steroid hormones produced primarily by the gonads and adrenal cortex and are involved in sexual development, reproductive function, muscle maintenance, and body fat distribution. Their biological effects are mediated through both genomic and non-genomic pathways.⁷² Hyperandrogenism is a defining feature of PCOS, while changes in maternal androgen levels have also been associated with NVP, suggesting potential overlap through pathways involving androgen metabolism, insulin resistance, inflammatory signalling, and neurotransmitter regulation.

In PCOS, hyperinsulinaemia can stimulate ovarian theca cells to increase androgen production, whereas elevated androgens may impair insulin receptor signalling, thereby establishing a reciprocal interaction that can exacerbate metabolic dysfunction.⁷³ Androgen excess may also affect the central nervous system, as experimental PCOS models have demonstrated reduced levels of monoamine neurotransmitters, including norepinephrine, 5-HT, and dopamine, suggesting a possible link between hyperandrogenism and altered neurochemical regulation.⁷⁴

During pregnancy, higher maternal levels of A4 and dehydroepiandrosterone sulfate (DHEAS) have been associated with NVP symptoms at different gestational stages, while vomiting scores have also been related to total and free T levels and the FAI.⁷⁵ These associations may be more pronounced in HG; moreover, changes in androgen metabolites, including increased androstanediol glucuronide and decreased A4, have been associated with disease severity.⁷⁶

Taken together, alterations in androgen signalling in PCOS and NVP intersect with insulin resistance, inflammatory responses, and neurotransmitter pathways. Long-standing hyperandrogenism and metabolic dysfunction in PCOS may therefore create a physiologically vulnerable background that modifies maternal responses to pregnancy-related hormonal changes. Associations between the severity of nausea and vomiting and circulating A4, DHEAS, and T provide indirect support for the potential involvement of androgen signalling; however, no study has directly demonstrated that androgen abnormalities mediate an increased risk of NVP in women with PCOS. A concise comparison of the baseline alterations in PCOS, the corresponding gestational changes in NVP, the proposed overlapping mechanisms, and the supporting literature is provided in Table 1.

Table 1.

Summary of the Potential Mechanistic Links Between PCOS and NVP

Biological Domain	Pre-Existing Features Associated with PCOS	Associated Changes in NVP	Proposed Overlapping Mechanisms	References
Neuroendocrine regulation	Infertile women with PCOS exhibit reduced serum levels of 5-HT and 5-HIAA, which are associated with altered LH secretion.	Women with HG exhibit elevated plasma 5-HT levels that correlate with PUQE scores. 5-HT may exert emetic effects through 5-HT3 receptors, vagal afferent fibres, and brainstem emetic pathways.	Pre-existing alterations in 5-HT regulation associated with PCOS may lower the response threshold for pregnancy-related emetic signals, resulting in a theoretical process of 5-HT sensitisation.	PCOS:28 NVP:30–32
Placental-derived factors	GDF15: Women with PCOS may exhibit altered GDF15 homeostasis. IGFBP7: Increased IGFBP7 expression is associated with insulin resistance, inflammation, and ovarian dysfunction.	GDF15: Circulating fetoplacental GDF15 levels are elevated in women with NVP, and its effects may be mediated through the brainstem GFRAL–RET signalling pathway. IGFBP7: In early pregnancy, women with HG exhibit elevated IGFBP7 levels associated with symptom severity; additionally, the 4q12 locus harbouring IGFBP7 has been linked to genetic susceptibility to HG.	Pre-existing metabolic abnormalities in PCOS may influence maternal sensitivity to gestational changes in GDF15 and IGFBP7, thereby altering the threshold for emetic responses.	PCOS:35–39,43–45 NVP:40–42,46,47
Appetite and energy balance	Ghrelin: Women with PCOS commonly exhibit reduced circulating ghrelin levels and blunted postprandial ghrelin suppression; these abnormalities are associated with insulin resistance and hyperandrogenism. Leptin: Women with PCOS often exhibit elevated leptin levels, which are associated with obesity, insulin resistance, and altered androgen-related indices.	Ghrelin: Women with HG may exhibit reduced ghrelin levels and elevated PYY and PP levels. Leptin: In women with NVP, the increase in leptin levels from early to mid-pregnancy is attenuated.	Abnormalities in appetite-regulating hormones and energy homeostasis in both conditions may impair metabolic adaptation during early pregnancy and influence appetite, gastrointestinal motility, and susceptibility to NVP.	PCOS:49–55,57–60 NVP:24,25,61
Gut microbiota–endocrine interactions	Women with PCOS may exhibit reduced gut microbial α-diversity and altered β-diversity, accompanied by an increased abundance of Gram-negative bacteria, a reduction in SCFA-producing bacteria, and LPS–TLR4-related chronic low-grade inflammation.	Alterations in gut microbial diversity and composition have been observed in women with NVP, and specific microbial taxa are associated with the risk of HG. However, severe vomiting and nutritional deprivation may also induce secondary alterations in the gut microbiota.	Shared gut microbial dysbiosis may influence NVP through the LPS–TLR4 inflammatory pathway and the SCFA–gut hormone–gut–brain axis; however, the causal direction remains unclear.	PCOS:63–67 NVP:68–71
Androgen signalling	Hyperandrogenism is a major feature of PCOS and is accompanied by an insulin–androgen feedback loop and possible alterations in monoamine neurotransmitters.	Gestational levels of A4, DHEAS, T, and the FAI are associated with NVP symptoms or vomiting scores. Women with HG may also exhibit elevated androstanediol glucuronide levels and reduced A4 levels.	Pre-existing hyperandrogenism and metabolic abnormalities in PCOS may affect neuroendocrine and metabolic adaptation during pregnancy, thereby increasing susceptibility to NVP.	PCOS:73,74 NVP:75,76

