Maternal serum adropin as a potential biomarker for placenta accreta spectrum: a cross-sectional study Authors: Seray Sırkıntı, Gulsum Uysal, Sule Gül Aydın, Merve Vay Bedir Published: 2026 ABSTRACT BackgroundAdropin is a regulatory peptide implicated in endothelial homeostasis, and reduced circulating levels have been linked to endothelial dysfunction and impaired vascular remodeling. This study aimed to compare maternal serum adropin levels in pregnancies complicated by placenta previa and/or placenta accreta spectrum (PAS) with those of healthy pregnant women and to evaluate the discriminative performance of adropin for abnormal placental invasion.MethodsThis cross-sectional study included 84 singleton pregnancies at ≥ 34 weeks of gestation: 42 women diagnosed with placenta previa and/or PAS and 42 healthy pregnant women who underwent cesarean delivery. Maternal preoperative serum adropin concentrations were quantified using a commercially available enzyme-linked immunosorbent assay (ELISA). Demographic, obstetric, ultrasonographic, and intraoperative findings were recorded. Receiver operating characteristic (ROC) curve analysis was performed to assess the discriminative ability of serum adropin levels for abnormal placental invasion.ResultsA total of 84 pregnant women were included in the study: 42 healthy controls, 30 patients with placenta previa without PAS, and 12 patients diagnosed with PAS.Maternal serum adropin concentrations were significantly reduced in pregnancies complicated by placenta previa and/or PAS compared with healthy controls (p < 0.001). No significant differences were observed between groups in terms of body mass index or routine laboratory parameters. ROC analysis demonstrated moderate discriminative performance, with an area under the curve of 0.778 (p < 0.001). An adropin cut-off value of ≤431.27 pg/mL yielded a sensitivity of 71.4% and a specificity of 78.6%.ConclusionMaternal serum adropin levels are significantly reduced in pregnancies complicated by placenta previa and/or PAS. Decreased adropin concentrations may reflect a disturbed metabolic–endothelial milieu associated with abnormal placental invasion. Adropin may serve as a supportive biomarker; however, findings should be interpreted cautiously given the cross-sectional design. Further prospective studies are required to clarify its potential clinical utility.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12884-026-09344-2. Introduction Placenta previa refers to placental implantation within the lower uterine segment, resulting in partial or complete coverage of the internal cervical os. It is a major cause of third-trimester vaginal bleeding and is linked to significant maternal and perinatal morbidity and mortality [1]. Another major obstetric complication is PAS which refers to abnormal placental adherence or invasion into the uterine wall due to partial or complete absence of the decidua basalis. According to the depth of invasion, PAS is classified as placenta accreta, increta, or percreta. The coexistence of placenta previa and PAS is associated with particularly severe maternal and perinatal outcomes [2]. Reducing morbidity and mortality in pregnancies complicated by placenta previa and PAS requires timely and accurate prenatal diagnosis, close surveillance, appropriate surgical planning, and a multidisciplinary management approach [3]. Moreover, increasing depth of placental invasion further complicates surgical management and significantly elevates the risk of severe maternal morbidity, including massive hemorrhage and peripartum hysterectomy [4]. Therefore, effective prenatal risk stratification and the identification of reliable biomarkers in pregnancies complicated by placenta previa with suspected PAS may facilitate safer obstetric management and improve clinical outcomes. In recent years, the peptide hormone adropin has attracted increasing attention due to its potential role in various systemic diseases. Adropin is a recently identified peptide encoded by the energy homeostasis–associated (ENHO) gene that plays a regulatory role in energy balance, inflammatory pathways, and endothelial function. It was first described by Kumar et al. in 2008 and is primarily expressed in the liver, brain, endocrine tissues, and cardiac muscle [5, 6]. Previous studies have demonstrated significantly reduced circulating adropin levels in conditions such as hypertension, gestational diabetes mellitus, polycystic ovary syndrome, metabolic syndrome, and cardiovascular diseases [5, 7–10]. In contrast, elevated adropin levels have been reported in patients with heart failure [11]. In addition, decreased adropin concentrations have been