Assessment of uteroplacental perfusion with 3D power Doppler for the early prediction of pre-eclampsia: a systematic review and meta-analysis

Introduction

Pre-eclampsia (PE) is a multisystem disorder that complicates about 3–5% of all pregnancies, and can lead to significant mortality and morbidity among mothers, fetuses, and newborns (1). The early identification of at-risk women enables the initiation of preventive interventions, such as low-dose aspirin, which has been shown to reduce the incidence of PE when initiated before 16 weeks of gestation (2,3). Further, the risk reduction for preterm PE is likely to be even greater, as PE has no curative treatment other than delivery, and is thus associated with a significant risk of iatrogenic prematurity (3). Thus, it is essential to correctly identify the risk of PE in early pregnancies, offering a window for interventions before irreversible placental damage occurs.

Abnormal placentation, primarily resulting from inadequate placental implantation and subsequent malperfusion (4-6), contributes significantly to the pathophysiology of PE (7), and is histologically characterized by infarction, aberrant placental villi development, decreased placental size, and the insufficient development of maternal decidual spiral arterioles (4). Maternal vascular malperfusion histopathological severity is positively associated with the severity of maternal-fetal PE symptoms, but is negatively associated with gestational age at delivery (1,8,9).

Current screening strategies comprise clinical risk factors, maternal biochemical markers, and ultrasound examinations (10). Numerous studies have used the uterine artery pulsatility index (UtA-PI) or the uterine artery resistance index (UtA-RI) as predictors of PE (11-14). Previous research has shown that first-trimester uterine artery Doppler has limited sensitivity when maintaining a false-positive rate of 10% (10). Thus, new uteroplacental perfusion markers need to be developed and validated.

Beyond ultrasonographic techniques, emerging magnetic resonance imaging (MRI) approaches have shown promise in quantifying placental dysfunction in PE. Specifically, diffusion-derived vessel density, a biomarker derived from diffusion-weighted MRI that captures microvascular abnormalities, has shown high diagnostic performance in differentiating between PE-associated placentas and normal pregnancies (15,16). However, the limited accessibility and higher costs of MRI currently restrict its utility for first-trimester screening, and Doppler remains more feasible for widespread clinical implementation.

The three-dimensional power Doppler (3D-PD) ultrasonographic technique enables the volumetric assessment of vascular networks in a defined region of interest (ROI). It has shown utility in assessing endometrial receptivity in infertility and quantifying tumor vascularity in oncology (17-19). For PE, 3D-PD may better reflect early placental malperfusion, and its uteroplacental vascularization indices (VIs) [i.e., the VI, flow index (FI), and vascularization flow index (VFI)], are more strongly correlated with histopathological markers of placental insufficiency than conventional Doppler, which relies on single-vessel waveforms (20-22).

Numerous studies have examined the efficacy of 3D-PD for PE prediction, and suggested that the uteroplacental perfusion indices might serve as predictors of future PE (23-31). One meta-analysis by Eastwood et al. (32) assessed the mean value of each index, and found that the first-trimester uteroplacental VIs were decreased in women who eventually developed PE (32). However, their meta-analysis was restricted to a small sample size of three primary investigations, and thus failed to fully encapsulate the research in this domain. Further, they did not assess the predictive performance of any indicator due to a lack of sufficient information from diagnostic studies. Consequently, our review sought to assess the uteroplacental VIs via 3D-PD in the first trimester for the purpose of predicting PE. We present this article in accordance with the PRISMA reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-158/rc).

Methods

The protocol for this review has been entered into International Prospective Register of Systematic Reviews (PROSPERO, CRD42024600360).

Data sources and search strategy

Two researchers systematically searched the PubMed, Web of Science, the Cochrane Library, and Embase databases to retrieve relevant articles published up to October 22, 2024. Similar search terms were employed without any restrictions on language. Each term was defined using a mix of medical subject headings (MeSH) terms and keywords like “placenta”, “ultrasonography”, and “PE” to maximize the number of articles identified (Table S1). The researchers also searched the applicable article reference lists manually. Following the removal of duplicates, the two researchers independently screened the titles and abstracts, eliminated extraneous research, and subsequently examined the full texts of the potentially relevant articles. Any disagreements were resolved by a third researcher.

