Fluorescent Lymphography–Guided Lymphadenectomy During Robotic Radical Gastrectomy for Gastric Cancer

摘要

Importance: Fluorescent imaging with indocyanine green can be used to visualize lymphatics. Peritumoral injection of indocyanine green may allow for visualization of every draining lymph node from a primary lesion on near-infrared imaging. Objectives: To evaluate the role of fluorescent lymphography using near-infrared imaging as an intraoperative tool for achieving complete lymph node dissection and compare the number of lymph nodes retrieved with the use of near-infrared imaging and the number of lymph nodes retrieved without the use of near-infrared imaging. Design, Setting, and Participants: This prospective single-arm study was conducted among 40 patients who underwent robotic gastrectomy between August 30, 2013, and July 21, 2014, at a single-center, tertiary referral teaching hospital. After propensity score matching, the results of these 40 patients were compared with the results of 40 historical control patients who underwent robotic gastrectomy without indocyanine green injection between January 1, 2012, and August 31, 2013. Statistical analysis was performed from January 1, 2015, to July 31, 2016. Interventions: Robotic gastrectomy with systemic lymphadenectomy and retrieval of lymph nodes under near-infrared imaging after peritumoral injection of indocyanine green to the submucosal layer 1 day before surgery. Main Outcomes and Measures: The primary outcome was the number of retrieved lymph nodes in each nodal station. Results: Among the 40 patients in the study (19 women and 21 men; mean [SD] age, 52.2 [11.7] years), no complications related to indocyanine green injection or near-infrared imaging were observed. On completion of the lymphadenectomy, the absence of fluorescent lymph nodes in the dissected area was confirmed. A mean (SD) total of 23.9 (9.0) fluorescent lymph nodes were recorded among a mean (SD) total of 48.9 (14.6) overall lymph nodes retrieved. The mean number of overall lymph nodes retrieved was larger in the near-infrared group than in the historical controls (48.9 vs 35.2; P < .001), with a significantly greater number of lymph nodes retrieved at stations 2, 6, 7, 8, and 9. In the near-infrared group, 5 patients exhibited lymph node metastases, and all metastatic lymph nodes were fluorescent. Conclusions and Relevance: This study's findings suggest that fluorescent lymphography may be useful intraoperatively for identifying and retrieving all necessary lymph nodes for a complete and thorough lymphadenectomy.