Diagnostic significance of the blue hue in dermoscopy of melanocytic lesions - A dermoscopic-pathologic study

In epiluminescence microscopy, the perception of a blue hue is generally co nsidered a clue to malignancy, especially in clinically equivocal melanocyt ic skin lesions. However, melanocytic nevi can seldom show a blue hue under dermoscopy. The aim of the current study was to evaluate the histopatholog ic correlates of the blue hue seen in dermoscopy, to clarify its significan ce and diagnostic value. From a series of 224 consecutive pigmented skin le sions submitted to surgical excision, we selected all the melanocytic skin lesions (n. 36), blue nevi excluded, characterized by the presence of a blu e hue dermoscopically. In agreement with recent refinement of dermoscopic s emeiology, all cases were further classified in cases showing blue areas an d cases showing blue-whitish veil by experts observers blinded to the final diagnosis. Histopathologically, the series included 23 (63.9%) melanocytic nevi and 13 (36.1%) melanomas. For each lesion, several histopathologic pa rameters related to both epidermal and dermal alterations were assessed. Bl ue areas were found in 21 melanocytic nevi and 7 melanomas, whereas blue-wh itish veil was found in 6 melanomas and 2 nevi. Careful dermoscopic-histopa thologic correlation demonstrated that blue areas are related to the presen ce of large amounts of melanin pigment, either within melanophages (in the context of areas of regression) or within pigmented melanocytes. in the sup erficial dermis. Conversely, the histopathologic correlate of the blue-whit ish veil resulted in the presence of an acanthotic epidermis with compact o rthokeratosis. overlying large amounts of melanin in the dermis. Such melan in was found not only within melanocytes but also in large clusters of mela nophages within areas of regression in the dermis. In conclusion, the major ity of melanocytic lesions characterized by the presence of blue areas were histopathologically diagnosed as melanocytic nevi whereas the presence of blue-whitish veil was highly indicative of malignant melanoma diagnosis (sp ecificity 91% vs. 9% of blue areas; sensitivity 75% vs. 25% of blue areas). Thus, these two features of blue hue under dermoscopy cannot be longer con sidered as synonymous in dermoscopy setting, being associated with differen t histopathologic alterations and different diagnostic information.