Magnetic resonance venography in consecutive patients with suspected deep vein thrombosis of the upper extremity: Initial experience

Purpose: To assess the feasibility and accuracy of two magnetic resonance (MR) venography methods in a consecutive series of patients with suspected deep vein thrombosis of the upper extremity (DVTUE). Material and Methods: Consecutive in- and outpatients who were referred for imaging of suspected DVTUE in a large teaching hospital during the period April 2001 to October 2002 were eligible for inclusion. All patients were scheduled to undergo contrast venography with the intention to perform additional MR venography. Both time-of-flight and gadolinium-enhanced 3D MR venography were scheduled. All MR imaging were interpreted independently by consensus of two experienced radiologists, who were blinded for contrast venography outcome. Patients were managed based on contrast venography only. Results: A total of 44 patients were eligible for inclusion. Thirteen patients were excluded (5 refused consent, 2 inability to gain venous access, 2 renal failure, 4 logistic reasons). Contrast venography was performed in 31 patients, and demonstrated DVTUE in 11 patients. MR imaging was not feasible in 10 patients (4 unable to lie flat, 3 claustrophobia, 1 too large for MR scanner, 1 osteosynthesis of shoulder, 1 pacemaker). The sensitivity and specificity of TOF MRV versus Gadolinium 3D MRV was 71% and 89% versus 50% and 80%, respectively. Conclusion: A high number of patients were unable to undergo MR venography in this setting. Contrast-enhanced MRV did not improve diagnostic accuracy. The clinical utility of MR venography in the setting of suspected DVTUE seems disappointing.