Prenatal prediction of neonatal respiratory distress syndrome using uteroplacental Doppler and placental microvascular perfusion indices: development of a risk model

Introduction

Neonatal respiratory distress syndrome (NRDS) is a leading cause of neonatal morbidity, neonatal intensive care unit (NICU) admission, and early mortality worldwide, affecting approximately 7% of newborn infants (1,2). Characterized by insufficient pulmonary surfactant production, dysfunctional surfactant activity, or delayed pulmonary maturation, NRDS occurs predominantly in preterm neonates, with its incidence increasing as gestational age decreases (3,4). Clinically, affected infants typically present within hours of birth with tachypnea, nasal flaring, grunting, and chest wall retractions (5). Without timely intervention, NRDS may rapidly progress to hypoxemia, hypercapnia, and respiratory failure, necessitating interventions such as surfactant replacement therapy, continuous positive airway pressure, or mechanical ventilation (6). Despite advances in perinatal care, including antenatal corticosteroids and postnatal surfactant therapy, NRDS continues to result in complications such as bronchopulmonary dysplasia, retinopathy of prematurity, necrotizing enterocolitis, and nosocomial infections, contributing to prolonged hospitalization and long-term neurodevelopmental impairment (7,8).

The timely identification of infants at high risk of NRDS is essential for optimizing perinatal decision-making. Current antenatal risk assessments for NRDS largely rely on gestational age and obstetric history but lack sufficient physiological specificity (9). Practical and scalable tools for risk stratification during the second trimester are limited, reducing opportunities for early preventative interventions such as corticosteroid administration, targeted delivery in tertiary centers, and early surfactant administration (10).

Emerging evidence suggests that uteroplacental hemodynamics may provide a biologically grounded framework for assessing fetal respiratory risk (11). Abnormal remodeling and increased uterine vascular resistance can impair placental perfusion, triggering detectable fetal adaptive responses (12-14). The increased second-trimester uterine artery pulsatility index (UtA-PI) is an established marker of increased resistance and adverse outcomes (15). Similarly, the cerebroplacental ratio (CPR) reflects the balance between fetal adaptation and placental afterload (16-18). However, the predictive performance of CPR has shown variability across different populations and clinical settings (19). Given the inconsistencies in the reported threshold values (20) and limited standalone value in certain subgroups (21), these markers may achieve greater prognostic accuracy when contextualized within maternal-fetal clinical profiles (22).

Beyond macrocirculation, three-dimensional power Doppler (3D-PD) imaging provides insights into microvascular density and low-velocity perfusion (23). Reduced placental indices have been associated with pre-eclampsia and fetal growth restriction (FGR) (24-26). Maternal clinical conditions, particularly hypertensive disorders of pregnancy (HDP), are known contributors to neonatal respiratory morbidity independent of gestational age (27,28). Previous studies have reported increased respiratory complications in neonates born to mothers with HDP, underscoring the need to integrate these comorbidities into predictive models (29,30).

In light of these findings, we hypothesized that a composite antenatal risk model integrating (I) uteroplacental macrovascular resistance (UtA-PI); (II) fetal adaptive hemodynamics (CPR); (III) placental microvascular perfusion [assessed by the vascularization flow index (VFI), supported by the vascularization index (VI) and flow index (FI)]; and (IV) key maternal clinical factors, such as hypertensive status and metabolic profile, would improve early discrimination of NRDS risk. Unlike previous studies that focused on isolated Doppler markers or placental disorders, our approach provides a multiscale physiological framework targeting neonatal respiratory outcomes. Further, translating this composite risk model into a clinically applicable nomogram may provide intuitive, bedside decision support for respiratory management, especially in resource-limited settings.

Accordingly, we conducted a retrospective cohort study of pregnant women undergoing mid-gestation ultrasound evaluations. We assessed the independent and combined predictive value of UtA-PI, CPR, 3D-PD-derived indices, and maternal clinical variables for neonatal NRDS—defined as the need for sustained respiratory support and NICU admission. Model performance was compared with single-indicator strategies, with evaluation of discrimination, calibration, and net clinical benefit. By linking antenatal placental hemodynamics to a clinically significant neonatal endpoint, this study sought to address a key gap in perinatal respiratory risk prediction and support the development of physiology-informed, pre-delivery risk stratification tools. We present this article in accordance with the TRIPOD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0143/rc).

Methods

Study population

A total of 610 pregnant women were initially screened for second-trimester ultrasonographic examinations at Xiangyang No. 1 People’s Hospital, Hubei University of Medicine between January 2021 and December 2024. After applying the exclusion criteria, 362 participants were included in the final analysis. The study cohort included 58 pregnant women whose neonates developed respiratory distress (the NRDS group) and 304 pregnant women with healthy neonates (the control group). During the screening process, 25 participants were excluded due to multiple gestations, and 30 were excluded as a result of preterm delivery before 34 weeks of gestation. Additionally, 20 participants were excluded following a prenatal diagnosis of fetal chromosomal abnormalities, while 25 were excluded due to major fetal structural anomalies detected during prenatal screening. Further, 35 participants were excluded due to suboptimal ultrasound image quality, and 113 were excluded due to incomplete clinical or neonatal outcome data or loss to follow-up.

Data collection included maternal clinical characteristics and medical history, macrovascular Doppler parameters, and slow-flow placental indices. The statistical analysis was performed as outlined in Figure 1.

Figure 1 Flowchart depicting the study workflow.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Xiangyang No. 1 People’s Hospital (No. 2021KYLX02; date of approval: January 14, 2022), and all patients provided written informed consent for fetal ultrasonography.