Articles were included in the meta-analysis if they met the following inclusion criteria: (I) examined singleton pregnancies; (II) used 3D-PD to assess uteroplacental VIs, including the VI, FI and VFI, at 11+0–13+6 weeks of gestation; and (III) the primary outcome (PE or non-PE) was reported, inclusive of all subtypes. Articles were excluded from the meta-analysis if they met any of the following exclusion criteria: (I) concerned animal research; (II) comprised a review, comment, case report, or conference abstract; (III) examined twin or multiple pregnancies; (IV) did not report the required outcome variables; and/or (V) lacked sufficient data to establish a two-by-two diagnostic table.

Data extraction and quality assessment

Two reviewers individually extracted information from the qualifying articles applying a pre-designed data collection form. The recorded data included the first author’s name, publication year, participant quantity and characteristics, type of study, and mean value of each index. The sensitivity and specificity values of the prognostic studies were obtained if available. Afterwards, a two-by-two diagnostic table was generated by calculating true-positive, true-negative, false-positive, and false-negative counts. In the absence of this information, the area under the curve (AUC) values of the receiver operating characteristic (ROC) curves and/or odds ratios were extracted. Relevant authors were contacted as necessary to obtain data.

Both investigators independently assessed the clinical applicability and the risk of bias of each article, applying an adapted Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool (33) (Table S2). Any disagreements were resolved by a third investigator after considering both views.

Analytical strategy and statistical analysis

This meta-analysis sought to examine:

• between-group differences to compare the mean values of the 3D-PD indices between women who subsequently developed PE and those who did not. To harmonize the data obtained from different ultrasound machines and parameter settings, we applied the standardized mean difference (SMD) as the pooled effect size metric;
• predictive accuracy to synthesize the predictive performance (e.g., sensitivity and specificity) of these indices for PE.

The statistical analysis was conducted using the following three software programs: Review Manager version 5.3 (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark), Meta-DiSc 1.4 (Ramóny Cajal Hospital, Madrid, Spain), and Stata version 14.0 (StataCorp, College Station, TX, USA).

Research heterogeneity included threshold effect and non-threshold effect heterogeneity. To evaluate the threshold effect over trials, we calculated the Spearman correlation coefficient between logit sensitivity and logit (1-specificity). A P value <0.05 indicated a threshold effect. The I-square (I²) statistics and the Cochran-Q test were calculated to measure non-threshold effect heterogeneity. Fixed-effect models were employed in the absence of heterogeneity (I²<50%; Cochran Q, P>0.10); otherwise random-effects models were used.

The summarized forest plots were presented, and the SMD of each index was calculated between the PE and non-PE groups. The sensitivity, specificity, 95% confidence interval (CI), negative likelihood ratio (LR–), and positive likelihood ratio (LR+) values of the individual studies were concurrently combined. The summarized AUC value was calculated after plotting the summary receiver operating characteristic (sROC) curve. A P value <0.05 indicated statistical significance. In addition, the sensitivity analysis was conducted by the systematic elimination of individual studies, and publication bias was assessed by plotting a Deeks funnel plot.

Results

Literature search

In total, 986 studies were retrieved, of which, 213 records were identified as duplicates and one record was retracted. Following the evaluation of the titles and abstracts, 43 full-text articles were considered eligible after 729 records were eliminated. Ultimately, this review included 11 studies comprising 9,622 participants in the qualitative synthesis and meta-analysis. Figure 1 shows the literature screening flow chart.

Figure 1 Flowchart for research selection using the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.

Study attributes

Table 1 summarizes the fundamental features of the 11 studies. The entire count of the participants in the PE groups varied from 6 to 163, while that in the control groups varied from 12 to 4,263. Of the 11 studies analyzed, 10 were prospective cohort studies in which the incidence of PE ranged from 1.4% to 12%. In another prospective case-control study conducted by Hashish et al. (30), the high-risk subgroup had a high PE incidence of 76%. Among the 11 included studies, five defined PE using the American College of Obstetrics and Gynecology criteria (36,37), five defined PE using the International Society for the Study of Hypertension in Pregnancy criteria (38), and one employed a composite definition. Of the 11 studies, 10 used placental vascularization indices (PVIs) of the entire placenta, including the PVI, Placental Flow Index (PFI), and Placental Vascularization Flow Index (PVFI). Three studies used placental bed vascularization indices (PBVIs) of the placental bed (i.e., the placenta-adjacent myometrium with a maximum thickness of 1 cm or less) (31), including the Placental Bed Vascularization Index (PBVI), Placental Bed Flow Index (PBFI), and Placental Bed Vascularization Flow Index (PBVFI). Six studies provided adequate data to evaluate the predictive value of the PVI and PFI, and five studies provided adequate data to evaluate the predictive value of the PVFI. Only two studies reported the sensitivity and specificity values for the PBVI, and only one study reported these values for the PBFI and PBVFI.

