Effectiveness of a fetal magnetic resonance imaging scoring system for predicting the prognosis of pernicious placenta previa: A retrospective study.

Background

The value of multiple magnetic resonance imaging (MRI) signs in predicting pernicious placenta previa (PPP) with placenta accreta spectrum disorders (PAS) is still controversial. This study aimed to investigate the value of a self-made fetal magnetic resonance imaging scoring system in predicting the different types of PAS in pernicious placenta previa and its associated risk of bleeding.

Methods

This retrospective study included 193 patients diagnosed with PPP based on MRI findings before delivery. Based on pathological and intraoperative findings, we divided patients into four groups: non-PAS, placental adhesion, placental implantation, and placenta percreta. Receiver operator characteristic curves of the MRI total score and placental implantation type were drawn using pROC packages in the R Studio environment, and cutoff values of each type were calculated, as well as diagnostic evaluation indexes, such as sensitivity, specificity, and the Youden index. Hemorrhage during surgery was compared between the groups.

Results

The boundary value between the non-PAS and placental adhesion was 5.5, that between placental adhesion and placental implantation was 11.5, and that between placental implantation and placenta percreta was 15.5 points. The respective specificities were 0.700, 0.869, and 0.958, and the respective sensitivities were 0.994, 0.802, and 0.577. The Youden indices were 0.694, 0.671, and 0.535, respectively. The median (minimum, maximum) quantities of hemorrhage during the operation in the non-PAS, placental adhesion, placental implantation, and placenta percreta groups were 225 (100, 3700), 600 (200, 6000), 1500 (300, 7000), and 3000 (400, 6300) ml, respectively. Hemorrhage was significantly different between the four groups (p < 0.001).

Conclusion

These results suggest that the proposed MRI scoring system could be an effective diagnostic tool for assessing PPP types and predicting the associated bleeding risk.