A comprehensive severity score for the morbidly adherent placenta: combining ultrasound and magnetic resonance imaging.

Abstract

Ultrasound (US) is the first-line imaging modality to assess the morbidly adherent placenta, but sensitivity and specificity are lacking. This investigation aims to improve diagnostic accuracy with a comprehensive score using clinical history, US, and magnetic resonance imaging (MRI). We conducted a retrospective cohort study of pregnant women who received both transvaginal US and MRI with suspicion for morbidly adherent placenta between 2009 and 2016. US was scored with the following metrics: (i) previa, (ii) hypervascularity, (iii) loss of retroplacental clear space and (iv) lacunae. MRI was evaluated for (i) intraparenchymal vessels, (ii) abnormally dilated vessels, (iii) fibrin deposition, (iv) placental bulge and (v) bladder dome irregularity. Bayesian analysis was used to estimate the probability of morbidly adherent placenta for a given score. Diagnostic testing parameters were calculated. Among the 41 women with concerning imaging, histologically identified disease was confirmed in 16. The probability of morbidly adherent placenta increased with the score. At the highest US score, the probability of disease was 63.7%. With the highest MRI score, the probability of adherent placentation was 90.5%. Combining the US and MRI findings had a sensitivity of 56% and a specificity of 92%. A combined scoring system using MRI and US may accurately identify patients at risk for morbidity associated with morbidly adherent placenta.