Clinical data collection

Maternal clinical data from January 2021 to December 2024 were systematically collected through reviews of electronic medical records and prenatal care files. The information collected included: baseline maternal indicators, such as maternal age, gestational age at the time of the second-trimester ultrasound, gravidity, and parity, as well as pregnancy-related and metabolic indicators. Pre-pregnancy body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Gestational hypertension was diagnosed as per the International Society for the Study of Hypertension in Pregnancy (ISSHP) criteria (31). Peeclampsia was diagnosed as per the ISSHP diagnostic criteria (32). Gestational diabetes mellitus (GDM) was diagnosed based on the oral glucose tolerance test according to the International Association of Diabetes and Pregnancy Study Groups criteria (33). Additional metabolic variables included hypertriglyceridemia, serum high-density lipoprotein-cholesterol, low-density lipoprotein-cholesterol, first-trimester fasting glucose and triglyceride levels. Other variables included the mode of labor onset (spontaneous labor vs. induced/elective labor); FGR, defined as an estimated fetal weight <10th percentile for gestational age combined with abnormal Doppler parameters (34); and perinatal infection, defined according to the World Health Organization as a bacterial infection of the genital tract or surrounding tissues occurring between the onset of rupture of membranes or labor and 42 days postpartum, accompanied by at least two of the following: pelvic pain, fever, abnormal vaginal discharge, foul-smelling vaginal discharge, or delayed uterine involution (35). The neonatal outcome indicators included gestational age at delivery, neonatal birth weight, 1-minute Apgar score ≤7, NICU admission rate, and diagnosis of NRDS based on clinical symptoms, chest X-ray findings, and the need for surfactant or respiratory support (8). All data were independently verified by two researchers to ensure completeness and accuracy, with discrepancies resolved through consultation with the attending physician.

Conventional ultrasound data collection

Conventional ultrasound examinations were uniformly performed using a Voluson E10 ultrasound system (GE Healthcare, Chicago, IL, USA) to ensure consistency in equipment and measurement standards. Spectral Doppler technology was used to measure the macrovascular hemodynamic parameters of the maternal-fetal-placental system: for the umbilical artery (UA), the parameters measured included the systolic/diastolic (S/D) ratio, pulsatility index (PI), and resistance index (RI), with measurements taken at the free segment of the umbilical cord (2–3 cm from the placental insertion); for the fetal middle cerebral artery (MCA), the parameters measured included the S/D ratio, PI, and RI, with measurements taken at the proximal 1/3 segment of the MCA (avoiding the area near the circle of Willis); for the UtA, the parameters measured included the S/D ratio, PI, and RI, with measurements taken at the intersection of the UtA and external iliac artery, and the average value of bilateral UtAs used for analysis. Additionally, the CPR was calculated as the ratio of MCA-PI to UA-PI, which reflects the balance of fetal cerebral and placental blood flow. Each parameter was measured consecutively three times, and the average value was used to reduce measurement error (Figure 2).

Figure 2 Spectral Doppler ultrasound: (A) Umb; (B) Rt MCA; (C) Lt UtA. Each with color Doppler image and spectral waveform, displaying hemodynamic parameters including PS, ED, S/D, PI, RI, MD, TAmax, and HR. ED, end-diastolic velocity; HR, heart rate; Lt UtA, left uterine artery; MD, mean velocity; PI, pulsatility index; PS, peak systolic velocity; RI, resistance index; Rt MCA, right middle cerebral artery; S/D, systolic/diastolic ratio; TAmax, time-averaged maximum velocity; Umb, umbilical artery.

Slow-flow ultrasound data collection

Slow-flow ultrasound data acquisition was performed using the same Voluson E10 ultrasound system and conventional parameters to quantify placental microcirculatory perfusion, targeting the microvascular indices (VI, FI, and VFI). Using the 3D-PD slow-flow imaging mode, the placental location was first identified via two-dimensional ultrasound, after which the region of interest (ROI) was adjusted to cover the entire placental parenchyma (avoiding the umbilical cord insertion site and fetal tissues). The 3D-PD system was adjusted to slow-flow mode (with a low pulse repetition frequency and high sensitivity to capture low-velocity microvascular blood flow). Placental volume datasets were acquired over a scanning time of 5–10 seconds to ensure stable image quality.

After data acquisition, the offline analysis was performed using VOCAL™ (Virtual Organ Computer-aided Analysis) software. The software automatically segmented the placental volume data and calculated three core microcirculatory indices: the VI (the percentage of vascular volume in the ROI, reflecting vascular density); the FI (the average flow intensity in the vascular region, reflecting blood flow velocity); and the VFI (the product of VI and FI, comprehensively reflecting vascular density and flow intensity). Each volume dataset was analyzed twice by a sonographer with 5 years of experience in fetal ultrasound, and the average value was used for subsequent analysis (Figure 3).

Figure 3 Placental imaging via slow-flow ultrasound technology. (A) Multiplanar and 3D reconstructed images illustrating slow-flow vascular distribution in the placenta. (B) ROI delineation and histogram analysis for quantifying the VI, FI, and VFI. 3D, three-dimensional; FI, flow index; MG, mean gray value; ROI, region of interest; VFI, vascularization flow index; VI, vascularization index.

