Abstract Background: The role of axillary lymph node dissection in the staging of patients with breast carcinoma is currently under evaluation. As a result of recent advances in minimally invasive techniques, an endoscopic approach to axillary lymph node dissection may be an attractive alternative to lymphadenectomy performed via a standard “open” axillary incision. The purpose of the present study was to evaluate the technical feasibility and safety of such an approach in human cadavers. Study Design: Twelve axillary dissections (right in seven, left in five) were performed in seven cadavers. A 2.5-cm incision was made along the lateral chest wall 12 cm inferior to the apex of the axilla. The subcutaneous axillary space was expanded with a balloon dissector, and exposure was maintained with a prototype external lift device. Endoscopic dissection was performed with three or four 5-mm ports inserted into this space. Histologic analysis was performed to document the number of lymph nodes removed. Results: The mean dissection time for endoscopic axillary lymphadenectomy was 56.7 ± 19 minutes (range, 30–90 minutes). Structures visualized endoscopically included the axillary vein (12 of 12 patients), the long thoracic nerve (12 of 12 patients), the thoracodorsal nerve (10 of 12 patients), and the intercostobrachial nerve (11 of 12 patients). An average of 9.9 ± 7.2 lymph nodes (range, 2–22 nodes) was removed from each axilla. Open exploration of the axilla after the endoscopic dissection confirmed no injuries to any neurovascular structures. Residual lymph nodes were removed from the axilla in 7 of 12 dissections (58%; average, 4.2 ± 4.0 nodes per specimen). Conclusions: These results suggest that endoscopic axillary dissection is feasible with currently available technology. Clinical trials in patients with breast carcinoma may be warranted to evaluate this technique further.