Display Settings:


Send to:

Choose Destination
See comment in PubMed Commons below
Eur Urol. 2007 Jul;52(1):98-105. Epub 2007 Jan 22.

Development and split-sample validation of a nomogram predicting the probability of seminal vesicle invasion at radical prostatectomy.

Author information

  • 1Cancer Prognostics and Health Outcomes Unit, University of Montreal, Montreal, Quebec, Canada.



Seminal vesicle preservation may have a beneficial impact on erectile and urinary function after radical prostatectomy (RP). We hypothesized that contemporary patients, at very low risk of seminal vesicle invasion (SVI), can be highly accurately identified with an equally contemporary SVI nomogram.


A nomogram predicting SVI was developed in a cohort study of 666 men diagnosed with 10 or more biopsy cores and treated with RP. Biopsy Gleason sum, prostate-specific antigen, clinical stage, and percentage of positive cores represented predictors in multivariable logistic regression models and formed the basis for the nomogram. The regression coefficient-based nomogram was then externally validated in a split-sample cohort of 230 patients. This cohort also served for a head-to-head comparison of external validity of the novel nomogram with two existing tools, namely Partin's SVI predictions and Koh's SVI nomogram.


Split-sample validation of the novel SVI nomogram demonstrated 79.2% accuracy versus 75.6% for Partin versus 77.7% for Koh. Our nomogram cutoff of 3% or less had 96.2% negative predictive value for identifying men at very low risk of SVI. This cutoff could have safely allowed omitting the removal of seminal vesicles in 45.2% of RP candidates, if incorrect classification of 1.7% of patients was judged acceptable.


The potential benefits of SV preservation can be safely accomplished in virtually 50% of contemporary men subjected to RP, if our tool's predictions are applied.

Comment in

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk