Imaging of the intrahepatic portal vein in children with extrahepatic portal vein thrombosis — Comparison of magnetic resonance imaging and retrograde portography

Abstract

PurposeExtrahepatic portal vein thrombosis (EPVT) is one major cause of portal hypertension in children. Surgical reinstallation of portal venous flow can be achieved in patients with patent intrahepatic portal venous system/Rex recess. Our study aimed to compare the ability of magnetic resonance imaging (MRI) and retrograde portography (RP) to assess patency of the intrahepatic portal venous system in children with EPVT. MethodsAll pediatric patients with EPVT who were examined with contrast enhanced MRI (1.5 T) and invasive RP between 2013 and 2017 were included in this retrospective study. Medical records were reviewed for demographic, biochemical and clinical data. Patency of the Rex recess as detected by MRI and RP was retrospectively reviewed. ResultsSixteen children (7.6 ± 5.0 years) with EPVT were included. Sensitivity, specificity, positive and negative predictive value for the detection of patent Rex recess by MRI compared to RP were 55%, 57%, 63% and 50%. Diagnostic accuracy was 56%. Diagnostic failure of MRI compared to RP was explained by the following: I. Problems differentiating collaterals from portal venous vessels II. Incapability showing dynamic blood flow in compromised portal venous flow III. Poor spatial resolution, especially in small children. ConclusionRP is a reliable method for the visualization of the Rex recess and the intrahepatic portal venous system in children with EPVT, whereas MRI has shown to be unsuitable for the assessment of the intrahepatic portal vein in these patients. In the preoperative setup, we recommend both procedures, RP and MRI for the visualization of the intrahepatic portal venous system, and the extrahepatic vessels, respectively. Level of evidenceLevel III.