Abstract
The objective of this study was to evaluate the appearance and the natural evolution of desmoid tumors on MR imaging, given histologic correlation. The MR images of 30 desmoids (20 primary and 10 recurrent) in 26 patients were scored for a multiplicity of morphological parameters, signal intensity (SI) on different pulse sequences, and behavior after contrast administration. Natural evolution was evaluated in 2 primary and 3 recurrent lesions, and correlated with evolution on histologic specimens. Desmoid tumors are mostly found in muscles of shoulder and hip girdle and are often fusiform with partially ill-defined margins. Rare subcutaneous desmoids have a more stellar morphology. Variable amounts of low-SI areas are present on all sequences. On T1-weighted images (T1-WI), most lesions are near homogeneous and isointense to muscle, whereas on T2-WI they are more heterogeneous with an overall SI equal to or slightly lower than fat. Histologic correlation reveals that SI on T2-WI cannot be explained solely by cellularity. After initial growth, spontaneous evolution of desmoids is characterized by shrinking and an increase in low-SI areas on T2-WI. While distal lesions shrink, the more recent lesions in asynchronous multicentric desmoids have a tendency to develop proximally in the same limb, and should not be confused with recurrences. Fast growth, extracompartmental spread, and bone involvement are often seen in recurrences. Follow-up MR imaging of desmoids indicates natural regression of desmoids and more aggressive behavior of recurrences, which may justify a more conservative therapeutic approach.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Author information
Authors and Affiliations
Additional information
Received 30 July 1996; Revision received 26 November 1996; Accepted 16 January 1997
Rights and permissions
About this article
Cite this article
Vandevenne, J., De Schepper, A., De Beuckeleer, L. et al. New concepts in understanding evolution of desmoid tumors: MR imaging of 30 lesions. Eur Radiol 7, 1013–1019 (1997). https://doi.org/10.1007/s003300050243
Published:
Issue Date:
DOI: https://doi.org/10.1007/s003300050243