Chronic liver disease: systemic and splanchnic venous flow mapping with optimized cine phase-contrast MR imaging validated in a phantom model and prospectively evaluated in patients.

Abstract

To validate magnetic resonance (MR) imaging cine phase-contrast blood flow mapping in vitro and in patients with chronic liver disease, with or without portal hypertension, and to assess the accuracy of azygos, splanchnic, and systemic blood flow measured with MR imaging in the detection of high-risk esophageal varices and compare these measurements with endoscopic evaluation, the reference standard. The local ethics committee approved this study. Patients gave written informed consent. Two phantoms were used to validate the MR imaging phase-contrast flow sequence. Patients with liver cirrhosis (n = 59), chronic liver disease without cirrhosis (n = 12), and nodular regenerative hyperplasia (NRH) (n = 11), and healthy control subjects with no liver disease (n = 25) were included. The patients underwent upper digestive system endoscopy. Mean abdominal aorta, portal venous, and azygos blood flow was measured on MR images, which were reviewed by two blinded observers to determine the presence and grade of esophageal varices. The reproducibility and intra- and interobserver variability of the blood flow measurements were assessed with intraclass correlation coefficients (ICCs). The performance of the MR blood flow measurements in staging high-risk varices was determined with receiver operating characteristic curve analysis. The correlation between MR visual analysis and endoscopic grading was assessed by using κ statistics. MR flow rate measurements had excellent correlations with actual flow values in vitro (ICC > 0.990 for phantoms 1 and 2). Mean aortic flow was significantly higher in patients with cirrhosis than in control subjects (P < .001). Mean azygos flow was significantly higher in patients with cirrhosis than in patients with chronic liver disease without cirrhosis (P = .005) and control subjects (P < .001). Low intra- and interobserver variability (ICC > 0.990 for each blood flow type) and high reproducibility (ICC > 0.850 for each blood flow type) were demonstrated. The optimal cutoff mean azygos flow value was 2.3 mL/sec for varices with grades of 2 or higher. MR imaging azygos flow measurement appears to be a promising technique for detecting high-risk esophageal varices in patients with portal hypertension.