Abstract
BACKGROUND: Postpartum hemorrhage (PPH) represents one of the most frequent and serious complications in placenta accreta spectrum (PAS), making prenatal risk stratification essential for optimizing obstetric management strategies and maternal outcomes. This study aimed to evaluate a novel prenatal ultrasound method for estimating placenta invasion area and to investigate its association with estimated blood loss (EBL in mL) during delivery and the adverse maternal outcomes in PAS. METHODS: This a retrospective cohort study measured PAS area by determining the length of "tramline sign" obliteration and its distance from the cervical os. Placental invasion was segmented into trapezoidal sections using three-dimensional (3D)-Crystal Vue imaging. Linear and multiple regression analyses were used to assess associations between estimated PAS area and EBL, PPH (EBL ≥ 1000 ml), severe PPH (EBL ≥ 2500 ml) and post-delivery transfusion need. RESULTS: Among 168 patients, 78 developed PPH and 90 did not. The PPH group > 2-fold higher PAS area vs. non-PPH (17.28 cm² vs. 7.36 cm²; P < 0.001). Linear regression analysis indicated each 1 cm² PAS area increase corresponded to 42.84 mL higher EBL (95% CI, 27.47-55.77; P < 0.001). PAS area independently predicted PPH (adjusted odds ratio (aOR) 1.08, 95% CI 1.04-1.13; P < 0.001) and severe PPH (aOR 1.03, 95% CI 1.02-1.04; P = 0.03). ROC analysis yielded PAS area cutoffs for PPH (10.13 cm(2); AUC 0.83 (0.76-0.89); P < 0.001) and severe PPH (10.57cm(2); AUC 0.83 (0.76-0.89); P < 0.001). Using these cutoffs, PAS area outperformed classic ultrasound signs in predicting PPH and severe PPH. CONCLUSION: 3D-Crystal Vue-derived PAS area estimation is clinically feasible and correlates with EBL and PPH risk in PAS patients.