Association of iron nutritional homeostasis and hepcidin gene polymorphisms with gestational diabetes mellitus

Abstract

Objective

This study aimed to investigate the relationship between iron homeostasis and gestational diabetes mellitus (GDM), focusing on hepcidin levels and its gene polymorphisms.

Methods

In this cross-sectional comparative study, a total of 166 pregnant women (69 with GDM and 97 with normal glucose tolerance) were enrolled. Clinical parameters and six hepcidin gene polymorphisms were analyzed. Statistical analyses included Spearman correlation, logistic regression, and parallel genotype-phenotype assessments in both groups with correction for multiple comparisons.

Results

Hepcidin, fasting, and postprandial glucose levels were significantly higher in the GDM group (p < 0.05). Genotype frequencies did not differ between groups. Within the GDM group, specific polymorphisms (rs10416533, rs7251432, rs55863037) showed nominal associations with ferritin, hemoglobin, and glucose levels; however, these did not survive strict correction for multiple testing. No such associations were observed in the control group.

Conclusion

Elevated hepcidin is confirmed in GDM. While certain hepcidin gene variants showed disease-specific, this study had limited statistical power to detect small to moderate genetic effects due to its modest sample size. The study underscores a potential link between iron homeostasis and GDM, highlighting hepcidin as a candidate for further mechanistic investigation. these preliminary and exploratory findings require validation in larger cohorts. These findings underscore the involvement of hepcidin in GDM-associated iron dysregulation and highlight it as a key candidate for future mechanistic and longitudinal studies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-026-08870-3.

Introduction

Gestational diabetes mellitus (GDM) is a metabolic disorder that adversely affects glucose tolerance during pregnancy. This constitutes an increasingly serious public health concern. In recent decades, the prevalence of GDM among Asian populations has ranged from 0.7% to 51% [1, 2, 3].A meta-analysis of 25 studies involving 79,064 Chinese pregnant women conducted in 2019 revealed that the prevalence of gestational diabetes mellitus (GDM) in China, according to the diagnostic criteria established by the International Association of Diabetes and Pregnancy Study Groups (IADPSG), has reached an alarming rate of 14.8%. This figure positions China at the forefront globally in terms of total GDM cases [4].Differences in race, diagnostic criteria, screening protocols, and population characteristics may contribute to significant disparities in prevalence, Women diagnosed with gestational diabetes mellitus (GDM) are at an increased risk of experiencing a range of adverse perinatal outcomes. These may include ketoacidosis, preeclampsia, polyhydramnios, macrosomia, urinary and genital tract infections, congenital malformations in the fetus, difficult labor, postpartum infections, neonatal hypoglycemia, respiratory distress syndrome, neonatal hypocalcemia, and hyperbilirubinemia [5, 6]. Furthermore, the progeny of patients with gestational diabetes mellitus (GDM) are at an elevated risk for developing insulin resistance, obesity, and diabetes [7, 8].

Research has indicated that iron homeostasis is linked to insulin resistance in women with gestational diabetes mellitus (GDM). This suggests that alterations in iron homeostasis may play a role in the pathogenesis of GDM by exacerbating stress-induced hyperglycemia [9]. Iron possesses the potential for toxicity, as it catalyzes the generation of reactive oxygen species. This process leads to oxidative stress reactions that result in damage to beta cells and functional impairment. Additionally, iron influences hypoxia-inducible factors, which can downregulate glucose transporters and diminish glucose tolerance [10]. Hepcidin is a peptide composed of 25 amino acids, synthesized by the liver and excreted by the kidneys [11], Hepcidin is a key regulator of iron homeostasis and controls the expression of iron transporters on the surfaces of intestinal cells and macrophages [12]. Hepcidin blocks iron from entering the bloodstream by binding iron transporter proteins on the plasma membranes of cells in various tissues and inducing their internalization and degradation [13]. Studies suggest that women with GDM have significantly elevated levels of hepcidin [14], Some polymorphisms in the promoter region of the hepcidin encoding gene (HAMP) are related to the low expression of this hormone, thereby increasing the level of serum iron [15, 16]. The pathogenesis of gestational diabetes mellitus (GDM) remains undetermined at present. Previous studies have proposed that the pathogenesis of GDM may be associated with genetic factors [17].

