Improvement of malignant/benign ratio in excised melanocytic lesions in the 'dermoscopy era': a retrospective study 1997-2001

Abstract Background Because of the many limitations of studies based on the diagnostic setting of excised lesions, the impact of dermoscopy (epiluminescence microscopy, dermatoscopy) in melanoma screening during practice remains to be established. Objectives We assumed that effects of the use of dermoscopy on some indicators of diagnostic performance in melanoma screening should be traceable retrospectively; therefore, we analysed the impact of routine dermoscopy use on the malignant ⁄ benign ratio in excised melanocytic lesions. Methods Preoperative and histological diagnosis of 3053 melanocytic lesions [319 melanomas (10Æ4%)] consecutively diagnosed and excised at the Department of Dermatology, University of Florence in the period 1997–2001 inclusive were retrieved. Six dermatologists who selected the lesions to excise and who performed preoperative diagnosis were divided into two groups according to their use of dermoscopy in routine activity (n ¼ 2 dermoscopy users and n ¼ 4 nonusers). The study period was divided into a predermoscopy period (1997), a shift phase (1998) and a dermoscopy period (1999–2001). Results During the study period, the malignant ⁄ benign ratio improved in dermoscopy users only (from 1 : 18 to 1 : 4Æ3, P ¼ 0Æ037). No significant difference was found for nonusers (from 1 : 11Æ8 to 1 : 14Æ4). Dermoscopy users were more likely to have a melanoma diagnosed within a series of excised lesions than nonusers, even taking into account potential confounders such as sex, age and study period by means of multivariate analysis (odds ratio 1Æ55, 95% confidence interval 1Æ17– 2Æ01). The percentage of problem naevi (naevi with architectural disorder with or without cytological atypia and Spitz or Reed naevi) over the total number of excised lesions was higher in dermoscopy users than in nonusers (year 2001, 51Æ6% vs. 40Æ9%, P ¼ 0Æ014). Similar findings were obtained after exclusion from the data set of lesions excised for cosmetic reasons. Conclusions The adoption of dermoscopy in routine melanoma screening is followed by an improvement of the malignant ⁄ benign ratio in excised lesions, suggesting a more appropriate selection of pigmented lesions referred to surgery. Because of the possible limitations of a retrospective study design, future confirmation of this finding by means of a prospective, randomized study is advisable. The introduction of dermoscopy in routine practice may have major implications in large-scale melanoma screening with cost savings and a reduction of the dermosurgery workload.