DIAGNOSTIC ACCURACY OF MAGNETIC RESONANCE IMAGING IN PRENATAL PLACENTA ACCRETA SPECTRUM: A PROSPECTIVE COHORT STUDY OF 412 CASES

Dr. Emily Thompson; MD; Prof. Robert Anderson; MD

Abstract

Placenta Accreta Spectrum (PAS) represents a life-threatening obstetric complication characterized by morbidly adherent placenta to the myometrium. While ultrasonography remains the primary screening modality, its sensitivity is often operator-dependent, particularly in posterior implantations. This prospective cohort study evaluates the diagnostic precision of Magnetic Resonance Imaging (MRI) in characterizing the depth and topography of placental invasion. By analyzing 412 high-risk pregnancies, we compare prenatal MRI findings with intraoperative observations and histopathological confirmation, the gold standard. Our data indicates that MRI provides superior visualization of parametrial involvement and thinned myometrium, which are critical for surgical mapping. The integration of T2-weighted imaging and Diffusion-Weighted Imaging (DWI) significantly enhances the identification of intra-placental dark bands, a hallmark of PAS. This research underscores the necessity of a standardized MRI grading system to improve surgical outcomes, reduce massive transfusion requirements, and facilitate planned multidisciplinary interventions. The findings advocate for MRI as a secondary diagnostic tool in inconclusive ultrasound cases to mitigate maternal morbidity.

Keywords: Placenta Accreta Spectrum, Magnetic Resonance Imaging, Prenatal Diagnosis, Maternal Morbidity, Obstetric Imaging, Placenta Percreta, Radiological Accuracy, Hysterectomy

The escalating global rate of cesarean deliveries has led to a concomitant rise in Placenta Accreta Spectrum (PAS), which now occurs in approximately 1 in 533 pregnancies. PAS is a leading cause of intractable postpartum hemorrhage and emergency peripartum hysterectomy, carrying a maternal mortality rate as high as 7% in some regions. Accurate prenatal diagnosis is the cornerstone of risk stratification, allowing for the mobilization of multidisciplinary teams, including urologists, interventional radiologists, and pelvic surgeons. Despite its clinical importance, the optimal imaging strategy remains debated. This study addresses a critical gap in obstetric radiology by quantifying the incremental diagnostic value of MRI. By providing a more precise topographical map of invasion, MRI enables surgeons to avoid placental transection and plan conservative management where appropriate. Enhancing diagnostic accuracy is essential for minimizing massive blood transfusions and improving localized healthcare resource allocation in high-risk obstetric cases.

To evaluate the diagnostic accuracy, sensitivity, and specificity of 1.5T Magnetic Resonance Imaging in the prenatal identification and staging of placenta accreta spectrum disorders in 412 high-risk patients.

This prospective observational study enrolled 412 pregnant women (mean age 32.4 ± 4.5 years) between 2019 and 2023 who were referred to a tertiary care center with suspected PAS based on prior ultrasonography or clinical risk factors (e.g., previous cesarean sections). MRI was performed between 28 and 34 weeks of gestation using a 1.5-Tesla scanner without intravenous gadolinium. Imaging protocols included T2-weighted HASTE sequences in three anatomical planes and Diffusion-Weighted Imaging (DWI). Two blinded senior radiologists independently reviewed the images for predefined markers: placental bulges, intra-placental T2 dark bands, abnormal placental vascularity, and myometrial thinning. Results were correlated with intraoperative findings and histopathological examination of the placenta or hysterectomy specimens. Statistical analysis utilized ROC curves, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Cohen’s kappa coefficient was calculated to assess inter-observer reliability.

Out of 412 participants, PAS was histologically confirmed in 154 cases (37.3%). MRI demonstrated an overall sensitivity of 91.5% (95% CI: 86.2%–95.4%) and a specificity of 88.4% (95% CI: 83.9%–92.0%). The diagnostic accuracy was calculated at 89.6%. For the subset of placenta percreta (n=42), MRI showed a significantly higher sensitivity (95.2%) compared to Doppler ultrasound (78.6%, p<0.01). The most reliable MRI sign was the presence of intra-placental T2 dark bands (Odds Ratio [OR]: 7.4; 95% CI: 4.2–12.9) and placental bulging (OR: 5.8; 95% CI: 3.1–10.5). Inter-observer agreement was substantial, with a Cohen’s kappa of 0.82. However, 30 false positives were recorded, primarily attributed to focal myometrial thinning in patients with multiple repeat cesarean sections. The Negative Predictive Value (NPV) for bladder invasion was exceptionally high at 97.4%, indicating MRI's strength in ruling out advanced pelvic organ involvement.

Magnetic Resonance Imaging demonstrates high diagnostic accuracy for the prenatal detection of PAS, particularly in staging higher-grade invasions such as placenta percreta. The high negative predictive value suggests that MRI is an effective tool for excluding parametrial involvement, thereby refining surgical planning and reducing the incidence of unexpected intraoperative hemorrhages. While ultrasound remains the frontline tool, MRI’s superior anatomical resolution provides critical topographical details that facilitate multidisciplinary surgical approaches. Future efforts should focus on standardized reporting protocols to further reduce the rate of false positives observed in cases of simple placenta previa. This study confirms that MRI is an indispensable adjunct in the management of morbidly adherent placenta, directly contributing to improved maternal outcomes through precise preoperative mapping.