Correlation of placental MR imaging signs and pathologic diagnosis of placenta accreta spectrum: Retrospective single center case series

Introduction: Alongside initial screening obstetric US, use of placental MRI has been increasing in the last few decades to aid with antenatal diagnosis and delivery planning in Placenta Accreta Spectrum (PAS). The aim of this study was to determine if the MRI pathophysiological sign subcategories described in the current literature can predict the severity of pathologic diagnosis. Methods: Institutional imaging records were reviewed for placental MRIs performed for suspicion of PAS in the last decade. Electronic health records were searched for patient history and pathology. The 59 MRI studies were reviewed using the 11 MRI signs described by the SAR and ESUR joint consensus statement. Further breakdown of the signs was divided by underlying pathophysiologic subcategories including gross morphologic, interface and tissue architecture signs. Results: Pathologic diagnosis yielded 34 cases: accreta 4/34, incerta 14/34, percreta 10/34 and normal 6/ 34. Of the accreta cases all of them demonstrated at least two interface and half of the cases had tissue architecture signs, 13/14 increta cases demonstrated interface signs and 12/14 demonstrated tissue architecture signs, 9/10 percreta cases had two interface and at least six demonstrated three tissue architecture signs. Statistical analysis showed significant difference between pathologic diagnosis and the number of positive interface signs with p = 0.02. Discussion: Interface signs were the most objective and sensitive MRI subcategory. Statistical analysis determined there was a significant difference between PAS diagnosis and number of interface signs present. This subcategory has the most overlap with classic US signs which are traditionally used before MRI referral. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

This study reviewed 59 placental MRIs for suspected PAS and found that interface signs were the most sensitive subcategory, showing significant differences between PAS diagnosis and the number of interface signs present.