Prognostic Value of Inflammatory Hematological Indices for In-Hospital Mortality After Stroke

TL;DR

This study found that certain blood-based inflammation markers can predict in-hospital death after stroke, with SIRI and NLR being the most useful.

Contribution

The study identifies SIRI and NLR as strong predictors of in-hospital mortality in stroke patients, regardless of stroke type.

Findings

NLR, SIRI, AISI, and SII were significantly higher in patients with unfavorable outcomes compared to those with favorable outcomes.

NLR and SIRI showed the strongest associations with in-hospital mortality in logistic regression analysis.

ROC analysis indicated strong predictive performance for NLR and SIRI, with AUC values of 0.885 and 0.867, respectively.

Abstract

Background and Objectives: This study aimed to assess the prognostic value of several inflammation-based hematological indices in patients with ischemic and hemorrhagic stroke and to evaluate whether their prognostic significance differs between stroke types. The analyzed indices included the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), systemic inflammatory response index (SIRI), systemic immune-inflammation index (SII), and aggregate index of systemic inflammation (AISI). Materials and Methods: This retrospective cohort study analyzed the medical records of stroke patients admitted to two hospitals in Shymkent, Kazakhstan. Hematological parameters were calculated from routine complete blood counts obtained on the third day of hospitalization. Nonparametric tests, univariable and multivariable logistic regression, and receiver operating characteristic (ROC) analysis were used to evaluate associations between inflammatory indices and in-hospital mortality. Results: A total of 199 patients who met the inclusion criteria were classified into three groups according to in-hospital outcome at discharge: (1) patients discharged alive (favorable outcome), (2) patients who died during hospitalization due to ischemic stroke (unfavorable ischemic stroke), and (3) patients who died during hospitalization due to hemorrhagic stroke (unfavorable hemorrhagic stroke). NLR, SIRI, AISI, and SII values were significantly higher in both unfavorable outcome groups compared with the favorable outcome group (p < 0.001, effect size r > 0.6). No statistically significant differences were observed between unfavorable ischemic and hemorrhagic stroke outcomes. In logistic regression analysis, NLR (OR = 1.65) and SIRI (OR = 2.36) showed the strongest associations with in-hospital mortality. ROC analysis demonstrated good predictive performance, with AUC values of 0.885 for NLR and 0.867 for SIRI. Conclusions: The inflammatory indices evaluated in this study were associated with stroke outcomes regardless of stroke subtype. Among them, SIRI and NLR showed the highest prognostic value. These indices may serve as accessible markers of disease severity but should not be considered independent clinical decision-making tools.

Figures