observed in obstetric disorders characterized by endothelial dysfunction and impaired placental perfusion, including preeclampsia and intrauterine growth restriction [12, 13]. Given the established involvement of adropin in endothelial homeostasis and inflammatory pathways, its potential role in abnormal placental implantation and invasion warrants investigation. The aim of the present study was to evaluate maternal serum adropin levels in pregnancies complicated by placenta previa and placenta accreta spectrum. To our knowledge, no previous study has evaluated the relationship between maternal serum adropin levels and placenta previa or PAS. Our findings offer new perspectives on the potential pathophysiological mechanisms involved in abnormal placental invasion. Methods Study design and population This cross-sectional study was conducted at the Department of Obstetrics and Gynecology, Adana City Training and Research Hospital, University of Health Sciences, between August 1, 2024 and June 1, 2025. The study aimed to evaluate and compare maternal serum adropin levels in singleton pregnancies at ≥ 34 weeks of gestation diagnosed with placenta previa and/or PAS. As no previous studies have evaluated serum adropin levels in placenta previa or PAS, effect size estimates were derived from a study assessing serum adropin concentrations in endometrial cancer. An a priori sample size calculation using G*Power (version 3.1.9.7) indicated that a minimum of 42 participants per group (total n = 84) would provide 80% power at a two-sided α level of 0.05 [14]. Patient selection Pregnant women aged 18–45 years in the reproductive period with singleton, viable third-trimester pregnancies diagnosed with placenta previa and/or PAS were included in the study. The control cohort comprised healthy third-trimester pregnant women with comparable demographic profiles who underwent elective cesarean section and had no obstetric or systemic comorbid conditions. Exclusion criteria included uterine anomalies, chronic systemic diseases, multiple pregnancies, and other obstetric complications such as preterm premature rupture of membranes, threatened preterm labor, fetuses followed for intrauterine growth restriction, and placental abruption. Patients with conditions that could affect adropin levels, including hypertensive disorders of pregnancy, gestational diabetes mellitus, fetal growth restriction, and other placental pathologies, were also excluded from the study. To ensure homogeneity between groups, patients with comparable body mass index values were preferentially selected. Clinical assessment and data collection In all included pregnancies, the presence or absence of placenta previa and PAS was evaluated cross-sectionally. Clinical and obstetric data were obtained from patient medical records and the hospital electronic database. PAS diagnosis was established based on standardized ultrasonographic criteria reported in the literature and was assessed by the same experienced operator to minimize interobserver variability [15–17]. Ultrasonographic findings suggestive of PAS included loss of the clear zone and normal retroplacental space, myometrial thinning, presence of abnormal placental lacunae, placental bulging, and hypervascularity at the uterine serosa–bladder interface [15–17]. All patients were evaluated by the same perinatologists using grayscale and color Doppler ultrasonography. When required based on placental location, transvaginal ultrasonography was additionally performed to improve diagnostic accuracy. These imaging findings were used for delivery planning, and the diagnosis of PAS was subsequently confirmed by intraoperative findings and, when hysterectomy was performed, by histopathological evaluation. Therefore, a consistent diagnostic framework was applied across patients, strengthening the validity of the ROC analysis. The recorded parameters included maternal age, body height and weight, gravidity, parity, number of prior abortions, history of cesarean section, previous uterine surgery or curettage, performance of peripartum hysterectomy, and application of intraoperative compression sutures. Histopathological findings were recorded for patients who underwent peripartum hysterectomy. In patients who underwent hysterectomy, the diagnosis of PAS was confirmed by histopathological examination. In conservatively managed cases, the diagnosis was based on clinical and visual intraoperative assessment during cesarean section. PAS diagnosis was based on standardized intraoperative findings, including abnormal placental adherence such as difficulty in placental separation, excessive bleeding, and the need for additional surgical