You have accessJournal of UrologyImaging/Radiology: Uroradiology I1 Apr 20102016 PREDICTING ROBOTIC PROSTATECTOMY DIFFICULTY FROM PREOPERATIVE MAGNETIC RESONANCE IMAGING Barry Mason, David Faleck, Abraham Hakimi, Marc Feder, Victoria Chernyak, Alla Rozenblit, and Reza Ghavamian Barry MasonBarry Mason More articles by this author , David FaleckDavid Faleck More articles by this author , Abraham HakimiAbraham Hakimi More articles by this author , Marc FederMarc Feder More articles by this author , Victoria ChernyakVictoria Chernyak More articles by this author , Alla RozenblitAlla Rozenblit More articles by this author , and Reza GhavamianReza Ghavamian More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2010.02.2049AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Pelvic magnetic resonance imaging (MRI) is often part of the preoperative evaluation of patients undergoing robot assisted laparoscopic prostatectomy (RALP). We sought to determine whether there are certain anatomic features that can reliably predict surgical difficulty. METHODS All patients who had a preoperative MRI in our RALP database from April 2008 thru May 2009 were reviewed. We recorded patient age, preoperative PSA, Gleason score, and body mass index (BMI). Anatomic factors included the MRI calculated prostate volume (PV) and bony pelvic measurements. A new parameter to estimate the pelvic volume is the pelvic cavity index (PCI), as a surrogate of the pelvic working space for the robotic arms. The PCI represents the pelvic inlet multiplied by the interspinous distance and divided by the pelvic depth. These parameters were analyzed for statistically significant correlations and used in multivariate regression analysis with estimated blood loss (EBL), operative time, surgical margin status, and transfusion rate as markers of surgical difficulty. RESULTS Of 100 patients that had a RALP in the study period, 76 had an MRI. Patients averaged 61 years old, a preoperative PSA of 8.6, a Gleason score of 6.4, and a BMI of 27. The average calculated prostate volume was 45 grams. The average operative time and EBL were 173 minutes and 144 ml, respectively. The prostatic transverse diameter (p=0.0218), PV (p=0.0000882), and ratio of PV to PCI (p=0.000156) were significantly correlated with operative time. The BMI (p=0.00316), prostate transverse diameter (p=0.0191), PV (p=0.0002), interspinous distance (p=0.0321), and PV to PCI ratio (p=0.0000509) were all significantly correlated with EBL. Using multiple linear regression analysis, the PV and the ratio of the PV to PCI were significantly predictive of more lengthy operative times. Similarly, the ratio of PV to PCI was predictive of increased EBL. However, multiple logistic regression analysis revealed that neither prostate volume nor PCI were predictive of positive surgical margin or transfusion rate, but BMI was significantly correlated to transfusion rate (p=.00025). CONCLUSIONS Patients with larger prostates and those with short, narrow pelvises can be predicted to have a more difficult robotic prostatectomy. Creating a ratio of the PV to PCI statistically predicts a lengthier and bloodier procedure. However, those factors do not predict positive surgical margins or risk of transfusion. From these analyses, the pelvic MRI continues to demonstrate its utility in the evaluation of a patient with prostate cancer. Bronx, NY© 2010 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 183Issue 4SApril 2010Page: e782 Advertisement Copyright & Permissions© 2010 by American Urological Association Education and Research, Inc.MetricsAuthor Information Barry Mason More articles by this author David Faleck More articles by this author Abraham Hakimi More articles by this author Marc Feder More articles by this author Victoria Chernyak More articles by this author Alla Rozenblit More articles by this author Reza Ghavamian More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...
Read full abstract