Background: It is mandatory that a new diagnostic method be validated by comparison with a well-established reference procedure before being introduced for use in clinical practice. In the field of pigmented skin lesions (PSL), clinical examination should be considered the reference procedure for new diagnostic methods, such as dermoscopy. However, it has not yet been established which is the best photographic procedure for obtaining the most informative clinical images to be used in a formal diagnostic setting.
Methods: In this study we investigated the diagnostic information provided by the two most popular methods currently used for clinical photographing of PSL: "contact" images obtained with a Dermaphot (Heine Ltd) at original x 10 magnification without oil application and "distant" images obtained with a macro objective from a distance of 10 cm.
Results: Two experienced dermatologists observed clinical images of a series of 57 PSL (11 melanomas, 31 melanocytic nevi, 10 pigmented basal cell carcinomas, and four other diagnoses). The degree of concordance between the diagnoses based on "contact" and "distant" images (melanoma/non-melanoma) was very good (k = 0.819). Regarding histology, the degree of concordance was better when the diagnosis was based on "contact" images (k = 0.54) than "distant" images (k = 0.47). In particular, "contact" images were superior to "distant" images for diagnosis of non-melanoma lesions (specificity of melanoma diagnosis 87.7% vs. 83.6%), but we found no difference in melanoma detection (73% of sensitivity for both methods).
Conclusions: Although the two photographic procedures appear to provide similar levels of diagnostic information, the "contact" method seems to provide a higher specificity for melanoma diagnosis.