Clinical Diagnosis has a High Negative Predictive Value in Evaluation of Malignant Skin Lesions

Abstract

Background. The increasing incidence of skin cancers in fair-skinned population and its relatively good response to treatment make its accurate diagnosis of great importance. We evaluated the accuracy of clinical diagnosis of malignant skin lesions by comparing the clinical diagnosis with histological diagnosis as the gold standard. Materials and Methods. In this retrospective study, we assessed all the pathology reports from specimens sent to a university hospital laboratory in 3 consecutive years from March 2008 to March 2010. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive and negative likelihood ratios were calculated for clinical diagnosis of malignant skin lesions stratified by their histological subtype. Results. A total 4,123 specimen were evaluated. The sensitivity and specificity for clinical diagnosis of malignancy were 90.48% and 82.85%, respectively, whereas the negative predictive value was shown to be 99.06%. The positive and negative likelihood ratios were 5.23 and 0.11, respectively. Conclusion. Pathological assessment of skin lesions remains the cornerstone of skin cancer diagnosis. The high NPV and the relatively low PPV indicate that clinical diagnosis is more efficient in ruling out malignancies rather than diagnosing them.