Elastic ultrasound data collection

Elastic ultrasound data were collected to assess placental tissue elasticity. The shear wave elastography function of the Voluson E10 system was used to measure both the central and marginal regions of the placenta (with the central region representing the main functional area of placental perfusion). For each region, the probe was placed perpendicular tothe maternal abdominal wall corresponding to the target placental area, and a 2 mm × 2 mm ROI was placed entirely within the placental parenchyma. The system automatically measured the mean elasticity (Emean) value of each ROI. Emean reflects the hardness of placental tissue, with higher values indicating stiffer placental tissue, which may be associated with impaired microcirculation. Each measurement was repeated three times, and the average Emean values for both the central and marginal regions were recorded for subsequent predictive analysis based on preliminary findings indicating an association with placental function and neonatal respiratory outcomes.

Statistical analysis

The statistical analyses were performed using R (version 4.2.0) and SPSS (version 26.0) software. Participants were allocated to the NRDS and non-NRDS groups. Continuous and categorical variables were compared using the Student’s t-test and Chi-square test, respectively. Univariate and multivariate logistic regression analyses were employed to identify independent predictors. Specifically, a P value threshold of <0.1 was used during the univariate screening to ensure that potentially significant predictors or confounders were not prematurely excluded, thereby enhancing the comprehensiveness of the final multivariate model. A nomogram was constructed based on the final model. Model performance was evaluated by the area under curve (AUC) of the receiver operating characteristic (ROC) curve and calibration plots. Internal validation was conducted via bootstrapping with 1,000 resamples. A P value <0.05 was considered statistically significant. Given the retrospective design, the initial sample size was determined by the availability of eligible records. To ensure the stability of the multivariate logistic regression model, we adhered to the events per variable (EPV) guideline. With 58 NRDS events and four independent predictors, our EPV was 14.5 (exceeding the recommended minimum of 10), indicating a sufficient sample size to prevent overfitting. Further, a post-hoc power analysis confirmed that the current sample size (n₁=58, n₂=304) provided adequate power (>80%) to detect the observed differences in key continuous predictors, using a two-tailed alpha (α) error probability of 0.05 and a beta (β) of 0.20.

Results

Study population and clinical characteristics

A total of 362 women met the inclusion criteria and were included in the final study cohort. Among them, 58 (16.0%) neonates developed NRDS and formed the case group, while the remaining 304 (84.0%) comprised the non-NRDS control group. The two groups were comparable in terms of maternal age and gestational age at scan. However, the NRDS group had a significantly higher pre-pregnancy BMI (25.8±3.5 vs. 23.1±2.9 kg/m², P<0.001) and a higher prevalence of gestational hypertension (25.9% vs. 7.9%, P<0.001), pre-eclampsia (20.7% vs. 3.6%, P<0.001), and GDM (41.4% vs. 22.0%, P=0.002). Notably, FGR was also more frequent in the NRDS group (29.3% vs. 5.9%, P<0.001). Regarding the mode of delivery, the proportion of induced or elective labor was significantly higher in the NRDS group than in the control group (60.3% vs. 25.0%, P<0.001). In addition, the incidence of perinatal infection (clinical chorioamnionitis or early-onset neonatal sepsis) was significantly higher in the NRDS group than in the non-distress control group (24.1% vs. 4.9%, P<0.001). Neonates in the NRDS group also had lower birth weights and were significantly more likely to be admitted to the NICU (Table 1).

Comparison of ultrasound parameters

The analysis revealed significant hemodynamic and perfusion alterations in the NRDS group. Notably, the UtA-PI was elevated (1.15±0.38 vs. 1.02±0.35, P<0.05). Additionally, the CPR was significantly lower in the NRDS group than in the control group (1.53±0.27 vs. 1.62±0.30, P=0.003). Crucially, the slow-flow 3D-PD indices also showed significant differences; both VI and VFI were substantially reduced in the NRDS group compared (P=0.009 and P=0.002, respectively) (Table 2).

Identification of potential predictors: univariate analysis

To screen for the variables associated with NRDS, univariate logistic regression was performed. The results identified a set of significant potential predictors. These included maternal clinical factors such as higher pre-pregnancy BMI, the presence of gestational hypertension, pre-eclampsia, FGR, induced/elective labor onset, GDM, and increased first-trimester fasting glucose and triglyceride levels. Among the ultrasound parameters, increased UtA-PI, reduced MCA-RI, a lower CPR, decreased VI and VFI, and increased placental elasticity (Emean) at the central region were all significantly associated with an increased risk of NRDS (Table 3).

Identification of potential predictors: multivariate analysis

The significant variables from the univariate analysis (P<0.1) were subsequently entered into a multivariate logistic regression model to identify independent predictors. Four factors remained independently and significantly associated with NRDS after adjustment for confounding variables: increased UtA-PI [odds ratio (OR) =1.52, 95% confidence interval (CI): 1.14–2.02, P=0.004), decreased CPR (OR =0.70, 95% CI: 0.57–0.86, P<0.001), decreased VFI (OR =0.75, 95% CI: 0.63–0.90, P=0.002), and the presence of gestational hypertension (OR =3.08, 95% CI: 1.44–6.59, P=0.004) (Table 4).

Nomogram for predicting neonatal respiratory distress

A nomogram was developed to predict the risk of NRDS by integrating the four independent predictors identified in the multivariate logistic regression model: UtA-PI, CPR, VFI, and gestational hypertension. Each predictor was assigned a dedicated score axis, with maximum scores varying across variables (UtA-PI: 0–100, CPR: 0–80, VFI: 0–90, and gestational hypertension: 0–60). The scoring process was straightforward: for each predictor, the corresponding score was determined by locating the patient’s measured value on the respective axis. These individual scores were then summed to obtain a total score (ranging from 0 to 330), which was mapped to the “Risk of NRDS” axis to estimate the predicted probability of NRDS (ranging from 0 to 0.8). For example, a pregnant woman with UtA-PI =1.4 (score =70), CPR =1.4 (score =20), VFI =4.5 (score =30), and gestational hypertension (score =60) would have a total score of 180, corresponding to an estimated NRDS risk of approximately 0.4 (Figure 4).

