• James C Alex.
• Section of Otolaryngology, Yale University School of Medicine, New Haven, Connecticut, USA. james.alex@lahey.org
• Laryngoscope. 2004 Jan 1; 114 (1): 2-19.

Objectives/HypothesisThe goals of the research study were to develop an easily mastered, accurate, minimally invasive technique of sentinel node radiolocalization with biopsy (SNRLB) in the feline model; to compare it with blue-dye mapping techniques; and to test the applicability of sentinel node radiolocalization biopsy in three head and neck tumor types: N0 malignant melanoma, N0 Merkel cell carcinoma, and N0 squamous cell carcinoma.Study DesignProspective consecutive series studies were performed in the feline model and in three head and neck tumor types: N0 malignant melanoma (43 patients), N0 Merkel cell carcinoma (8 patients), and N0 squamous cell carcinoma (20 patients).MethodsThe technique of sentinel node radiolocalization with biopsy was analyzed in eight felines and compared with blue-dye mapping. Patterns of sentinel node gamma emissions were recorded. Localization success rates were determined for blue dye and sentinel node with radiolocalization biopsy. In the human studies, all patients had sentinel node radiolocalization biopsy performed in a similar manner. On the morning of surgery, each patient had sentinel node radiolocalization biopsy of the sentinel lymph node performed using an intradermal or peritumoral injection of technetium Tc 99m sulfur colloid. Sentinel nodes were localized on the skin surface using a handheld gamma detector. Gamma count measurements were obtained for the following: 1) the "hot" spot/node in vivo before incision, 2) the hot spot/node in vivo during dissection, 3) the hot spot/node ex vivo, 4) the lymphatic bed after hot spot/node removal, and 5) the background in the operating room. The first draining lymph node(s) was identified, and biopsy of the node was performed. The radioactive sentinel lymph node(s) was submitted separately for routine histopathological evaluation. Preoperative lymphoscintigrams were performed in patients with melanoma and patients with Merkel cell carcinoma. In patients with head and neck squamous cell carcinoma, the relationship between the sentinel node and the remaining lymphatic basin was studied and all patients received complete neck dissections. The accuracy of sentinel node radiolocalization with biopsy, the micrometastatic rate, the false-negative rate, and long-term recurrence rates were reported for each of the head and neck tumor types. In the melanoma study, the success of sentinel node localization was compared for sentinel node radiolocalization biopsy, blue-dye mapping, and lymphoscintigraphy. In the Merkel cell carcinoma study, localization rates were evaluated for sentinel node radiolocalization biopsy and lymphoscintigraphy. In the head and neck squamous cell carcinoma study, the localization rate of sentinel node radiolocalization biopsy and the predictive value of the sentinel node relative to the remaining lymphatic bed were determined. All results were analyzed statistically.ResultsAcross the different head and neck tumor types studied, sentinel node radiolocalization biopsy had a success rate approaching 95%. Sentinel node radiolocalization biopsy was more successful than blue-dye mapping or lymphoscintigraphy at identifying the sentinel node, although all three techniques were complementary. There was no instance of a sentinel node-negative patient developing regional lymphatic recurrence. In the head and neck squamous cell carcinoma study, there was no instance in which the sentinel node was negative and the remaining lymphadenectomy specimen was positive.ConclusionIn head and neck tumors that spread via the lymphatics, it appears that sentinel node radiolocalization biopsy can be performed with a high success rate. This technique has a low false-negative rate and can be performed through a small incision. In head and neck squamous cell carcinoma, the histological appearance of the sentinel node does appear to reflect the regional nodal status of the patient.

34 keywords...

hide…