Detection of prostate cancer in males with prostatism
Version of Record online: 20 JUL 2006
Copyright © 1994 Wiley-Liss, Inc., A Wiley Company
Volume 25, Issue 3, pages 132–140, September 1994
How to Cite
Lepor, H., Owens, R. S., Rogenes, V. and Kuhn, E. (1994), Detection of prostate cancer in males with prostatism. Prostate, 25: 132–140. doi: 10.1002/pros.2990250304
- Issue online: 20 JUL 2006
- Version of Record online: 20 JUL 2006
- Manuscript Accepted: 12 JAN 1994
- Manuscript Received: 29 JUN 1993
- benign prostatic hyperplasia;
- prostate specific antigen;
- prostate specific antigen density;
- detection of prostate cancer
The present study was designed to compare the prostate cancer detection rate, sensitivity, specificity, and positive predictive value of digital rectal examination (DRE) and serum prostatic specific antigen (PSA) in a consecutive cohort of males presenting to a single institution with clinically significant prostatism. The study population was comprised of 224 consecutive males with clinically significant prostatism referred to the Prostate Center at the Medical College of Wisconsin between June 1990 and December 1991. Subjects were considered to have clinically significant prostatism if they elected to pursue medical or surgical therapy following exclusion of carcinoma of the prostate.
The initial examination consisted of a Boyarsky symptom score assessment, DRE, uroflowmetry, postvoid residual determination, serum PSA level, and transrectal prostatic ultrasonography. Subjects with an abnormality on DRE or serum PSA > 4 ng/dl were advised to undergo transrectal prostatic biopsy. Of the 224 subjects, 40 (17.9%) had an abnormal DRE and 57 (25.4%) had an elevated serum PSA > 4 ng/dl. The overall detection rate of prostate cancer in the study population was 6.7%. The prostate cancer detection rates for PSA alone and DRE alone were 5.8% and 5.3%, respectively. The sensitivity, specificity, and positive predictive values of PSA alone were 86.7%, 80.9%, and 25.0% and of DRE alone 80.0%, 86.3%, and 30.0%, respectively.
Receiver operator characteristic (ROC) curves were constructed for the entire study population in order to compare the screening measures serum PSA and PSA density. The area under the curves was 0.88 for both tests, indicating that these screening tests for prostate cancer were not significantly different. The present study demonstrated that males with clinically significant prostatism represent a high risk cohort for detecting prostate cancer. DRE and PSA are equally effective measures for detecting prostate cancer. PSA density does not offer any advantage over serum PSA in screening for prostate cancer, except in the subset of patients with a normal DRE and serum PSA levels between 4.0 and 9.9 ng/dl. © 1994 Wiley-Liss, Inc.