Keywords:
Observer performance, MR, Obstetrics (Pregnancy / birth / postnatal period), Genital / Reproductive system female, Abdomen, Obstetrics
Authors:
M. Stewart, D. Richmond, S. Esler, S. Mooney, N. Israelsohn, L. Churilov, N. Yang; VIC/AU
DOI:
10.1594/ranzcr2018/R-0006
Methods and materials
A retrospective analysis was performed.
Expectant mothers undergoing prenatal MRI for suspected PAD were identified over a 10-year period,
from 1 January 2007 to 31 December 2016.
Patients were included in the analysis where their MRIs were reported by two participating radiologists (NY,
SE).
Both radiologists re-reviewed the MRIs images independently,
blinded to the previous reports as well as the surgical and histological outcomes.
Evaluation included the presence or absence of PAD,
severity (accreta,
increta,
percreta) and various recognised MRI features of PAD.
Intraobserver agreement was calculated using Cohen's kappa coefficient (κ).
The ability of the radiologist to correctly diagnose the severity of placental invasion (using the histopathology as the gold standard) was determined by comparing area under curve (AUC) of the receiver operating characteristic (ROC) using the chi squared test.