Obstetric ultrasound (US) is the most widely used instrument for the prenatal diagnosis of placental accreta spectrum (PAS), with a variable efficacy and frequent false positives and negatives. We report the diagnostic effectiveness of US in PAS, using a “pro-form” for US report and analyse the factors associated with inaccurate diagnoses. A retrospective, observational study conducted at a tertiary obstetric centre, between 2017-2020. Patients with PAS suspicion, in whom ultrasonographic 2D and Doppler were performed by a perinatologist trained in PAS, were included. The ultrasound images were evaluated looking for the predictive capacity of the US, taking into account the signs included in it. 101 patients were included. The diagnosis was confirmed by intraoperative findings and-or histopathological analysis in 51 of them. The prenatal ultrasound analysis, using the European pro-form criteria, showed a sensitivity of 88.37% with a specificity of 79.31%. The most-reported grayscale ultrasound parameters were the loss of “clear zone” 72,5%, followed by abnormal placenta lacunae presence 68.6%, and myometrial thinning. Moreover, the most frequent colour Doppler ultrasound parameters were the sub placental hypervascularity 64,7% followed by uterovesical hypervascularity 56,9%. 12 patients with intraoperative or histological diagnosis of PAS, missed the prenatal ultrasound diagnosis, 4 of them had placenta posterior, and 2 others had giant myomatosis. Prenatal ultrasonography is a useful tool for the detection of PAS. The European pro forma for ultrasonographic assessment of patients with PAS is a useful tool that allows a uniform record of findings facilitating communication de interdisciplinary management groups.