We investigate the correlation between polymorphisms in the hepcidin gene and gestational diabetes mellitus by developing a custom single nucleotide polymorphism (SNP) array that encompasses eight specific hepcidin gene variants. Utilizing these polymorphisms, we assess the risk of iron deficiency during pregnancy and have submitted a patent application for this innovation (Patent application number: 202411649734.2).The identified genetic polymorphisms are as follows: rs117345431, rs149358257, rs10416533, rs10421768, rs7251432, rs8101606, rs55863037, and rs2293689.Based on the aforementioned background, this study aims to investigate the correlation between iron nutritional homeostasis and the incidence of gestational diabetes mellitus (GDM), as well as to explore the role of hepcidin gene polymorphisms in the onset and progression of GDM. To our knowledge, this is among the first studies to comprehensively explore the association between hepcidin gene polymorphisms and both the occurrence and progression of GDM.

Materials and methods

Study participants

A cross-sectional comparative study was conducted among pregnant women at 22–28 weeks gestation, recruiting those with and without GDM between January and April 2024, with a total of 186 women initially enrolled. Following the application of inclusion and exclusion criteria, 20 cases were excluded (e.g., due to transfer to other hospitals), resulting in a final study population of 166 participants, comprising 69 women diagnosed with gestational diabetes mellitus (GDM) and 97 normoglycemic pregnant women. Biochemical and genetic data were derived from routine prenatal testing samples collected at 22–28 weeks gestation, with all clinical records systematically reviewed between May and September 2024.

This study received approval from the Ethics Review Committee of Xiamen University Affiliated Women’s and Children’s Hospital (approval number: KY-2023-067-H01) and adhered to the principles outlined in the Helsinki Declaration. Written informed consent was obtained from all participants prior to their inclusion in the study.During data collection, researchers temporarily accessed identifiable information to link clinical data. However, all direct identifiers (names, IDs) were immediately removed after data verification, and anonymized codes were used for analysis. The original identifiers were stored securely and accessible only to the ethics committee.

All subjects had no prior history of other types of diabetes. Each pregnant participant was regularly monitored and evaluated by obstetrician-gynecologists from the Obstetrics and Gynecology Outpatient Department at Xiamen University Affiliated Women and Children’s Hospital. According to the Guidelines for the Diagnosis and Treatment of Pregnancy Complicated by Diabetes (2014) [18], the diagnostic standard for a 75 g Oral Glucose Tolerance Test (OGTT) is recommended as follows: blood glucose levels should be measured before sugar intake, and at 1 h and 2 h post-ingestion. The three blood glucose values must be below 5.1, 10.0, and 8.5 mmol/L (92, 180, and 153 mg/dL), respectively. Any blood glucose level that meets or exceeds these thresholds is diagnosed as Gestational Diabetes Mellitus (GDM). Exclusion criteria include patients with a history of kidney disease, liver disease, multiple pregnancies, iron deficiency anemia, maternal age under 15 years, fetal congenital or chromosomal abnormalities, family history of diabetes mellitus, chronic hypertension, preeclampsia, acute or chronic inflammatory conditions or infectious diseases.Our study population has not received treatment with iron supplements, folate supplements, vitamin B12 supplements or any known medications that may influence carbohydrate metabolism; furthermore all participants are non-smokers.The study design and participant flow are summarized in Supplementary Figure S1.

Clinical biochemical index testing

Blood plasma and whole blood samples are collected from the vein between 22 and 28 weeks of gestation, coinciding with the period designated for glucose tolerance testing in pregnant women. These samples are then placed separately into serum separation tubes and EDTA tubes, after which they are stored at -80 °C for subsequent analysis.The Mindray BC-5390CRP blood analyzer (Mindray Diagnostic Kit, Shenzhen, China) is utilized to assess whole blood hemoglobin levels. The Bio Rad hemoglobin detection system measures the concentration of glycated hemoglobin in whole blood. Serum ferritin levels (ng/mL) are determined using the COBAS 8000 automatic analyzer (Roche, Germany), employing electrochemiluminescence technology. Additionally, serum glucose levels (mmol/L) are measured using the hexokinase method. For hepcidin quantification, a commercially available enzyme-linked immunosorbent assay (ELISA) kit was employed (BBI Life Sciences Co., Ltd., China; Product No.: D711417), with inter-assay and intra-assay coefficients of variation both below 10%.

