Objective: The objective of this study was to evaluate the ability of transrectal surface coil MR imaging to reveal local recurrence of malignancy in men who have had radical prostatectomy for prostatic adenocarcinoma.
Subjects and methods: We performed prospective analysis of 41 men who had undergone radical prostatectomy (range of time since surgery, 8 months to 5 years; mean, 26 months), 35 of whom had clinical suspicion of recurrent prostatic cancer and the remaining six of whom had no clinical evidence of recurrent prostatic cancer (controls). Our imaging used a transrectal surface coil on a 1.5-T MR scanner. Sagittal and axial fat-saturated T2-weighted fast spin-echo as well as axial T1-weighted unenhanced and gadolinium-enhanced MR images of the prostatic bed were acquired in all patients. Thirty-one of the 35 men with clinical suspicion of recurrent prostatic cancer had elevated prostate-specific antigen (PSA) levels (> or = 0.4 ng/ml), and 22 of these 31 men had a palpable prostatic bed module or induration. The four of 35 men with clinical suspicion of recurrent prostatic cancer who had PSA levels less than 0.4 ng/ml had a palpable prostatic bed nodule or induration. Transrectal biopsy of the prostatic bed was directed by digital palpation or transrectal sonography in all 35 men with clinical suspicion of recurrent malignancy.
Results: Thirty-one of the 35 men who had clinical suspicion of local recurrence of prostatic cancer had a soft-tissue nodule revealed in the prostatic bed by transrectal surface coil MR imaging. Compared with the adjacent muscle, all nodules were isointense on the T1-weighted images, hyperintense on the T2-weighted images, and enhanced with gadolinium administration. The 22 patients who had an abnormal MR scan and a palpable nodule or induration and the nine patients with elevated PSA levels, no palpable abnormality, and an abnormality revealed by MR imaging underwent transrectal biopsy; all had recurrent prostatic cancer proven by histology in the four patients with a palpable prostatic bed nodule or induration and normal PSA levels, MR imaging showed no distinct soft-tissue nodule or area of enhancement in the prostatic bed; transrectal biopsy of the palpable nodule or induration yielded fibrosis but no malignancy in all four patients. In the six control patients with no clinical evidence of local recurrence, MR imaging revealed no evidence of recurrent malignancy; all six control patients continue to have no clinical evidence of recurrent prostatic cancer with a minimum follow-up of 22 months. Thus, the sensitivity of MR imaging in revealing local recurrence of prostatic cancer was 100% (95% confidence interval = 89-100%), and the specificity also was 100% (95% confidence interval = 69-100%). The kappa coefficient was 1.0 (p < .001).
Conclusion: MR imaging with a transrectal surface coil is a useful imaging tool to evaluate men who have undergone radical prostatectomy and are suspected of having local recurrence of malignancy in the prostatic bed.