To evaluate the clinical usefulness of power Doppler imaging (PDI), we compared this method to gray-scale transrectal ultrasound (TRUS) in the detection of prostate cancer. A total of 101 men with abnormally high serum prostate specific antigen (PSA) levels and/or abnormal digital rectal examination (DRE) findings were assessed using TRUS and PDI. Random systematic sextant and bilateral far lateral prostate biopsies were performed in all cases. In addition, when TRUS revealed a hypoechoic lesion or PDI revealed a hypervascular lesion (HVL), these lesions were directly biopsied. Of the 101 patients, 48 (47.5%), 42 (41.5%) and 42 (41.5%) were suspicious of having prostate cancer by DRE, TRUS and PDI, respectively. Prostate needle biopsy revealed prostate cancer in 39 patients (38.6%) and benign prostatic diseases in 62 patients (61.4%). If prostate needle biopsy was avoided when PDI was negative, then PDI eliminated the need for biopsy in 59 of the 101 patients (rate of biopsy procedures saved: 58.4%) and missed only 8 (13.6%) prostate cancers. Moreover, in 63 patients with intermediate PSA (3-10 ng/ml), the rate of biopsy procedures saved by DRE, TRUS, and PDI was 60.3%, 65.1%, and 68.3%, respectively, and the rate of cancers missed was 26.3%, 19.5%, and 14.0%, respectively. In a total of 826 specimens of TRUS-guided prostate biopsy, 126 (15.3%) specimens had adenocarcinoma. Site by site based analysis of the present series revealed 34.1% of prostate cancer sites were isoechoic and hypervascular. On a site by site basis, PDI had better sensitivity, specificity, positive predictive value and negative predictive value than TRUS. In 48 patients without abnormal DRE findings, on a site by site basis, the sensitivities of TRUS and PDI were 22.9% and 34.4%, respectively. Gleason score was associated with a positive rate of PDI on both a patient basis and site by site basis. From these results, on a patient basis, we conclude that PDI was helpful in the indication for prostate biopsy for all patients or patients with intermediate PSA level. On a site by site basis, PDI may be able to select prostate cancer sites at biopsy, in particular in patients without abnormal DRE findings.