Objectives: To determine if the use of magnetic resonance imaging (MRI) changes the diagnosis of placenta accreta spectrum (PAS) made on prenatal ultrasound (US) leading to an improvement in clinical outcomes.
Methods: This was a retrospective chart review of all patients with evidence of PAS on US from 2012 to 2018 in one tertiary care medical center with subsequent use of MRI of the uterus to confirm diagnosis. The type of PAS classified by imaging was compared between US and MRI, with a final diagnosis made using histology. Outcomes that were analyzed included the following: 1) MRI correctly changed diagnosis, 2) MRI incorrectly downgraded diagnosis, 3) MRI incorrectly upgraded diagnosis, and 4) MRI did not change diagnosis. A T-test and Chi-squared test were performed to compare the clinical outcomes of patients with an upgraded diagnosis by MRI to those whose diagnosis was downgraded or stayed the same.
Results: Forty-one patients received an MRI to validate the diagnosis of PAS after ultrasound and are included in the analysis. MRI changed the diagnosis in 36.6% (15/41) patients, correctly changing the diagnosis in 22% (9/41) and incorrectly upgrading the diagnosis in 14.6% (6/41). Patients whose diagnosis was upgraded by MRI (either correctly or incorrectly) were more likely to deliver earlier compared to those who were either downgraded or had no change in their diagnosis [33. 2 ± 3. 5 weeks vs 35. 2 ± 2. 9 weeks, p = 0.05]. Patients who were upgraded were more likely to have interventional radiology and/or urology involvement at the time of delivery [91.7% (11/12) vs 25. 9% (7/27), p = 0.001]. There were no complications from these procedures.
Conclusion: The use of MRI incorrectly changed the diagnosis as much as it correctly changed the diagnosis of PAS after US. MRI should not be used routinely as a clinical adjunct to ultrasound in the diagnosis of placenta accreta spectrum.
Keywords: Placenta accreta spectrum; cesarean; hysterectomy; increta; magnetic resonance imaging; percreta; ultrasound.