High prevalence of intrapelvic parasitic arteries in patients with placenta accreta spectrum: A case-control study using unenhanced magnetic resonance angiography.

PURPOSE: To compare the prevalence of enlarged ovarian and intrapelvic parasitic arteries to the gravid uterus between cases of placenta accreta spectrum (PAS) and those with normal placentation using unenhanced magnetic resonance (MR) angiography. METHODS: Unenhanced time-of-flight MR angiography was performed in 12 consecutive women with PAS (mean age, 34 years; range, 23-42 years) and 24 women with normal placentation (mean age, 31 years; range, 24-42 years) in their third trimester and reviewed by two independent observers. The consensus reading served as the reference standard. Findings of pelvic arteriography performed at cesarean hysterectomy were reviewed in all cases of PAS. The prevalence of enlarged ovarian and intrapelvic parasitic arteries was compared using Fisher's exact test. The interobserver agreement was assessed with Kappa statistics. RESULTS: The prevalence of enlarged ovarian arteries was not significantly different between cases of PAS and normal placentation (17% [4/24 pelvic sides] vs. 4% [2/48 pelvic sides], P = .091). The prevalence of intrapelvic parasitic arteries was significantly higher in cases of PAS than in those with normal placentation (67% [16/24 pelvic sides] vs. 0% [0/48 pelvic sides], P < .0001). On a patient-by-patient basis, the intrapelvic parasitic artery was frequently present in women with PAS (92% [11/12 patients]). The Kappa values were 0.915 and 0.852 for detecting enlarged ovarian and intrapelvic parasitic arteries, respectively, indicating excellent interobserver agreement. CONCLUSIONS: The development of intrapelvic parasitic arteries was an anomalous phenomenon observed on unenhanced MR angiography in the majority of women with PAS but was not observed in those with normal placentation.