Effect of Inflammation and Benign Prostatic Hyperplasia on Elevated Serum Prostate Specific Antigen Levels

Robert B. Nadler, Peter A. Humphrey, Deborah S. Smith, William J. Catalona, Timothy L. Ratliff*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

368 Scopus citations

Abstract

Purpose: We quantify the causes of elevated serum prostate specific antigen (PSA) concentrations in men whose prostate biopsies repeatedly showed no cancer. Materials and Methods: The effects of prostate volume, inflammation, echogenicity on ultrasound and calculi were examined in a large PSA-based screening population of 148 men with serum PSA concentrations greater than 4.0 ng./ml., findings suspicious for cancer on digital rectal examination and multiple negative biopsies. These men were selected and compared to 64 men with suspicious rectal examinations, multiple negative biopsies and serum PSA concentrations of 4.0 ng./ml. or less. Results: The high PSA group had larger prostates (68 versus 33 cc, p = 0.0001) and significantly more subclinical prostatic inflammation. Acute and chronic inflammation was more prevalent in the high PSA group (63 percent versus 27 percent, p = 0.0001 and 99 percent versus 77 percent, p = 0.0001, respectively). A simultaneous regression analysis showed that prostatic size accounted for 23 percent, inflammation 7 percent, prostatic calculi 3 percent and nonisoechoic ultrasound lesions 1 percent of the serum PSA variance. Conclusions: Prostate volume and inflammation are the most important factors contributing to serum PSA elevation in men without clinically detectable prostate cancer.

Original languageEnglish (US)
Pages (from-to)407-413
Number of pages7
JournalThe Journal of urology
Volume154
Issue number2
DOIs
StatePublished - Aug 1995

ASJC Scopus subject areas

  • Urology

Fingerprint

Dive into the research topics of 'Effect of Inflammation and Benign Prostatic Hyperplasia on Elevated Serum Prostate Specific Antigen Levels'. Together they form a unique fingerprint.

Cite this