Quality assessment

The research quality was evaluated using an adapted QUADAS-2 (33) tool. As Figure 2 shows, the 11 studies were of moderate quality. In relation to the risk of bias of the index test domain, none of the thresholds were prespecified. In relation to the flow and timing domain, nine investigations excluded some individuals from the final analysis for various reasons. The majority of the studies did not provide information as to whether the research was conducted in a blinded manner. Additionally, one study (35) included high-risk pregnancies as the participants, while another case-control study (30) categorized the participants into high-risk and low-risk groups.

Figure 2 Quality assessment outcomes using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool. Ratings (A) and percentages (B) of the included studies with a low, unclear, or high risk of bias.

Differences in the uteroplacental VIs between the PE and non-PE groups

The case-control study by Hashish et al. (30), compared the VIs between high-risk and low-risk groups, irrespective of the ultimate PE diagnosis. Consequently, its data could not be harmonized for PE/non-PE group contrasts and it was excluded from this part of the meta-analysis.

As detailed in Table 1, this part of the meta-analysis examined nine original studies on the PVI, seven on the PFI and PVFI, and three on the PBVI. As significant heterogeneity was found in each index (I²=97%>75%, Cochrane Q=252.93, P<0.1), a random-effects model was employed. As Figure 3 shows, the PE groups had significantly decreased mean PVI, PFI, PVFI, and PBVI values, and the SMDs between the PE and non-PE groups were –0.94 for the PVI, –0.94 for the PFI, –0.76 for the PVFI, and –0.53 for the PBVI (P<0.05).

Figure 3 Forest plots applying random-effects models were used to compare mean the uteroplacental vascularization indices in the PE and non-PE groups. CI, confidence interval; NPE, non-PE; PBVI, Placental Bed Vascularization Index; PE, pre-eclampsia; PFI, Placental Flow Index; PVFI, Placental Vascularization Flow Index; PVI, Placental Vascularization Index; SD, standard deviation.

A sensitivity analysis was conducted via the systematic removal of individual articles to investigate the source of heterogeneity in the PVIs. The results indicated that the studies by Sweed et al. (25) and Abdallah et al. (28) might be the principle source of heterogeneity among the included research. After excluding the two studies, the I² reduced to 36% with P=0.16>0.1, thus a re-analysis was performed by applying a fixed-effects model (Figure S1). Consistent with previous findings, it remained evident that the PE groups had significantly decreased values of PVIs in early pregnancies, and the SMD values between the two groups were –0.36 for the PVI, –0.50 for the PFI, and –0.39 for the PVFI (P<0.05).

Diagnostic performance of the uteroplacental VIs in predicting PE

As detailed in Table 1, this part of the meta-analysis examined six original studies on the PVI and PFI, and five studies on the PVFI. The pooled results for the three indices all exhibited acceptable predictive accuracy with a P value <0.05. We could not assess the predictive efficacy of the PBVIs (i.e., the PBVI, PBFI, and PBVFI) as only Hannaford et al.’s (31) study included the relevant data.

Threshold effect

The Meta-DiSc 1.4 program (Ramóny Cajal Hospital, Madrid, Spain) was used for the data analysis. The Spearman correlation coefficient of the PVI between the logarithm of sensitivity and the logarithm of (1-specificity) was –0.036 (P=0.939>0.05), while that of the PFI was –0.357 (P=0.432>0.05), and that of PVFI was 0.429 (P=0.397>0.05), illustrating the absence of threshold effects in this part of the analysis. Moreover, the lack of a “shoulder-arm” shape in the sROC curve further proved the absence of a threshold effect.

Non-threshold effect

Using the Cochran-Q test, we determine the diagnostic odds ratio (DOR) for the PVI (Cochran Q=67.25, P<0.001), PFI (Cochran Q=34.76, P<0.001), and PVFI (Cochran Q=44.04, P<0.001), and the results revealed heterogeneity resulting from a non-threshold effect. Moreover, the I² for the DOR, sensitivity, specificity, LR–, and LR+ of each indicator were greater than 75%, indicating substantial heterogeneity. Thus, the following meta-analysis was conducted using a random-effects model.

