Sentinel lymph node detection in early stage cervical cancer: a prospective study comparing preoperative lymphoscintigraphy, intraoperative gamma probe, and blue dye.

OBJECTIVE The objective of this prospective study was to determine the feasibility of sentinel lymph node (SLN) detection in patients with cervical cancer using lymphoscintigraphy (LS), gamma probe, and blue dye. METHODS A total of 32 patients with early stage cervical cancer (FIGO IA2-IIA) who were treated with total abdominal hysterectomy and bilateral pelvic and paraortic lymphadenectomy underwent SLN biopsy. LS was performed on all the patients following the injection of 74 MBq technetium-99m-nanocolloid pericervically. The first appearing persistent focal accumulation on either dynamic or static images of LS was considered to be an SLN. Blue dye was injected just prior to surgical incision in 16 patients (50%) at the same locations as the radioactive isotope injection. During the operation, blue-stained node(s) were excised as SLNs. For gamma probe, a lymph node was accepted as an SLN, if its ex vivo radioactive counts were at least 10-fold above background radioactivity. SLNs, which were negative by routine hematoxylin and eosin (H&E) examination, were histopathologically reevaluated for the presence of micrometastases by step sectioning and immunohistochemical staining with pancytokeratin. RESULTS At least one SLN was identified for each patient by gamma probe. Intraoperative gamma probe was the most sensitive method with a technical success rate of SLN detection of 100% (32/32), followed by LS 87.5% (28/32) and blue dye 68.8% (11/16), respectively. The average number of SLNs per patient detected by gamma probe was 2.09 (range 1-5). The localizations of the SLNs were external iliac 47.8%, obturatory 32.8%, common iliac 9%, paraaortic 4.4%, and paracervical 6%. Micrometastases, not detected by routine H&E were found by immunohistochemistry in one patient. On the basis of the histopathological analysis, the negative predictive value for predicting metastases was 100%, and there were no false-negative results. CONCLUSIONS Preoperative LS with radiocolloids, intraoperative lymphatic mapping with blue dye and gamma probe are all feasible methods comparable with each other for SLN detection in early stage cervical cancer patients, but gamma probe is the most useful method in terms of technical success.