Compensated hepatitis C: unenhanced MR imaging correlated with pathologic grading and staging

Abdom Imaging. 2008 Jan-Feb;33(1):58-64. doi: 10.1007/s00261-007-9203-7.

Abstract

Background: We prospectively examined unenhanced MR imaging findings in relation to pathologic fibrosis, inflammation and steatosis in patients with compensated chronic hepatitis C viral infection (HCV).

Methods: Unenhanced MRI at 1.5 T was obtained within one month of core liver biopsy in 64 consecutive candidates for antiviral therapy for compensated HCV. Two pathologists independently graded inflammatory activity index (HAI) and steatosis, and staged fibrosis (grades 0-6). Morphologic MRI findings of cirrhosis, periportal lymph nodes, and MR fat signal ratio from dual gradient echo images were assessed independently by two radiologists blinded to clinical data. MRI and laboratory liver function results were correlated with pathologic results, using Spearman correlation coefficient and stepwise multiple regression.

Results: MR fat signal ratio correlation coefficient with pathologic steatosis was 0.71 (p < 0.0001). Coefficients with fibrosis stage were highest for surface nodularity (r (s) = 47, p < 0.0001) and expanded gallbladder fossa (r (s) = 0.42, p = 0.0006). Coefficients with HAI were highest for lymph node size (r (s) = 0.355, p = 0.0040), surface nodularity (r = 0.47, p < 0.0001), expanded gallbladder fossa (r = 0.332, p = 0.0073), and caudate/right lobe ratio (r = 0.326, p = 0.0110). Combined lab and MRI variables provided the best prediction of fibrosis stage (r (2) = 0.656) and HAI (r (2) = 0.597).

Conclusions: A combination of MRI and laboratory findings was most predictive of fibrosis and inflammation.

MeSH terms

  • Adult
  • Aged
  • Biopsy
  • Fatty Liver / pathology
  • Female
  • Fibrosis / pathology
  • Hepatitis C, Chronic / pathology*
  • Humans
  • Inflammation / pathology
  • Liver Function Tests
  • Magnetic Resonance Imaging / methods*
  • Male
  • Middle Aged
  • Neoplasm Staging
  • Prospective Studies
  • Regression Analysis