Predictive performance of the indicators

The aggregated sensitivity, specificity, LR+, LR–, Q*, AUC, and DOR of the PVI (Figure 4) were 0.78 (95% CI: 0.73–0.82), 0.82 (95% CI: 0.81–0.83), 3.86 (95% CI: 2.61–5.70), 0.31 (95% CI: 0.14–0.70), 0.8256, 0.8947, and 14.01 (95% CI: 3.92–50.11), those of the PFI (Figure S2) were 0.68 (95% CI: 0.62–0.73), 0.82 (95% CI: 0.81–0.84), 4.06 (95% CI: 2.56–6.44), 0.42 (95% CI: 0.24–0.73), 0.8241, 0.8933, and 9.64 (95% CI: 4.17–22.28), and those of the PVFI (Figure S3) were 0.83 (95% CI: 0.79–0.87), 0.75 (95% CI: 0.74–0.77), 3.71 (95% CI: 2.97–4.64), 0.22 (95% CI: 0.07–0.66), 0.8207, 0.8900, and 20.97 (95% CI: 4.79–91.77).

Figure 4 Forest plots of the (A) sensitivity, (B) specificity, (C) diagnostic odds ratio, and (D) the sROC curves of the PVI studies. AUC, area under the curve; CI, confidence interval; OR, odds ratio; PVI, Placental Vascularization Index; SE, standard error; sROC, summary receiver operating characteristic.

Sensitivity analysis

The data were imported into Stata 14.0 software (StataCorp, College Station, TX, USA) for the sensitivity analysis (Figure S4). The results showed that Dar et al.’s study (24) exhibited a strong sensitivity for the PVI studies, while Abdallah et al.’s study (28) exhibited a strong sensitivity for the PFI studies. The remaining studies would not make the results of PVI and PFI studies sensitive. No individual study significantly influenced the pooled estimates of the PVFI. Taken together, the results for predictive performance were comparatively stable.

Publication bias

Deeks publication bias test revealed a P value >0.05, and the funnel plots for the three indicators were all symmetric (Figure S5), implying the absence of significant publication bias in this meta-analysis.

Conflicts

None of the studies declared conflicts of interest related to financial support, institutional affiliations, or non-financial relationships relevant to their studies.

Discussion

The aggregated findings revealed that pregnant women who eventually developed PE had statistically lower mean values of the uteroplacental VIs obtained from 3D-PD in the first trimester. In addition, the PVI, PFI, and PVFI showed acceptable predictive accuracy, with moderate AUC values of 0.8947, 0.8933, and 0.8900, respectively, indicating that the decrease in the PVIs may predict the subsequent onset of PE.

Current early screening techniques for PE rely on a synthesis of maternal biochemical markers, clinical risk factors, and uterine artery Doppler (1). While conventional two-dimensional Doppler parameters (e.g., UtA-PI and UtA-RI) continue to be widely used to assess placental perfusion, their reliance on operator-dependent vessel sampling and single-waveform analysis limits their reproducibility and spatial resolution. Conversely, 3D-PD indices (i.e., the VI, FI, and VFI) provide volumetric assessments of placental perfusion, capturing the entire vascular network in a defined ROI. This reduces sampling biases and improves inter-observer reliability. Clinically, 3D-PD indices may better reflect early placental malperfusion, which is a key precursor to PE. Our pooled results showed that the PVI had moderate predictive accuracy (AUC: 0.8947) in the first trimester, compared to the UtA-PI (AUC <0.6) in previous studies (10,39), suggesting its superior discriminatory power. However, ROI placement and machine settings need to be standardized if these techniques are to be implemented in clinical practice.

For each index, the VI signifies the percentage of color voxels in the total voxels, the FI represents the ratio of the signal intensity of the entire color voxels to the color voxel count, while the VFI represents the ratio of the signal intensity of the color voxels to the entire tissue voxels (25). Consequently, low PVI values indicate fewer placental vessels. A decrease in the PFI indicates a decrease in placental blood flow resulting from increased vascular resistance. A decreased PVFI level indicates a reduction in both the quantity and blood flow of the vessels in the placenta. The decline in vessel quantity, velocity, and intensity, and placental perfusion, represented by a decrease in each of the three indexes, respectively, represents uteroplacental malperfusion and is consistent with the pathology of PE.

Contrary to our review, Neto et al. (23) found no significant difference in the PFI and PVFI between 11⁺⁰ to 13⁺⁶ weeks in the PE and non-PE groups. The research by González-González et al. (29) and Hannaford et al. (31) similarly showed no statistically significant differences in the PVIs (i.e., the PVI, PFI, and PVFI) between the two groups. The limited sample sizes of these studies might explain the conflicting data with the pooled results.

