This article is more than 5 years old. Information may no longer be current.

First step of two-step dermoscopy algorithm highly sensitive, specific

The first step of the two-step dermoscopy algorithm displayed a high level of sensitivity, specificity and accuracy, according to recently published study results.

Introduced in 2001 by a panel on the virtual Consensus Net Meeting on Dermoscopy, the two-step dermoscopy algorithm has undergone minor modifications, the researchers wrote.

The researchers conducted a retrospective study of all skin lesions evaluated through the two-step dermoscopy algorithm between Jan. 1, 2001, and Dec. 31, 2010, at a high-risk dermatology clinic at Sloan Kettering Cancer Center. Classifications of melanocytic or nonmelanocytic were given to prebiopsy and histopathology diagnoses.

Histopathological lesion classification was used to estimate diagnostic accuracy, sensitivity, specificity, positive predictive value and negative predictive value for the first step of the algorithm, which were the study’s main outcomes.

There were 9,168 lesions biopsied during the study period. In the first step of the algorithm, sensitivity for correctly identifying melanocytic lesions was 85%, whereas specificity was 94%. There were 667 lesions (approximately 7%) that had discordant clinical and histopathological classifications, including 415 (4.5%) being false-positive lesions, which were clinically misclassified as melanocytic. Additionally, 252 (2.7%) were false-negative, which were clinically classified as nonmelanocytic, according to the researchers.

Interdermal nevus misclassified as basal cell carcinoma (BCC) and nonmelanocytic lesions misclassified as melanocytic since they mimic melanoma were common classification errors. Eight melanomas (2.1% of 381 melanomas) were misclassified as nonmelanocytic, with various clinical diagnoses including BCC and squamous cell carcinoma. – by Bruce Thiel

Disclosure: The researchers report no relevant financial disclosures.