The importance of combining xylene clearance and immunohistochemistry in the accurate staging of colorectal carcinoma.

The prognosis of colorectal carcinoma relies heavily on pathological staging which includes the metastatic state of lymph nodes. Colorectal resectates from 47 patients (41 with colorectal carcinoma and six with non-malignant disease) were entered into a study to assess the best method for detecting metastases in lymph nodes. The maximum number of lymph nodes was harvested at an initial careful examination of the specimen. Subsequently, the pericolic and perirectal fat was dissected out, dehydrated in alcohol, cleared in xylene and further lymph nodes were recovered. Both sets of lymph nodes were examined by the standard histological method and subsequently stained immunohistochemically for cytokeratins (CK). The mean number of lymph nodes recovered at the initial dissection from all 47 cases was 6.7, this was raised to 58.2 after xylene clearance, ie an average of 51.5 lymph nodes were not recovered by traditional methods. At the initial dissection no epithelial cells were detected in any of the lymph nodes from the nonmalignant cases or 25 of the malignant cases. In the other 16 cases, epithelial cells were detected by H&E in 38 lymph nodes. Thus the initial staging was 3 Dukes A, 22 Dukes B and 16 Dukes C. After immunohistochemistry, eight additional cases (originally staged Dukes B) showed epithelial cells in the lymph nodes, these were chiefly occult invasion, raising the involved lymph nodes number to 70. After xylene clearance and applying the CK staining, an additional 135 lymph nodes were found to be involved, thus the overall number of involved lymph nodes was increased to 205. The combined technique changed the Dukes staging in 12 out of 41 cases of colorectal carcinoma, resulting finally in 3 Dukes A, 10 Dukes B and 28 Dukes C, ie 55% of Dukes B become Dukes C.