Lisette van Dam

Detection of UEDVT by MR-NCTI 5 91 SPAIR sequences in all patients when applied in the diagnostic management of suspected UEDVT. A drawback of such practice is the longer image acquisition time (30 minutes) compared to that of performing only one of the sequences (MRDTI) in the lower extremities (10 minutes). Moreover, it is important that experience in performing and image reading of these techniques is gained before it can be used for the diagnosis of UEDVT. Strengths of this study include its prospective design. The MR-NCTI scans were performed in different centers across different countries and using MR scanners of different manufactures. This, together with the excellent interobserver agreement, supports the wide applicability of this technique. We were able to prove its accuracy for the diagnosis of UEDVT in an adequate patient sample and also included the subgroup of patients with inconclusive ultrasound, for which the use of MR-NCTI may be particularly relevant. As MRI is associated with higher costs compared to ultrasonography, it should not be used as first line imaging test. Instead, we suggest that MR-NCTI could potentially serve as a second-line imaging test in patients with high clinical suspicion for UEDVT but inconclusive ultrasound. Since it was not the aim of the current study to assess the accuracy of MR-NCTI in this particular setting, future studies to the safety of this technique to exclude UEDVT in case of an inconclusive ultrasound are needed. In conclusion, MR-NCTI was accurate for the diagnosis of UEDVT and had an excellent interobserver agreement. Future studies should determine whether this technique can replace venography as the second-line imaging test in patients with an inconclusive CUS.