Interobserver agreement on dermoscopic features of pigmented basal cell carcinoma

BACKGROUND. A dermoscopic method based on the absence of a pigment network and the presence of at least one of six positive features has been described for diagnosis of pigmented basal cell carcinoma (BCC). OBJECTIVE. To evaluate the observers' global agreement and interobserver agreement on each dermoscopic parameter of the method recently proposed. METHODS. Dermoscopic images of 56 pigmented BCCs were examined by five observers with different degrees of experience in dermoscopy. RESULTS. An overall full agreement was reached for the absence of pigment network (k = 1). Very good agreement was detected for the presence of spoke wheel areas (k = 0.85) and arborizing vessels (k = 0.72), and good agreement was shown for ulceration (k = 0.49) and multiple blue-gray globules (k = 0.41). No agreement was identified on large blue-gray ovoid nests (k = 0.28) and leaflike areas (k = 0.26). CONCLUSION. We confirm the reproducibility of the method and show that ulceration, spoke wheel areas, and arborizing telangiectases represent the most robust positive parameters.