Figure 4 Nomogram for predicting perinatal adverse event risk using the UtA-PI, CPR, VFI, and gestational hypertension. CPR, cerebroplacental ratio; UtA-PI, uterine artery pulsatility index; VFI, vascularization flow index.

ROC plot of the nomogram

An ROC plot was used to evaluate the ability of the nomogram to distinguish between pregnant women whose neonates developed NRDS and those who did not. The x-axis represents the false-positive rate (1 − specificity), and the y-axis represents the true-positive rate (sensitivity). The AUC of the nomogram was 0.849 (95% CI: 0.745–0.871). At a sensitivity of 82%, the nomogram achieved a specificity of 78%, while a specificity of 85%, corresponded to a sensitivity of 75% (Figure 5).

Figure 5 Receiver operating characteristic curve for model prediction, with the AUC. AUC, area under the curve; CI, confidence interval.

Calibration curves of the nomogram

Calibration curves were used to assess the agreement between the nomogram’s predicted probabilities and the observed incidence of NRDS. The blue line represents the ideal calibration scenario, indicating perfect agreement between the predicted and actual outcomes. The red line represents the apparent calibration (i.e., the model performance in the original study cohort), while the green line represents the bias-corrected calibration after 500 bootstrap resamples (to account for potential overfitting). The mean absolute error between predicted and actual risks was 0.046 (n=362). For example, when the nomogram predicted an NRDS risk of 0.3, the actual observed incidence was approximately 0.28–0.32 (Figure 6).

Figure 6 Calibration curve of the predictive model, plotting observed probability against predicted probability. The dashed, red dashed, and blue solid lines represent ideal, apparent, and bias-corrected calibration, respectively.

Decision curve of the nomogram

A decision curve analysis (DCA) was performed to evaluate the clinical utility of the nomogram by comparing its net benefit to two extreme strategies: “treat all” (intervening in every pregnant woman) and “treat none” (no intervention for any woman). The x-axis represents the threshold probability, defined as the minimum NRDS risk at which a clinician would recommend intervention. The y-axis represents the net benefit, calculated as the number of correctly identified high-risk cases minus the number of unnecessary interventions (false positives). The nomogram (red curve) demonstrated a higher net benefit than both “treat all” (blue curve) and “treat none” (green curve) across a threshold probability range of 0.05–0.75. For example, at a threshold probability of 0.3, the nomogram’s net benefit was 0.25, compared to 0.12 for “treat all” and 0.03 for “treat none” (Figure 7).

Figure 7 Decision curve analysis of the prediction nomogram, showing the standardized net benefit across different high-risk thresholds. The blue line represents the nomogram, with gray and black lines denoting “All” (intervening for all) and “None” (intervening for none) strategies, alongside the corresponding cost-benefit ratios.

Discussion

This study successfully developed and validated a clinically applicable nomogram for the prediction of NRDS using second-trimester assessments. The model’s strength lies in its integrative approach, combining a key maternal comorbidity (gestational hypertension), a marker of uteroplacental perfusion (UtA-PI), an indicator of fetal hemodynamic adaptation (CPR), and a direct measure of placental microvascularization (VFI) derived from slow-flow imaging. The model demonstrated robust predictive performance, as evidenced by an AUC of 0.849, strongly supporting our hypothesis that the comprehensive evaluation of the maternal-placental-fetal axis during the mid-trimester can effectively stratify the risk of subsequent NRDS.

A key finding of our study was the independent predictive value of VFI in predicting postnatal outcomes, particularly NRDS. While conventional Doppler parameters such as UtA-PI and CPR provide crucial insights into vascular resistance and fetal compensation, they primarily assess macrocirculation (36,37). Conversely, our study uniquely integrated 3D-PD slow-flow imaging, adding a fundamental and complementary layer of information by quantifying the functional placental microvasculature. This approach enables a more precise assessment of placental health, with VFI representing an integrated measure of vascularity and blood flow velocity within the placental volume.

Previous research on VFI has largely focused on its ability to predict adverse pregnancy outcomes such as FGR and pre-eclampsia, typically in the prenatal period (38,39). Our study extends these findings by demonstrating that VFI is also a valuable predictor of NRDS after birth. The significant reduction in VFI observed in the NRDS group, even after adjustment for maternal clinical confounders, including perinatal infection, suggests an intrinsic impairment of placental villous perfusion and density that is independently associated with the risk of NRDS. This finding is particularly important, as microcirculatory dysfunction, detectable via slow-flow technology, has not been extensively studied in the context of postnatal respiratory maladaptation. Our results indicate that microcirculatory deficiency is a core component of placental insufficiency that independently contributes to the risk of fetal compromise and postnatal respiratory distress. This provides a unique diagnostic window for placental health that extends beyond traditional resistance indices, emphasizing the utility of VFI in postnatal risk prediction, an area that remains underexplored in the current literature.

Our findings are consistent with those of Sato et al. (40) and Mihu et al. (41), who highlighted that microcirculatory deficiencies, captured effectively by 3D-PD, are key to understanding placental insufficiency. However, unlike previous studies, which focused on prenatal conditions, we demonstrated that VFI can also serve as an independent predictor for postnatal outcomes, specifically NRDS. Similarly, using a different imaging modality, He et al. (42) demonstrated that diffusion-derived vessel density from intravoxel incoherent motion magnetic resonance imaging could serve as a novel biomarker of placental dysfunction in early pre-eclampsia, further supporting the critical role of microvascular assessment in pregnancy complications. Thus, this study represents a valuable contribution to the growing body of research on placental health, providing novel insights into its role in postnatal respiratory maladaptation.

