OBJECTIVETo identify the predictors of the positive results of transrectal ultrasound (TRUS)-guided biopsy for prostate cancer.

METHODSWe performed univariate and multivariate logistic regression analyses on the relevant data on 385 male patients that underwent TRUS-guided biopsy for prostate cancer, including such potential predictors as age, body mass index (BMI), symptoms, results of digital rectal examination (DRE), tPSA, fPSA, free/total PSA ratio (f/tPSA), prostate volume (PV), and PSA density (PSAD) for identification of the risk factors related to the positive rate of biopsy. Then we constructed a scoring system as a tool for predicting prostate cancer in repeat biopsies and determined the sensitivity of the system by calculating the false positive rate using the receiver operating characteristic curve.

RESULTSAmong the 385 patients, 139 (36.1%) were diagnosed with prostate cancer. On multivariate analysis, age (P < 0.01), DRE (P < 0.01), tPSA (P < 0.01), fPSA (P < 0.01), f/tPSA (P < 0.01), PV (P < 0.01), and PSAD (P < 0.01) were all significant predictors of prostate cancer. Multivariate logistic regression analysis showed age, tPSA, f/tPSA, PV, and PSAD to be independent predictors, with ORs and 95% CIs of 1.07 (1.05-1.16), 1.05 (1.02-1.15), 0.97 (0.86-0.99), 0.98 (0.87-0.96), and 1.79 (1.48-2.06), respectively. Moreover, patients with the risk score of 3-5 had a significantly higher rate of prostate cancer than those with 0-2 (64% vs 11%, P < 0.001).

CONCLUSIONThe scoring system on the key predictors of prostate cancer can help urologists to identify the men in need of prostatic biopsy.