[What may be expected from endorectal ultrasonography and magnetic resonance imaging in the assessment of local extension of cancer of the prostate?]

Abstract

30 patients presenting with a clinical stage A or B cancer underwent TRUS and MRI with a body coil prior to radical prostatectomy. Imaging and pathology were correlated to assess the limits of TRUS and MRI in pre-operative staging of prostatic carcinoma. Results showed that TRUS and MRI had a sensitivity of 42 and 66% and a specificity of 89 and 94% respectively, for the diagnosis of macroscopic capsular effraction. Seminal vesicle invasion was diagnosed by TRUS with a sensitivity of 16% and a specificity of 81%. With MRI, the sensitivity and the specificity were 85 and 94% if only gross infiltration was considered. Sensitivity dropped to 45% if microscopic invasion was included. We conclude that: (1) postero-lateral venous plexus are more easily delineated by MRI, permitting a more accurate diagnosis of capsular penetration. (2) Seminal vesicle invasion cannot be diagnosed by TRUS as the echostructure of the cancer and the caudal junction of the seminal vesicles and vas deferens are similar. (3) A normal caudal junction on TRUS or a normal bilateral hypersignal of the seminal vesicles on MRI are associated with seminal vesicle invasion in less than 5% of cases. (4) A normal caudal junction on TRUS eliminates false positive cases of MRI related to non tumoral hyposignals of the seminal vesicles.