Cardiovascular

Association Between Multiple Inflammation-Derived Indices and the Risk of Post-Stroke Epilepsy.

TL;DR

Systemic immune-inflammation index (SIRI) showed a relatively strong association with PSE risk and moderate discriminative ability in the stroke population, and may serve as an auxiliary reference indicator for early risk stratification of post-stroke epilepsy.

Key Findings

Each one-standard-deviation increase in SIRI was associated with a 172% increased risk of post-stroke epilepsy.

  • OR = 2.72, FDR-adjusted p < 0.001
  • Quartile-stratified analysis revealed a clear dose-response relationship (trend p < 0.001)
  • SIRI was calculated from routine hematological parameters collected within 24 hours of admission
  • The association was assessed using multivariable logistic regression models

SIRI demonstrated the best discriminative ability for PSE among the six inflammatory indices examined.

  • AUC = 0.693, 95% CI: 0.654–0.732
  • SIRI outperformed NLR, DNLR, MLR, NPR, and NMLR in ROC analysis
  • Discriminative ability was also assessed using NRI and DCA, with SIRI performing relatively better on all measures
  • The discriminative ability was characterized as 'moderate'

The association between SIRI and PSE risk was consistent across different patient subgroups.

  • Subgroup analyses were conducted to verify the stability of the results
  • Results indicated consistency across different subgroups
  • Restricted cubic spline regression and E-value analysis further supported the robustness of the association

The study included 317 PSE patients matched 1:1 with 317 stroke patients without PSE using propensity score matching.

  • Data were extracted from hospitalized stroke patients at Hefei Second People's Hospital between January 2018 and December 2025
  • Propensity score matching balanced confounding factors including sex, age, and underlying diseases
  • This was a retrospective case-control study design
  • Six inflammatory indices were calculated: SIRI, NLR, DNLR, MLR, NPR, and NMLR

Nonlinear relationships and threshold effects between inflammatory indices and PSE risk were explored using restricted cubic spline regression.

  • Restricted cubic spline regression was used alongside standard multivariable logistic regression
  • E-value analysis was performed to assess robustness against unmeasured confounding
  • Decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were used to assess incremental value of each index

SIRI's strength of association with PSE was superior to that of all other inflammation-derived indices studied.

  • The six indices compared were SIRI, NLR, DNLR, MLR, NPR, and NMLR
  • All indices were derived from routine hematological parameters collected within 24 hours of admission
  • The authors noted that 'simple and efficient risk assessment tools remain lacking in clinical practice' for PSE

What This Means

This research suggests that a blood-test-derived measure of immune system activity called the Systemic Immune-Inflammation Index (SIRI) is meaningfully linked to the risk of developing epilepsy after a stroke — a condition known as post-stroke epilepsy (PSE). The study compared 317 stroke patients who developed PSE with 317 stroke patients who did not, carefully matching the two groups by age, sex, and other health conditions. The researchers found that higher SIRI values at hospital admission were strongly associated with greater PSE risk, and this relationship appeared consistent across different patient groups. SIRI outperformed five other commonly used inflammation markers in predicting PSE risk. The SIRI is calculated from routine blood tests that are already taken when patients arrive at the hospital, meaning it would require no additional testing to implement. This makes it a potentially practical and low-cost tool for identifying stroke patients who may be at higher risk of developing epilepsy later on. However, while the association was statistically robust, the test's ability to distinguish between those who will and won't develop PSE was only moderate (AUC of 0.693), meaning it would not be sufficient as a standalone diagnostic tool. This research suggests that inflammation-related blood markers, particularly SIRI, could serve as useful supplementary indicators to help clinicians flag high-risk patients for closer monitoring after stroke. The authors caution that the findings come from a single hospital's retrospective data and that prospective studies are needed before SIRI can be routinely recommended for PSE risk stratification in clinical practice.

Have a question about this study?

Citation

Wang Y, Gao Y, Wang L. (2026). Association Between Multiple Inflammation-Derived Indices and the Risk of Post-Stroke Epilepsy.. CNS neuroscience &amp; therapeutics. https://doi.org/10.1002/cns.71143