Lisette van Dam

Chapter 6 98 computed tomography (CT) venography or MRI and 35 patients with confirmed non-symptomatic chronic PVT (partial or occlusive), i.e. unchanged chronic thrombi on 2 serial imaging tests with an at least 3-month interval. The study protocol and its amendments were approved by the institutional review board of the Leiden University Medical Center, Leiden, the Netherlands, and the University of Insubria, Varese, Italy. All patients will be asked for written informed consent. For this first phase of the Rhea study, we aimed to select and develop MR-NCTI sequences that would be accurate for differentiation of acute from chronic PVT. First, a literature search for different MRI sequences for portal vein and acute thrombosis imaging was performed. The literature search was conducted in PubMed for papers published in English and in humans on December 16 th 2019 (search strategy detailed in Appendix A ). Because imaging of the abdomen is hampered by intestinal movements and gas, search terms included techniques with correction for respiratory motion artefacts, good spatial and contrast resolution and fat suppression. The most promising sequences were adjusted for abdominal imaging and tested in a healthy volunteer using a MRI 3.0 Tesla unit (Philips Ingenia, Philips Medical Systems, Best, TheNetherlands). A55cmreceive-onlybodymulti coil (combinationof posterior and anterior coil) was used. Image assessment involved acquiring images in the coronal and axial plane with standard image reconstruction techniques. To gain knowledge of the performance of the sequence to detect fresh blood clots in addition to the visualisation of the venous anatomy, two tests coagulation tubes were attached to the abdomen of the volunteer; one was filled with water (control) and one with clotted blood of a healthy volunteer, prepared and stored at room temperature at least 48 hours before each scanning session. The test sequence scan parameters were adjusted until adequate image quality of the veins with high signal intensity for thrombus in the coagulation tube was achieved. MR-NCTI sequences showing the best image quality and contrast resolution were tested in three patients with confirmed PVT, including one patient with acute PVT and two patients with chronic PVT. Scan optimization was performed until a clear distinct signal intensity was achieved between acute and chronic PVT. Finally, the MRI images of the three PVT patients were evaluated by an expert panel, consisting of two radiologists with over 20 years of experience with vascular MRI (LK and MW), one radiology technician with over 20 years of experience with vascular MRI acquisition (GH), one internist with 6 years (FK) and one researcher with 2 years (LD) of experience with vascular MRI interpretation. The evaluation