Clinical Question: Does use of a dermoscope improve the accuracy of melanoma diagnosis?

Setting: Outpatient (any)

Study Design: Diagnostic test evaluation

Synopsis: Dermoscopy, also called dermatoscopy, involves the use of a low-power, handheld microscope (typically 10×) used with or without oil immersion to view suspicious skin lesions (http://www.dermoscopy.org/default.asp). Previous studies have shown that, in expert hands, dermoscopy increases the accuracy of diagnosis. In this study, 61 physicians (35 primary care physicians, 10 dermatologists, and 16 dermatology trainees) were presented with unaided photos of 40 lesions and asked to use their standard clinical judgment to make a diagnosis of melanoma or nonmelanoma. They were then presented with dermascopic images of the same lesions four times in succession and asked to apply four different sets of standard, validated criteria for diagnosing melanoma (the ABCD rule, the Menzies method, a 7-point checklist, and pattern analysis). They were instructed in each of these methods using a CD-ROM. Interrater reliability (how well physicians agreed with one another) for key melanoma diagnostic features was fair, with kappas (a major of agreement where zero is perfect disagreement and 1.0 is perfect agreement) in the range from 0.21 to 0.56. The unaided eye, using a standard photo of the lesion, was 61 percent sensitive and 85 percent specific with 73 percent diagnostic accuracy. The dermoscopic photo used in conjunction with each of the four sets of criteria were more sensitive (68 percent for pattern analysis, 77 percent for the ABCD rule, 81 percent for the 7-point checklist, and 85 percent for the Menzies method). The specificity of pattern analysis is similar to the clinical examination (85 percent). The specificity of all other methods ranged from 73 to 80 percent. In this situation, sensitivity is more important than specificity; you do not want to miss any cancers, and the biopsy is a more specific test that, it is hoped, will sort out the nonmalignant tumors.

Bottom Line: Dermoscopy, especially when used in conjunction with the Menzies method, is more sensitive than the unaided eye in diagnosing melanoma. This technique is not widely used in the United States and deserves further study. We do not know whether it was the validated criteria or the use of dermoscopy that made the intervention more sensitive. Note that this study does not indicate if patients diagnosed using dermoscopy live longer or have better lives: it is not a POEM. Rather, it is a surrogate marker of relevant outcomes, if one assumes that more accurate diagnosis of melanoma leads to better outcomes. (Level of Evidence: 2b)