(985) Magnetic resonance imaging predicts severe placenta accreta spectrum (PAS) resulting in hysterectomy

Recent guidelines have identified magnetic resonance (MR) placenta accreta spectrum (PAS) markers. The purpose of this study is to evaluate these markers ability to predict need for cesarean hysterectomy due to PAS.

Study Design: This was a single-center retrospective study of pregnancies referred with suspected PAS between September 2019-February 2023. MR reports were prospectively interpreted by two investigators, and MR parameters were extracted from the clinical reports, blinded to clinical outcomes. Medical records were reviewed to determine surgical outcomes, and patients were grouped according to the outcome of hysterectomy. Patient demographics were compared between those who required hysterectomy and those that did not using T-tests and Fisher’s Exact test. Univariate logistic regression models and area under the receiver operating characteristics curve (AUC) for MRI variables were generated.

Results: Of 82 pregnant patients with suspicion for PAS and available MR scan, 41 (50%) required cesarean hysterectomy for PAS. No differences in age, race, or gestational age at MR were found (Table 1A). Eleven MR parameters were significantly predictive of the outcome of hysterectomy, with greatest dimension of invasion by MR and radiologist impression the most predictive (Table 1B).

Conclusion: MR parameters correlate to severe PAS requiring cesarean hysterectomy, especially greatest dimension of invasion, radiologist impression of invasion, and placental heterogeneity. Further studies with standardized MR reporting of PAS are needed to compare the predictive value to ultrasound, develop an MRI predictive index for severe PAS, and ultimately automate these processes for reproducibility.