81 - Assessment of Pre-procedure MR Angiography Compared to Iliac Artery Cone-beam CT for the Detection of Prostatic Arteries and Pelvic Collateral Vasculature

Purpose: Effective and safe prostatic artery embolization (PAE) requires a detailed understanding of pelvic arterial anatomy. Prior studies demonstrated that MR angiography (MRA) can successfully identify the prostatic arteries (PAs) and reduce the procedure time {1,2}. However, a direct comparison of pre-procedure MRA to PAE using intra-procedure internal iliac artery (IIA) cone-beam CT (CBCT) has not been previously reported. This evaluation is important to determine if pre-procedure MRA is warranted before PAE.

Materials and Methods: A retrospective cohort of 35 patients underwent PAE between April 2023 and February 2024 utilizing both pre-procedure MRA and IIA CBCT. Images were reviewed by two board-certified interventional radiologists in a blinded manner. For each imaging modality, the number and origins of the prostatic arteries (PA) on each side and the presence of penile and rectal collateral pathways were recorded. Additional qualitative parameters (e.g. image quality and confidence in identifying PAs) were scored on a Likert scale. A McNemar test was used to compare the detectability of the PAs origins and collaterals between MRA and CBCT. Qualitative parameters were compared across imaging modalities using a Wilcoxon signed-rank test.

Results: Mean patient age was 77.5 years (std dev: 6.9), and the mean prostate size of 82.9 ml (std dev: 34.2). Of the 99 PA origins identified using CBCT, 87 were recognized on MRA (sensitivity 87.9%, p >0.5). Of the 52 collateral vessels identified on CBCT, only 18 were recognized on MRA (sensitivity 34.6%, p< 0.00001). A few additional PA origins and collaterals were identified on MRA that were not seen on CBCT, but these were often not recapitulated on DSA (3/11 for PA origins and 3/5 for collaterals). Image quality and confidence in identifying all feeding vessels were significantly lower in the MRA data set compared to the CBCT data set (p< 0.00001 and p=0.015 respectively).

Conclusion: IIA CBCT performed at the time of PAE demonstrated improved image quality with higher sensitivity for identifying PA origins and important collateral vasculature when compared to MRA. The additional time, cost, and expertise needed for the acquisition and interpretation of pre-procedure MRA may not be justified to evaluate pelvic arterial anatomy for procedure planning.