Background & objectives: The characteristics of prostate specific antigen (PSA) for trans-rectal ultrasonography guided prostate biopsy in men with lower urinary tract symptoms (LUTS) are not well defined. This study was carried out to analyse the threshold of PSA for biopsy in symptomatic men in India.
Methods: From January 2000 to June 2011, consecutive patients who had digital rectal examination (DRE) and PSA testing done for LUTS were included in this study. PSA was done with ELISA technique. Patients with acute or chronic prostatitis, prostatic abscess, history of surgery on prostate within the previous three months and patients on 5α-reductase inhibitors or on urethral catheter were excluded.
Results: Of the 4702 patients evaluated, 70.9 per cent had PSA of less than 4 ng/ml and 29.1 per cent had PSA of more than 4 ng/ml. Of these, 875 men with a mean age of 65.72±7.4 (range 50-75 yr) had trans rectal ultrasonography (TRUS) guided biopsy. Twenty five men had biopsy at PSA level of <4 ng/ml due to positive DRE, 263 at 4.1-10ng/ml, 156 at 10.1-20 ng/ml and 431 at >20 ng/ml. Positive predictive value of PSA in ranges of 4.1-10, 10.1-20, >20 ng/ml was 15.2, 24 and 62.6 per cent, respectively with negative DRE. PSA cut-off to do biopsy was derived by ROC curve as 5.82 ng/ml for all the men. When the subjects were further stratified on the basis of DRE findings, a cut-off of 5.4 ng/ml was derived in men with normal DRE.
Interpretation & conclusions: A cut-off for biopsy in symptomatic men with negative DRE could safely be raised to 5.4 ng/ml, which could avoid subjecting 10 per cent of men to undergo unnecessary biopsy.