Cross-sectional imaging patterns of desmoplastic fibroma

Eur Radiol. 2001;11(7):1105-10. doi: 10.1007/s003300000739.

Abstract

The aim of this study was to work out the cross-sectional imaging characteristics of desmoplastic fibroma (DF). In 3 patients with histologically proven DF, the imaging characteristics obtained with cross-sectional techniques were reviewed retrospectively. Radiographs and CT scans were available in all patients, and plain and contrast-enhanced MR examinations in 2 patients. Compared with conventional radiographs, CT allowed more accurate assessment of the extent of bone destruction including cortical breakthrough and articular invasion. Intramedullary tumor growth and soft tissue extension was best detected with MRI. Apart from heterogeneity on MR images, DF displayed nonspecific low signal intensity on unenhanced T1-weighted images and an intermediate to high signal intensity including areas of low intensity on T2-weighted images. Desmoplastic fibroma showed a distinct, inhomogeneous gadolinium enhancement. Although cross-sectional imaging features of DF are nonspecific, some MR characteristics, such as inhomogeneous contrast enhancement and the presence of low-intensity regions on T2-weighted images, are helpful in determining the differential diagnosis. Cross-sectional imaging of DF is useful for local staging of the tumor because it provides valuable information about the extent of bone destruction as well as medullary and extraosseous spread.

Publication types

  • Case Reports
  • Comparative Study

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Bone Neoplasms / diagnostic imaging*
  • Bone Neoplasms / pathology*
  • Bone and Bones / diagnostic imaging
  • Bone and Bones / pathology
  • Cross-Sectional Studies
  • Female
  • Fibroma, Desmoplastic / diagnostic imaging*
  • Fibroma, Desmoplastic / pathology*
  • Humans
  • Magnetic Resonance Imaging*
  • Male
  • Middle Aged
  • Neoplasm Invasiveness
  • Neoplasm Recurrence, Local
  • Retrospective Studies
  • Tomography, X-Ray Computed*