The correlation between the systemic inflammation response index and 6-month readmission risk in patients with hypertensive heart disease-related heart failure and its predictive model.

OBJECTIVE

To investigate the association between the Systemic Inflammation Response Index (SIRI) and the risk of 6-month readmission in patients with hypertensive heart disease-related heart failure (HHD-HF), and develop and validate a nomogram prediction model integrating SIRI.

METHODS

A retrospective cohort of 158 hypertensive heart disease-related heart failure (HHD-HF) patients (June 2022-December 2024) was enrolled. Baseline clinical data, laboratory indicators, and cardiac structural and functional parameters were collected. Patients were split 7:3 into training (n = 111) and validation (n = 47) sets. LASSO regression was used for variable selection, and a multivariate logistic regression model incorporating SIRI was constructed and visualized as a nomogram. The model's discriminative ability, calibration, and clinical utility were evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Additionally, SHapley Additive exPlanations (SHAP) analysis was employed to interpret the contribution and directionality of SIRI and other predictors at both global and individual levels. Spearman rank correlation was used to explore the associations between SIRI and cardiac structural parameters, renal function, and comorbidities.

RESULTS

The LVMI (OR = 1.02, p = 0.046), diabetes history (OR = 5.62, p = 0.028), and SIRI (OR = 2.88, p = 0.014) as independent predictors of 6-month readmission in HHD-HF patients. The nomogram showed discrimination in training and validation cohorts (AUC 0.950 and 0.948, respectively). SHAP analysis confirmed SIRI as the dominant contributor (mean |SHAP| ≈ 1.4). SIRI positively correlated with readmission risk (R = 0.610) and LVMI (R = 0.413), and negatively eGFR (R = -0.446, all p < 0.001). In the total cohort, SIRI alone yielded an AUC of 0.863 (cutoff 1.71). Compared with other markers in the training cohort, SIRI significantly outperformed CRP (p = 0.001), NLR (p < 0.001), and BNP (p < 0.001). Subgroup analysis by LVEF showed consistent performance of SIRI in HFrEF (AUC 0.868) and HFpEF (AUC 0.829, P for difference = 0.589), and SIRI remained an independent predictor in both phenotypes (HFrEF: OR = 3.65, p = 0.025; HFpEF: OR = 3.33, p = 0.038).

CONCLUSION

This nomogram incorporating LVMI, diabetes history, and SIRI effectively predicts 6-month readmission in HHD-HF patients, with discriminative ability and good calibration.