Divergent Inflammatory Trajectories: Independent Predictive Value of Admission Neutrophil-to-Lymphocyte and Lymphocyte-to-Monocyte Ratios for Acute Ischemic Stroke Severity in a Southeast Asian Cohort

Abstract

Background: Systemic sterile inflammation acts as a critical pathophysiological driver in the acute phase of ischemic stroke, mediating secondary brain injury through blood-brain barrier disruption and microvascular thrombosis. While the Neutrophil-to-Lymphocyte Ratio (NLR) and Lymphocyte-to-Monocyte Ratio (LMR) have emerged as potential biomarkers in Western populations, their independent prognostic utility and specific diagnostic thresholds within Southeast Asian populations remain under-explored. This region presents unique challenges due to potential variations in baseline hematological profiles driven by genetic polymorphisms and environmental factors. This study aims to elucidate the association between admission NLR and LMR levels and the severity of Acute Ischemic Stroke (AIS) and to determine optimal population-specific prognostic cut-offs.

Methods: We conducted a retrospective cross-sectional comparative study involving 128 patients with confirmed AIS admitted to Wangaya Regional General Hospital, Indonesia, between January 2025 and August 2025. Patients were stratified based on admission National Institutes of Health Stroke Scale (NIHSS) scores into a Mild Group (NIHSS ≤ 6, n=64) and a Moderate-Severe Group (NIHSS > 6, n=64). Infection was strictly excluded using clinical and radiological criteria independent of admission leukograms to prevent circular bias. Receiver Operating Characteristic (ROC) curve analysis was performed to determine diagnostic accuracy and identify optimal cut-offs. To address potential multicollinearity between NLR and LMR, two separate multivariate binary logistic regression models were constructed to determine independent predictors of severity.

Results: The study population had a mean age of 60.5 years. The Moderate-Severe Group exhibited significantly higher NLR (6.12 ± 3.41 vs. 2.85 ± 1.20; p < 0.001) and lower LMR (2.15 ± 0.92 vs. 4.22 ± 1.50; p < 0.001) compared to the Mild Group. ROC analysis identified optimal cut-offs of ≥ 4.82 for NLR (AUC: 0.782; Sensitivity: 76.6%) and ≤ 2.89 for LMR (AUC: 0.724; Sensitivity: 71.9%). In the multivariate analysis Model 1, NLR ≥ 4.82 remained an independent predictor of severity (Adjusted Odds Ratio [aOR]: 4.12; 95% CI: 1.78–9.54; p = 0.001). In the separate Model 2, LMR ≤ 2.89 was also confirmed as an independent predictor (aOR: 2.85; 95% CI: 1.24–6.55; p = 0.014).

Conclusion: Elevated NLR and reduced LMR at admission are robust, independent indicators of stroke severity in this Indonesian cohort. These accessible hematological biomarkers reflect the divergent trajectories of post-ischemic neuroinflammation—innate immune hyperactivity and adaptive immune exhaustion. They provide a cost-effective method for risk stratification in resource-limited settings, warranting their integration into routine initial assessment protocols.