VP44.09: Predictive value of individual ultrasound findings for second or third trimester diagnosis of placenta accreta spectrum in a retrospective population

The purpose of this work was to evaluate individual predictive value and interobserver variability for a comprehensive list of second and third trimester ultrasound findings for antenatal diagnosis of placenta accreta spectrum. A systematic review was performed for ultrasound diagnosis of placenta accreta spectrum and ultrasound findings were consolidated into a single protocol. Ultrasound images of histologically proven placenta accreta spectrum subjects (n = 40) and non-accreta spectrum controls (n = 195) were reviewed by two observers for each of the ultrasound characteristics. Area under the receiver operating characteristics curves were calculated for each ultrasound parameter used as a dichotomous test. Interobserver variability was examined using Cohen's kappa. 638 articles were identified and, after review, thirty-five distinct ultrasound parameters were extracted. ROC curves were created for all ultrasound parameters (figure 1). The placental lacunae (96.3%) and their irregularity (94.1%) of were the most sensitive markers and visualisation of extrauterine placental tissue (97.5%) and parameterial invasion (100%) or hypervascularity (99%) were the most specific. There was a large amount of interobserver variability with only uterovesicular border hypervascularity (kappa = 0.528) having a Cohen's kappa of greater than 0.5. While some ultrasound parameters have excellent value for prediction of placenta accreta spectrum there remains significant interobserver variability. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.