interventions. Laboratory analysis Blood samples were obtained at the time of delivery, and all samples were collected at ≥ 34 weeks of gestation. Venous blood samples were obtained in the preoperative period following at least 6 h of fasting. Blood samples were kept in an upright position for approximately 30 min and then centrifuged at 3000 rpm for 10 min in the biochemistry laboratory of Adana City Training and Research Hospital. Following centrifugation, the serum fraction was carefully separated, transferred into appropriately labeled Eppendorf tubes, and preserved at − 80 °C until further analysis. Samples were transported to a certified external laboratory (Baran Medical, Ankara, Türkiye) on dry ice at − 78.5 °C. Serum human adropin concentrations were measured using a commercially available ELISA kit (Elabscience, catalog no. E-EL-H5307) according to the manufacturer’s instructions. Following incubation at room temperature, optical density was determined at 450 nm using a microplate reader (BioTek ELx800) in conjunction with an automated plate washer (BioTek ELx50). Adropin concentrations were calculated by comparing sample optical densities with a standard calibration curve. The assay detection limit was 7.5 pg/mL, with a reference range of 12.5–800 pg/mL. Samples with concentrations exceeding the upper limit of the assay (> 800 pg/mL) were reanalyzed after dilution at a 1:2 ratio with phosphate-buffered saline (PBS). The measured values obtained after dilution were multiplied by the dilution factor to calculate the final concentrations. Statistical analysis Statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables were reported as mean ± standard deviation when normally distributed and as median (range) when non-normally distributed, while categorical variables were presented as counts and percentages. The normality of data distribution was assessed using the Shapiro Wilk test. Intergroup comparisons were conducted using the independent samples t-test or Mann–Whitney U test for continuous variables, and the chi-square test for categorical variables, as appropriate. The discriminative performance of serum adropin levels for detecting PAS was evaluated using receiver operating characteristic (ROC) curve analysis. In addition, a multivariable logistic regression analysis was performed to assess the independent association between serum adropin levels and PAS, adjusting for clinically relevant covariates.Statistical significance was defined as a two-sided p value < 0.05. Results A total of 84 pregnant women were included in the study: 42 healthy controls, 30 patients with placenta previa without PAS, and 12 patients diagnosed with PAS. The baseline demographic and obstetric characteristics of the participants are presented in Table 1. There were no significant differences among the three groups regarding maternal age, BMI, previous abdominal surgery, hemoglobin level, platelet count, white blood cell count, or lipid profile parameters including HDL, LDL, triglycerides, and total cholesterol (all p > 0.05) (Table 2). Table 1Maternal demographic and obstetric characteristics of the study groupsVariableControl group (n = 42)Placenta previa group (n = 30)Placenta Accreta Spectrum group (n = 12)p valueAge (years)30.09 ± 5.7831.36 ± 6.8732.16 ± 6.070.510Body mass index (kg/m²)29.95 ± 4.7729.70 ± 4.1327.12 ± 2.840.306Gravidity3 [1–15]3 [1–8]5[2–8]0.027Parity2 [0–5]1 [0–5]2.5[1–6]0.018Abortions0 [0–11]0 [0–2]1[0–4]0.321Living children1.5 [0–5]2 [0–5]2.5[1–6]0.033Previous cesarean deliveries1 [0–4]1 [0–5]2[1–4]0.033Previous curettage0 [0–11]0 [0–2]1[0–4]0.321Previous abdominal surgeries1.2 ± 0.991.66 ± 1.372.16 ± 1.020.560Placental location < 0.001 Anterior33 (67.3%)ᵃ9 (30.0%)ᵇ7 (58.3%)ᵃᵇPosterior9 (32.7%)ᵃ21 (70.0%)ᵇ5 (41.7%)ᵃᵇPreoperative ultrasonography < 0.001 Invasion present0 (0%)ᵃ30 (71.4%)ᵇ12 (100%)ᵇInvasion absent42 (100%)ᵃ0 (0%)b0 (0%)bIntraoperative assessment < 0.001 Invasion present2 (4.8%)ᵃ20 (66.7%)ᵇ12 (100%)ᵇInvasion absent40 (95.2%)ᵃ10 (33.3%)ᵇ0 (0%)ᵇIntraoperative additional interventions < 0.001 None40 (95.2%)ᵃ10 (33.3%)ᵇ0 (0%)ᶜCompression sutures2 (4.8%)ᵃ2 (4.8%)ᵃ1 (8.3%)ᶜHysterectomy0 (0%)ᵃ5 (16.7%)ᵇ11 (91.7%)ᶜData are presented as mean ± standard deviation or median (minimum–maximum) for continuous variables and as number (%) for categorical variablesIntergroup comparisons were performed using one-way ANOVA or Kruskal–Wallis test for continuous variables and chi-square test for categorical variablesDifferent superscript letters (a, b, c) indicate statistically significant differences between groups (post hoc analysis)PAS placenta accreta spectrumA p value < 0.05 was considered statistically significant Table 