Semi-quantitative assessment of the distribution of skin lesions in patients with psoriasis and psoriasis arthritis.

Background: We propose that the distribution of skin lesions in psoriasis may be assessed using parametric maps on a pixel-by-pixel basis.

Material And Methods: We processed 428 patient-drawn self-descriptions of the psoriasis lesions on a supplied body template. We compared 195 patients with a confirmed diagnosis of psoriatic arthritis (PsA) with 89 who had this diagnosis rejected (Psor). Additionally, 28 Psor cases supplied drawings performed after 3 weeks of climate therapy (PsorCT) to test the treatment efficacy. The drawings were scanned, lesion areas were segmented, followed by construction of parametric maps of lesion distributions and calculation of statistical differences between groups.

Results And Discussion: In PsorCT, the lesions occupied 11.2% (0-42%) [median (min.-max.)] of the body area. The area decreased to 2.4% (6-11%) after heliotherapy. The differences were statistically significant for all the areas studied and spread evenly over the body surface. PsA had a relatively low psoriasis lesion occupancy of 2.5% (0-42%) compared with Psor 9.8% (0-34%), which is attributed to the difference in recruitment. Correcting for this, we demonstrate a clear tendency for the head, palms, feet, groin and nails to be preferred lesion sites in PsA in contrast to psoriasis.

Conclusion: Pixel-based analysis of self-reported skin lesion distributions is a powerful tool to assess systematic differences due to treatment or disease variants.