Assessment of sentinel Lymph node mapping with different volumes of Indocyanine green in early-stage ENdometrial cancer: the ALIEN study.

ABSTRACT

OBJECTIVE: To evaluate the impact of different volumes of indocyanine green (ICG) on the detection rate and bilateral mapping of sentinel lymph nodes in patients with apparent uterine-confined endometrial cancer. METHODS: All patients who underwent surgical staging with sentinel node mapping in six reference centers were included. Two different protocols of ICG intracervical injection were used (1) 2 mL group total volume of 2 mL injected superficially; (2) 4 mL group total volume of 4 mL, 2 mL deeply and 2 mL superficially. Logistic regression was used to analyze factors that could influence dye migration and detection rates. A sensitivity analysis was carried out to determine how independent variables could affect the sentinel node detection rate. RESULTS: Of 442 eligible patients, 352 were analyzed (172 in the 2 mL group and 180 in the 4 mL group). The bilateral detection rates of the 2 mL and 4 mL groups were 84.9% and 86.1%, respectively (p=0.76). The overall detection rate was higher with a volume of 4 mL than with 2 mL (97.8% vs 92.4%, respectively; p=0.024). In the univariate analysis the rate of bilateral mapping fell from 87.5% to 73.5% when the International Federation of Gynecology and Obstetrics (FIGO) 2009 tumor stage was >IB (p=0.018). In the multivariate analysis, for both overall and bilateral detection rates a statistically significant difference emerged for the volume of ICG injected and FIGO 2009 stage >IB. Increasing body mass index was associated with worse overall detection rates on univariate analysis (p=0.0006), and significantly decreased from 97% to 91% when the body mass index exceeded 30 kg/m2 (p=0.05). CONCLUSIONS: In patients with early-stage endometrial cancer, a volume of 2 mL ICG does not seem to compromise the bilateral detection of sentinel lymph nodes. In women with obesity and FIGO 2009 stage >IB, a 4 mL injection should be preferred.