CORONADO, CALIF. — Dermatoscopy can identify melanomas as small as 3 mm, but should be combined with a careful exam for the best diagnosis, Dr. James W. Steger said at an update on melanoma sponsored by the Scripps Clinic.
Researchers evaluated 349 consecutive patients who had 375 suspicious lesions requiring biopsy. Of these, 161 were 6 mm or smaller and 13 were melanomas. Clinical diagnosis alone detected 10 of 13, for a sensitivity of 77% and a specificity of 74%. Dermatoscopy alone also detected 10 of 13. Clinical and dermatoscopy criteria combined detected all 13 (Eur. J. Dermatol. 2002;12:573–6).
In a follow-up study, the researchers compared clinical exam with dermatoscopy for diagnosing 203 sequential pigmented lesions smaller than 3 mm in diameter (Br. J. Dermatol. 2006;155:570–3). In this study, 10 of 23 melanomas were diagnosed by clinical exam alone while dermatoscopy using Menzies score picked up 19 of the 23, which means that, for “very small melanoma [3 mm and under] the diagnostic rate of dermatoscopy is about double what it is for the naked eye,” said Dr. Steger, chair of the department of dermatology at Naval Medical Center San Diego. He then discussed two easy screening algorithms in dermatoscopy.
The first is the three-color test. After review of 74 pigmented lesions referred for excision, the most powerful criterion correlating with a histopathologic diagnosis of melanoma was the presence of three or more colors seen in the lesion on dermatoscopy. Sensitivity was 92%. Specificity was only 51% (Br. J. Dermatol. 2002;146:481–4).
“That's okay, since this is a screening technique,” Dr. Steger said.
The second algorithm is the three-criteria checklist. Criteria include asymmetry of color or dermoscopic structures, atypical pigment network, a “tennis net-like” pattern of irregular holes and thick lines, and the presence of any type of blue or white colors (Dermatology 2004;208:27–31).
Six nonexperts underwent 1 hour of training and applied the criteria to 231 consecutively excised pigmented lesions. Results were compared with those of an expert who used dermatoscopy with the pattern analysis method of diagnosis.
The nonexperts had a sensitivity of 96% and a specificity of 33%. The expert had a sensitivity of 90% and specificity of 94% using dermatoscopy.
The most powerful criterion correlating with the diagnosis is the presence of three or more colors in the lesion. Courtesy Dr. James W. Steger