Frontiers in Cardiovascular Medicine (Nov 2024)

Integration of the systemic inflammatory response index with pulse pressure enhances prognostication of cardiovascular mortality in the general population of the United States: insights from the NHANES database

  • Jie An,
  • Zikan Zhong,
  • Bingquan Xiong,
  • Dandan Yang,
  • Youquan Li,
  • Ya Luo,
  • Hao Li,
  • Yang Jiao,
  • Genqing Zhou,
  • Min Xu,
  • Shaowen Liu,
  • Jie Li

DOI
https://doi.org/10.3389/fcvm.2024.1439239
Journal volume & issue
Vol. 11

Abstract

Read online

BackgroundThe prognostic significance of utilizing both the systemic inflammatory response index (SIRI) and pulse pressure (PP) collectively in assessing cardiovascular mortality (CVM) across populations remains to be elucidated.MethodsMultivariate Cox proportional hazards analysis investigated the SIRI, PP, and CVM association. Receiver operating characteristic (ROC) curves evaluated the predictive performance of the combined SIRI and PP for CVM in the broader demographic. Subsequently, the area under the ROC curve (AUC) was compared using the Z-test, and a novel nomogram was developed to assess its accuracy in predicting CVM. Restricted cubic spline (RCS) was used to evaluate the association between SIRI and PP.ResultsThe study involved 19,086 NHANES database individuals, with 9,531 males (49.94%). During the follow-up period, 456 CVM instances (2.39%) occurred. Multivariate Cox proportional hazards analysis revealed both the SIRI [adjusted hazard ratio (HR) 1.16, P < 0.001] and PP (HR = 1.01, P = 0.004) as independent CVM predictors. A 0.1-unit SIRI increase and 10 mmHg PP escalation correlated with 2% (adjusted HR = 1.02, P < 0.001) and 7% (adjusted HR = 1.07, P = 0.004) CVM enhancements, respectively. The combined SIRI and PP area under the curve was 0.77, ranging from 0.77 to 0.79 in female cohorts, non-smokers, and non-pathological contexts. High SIRI and PP, either high SIRI or PP, were associated with 3 and 2 times the CVM risk compared to low SIRI and PP. Adding the SIRI and PP to general risk factors improved CVM predictive efficacy (Z = 4.17, P < 0.001). The novel nomogram's concordance index was 0.90, indicating excellent discrimination. The predicted probabilities’ calibration plot aligned with actual CVM rates at 1, 5, and 10 years. RCS showed an S-shaped relationship between SIRI and PP.ConclusionsIntegrating the SIRI with PP demonstrates substantial predictive efficacy for CVM within the broader United States community, notably in female cohorts, non-smokers, and non-pathological contexts.

Keywords