Detection of prostate cancer by TURP or open surgery in patients with previously negative transrectal prostate biopsies

Objectives. To evaluate retrospectively the effectiveness of transurethral resection of the prostate (TURP) in diagnosing prostate cancer in patients with obstructive voiding symptoms and a history of negative transrectal prostate biopsy but elevated prostate-specific antigen (PSA) and/or abnormal digital rectal examination (DRE). Methods. In 1189 consecutive patients undergoing TURP or open prostatectomy between 1994 and 2000 for obstructive voiding symptoms, we identified 445 patients (37.4%) with at least one previous set of transrectal prostate biopsies because of an elevated PSA level and/or abnormal DRE findings. The probability to detect prostate cancer by TURP (n = 423; 95%) or open surgery (n = 22; 5%) was investigated overall, as well as related to patient age, PSA level, DRE findings, number of previous biopsies, time from biopsy to surgery, and weight of resected tissue. Results. The mean number of preoperative negative biopsies per patient was 1.6 (range I to 8). The mean patient age was 69 years (range 48 to 89). The median PSA level and resection weight was 8.64 ng/mL and 32 g, respectively. Ninety-seven patients (21.8%) had abnormal DRE findings. Overall, prostate cancer was detected in 35 patients (7.9%). The cancer incidence was 5.5% (19 of 348) in patients with a normal DRE compared with 16.5% (16 of 97) in patients with an abnormal DRE (P < 0.001; Fisher's exact test). The cancer rate was also related to age; other subgroups showed no statistically significant differences regarding cancer incidence. Conclusions. In patients with previously negative biopsies, the diagnostic yield of TURP is low. Therefore, TURP for diagnostic purposes only cannot be recommended. However, in patients with an abnormal DRE and obstructive symptoms, surgery should be preferred over alternative treatment options. (C) 2003 Elsevier Inc.