DNA isolation and genotyping utilizing matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF-MS). After thawing the frozen EDTA anticoagulant venous blood at room temperature, genomic DNA extraction is performed using the Ezup column animal tissue genomic DNA extraction kit (BioNTech Co., Ltd.). Specific steps should be followed as outlined in the reagent manual. For electrophoresis detection, 5 µl of the DNA solution is mixed with 1% agarose and 1X TAE buffer solution, and subjected to a voltage range of 120–180 V. The presence of a single band indicates that the DNA is intact and not degraded, while a clear band suggests that the concentration meets PCR requirements. To assess concentration and purity, a spectrophotometer is utilized; specifically, 1 µl of the sample is measured for its OD value. An OD ratio of 260/280 between 1.7 and 2.0 signifies good DNA quality, with protein contamination indicated by values below 1.7 and RNA contamination suggested by values above 2.0. Generally speaking, minor levels of protein or RNA contamination do not adversely affect standard PCR procedures.For gene typing utilizing the MassARRAY^® MALDI-TOF System, detailed protocols and primer sequences can be referenced in previously filed patents (Patent application number: 202411649734.2) .

Statistical analysis

The data were analyzed using SPSS statistical software version 23. A p-value of less than 0.05 was considered statistically significant. Student’s t-test was employed to compare age, fasting plasma glucose (FPG), oral glucose tolerance test (OGT), ferritin, hemoglobin levels, and hepcidin levels between the gestational diabetes mellitus (GDM) group and the control group. Numerical variables are presented as mean ± standard deviation (mean ± SD). The Hardy-Weinberg equilibrium was assessed using the chi-square test in both case and control groups. The frequencies and genotypes of the candidate alleles were compared utilizing either the chi-square test or Fisher’s exact test. Conversely, categorical variables are expressed as counts (n) and percentages (%). Odds ratios (OR) along with 95% confidence intervals (CI) were calculated through logistic regression analysis with age included as a covariate. The Mann-Whitney U test was applied to compare the differences in ferritin, hemoglobin levels, hepcidin levels, and HbA1c levels among GDM patients categorized by different candidate genotypes. Given the exploratory nature of the genotype-phenotype analyses and the multiple comparisons performed, a Bonferroni correction was applied to adjust the significance level for all within-group pairwise genotype comparisons for each SNP. The corrected p-value is reported as p-Bonferroni.We could not calculate a priori sample size because there is currently no consensus on the hepcidin reference intervals in normal pregnancies, and women with diabetes.

Results

Demographic and clinical characteristics of the study subjects

We selected 186 pregnant women aged between 22 and 28 weeks who received regular prenatal care at Xiamen Maternal and Child Health Hospital from January to April 2024. Following the established inclusion and exclusion criteria, 20 cases were excluded due to transfers to other hospitals, resulting in a final study population of 166 pregnant women. Based on the Oral Glucose Tolerance Test (OGTT), participants were categorized into two groups: the Gestational Diabetes Mellitus (GDM) group, which comprised 69 cases, and the non-GDM group with 97 cases. A comparative analysis of general information and routine biochemical indicators was conducted for both groups. No significant differences were observed in age, ferritin levels, or hemoglobin concentrations between the two groups (P > 0.05). However, statistically significant differences were identified in hepcidin levels, fasting plasma glucose (FPG), and postprandial blood glucose (PBG) between the two groups (P < 0.05), as detailed in Table 1.

Table 1.

Demographic and clinical characteristics of the study subjects

Participant Characteristics	Non-GDM	GDM	p
Number	97	69
Age (years)	30.9381 ± 3.9206	30.913 ± 3.48838	0.966
Ferritin(ng/mL )	38.067 ± 25.65503	38.1943 ± 22.87633	0.974
Hemoglobin(g/L)	118.6701 ± 8.13393	120.1884 ± 9.59055	0.273
Hepcidin(ng/mL)	63.5116 ± 40.20717	86.2828 ± 56.54005	0.005*
Fasting blood glucose (mmol/L)	4.3597 ± 0.36235	5.0165 ± 0.77631	< 0.001*
Blood glucose (one hour) (mmol/L)	7.7329 ± 1.36231	10.5662 ± 1.31594	< 0.001*
Blood glucose (two hour)(mmol/L)	6.7012 ± 1.27226	9.2207 ± 1.51126	< 0.001*

