Limits of day-before lymphoscintigraphy to localize sentinel nodes in women with cervical cancer

Background: Lymph node status in cervical cancer is a major prognostic factor. Sentinel lymph node (SN) biopsy using radiocolloid and blue dye labeling and preoperative lymphoscintigraphy has emerged as a potential alternative to systematic lymphadenectomy. The aim of this study was to evaluate the contribution of preoperative lymphoscintigraphy to SN biopsy. Methods: Between April 2001 and December 2005, 71 of 77 patients with cervical cancer (38 patients with stages IA or IB1, and 39 patients with stage IB2, IIA or IIB) underwent laparoscopic SN procedure using radiocolloid and blue dye with day-before lymphoscintigraphy. The SN identification rates and false-negative rates were studied. Results: Seventy patients underwent a combined technique and the last patient a radiocolloid technique alone due to blue dye allergic reaction. Detection rate of lymphoscintigraphy was 84.5% (60/71), with 1.4 sentinel nodes per patient. Three of 11 patients (27.3%) with no SN on lymphoscintigraphy had at least one SN during surgery. Sixteen of 27 patients (59.3%) with solitary SN on lymphoscintigraphy had multiple SNs. Nine of 35 patients (25.7%) with unilateral SNs on lymphoscintigraphy had bilateral SNs at surgery (kappa = 0.44 [0.19-0.64]). When categorized into < 2 and >= 2 sentinel nodes, the correlation between lymphoscintigraphic and surgical detection was poor (kappa = 0.05 [0.0-0.18]). Conclusion: SN biopsy is a feasible and accurate method to stage early cervical cancer. However, day-before lymphoscintigraphy is poorly correlated to surgical SN mapping.