The correlation of serum prostate specific antigen(PSA) and prostatic volume measured by transrectal ultrasonography in the diagnosis of benign prostatic hyperplasia

Ha, Bong Dal; Lee, Sang Sung; Park, Choal Hee; Lee, Sung Choon; Lee, Sung Moon; Woo, Sung Goo

Abstract

We investigated a correlation of serum prostate specific antigen(PSA) and preoperative prostatic volume measured by transrectal ultrasonography (TRUS) in 66 patients with pathologically con- firmed benign prostatic hyperplasia (BPH), as aids in the early detection of prostate cancer and in the improvement of specificity in BPH. The preoperative serum PSA value was also measured in 18 patients with pathologically confirmed prostate cancer and 15 men without prostatic disease as normal control using Tandem-R PSA radioimmunometric assay. There was a highly significant correlation between natural logarithm transformed PSA(In PSA) and non-transformed prostatic volume by linear regression analysis(r=0.6668, p<0.0001). Given above data, we determined a volume adjusted 95th percentile upper confidence limit for n PSA as an appropriate cutoff between normal and abnormal PSA value, according to prostatic volume in BPH. In the patients with BPH and prostate cancer, sensitivity, specificity, diagnostic accuracy and positive predictive value were 94%, 72%, 68% and 48%, respectively. A statistical results by the volume-adjusted cutoff for in PSA was similar to that at 10 ng/ml as cutoff for PSA value. Above 40ml in prostatic volume, however, a false positive rate was decreased in inverse proportion to prostatic volume in BPH. In summary, the prostatic volume must be considered in BPH patients with elevated serum PSA value, negative digital rectal examination and negative TRUS of prostate. And when the volume adjusted PSA cutoff is clinically applied to BPH patients for the early detection of prostate cancer, we thought that the false positive rate can be decreased in BPH patients.