Two studies reported a negative correlation between the VIs and maternal body mass index (BMI) (26,34). In addition, Hafner et al. (34) reported that the PFI and PBFI were significantly correlated with the thickness of the maternal abdominal wall, and the maternal BMI significantly negatively affected the flow indices but only minimally affected the VIs (i.e., the PVI, PBVI, PVFI, and PBVFI). This indicates that in obese women, the flow indicators are mostly depth-dependent and comparatively inaccurate. Conversely, the VIs remain relatively independent from maternal obesity and are more reliable. Moreover, compared to the PVIs (i.e., the PVI, PFI, and PVFI), the PBVI exhibited minimal intra-observer variability and was less affected by the maternal BMI, indicating that the PBVI is the most reliable measurement and better reflects uteroplacental perfusion. Only three studies (26,31,34) examined the PBVIs. In addition, only two (23,28) of the three articles reported the sensitivity and specificity values. Thus, we were unable to summarize the predictive value of the PBVI for PE.

According to the sensitivity analysis, the studies of Dar et al. (24) and Sweed et al. (25), were the primary contributors of heterogeneity in the prediction accuracy analysis. This might be due to several reasons, including the use of distinct techniques to measure the VIs, various machine settings, and regional population variations. Previous studies have shown that machine parameters alone can significantly affect the outcomes of 3D-PD (40,41).

Previous research has established that the predictive performance of any single factor, such as clinical risk or maternal serum biomarkers, is modest (42,43). Similarly, any individual 3D-PD index should not serve as a single-marker prediction model. González-González et al. (29) found that the incorporation of the VI, FI, or VFI into an existing prediction model based on baseline maternal characteristics did not result in any significant improvement in the AUC. Eastwood et al. (35) also integrated each index into the current model without achieving a notable increase in the AUC; nonetheless, both the Net Reclassification Index (NRI) and Integrated Discrimination Improvement (IDI) showed that combining the VI and VFI with established clinical risk factors significantly improved the accurate categorization of PE and non-PE cases. This inconsistency in results might be due to the limitation of the AUC, which cannot typically be used to interpret slight statistical changes, and the correlation between the magnitude of model improvement and the baseline model performance (44).

In previous studies, only one systematic review (32), comprising three original studies, evaluated the mean values of the PVIs (i.e., the PVI, PFI, and PVFI). Our review was innovative for a number of reasons. First, while not the first systematic review of PVIs obtained from 3D-PD, our review was the first to summarize their predictive value in identifying the potential risk of subsequent PE in early pregnancies. Second, our study included more than three times the amount of literature than that of the previous systematic review, enabling a more comprehensive evaluation of the comparison of the mean PVIs in PE and normal pregnancies. Third, we also assessed the placental bed flow perfusion in both groups by analyzing the quantified PBVI.

Limitations

This analysis had certain limitations. First, the findings have to be interpreted with caution given the small number of included studies and their considerable heterogeneity. Second, to minimize the effects of different machines and parameter settings across studies, we applied the SMD for the group comparisons, which is unitless and does not intuitively reflect the magnitude of difference in a certain index, thus precluding the recommendation of summarized cut-off values for each indicator. Third, Hashish et al.’s study adopted a case-control design, and the incidence of PE in its high-risk subgroup was significantly higher than that in the general pregnancy population, which might have inflated the overall predictive values. Additionally, due to the paucity of research, we were unable to conduct a subgroup analysis or assess the value of the PBVI in predicting PE, despite its potential reliability. Further, the uteroplacental VIs alone have limited predictive value; however, this review did not assess their contribution in prediction models. Finally, the included studies varied in adjusting for confounders, such as smoking status, which was not consistently reported across studies.

Conclusions

Our review confirmed that women who subsequently develop PE tend to have low PVI and PBVI values in the first trimester. The PVI, PFI, and PVFI are all valuable non-invasive predictors of PE during early gestation. More clinical trials examining the predictive value of the PBVI for PE, as well as the contribution of incorporating VIs into predictive models, need to be conducted in the future. Further, different ultrasonography machines and diverse populations will require unique reference values. In future research on 3D-PD-derived indices, defined qualitative analysis processes should be established to improve the reliability and repeatability of the findings.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-158/rc

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-158/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.

