Developing and validating the Cutaneous WARTS (CWARTS) diagnostic tool: a novel clinical assessment and classification system for cutaneous warts

Abstract

Background : The clinical appearance of cutaneous warts is highly variable and not standardized. Objectives : To develop and validate a reproducible clinical tool for the standardized assessment of cutaneous warts to distinguish these lesions accurately. Methods : Nine morphological characteristics were defined and validated regarding intra- and interobserver agreement. Based on literature and semistructured interviews, a systematic dichotomous assessment tool, the Cutaneous WARTS (CWARTS) diagnostic tool was developed. The validation consisted of two independent parts performed with photographs from the recent WARTS-2 trial. In part A, the CWARTS diagnostic tool was tested by 28 experienced physicians who assessed photographs of 10 different warts to investigate interobserver concordance. In part B, morphological characteristics were validated by masked and independent scoring of 299 photographs by six different observers. Part B also entailed reassessment of the photographs after at least 1 week. The primary outcome measurement was the intraclass correlation coefficient (ICC). Results : Presence of black dots (capillary thrombosis) had the greatest ICC (0·85) for interobserver agreement in part A, followed by arrangement (0·65), presence of border erythema (0·64) and sharpness of the border (0·60). In part B, results were similar for interobserver agreement with presence of black dots having the highest ICC (0·68), followed by border erythema (0·64), arrangement (0·58) and colour (0·55). For intraobserver agreement, presence of black dots had the highest agreement (0·70), followed by presence of border erythema (0·694) and colour (0·59). Conclusions : Wart phenotype can be reliably assessed using the CWARTS diagnostic tool.