1Maternal demographic and obstetric characteristics of the study groupsVariableControl group (n = 42)Placenta previa group (n = 30)Placenta Accreta Spectrum group (n = 12)p valueAge (years)30.09 ± 5.7831.36 ± 6.8732.16 ± 6.070.510Body mass index (kg/m²)29.95 ± 4.7729.70 ± 4.1327.12 ± 2.840.306Gravidity3 [1–15]3 [1–8]5[2–8]0.027Parity2 [0–5]1 [0–5]2.5[1–6]0.018Abortions0 [0–11]0 [0–2]1[0–4]0.321Living children1.5 [0–5]2 [0–5]2.5[1–6]0.033Previous cesarean deliveries1 [0–4]1 [0–5]2[1–4]0.033Previous curettage0 [0–11]0 [0–2]1[0–4]0.321Previous abdominal surgeries1.2 ± 0.991.66 ± 1.372.16 ± 1.020.560Placental location < 0.001 Anterior33 (67.3%)ᵃ9 (30.0%)ᵇ7 (58.3%)ᵃᵇPosterior9 (32.7%)ᵃ21 (70.0%)ᵇ5 (41.7%)ᵃᵇPreoperative ultrasonography < 0.001 Invasion present0 (0%)ᵃ30 (71.4%)ᵇ12 (100%)ᵇInvasion absent42 (100%)ᵃ0 (0%)b0 (0%)bIntraoperative assessment < 0.001 Invasion present2 (4.8%)ᵃ20 (66.7%)ᵇ12 (100%)ᵇInvasion absent40 (95.2%)ᵃ10 (33.3%)ᵇ0 (0%)ᵇIntraoperative additional interventions < 0.001 None40 (95.2%)ᵃ10 (33.3%)ᵇ0 (0%)ᶜCompression sutures2 (4.8%)ᵃ2 (4.8%)ᵃ1 (8.3%)ᶜHysterectomy0 (0%)ᵃ5 (16.7%)ᵇ11 (91.7%)ᶜData are presented as mean ± standard deviation or median (minimum–maximum) for continuous variables and as number (%) for categorical variablesIntergroup comparisons were performed using one-way ANOVA or Kruskal–Wallis test for continuous variables and chi-square test for categorical variablesDifferent superscript letters (a, b, c) indicate statistically significant differences between groups (post hoc analysis)PAS placenta accreta spectrumA p value < 0.05 was considered statistically significant Maternal demographic and obstetric characteristics of the study groups Data are presented as mean ± standard deviation or median (minimum–maximum) for continuous variables and as number (%) for categorical variables Intergroup comparisons were performed using one-way ANOVA or Kruskal–Wallis test for continuous variables and chi-square test for categorical variables Different superscript letters (a, b, c) indicate statistically significant differences between groups (post hoc analysis) PAS placenta accreta spectrum A p value < 0.05 was considered statistically significant Table 2Comparison of preoperative laboratory parameters between study groupsParameterControl group (n = 42)Placenta previa group (n = 30)PAS group (n = 12)p valueHemoglobin (g/dL)11.22 ± 1.3611.27 ± 1.2911.09 ± 1.050.917Platelet count (×10³/µL)261.78 ± 71.41236.20 ± 62.43240.83 ± 37.560.228White blood cell count (×10³/µL)10.75 [6.40–29.80]10.50 [6–15.40]9.57[6-15.50]0.193Serum adropin (pg/mL)674.44 [93.19–1902.07]257.03 [75.47–959.64]335[67.53-1202.21]< 0.001HDL cholesterol (mg/dL)70.50 ± 12.0863.88 ± 15.6665.33 ± 16.550.144LDL cholesterol (mg/dL)149.02 ± 35.73143.85 ± 32.59149.88 ± 35.100.810Triglycerides (mg/dL)214 [69–463]227 [166–766]199[106–374]0.544Total cholesterol (mg/dL)239.95 ± 47.74225.20 ± 65.90234.55 ± 43.520.557Hb hemoglobin, PLT platelet, WBC white blood cell, HDL high-density lipoprotein, LDL low-density lipoprotein, TG triglycerides Table 2Comparison of preoperative laboratory parameters between study groupsParameterControl group (n = 42)Placenta previa group (n = 30)PAS group (n = 12)p valueHemoglobin (g/dL)11.22 ± 1.3611.27 ± 1.2911.09 ± 1.050.917Platelet count (×10³/µL)261.78 ± 71.41236.20 ± 62.43240.83 ± 37.560.228White blood cell count (×10³/µL)10.75 [6.40–29.80]10.50 [6–15.40]9.57[6-15.50]0.193Serum adropin (pg/mL)674.44 [93.19–1902.07]257.03 [75.47–959.64]335[67.53-1202.21]< 0.001HDL cholesterol (mg/dL)70.50 ± 12.0863.88 ± 15.6665.33 ± 16.550.144LDL cholesterol (mg/dL)149.02 ± 35.73143.85 ± 32.59149.88 ± 35.100.810Triglycerides (mg/dL)214 [69–463]227 [166–766]199[106–374]0.544Total cholesterol (mg/dL)239.95 ± 47.74225.20 ± 65.90234.55 ± 43.520.557Hb hemoglobin, PLT platelet, WBC white blood cell, HDL high-density lipoprotein, LDL low-density lipoprotein, TG triglycerides Comparison of preoperative laboratory parameters between study groups Hb hemoglobin, PLT platelet, WBC white blood cell, HDL high-density lipoprotein, LDL low-density lipoprotein, TG triglycerides Gravidity, parity, number of living children, and number of prior cesarean deliveries differed among groups, with higher values observed in the PAS group compared with controls (p < 0.05). However, abortion history and prior curettage procedures were comparable between groups. Placental location differed significantly among groups, with posterior placentation being more frequent in the placenta previa group compared with controls (p < 0.001). Preoperative ultrasonographic evaluation demonstrated placental invasion in 71.4% of the placenta previa group and in all patients in the PAS