Abstract
OBJECTIVE: To evaluate the predictive utility of peripheral blood eosinophil percentage (Eos%) for subclinical pulmonary impairment in patients with chronic rhinosinusitis (CRS) and to investigate its association with coagulation parameters. METHODS: This single-center, retrospective study included 303 non-asthmatic patients with CRS, who were stratified into an uncomplicated CRS group (n = 241) and a subclinical pulmonary impairment group (n = 62). A cohort of 41 healthy individuals served as controls. Data on peripheral blood counts (including Eos%), coagulation indices (e.g., APTT), and spirometric parameters were collected and compared. Multivariable logistic regression analysis was performed to identify independent predictors of pulmonary impairment, and receiver operating characteristic (ROC) curve analysis was used to assess the predictive performance of significant variables. RESULTS: Patients in the pulmonary impairment subgroup exhibited a significantly higher mean Eos% than those in the uncomplicated CRS group (5.94% vs. 3.51%, P < 0.001). Eos% was identified as an independent predictor of subclinical pulmonary impairment (Odds Ratio [OR] = 1.133, 95% Confidence Interval [CI]: 1.052–1.220, P < 0.001). ROC analysis revealed a moderate predictive utility for Eos%, with an Area Under the Curve (AUC) of 0.677 (95% CI: 0.603–0.751). Notably, APTT levels did not differ significantly between the two CRS subgroups (P = 0.70), and no significant interaction effect between Eos% and APTT on pulmonary impairment was found. Furthermore, both Eos% and platelet counts correlated with sinonasal symptom severity, but only Eos% was associated with pulmonary impairment. CONCLUSION: Peripheral Eos% demonstrates moderate utility as a biomarker for risk stratification of subclinical pulmonary impairment in non-asthmatic CRS patients. While coagulation abnormalities are present in the overall CRS population, they do not specifically associate with lower airway dysfunction, suggesting they may reflect a systemic inflammatory state rather than a direct interactive mechanism driving pulmonary injury. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12890-026-04196-0.