Abbreviations: PCOS, polycystic ovary syndrome; NVP, nausea and vomiting of pregnancy; HG, hyperemesis gravidarum; 5-HT, 5-hydroxytryptamine; 5-HIAA, 5-hydroxyindoleacetic acid; LH, luteinising hormone; PUQE, Pregnancy-Unique Quantification of Emesis; GDF15, growth differentiation factor 15; IGFBP7, insulin-like growth factor-binding protein 7; GFRAL, glial cell line-derived neurotrophic factor family receptor alpha-like; RET, rearranged during transfection receptor tyrosine kinase; PYY, peptide YY; PP, pancreatic polypeptide; SCFA, short-chain fatty acids; LPS, lipopolysaccharide; TLR4, Toll-like receptor 4; A4, androstenedione; DHEAS, dehydroepiandrosterone sulfate; T, testosterone; FAI, free androgen index.

Strengths and Limitations

This review synthesises potential mechanistic links between PCOS and NVP across multiple biological domains, including neuroendocrine regulation, placental-derived factors, appetite and energy metabolism, gut microbiota, and androgen-related pathways. Whereas previous studies have generally focused on individual hormones or isolated pathways, integrating evidence across these systems provides a broader perspective on the potential relationship between the two conditions. Pre-existing endocrine and metabolic disturbances in women with PCOS may alter their adaptation to the rapid physiological changes of early pregnancy and thereby contribute to greater susceptibility to or severity of NVP. Interactions among metabolic, inflammatory, neurotransmitter, and microbiota-related factors may further modify this response. By bringing these pathways together, this review provides a conceptual basis for understanding the potential association between PCOS and NVP and for guiding future research.

Several limitations should nevertheless be acknowledged. First, direct evidence linking PCOS to NVP remains limited, and much of the proposed mechanistic framework is based on evidence derived separately from PCOS and NVP populations rather than from studies directly examining their interaction. Second, many of the available studies are cross-sectional or observational, limiting causal inference and leaving the interactions among different pathways incompletely understood. Third, heterogeneity in study populations, diagnostic criteria, and measurement methods may contribute to inconsistencies across studies. Although proposed mechanisms involving the gut microbiota–brain axis and GDF15-related signalling are biologically plausible, their specific roles in linking PCOS to NVP have not been validated in prospective longitudinal or integrative multi-omics studies. Future research should therefore incorporate prospective designs, direct mechanistic investigations, and multi-omics approaches to clarify temporal relationships, causal pathways, and interactions among the proposed mechanisms.

Conclusion

This review summarises the possible associations between PCOS and NVP from endocrine and metabolic perspectives. Available evidence suggests that the two conditions share alterations across multiple biological systems, including neuroendocrine regulation, placental factors, energy metabolism, gut microbiota, and androgen-related pathways. In PCOS, longstanding hormonal and metabolic disturbances may alter maternal responses to early pregnancy and thereby increase the likelihood and severity of NVP. Overall, the endocrine and metabolic features associated with PCOS may contribute to the development and progression of NVP. However, because most of the supporting evidence is derived from indirect inference, these pathways should currently be regarded as hypothesis-generating theoretical mechanisms rather than established causal mechanisms.

Accordingly, the clinical and translational relevance of these potential mechanisms requires further validation. Future studies should systematically evaluate the relationships of the candidate pathways and related markers discussed above with the occurrence and severity of NVP within the broader endocrine and metabolic context of PCOS. Factors with independent predictive value could then be identified to provide a basis for future risk stratification. Studies should also determine whether preconception optimisation of modifiable endocrine and metabolic abnormalities reduces the risk or severity of NVP. Medication regimens should be assessed individually before conception and during early pregnancy, with attention to gastrointestinal adverse effects that may overlap with NVP symptoms and the need for timely evidence-based treatment after symptom onset. Targeted clinical surveillance of pregnant women with PCOS could incorporate both their pre-existing endocrine and metabolic abnormalities and the progression of NVP symptoms, with particular attention to glycaemic status, symptom severity, changes in body weight and nutritional intake, and, when symptoms worsen, hydration and electrolyte status. This approach may facilitate the early identification of women who are at risk of progressing to severe NVP or HG. Prospective longitudinal studies extending from preconception through the first trimester are particularly needed to characterise changes in neuroendocrine markers such as 5-HT, placental-related factors such as GDF15 and IGFBP7, hormones involved in appetite and energy balance, gut microbial features, and androgen-related indices, thereby clarifying their temporal patterns, predictive value, and potential clinical applicability.

Funding Statement

(1) The National key R&D Program of China (2019YFC1709500); (2) The National Collaboration Project of Critical Illness by Integrating Chinese Medicine and Western Medicine; (3) Project of Heilongjiang Province Innovation Team “TouYan” (LH2019H046); (4) Heilongjiang Provincial Clinical Research Centre for Ovary Diseases (LC2020R009); (5) Traditional Chinese Medicine Research Project of Heilongjiang Administration of Traditional Chinese Medicine (ZHY2022-124); (6) The project of Evidence-based capacity in Traditional Chinese Medicine (TCM Sci-Tech Internal Letter [2023] No. 24).

Data Sharing Statement

Data sharing does not apply to this article as no new datasets were generated or analysed in this study.

Ethics Statement

Ethical review and approval were not required for this study because it is a review article based on previously published data.

Author Contributions

Zhuwei Gao and Jiaxing Feng contributed to the work equally and should be regarded as co-first authors. All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare no conflicts of interest.