Validity of the histopathological criteria used for diagnosing dysplastic naevi. An interobserver study by the pathology subgroup of the EORTC Malignant Melanoma Cooperative Group

Eur J Cancer. 1993;29A(6):831-9. doi: 10.1016/s0959-8049(05)80419-8.


Ten (dermato)pathologists studied 50 cutaneous melanocytic lesions including common naevocellular naevi, dysplastic naevi (DN), melanomas in situ and invasive primary melanomas, with emphasis on the histological criteria of DN. Using a standardised form, 20 defined histopathological features were scored (semi)quantitatively. Concordance of diagnosis, efficacy and reproducibility of features were investigated. DN were distinguished well from the other entities (mean Po 0.87). Agreement on the degree of atypia of DN was low. The reproducibility of the scoring was best for the following features: irregular nests, lymphohistiocytic infiltrate, marked junctional proliferation and large nuclei. The overall values of these features to discriminate between DN and non-DN were better than for the other features studied. Using the presence of at least three of the four features as a condition for the diagnosis of DN, values for sensitivity, specificity and positive and negative predictive values were 0.86, 0.91, 0.96 and 0.73, respectively. On the basis of the results these features seem best suited as histological criteria for the diagnosis of DN.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Diagnosis, Differential
  • Dysplastic Nevus Syndrome / diagnosis
  • Dysplastic Nevus Syndrome / pathology*
  • Humans
  • Observer Variation
  • Reproducibility of Results
  • Skin Neoplasms / diagnosis
  • Skin Neoplasms / pathology*