Clinical, dermoscopic and pathologic correlation of pigmentary lesions observed in a dermoscopy service in the year 2003

Aim. The aim of the study is to evaluate the role of dermoscopy in diagnosing pigmentary skin lesions, and to improve the early diagnosis of cutaneous malignant melanoma. Methods. A total of 520 pigmentary skin lesions from 349 consecutive patients were evaluated using a video microscope. Clinical diagnosis and dermoscopy performance in recognising histological features of pigmentary skin lesions were compared on 79 surgically excised lesions. Dermoscopy sensitivity and specificity in detecting melanoma were also evaluated and compared, using K statistics, with clinical examination performance in detecting melanoma. Results. Overall agreement for dermoscopy was 82.3% (82.1% and 83.3% for melanocytic and non melanocytic lesions, respectively). Overall agreement for clinical examination was 58.2% (53.7% and 83.3% for melanocytic and non melanocytic lesions, respectively). Dermoscopy sensitivity in detecting melanoma proved to be 91.6%, specificity 98.2%. Clinical examination sensitivity and specificity in detecting melanoma proved to be 66.6% and 94.5%, respectively. K statistic values (agreement with histological diagnosis) were 0.87 and 0.63 for dermoscopy and clinical examination respectively. Conclusions. Our study confirms that dermoscopy is helpful to the clinician in diagnosing cutaneous melanoma and in deciding which pigmented lesions need surgical excision as well.