Although a wide range of risk factors for coronary heart disease have been identified from population studies, these measures, singly or in combination, are insufficiently powerful to provide a reliable, noninvasive diagnosis of the presence of coronary heart disease. Here we show that pattern-recognition techniques applied to proton nuclear magnetic resonance (1H-NMR) spectra of human serum can correctly diagnose not only the presence, but also the severity, of coronary heart disease. Application of supervised partial least squares-discriminant analysis to orthogonal signal-corrected data sets allows >90% of subjects with stenosis of all three major coronary vessels to be distinguished from subjects with angiographically normal coronary arteries, with a specificity of >90%. Our studies show for the first time a technique capable of providing an accurate, noninvasive and rapid diagnosis of coronary heart disease that can be used clinically, either in population screening or to allow effective targeting of treatments such as statins.

Although a wide range of risk factors for coronary heart disease have been identified from population studies, these measures, singly or in combination, are insufficiently powerful to provide a reliable, noninvasive diagnosis of the presence of coronary heart disease. Here we show that pattern-recognition techniques applied to proton nuclear magnetic resonance (1H-NMR) spectra of human serum can correctly diagnose not only the presence, but also the severity, of coronary heart disease. Application of supervised partial least squares-discriminant analysis to orthogonal signal-corrected data sets allows >90% of subjects with stenosis of all three major coronary vessels to be distinguished from subjects with angiographically normal coronary arteries, with a specificity of >90%. Our studies show for the first time a technique capable of providing an accurate, noninvasive and rapid diagnosis of coronary heart disease that can be used clinically, either in population screening or to allow effective targeting of treatments such as statins.

The application of chemometric methods to 1H NMR spectroscopic data has been documented for pathophysiological processes. In this study we show the application of 1H NMR-based metabonomics to investigate a relationship between serum metabolic profiles and hypertension. Although hypertension can be defined using blood pressure measurements, the underlying aetiology and metabolic effects are not so readily identified. Serum profiles for patients with low/normal systolic blood pressure (SBP < or = 130 mm Hg; n = 28), borderline SBP (131-149 mm Hg; n = 19) and high SBP (> or = 150 mm Hg; n = 17) were acquired using 1H NMR spectroscopy. Orthogonal signal correction followed by principal components analysis were applied to these NMR data in order to facilitate interpretation, and the resulting chemometric models were validated using Soft Independent Modelling of Class Analogy. Using 1H NMR-based metabonomics, it was possible to distinguish low/ normal SBP serum samples from borderline and high SBP samples. Borderline and high SBP samples, however, were indiscriminate from each other. Our preliminary results showed that there was a relationship between serum metabolic profiles and blood pressure which, in part, was due to lipoprotein particle composition differences between the samples. Furthermore, our results indicated that serum pathology associated with blood pressure is apparent at SBP values > 130 mm Hg, which the WHO and ISH currently define as the limit between normal and high-normal.

1H nuclear magnetic resonance (NMR)-based metabonomics is a well-established technique used to analyse and interpret complex multiparametric metabolic data, and has a wide number of applications in the development of pharmaceuticals. However, interpretation of biological data can be confounded by extraneous variation in the data such as fluctuations in either experimental conditions or in physiological status. Here we have shown the novel application of a data filtering method, orthogonal signal correction (OSC), to biofluid NMR data to minimise the influence of inter- and intra-spectrometer variation during data acquisition, and also to minimise innate physiological variation. The removal of orthogonal variation exposed features of interest in the NMR data and facilitated interpretation of the derived multivariate models. Furthermore, analysis of the orthogonal variation provided an explanation of the systematic analytical/biological changes responsible for confounding the original NMR data.

Toll-like receptor-4 (TLR4) signaling augments chemokine-induced neutrophil migration by modulating cell surface expression of chemokine receptors Jie Fan & Asrar B. Malik Nat. Med. 9, 315-321 (2003). Fig. 1d: the 'MIP-2' label over lane 4 should be over lane 5. The correct figure is below. We regret the errors.