NEAUA Main Site   | Past & Future Meetings
The New England Section of the American Urological Association
Meeting
Home
Accreditation
Information
Preliminary
Program
Registration
Information
Housing
Information
Exhibitors &
Sponsors
Local Area
Attractions

Back to 2016 Annual Meeting


Limited accuracy of pre-operative MRI for evaluation of extra-prostatic extension and seminal vesicle invasion in prostate cancer patients undergoing radical prostatectomy
Urszula Kowalik, MD, Peter A. Holoch, MD, Scott D. Perrapato, DO, Brian Irwin, MD, Mark K. Plante, MD, FRCS(C), FACS.
University of Vermont Medical Center, Burlington, VT, USA.

Background: MRI is increasingly being used to stage patients with prostate cancer to better define disease state in patients deemed candidates for radical prostatectomy. We evaluated the accuracy and predictive value (NPV and PPV) of pre-operative MRI for extra-prostatic extension (EPE) and seminal vesicle invasion (SVI) based on post-operative pathology results in a sample of prostate cancer patients who underwent radical prostatectomy.
Methods: Between 2013 and 2015, 224 patients underwent robotic-assisted laparoscopic prostatectomy (RALP) at the University of Vermont Medical Center. Of those, 62 underwent a pre-operative dedicated prostate 3T MRI. Indications for MRI included both patients who met criteria for pre-operative staging imaging by NCCN guidelines and patients with a prior biopsy and persistently elevated or rising PSA. One patient was excluded, having had the MRI within 4 weeks of biopsy, as the evaluation of EPE could have been confounded by immediate post-biopsy changes. MRI reports were considered positive for EPE or SVI if findings were reported as suspicious or positive. EPE and SVI reported on whole-mount pathology reports were used to determine the negative predictive value and accuracy of MRI.
Results: Of the 61 patients, 9 were found to have EPE on MRI. Comparison with whole-mount specimens revealed a NPV of 38%, with a sensitivity of 18% and specificity of 90%. Accuracy of MRI for predicting EPE was calculated at 44%. Regarding SVI, the NPV was 69%, with a sensitivity of 30% and specificity of 95%. Although our sample of patients with positive EPE on MRI was small (n=9), the PPV was 78%.
Conclusions: Our findings suggest that MRI prediction of EPE may not be as universally accurate as described in prior reports where NPV has been estimated as high as 95%. This may be related to the relatively recent adoption of PI RADS criteria and new protocols for prostate MR at our institution. While there may be a role for MRI in the diagnosis of significant prostate cancer, its role in the pre-operative evaluation of patients undergoing RALP warrants further investigation.


Back to 2016 Annual Meeting


Old Port
Bay by Holiday Inn
Lighthouse
© 2022 New England Section of the American Urological Association. All Rights Reserved. Privacy Policy.
Banner photo courtesty of Holiday Inn By the Bay © InterContinental Hotels.