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. Chappell LC, Cluver CA, Kingdom J, Tong S. Pre-eclampsia. Lancet 2021;398:341-54. [Crossref] [PubMed]
2. Duley L, Meher S, Hunter KE, Seidler AL, Askie LM. Antiplatelet agents for preventing pre-eclampsia and its complications. Cochrane Database Syst Rev 2007;CD004659. [Crossref] [PubMed]
3. Rolnik DL, Nicolaides KH, Poon LC. Prevention of preeclampsia with aspirin. Am J Obstet Gynecol 2022;226:S1108-19. [Crossref] [PubMed]
4. Ernst LM. Maternal vascular malperfusion of the placental bed. APMIS 2018;126:551-60. [Crossref] [PubMed]
5. Khong TY, Mooney EE, Ariel I, Balmus NC, Boyd TK, Brundler MA, et al. Sampling and Definitions of Placental Lesions: Amsterdam Placental Workshop Group Consensus Statement. Arch Pathol Lab Med 2016;140:698-713. [Crossref] [PubMed]
6. Davis DL, Lechner AC, Chapel DB, Slack JC, Carreon CK, Quade BJ, Parra-Herran C. Outcome-Based Risk Stratification Model for the Diagnosis of Placental Maternal Vascular Malperfusion. Mod Pathol 2024;37:100370. [Crossref] [PubMed]
7. Gibbins KJ, Silver RM, Pinar H, Reddy UM, Parker CB, Thorsten V, Willinger M, Dudley DJ, Bukowski R, Saade GR, Koch MA, Conway D, Hogue CJ, Stoll BJ, Goldenberg RL. Stillbirth, hypertensive disorders of pregnancy, and placental pathology. Placenta 2016;43:61-8. [Crossref] [PubMed]
8. Baltajian K, Hecht JL, Wenger JB, Salahuddin S, Verlohren S, Perschel FH, Zsengeller ZK, Thadhani R, Karumanchi SA, Rana S. Placental lesions of vascular insufficiency are associated with anti-angiogenic state in women with preeclampsia. Hypertens Pregnancy 2014;33:427-39. [Crossref] [PubMed]
9. Weiner E, Feldstein O, Tamayev L, Grinstein E, Barber E, Bar J, Schreiber L, Kovo M. Placental histopathological lesions in correlation with neonatal outcome in preeclampsia with and without severe features. Pregnancy Hypertens 2018;12:6-10. [Crossref] [PubMed]
10. Townsend R, Khalil A, Premakumar Y, Allotey J, Snell KIE, Chan C, Chappell LC, Hooper R, Green M, Mol BW, Thilaganathan B, Thangaratinam S. IPPIC Network. Prediction of pre-eclampsia: review of reviews. Ultrasound Obstet Gynecol 2019;54:16-27. [Crossref] [PubMed]
11. Crovetto F, Figueras F, Triunfo S, Crispi F, Rodriguez-Sureda V, Dominguez C, Llurba E, Gratacós E. First trimester screening for early and late preeclampsia based on maternal characteristics, biophysical parameters, and angiogenic factors. Prenat Diagn 2015;35:183-91. [Crossref] [PubMed]
12. Akolekar R, Syngelaki A, Poon L, Wright D, Nicolaides KH. Competing risks model in early screening for preeclampsia by biophysical and biochemical markers. Fetal Diagn Ther 2013;33:8-15. [Crossref] [PubMed]
13. Arakaki T, Hasegawa J, Nakamura M, Hamada S, Muramoto M, Takita H, Ichizuka K, Sekizawa A. Prediction of early- and late-onset pregnancy-induced hypertension using placental volume on three-dimensional ultrasound and uterine artery Doppler. Ultrasound Obstet Gynecol 2015;45:539-43. [Crossref] [PubMed]
14. Mendoza M, Garcia-Manau P, Arévalo S, Avilés M, Serrano B, Sánchez-Durán MÁ, Garcia-Ruiz I, Bonacina E, Carreras E. Diagnostic accuracy of first-trimester combined screening for early-onset and preterm pre-eclampsia at 8-10 compared with 11-13 weeks' gestation. Ultrasound Obstet Gynecol 2021;57:84-90. [Crossref] [PubMed]
15. 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]
16. Li CY, Chen L, Ma FZ, Chen JQ, Zhan YF, Wáng YXJ. High performance of the diffusion magnetic resonance imaging biomarker diffusion-derived 'vessel density' (DDVD) for separating placentas associated with pre-eclampsia from placentas in normal pregnancy. Quant Imaging Med Surg 2025;15:1-14. [Crossref] [PubMed]
