Assessment of the risk of positive surgical margins with pelvic phased-array magnetic resonance imaging in patients with clinically localized prostate cancer: a prospective study.

OBJECTIVES

We assessed magnetic resonance imaging (MRI) performance in the prediction of positive surgical margins (PSMs) before radical prostatectomy in a prospective study correlating the MRI results and pathologic findings.

METHODS

Between January 1995 and December 1999, 176 patients (mean age 64.2 years, range 49 to 75), with localized prostate cancer (49 with Stage T1 and 127 with Stage T2) underwent preoperative MRI with a pelvic phased-array coil (Tesla-1, Siemens) at a mean interval of 35 days after randomized transrectal biopsies. The mean preoperative prostate-specific antigen level was 10.9 ng/mL (range 1.2 to 39). The MRI studies and specimen analysis were performed by one radiologist unaware of the clinical and biopsy findings and by one pathologist, respectively. Multivariate analysis was performed to compare the predictive value of MRI staging, prostate-specific antigen value, and preoperative Gleason score to identify the PSM rate.

RESULTS

Of the 176 patients, 131 (74%) had Stage T2 disease by MRI and 45 (26%) Stage T3 disease by MRI. Pathologic staging showed 103 with pT2 and 73 with pT3. Overall, the PSM rate of the series was 18%. The PSM rate was 13.7% and 31% for patients with T2 and T3 disease by MRI, respectively. For the T3 MRI cases, the PSM rate was 2.32-fold higher. MRI staging, like the prostate-specific antigen value, was a predictive factor of PSMs (P = 0.05).

CONCLUSIONS

The results of this study show that preoperative MRI staging with the phased-array coil may be helpful in predicting the PSM risk in radical prostatectomy candidates with clinically localized prostate cancer.