Lisette van Dam

Chapter 5 90 Current guidelines recommend CUS combined with doppler ultrasonography as the first line imaging test in patients with suspected UEDVT, due to its availability, relatively low cost and non-invasive nature. 3 Since the diagnosis of UEDVT can be difficult as deep axillary and retro-clavicular areas cannot be well visualized nor compressed due to overlying (bone) structures, both CUS and doppler ultrasonography can be used to confirm UEDVT in the presence of non- compressibility of a venous segment and/or in the absence of a color or doppler signal within the lumen of the vein or to exclude UEDVT in the absence of these findings. Moreover, follow-up imaging including repeat CUS combined with doppler, contrast-venography or CT venography is recommended in patients with high clinical suspicion but negative ultrasound. 3 Previously, MR venography (time-of-flight and contrast-enhanced) has been evaluated as alternative in the diagnostic management of UEDVT, but was not safe to exclude UEDVT (sensitivity of 71% (95%CI 29-96%) and 50% (95% CI 12-88%) and specificity of 89% (95% CI 52-100%) and 80% (95%CI 44-97%), respectively). 21 MR-NCTI has the advantage of direct thrombus visualization without the use of a contrast agent as the technique is based on the intrinsic contrast of fresh thrombus itself. 7,11-14,22-26 3D TSE-SPAIR has some advantages over MRDTI sequences, including a higher spatial resolution of the vessel wall and less inflow artefacts in the arteries. 14 The two techniques were found to be potentially feasible for the diagnosis of UEDVT which was confirmed in this study. 14 We found a sensitivity and specificity of MR-NCTI that are comparable to that of MRDTI in the diagnosis of recurrent ipsilateral DVT of the leg, and for which the safety to exclude recurrent ipsilateral DVT of the leg was confirmed in an outcome study. 9,11 Notably, MR-NCTI missed the diagnosis of UEDVT in 2 patients in our study. In one patient, the MR- NCTI was performed after 48 hours of anticoagulant therapy and the anatomy was particular complex with a hypoplastic jugular vein, which may have contributed to a false negative reading by the experts. In the other case, no straightforward explanation was identified. A limitation of the study is that 3D TSE-SPAIR sequence was not performed in all patients. Also, MR image quality of 8 patients was deemed insufficient to provide a definite diagnosis. Direct thrombus imaging seems more challenging in the upper arms and clavicular areas than in the lower extremities, because of the vascular orientation and image artefacts due to respiratory motion and cardiac- and vascular pulsation, limiting the image and contrast quality of the MRDTI scan. Therefore, we recommend using the combination of MRDTI and 3D TSE-