You have accessJournal of UrologyProstate Cancer: Staging II1 Apr 2014MP42-07 COMPUTERIZED TRANSRECTAL ULTRASOUND TARGETED BIOPSIES: PREOPERATIVE PREDICTION OF THE RADICAL PROSTATECTOMY GLEASON GRADE Tillmann Loch, Annemie Loch, Björn Grabski, Leif Baeurle, Udo Paul, and Theodore Tokas Tillmann LochTillmann Loch More articles by this author , Annemie LochAnnemie Loch More articles by this author , Björn GrabskiBjörn Grabski More articles by this author , Leif BaeurleLeif Baeurle More articles by this author , Udo PaulUdo Paul More articles by this author , and Theodore TokasTheodore Tokas More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2014.02.1184AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail Introduction and Objectives The accuracy of the random prostate biopsy to predict the prostatectomy GS varies tremendously in the literature. In a meta-analysis the accuracy ranges from 53% to 67% with a mean value of 63%. Tumor upgrading in radical prostatectomy specimens approximates 30% (range: 6-36%). In low grade tumors, random biopsy accuracy reaches 62% with a tumor upgrading rate of 38%. The aim of this study is to assess the accuracy preoperative GS prediction of computerized transrectal ultrasound (C-TRUS/ANNA), with six targeted biopsy cores. Methods We reviewed 164 patients from August 2004 until April 2013 with positive C-TRUS/ANNA targeted prostate biopsies and subsequently underwent radical prostatectomy, at our institution. Of them, 82 had also a history of negative prior systematic biopsies (1-10 sessions, 6-70 cores). The biopsy and radical prostatectomy differentiations were assessed and compared. Results The patients had a median age of 69 years (range: 51-77) and a median PSA value of 7.86 ng/mL (range: 1.12-91). Gleason scores were correctly predicted in 120 patients (60%). Tumor upgrading appeared in 76 patients (38%), and upgrading, more than one grade, only in 11 patients (6.67%). Downgrading occured in only 8 patients (4.85%). According to primary Gleason grade, 135 (82.32%) patients were correctly predicted, 23 (14.02%) patients had a higher and 6 (3.66%) patients had a lower Gl grade in the RRP specimen. These results (see table) can be expected, since the C-TRUS/ANNA system recognizes variations in pixel architecture of a TRUS digital image, and, in that way, mainly detects the predominant (primary) Gl grade, especially if the secondary grade is of low percentage. Conclusions By taking only six targeted biopsies, C-TRUS/ANNA could predict tumor differentiation better than by taking 12 random systematic biopsies. Most importantly, it can predict differentiation of tumors that could not be detected, at all, by random systematic biopsies, with the same accuracy. RRP Gl Primary Grade 3 (%) RRP Gl Primary Grade 4 (%) RRP Gl Primary Grade 5 (%) Biopsy Gl Primary Grade 3 (n=142) 121 (85.21) 16 (11.27) 5 (3.52) Biopsy Gl Primary Grade 4 (n=14) 4 (28.57) 8 (57.14) 2 (14.29) Biopsy Gl Primary Grade 5 (n=8) - 2 (25) 6 (75) © 2014FiguresReferencesRelatedDetails Volume 191Issue 4SApril 2014Page: e470-e471 Advertisement Copyright & Permissions© 2014MetricsAuthor Information Tillmann Loch More articles by this author Annemie Loch More articles by this author Björn Grabski More articles by this author Leif Baeurle More articles by this author Udo Paul More articles by this author Theodore Tokas More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...