The purpose of the present work was two-fold: 1) to evaluate the predictive value of the sentinel lymph node (sLN) versus the axillary-node status in patients with T1-T2 breast cancer, and 2) to form an experimental basis for a randomized trial in which one group of patients with non-metastatic sLN will not have axillary dissection. Of a group of 284 patients considered for this analysis, 264 had a T1 cancer (16 T1a, 37 T1b and 211 T1c), while 20 had a T2 cancer; 243 patients were in clinical stage N0 and 41 were N1. All patients underwent lymphoscintigraphy 18 hr before surgery: 10 MBq in 0.15 mL of 99mTc-human albumin nanocolloids (particle size between 50-80 nm) was injected subdermally at the cutaneous projection of the tumor. Static gamma-camera images were acquired every 10-15 minutes until scintigraphic identification of the sLN. At surgery, 1-2 mL of Patent-Blue Violet was injected subdermally, and the sLN was searched by gamma-probe and by the dye method. The surgically isolated sLN was processed for intraoperative Hematoxylin & Eosin (H&E) histology, then for delayed histological and immunohistochemical examinations. The sLN was successfully identified by the combined radioisotopic procedure and Patent-Blue dye technique in 278/284 cases (97.9%). The Patent-Blue dye technique alone identified fewer sLNs than the radioisotopic procedure alone (56.3% versus 97.2%). Analysis of the predictive value of the sLN as to the status of axillary lymph nodes was limited to 197 patients undergoing standard axillary dissection irrespective of the sLN status. Overall, 63/191 (33%) identified sLNs were metastatic, the sLN alone being involved in 37/63 (58.7%) patients; a positive axilla status with negative sLN was found in 10/73 patients with metastatic involvement (13.7% false-negative rate). In conclusion, subdermal lymphoscintigraphy was confirmed to be an effective technique for sLN mapping; the addition of Patent-Blue dye minimally improved intra-surgical identification of the sLN. There was a high, but not absolute, correlation between a negative sLN and a negative axilla.