Outcomes of pregnancies at high-risk for placenta accreta spectrum following negative diagnostic imaging

Objectives To assess the pretest and negative post-test probability for placenta accreta spectrum (PAS) in a group of patients with high-risk clinical factors. Methods We included patients with suspected and/or confirmed PAS at our institution over 8 years. Sonography performed by maternal-fetal medicine specialists, and selected patients underwent MRI. Imaging was considered positive if either sonography or MRI suggested PAS. Histopathology was the gold standard for diagnosis of PAS. We assessed the pretest and negative imaging-test probability, and resources required. Results We identified 82 high-risk patients with the following: (1) a history of >= 1 cesarean section and/or intrauterine gynecologic procedure and placenta previa in the index pregnancy; (2) a history of >3 cesarean deliveries and/or gynecologic procedures regardless of placental location; (3) prior PAS disorder, or retained placenta requiring manual extraction and/or curettage, complicated by postpartum hemorrhage; and (4) suspected cesarean section scar pregnancy. Histopathology confirmed PAS in 52 patients, with pretest probability of 63%. Imaging correctly identified 44/50 cases with PAS, and excluded this condition in 24/30 cases. Thus, the positive and negative post-test probability for PAS following negative imaging was 88 and 20%, respectively. Of the six patients with false-negative imaging, all had either surgical complications or required care beyond that for routine cesarean section. Conclusions Although diagnostic imaging is sensitive, the negative posttest probability remains high in women with high pretest probability for PAS. Therefore, women at high risk for PAS should be managed in experienced centers by a multidisciplinary team even if imaging is negative.