Effect of lesion size on the diagnostic performance of dermoscopy in melanoma detection

Abstract

Background: Dermoscopy is able to correctly classify a higher number of melanomas than naked-eye examination. Little is known however about factors which may influence the diagnostic performance during practice. The aim of the study was to analyze the effect of size of the lesion on diagnostic performance of dermoscopy in melanoma detection.

Methods: Eight dermatologists examined clinical and, separately, clinical and dermoscopic (combined examination) images of 200 melanocytic lesions previously excised [64 melanomas, 24 in situ and 40 invasive (median thickness 0.30 mm) and 136 melanocytic nevi]. After examination, diagnostic performance was analyzed in accordance with the major diameter of the lesions divided into 3 groups, i.e. small (less than 6 mm), intermediate (between 6 and 9 mm) and large (10 mm or more) lesions. These groups were shown to be highly comparable concerning the microstaging of melanomas (median thickness value 0.30, 0.22 and 0.32 mm, respectively).

Results: Dermoscopy increased the diagnostic performance of naked-eye examination of both intermediate and large lesions [sensitivity value: +19.3 (p = 0.002) and +10.3 (p = 0.007); diagnostic accuracy value: +7.4 (p = 0.004) and +6.1 (p = 0.07)]. On the contrary, no statistically significant increase was found dealing with small lesions (sensitivity +3.7, p = 0.66; diagnostic accuracy -1.7, p = 0.55).

Conclusions: The diagnostic improvement associated with the addition of dermoscopy to naked-eye examination is influenced by the size of the lesion, i.e. it is lacking with lesions up to 6 mm in diameter. The optimized use of dermoscopy in melanoma detection is obtained dealing with melanocytic lesions 6 mm in diameter or larger.