Assessment of Sensitivity and Specificity of Ultrasound and Magnetic Resonance Imaging in the Diagnosis of Placenta Accreta.

Objective:  To assess and compare the sensitivity and specificity of ultrasonography and magnetic resonance imaging in the diagnosis of placenta accreta in patients with placenta previa.

Methods:  This retrospective cohort study included 37 women, and was conducted between January 2013 and October 2015; 16 out of the 37 women suffered from placenta accreta. Histopathology was considered the gold standard for the diagnosis of placenta accreta; in its absence, a description of the intraoperative findings was used. The associations among the variables were investigated using the Pearson chi-squared test and the Mann-Whitney -test.

Results:  The mean age of the patients was 31.8 ± 7.3 years, the mean number of pregnancies was 2.8 ± 1.1, the mean number of births was 1.4 ± 0.7, and the mean number of previous cesarean sections was 1.2 ± 0.8. Patients with placenta accreta had a higher frequency of history of cesarean section than those without it (63.6% versus 36.4% respectively;  < 0.001). The mean gestational age at birth among women diagnosed with placenta previa accreta was 35.4 ± 1.1 weeks. The mean birth weight was 2,635.9 ± 374.1 g. The sensitivity of the ultrasound was 87.5%, with a positive predictive value (PPV) of 65.1%, and a negative predictive value (NPV) of 75.0%. The sensitivity of the magnetic resonance imaging was 92.9%, with a PPV of 76.5%, and a NPV of 75.0%. The kappa coefficient of agreement between the 2 tests was 0.69 (95% confidence interval [95%CI]: (0.26-1.00).

Conclusion:  The ultrasound and the magnetic resonance imaging showed similar sensitivity and specificity for the diagnosis of placenta accreta.