group, whereas no invasion was detected in the control group (p < 0.001). Intraoperative findings confirmed this distribution.Regarding intraoperative management, additional surgical interventions were significantly more common in the PAS group. Compression sutures were applied more frequently in placenta previa patients, whereas peripartum hysterectomy was predominantly performed in the PAS group (p < 0.001). Among hysterectomy cases, histopathological examination demonstrated placenta accreta, increta, and percreta in varying proportions. Median preoperative maternal serum adropin levels differed significantly among the three groups (p < 0.001). Adropin levels were highest in the control group (median 674.44 pg/mL), markedly lower in the placenta previa group (median 257.03 pg/mL), and similarly reduced in the PAS group (median 335.93 pg/mL). No significant difference was observed between placenta previa and PAS groups, whereas both groups showed significantly lower values compared with controls. ROC curve analysis demonstrated moderate discriminative performance of maternal serum. adropin levels for identifying abnormal placentation (control vs. placenta previa + PAS), with. an area under the curve (AUC) of 0.778 ((95% CI: 0.678–0.878, p < 0.001). An adropin cut-off value of ≤ 431.27. pg/mL yielded a sensitivity of 71.4% and a specificity of 78.6%. The positive predictive value. and negative predictive value were 76.9% and 73.3%, respectively (Table 3, Fig.1). Table 3Discriminative performance of serum adropin levels for identifying abnormal placentation (controls vs. placenta previa + placenta accreta spectrum)ParameterCut-off valueAUC (95% CI)p valueSensitivity (%)Specificity (%)PPV (%)NPV (%)Serum adropin level≤ 431.27 pg/mL0.778 (0.678–0.878)< 0.00171.478.676.973.3 AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negative predictive value Table 3Discriminative performance of serum adropin levels for identifying abnormal placentation (controls vs. placenta previa + placenta accreta spectrum)ParameterCut-off valueAUC (95% CI)p valueSensitivity (%)Specificity (%)PPV (%)NPV (%)Serum adropin level≤ 431.27 pg/mL0.778 (0.678–0.878)< 0.00171.478.676.973.3 AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negative predictive value Discriminative performance of serum adropin levels for identifying abnormal placentation (controls vs. placenta previa + placenta accreta spectrum) AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negative predictive value Fig. 1Receiver operating characteristic (ROC) curve demonstrating the discriminative performance of maternal serum adropin levels for identifying abnormal placentation (control group vs. placenta previa and placenta accreta spectrum). The area under the curve (AUC) was 0.778 (95% CI: 0.678–0.878), indicating moderate diagnostic accuracy Receiver operating characteristic (ROC) curve demonstrating the discriminative performance of maternal serum adropin levels for identifying abnormal placentation (control group vs. placenta previa and placenta accreta spectrum). The area under the curve (AUC) was 0.778 (95% CI: 0.678–0.878), indicating moderate diagnostic accuracy In addition, subgroup ROC analysis showed that serum adropin levels did not. significantly differentiate PAS from placenta previa (AUC = 0.542, 95% CI: 0.315–0.768, p = 0.676) A cut-off value of > 412.08 pg/mL yielded a sensitivity of 50.0% and a specificity of 76.7% (Table 4, Fig. 2). Table 4Discriminative performance of serum adropin levels for differentiating placenta accreta spectrum (PAS) from placenta previaParameterCut-off valueAUC (95% CI)p valueSensitivity (%)Specificity (%)PPV (%)NPV (%)Serum adropin level> 412.08 pg/mL0.542 (0.315–0.768)0.67650.076.746.279.3AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negativepredictive value Table 4Discriminative performance of serum adropin levels for differentiating placenta accreta spectrum (PAS) from placenta previaParameterCut-off valueAUC (95% CI)p valueSensitivity (%)Specificity (%)PPV (%)NPV (%)Serum adropin level> 412.08 pg/mL0.542 (0.315–0.768)0.67650.076.746.279.3AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negativepredictive value Discriminative performance of serum adropin levels for differentiating placenta accreta spectrum (PAS) from placenta previa AUC area under the curve, CI confidence interval, PPV positive predictive value, NPV negativepredictive value Fig. 2Receiver operating characteristic (ROC) curve demonstrating the discriminative performance of maternal serum adropin levels for differentiating placenta accreta spectrum (PAS) from placenta previa. The area under the curve (AUC) was 0.542 (95% CI: 0.315–0.768), indicating