Diagnostic Power of Circulatory Metabolic Biomarkers As Metabolic Syndrome Risk Predictors in Community-Dwelling Older Adults in Northwest of England (A Feasibility Study) Publisher Pubmed



Hassannejad R¹ ; Sharrouf H² ; Haghighatdoost F¹ ; Kirk B^{2, 3, 4} ; Amirabdollahian F² Show Affiliations
Source: Nutrients Published:2021

Abstract

Background: Metabolic Syndrome (MetS) is a cluster of risk factors for diabetes and cardiovascular diseases with pathophysiology strongly linked to aging. A range of circulatory metabolic biomarkers such as inflammatory adipokines have been associated with MetS; however, the diagnostic power of these markers as MetS risk correlates in elderly has yet to be elucidated. This cross-sectional study investigated the diagnostic power of circulatory metabolic biomarkers as MetS risk correlates in older adults. Methods: Hundred community dwelling older adults (mean age: 68.7 years) were recruited in a study, where their blood pressure, body composition and Pulse Wave Velocity (PWV) were measured; and their fasting capillary and venous blood were collected. The components of the MetS; and the serum concentrations of Interleukin-6 (IL-6), Tumor Necrosis Factor-α (TNF-α), Plasminogen Activator Inhibitor-I (PAI-I), Leptin, Adiponectin, Resistin, Cystatin-C, C-Reactive Protein (CRP), insulin and ferritin were measured within the laboratory, and the HOMA1-IR and Atherogenic Index of Plasma (AIP) were calculated. Results: Apart from other markers which were related with some cardiometabolic (CM) risk, after Bonferroni correction insulin had significant association with all components of Mets and AIP. These associations also remained significant in multivariate regression. The multivariate odds ratio (OR with 95% confidence interval (CI)) showed a statistically significant association between IL-6 (OR: 1.32 (1.06–1.64)), TNF-α (OR: 1.37 (1.02–1.84)), Resistin (OR: 1.27 (1.04–1.54)) and CRP (OR: 1.29 (1.09–1.54)) with MetS risk; however, these associations were not found when the model was adjusted for age, dietary intake and adiposity. In unadjusted models, insulin was consistently statistically associated with at least two CM risk factors (OR: 1.33 (1.16–1.53)) and MetS risk (OR: 1.24 (1.12–1.37)) and in adjusted models it was found to be associated with at least two CM risk factors and MetS risk (OR: 1.87 (1.24–2.83) and OR: 1.25 (1.09–1.43)) respectively. Area under curve (AUC) for receiver operating characteristics (ROC) demonstrated a good discriminatory diagnostics power of insulin with AUC: 0.775 (0.683–0.866) and 0.785 by cross validation and bootstrapping samples for at least two CM risk factors and AUC: 0.773 (0.653–0.893) and 0.783 by cross validation and bootstrapping samples for MetS risk. This was superior to all other AUC reported from the ROC analysis of other biomarkers. Area under precision-recall curve for insulin was also superior to all other markers (0.839 and 0.586 for at least two CM risk factors and MetS, respectively). Conclusion: Fasting serum insulin concentration was statistically linked with MetS and its risk, and this link is stronger than all other biomarkers. Our ROC analysis confirmed the discriminatory diagnostic power of insulin as CM and MetS risk correlate in older adults. © 2021 by the authorLicensee MDPI, Basel, Switzerland.