Axillary lymph node dissection is an important part of the surgical treatment of breast cancer as a staging procedure. Recent progressive advances in early detection have led to the treatment of small primary carcinomas; thus, a great number of axillary dissections show completely negative lymph nodes. The sentinel node (SN) concept, developed for melanoma patients, seems to be similarly valid in breast cancer and has the potential to change the standard surgical approach in these patients. To verify the accuracy of lymphoscintigraphic method associated with radioguided biopsy of the sentinel node in a large series of patients, we studied 382 patients with operable breast cancer. Lymphoscintigraphy (LS) was performed the day before surgery; three different-sized ranges of 99mTechnetium-labeled colloid particles were injected either by subdermal or peritumoral administration. Planar scans were registered in anterior and oblique projections, and a cutaneous marker was placed over the skin corresponding to the SN as visualized. SNs were localized and removed during surgery, using a gamma-detecting probe (GDP); total axillary dissection was then performed. In 54 patients, blue dye was also administrated in the tumor bed immediately after excision of the primary. LS identified at least one SN in 377 of 382 cases (98.7%). Axillary SN was localized in 371 cases (97.1%). The overall concordance between SN status and other axillary nodes was 96.8% (359 of 371). Localization of the SN was easier when large-size particles of colloidal albumin were injected in a volume of 0.4 ml. GDP successfully localized SN in 54/54 cases (100%), while blue dye identified SN in 37/54 patients (68.5%). In 33 of 37 cases (89%) the dye and LS identified the same node. LS and GDP-guided surgery provide accurate identification and removal of the SN, particularly when large-size radiolabeled colloids are injected in a small volume.