DS23: Inconsistencies in lesion removal in secondary care dermatology

Areej Karim, Daniel Shaw and Manu Shah

East Lancashire Hospitals NHS Trust, Blackburn, UK

Diagnostic accuracy in identifying and removing skin lesions is essential to conserve scarce healthcare resources, reduce unnecessary surgery and reduce psychological distress to patients. Clinician accuracy can be assessed by calculating benign to malignant ratios and diagnostic concordance. Data from 274 patients operated on in secondary care dermatology with a skin lesion(s) over a 4-month period were collected. Multiple parameters were recorded, including type of procedure performed and concordance between suspected clinical diagnosis and final histological diagnosis. Benign to malignant ratio for skin lesions was calculated, as well as the ratio between benign and malignant plus premalignant lesions. Data were analysed according to individual clinician. In total, 296 procedures were studied. Ten were excluded owing to inadequate data. The most common procedure performed was diagnostic punch biopsy (n = 125; 42·2%) followed by full excision (n = 116; 39·2%) and double curettage and cautery (n = 53; 18·0%). Lesions were defined as benign, malignant (melanoma, squamous cell carcinoma, basal cell carcinoma, rare cancers) or premalignant (Bowen disease, actinic keratosis, melanoma in situ). Dysplastic naevi were defined as premalignant for the purpose of this study. Overall, 117 (39·5%) lesions excised were benign, 72 (24·3%) were premalignant and 107 (36·2%) were malignant. Overall concordance between suspected and histological diagnosis was 48%. Benign to malignant ratio of lesions was 1·67. Benign to malignant + premalignant ratio was 0·66. There was significant variation between clinicians. Clinicians with a lower diagnostic concordance had a higher benign to malignant ratio. The benign to malignant ratio varied between clinicians from 1·24 to 4·8 and the benign to malignant + premalignant ratio varied between clinicians from 0·48 to 1·07. Overall, the number of skin lesions needing to be removed to discover a malignancy was 2·66. This study has shown differences in diagnostic accuracy between clinicians. There was a significant variation between clinicians’ benign to malignant ratio regarding operations on lesions. It may be possible to reduce the number of procedures on benign lesions through better diagnostic skills with dermoscopy and upskilling clinicians. Dermatology departments should strive to provide more education and training on lesion recognition and to reduce the differences between individuals. Mentoring and regular feedback may be helpful.