Pernicious placenta previa (PPP) complicated by placenta accreta spectrum (PAS) is associated with severe maternal morbidity and remains challenging to diagnose accurately before delivery. To evaluate the diagnostic performance of combining coagulation biomarkers (D-dimer D-D, fibrinogen FIB) with three-dimensional Doppler ultrasound vascular indices for detecting PAS in PPP. In this prospective cohort study, 100 pregnant women with PPP were enrolled (PAS group: n = 66; control group: n = 34). D-D and FIB levels were measured in the second and third trimesters, alongside ultrasound-derived vascular indices, including vascularization index (VI), flow index (FI), and vascular flow index (VFI). Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis, with intraoperative findings and/or histopathological confirmation as the reference standard. Patients with PAS exhibited significantly higher intraoperative blood loss, preterm birth rates, and transfusion requirements compared with controls (all P < .05). In the second trimester, both D-D and FIB levels were elevated in PAS ( P < .05). In the third trimester, D-D levels further increased, whereas FIB levels decreased significantly ( P < .05). Ultrasound vascular indices (VI, FI, and VFI) were consistently higher in the PAS group ( P < .05). The combined model integrating D-D, FIB, and VI/FI/VFI demonstrated improved diagnostic performance, particularly in the third trimester, achieving an area under the curve of 0.952 with a sensitivity of 90.91% and specificity of 91.18%. Subgroup analysis showed that D-D levels increased and FIB levels decreased with increasing PAS severity, whereas ultrasound vascular indices did not differ significantly among PAS subtypes. The integration of coagulation biomarkers with Doppler-derived vascular indices significantly improves diagnostic accuracy for PAS in PPP. This multimodal approach provides a clinically useful tool for preoperative risk stratification and surgical planning.
Gao et al. (Fri,) studied this question.