A metabolomic profile is associated with the risk of incident coronary heart disease

Am Heart J. 2014 Jul;168(1):45-52.e7. doi: 10.1016/j.ahj.2014.01.019. Epub 2014 Apr 4.


Background: Metabolomics, defined as the comprehensive identification and quantification of low-molecular-weight metabolites to be found in a biological sample, has been put forward as a potential tool for classifying individuals according to their risk of coronary heart disease (CHD). Here, we investigated whether a single-point blood measurement of the metabolome is associated with and predictive for the risk of CHD.

Methods and results: We obtained proton nuclear magnetic resonance spectra in 79 cases who developed CHD during follow-up (median 8.1 years) and in 565 randomly selected individuals. In these spectra, 100 signals representing 36 metabolites were identified. Applying least absolute shrinkage and selection operator regression, we defined a weighted metabolite score consisting of 13 proton nuclear magnetic resonance signals that optimally predicted CHD. This metabolite score, including signals representing a lipid fraction, glucose, valine, ornithine, glutamate, creatinine, glycoproteins, citrate, and 1.5-anhydrosorbitol, was associated with the incidence of CHD independent of traditional risk factors (TRFs) (hazard ratio 1.50, 95% CI 1.12-2.01). Predictive performance of this metabolite score on its own was moderate (C-index 0.75, 95% CI 0.70-0.80), but after adding age and sex, the C-index was only modestly lower than that of TRFs (C-index 0.81, 95% CI 0.77-0.85 and C-index 0.82, 95% CI 0.78-0.87, respectively). The metabolite score was also associated with prevalent CHD independent of TRFs (odds ratio 1.59, 95% CI 1.19-2.13).

Conclusion: A metabolite score derived from a single-point metabolome measurement is associated with CHD, and metabolomics may be a promising tool for refining and improving the prediction of CHD.

MeSH terms

  • Adult
  • Coronary Disease / blood*
  • Coronary Disease / epidemiology*
  • Female
  • Follow-Up Studies
  • Humans
  • Incidence
  • Lipids / blood*
  • Magnetic Resonance Spectroscopy / methods*
  • Male
  • Metabolomics / methods*
  • Middle Aged
  • Netherlands / epidemiology
  • Odds Ratio
  • Predictive Value of Tests
  • Prognosis
  • Prospective Studies
  • Risk Factors
  • Time Factors
  • Young Adult


  • Lipids