Ex vivo sentinel lymph node mapping in colorectal cancer.

Introduction: Sentinel lymph node mapping has been used in clinical work in malignant melanoma and breast cancer and shown an advantage over routine regional lymphadenectomy. The technique has been applied to colorectal cancer, but concerns over accuracy and high false-negative rates have restricted its use in the routine clinical setting. Most published series have used the in vivo technique and only three studies have been published in which the ex vivo technique was used. The aim of this study was to report the results of a larger study of ex vivo sentinel node mapping.

Methods: All patients with colorectal cancer were considered for the trial, except patients who received preoperative radiotherapy for rectal cancer. All specimens were examined in the operating room within 30 minutes of resection. After opening the bowel, 0.5 ml of patent blue dye was injected submucosally at four sites immediately adjacent to the tumor (2 ml). The pathologic examination of the sentinel nodes and of an equal number of nonsentinel nodes consisted of standard hematoxylin and eosin sectioning, followed by multiple sectioning for further hematoxylin and eosin staining and immunohistochemistry if initial samples did not show tumor metastases.

Results: A total of 58 tumors in 57 patients were studied. One or more sentinel nodes were found in relation to 56 tumors, with one of the two failures being attributed to gross mesenteric metastases obstructing lymphatic flow. A mean of 2.93 (0-8) sentinel nodes were found per patient. There was concordance between the sentinel nodes and nonsentinel nodes in 43 patients (76.8 percent). There were nine false-negative sentinel nodes (16 percent). Two patients were upstaged by detailed pathologic examination of the sentinel nodes (micrometastases), and in a further two patients the sentinel node was the only positive node on simple hematoxylin and eosin sectioning.

Conclusions: The technique of ex vivo sentinel node mapping is feasible and accurate in defining sentinel nodes in colorectal cancer. There is, however, a significant false-negative rate making the sentinel nodes not representative of the lymph node basin. This precludes the use of this technique in routine clinical practice. There may be a role in a research setting to help define the prognostic significance of micrometastases.