Gastrointestinal leiomyoma and leiomyosarcoma: CT differentiation

Research output: Contribution to journalArticlepeer-review

37 Scopus citations

Abstract

Purpose: Our goal was to evaluate CT efficacy in differentiating gastrointestinal leiomyoma and leiomyosarcoma. Method: We retrospectively reviewed CT scans of 45 patients (21 men, 24 women, mean age 55 years) with surgically proven gastrointestinal leiomyomas (n = 21) and leiomyosarcomas (n = 24) with respect to size, contour, enhancing pattern, mesenteric fat infiltration, calcification, ulceration, regional lymphadenopathy, direct invasion, distant metastasis, and growth pattern after visual inspection by two radiologists in agreement. On the basis of these CT features, subjective diagnosis was also categorized into three groups (Group I: probably benign, Group II: probably malignant, Group III: diagnostic malignant). The results were compared with the final histopathologic diagnosis. Results: In addition to the features of direct invasion and distant metastasis suggesting diagnostic malignancy, the CT features favoring malignancy with statistical significance included larger size (>5 cm), lobulated contour, heterogeneous enhancement, mesenteric fat infiltration, ulceration, regional lymphadenopathy, and exophytic growth pattern (p < 0.005). However, calcification was not significant in differentiating the two entities (p = 0.25163). A subjective analysis revealed 89% sensitivity, 85% specificity, and 87% accuracy for diagnosis of leiomyosarcoma. Conclusion: CT features are useful in differentiating leiomyoma and leiomyosarcoma in gastrointestinal tract.

Original languageEnglish
Pages (from-to)69-74
Number of pages6
JournalJournal of Computer Assisted Tomography
Volume22
Issue number1
DOIs
StatePublished - 1998

Keywords

  • Computed tomography
  • Gastrointestinal tract, diseases
  • Gastrointestinal tract, neoplasms
  • Leiomyoma
  • Leiomyosarcoma

Fingerprint

Dive into the research topics of 'Gastrointestinal leiomyoma and leiomyosarcoma: CT differentiation'. Together they form a unique fingerprint.

Cite this