Gender-specific MetS prediction using pathophysiological determinants: Beyond diagnostic constraints

Abstract

Metabolic syndrome (MetS), constituted by obesity, hyperglycemia, hypertension, and dyslipidemia, is a growing public health concern due to its associated risk with cardiovascular and other metabolic disorders. Early-stage detection and prevention of MetS are key factors in an effective management strategy. Thus, the current study aimed to develop a risk score for MetS progression based on features pertaining to its pathophysiology, that can facilitate early detection and treatment strategies. A multivariate logistic regression model was developed using a representative feature from MetS pathophysiological pathways of inflammation, endothelial dysfunction, and hepatic dysregulation. The model was built on NHANES dataset and validated in Chinese and Indian datasets. The model performance evaluated using ROC resulted in an AUC of 0.81 in training and 0.87 and 0.79 for Chinese and Indian validation datasets, respectively. In conclusion, a MetS predictive model built on three MetS pathophysiological determinants was developed and validated in distinct datasets. HighlightsO_LIA MetS prediction model, trained on NHANES data, is validated with two datasets of varied ethnicity. C_LIO_LIThe three parameter model, independent of MetS diagnostic criteria, is reflective of inflammation, endothelial dysfunction, and hepatic dysregulation. C_LIO_LIThe model achieved an AUC of 0.81 in NHANES data; 0.87 and 0.79 for Chinese and Indian validation datasets respectively. C_LI