Development and Validation of a Novel Nomogram to Predict the Impact of the Polymorphisms of the Variants of ICAM-1 Gene on the Prognosis of Ischemic Cardiomyopathy.

Object: This study investigated the correlation between polymorphisms of the Methods: The current study included totally 576 patients with ICM. All patients are randomly divided into training group with 399 patients and validation group with 177 patients. The prognostic model was constructed by using the data of training group. Univariable Cox-regression analysis was performed, including clinical and gene variants, then used the least absolute shrinkage and selection operator (LASSO) regression model to optimize feature selection. Furthermore, multivariate Cox-regression was applied to build the prognostic nomogram model, which included clinical and gene features chosen by the LASSO regression model. Following that, the receiver operating characteristic (ROC) curve, C-index, calibration plot analyses and decision curve analysis (DCA) were carried out to evaluate the discrimination ability, consistency, and clinical utility of the prognostic model. Results: Predicting factors rs281430, ventricular arrhythmia, treating by PCI or CABG, use of β-blockers, heart rate (HR), serum sodium level, left ventricular end-diastolic diameter (LVDD) were the risk factors of the prognosis of ICM, incorporated these factors into the prognostic nomogram model. The constructed nomogram performed well in discrimination ability, as observed by the ROC and C-index. Furthermore, as shown by calibration curves, our nomogram's predicted probabilities were highly consistent with measured values. With threshold probabilities, DCA suggested that our nomogram could be useful in the clinic. Conclusion: rs281430 mutation (from AA genotype to AG or GG genotype) is a risk factor for ICM patients to have a higher survival probability; the survival probability of ICM patients with the mutant genotype (AG or GG) is lower than those with the wild genotype (AA).