Proton NMR analysis of plasma is a weak predictor of coronary artery disease

Area of Science:

• Biochemistry
• Medical Diagnostics
• Cardiovascular Medicine

Background:

• Previous studies indicated high accuracy (>90%) of 1H-NMR spectra analysis in predicting advanced coronary artery disease (CAD) using blood sera.
• The prior analysis primarily focused on lipid regions, potentially overlooking confounders like gender and drug treatments that influence lipid composition.

Purpose of the Study:

• To evaluate the predictive capability of 1H-NMR spectroscopy for angiographically defined CAD in plasma samples.
• To assess the impact of statin treatment on the accuracy of 1H-NMR based CAD prediction.

Main Methods:

• Multivariate analysis of 1H-NMR spectra from plasma samples of male patients with normal coronary arteries (NCAs) or CAD.
• Patients were stratified based on statin treatment status (untreated vs. treated).

Main Results:

• Predictive accuracy for CAD was 80.3% in untreated patients and significantly lower at 61.3% in statin-treated patients.
• High confidence predictions (>99%) were achieved for only 36.2% of untreated cases versus 6.2% of treated cases.
• The diagnostic performance of 1H-NMR for CAD detection with >99% confidence was considerably weaker than that of coronary angiography.

Conclusions:

• 1H-NMR spectroscopy shows limited reliability for predicting CAD, particularly in patients undergoing statin therapy.
• Statin treatment significantly confounds the predictive accuracy of 1H-NMR analysis for coronary artery disease.
• Coronary angiography remains the gold standard for definitive CAD diagnosis due to the low confidence levels observed with 1H-NMR.