The effectiveness of cauterization in preventing recurrence seems to be comparable to surgical treatment. However, we suggest endoscopic cauterization as the treatment of choice for third and fourth branchial pouch sinuses because of the lower morbidity rate.

Almost a quarter of head and neck melanomas metastasize outside clinically predicted neck levels. Neck surgery guidelines of The Netherlands Cancer Institute provide for a smaller number of discordant sentinel nodes.

The sentinel node procedure failed to identify involvement in 5.7% of the patients with lymph node metastases. Half of the false-negative biopsies took place in the first year after the procedure was introduced, illustrating the existence of a learning period.

SPECT/CT provided relevant additional information in 16 (46%) of the 35 patients. Routine use of SPECT/CT in addition to conventional lymphoscintigraphy is recommended in melanoma patients undergoing lymphatic mapping.

The sentinel lymph node biopsy was a valuable addition to our diagnostic armamentarium. The procedure improves staging, results in better prognostic information that we can share with our patients, and increases the chance of survival in node-positive patients. But, there is room for further improvement. How well does the procedure do what it is supposed to do, and how does one examine this? The main question is the following: if there is a metastasis in the nodal basin, how often does the sentinel node biopsy reveal it? This is the sensitivity. And, how often is there a recurrence in the nodal basin in a situation that no node dissection was carried out because the sentinel node was tumorfree? If this happens, the procedure was false-negative. A falsenegative result means that a metastasis was present in the nodal basin, but the sentinel node biopsy did not find it. The false-negative rates that have been published in the past few years are not good. For instance, the false-negative rate in Morton's multicenter selective lymphadenectomy Trial now stands at 20% [1]. This means that one of every five patients with nodal involvement from melanoma is missed by the sentinel node procedure.There are many potential reasons for false-negative procedures. Perhaps the concept that melanoma disseminates through the lymphatic system in a step-wise fashion from one node to the next is not watertight. Rather than to disseminate in clusters, melanoma tends to disseminate as individual cells that perhaps can pass through a node or narrow collateral lymphatics and settle in the next node downstream. The lack of a commonly accepted definition of a sentinel node implies that one man's sentinel node is not necessarily somebody else's [2]. Analysis of 33 patients with a false-negative sentinel node biopsy at the Melanoma Institute Australia showed that there were process deficiencies in nuclear medicine for 31% of these cases, in pathology for 31%, and in surgery for 13% [3]. The lymph flow may be diverted to another node-a ''neo''sentinel node-because the original sentinel node is largely replaced by metastatic disease or because the afferent lymph vessel is blocked by in-transit metastases. Lymph drainage from a melanoma is variable. Reproducibility studies show drainage to another node in 12-15% of the patients in whom lymphoscintigraphy is done twice [4]. False-negativity has also been found to be associated with increasing age, lower Breslow thickness, higher Breslow thickness, ulceration, and less lymphovascular invasion. A complete regression of a metastasis in the sentinel node but not in a subsequent node may also be a potential cause for failure [3].How does the false-negative rate for melanoma compare to that for other cancers? A recent study of 153 patients who underwent sentinel lymph node biopsy for Merkel cell carcinoma revealed a false-negative rate of 15% [5]. The median duration of follow-up was 41 months. The false-negative rate was 19.2% in a series of 92 penile cancer patients with a median follow-up duration of...