no significant discriminative ability Receiver operating characteristic (ROC) curve demonstrating the discriminative performance of maternal serum adropin levels for differentiating placenta accreta spectrum (PAS) from placenta previa. The area under the curve (AUC) was 0.542 (95% CI: 0.315–0.768), indicating no significant discriminative ability To further evaluate potential confounding factors, a multivariable logistic regression analysis was performed including serum adropin levels, previous cesarean delivery, and placental location (Table 5). Table 5Multivariable logistic regression analysis for factors associated with placenta accreta spectrum (PAS) among patients with placenta previaVariableAdjusted OR95% CIp valueSerum adropin (pg/mL)1.0010.999–1.0030.260Previous cesarean delivery1.3090.731–2.3430.364Placental location (anterior vs. posterior)0.3660.086–1.5560.174OR odds ratio, CI confidence interval, PAS placenta accreta spectrumReference categories: placenta previa for outcome; posterior placental location for placental location variable Table 5Multivariable logistic regression analysis for factors associated with placenta accreta spectrum (PAS) among patients with placenta previaVariableAdjusted OR95% CIp valueSerum adropin (pg/mL)1.0010.999–1.0030.260Previous cesarean delivery1.3090.731–2.3430.364Placental location (anterior vs. posterior)0.3660.086–1.5560.174OR odds ratio, CI confidence interval, PAS placenta accreta spectrumReference categories: placenta previa for outcome; posterior placental location for placental location variable Multivariable logistic regression analysis for factors associated with placenta accreta spectrum (PAS) among patients with placenta previa OR odds ratio, CI confidence interval, PAS placenta accreta spectrum Reference categories: placenta previa for outcome; posterior placental location for placental location variable In this model, serum adropin was not found to be an independent predictor of PAS (OR: 1.001, 95% CI: 0.999–1.003, p = 0.260). Neither previous cesarean delivery nor placental location reached statistical significance, although placental location showed a trend toward association (OR: 0.366, 95% CI: 0.086–1.556, p = 0.174). Discussion This study demonstrated significantly reduced maternal serum adropin concentrations in pregnancies affected by placenta previa and/or PAS compared with healthy controls. This finding is consistent with experimental evidence demonstrating that adropin enhances endothelial function through endothelial nitric oxide synthase (eNOS)–mediated nitric oxide (NO) production. Accordingly, reduced adropin concentrations observed in the PAS group may reflect impaired endothelial homeostasis, which constitutes a key pathophysiological component of abnormal placental invasion [5]. Earlier clinical investigations have demonstrated a relationship between reduced circulating adropin concentrations and vascular disorders. Wu et al. examined the association between serum adropin concentrations and the angiographic extent of coronary atherosclerosis in diabetic and non-diabetic individuals. They reported significantly reduced adropin levels in patients with type 2 diabetes mellitus, independent of established cardiovascular risk factors [18]. Furthermore, adropin concentrations showed an inverse and independent relationship with the extent of coronary atherosclerosis, indicating a possible role in cardiovascular risk stratification. Supporting these findings, Lovren et al. showed that adropin enhances eNOS activity and NO bioavailability in endothelial cell culture models, thereby preserving endothelial integrity and vascular homeostasis [5]. Collectively, these data underscore the role of adropin as a critical regulator of vascular function. In obstetric pathology, Karaca et al. reported significantly reduced serum adropin levels in patients with preeclampsia, particularly in cases characterized by severe endothelial dysfunction [12]. In that study, adropin levels decreased as disease severity increased, whereas other endothelial dysfunction markers, including autotaxin and lysophosphatidic acid, were elevated. The authors suggested that reduced adropin concentrations may serve as a biochemical indicator of impaired NO synthesis, diminished vascular reactivity, and compromised placental perfusion. The significantly lower adropin levels observed in the PAS group in our study may be explained by similar pathophysiological mechanisms involving endothelial dysfunction and abnormal placental vascular remodeling. Similarly, Çelik et al. reported significantly lower maternal and cord blood adropin levels in pregnancies complicated by gestational diabetes mellitus. In