Gastrointestinal stromal tumor. Correlation of computed tomography findings with tumor grade and mortality

J Comput Assist Tomogr. 2003 Sep-Oct;27(5):792-8. doi: 10.1097/00004728-200309000-00018.

Abstract

Objectives: To assess computed tomography (CT) findings to assist in the distinction of low- and high-grade gastrointestinal stromal tumors (GISTs) and to estimate their relative risk (RR) for mortality.

Methods: Sixty-nine patients with clinicopathologically and immunohistochemically proven GISTs, including 44 patients with low-grade tumors and 25 with high-grade tumors, who underwent dual-phase CT for initial examination were included. Images were assessed retrospectively for tumor size, location, epicenter, types of surface, boundary, presence of invasion, enhancement pattern, hepatic metastasis, and peritoneal dissemination. Their RR for mortality was estimated by using a multiple logistic regression model.

Results: Statistically significant CT findings favoring a diagnosis of high-grade GIST and affecting the 5-year survival rate included a lesion larger than 11.1 cm (median + 1 SD), irregular surface, unclear boundary, presence of invasion, heterogeneous enhancement, hepatic metastasis, and peritoneal dissemination. Multivariate analysis showed RRs for mortality in lesions larger than 11.1 cm (RR=3.9), with the presence of wall invasion (RR=5.1), and with hepatic metastasis (RR=11.3), respectively.

Conclusions: The CT features that suggest a high-grade GIST and predict poor outcome include hepatic metastasis, presence of wall invasion, and lesions larger than 11.1 cm.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Digestive System / pathology
  • Female
  • Gastrointestinal Neoplasms / diagnostic imaging*
  • Gastrointestinal Neoplasms / mortality*
  • Gastrointestinal Neoplasms / pathology
  • Humans
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Prognosis
  • Risk Assessment
  • Stromal Cells / pathology
  • Survival Rate
  • Tomography, X-Ray Computed* / methods