Objective: To eliminate the need for per-operative frozen section analysis by using pre-operative serum prostate-specific antigen (PSA), clinical stage and Anderson biopsy grade to predict lymph node metastasis.
Patients and methods: Between 1988 and 1994, 214 patients underwent bilateral pelvic lymphadenectomy before an intended retropubic radical prostatectomy for clinically confined prostate carcinoma. Preoperative serum PSA, clinical stage and Anderson biopsy grade were analysed to investigate whether they were able to predict the results of per-operative frozen sections of pelvic lymph nodes.
Results: Serum PSA level was the best predictor of a frozen section which was positive for lymph node metastasis, followed by biopsy grade. Clinical T-category combined with PSA level predicted lymph node status less well than did biopsy grade and had no additional value when combined with biopsy grade and PSA. Pre-operative serum PSA combined with biopsy grade predicted a negative frozen section result of bilateral pelvic lymphadenectomy with 95% certainty in at least 17% of patients undergoing radical prostatectomy for clinically confined prostate cancer. Frozen section analysis had a false negative rate of 4.7% when compared with histological staging. Therefore, combining those patients with an estimated 95% chance of a negative frozen section with the group of false negatives, 11 in 214 patients (5.1%) would have undergone radical prostatectomy while having micrometastases.
Conclusions: Selecting men for whom the per-operative frozen section analysis could be omitted may save operating time and allow more efficient use of facilities.