Probabilistic issues with sentinel lymph nodes in malignant melanoma.

Objectives: To address issues of probability for sentinel lymph node results in melanoma and provide details about the probabilistic nature of the numbers of sentinel nodes as well as to address how these issues relate to tumor thickness and patient outcomes.

Methods: Analysis of the probability of observing sentinel node metastases uses the discrete exponential probability distribution to address the number of observed positive sentinel nodes. In addition, mathematical functions derived from survival analysis are used. Data are then chosen from the literature to illustrate the approach and to derive results.

Results: Observations about the numbers of positive and negative sentinel nodes closely follow discrete exponential probability distributions, and the relationship between the probability of a positive sentinel node and tumor thickness follows closely a function derived from survival analysis. Sentinel node results relate to tumor thickness as well as to the total number of nodes harvested but fall short of identifying all those who eventually develop metastatic melanoma.

Conclusions: Probability analyses provide useful insight into the success and failure of the sentinel node biopsy procedure in patients with melanoma.