Accuracy in skin lesion diagnosis and the exclusion of malignancy

Abstract

The accuracy of clinical diagnosis of skin lesions has important ramifications for treatment selection and importantly prioritisation for treatment. The objective of this study was to assess the accuracy of diagnosis of skin lesions within our department with an emphasis placed on whether there were any negative consequences of a missed malignant diagnosis. The study was conducted retrospectively. Accuracy of diagnosis was judged on 2 criteria. The first, if the clinical diagnosis matched the histological diagnosis. The second, if the malignancy was diagnosed correctly. 1186 lesions were excised. 57% of patients were female and the mean age was 56 (range 6-94). 25% were invasive malignancies. Clinical diagnosis was correct in 700 (66%) cases. 89% BCCs and 33% of SCCs excised were correctly diagnosed preoperatively. Misdiagnosis of BCCs or SCCs as benign was associated with a stastically significant delay in treatment (BCC 6.2 vs 10.7 weeks, p=0.02) (SCC 3.7 vs 9.5 weeks p=0.004). 100% of correctly diagnosed vs 79% of misdiagnosed SCCs were completely excised. The sensitivity and specificity of the diagnosis of MM were 87% and 97.7% respectively. The mean waiting time for patients correctly diagnosed preoperatively was 2.4 weeks vs 3 weeks (p=0.39). For malignant diagnoses sensitivity was 91%, specificity 84%, PPV 65% and NPV 96%. Misdiagnosis of skin lesions results in delays in treatment and may increase the rate of incomplete excision. The high NPV rate suggests that few malignancies are missed but those that are may have serious consequences if discharged untreated.