Superior Diagnostic Performance in a Defined Multi-Marker Panel for Acute Myocardial Infarction (AMI)
In a study of 795 differentially expressed lipid metabolites, 3-hydroxyoleylcarnitine (CarnitineC18:1-OH) was selected by machine learning algorithms (LASSO and SVM-RFE) as one of only three lipid molecules to form a diagnostic panel for acute myocardial infarction (AMI). This panel demonstrated a high predictive ability in discriminating AMI patients from those with unstable angina (UA), achieving an area under the curve (AUC) of 0.9 in both discovery and validation sets [1]. This performance establishes the compound as a high-value component of a minimal, high-accuracy biomarker panel, a specific context where other acylcarnitines, which were not selected from the 25 differentially expressed candidates, cannot substitute.
| Evidence Dimension | Diagnostic Accuracy (AUC) for AMI discrimination |
|---|---|
| Target Compound Data | AUC = 0.9 (as part of a 3-lipid panel including CarnitineC18:1-OH, CarnitineC18:2-OH, and FFA (20:1)) |
| Comparator Or Baseline | 25 other differentially expressed lipid metabolites from UPLC-MS/MS analysis that were not selected by LASSO/SVM-RFE for the final model. |
| Quantified Difference | The three-metabolite model containing CarnitineC18:1-OH achieved a final AUC of 0.9, outperforming models based on other candidate metabolites. |
| Conditions | UPLC-MS/MS quantitative lipidomics; human plasma samples from AMI, unstable angina (UA), and healthy control (HC) groups; discovery (n=?) and validation (n=?) sets. |
Why This Matters
For a procurement scientist, this justifies selecting this specific compound for developing or validating a high-accuracy AMI diagnostic assay, as its performance within a specific panel is empirically quantified rather than assumed.
- [1] Liu, J., et al. (2022). Plasma Quantitative Lipid Profiles: Identification of CarnitineC18:1-OH, CarnitineC18:2-OH and FFA (20:1) as Novel Biomarkers for Pre-warning and Prognosis in Acute Myocardial Infarction. Frontiers in Cardiovascular Medicine, 9, 848840. View Source