17. Ke X, Liang XF, Lin YH, Wang F. Pregnancy prediction via ultrasound-detected endometrial blood for hormone replacement therapy-frozen embryo transfer: a prospective observational study. Reprod Biol Endocrinol 2023;21:112. [Crossref] [PubMed]
18. Zang Z, Lyu J, Yan Y, Zhong M, Zhang Q, Zhang G, Li Y, Yan J. Subendometrial blood flow detected by Doppler ultrasound associates with pregnancy outcomes of frozen embryo transfer in patients with thin endometrium. J Assist Reprod Genet 2024;41:2625-33. [Crossref] [PubMed]
19. Wang H, Yan B, Yue L, He M, Liu Y, Li H. The Diagnostic Value of 3D Power Doppler Ultrasound Combined With VOCAL in the Vascular Distribution of Breast Masses. Acad Radiol 2020;27:198-203. [Crossref] [PubMed]
20. Mathewlynn S, Collins SL. Volume and vascularity: Using ultrasound to unlock the secrets of the first trimester placenta. Placenta 2019;84:32-6. [Crossref] [PubMed]
21. de Vos ES, Koning AHJ, Steegers-Theunissen RPM, Willemsen SP, van Rijn BB, Steegers EAP, Mulders AGMGJ. Assessment of first-trimester utero-placental vascular morphology by 3D power Doppler ultrasound image analysis using a skeletonization algorithm: the Rotterdam Periconception Cohort. Hum Reprod 2022;37:2532-45. [Crossref] [PubMed]
22. Huster KM, Haas K, Schoenborn J, McVean D, Odibo AO. Reproducibility of placental volume and vasculature indices obtained by 3-dimensional power Doppler sonography. J Ultrasound Med 2010;29:911-6. [Crossref] [PubMed]
23. Neto RM, Ramos JG. 3D power Doppler ultrasound in early diagnosis of preeclampsia. Pregnancy Hypertens 2016;6:10-6. [Crossref] [PubMed]
24. Dar P, Gebb J, Reimers L, Bernstein PS, Chazotte C, Merkatz IR. First-trimester 3-dimensional power Doppler of the uteroplacental circulation space: a potential screening method for preeclampsia. Am J Obstet Gynecol 2010;203:238.e1-7. [Crossref] [PubMed]
25. Sweed MS, El-Bishry GA, Abou-Gamrah AE, Abdel-Hamid MM, Nasrel-Din EA, El-Hawwary GE. First-trimester 3D power doppler of uteroplacental circulation and placental volume for the prediction of preeclampsia: A prospective cohort study. Donald School J Ultrasound Obstet Gynecol 2021;15:109-13. [Crossref]
26. Hafner E, Metzenbauer M, Stümpflen I, Waldhör T. Measurement of placental bed vascularization in the first trimester, using 3D-power-Doppler, for the detection of pregnancies at-risk for fetal and maternal complications. Placenta 2013;34:892-8. [Crossref] [PubMed]
27. 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]
28. Abdallah A, Khairy M, Tawfik M, Mohamed S, Abdel-Rasheed M, Salem S, Khalifa E. Role of first-trimester three-dimensional (3D) power Doppler of placental blood flow and 3D placental volume in early prediction of pre-eclampsia. Int J Gynaecol Obstet 2021;154:466-73. [Crossref] [PubMed]
29. González-González NL, González Dávila E, Padrón E, Armas Gonzalez M, Plasencia W. Value of Placental Volume and Vascular Flow Indices as Predictors of Early and Late Preeclampsia at First Trimester. Fetal Diagn Ther 2018;44:256-63. [Crossref] [PubMed]
30. Hashish N, Hassan A, El-Semary A, Gohar R, Youssef MA. Could 3D placental volume and perfusion indices measured at 11-14 weeks predict occurrence of preeclampsia in high-risk pregnant women? J Matern Fetal Neonatal Med 2015;28:1094-8. [Crossref] [PubMed]
31. Hannaford KE, Tuuli M, Goetzinger KR, Odibo L, Cahill AG, Macones G, Odibo AO. First-trimester 3-dimensional power Doppler placental vascularization indices from the whole placenta versus the placental bed to predict preeclampsia: does pregnancy-associated plasma protein a or uterine artery Doppler sonography help? J Ultrasound Med 2015;34:965-70. [Crossref] [PubMed]
