In colorectal cancer the sentinel node dissection may help to identify any unusual mesenteric lymphatic drainage pattern from the primary tumor site (ex/skip metastases); assuming that accurate pathological staging is critical for therapeutic decisions we are conducing a study to evaluate the feasibility of the sentinel node technique in colorectal neoplasms and its overall accuracy in predicting regional lymph nodes metastases for appropriate staging. From February 2001 to September 2004 we included in this study 30 patients with rectal lesions or degenerate colonic polyps not radically excised by endoscopy. Lymphatic mapping was performed with low molecular weight albumin colloid labelled with 500Mci of 99mTc in a 2 ml volume and injected submucosally by an endoscopic route at the four cardinal points around the tumor, the afternoon before the surgical procedure, both in case of colonic or rectal lesions. Scintigraphic images were obtained with a gamma camera fitted with a general purpose collimator. The day of the intervention, a hand held gamma detecting probe (Scintiprobe m100, Pol-Hi-Tech, Italy) was employed to detect the "hot" nodes, in vivo and ex vivo. These lymph nodes were tagged with a stitch in vivo; the specimen was removed by a standard resection and SLN were dissected ex vivo and sent separately for pathological examination. In case of rectal lesions, the sentinel nodes were searched ex vivo into mesorectal fat in case. All lymph nodes, including blue or hot ones, were embedded separately for preparation of paraffin sections and haematoxylin and eosin staining. Sentinel lymph node were submitted to multi-seriate sections in order to look for micrometastases. Using the radioactive tracer, sentinel lymph nodes were successfully identified in 27 out of 30 patients. Concordance between SLNs and nodal status was observed in 23 out of 27 cases (85%); two patients (7.4%) were upstaged, as SLN was the only site of metastases. In another two cases we observed no concordance between negative sentinel node and non sentinel nodes (false negative rate, 7.4%). Starting from this experience we are proposing a multicentric trial concerning the value of sentinel node technique in rectal cancer and in early colorectal cancers detected by screening programs.