their study, maternal adropin concentrations were negatively correlated with fasting glucose levels, suggesting that reduced adropin may reflect impaired metabolic and endothelial homeostasis during pregnancy [8]. Further evidence supporting the placental relevance of adropin comes from studies in intrauterine growth restriction (IUGR). Aydın et al. demonstrated significantly lower adropin levels in both mothers and neonates with IUGR compared with healthy controls [13], suggesting a role for adropin in fetal–placental circulation and growth regulation. These findings reinforce the concept that adropin may represent a shared biomarker across obstetric conditions associated with impaired placental function and adverse perinatal outcomes. Beyond obstetric disorders, reduced adropin levels have been consistently reported in women with polycystic ovary syndrome (PCOS), with significant negative correlations observed between adropin concentrations and body mass index, waist circumference, fasting glucose, insulin levels, insulin resistance and atherogenic lipid profiles [9, 10]. These findings suggest that adropin reflects systemic metabolic and insulin-resistant states rather than disease-specific processes alone. In line with this concept, evidence from endometrial cancer studies has shown that serum adropin levels are significantly reduced and exhibit high sensitivity and specificity for disease prediction, outperforming traditional metabolic parameters such as BMI and insulin resistance indices [14]. Notably, adropin levels were not associated with tumor stage, histologic subtype, grade, or depth of myometrial invasion, indicating that adropin may reflect an underlying metabolic endothelial milieu rather than disease extent. In the present study, the median maternal serum adropin level in patients with placenta previa and/or PAS was 257.03 pg/mL. ROC analysis demonstrated moderate discriminative performance for identifying abnormal placental invasion, with a cut-off value of ≤ 431.27 pg/mL. The observed reduction in adropin levels may partly reflect pregnancy-related inflammatory and metabolic adaptations; however, the statistically significant discriminative performance suggests that decreased adropin concentrations are not merely a physiological finding of pregnancy. These findings indicate that reduced adropin levels may be associated with abnormal placentation rather than the degree of myometrial invasion. Importantly, the initial aim of the present study was to evaluate maternal serum adropin levels in pregnancies at ≥ 34 weeks of gestation, specifically focusing on patients with placenta previa and healthy controls scheduled for delivery in the third trimester. Within this framework, we identified a significant reduction in adropin levels in pregnancies complicated by placenta previa compared with normal pregnancies. However, this reduction did not significantly differ between placenta previa and PAS cases, suggesting that adropin levels may reflect abnormal placental localization in general rather than the severity of trophoblastic invasion. Consistent with these observations, subgroup ROC analysis demonstrated that adropin did not have a statistically significant discriminative performance in differentiating PAS from placenta previa alone. This represents a major limitation of the study, as the inability to distinguish between these two clinically relevant conditions substantially reduces the specificity and potential clinical applicability of adropin as a diagnostic biomarker. Furthermore, multivariable logistic regression analysis revealed that serum adropin was not independently associated with PAS after adjustment for clinically relevant factors, including placental location and previous cesarean delivery. Taken together, these findings indicate that adropin cannot be considered a reliable independent biomarker for PAS and appears to reflect a more generalized alteration in placental physiology rather than the depth of myometrial invasion. Consistent with previous studies reporting reduced adropin levels in metabolic and vascular disorders, our findings support the concept that adropin reflects the metabolic–endothelial milieu associated with abnormal placental implantation. However, given the lack of independent association and limited discriminatory capacity, adropin should not be considered a standalone diagnostic tool in clinical practice. Instead, its potential role, if any, may be limited to a supportive or adjunctive biomarker alongside established imaging modalities. In our clinical setting, placental invasion is routinely assessed with high accuracy by experienced perinatologists using standardized