Intraoperative frozen section assessment of pelvic lymph nodes during radical prostatectomy is of limited value

Abstract

To evaluate the accuracy of frozen section (FS) assessment of pelvic lymph nodes (PLNs) during radical prostatectomy (RP) in a large contemporary cohort; and to analyse the contribution of FS to surgical decision making in this setting. During a 4-year period at a single institution, RPs with PLN dissection (PLND) were reviewed. The number and size of the PLNs, and the size of metastases were measured. FS was performed on 349 bilateral PLNDs. Overall, 28 (8%) cases were positive for metastasis, 11 of which were detected by FS (39%). The 17 false negatives, all of which contained metastases smaller than 5 mm, were due to failure to identify and freeze the positive PLNs (11), failure to section at the level of the metastatic tumour (four), or interpretative error (two). The sensitivity was not affected by the number of sampled nodes. The size of metastasis was the determining factor for the accuracy of FS, with metastases of ≥ 5 mm having a sensitivity of 100%, and metastases of < 5 mm having a sensitivity of 10%. Among the 11 true positives, RP was aborted in eight cases and continued in three. During the same period, 261 PLNDs were performed without FS, and 18 (6.9%) had metastases. FS is highly accurate in detecting large, grossly evident metastases, but performs poorly on micrometastases. It is recommended that a two-step approach applied to routine FS starting with a careful gross examination followed by FS for only grossly suspicious PLNs.