A prospective study was carried out to evaluate the possible impact of prostatic manipulation upon the serum prostate specific antigen (PSA) in a cohort of 98 patients prior to treatment of adenocarcinoma of the prostate. The PSA levels were assessed before and 1 hour after examination under anaesthetic (EUA). There were 87 complete pairs of observations. In 71 patients there was either no change (n = 2) or a rise (n = 69); in 16 there was a fall. The median change in the group as a whole was +2.0 ng/ml (range -10.8-36.2). The average percentage rise in PSA was 42.3% (median 17.4%) with the largest percentage changes seen in those with the lowest initial PSA levels, with those in the range < 4 ng/ml exhibiting a 66.7% rise and those with > 35 ng/ml a 3.2% rise (median values). An attempt is made to express these changes in a clinical sense by allocating patients (pre-EUA) to one of four ranges of values as follows, 0-4.0, 4.1-10.0, 10.1-35.0 and > 35 ng/ml and observing the frequency with which they climbed to a higher category after EUA. This occurred in 23%, 50%, and 15% of patients respectively. The changes in PSA were statistically significant but their clinical significance might depend upon the clinical setting. Thus no patient would have had their management changed as a result of such a rise in PSA, but this might assume greater importance during the months of follow-up after radiation therapy, where a rise in PSA after prostatic manipulation might result in undue patient anxiety and unnecessary investigations.(ABSTRACT TRUNCATED AT 250 WORDS)