Sentinel lymph node mapping for adenocarcinoma of the colon does not improve staging accuracy.

PURPOSE: This study was designed to: determine the efficacy of sentinel lymph node mapping in patients with intraperitoneal colon cancer; and create an algorithm to predict potential survival benefit by using best-case estimates in favor of sentinel node mapping and lymph node ultraprocessing techniques.
METHODS: Forty-one patients with intraperitoneal colon cancer undergoing colectomy with curative intent were studied prospectively. After mobilization of the colon and mesentery, 1 to 2 ml of isosulfan blue dye was injected subserosally around the tumor. The first several nodes highlighted with blue dye were identified as sentinel nodes. Additional nodes were identified by the pathologist in routine fashion by manual dissection of the mesentery. All nodes were processed in routine fashion by bivalving and hematoxylin and eosin staining. To create an algorithm to predict potential survival benefit of sentinel node mapping and lymph node ultraprocessing techniques, assumptions were made using data from the literature. All bias was directed toward success of the techniques.
RESULTS: Three of 41 patients (7 percent) did not undergo injection of dye and were excluded from further analysis. Stage of disease in the remaining 38 patients was: I, n = 10 (26 percent); II, n = 15 (39 percent); III, n = 11 (29 percent); IV, n = 2 (5 percent). At least one sentinel node was identified in 30 of 38 patients (79 percent). The median number of sentinel nodes identified was two (range, 1-3). Median total nodal retrieval was 14 (range, 7-45). All nodes were negative in 26 of 38 patients (68 percent). Sentinel nodes and nonsentinel nodes were positive in 2 of 38 patients (5 percent). Sentinel nodes were the only positive nodes in 1 of 38 patients (3 percent). Sentinel nodes were negative and nonsentinel nodes were positive in 9 of 38 patients (24 percent). Thus, sentinel node mapping would have potentially benefited only 3 percent, and failed to accurately identify nodal metastases in 24 percent of the patients in our study. To create a survival benefit algorithm, we assumed the following: combined fraction of Stage I and II disease (0.5); fraction understaged by bivalving and hematoxylin and eosin staining that would have occult positive nodes by more sophisticated analysis (0.15); fraction of occult positive nodes detected by sentinel node mapping (0.9); and survival benefit from chemotherapy (0.33). Thus, the fraction of patients benefiting from sentinel lymph node mapping and lymph node ultraprocessing techniques would be 0.02 (2 percent).
CONCLUSIONS: Sentinel node mapping with isosulfan blue dye and routine processing of retrieved nodes does not improve staging accuracy in patients with intraperitoneal colon cancer. Even using best-case assumptions, the percentage of patients who would potentially benefit from sentinel lymph node mapping is small.