Metabolomic approach by 1H NMR spectroscopy of serum for the assessment of chronic liver failure in patients with cirrhosis

J Proteome Res. 2011 Jul 1;10(7):3239-45. doi: 10.1021/pr200265z. Epub 2011 Jun 2.

Abstract

Assessment of chronic liver failure (CLF) in cirrhotic patients is needed to make therapeutic decisions. A biological score is usually performed, using the Model for End-Stage Liver Disease (MELD), to evaluate CLF. Nevertheless, MELD does not take into account metabolic perturbations produced by liver-function impairment. In contrast, metabolomics can investigate many metabolic perturbations within biological systems. The purpose of this study was to assess whether metabolomic profiles of serum, obtained by proton NMR spectroscopy from cirrhotic patients, are affected by the severity of CLF. An orthogonal projection to latent-structure analysis was performed to compare MELD scores and NMR spectra of 124 patients with cirrhosis. The statistical model obtained showed a good explained variance (R(2)X = 0.87 and R(2)Y = 0.86) and a good predictability (Q(2)Y = 0.64). Metabolomic profiles showed significant differences regarding various metabolites depending of severity of CLF: levels of high-density lipoprotein and phosphocholine resonances were significantly higher in patients with mild CLF compared to severe CLF. Other metabolites such as lactate, pyruvate, glucose, amino acids, and creatinine were significantly higher in patients with severe CLF than mild CLF. Our conclusion is that metabolomic NMR analysis provides new insights into metabolic processes related to the severity of hepatic function impairment in cirrhosis.

MeSH terms

  • End Stage Liver Disease / blood*
  • End Stage Liver Disease / etiology
  • End Stage Liver Disease / pathology
  • Female
  • Humans
  • Liver / metabolism*
  • Liver / pathology
  • Liver Cirrhosis / blood*
  • Liver Cirrhosis / complications
  • Liver Cirrhosis / pathology
  • Liver Function Tests
  • Magnetic Resonance Spectroscopy / methods*
  • Magnetic Resonance Spectroscopy / statistics & numerical data
  • Male
  • Metabolomics / methods*
  • Middle Aged
  • Models, Statistical
  • Prognosis
  • ROC Curve
  • Research Design*
  • Serum / chemistry*
  • Severity of Illness Index*