Refine
Year of publication
- 2018 (2) (remove)
Document Type
- Article (2)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2) (remove)
Keywords
- Diagnostic markers (2) (remove)
Institute
The use of cardiac troponins (cTn) is the gold standard for diagnosing myocardial infarction. Independent of myocardial infarction (MI), however, sex, age and kidney function affect cTn levels. Here we developed a method to adjust cTnI levels for age, sex, and renal function, maintaining a unified cut-off value such as the 99th percentile. A total of 4587 individuals enrolled in a prospective longitudinal study were used to develop a model for adjustment of cTn. cTnI levels correlated with age and estimated glomerular filtration rate (eGFR) in males/females with rage = 0.436/0.518 and with reGFR = −0.142/−0.207. For adjustment, these variables served as covariates in a linear regression model with cTnI as dependent variable. This adjustment model was then applied to a real-world cohort of 1789 patients with suspected acute MI (AMI) (N = 407). Adjusting cTnI showed no relevant loss of diagnostic information, as evidenced by comparable areas under the receiver operator characteristic curves, to identify AMI in males and females for adjusted and unadjusted cTnI. In specific patients groups such as in elderly females, adjusting cTnI improved specificity for AMI compared with unadjusted cTnI. Specificity was also improved in patients with renal dysfunction by using the adjusted cTnI values. Thus, the adjustments improved the diagnostic ability of cTnI to identify AMI in elderly patients and in patients with renal dysfunction. Interpretation of cTnI values in complex emergency cases is facilitated by our method, which maintains a single diagnostic cut-off value in all patients.
Based on increasing evidence suggesting that MS pathology involves alterations in bioactive lipid metabolism, the present analysis was aimed at generating a complex serum lipid-biomarker. Using unsupervised machine-learning, implemented as emergent self-organizing maps of neuronal networks, swarm intelligence and Minimum Curvilinear Embedding, a cluster structure was found in the input data space comprising serum concentrations of d = 43 different lipid-markers of various classes. The structure coincided largely with the clinical diagnosis, indicating that the data provide a basis for the creation of a biomarker (classifier). This was subsequently assessed using supervised machine-learning, implemented as random forests and computed ABC analysis-based feature selection. Bayesian statistics-based biomarker creation was used to map the diagnostic classes of either MS patients (n = 102) or healthy subjects (n = 301). Eight lipid-markers passed the feature selection and comprised GluCerC16, LPA20:4, HETE15S, LacCerC24:1, C16Sphinganine, biopterin and the endocannabinoids PEA and OEA. A complex classifier or biomarker was developed that predicted MS at a sensitivity, specificity and accuracy of approximately 95% in training and test data sets, respectively. The present successful application of serum lipid marker concentrations to MS data is encouraging for further efforts to establish an MS biomarker based on serum lipidomics.