Clinical prediction study on the risk of atrial fibrillation in hypertensive patients based on metabolism, inflammation, and gender differences.

This study aimed to explore the risk factors for atrial fibrillation (AF) within one year after discharge in hypertensive patients and to construct a corresponding predictive model. This single-center, retrospective study included 566 patients admitted with hypertension. Patients were divided into two groups: those who developed AF within one year after discharge and those who did not. Variables were selected for multivariate regression analysis using univariate regression and variance inflation factor (VIF) analysis. Subgroup analysis was performed by gender to explore the predictive value of the variables, and a nomogram was constructed. The total sample was randomly divided into a training set and a validation set (7:3 ratio). The discrimination and calibration of the predictive model were evaluated using receiver operating characteristic (ROC) and calibration curves. Patients who developed AF within one year had significantly higher levels of white blood cells (WBC), neutrophils (NEUT), lymphocytes (LYMPH), creatinine (Scr), fasting blood glucose (FBG), triglycerides (TG), lipoprotein(a) [Lp(a)], glycated hemoglobin (HbA1c), neutrophil-to-lymphocyte ratio (NLR), and triglyceride-glucose (TyG) index compared to those who did not (P < 0.05). Males, smokers, and diabetic patients were more prevalent in the AF group (P < 0.05). Logistic regression analysis showed that male gender, Lp(a), HbA1c, NLR, and the TyG index were independent predictors of AF within one year after discharge in hypertensive patients. The nomogram constructed showed an area under the ROC curve (AUC) of 0.793 in the training set and 0.740 in the validation set. The calibration curves indicated good fit (P = 0.726 in the training set; P = 0.489 in the validation set). Male, Lp(a), HbA1c, NLR, and the TyG index are independent risk factors for AF within one year of discharge in hypertensive patients. The nomogram model constructed has high predictive accuracy. This study suggests that individualized management strategies should be employed based on these risk factors in clinical practice.