GP2d - Novel Classification System and Surgical Algorithm for Facial Congenital Melanocytic Nevi

Background: Congenital melanocytic nevus (CMN) is characterized by pigmented skin patches. Large CMN affects 1 in 20,000 newborns globally and runs the risk of melanoma transformation. A classification system for CMN patterns has been previously established for the trunk and extremities. However, there is no classification system for facial CMN to date. This study proposes a facial CMN classification scheme that can help guide management decisions.

Methods: Facial CMN images were retrieved from public repositories. The outline of the facial CMNs were extracted and transposed onto a standardized 3D model. To identify distinct patterns, the CMN images of anatomic sub-regions of the face were clustered and heat-maps were generated. The clinical features of the CMNs were standardized based on previously published criteria.

Results: The study identified 291 images of facial CMNs. The CMNs could be categorized into seven distinct patterns based on anatomic distribution. Each CMN could be further characterized according to estimated size category, color heterogeneity, rugosity, nodularity, and degree of hypertrichosis.

Discussion: The study proposes a novel classification scheme for facial CMN to characterize the variability in the anatomic distribution. Researchers further aim to test the validity of their scheme on a CMN image dataset from collaborating institutions.

Conclusion: A better understanding of the patterns of facial CMNs can glean insights into the embryonic pathways of melanocytic migration on the face. Grouping of facial CMNs by patterns may guide the development of treatment strategies to best preserve the esthetics and functionality of a child’s face.