ABSTRACT
Scintigraphic identification of the sentinel lymph node is achievable in nearly all patients with malignant melanoma. However, in a very small number of cases the sentinel node fails to be detected, and sometimes recurrence appears during follow-up in patients who had previously tested negative for metastatic disease. The purpose of this study was to review our experience in order to isolate the reasons for erroneous sentinel lymph node identification. The evaluation involved 435 consecutive malignant melanoma patients with AJCC stages I and II (clinically negative nodes) and Breslow thickness >0.76 mm. Lymphoscintigraphy was performed the day before surgery by intradermal administration of technetium-99m labelled nanocolloid. Dynamic and static images were obtained. The sentinel node was intraoperatively identified with the aid of patent blue dye and a hand-held gamma probe. After removal, routine histopathological examination with haematoxylin-eosin (H-E) and immunohistochemistry with S 100 and HMB45 (IHC) were performed. In those patients who developed regional recurrences during follow-up, sentinel nodes were further evaluated by reverse transcriptase-polymerase chain reaction (RT-PCR). Lymphoscintigraphy visualised at least one sentinel node in 434 out of 435 patients (99.8%). Uptake in in-transit sentinel lymph nodes was observed in 32 patients (7.4%). During surgery, localisation and removal of sentinel nodes was successful in 430/435 patients (98.8%). A total of 790 sentinel lymph nodes were harvested, with a mean of 1.8 per patient. Routine histopathological examination with H-E or IHC revealed metastatic disease in 72 patients (16.8%). During a mean follow-up of 26 months, seven of those patients with a negative sentinel node developed regional lymph node metastases. In five of them RT-PCR was positive for micrometastases within the sentinel node. In conclusion, erroneous sentinel lymph node identification can be due to changes in the surgical team, difficult lymph node location or absence of a thorough histological study. Nevertheless, it is not possible to explain completely why, in a very small percentage of cases, the sentinel node is erroneously identified.