The synergistic interplay among the predictors in our model closely reflects the established pathophysiological cascade of progressive placental insufficiency. Increased UtA-PI represents the “upstream” event of impaired uteroplacental perfusion, often resulting from defective spiral artery remodeling (43). This initial insult triggers the “mid-stream” fetal compensatory mechanism of brain-sparing, reflected by a decreased CPR, which prioritizes blood flow to vital organs (44). However, when the placental pathology extends to the microvascular level, as captured by a low VFI, the efficiency of gas and nutrient exchange is severely compromised. This multi-parametric model thus captures the progression from maternal vascular malperfusion to fetal hemodynamic redistribution and, ultimately, to microcirculatory failure, thereby providing a more comprehensive risk assessment than any single parameter alone.

It is critical to contextualize the specific pathophysiology of respiratory distress observed in our cohort. Because our study protocol excluded neonates delivered before 34 weeks of gestation, the population consisted entirely of late-preterm and term infants. Consequently, the pathogenesis of NRDS observed in this study is less consistent with primary extreme lung immaturity—the classic absolute surfactant deficiency typically seen in extreme prematurity. Instead, the respiratory morbidity observed in this study aligns more closely with secondary respiratory maladaptation, which includes secondary surfactant dysfunction, delayed lung fluid clearance, and acute lung injury akin to acute respiratory distress syndrome (or “shock lungs”). This secondary dysfunction is closely intertwined with the pathophysiological cascade described above; chronic fetal hypoxia, fetal stress, and potential systemic inflammatory responses are directly triggered by the altered uteroplacental hemodynamics and microcirculatory ischemic disease, which our model objectively captures via increased UtA-PI and reduced VFI.

The inclusion of gestational hypertension as a powerful independent predictor (OR =3.08) in our model underscores the critical role of the maternal clinical context in assessing pregnancy outcomes. The identification of pre-eclampsia and FGR as significant clinical features in our NRDS cohort further reinforces the “pathogenic scenario” of placental insufficiency. Gestational hypertension and pre-eclampsia are widely recognized as key drivers of placental dysfunction, contributing to endothelial damage and an exacerbated ischemic environment (45,46). The high incidence of FGR in the NRDS group (29.3%) aligns with our findings of reduced VFI, collectively indicating chronic placental microcirculatory failure. Further, the significantly higher rate of induced or elective labor (60.3%) in the NRDS group reflects the clinical necessity for iatrogenic delivery in cases of severe placental compromise. This suggests that the increased respiratory risk is likely a combined result of both the underlying placental pathology and the higher frequency of early, non-spontaneous deliveries.

Our findings align with those of Melchiorre et al. (47), who found a similar association between maternal hypertension and placental insufficiency. However, our study integrated imaging biomarkers, which, when combined with maternal characteristics such as hypertension, provide a more comprehensive and individualized risk assessment. This approach facilitates the detection of subtle placental changes that may not be detected by traditional clinical measurements alone, offering a more nuanced understanding of the relationship between maternal health and fetal wellbeing. In comparison to studies focusing primarily on the maternal clinical context (48-50), our findings highlight the synergistic value of combining maternal characteristics with advanced imaging techniques. This suggests that imaging biomarkers, when interpreted in conjunction with maternal clinical factors, provide a more robust and personalized evaluation of pregnancy risk. This approach may enable more accurate early detection and improve risk stratification, ultimately guiding clinical decision-making and targeted interventions for high-risk pregnancies.

The nomogram developed to predict NRDS risk integrates four independent predictors identified through multivariate logistic regression: UtA-PI, CPR, VFI, and gestational hypertension. Each predictor contributes a corresponding score, determined by mapping the patient’s measured value to its respective scale. The total score, ranging from 0 to 330, is then converted into an estimated probability of NRDS, providing clinicians with a straightforward, quantitative assessment of risk. This novel scoring system simplifies the complex, multi-faceted pathophysiology of NRDS into a practical, user-friendly tool, enabling efficient and intuitive risk evaluation.

This approach represents a significant advance in personalized medicine, as it moves beyond the traditional binary classification of pregnancy risk (normal/abnormal) to a more continuous and granular risk assessment. For example, a pregnant woman with specific values for UtA-PI, CPR, VFI, and gestational hypertension can be assigned a total score that directly corresponds to her likelihood of developing NRDS. The clinical utility of our nomogram lies in its ability to provide actionable insights, enabling clinicians to tailor management strategies for high-risk pregnancies. Patients identified as high-risk by the nomogram may benefit from targeted interventions such as enhanced monitoring, scheduled delivery at specialized centers, and timely administration of antenatal corticosteroids, potentially reducing the incidence and severity of NRDS and improving neonatal outcomes.

The nomogram demonstrated excellent performance in predicting NRDS risk, as shown by the ROC plot with an AUC of 0.849 (95% CI: 0.745–0.871), indicating strong discriminative ability. At a sensitivity of 82%, it identified most high-risk cases, with a specificity of 78%. When specificity was prioritized at 85%, sensitivity remained high at 75%, showcasing the nomogram’s robustness in balancing detection against unnecessary interventions. The calibration curves further confirmed the model’s reliability, with a mean absolute error of only 0.046, indicating strong agreement between the predicted and actual outcomes. The calibration plots showed that the predicted risk closely matched the observed NRDS incidence, even after bias correction.