Non-GDM, normal pregnancy controls; and GDM, gestational diabetes

P values, t-test independent variables

*:P<0.05 was considered to be statistically significant

The genetic phenotype is correlated with gestational diabetes

This study successfully genotyped eight polymorphic loci. Two SNPs (rs117345431 and rs149358257) were monomorphic (wild-type) in all 166 specimens and were excluded from further association analyses. The six genotypes included in the statistical analysis were as follows: rs10416533 (T > C), rs10421768 (A > G), rs7251432 (A > G), rs8101606 (A > C), rs55863037 (G > A), and rs2293689 (C > T). The results indicated that all SNPs conformed to Hardy-Weinberg equilibrium (HWE) with a P-value greater than 0.05, demonstrating the absence of selective bias, population stratification, or genotyping errors within the study population [19]. As presented in Table 2, there was no statistically significant difference in the genotype frequencies of the six gene polymorphisms between individuals with gestational diabetes mellitus (GDM) and those without (non-GDM) (P > 0.05). Figure 1 illustrates the genotype frequency distribution.

Table 2.

Differential analysis of gene polymorphism between GDM group and normal pregnancy group

variant	Genotype	Non-GDM	GDM	OR(95%CI)	P	OR 95%CI*	P*	HWE
rs10416533	TT	71	49	-	-		0.897	0.257
	CT	23	17	1.071(0.519,2.211)	0.853	0.687(0.132,3.568)	0.655
	CC	3	3	1.449(0.281,7.479)	0.658	0.734(0.129,4.158)	0.726
rs10421768	AA	95	68	-	-			0.906
	AG	2	1	0.699(0.062,7.86)	0.771	1.429(0.126,16.218)	0.773
	GG	0	0	-	-	-	-
rs7251432	AA	32	26	-	-	-	0.813	0.355
	AG	51	34	0.821(0.418,1.612)	0.566	1.276(0.473,3.441)	0.631
	GG	14	9	0.791(0.296,2.117)	0.641	1.039(0.404,2.669)	0.937
rs8101606	AA	95	68	-	-	-		0.906
	AC	2	1	0.699(0.062,7.86)	0.771	1.429(0.126,16.218)	0.773
	CC	0	0	-	-	-	-
rs55863037	GG	53	40	-	-	-	0.615	0.297
	AG	37	22	0.788(0.404,1.537)	0.484	0.751(0.244,2.317)	0.619
	AA	7	7	1.325(0.43,4.082)	0.624	0.585(0.179,1.909)	0.374
rs2293689	CC	76	57	-	-	-	0.858	0.532
	CT	20	12	0.8(0.362,1.769)	0.582	0.799(0.361,1.768)	0.58
	TT	1	0	0	1	-	1

*The adjusted statistical data is calculated using age adjusted logistic regression

Fig. 1.

Analysis of genetic polymorphisms in the gestational diabetes mellitus group and the normal pregnancy group

The bar chart compares the genotype frequency distribution of six hepcidin gene polymorphisms (rs10416533, rs10421768, rs7251432, rs8101606, rs55863037, rs2293689) between women with gestational diabetes mellitus (GDM) and normal glucose-tolerant controls (Non-GDM). No statistically significant differences in genotype frequencies were observed between the two groups for any of the tested single nucleotide polymorphisms (SNPs).

Correlation analysis between hepcidin gene polymorphisms and clinical indicators

Given the lack of association between SNPs and GDM risk, we performed exploratory analyses to assess whether specific genotypes were associated with variations in clinical indicators within each group.

In the GDM group, three SNPs (rs10416533, rs7251432, rs55863037) showed nominal associations (uncorrected P < 0.05) with certain indicators (e.g., ferritin, hemoglobin, 1-hour glucose) (Fig. 2). However, none of these pairwise comparisons remained statistically significant after applying a strict Bonferroni correction for multiple testing (all adjusted P > 0.05). Detailed results of these genotype-phenotype associations and pairwise comparisons are provided in Supplementary Table S1-S2.

Fig. 2.

