Dermoscopic criteria for melanoma in situ are similar to those for early invasive melanoma

BACKGROUND. Dermoscopy is a noninvasive technique that increases the diagnostic accuracy of pigmented skin lesions, particularly improving the diagnosis of patients with cutaneous melanoma in situ (CMIS) and early invasive melanoma. To establish reliable and reproducible dermoscopic criteria for the diagnosis of CMIS, the authors conducted a retrospective clinical study of 37 patients with CMIS and 53 patients with invasive cutaneous melanomas (ICM). METHODS. The 37 patients with CMIS were divided into three groups: Those with CMIS lesions measuring ≤ 5 mm in greatest dimension (8 patients), those with CMIS lesions measuring from > 5 mm to ≤ 10 mm in greatest dimension (20 patients), and those with CMIS lesions measuring > 10 mm in greatest dimension (9 patients). The 53 patients with ICM were divided into two groups according to Breslow index: Those with 1CM lesions measuring ≤ 0.75 mm in tumor thickness (19 patients) and those with ICM lesions measuring > 0.75 mm in tumor thickness (34 patients). Lesions were examined with a dermatoscope and were photographed at a magnification of × 10. Dermoscopic criteria were evaluated from examination of the photomicrographs. RESULTS. Blue-whitish veil, gray-blue areas, black dots, and irregular extensions and branched streaks were the most relevant dermoscopic criteria for CMIS and were present in 78%, 76%, 73%, and 62% of lesions, respectively. Brown globules, irregular pigment network, pseudopods, and depigmentation were present in 57%, 54%, 54%, and 51% of CMIS lesions, respectively. White scar-like areas and linear and/or dotted vascular patterns, two criteria that are associated frequently with ICM, were not found in our patients with CMIS. No clinically significant differences were observed between the three groups of CMIS patients. CONCLUSIONS. Dermoscopic criteria for CMIS were similar to those for 1CM, although white scar-like areas and linear and/or dotted vascular patterns were observed only in patients with 1CM. Dermoscopic criteria appeared to be independent of CMIS lesions size.