DCA highlighted the nomogram’s clinical utility, demonstrating a higher net benefit than both “treat all” and “treat none” strategies across threshold probabilities ranging from 0.05 to 0.75. At a threshold probability of 0.3, commonly used for intervention, the nomogram achieved a net benefit of 0.25, compared with 0.12 for “treat all” and 0.03 for “treat none”, supporting its superior clinical effectiveness in guiding risk-based interventions. To summarize, the nomogram is a reliable, accurate, and clinically useful tool, offering a personalized approach to NRDS risk assessment that may improve decision-making while reducing unnecessary interventions.

Several limitations of this study warrant consideration. First, the retrospective, single-center design may introduce selection bias and limit the generalizability of our findings; thus, external validation in prospective, multicenter cohorts is essential. Second, the acquisition and analysis of 3D-PD volumes are operator-dependent and require specific expertise, which may pose challenges for widespread implementation. Finally, while the sample size, particularly the number of NRDS cases, was sufficient for initial model development, larger cohorts would be valuable in future studies to enhance model stability and enable more complex analyses.

Despite these limitations, our study has several notable strengths. Its primary innovation lies in the synergistic integration of traditional spectral Doppler with advanced slow-flow imaging to capture both macrovascular resistance and microvascular perfusion in the same predictive framework. Further, the integration of these detailed ultrasound parameters with a key clinical variable provides a comprehensive risk profile. The use of rigorous statistical methods, including bootstrap validation, further supports the robustness and clinical potential of the derived nomogram.

In summary, this study demonstrated that a multi-parametric assessment of the fetoplacental unit in the second trimester, integrating conventional Doppler with slow-flow imaging technologies, can effectively predict the risk of NRDS. The developed nomogram provides a comprehensive, visual, and clinically actionable tool for early risk stratification. The identification of high-risk pregnancies well before the onset of clinical symptoms creates a critical window for enhanced monitoring and personalized intervention strategies with the ultimate aim of improving perinatal outcomes. Future research should focus on the external validation of this model and on investigating whether its implementation in clinical practice leads to a measurable reduction in neonatal morbidity.

Conclusions

Integrating second-trimester Doppler indices, slow-flow placental perfusion (e.g., VFI), and key maternal factors enables early risk stratification for NRDS via a practical nomogram. Prospective multicenter studies are needed to confirm the generalizability and clinical impact of our model.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0143/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0143/dss