Correlation Analysis between Genotypes and Laboratory Indicators in Patients with Gestational Diabetes Mellitus (GDM). A Box plot showing serum ferritin levels (ng/mL) across different genotypes of rs10416533 (TT, CT, CC) within the GDM cohort. B-1, B-2 Box plots illustrating (B-1) serum ferritin levels (ng/mL) and (B-2) hemoglobin levels (g/L) across genotypes of rs7251432 (AA, AG, GG) in GDM patients. C-1, C-2, C-3 Box plots displaying (C-1) one-hour postprandial blood glucose (mmol/L), (C-2) serum ferritin (ng/mL), and (C-3) hemoglobin (g/L) levels across genotypes of rs55863037 (GG, AG, AA) in GDM patients.P values of < 0.05 were considered to denote nominal statistical significance prior to correction for multiple comparisons

In the non-GDM control group, identical analyses revealed no statistically significant associations (all P > 0.05) between any of the six SNPs and the measured clinical indicators (Supplementary Table S3).

Discussion

During pregnancy, iron requirements increase substantially due to expanded maternal blood volume and fetal-placental demands [20]. The relationship between iron status and gestational diabetes mellitus (GDM) remains inconsistent across studies [21–23].Although serum ferritin is commonly used to assess iron stores, it is an acute-phase reactant that may be elevated in response to inflammation, limiting its reliability [14, 22].In contrast, hepcidin, the central regulator of systemic iron homeostasis, has emerged as a more specific biomarker of iron status [24].It controls iron absorption, recycling, and mobilization by inhibiting ferroportin, and plays a key role in placental iron transfer during pregnancy [25, 26]. Previous studies have reported elevated hepcidin levels in women with GDM compared to those with normal glucose tolerance [27]. and higher iron status has been associated with greater GDM risk, a status often marked by increased hepcidin and ferritin [28].Consistent with prior research, we confirmed elevated hepcidin in GDM patients compared to controls, while ferritin levels showed no significant intergroup difference. These findings support a link between altered iron homeostasis and GDM, warranting careful clinical monitoring of iron status.

Hepcidin gene polymorphisms: linking genetic variation to phenotypic heterogeneity in GDM

Hepcidin expression is modulated by polymorphisms in its encoding gene (HAMP), with specific variants associated with susceptibility to iron metabolism disorders [29, 30]. For instance, certain HAMP promoter variants are linked to the severity of iron overload in hereditary hemochromatosis 32. This study provides, to our knowledge, a systematic investigation into the association between HAMP gene polymorphisms and GDM. We found no significant difference in the genotype frequencies of the six investigated SNPs between the GDM and control groups, suggesting that these specific variants may not be major genetic risk factors for GDM development. This observation aligns with the multifactorial and polygenic nature of GDM, where genetic susceptibility is likely subtle and heterogeneous.

We next performed exploratory analyses to assess whether specific HAMP variants were associated with phenotypic variation within the GDM cohort. Prior to multiple testing correction, three SNPs (rs10416533, rs7251432, rs55863037) exhibited nominally significant associations with parameters of iron and glucose metabolism. For example, uncorrected analyses suggested that carriers of the rs7251432 GG genotype had nominally higher hemoglobin levels, while rs55863037 AG carriers showed a distinct metabolic profile (elevated 1-hour postprandial glucose but lower ferritin and hemoglobin).

These phenotypic links are biologically plausible: hepcidin indirectly influences hemoglobin synthesis and erythropoiesis by modulating plasma iron availability [31]. while iron homeostasis imbalance is closely tied to insulin resistance and glucose intolerance, with animal models showing that high-iron diets exacerbate stress-induced hyperglycemia and hepatic oxidative damage [27, 32]. However, it is critical to emphasize that these observations derive from uncorrected comparisons and do not constitute evidence of true associations. Critically, these patterns were absent in the non-GDM group and did not survive Bonferroni correction. This pattern, if validated in future adequately powered studies, might suggest that the phenotypic influence of these variants becomes discernible only within a dysmetabolic environment. However, given the exploratory nature of these analyses and the lack of statistical significance after correction, such speculation remains hypothetical. In other words, if confirmed in future studies, these HAMP variants might hypothetically act as phenotypic modifiers rather than primary risk alleles, influencing metabolic and iron-related traits in the context of established GDM. The context-dependent influence of genetic variants is evident in other metabolic disorders, as shown by the association between RETN gene polymorphisms and Polycystic Ovary Syndrome (PCOS) in an Iranian population. Given that PCOS shares insulin resistance with GDM, this example highlights how genetic background can modulate disease expression within specific ethnic contexts [33]. This parallel underscores the importance of considering population-specific genetic architecture when interpreting the potential modifier role of HAMP variants in GDM.The observation that the rs7251432 GG genotype was nominally associated with higher hemoglobin levels only in the GDM group, for instance, warrants further investigation to determine whether this variant might influence anemia risk or iron utilization in diabetic pregnancy, a hypothesis that requires rigorous testing in independent cohorts.

