J-shaped associations of pan-immune-inflammation value and systemic inflammation response index with stroke among American adults with hypertension: evidence from NHANES 1999-2020.

Abstract

Stroke, a leading cause of death and disability worldwide, is primarily ischemic and linked to hypertension. Hypertension, characterized by systemic chronic inflammation, significantly increases stroke risk. This study explores the association of novel systemic inflammatory markers (SII, PIV, SIRI) with stroke prevalence in hypertensive U.S. adults using NHANES data. We analyzed data from hypertensive participants in the NHANES 1999-2020 survey, excluding those under 20, pregnant, or with missing data, resulting in 18,360 subjects. Systemic inflammatory markers (SII, PIV, SIRI) were calculated from blood counts. Hypertension and stroke status were determined by self-report and clinical measurements. Covariates included sociodemographic, lifestyle, and medical history factors. Weighted statistical analyses and multivariate logistic regression models were used to explore associations, with adjustments for various covariates. Ethical approval was obtained from the NCHS Ethics Review Board. In a cohort of 18,360 hypertensive individuals (mean age 56.652 years), 7.25% had a stroke. Stroke patients were older, had lower PIR, and were more likely to be female, single, less educated, smokers, non-drinkers, physically inactive, and have diabetes and CHD. Multivariate logistic regression showed that SII was not significantly associated with stroke. However, PIV and SIRI were positively associated with stroke prevalence. Each unit increase in lnPIV increased stroke odds by 14% (OR = 1.140, p = 0.0022), and lnSIRI by 20.6% (OR = 1.206, p = 0.0144). RCS analyses confirmed J-shaped associations for lnPIV and lnSIRI with stroke. Stratified analyses identified gender and smoking as significant effect modifiers. Smoking was significantly associated with elevated PIV, SIRI, and SII levels, especially in current smokers. Elevated PIV and SIRI levels significantly increase stroke prevalence in hypertensive individuals, notably among males and smokers. A predictive model with PIV, SIRI, and sociodemographic factors offers strong clinical utility.