The diagnostic criteria of pityriasis rosea and Gianotti-Crosti syndrome - a protocol to establish diagnostic criteria of skin diseases.

We established and validated diagnostic criteria for pityriasis rosea and Gianotti-Crosti syndrome. In this paper, we compare and contrast both diagnostic criteria to formulate a protocol in establishing diagnostic criteria for other dermatological diseases. The diagnostic criteria are similar in employing clear dividing lines and conjunctions ('and/or') to assure high reliability. Both sets of criteria should be applicable for all ethnic groups. Spontaneous remission is not included, so diagnosis is not delayed while waiting for disease remission. Laboratory investigations are not enlisted, so that the criteria can be used in medical care systems in different parts of the world. The diagnostic criteria are different in that pathognomonic clinical manifestations exist for pityriasis rosea, such as the herald patch and the orientation of lesions along the lines of skin cleavages. These features, however, score low for sensitivity. These specific manifestations are not seen in Gianotti-Crosti syndrome. Such differences led to different categorisation of clinical features. Atypical variants are more common for pityriasis rosea. The diagnostic criteria for pityriasis rosea therefore do not include a list of differential diagnoses, while diagnostic criteria for Gianotti-Crosti syndrome do. Using this comparison, we constructed a protocol to establish diagnostic criteria for other skin diseases. We advocate the need to justify the establishment of diagnostic criteria, that multiple diagnostic criteria for the same disease should be avoided, that diagnostic criteria should be compatible with the disease classification if applicable, and that the scope should be well-delineated with regard to clinical variants. We outline the need for validation studies to assess the criteria-related validity, test-retest intra-clinician reliability, and inter-clinician reliability. We emphasise that the establishment of diagnostic criteria should not be a generic process. We also highlight limitations of diagnostic criteria, and emphasise that no diagnostic criteria can replace the bedside experience of clinicians.