Magnetic resonance venography with a blood pool contrast medium

Introduction Non-invasive venography to cover the entire neck and thoracic central venous system in one examination is challenging. MR venography is ideal for coverage but the technique is difficult as the majority of contrast media are extracellular agents that rapidly leave the vascular pool. Time resolved serial acquisition methods have improved but image quality may be suboptimal and they are effectively limited to one body region . This could be overcome using a blood pool agent such as gadofosveset trisodium (Vasovist® , Schering) which can be imaged using both a conventional first pass (FP) technique and then repeatedly in the “steady state” after the redistribution of the agent throughout the vascular system[1,2,3]. The aim of this work was to evaluate the diagnostic performance of the steady-state (SS) images against the conventional first pass time resolved technique for MR venography of the central and upper throracic major veins. Methods A prospective ethically approved and Clinical Trial Authorised (2007-002730-11) open-label feasibility study was undertaken from August 2008. Expected final sample size is 30 patients. Patients over 18 years old who have been routinely referred for MRV of central veins were recruited. Informed consent was obtained from all participants. Intravenous Vasovist® 0.12ml/kg body weight (maximum 10mls) is given at 0.8ml/s followed by a saline flush of 20ml at 2ml/s. FP and SS imaging were performed using a 1.5T GE MRI system with an 8 channel cardiac receive array. FP parameters: FOV 40x40cm; slices 42x2.6mm; matrix 418x256x0.75NEX; flip angle 30°;ASSET factor 2; temporal resolution 10sec, 30 phases acquired. SS parameters: FOV 40x40cm; slices 64 x1.6mm;matrix 512x512x2.0 NEX interpolated to 1024 x1024; flip angle 30°; acquisition time 4.25min. Each set of images were assessed independently in randomised fashion. FP images were used as the reference standard for correct diagnosis. Nine venous segments were assessed – superior vena cava (SVC), left and right branches of brachiocephalic, subclavian, internal jugular, and axillary veins. Four parameters were evaluated: 1. image quality in terms of vessel conspicuity using a five-point scale (excellent, good, moderate, poor, very poor); 2. presence of artefacts using a three-point scale (none, mild, major); 3. presence of stenosis using a six-point scale (no: 0%, mild: 1-30%, mild to moderate: 31-50%, moderate: 51-75%, severe: 76-99%, total occlusion); 4. presence of thrombosis using a three-point scale (no, partial, complete). Images were scored by consensus among 3 experienced consultant radiologists.