Pitfalls and Pearls of Prostate Imaging and Interpretation

Abstract

Multiparametric prostate magnetic resonance imaging (mpMRI) has become a mainstay for the detection, characterization, staging, and monitoring of prostate cancer. While the technical specifications, appearance of the normal prostate, assessment (including staging), and reporting of mpMRI have become well documented and codified, reading mpMRI of the prostate is not trivial (Turkbey et al. Eur Urol 76(3):340–351, 2019). Fortunately, many of the “pitfalls and pearls” of mpMRI of the prostate are well documented (Rosenkrantz and Taneja, AJR Am J Roentgenol 202(1):109–120, 2014) Pitfalls can be roughly grouped into three main categories: technical challenges related to diffusion-weighted imaging, normal anatomic structures that may be mistaken for tumor, and noncancerous abnormalities that can mimic tumor. Pearls generally fall into three groups as well: differentiating tumor from benign or normal structures, suspicious findings not explicit in PI-RADS categorization, and extra-prostatic structures. Common pitfalls include anatomic distortion of high b-value diffusion-weighted images, lack of suppression of benign prostate tissue on standard high b-value diffusion-weighted images, suboptimal windowing of the ADC map, the central zone, thickening of surgical capsule, the periprostatic venous plexus and neurovascular bundle, post-biopsy hemorrhage, stromal BPH nodules, acute and chronic prostatitis including post-inflammatory scars and atrophy, and granulomatous prostatitis. Important pearls include anterior fibromuscular stroma and central zone: volume averaging of the surgical capsule at the apex; ejaculatory ducts; tumors are homogeneous; dynamic contrast as a saving grace; the diffusely heterogeneous prostate; how wedge-shaped is “wedge-shaped”?; bladder inlet; seminal vesicle atrophy versus involvement; and non-prostate cancers.