Contribution of lymphoscintigraphy to intraoperative sentinel lymph node detection in early cervical cancer: Analysis of the prospective multicenter SENTICOL cohort.

PURPOSE:To evaluate the contribution of preoperative lymphoscintigraphy to intraoperative lymphatic mapping (ILM) in early cervical cancer METHODS:We conducted an ancillary analysis of the multicenter prospective SENTICOL study in early cervical cancer. Radiocolloid was injected intracervically on the day before (long protocol) or morning of (short protocol) surgery, lymphoscintigraphy was performed, and the results of a centralized image review were communicated to the surgeons. ILM was performed on combined radioactivity/patent blue detection. Sentinel lymph nodes (SLNs) were electively sampled before routine bilateral pelvic lymphadenectomy by laparoscopy. RESULTS:Of 139 patients in the modified intention-to-diagnose analysis, 114 had centrally reviewed lymphoscintigrams, which showed 352 SLNs in 100 patients. Lymphoscintigraphy and ILM detection rates were 87.8% and 97.8%, respectively. Agreement between lymphoscintigraphy and ILM was low for the number of SLNs (κ=0.23; -0.04; 0.49) and bilateral SLNs (κ=0.36; 0.2; 0.52). No patient without SLNs by ILM had SLNs by lymphoscintigraphy. Lymphoscintigraphy identified substantial proportions of unusual drainage pathways. No patients with metastatic nodes had SLNs by lymphoscintigraphy but not by ILM in the relevant territory. In 1 of the 2 patients with false-negative SLN results, SLNs were bilateral by lymphoscintigraphy and unilateral by ILM. CONCLUSION:Although the detection rate was lower by lymphoscintigraphy than by ILM, the substantial proportions of SLNs in unusual territories provided valuable guidance for the surgical exploration. Awareness of the limited agreement between lymphoscintigraphic and surgical detection might help surgeons decrease the false-negative rate.