ultrasonographic criteria. Therefore, the primary aim of the present study was not to replace established imaging methods, but rather to explore whether serum adropin levels could provide additional information in distinguishing pregnancies with abnormal placentation from healthy controls. Considering the relatively low prevalence of PAS and the presence of comorbid conditions among referred patients, constructing a balanced three-group design was not initially feasible. Accordingly, this study should be regarded as an exploratory analysis aimed at generating hypotheses rather than providing definitive conclusions. Placental location differed significantly among groups, with posterior placentation being more frequent in the placenta previa group compared with controls. However, a clear mechanistic explanation for this finding could not be established, and the relationship between placental location and adropin levels was not methodologically predefined. Future prospective studies designed specifically to investigate this relationship may help to clarify its potential significance. Several limitations of this study warrant careful consideration. First, maternal serum adropin concentrations were assessed exclusively under fasting conditions, although adropin levels are known to be influenced by metabolic status. Second, all measurements were performed in the third trimester, after the diagnosis of placenta previa and/or PAS, representing near preoperative values. This timing limits the ability to assess the predictive value of adropin in earlier stages of pregnancy and reduces its potential utility as an early diagnostic biomarker. Third, the sample size was relatively limited, particularly within the PAS subgroup, which included a small number of patients. Given the low prevalence of PAS, this limitation was difficult to overcome; however, it may have significantly reduced the statistical power to detect meaningful differences between PAS and placenta previa. Finally, the cross-sectional design of the study further limits causal inference and restricts the ability to evaluate temporal changes in adropin levels throughout pregnancy. Adropin concentrations may vary across gestational ages and in the presence of comorbid conditions, which were excluded in this study to ensure group homogeneity. Therefore, the findings cannot be generalized to earlier gestational periods or to more heterogeneous patient populations. Despite these limitations, a major strength of this study is the comprehensive evaluation of placental invasion using standardized ultrasonographic criteria confirmed by intraoperative findings and histopathological examination when available. To the best of our knowledge, this is the first study to specifically evaluate maternal serum adropin levels in pregnancies complicated by placenta previa and PAS. Taken together, the present findings provide preliminary evidence that decreased maternal serum adropin levels are associated with abnormal placentation. However, these results should be interpreted as exploratory and hypothesis-generating. Larger, prospective studies incorporating earlier gestational time points and more diverse clinical populations are required to better define the diagnostic and prognostic value of adropin and to determine whether it has any meaningful role in clinical practice. Conclusion In summary, our findings indicate that maternal serum adropin concentrations are significantly reduced in pregnancies complicated by placenta previa and/or placenta accreta spectrum compared with uncomplicated pregnancies. These results suggest that decreased adropin levels may reflect a disturbed metabolic–endothelial milieu associated with abnormal placental implantation rather than the severity of trophoblastic invasion. Although ROC analysis demonstrated moderate discriminative performance for identifying abnormal placentation, adropin did not show sufficient ability to differentiate PAS from placenta previa and was not independently associated with PAS. Therefore, adropin cannot be considered a reliable standalone diagnostic biomarker for invasive placentation.Given the cross-sectional design, limited sample size—particularly within the PAS subgroup—and late gestational sampling, these findings should be interpreted with caution and regarded as exploratory and hypothesis-generating. Further large-scale, prospective studies incorporating earlier gestational time points are required to better define the temporal dynamics of adropin and to determine whether it has any meaningful role in the clinical evaluation of abnormal placentation. Supplementary Information Supplementary Material 1. Supplementary Material 1.