FC-50 Results of repeat intradermal testing in 135 dogs

Objective assessment of canine atopic dermatitis severity is very difficult and only a few scoring systems have been developed. The most commonly used is the Canine Atopic Dermatitis Extent and Severity Scoring Index (CADESI), adapted from the human SCORAD. Despite wide use of this score in clinical trials, no validation has been performed to our knowledge. The aim of this study was to determine the reliability of the CADESI in clinical practice. First, a set of 28 pictures taken from dogs diagnosed with atopic dermatitis was scored by six different investigators for three items: erythema, lichenification and excoriation. Next, 23 dogs with clinical signs compatible with atopic dermatitis were graded by two investigators using the CADESI. Erythema, lichenification and excoriation were assessed on 39 areas. With the pictures, significant correlations (Spearman's r , P  < 0.05) were found for each combination of investigators for erythema and lichenification, but only in 10 of 15 combinations for excoriation. Interobserver agreement ranged between poor and fair (0.221 < Cohen's κ <0.508, mean = 0.395). For the living animals, significant correlations ( P  < 0.0001), but poor interobserver agreement, were found for the three items (κ_erythema = 0.366, κ_{lichenification} = 0.385, and κ_excoration = 0.226). A significant correlation ( P  < 0.05) was found for each location, and interobserver agreement varied between very poor and good (0.16 < κ < 0.66). These results suggest that erythema and lichenification are reliably assessed, but that grading excoriation is more difficult. Also, the assessment of severity varied depending on the site studied.
Funding: Self-funded.