Introduction. Our objective was to elucidate the overall diagnostic accuracy of ultrasound in detecting the severity of abnormally invasive placentation (AIP). Material and methods. Medline, Embase, CINAHL and The Cochrane databases were searched. The ultrasound signs explored were: loss of hypoechoic (clear) zone in the placental-uterine interface, placental lacunae, bladder wall interruption, myometrial thinning, focal exophitic mass, placental lacunar flow, subplacental vascularity, and uterovesical hypervascularity. Results. Twenty studies (3209 pregnancies) were included. Ultrasound had an overall good diagnostic accuracy in identifying the depth of placental invasion with sensitivities of 90.6%, 93.0%, 89.5%, and 81.2% for placenta accreta, increta, accreta/increta, and percreta, respectively; the corresponding specificities were 97.1%, 98.4%, 94.7%, and 98.9%. Placental lacunae had sensitivities of 74.8%, 88.6%, and 76.3% for the detection of placenta accreta, increta, and percreta, respectively. Sensitivity and specificity of loss of the clear zone in identifying placenta accreta were 74.9% and 92.0%, whereas the corresponding figures for placenta increta and percreta were 91.6% and 76.9%, and 88.1% and 71.1%. Lacunar flow had sensitivities of 81.2%, 84.3%, and 45.2% for the detection of placenta accreta, increta, and percreta respectively; the corresponding figures for specificity were 84.0%, 79.7%, and 75.3%. Sensitivity of uterovesical hypervascularity was low for the detection of placenta accreta (12.3%) but high for placenta increta (94.4%) and percreta (86.2%); the corresponding figures for specificity were 90.8%, 88.0% and 88.2%, respectively. Conclusions. Ultrasound has an overall good diagnostic accuracy in recognizing the depth and the topography of placental invasion.