Lower Lymphocyte-Albumin-Neutrophil Ratio Provides Complementary Clinical Information for Identifying Patients with Prolonged Hospitalization Following Acute Exacerbation of Chronic Obstructive Pulmonary Disease: A Retrospective Cohort Study

Abstract Background Acute exacerbations of COPD (AECOPD) are leading causes of hospitalization, healthcare expenditure, and disease progression. Prolonged hospitalization is associated with increased costs, functional decline, and adverse outcomes. Simple biomarkers from routine laboratory tests may complement multidimensional tools in assessing exacerbation risk during initial evaluation. The lymphocyte-albumin-neutrophil ratio (LANR) integrates inflammatory activity, immune competence, and nutritional status, whereas the systemic immune-inflammation index (SII) reflects systemic inflammatory burden. Whether LANR provides information beyond conventional biomarkers for identifying patients at risk of prolonged hospitalization remains uncertain. Methods This retrospective cohort study included 166 consecutive patients hospitalized for AECOPD at Shenzhen Second People's Hospital between December 2023 and April 2025. The primary outcome was prolonged hospitalization (> 7 days). Clinical characteristics, laboratory variables, and inflammatory biomarkers were extracted from electronic medical records. Multivariable logistic regression analyses were performed using clinically prespecified covariates. Model discrimination was evaluated using receiver operating characteristic (ROC) curves, bootstrap confidence intervals, and Brier scores. Restricted cubic spline analyses explored potential non-linear associations between LANR and prolonged hospitalization. Exploratory analyses of continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were performed to assess incremental clinical information. Results Among 166 patients hospitalized for AECOPD, 98 (59.0%) experienced prolonged hospitalization (> 7 days). LANR was lower in patients with prolonged hospitalization than in those hospitalized for ≤ 7 days [7.39 (IQR 3.83–11.75) vs. 10.97 (7.59–18.40), FDR-adjusted P = 0.003]. After adjustment for age, sex, body mass index, smoking exposure, fever, arterial pH, hs-CRP, and SII, higher LANR remained independently associated with lower odds of prolonged hospitalization (adjusted OR per 1-unit increase, 0.942; 95% CI, 0.891–0.995; P = 0.032), whereas SII was not independently associated (adjusted OR per 1000-unit increase, 0.961; 95% CI, 0.784–1.177; P = 0.700). The clinical base model yielded an AUC of 0.658 (bootstrap 95% CI, 0.572–0.742), which increased to 0.694 (0.609–0.776) after addition of LANR, with a corresponding reduction in the Brier score from 0.224 to 0.213. The Youden-derived LANR threshold of 9.595 provided 67.3% sensitivity and 64.6% specificity; patients with LANR ≤ 9.595 had higher odds of prolonged hospitalization (OR, 3.766; 95% CI, 1.945–7.293; P < 0.001). Thirty-day readmission occurred in 10 patients (6.1%), and neither LANR nor SII showed a significant association with this exploratory outcome. Conclusions Lower LANR was independently associated with prolonged hospitalization in patients hospitalized for AECOPD. Although its incremental contribution beyond routinely available clinical variables was modest, LANR integrates inflammatory, immune, and nutritional information available from routine laboratory testing and may serve as a practical adjunct to early clinical assessment. Prospective multicenter studies are warranted to validate these findings.