Main findings

This study confirms elevated hepcidin levels in women with gestational diabetes mellitus (GDM) and reveals a complex relationship with its genetic underpinnings. The investigated hepcidin gene single nucleotide polymorphisms (SNPs) were not associated with the risk of developing GDM per se. However, in uncorrected exploratory analyses within the GDM cohort, specific variants (e.g., rs7251432, rs55863037) showed nominal associations with iron and glucose parameters. These trends were absent in controls and did not survive multiple testing correction. This pattern, if confirmed in larger studies, would suggest that these variants are not primary risk factors for GDM but might, hypothetically, modulate phenotypic expression once hyperglycemia is established. Such interpretation remains speculative at this stage.

Clinical implications and significance

The findings carry several clinically relevant implications. First, hepcidin may serve as a potential biomarker for dysregulated iron metabolism within the context of GDM, a hypothesis that warrants further longitudinal investigation to establish its predictive or diagnostic utility.Second, genetic background may contribute to the phenotypic heterogeneity observed in GDM. For instance, the trend linking the rs7251432 GG genotype to higher hemoglobin levels specifically in GDM patients suggests that genetic factors could partially explain variations in anemia risk or iron utilization efficiency within this population, providing a rationale for further research into personalized antenatal care strategies.

Limitations and future directions

This study has several limitations. First and foremost, the modest sample size, particularly for stratified genetic analyses, limits statistical power. The study was not powered to detect small to moderate genetic effects, and the very small number of minor homozygotes for some SNPs (e.g., only 3 carriers of the rs10416533 CC genotype in the GDM group) substantially limits the reliability and interpretability of genotype-phenotype comparisons involving these variants. Consequently, the negative association findings between the SNPs and GDM risk, as well as the nonsignificant genotype-phenotype associations after correction, should be interpreted with caution. The exploratory within-GDM associations, while biologically plausible, require confirmation in larger, adequately powered cohorts. Second, its single-center, cross-sectional design limits causal inference. Third, although multiple comparison corrections were applied, the observed genotype-phenotype associations remain exploratory. Furthermore, unmeasured confounders, such as detailed dietary iron intake and systemic inflammation markers, could introduce residual bias.

Future research should focus on: (1) validating the potential modifier effects of these genetic variants on GDM phenotypes in larger, multi-center, prospective cohorts with adequate power; (2) incorporating comprehensive lifestyle and dietary data to better assess potential confounders; and (3) elucidating the specific molecular mechanisms through which these variants might influence hepcidin biology and the iron-glucose axis using cellular or animal models.

Conclusion

In summary, this study reinforces the central role of hepcidin in the dysregulation of iron metabolism in GDM. It provides novel, preliminary, and exploratory evidence generated from a cohort of modest size. In exploratory analyses, specific hepcidin gene variants showed nominal associations with metabolic phenotypes in women with established GDM, a finding that requires validation in independent, statistically powered cohorts before any definitive conclusions can be drawn.These findings contribute to the understanding of the “iron-glucose axis” in GDM pathogenesis and lay a theoretical foundation for future research. They also underscore the importance of a comprehensive assessment of iron homeostasis in the clinical management of GDM. However, these potential associations, along with the negative genetic association results, require validation in larger, statistically powered studies.

Authors’ contributions

Juan Li, Xiaoqian Su and Wei Wei contributed equally to this work. Huiming Ye and Hongyi Yang conceived and designed the study. Yuzhan Xu, Yan Ni, Danzeng Baimu and Honghong Yan contributed to data collection, analysis, and interpretation. Juan Li, Xiaoqian Su and Wei Wei drafted the manuscript. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the Major Science and Technology Project of Fujian Provincial Health Commission (2021ZD01006, founded by Xiamen Municipal Health Commission), the Natural Science Foundation of Xiamen, China (No. 3502Z20227139), and the Xiamen Major Science and Technology Program (Milestone Project)(No. 3502Z20254028).

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the Ethics Committee of Women and Children’s Hospital of Xiamen University China (approval number: KY-2023-067-H01). All participating patients signed written informed consent. The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Competing interests

The authors declare no competing interests.