BACKGROUND: To assess whether the severity of lower urinary tract symptoms (LUTS) is associated with the risk of prostate cancer (PCa) detection via a multi (≥12)-core prostate biopsy. METHODS: From January 2004 to May 2011, 3,107 patients underwent transrectal ultrasound (TRUS) prostate biopsies due to elevated PSA levels ranging between 3 and 20 ng/ml or abnormal digital rectal exams (DREs). Multivariate logistic analysis was used to assess the potential association of LUTS and PCa detection via biopsy. The predictive accuracy of the multivariate model was assessed based on the receiver operating characteristics-derived area under the curve. RESULTS: The median International Prostate Symptom Score (IPSS) was 11, and the mean PSA was 6.81 ng/ml. Of the total subjects, PCa was detected from biopsy in 931 (30.0%) patients. In a comparison of 1,465 patients with IPSS ≥ 11 and 1,642 patients with IPSS <11, those with a higher IPSS were older, had higher PSA and had a larger prostate, but there were no significant differences in the PCa detection rates. However, in multivariate analysis incorporating other associated variables, a higher IPSS was significantly associated with lower odds of PCa detection (P = 0.016). Nevertheless, addition of the IPSS did not significantly increase the accuracy of the multivariate model devised for the detection of PCa (P = 0.098). CONCLUSIONS: Although PCa was detected less commonly among men with higher LUTS, LUTS may not provide additional prognostic information beyond that which can be obtained via previously established prognostic factors.
BACKGROUND: To assess whether the severity of lower urinary tract symptoms (LUTS) is associated with the risk of prostate cancer (PCa) detection via a multi (≥12)-core prostate biopsy. METHODS: From January 2004 to May 2011, 3,107 patients underwent transrectal ultrasound (TRUS) prostate biopsies due to elevated PSA levels ranging between 3 and 20 ng/ml or abnormal digital rectal exams (DREs). Multivariate logistic analysis was used to assess the potential association of LUTS and PCa detection via biopsy. The predictive accuracy of the multivariate model was assessed based on the receiver operating characteristics-derived area under the curve. RESULTS: The median International Prostate Symptom Score (IPSS) was 11, and the mean PSA was 6.81 ng/ml. Of the total subjects, PCa was detected from biopsy in 931 (30.0%) patients. In a comparison of 1,465 patients with IPSS ≥ 11 and 1,642 patients with IPSS <11, those with a higher IPSS were older, had higher PSA and had a larger prostate, but there were no significant differences in the PCa detection rates. However, in multivariate analysis incorporating other associated variables, a higher IPSS was significantly associated with lower odds of PCa detection (P = 0.016). Nevertheless, addition of the IPSS did not significantly increase the accuracy of the multivariate model devised for the detection of PCa (P = 0.098). CONCLUSIONS: Although PCa was detected less commonly among men with higher LUTS, LUTS may not provide additional prognostic information beyond that which can be obtained via previously established prognostic factors.