Deep venous thrombosis: Gradient-recalled-echo MR imaging changes over time - Experience in 10 patients

Charles E. Spritzer, Priscilla Trotter, Henry Dirk Sostman

Research output: Contribution to journalArticlepeer-review

26 Scopus citations


PURPOSE: To determine the gradient-recalled-echo (GRE) magnetic resonance (MR imaging changes caused by intraluminal thrombus of the lower extremities over time and to establish GRE MR imaging criteria to help distinguish acute from previous deep venous thrombosis. MATERIALS AND METHODS: Four women and six men, aged 38-76 years, underwent GRE MR imaging two to four times after the original diagnosis of deep venous thrombosis; eight also underwent confirmatory ultrasonography at that time. Follow-up was 3 months to 1 1/4 years. All patients were treated with anticoagulants. Evidence of thrombus, signal intensity characteristics, and MR imaging appearance were evaluated. RESULTS: Acute thrombosis was identified as an occlusion or prominent filling defect within the vessel. The luminal diameter of thrombosed veins was equal to or larger than that of a corresponding, unaffected vessel in each case. Acute thrombus signal intensity tended to be decreased initially and increased over time. Residual changes due to thrombosis included the web formation (n = 3), luminal narrowing (n = 4), and wall thickening and/or slow flow (n = 5). Vessels in three patients reverted to normal. CONCLUSION: GRE MR imaging demonstrated progressive changes of venous thrombus over time. These criteria may help distinguish acute deep venous thrombosis from the residual changes of previous thrombosis.

Original languageEnglish (US)
Pages (from-to)631-639
Number of pages9
Issue number3
StatePublished - Jan 1 1998


  • Thrombosis, MR
  • Veins, MR
  • Veins, thrombosis

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology


Dive into the research topics of 'Deep venous thrombosis: Gradient-recalled-echo MR imaging changes over time - Experience in 10 patients'. Together they form a unique fingerprint.

Cite this