PURPOSE: To describe the lymphatic drainage patterns of the human eyelids. METHODS: Twenty-eight consenting patients who underwent unilateral eyelid surgery at McMaster University between March 2001 and July 2003 had their contralateral eyelids injected with 0.2 ml (0.250 mCi) of Tc 99 m sulphur colloid. The patients were divided into 1 of 5 injection sites of the eyelid, namely upper lateral, upper medial, medial canthus, lower medial, and lower lateral. Lymphoscintigraphy was performed between 2 and 6 hours later with a conventional planar gamma camera. Nodes in the head and neck were identified. In 15 patients, the right eye was injected, and in 13 patients, the left eye was injected. RESULTS: Three patients had no nodes that were identifiable. The remaining 25 patients had at least one node identified. In 11 patients, more than one node was identified. In 18 patients, the preauricular node was most intense and recognized first. Regardless of location on the eyelid, the sentinel node was most commonly the preauricular node. CONCLUSIONS: These results conflict with previously described classic drainage patterns of the eyelid lymphatics. In 72% (18/25) of cases, the first-order sentinel node was the preauricular node, regardless of location of the injection site on the eyelid. Many individuals did not fit the classic drainage patterns.
PURPOSE: To describe the lymphatic drainage patterns of the human eyelids. METHODS: Twenty-eight consenting patients who underwent unilateral eyelid surgery at McMaster University between March 2001 and July 2003 had their contralateral eyelids injected with 0.2 ml (0.250 mCi) of Tc 99 m sulphur colloid. The patients were divided into 1 of 5 injection sites of the eyelid, namely upper lateral, upper medial, medial canthus, lower medial, and lower lateral. Lymphoscintigraphy was performed between 2 and 6 hours later with a conventional planar gamma camera. Nodes in the head and neck were identified. In 15 patients, the right eye was injected, and in 13 patients, the left eye was injected. RESULTS: Three patients had no nodes that were identifiable. The remaining 25 patients had at least one node identified. In 11 patients, more than one node was identified. In 18 patients, the preauricular node was most intense and recognized first. Regardless of location on the eyelid, the sentinel node was most commonly the preauricular node. CONCLUSIONS: These results conflict with previously described classic drainage patterns of the eyelid lymphatics. In 72% (18/25) of cases, the first-order sentinel node was the preauricular node, regardless of location of the injection site on the eyelid. Many individuals did not fit the classic drainage patterns.