Short- and Long-Term Reproducibility of Intrahepatic Lipid Quantification by 1H-MR Spectroscopy and CT in Obesity

J Comput Assist Tomogr. 2016 Sep-Oct;40(5):678-82. doi: 10.1097/RCT.0000000000000423.

Abstract

Objective: This study aimed to assess short- and long-term reproducibility of intrahepatic lipid (IHL) quantification by proton magnetic resonance spectroscopy (H-MRS) and computed tomography (CT).

Methods: Sixteen obese subjects underwent H-MRS using a single-voxel point-resolved single-voxel spectroscopy sequence at 3 T and noncontrast single-slice CT of the liver. Measurements were repeated after 6 weeks and 6 months. Clinical parameters (weight, activity, serum lipids) were collected. Short-term (baseline to 6 weeks) and long-term (baseline to 6 months) reproducibility of IHL was assessed by coefficient of variance (CV), SD, and intraclass correlation coefficient (ICC).

Results: Short-term reproducibility and long-term reproducibility of H-MRS were as follows: CV, 5.9% to 18.8%; SD, 0.7 to 1.9; and ICC, 0.998 to 0.995 (95% confidence interval, 0.942-0.999). Short-term reproducibility and long-term reproducibility of CT were as follows: CV, 4.4% to 14.2%; SD, 2.4 to 8.7; and ICC, 0.766 to 0.982 (95% confidence interval, 0.271-0.994). There was no significant change in clinical parameters (P > 0.3).

Conclusions: Proton magnetic resonance spectroscopy and CT are reproducible methods for short- and long-term quantification of IHL content. Our results can guide sample size calculations for interventional and longitudinal studies.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Biomarkers / metabolism
  • Female
  • Humans
  • Lipid Metabolism*
  • Liver / diagnostic imaging*
  • Liver / metabolism*
  • Longitudinal Studies
  • Male
  • Middle Aged
  • Molecular Imaging / methods
  • Obesity / diagnostic imaging*
  • Obesity / metabolism*
  • Proton Magnetic Resonance Spectroscopy / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity

Substances

  • Biomarkers