Abstract
BACKGROUND: Stroke-associated pneumonia (SAP) is a severe complication of acute ischemic stroke (AIS). While post-stroke immunosuppression and metabolic dysregulation are implicated in its pathogenesis, biomarkers integrating these pathways are needed for early risk stratification. We evaluated the lymphocyte-to-HDL-cholesterol ratio (LHR), a marker of immune-metabolic status, as a predictor for SAP. METHODS: We retrospectively analyzed 978 consecutive AIS patients from a single-center cohort. The primary outcome, SAP, was diagnosed based on modified Centers for Disease Control and Prevention criteria within the first 7 days of hospitalization. The relationship between the LHR, calculated from baseline measurements on admission, and SAP was assessed using smooth curve fitting and multivariable logistic regression, adjusting for potential confounders (eg, age, sex, key comorbidities, and thrombolysis treatment). The association's robustness was tested via subgroup and sensitivity analyses. RESULTS: SAP occurred in 9.2% (90/978) of patients. A distinct nonlinear, L-shaped association was found: SAP risk increased sharply as LHR fell below an empirically identified threshold of approximately 1.0, while remaining consistently low at higher LHR values. In the fully adjusted model, a low LHR (<1.0) was independently associated with a nearly twofold increased risk of SAP (adjusted OR 1.98; 95% CI, 1.20-3.24; p < 0.05). The association was robust across major subgroups (eg, by age), though a trend toward a stronger effect was observed in male patients (p for interaction = 0.06). CONCLUSION: A low admission LHR (<1.0) serves as a novel, independent predictor of SAP in AIS patients, exhibiting a clinically relevant data-driven threshold effect. As a cost-effective and readily available biomarker reflecting the interplay between immunity and metabolism, LHR holds significant potential for early risk stratification using this empirically derived cut-off, particularly in resource-limited settings.