The purpose of this study was to investigate whether a computer-assisted gamma probe with adjustable collimation could aid in the detection of sentinel nodes (SNs) and to analyse the patterns of recurrence and prognosis in SN-positive and SN-negative cases. We analysed 385 SN biopsies. The SN identification rate was 87.2% using preoperative lymphoscintigraphy and blue dye, 93.9% using preoperative lymphoscintigraphy, blue dye and different probes, and 100% using preoperative lymphoscintigraphy, blue dye and a computer-assisted probe with adjustable collimation. The computer-assisted probe was particularly advantageous in cases where the melanoma was located very close to the SN and in cases of deep-seated nodes or nodes with low uptake, due to the possibility of changing the collimation during the procedure. The SN-positive rate according to the thickness of the primary melanoma was 1.7% for melanomas < or = 1 mm in thickness and 27.5% for melanomas > or = 1 mm. In 4.9% of cases we identified nodes outside the regional nodal basin. In one case we found a micrometastasis in a blue and hot interval node of the lateral abdominal wall. Analysing the node counts registered by the computer-assisted probe, we verified that the blue-positive node for tumour metastases was not the most radioactive node in the field in six out of 52 positive cases (11.5%). Distant metastases were present in 2.0% of SN-negative patients, and in 24% of SN-positive patients (P < 0.001). Highly statistically significant differences were found between SN-negative and SN-positive patients in both the 3 year disease-free survival (86.3% versus 49.2%) and the 3 year disease-specific survival (92.3% versus 77.1%) (P < 0.001).
Copyright 2004 Lippincott Williams & Wilkins