Proposal of a histopathological predictive rule for the differential diagnosis between American tegumentary leishmaniasis and sporotrichosis skin lesions

Br J Dermatol. 2012 Oct;167(4):837-46. doi: 10.1111/j.1365-2133.2012.11012.x. Epub 2012 Sep 5.

Abstract

Background: American tegumentary leishmaniasis (ATL) and sporotrichosis exhibit similar histopathology and low frequencies of microorganism detection.

Objectives: This study seeks to identify microscopic alterations that can distinguish between these diseases.

Methods: Haematoxylin and eosin stained slides of 171 ATL and 97 sporotrichosis samples from active cutaneous lesions were examined for histopathological alterations. The lesions were diagnosed by isolating the agent (which was not visible) in culture. An intuitive diagnosis was assigned to each slide. The strength of the association between the histopathological findings and the diagnosis was estimated by an odds ratio, and each finding was graded according to a regression model. A score was assigned to each sample based on the histopathological findings. A study of the interobserver reliability was performed by calculating kappa coefficients of the histopathological findings and intuitive diagnoses.

Results: The markers 'macrophage concentration', 'tuberculoid granuloma' and 'extracellular matrix degeneration' were associated with ATL. 'Suppurative granuloma', 'stellate granuloma', 'different types of giant cells', 'granulomas in granulation tissue' and 'abscess outside the granuloma' were associated with a diagnosis of sporotrichosis. 'Macrophage concentration' and 'suppurative granuloma' had the highest (substantial and almost perfect, respectively) reliability. The regression model score indicated 92.0% accuracy. The intuitive diagnosis had 82.5% diagnostic accuracy and substantial reliability.

Conclusions: Taking into account the clinical and epidemiological context, some histopathological alterations might be useful for the differential diagnosis between ATL and sporotrichosis cutaneous lesions in cases in which the aetiological agent is not visible.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Brazil
  • Cross-Sectional Studies
  • Dermatomycoses / diagnosis*
  • Diagnosis, Differential
  • Granuloma / pathology
  • Humans
  • Leishmaniasis, Cutaneous / diagnosis*
  • Logistic Models
  • Macrophages / pathology
  • Predictive Value of Tests
  • Reproducibility of Results
  • Skin Diseases, Parasitic / diagnosis*
  • Sporotrichosis / diagnosis*
  • Staining and Labeling