No abstract

represented pregnancies with PAS who delivered by cesarean-hysterectomy and had a first trimester ultrasound with LGSI. Controls represented pregnancies with persistent placenta previas without PAS, who delivered by cesarean-section without post-partum hemorrhage and had a first trimester ultrasound with LGSI. Sonographic images were reviewed by an investigator blinded to pregnancy outcome and sonography reports. Images were reviewed for presence of abnormal utero-placental interface, increased uterovesical hypervascularity, and placental lacunae, with or without swirling on gray scale or color Doppler. Definitions of ultrasound markers followed national task force on PAS. RESULTS: 21 cases and 46 controls met the inclusion criteria. Lacunae were present in 18/21 (85.7%) cases and in 7/46 (15.2%) controls (OR 33.4; 95% CI 7.7-144.4, p<.001). Presence of 3 or more lacunae were 100% predictive of PAS with an average of 5.0 lacunae in cases compared to 1.3 in controls (p¼.019). The average size of the lacunae was 9.88AE3.7mm in cases and 4.41AE0.56 mm in controls (p<.001). Lacunae swirling on gray scale or color Doppler were 100% predictive of PAS (p<.001). Presence of an abnormal uteroplacental interface was also 100% predictive of PAS (p<.001). Uterovesical hypervascularity was present in 14/14 cases and only 1/12 controls (p<.001). CONCLUSION: To our knowledge this is the first study evaluating the predictive value of PAS markers in early gestation in pregnancies with LGSI. In pregnancies with LGSI, presence of 3 or more lacunae or abnormal utero-placental interface in the first trimester is highly predictive of PAS.