Pelvic Lymph Node Metastases in Prostate Cancer: Preoperative Detection With Dynamic Contrast-Enhanced Magnetic Resonance Imaging Compared With Postoperative Pathologic Result of Pelvic Lymph Node Dissection

Hoyoung Ryu, Byeongdo Song, Jinho Hwang, Sung Kyu Hong, Seok-Soo Byun, Sang Eun Lee, Sangchul Lee
2017 The Korean Journal of Urological Oncology  
Purpose: The aim of this study is evaluating the accuracy of preoperative magnetic resonance imaging (MRI) in patients who underwent pelvic lymph node dissection (PLND). Materials and Methods: The medical records of 1,528 patients who underwent radical prostatectomy and PLND from 2003 to 2017 in Seoul National University Bundang Hospital were retrospectively reviewed. We evaluated the various clinicopathologic variables including preoperative MRI findings and pathologic lymph node (LN)
more » ... s. The prediction model for pathologic LN metastasis was assessed using univariate and multivariable logistic regression analyses and areas under receiver operating characteristic (ROC) curves. Results: The mean age of our cohort was 66.4±6.7 years. Positive LN finding of preoperative MRI finding was observed in 9.4% (145 of 1,528) of patients. 5.3% (81 of 1,528) of patients had confirmed final pathologic LN metastases. Sensitivity and specificity of preoperative MRI were 30.8% and 91.7%, respectively. Multivariable analysis showed that preoperative MRI findings, clinical stage and biopsy Gleason score were independent significant predictors for pathologic LN metastasis (p<0.001, p=0.002, and p<0.001, respectively). Prediction model using preoperative MRI findings and National Comprehensive Control Network risk stratification showed fair accuracy using ROC analysis. Conclusions: Preoperative MRI findings for pathologic LN metastasis showed limited prediction value. A large-scale, multicenter, prospective study is needed to fully evaluate the clinical significance of preoperative MRI. (Korean J Urol Oncol 2017;15:158-164)
doi:10.22465/kjuo.2017.15.3.158 fatcat:x5k6j6c3xfgjzfldj7friqzfm4