Funding: This study was supported by Hubei Provincial Natural Science Foundation (No. 2025AFB845) and Graduate Innovation and Entrepreneurship Fund of Wuhan University of Science and Technology (No. JCX2024044).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0143/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethical Committee of Xiangyang No. 1 People’s Hospital (No. 2021KYLX02; date of approval: January 14, 2022), and all patients provided written informed consent for fetal ultrasonography.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Edwards MO, Kotecha SJ, Kotecha S. Respiratory distress of the term newborn infant. Paediatr Respir Rev 2013;14:29-36; quiz 36-7. [Crossref] [PubMed]
2. Polin RA, Carlo WACommittee on Fetus and Newborn. American Academy of Pediatrics. Surfactant replacement therapy for preterm and term neonates with respiratory distress. Pediatrics 2014;133:156-63. [Crossref] [PubMed]
3. De Luca D. Respiratory distress syndrome in preterm neonates in the era of precision medicine: A modern critical care-based approach. Pediatr Neonatol 2021;62:S3-9. [Crossref] [PubMed]
4. Rubarth LB, Quinn J. Respiratory Development and Respiratory Distress Syndrome. Neonatal Netw 2015;34:231-8. [Crossref] [PubMed]
5. Mohammadi A, De Luca D, Gauda EB. Characteristics, triggers, treatments, and experimental models of neonatal acute respiratory distress syndrome. Am J Physiol Lung Cell Mol Physiol 2025;328:L512-25. [Crossref] [PubMed]
6. Cresi F, Maggiora E, Lista G, Dani C, Borgione SM, Spada E, Ferroglio M, Bertino E, Coscia AENTARES Study Group. Effect of Nasal Continuous Positive Airway Pressure vs Heated Humidified High-Flow Nasal Cannula on Feeding Intolerance in Preterm Infants With Respiratory Distress Syndrome: The ENTARES Randomized Clinical Trial. JAMA Netw Open 2023;6:e2323052. [Crossref] [PubMed]
7. Soll R, Ozek E. Multiple versus single doses of exogenous surfactant for the prevention or treatment of neonatal respiratory distress syndrome. Cochrane Database Syst Rev 2009;CD000141.
8. Sweet D, Bevilacqua G, Carnielli V, Greisen G, Plavka R, Saugstad OD, Simeoni U, Speer CP. Valls-I-Soler A, Halliday H; Working Group on Prematurity of the World Association of Perinatal Medicine; European Association of Perinatal Medicine. European consensus guidelines on the management of neonatal respiratory distress syndrome. J Perinat Med 2007;35:175-86. [Crossref] [PubMed]
9. Pinette MG, Wax J. Risk assessment for neonatal respiratory distress syndrome with FLM II combined with gestational age. Am J Obstet Gynecol 2005;193:1879-author reply 1880. [Crossref] [PubMed]
10. Singh R, Munian LP, Memela NA. Management of neonates with respiratory distress syndrome in resource-limited settings. S Afr Fam Pract (2004) 2024;66:e1-7. [Crossref] [PubMed]
11. Menuet C, Wlodek ME, Fong AY, Allen AM. Respiratory modulation of sympathetic nerve activity is enhanced in male rat offspring following uteroplacental insufficiency. Respir Physiol Neurobiol 2016;226:147-51. [Crossref] [PubMed]
12. Basak S, Duttaroy AK. Maternal PUFAs, Placental Epigenetics, and Their Relevance to Fetal Growth and Brain Development. Reprod Sci 2023;30:408-27. [Crossref] [PubMed]
13. Chen S, Shenoy A. Placental Pathology and the Developing Brain. Semin Pediatr Neurol 2022;42:100975. [Crossref] [PubMed]
14. Silva PIP, Perez M. Prenatal Ultrasound Diagnosis of Biometric changes in the Brain of Growth Restricted Fetuses. A Systematic Review of Literature. Rev Bras Ginecol Obstet 2021;43:545-59.
15. Ramos DR, Araujo Júnior E, Petrini CG, Dulgheroff FF, Caldas TMRDC, Peixoto AB. Increased pulsatility index of uterine artery Doppler between 26 and 28 weeks of gestation and adverse perinatal outcomes. J Matern Fetal Neonatal Med 2022;35:4810-7. [Crossref] [PubMed]
16. Di Mascio D, Rizzo G, Buca D, D'Amico A, Leombroni M, Tinari S, Giancotti A, Muzii L, Nappi L, Liberati M, D'Antonio F. Comparison between cerebroplacental ratio and umbilicocerebral ratio in predicting adverse perinatal outcome at term. Eur J Obstet Gynecol Reprod Biol 2020;252:439-43. [Crossref] [PubMed]
17. Kalafat E, Khalil A. Clinical significance of cerebroplacental ratio. Curr Opin Obstet Gynecol 2018;30:344-54. [Crossref] [PubMed]
18. Moreta D, Vo S, Eslick GD, Benzie R. Re-evaluating the role of cerebroplacental ratio in predicting adverse perinatal outcome. Eur J Obstet Gynecol Reprod Biol 2019;242:17-28. [Crossref] [PubMed]
19. McMaster-Fay RA. Cerebroplacental ratio assessment in low-risk population. Am J Obstet Gynecol 2019;221:659. [Crossref] [PubMed]
20. Elmes C, Phillips R. Systematic review evaluating the efficacy of the cerebroplacental ratio (CPR) in saving babies lives. Ultrasound 2022;30:184-93. [Crossref] [PubMed]
21. Cardinali F, Panunzi C, D'Antonio F, Khalil A, Spinillo A, Arossa A, Familiari A, Pagani G, Resta S, Rizzo G. Role of Cerebroplacental Ratio in Predicting the Outcome of Pregnancies Complicated by Diabetes. Fetal Diagn Ther 2024;51:55-65. [Crossref] [PubMed]
22. Karge A, Lobmaier SM, Haller B, Kuschel B, Ortiz JU. Value of Cerebroplacental Ratio and Uterine Artery Doppler as Predictors of Adverse Perinatal Outcome in Very Small for Gestational Age at Term Fetuses. J Clin Med 2022;11:3852. [Crossref] [PubMed]
23. Schiffer VMMM, Pellaers D, Hoenen JLJM, van Kuijk SMJ, Spaanderman MEA, Al-Nasiry S. Feasibility of three dimensional power Doppler ultrasonography methods to assess placental perfusion. Eur J Obstet Gynecol Reprod Biol 2020;254:321-8. [Crossref] [PubMed]
24. Schiffer VMMM, Pellaers D, Hoenen JLJM, van Kuijk SMJ, Spaanderman MEA, Al-Nasiry S. Feasibility of three dimensional power Doppler ultrasonography methods to assess placental perfusion. Eur J Obstet Gynecol Reprod Biol 2020;254:321-8. [Crossref] [PubMed]
25. Liu Y, Gao X, Shi D, Wang Y, Qi X, Wang N. Changes of placental three-dimensional power Doppler ultrasonography in third trimester among hypertensive disorders complicating pregnancy. Biomol Biomed 2023;23:894-901. [Crossref] [PubMed]
