Medical Calculators
Ulcerative PG Diagnostic Corridor
Delphi consensus criteria and the PARACELSUS score, scored together, with unanswered findings kept explicit and each framework reported separately.
Decision support, not a diagnosis. Both instruments were derived and validated for the ulcerative subtype only, and neither replaces assessment of mimickers by an experienced clinician. A threshold-met result does not establish PG or authorize treatment. Nothing entered here is stored or transmitted. Use a non-identifying case code. Evidence reviewed August 2026.
Look at the wound
0 of 7 answeredErythematous to livid rim at the advancing edge. Present in about 98 percent of PG cases in the PARACELSUS derivation cohort.
Delphi minor criterion 5 requires peripheral erythema, an undermined border, and tenderness together.
The edge overhangs the bed and a probe passes beneath intact skin.
Record examination tenderness independently of the patient's spontaneous pain score.
Bizarre, geographic or serpiginous outline rather than the round or oval shape of a vascular ulcer.
Delphi minor 6. Both parts are required.
Delphi minor 7. Highly discriminating in the derivation data, and easy to miss unless you look at old sites.
Ask the patient
0 of 5 answeredPARACELSUS defines a progressive course as an ulcer developing or progressing within 6 weeks.
PARACELSUS extreme-pain criterion. This does not substitute for examination tenderness in Delphi minor criterion 5.
PARACELSUS localization criterion, including surgical sites, needle sticks, debridement, and stoma sites.
Delphi minor criterion 2. Record a lesion induced or worsened by trauma; the ulceration should extend beyond the area of trauma.
Delphi minor 4. Requires a specific history of the primary lesion and its speed of breakdown.
Records & exclusion workup
0 of 4 answeredDelphi minor 3 counts only these two. PARACELSUS counts any associated systemic disease.
Haematologic malignancy, myelodysplastic syndrome, monoclonal gammopathy especially IgA, solid tumour, hidradenitis suppurativa, autoinflammatory syndromes. Feeds PARACELSUS only.
Delphi minor 1. Bacterial, mycobacterial and fungal. Surface swabs do not count.
PARACELSUS major criterion. Vascular arterial and venous, vasculitis, calciphylaxis, malignancy, exogenous or factitial, infectious.
Tissue & time
0 of 4 answeredDelphi requires a neutrophilic infiltrate in a biopsy from the ulcer edge. This tool does not recommend whether to biopsy; use clinical judgement to balance evaluation of mimickers against pathergy risk.
PARACELSUS response criterion. Record separately from the more specific Delphi one-month ulcer-size criterion.
Delphi minor criterion 8. Cannot be known before treatment starts, so an unresolved answer is appropriate when treatment has not yet been observed.
PARACELSUS additional criterion. This is recorded separately from Delphi's ulcer-edge neutrophilic-infiltrate requirement.
PARACELSUS corridor
Delphi consensus
Unanswered information
Completeness list only. It does not estimate diagnostic yield, rank tests, or recommend biopsy or treatment.
Reading the two together
Still open. At least one framework remains indeterminate because scoring information is unanswered.
Ulcer-edge histology is unresolved. Delphi cannot be met without its major criterion. Biopsy can also help evaluate mimickers, but this calculator does not recommend whether or when to biopsy; balance the clinical need against pathergy risk.
Where these numbers come from+−
| Instrument & threshold | Reference standard population | Sens | Spec | AUC | PPV |
|---|---|---|---|---|---|
| PARACELSUS ≥ 10 | Jockenhöfer 2019 derivation, 60 lower-extremity PG vs 50 venous leg ulcers | 100% | 100% | n/r | n/r |
| PARACELSUS ≥ 10 | Moelleken 2026, 1403 ulcers from 14 centres in 7 countries, 180 PG across 85 wound entities | 100.0% | 93.2% | n/r | 68.4% |
| PARACELSUS > 10 (≥ 11) | Moelleken 2026, same cohort | 98.3% | 96.8% | n/r | 81.9% |
| PARACELSUS ≥ 10 | Ji-Xu 2023, ulcerative PG cases vs deliberately selected mimicker controls | 95.9% | 4.7% | 0.55 | n/r |
| PARACELSUS ≥ 10 | Haag 2021, expert-confirmed PG cohort only, no controls | 89% | n/a | n/r | n/r |
| Delphi, major + 4 of 8 minor | Maverakis 2018 derivation, 65 published ulcerative PG vs 48 mimickers, k-fold CV with imputation | 86% | 90% | n/r | n/r |
| Delphi, major + 4 of 8 minor | Ji-Xu 2023, cases vs mimickers, complete-case analysis | 88.0% | 90.5% | 0.89 | n/r |
| Delphi, major + 4 of 8 minor | Haag 2021, expert-confirmed PG cohort only | 74% | n/a | n/r | n/r |
| Su 2004, for context | Ji-Xu 2023 | 86.2% | 69.6% | 0.74 | n/r |
Performance depends on the comparator population. PARACELSUS specificity was 93.2 to 96.8 percent in the broad mixed-wound cohort studied by Moelleken and 4.7 percent in the selected mimicker controls studied by Ji-Xu. These estimates are not interchangeable and should not be applied as patient-specific probabilities.
Agreement is limited. Ji-Xu reported a Cohen kappa of 0.15 between Delphi and PARACELSUS. Haag found that all three frameworks agreed for 72 percent of an expert-confirmed PG cohort, with a Fleiss kappa of 0.44. A discordant result should prompt review of the exact item definitions and the differential diagnosis.
Histopathology has different roles in the frameworks. Delphi requires an ulcer-edge biopsy showing a neutrophilic infiltrate; PARACELSUS separately awards 1 point for suppurative inflammation. A negative or nonspecific biopsy should be interpreted in clinical context rather than assumed to be a false negative.
References
- Maverakis E, Ma C, Shinkai K, et al. Diagnostic criteria of ulcerative pyoderma gangrenosum: a Delphi consensus of international experts. JAMA Dermatol. 2018;154(4):461-466.
- Jockenhöfer F, Wollina U, Salva KA, Benson S, Dissemond J. The PARACELSUS score: a novel diagnostic tool for pyoderma gangrenosum. Br J Dermatol. 2019;180(3):615-620.
- Haag C, Hansen T, Hajar T, et al. Comparison of three diagnostic frameworks for pyoderma gangrenosum. J Invest Dermatol. 2021;141(1):59-63.
- Min MS, Kus K, Wei N, et al. Evaluating the role of histopathology in diagnosing pyoderma gangrenosum using Delphi and PARACELSUS criteria. Br J Dermatol. 2022;186(6):1035-1037.
- Ji-Xu A, Liakos W, Merleev A, Brüggen MC, Nelson CA. Assessing the discriminatory ability of diagnostic criteria for ulcerative pyoderma gangrenosum and its mimickers. JAMA Dermatol. 2023;159(3):337-338.
- Moelleken M, Ortega-Loayza AG, Busch D, et al. Validation of PARACELSUS score performance for the diagnosis of pyoderma gangrenosum: an international multicenter study with 1403 cases. J Am Acad Dermatol. 2026;94(6):1738-1746.