The role of pattern analysis and the ABCD rule of dermoscopy in the detection of histological atypia in melanocytic naevi.

Abstract

Clinical features of melanocytic naevi correlate poorly with the presence, histopathologically, of architectural disorder and cytological atypia, making the detection of histological atypia by means of macroscopic appearance unreliable. The aim of this study was to investigate the diagnostic effectiveness of dermoscopy in the non-invasive detection of histological atypia in naevi. Observers blinded for histological diagnosis classified a series of 168 melanocytic naevi as common or atypical on the basis of their clinical features and on their dermoscopic profile. The diagnostic performance of both methods compared with the true (histopathological) diagnosis was assessed. Dermoscopy using pattern analysis showed better results than clinical examination in the non-invasive detection of naevi with architectural disorder with or without cytological atypia (diagnostic accuracy 45% vs. 28%). A statistically significant difference in the frequency of dermoscopic parameters between atypical and common naevi was found for atypical pigment network (39% vs. 17%, P = 0.001) and dermoscopic regression structures (13% vs. 2%, P = 0.008). Dermoscopic features, which best predicted histological atypia in naevi, were regression structures (white scar-like areas or peppering), irregular vascular pattern and grey-blue areas (positive predictive values 83%, 83% and 73%, respectively). In contrast, no statistically significant difference in the mean values of the ABCD score between common and atypical naevi was found. The best diagnostic performance of dermoscopy by means of the ABCD rule (cut-off point of 4.0 of total dermoscopy score) was not dissimilar to that of clinical diagnosis (diagnostic accuracy 30%). Dermoscopy by means of pattern analysis enhances the diagnostic accuracy of dermatologists in the prediction of histological atypia in melanocytic naevi as compared with clinical examination alone.