32. Eastwood KA, Patterson C, Hunter AJ, McCance DR, Young IS, Holmes VA. Evaluation of the predictive value of placental vascularisation indices derived from 3-Dimensional power Doppler whole placental volume scanning for prediction of pre-eclampsia: A systematic review and meta-analysis. Placenta 2017;51:89-97. [Crossref] [PubMed]
33. Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, Leeflang MM, Sterne JA, Bossuyt PM. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011;155:529-36. [Crossref] [PubMed]
34. Hafner E, Metzenbauer M, Stümpflen I, Waldhör T, Philipp K. First trimester placental and myometrial blood perfusion measured by 3D power Doppler in normal and unfavourable outcome pregnancies. Placenta 2010;31:756-63. [Crossref] [PubMed]
35. Eastwood KA, Hunter AJ, Patterson CC, Mc Cance DR, Young IS, Holmes VA. Placental vascularization indices and prediction of pre-eclampsia in high-risk women. Placenta 2018;70:53-9. [Crossref] [PubMed]
36. Gestational Hypertension and Preeclampsia. ACOG Practice Bulletin, Number 222. Obstet Gynecol 2020;135:e237-60. [Crossref] [PubMed]
37. Cuenca-Gómez D, De Paco Matallana C, Rolle V, Mendoza M, Valiño N, Revello R, Adiego B, Casanova MC, Molina FS, Delgado JL, Wright A, Figueras F, Nicolaides KH, Santacruz B, Gil MM. Comparison of different methods of first-trimester screening for preterm pre-eclampsia: cohort study. Ultrasound Obstet Gynecol 2024;64:57-64. [Crossref] [PubMed]
38. Magee LA, Brown MA, Hall DR, Gupte S, Hennessy A, Karumanchi SA, Kenny LC, McCarthy F, Myers J, Poon LC, Rana S, Saito S, Staff AC, Tsigas E, von Dadelszen P. The 2021 International Society for the Study of Hypertension in Pregnancy classification, diagnosis & management recommendations for international practice. Pregnancy Hypertens 2022;27:148-69. [Crossref] [PubMed]
39. Velauthar L, Plana MN, Kalidindi M, Zamora J, Thilaganathan B, Illanes SE, Khan KS, Aquilina J, Thangaratinam S. First-trimester uterine artery Doppler and adverse pregnancy outcome: a meta-analysis involving 55,974 women. Ultrasound Obstet Gynecol 2014;43:500-7. [Crossref] [PubMed]
40. Raine-Fenning NJ, Nordin NM, Ramnarine KV, Campbell BK, Clewes JS, Perkins A, Johnson IR. Evaluation of the effect of machine settings on quantitative three-dimensional power Doppler angiography: an in-vitro flow phantom experiment. Ultrasound Obstet Gynecol 2008;32:551-9. [Crossref] [PubMed]
41. Jones NW, Hutchinson ES, Brownbill P, Crocker IP, Eccles D, Bugg GJ, Raine-Fenning NJ. In vitro dual perfusion of human placental lobules as a flow phantom to investigate the relationship between fetoplacental flow and quantitative 3D power doppler angiography. Placenta 2009;30:130-5. [Crossref] [PubMed]
42. Tan MY, Wright D, Syngelaki A, Akolekar R, Cicero S, Janga D, Singh M, Greco E, Wright A, Maclagan K, Poon LC, Nicolaides KH. Comparison of diagnostic accuracy of early screening for pre-eclampsia by NICE guidelines and a method combining maternal factors and biomarkers: results of SPREE. Ultrasound Obstet Gynecol 2018;51:743-50. [Crossref] [PubMed]
43. Widmer M, Cuesta C, Khan KS, Conde-Agudelo A, Carroli G, Fusey S, Karumanchi SA, Lapaire O, Lumbiganon P, Sequeira E, Zavaleta N, Frusca T, Gülmezoglu AM, Lindheimer MD. Accuracy of angiogenic biomarkers at ⩽20weeks' gestation in predicting the risk of pre-eclampsia: A WHO multicentre study. Pregnancy Hypertens 2015;5:330-8. [Crossref] [PubMed]
44. Leening MJ, Vedder MM, Witteman JC, Pencina MJ, Steyerberg EW. Net reclassification improvement: computation, interpretation, and controversies: a literature review and clinician's guide. Ann Intern Med 2014;160:122-31. [Crossref] [PubMed]