Transrectal ultrasound versus magnetic resonance imaging for detection of rectal wall invasion by prostate cancer

BACKGROUND. This study compared the accuracy of transrectal ultrasound (TRUS) versus magnetic resonance imaging (MRI) in the detection of rectal wall involvement by prostate cancer in patients undergoing salvage total pelvic exenteration (TPE) or cystoprostatectomy. METHODS. We identified 16 patients who underwent TPE and 24 patients who underwent cystoprostatectomy for locally advanced prostate cancer as salvage procedures with palliative intent. Patients were examined by TRUS, MRI, or both within the month preceding surgery. Histologic evidence of rectal involvement with prostate cancer was considered the gold standard diagnostic criterion in patients undergoing TPE. Among patients undergoing cystoprostatectomy, posterior prostatic surgical margins and clinical evidence of rectal wall recurrence during a median follow-up duration of 18.6 months were considered the gold standard. The sensitivity, specificity, and overall accuracy with which TRUS and MRI detected rectal wall involvement were compared. RESULTS. Fifteen (93.7%) of the patients who underwent TPE had histologically-proven rectal wall involvement with prostate cancer. Rectal and perineal recurrence developed 10 months after surgery in 1 (4.1%) patient in the cystoprostatectomy group. The sensitivity, specificity, and overall accuracy of TRUS were: 92.9 (66.1-99.8), 87.0 (66.4-97.2), and 89.2 (74.6-97.0), respectively. The sensitivity, specificity, and overall accuracy of MRI were: 54.6 (23.4-83.3), 100 (76.8-100.0), and 80 (59.3-93.2), respectively. CONCLUSIONS. TRUS is a highly sensitive diagnostic modality for rectal wall involvement in patients with locally advanced prostate cancer. Although MRI is very specific, it cannot reliably rule out rectal involvement in the presence of a positive TRUS.