Expert Consensus Defines Pyoderma Gangrenosum Phenotypes

Key Takeaways

• PG-expert dermatologists in a modified Delphi exercise agreed on a descriptive classification above their prespecified consensus threshold.
• The agreed vocabulary separates PG associated with autoinflammatory syndromes from nonsyndromic PG.
• Within nonsyndromic PG, associated disease and drug exposure define phenotypes, while special-site involvement and extracutaneous manifestations modify their presentation.

Pyoderma gangrenosum (PG) is a rare neutrophilic dermatosis characterized by sterile ulcerative skin lesions. It occurs alongside varied comorbidities and clinical features, but standardized guidance for naming its phenotypes has been lacking. A shared vocabulary could clarify the distinction between an associated condition and a feature of an individual presentation. That terminology gap prompted investigators to seek expert agreement.

Investigators developed the JAMA Dermatology pyoderma gangrenosum phenotype classification through a modified Delphi expert consensus exercise, not a patient treatment trial or clinical practice guideline. The panel comprised 23 board-certified dermatologists with PG expertise who were Medical Dermatology Society members. They completed five anonymous, iterative online survey rounds between December 2023 and July 2025. Before voting began, nonvoting researchers searched PubMed from inception for English-language literature on PG clinical associations and comorbidities. Panelists rated proposed phenotypes and disease modifiers against a prespecified consensus threshold. Later rounds presented anonymized comments and aggregate results, and the final round tested the assembled framework.

Agreement on the final framework reached 83%, exceeding the prespecified threshold of 70% or greater. The classification distinguished PG with autoinflammatory syndromes from nonsyndromic PG. The nonsyndromic phenotypes were inflammatory bowel disease–associated PG; PG associated with hematologic cancers and blood dyscrasias; drug-induced PG; and other PG, including idiopathic cases. These labels distinguish syndromic disease from the associations used to describe nonsyndromic presentations.

Special-site involvement and manifestations beyond the skin were treated as modifiers rather than additional phenotypes. Involvement of the head or neck, genitals, or peristomal skin was designated a modifier, as were extracutaneous manifestations. These features describe where or how PG presents within the agreed vocabulary.

Expert agreement on descriptive terminology does not establish that using the classification improves diagnostic accuracy, treatment selection, treatment response, or patient outcomes. According to the authors, consistent nomenclature may inform future clinical guidelines and support more uniform reporting of epidemiologic findings and outcomes among people with PG. Those prospective uses do not make the classification a current treatment guideline.

The authors concluded that the shared classification differentiates autoinflammatory syndrome–associated from nonsyndromic PG while allowing modifiers to describe presentation.

Clinician Questions

What kinds of publications informed the pyoderma gangrenosum phenotype classification?

Nonvoting researchers searched English-language publications on pyoderma gangrenosum clinical associations and comorbidities, identifying observational studies, narrative reviews, systematic reviews, and meta-analyses.

Does the pyoderma gangrenosum phenotype classification recommend treatments for individual phenotypes?

Investigators proposed descriptive phenotype and disease-modifier nomenclature, not an algorithm assigning drugs or regimens to individual pyoderma gangrenosum phenotypes. Its possible use in future clinical guidelines remains prospective.