26. Sun W, Liu J, Zhang Y, Cai A, Yang Z, Zhao Y, Wang Y, Cao Z, Wei Q. Quantitative assessment of placental perfusion by three-dimensional power Doppler ultrasound for twins with selective intrauterine growth restriction in one twin. Eur J Obstet Gynecol Reprod Biol 2018;226:15-20. [Crossref] [PubMed]
27. Ushida T, Nakamura N, Nakatochi M, Kobayashi Y, Sato Y, Iitani Y, Imai K, Nakano-Kobayashi T, Hayakawa M, Kajiyama H, Kotani TNeonatal Research Network of Japan. Impact of hypertensive disorders of pregnancy on respiratory outcomes in extremely and very preterm infants: A population-based study in Japan. Pregnancy Hypertens 2022;29:54-60. [Crossref] [PubMed]
28. Tian T, Wang L, Ye R, Liu J, Ren A. Maternal hypertension, preeclampsia, and risk of neonatal respiratory disorders in a large-prospective cohort study. Pregnancy Hypertens 2020;19:131-7. [Crossref] [PubMed]
29. Bernardes TP, Zwertbroek EF, Broekhuijsen K, Koopmans C, Boers K, Owens M, Thornton J, van Pampus MG, Scherjon SA, Wallace K, Langenveld J, van den Berg PP, Franssen MTM, Mol BWJ, Groen H. Delivery or expectant management for prevention of adverse maternal and neonatal outcomes in hypertensive disorders of pregnancy: an individual participant data meta-analysis. Ultrasound Obstet Gynecol 2019;53:443-53. [Crossref] [PubMed]
30. Bromfield SG, Ma Q, DeVries A, Inglis T, Gordon AS. The association between hypertensive disorders during pregnancy and maternal and neonatal outcomes: a retrospective claims analysis. BMC Pregnancy Childbirth 2023;23:514. [Crossref] [PubMed]
31. Garovic VD, Dechend R, Easterling T, Karumanchi SA, McMurtry Baird S, Magee LA, Rana S, Vermunt JV, August PAmerican Heart Association Council on Hypertension. Council on the Kidney in Cardiovascular Disease, Kidney in Heart Disease Science Committee; Council on Arteriosclerosis, Thrombosis and Vascular Biology; Council on Lifestyle and Cardiometabolic Health; Council on Peripheral Vascular Disease; and Stroke Council. Hypertension in Pregnancy: Diagnosis, Blood Pressure Goals, and Pharmacotherapy: A Scientific Statement From the American Heart Association. Hypertension 2022;79:e21-41. Erratum in: Hypertension 2022;79:e70.
32. Phipps EA, Thadhani R, Benzing T, Karumanchi SA. Pre-eclampsia: pathogenesis, novel diagnostics and therapies. Nat Rev Nephrol 2019;15:275-89. [Crossref] [PubMed]
33. ACOG Practice Bulletin No. 190: Gestational Diabetes Mellitus. Obstet Gynecol 2018;131:e49-64. [Crossref] [PubMed]
34. Fetal Growth Restriction. ACOG Practice Bulletin, Number 227. Obstet Gynecol 2021;137:e16-28.
35. WHO Guidelines Approved by the Guidelines Review Committee. WHO Recommendations for Prevention and Treatment of Maternal Peripartum Infections. Geneva: World Health Organization; 2015.
36. Taylor TJ, Quinton AE, de Vries BS, Hyett JA. Uterine Artery Pulsatility Index Assessment at <11 Weeks' Gestation: A Prospective Study. Fetal Diagn Ther 2020;47:129-37. [Crossref] [PubMed]
37. Bonnevier A, Maršál K, Brodszki J, Thuring A, Källén K. Cerebroplacental ratio as predictor of adverse perinatal outcome in the third trimester. Acta Obstet Gynecol Scand 2021;100:497-503. [Crossref] [PubMed]
38. Noguchi J, Hata K, Tanaka H, Hata T. Placental vascular sonobiopsy using three-dimensional power Doppler ultrasound in normal and growth restricted fetuses. Placenta 2009;30:391-7. [Crossref] [PubMed]
39. Odibo AO, Goetzinger KR, Huster KM, Christiansen JK, Odibo L, Tuuli MG. Placental volume and vascular flow assessed by 3D power Doppler and adverse pregnancy outcomes. Placenta 2011;32:230-4. [Crossref] [PubMed]
40. Sato M, Noguchi J, Mashima M, Tanaka H, Hata T. 3D power Doppler ultrasound assessment of placental perfusion during uterine contraction in labor. Placenta 2016;45:32-6. [Crossref] [PubMed]
41. Mihu CM, Drugan T, Mihu D. Contribution of 3D power Doppler ultrasound to the evaluation of placental circulation in normal pregnancies and pregnancies complicated by preeclampsia. J Perinat Med 2012;40:359-64. [Crossref] [PubMed]
42. He J, Chen C, Xu L, Xiao B, Chen Z, Wen T, Wáng YXJ, Liu P. Diffusion-Derived Vessel Density Computed From a Simplified Intravoxel Incoherent Motion Imaging Protocol in Pregnancies Complicated by Early Preeclampsia: A Novel Biomarker of Placental Dysfunction. Hypertension 2023;80:1658-67. [Crossref] [PubMed]
43. de Ganzo Suárez T, de Paco Matallana C, Plasencia W. Spiral, uterine artery doppler and placental ultrasound in relation to preeclampsia. Best Pract Res Clin Obstet Gynaecol 2024;92:102426. [Crossref] [PubMed]
44. Bendall A, Schreiber V, Crawford K, Kumar S. Predictive utility of the fetal cerebroplacental ratio for hypoxic ischaemic encephalopathy, severe neonatal morbidity and perinatal mortality in late-preterm and term infants. Aust N Z J Obstet Gynaecol 2023;63:491-8. [Crossref] [PubMed]
45. Redman CW, Sargent IL. Pre-eclampsia, the placenta and the maternal systemic inflammatory response--a review. Placenta 2003;24 Suppl A:S21-7.
46. Zhou M, Feng Y, Zhang C, Tian X, Li M, Zheng Y. Relationship Between Glucose/Lipid Metabolism and Placental Biomarkers in Gestational Diabetes and Preeclampsia. Diabetes Metab Syndr Obes 2025;18:691-702. [Crossref] [PubMed]
47. Melchiorre K, Giorgione V, Thilaganathan B. The placenta and preeclampsia: villain or victim? Am J Obstet Gynecol 2022;226:S954-62. [Crossref] [PubMed]
48. Liu L, Sun D. Pregnancy outcomes in patients with primary antiphospholipid syndrome: A systematic review and meta-analysis. Medicine (Baltimore) 2019;98:e15733. [Crossref] [PubMed]
49. Peris M, Crompton K, Shepherd DA, Amor DJ. The association between human chorionic gonadotropin and adverse pregnancy outcomes: a systematic review and meta-analysis. Am J Obstet Gynecol 2024;230:118-84. [Crossref] [PubMed]
50. Yan X, Zheng W, Wang J, Yuan X, Li G. Low-dose aspirin for the prevention of preterm birth in nulliparous women: systematic review and meta-analysis. BMC Pregnancy Childbirth 2024;24:260. [Crossref] [PubMed]