Nasal and systemic interleukin-1β and interleukin-3 in chronic obstructive pulmonary disease: association with upper airway symptom burden rather than with disease severity

Chronic obstructive pulmonary disease (COPD) is accompanied by inflammation that is not confined to the lower airways, and upper airway symptoms are common. It is not known whether nasal cytokine concentrations reflect local upper airway inflammation or simply mirror overall disease severity. We measured nasal and blood interleukin-1β (IL-1β) and interleukin-3 (IL-3) in patients with COPD stratified by upper airway symptom burden, and tested whether any association with severity was independent of age and airflow limitation. In this cross-sectional study, 70 clinically stable outpatients with COPD were stratified by the nasal subdomain of the 22-item Sino-Nasal Outcome Test (SNOT-22) into high (≥ 6 points; n = 35) and low (< 6 points; n = 35) upper airway symptom groups. Blood and nasal cytokines were measured by sandwich enzyme-linked immunosorbent assay. Between-group comparisons were explicitly unadjusted; correlations were then repeated as partial rank correlations adjusted for age and for age plus GOLD stage. Binary logistic regression for frequent hospitalisation (≥ 2 admissions per year) was restricted to three predictors to respect the number of events, cytokines were modelled per 1 standard deviation (SD), and Firth penalised regression was used as a sensitivity analysis. A further sensitivity analysis modelled the total SNOT-22 score as a continuous outcome. In unadjusted analyses, the high upper airway symptom group had higher blood IL-1β (median 2458 [IQR 1966–3805] versus 1287 [867–1619] pg/L, p < 0.001) and higher nasal IL-1β (2145 [1776–2540] versus 1722 [1093–2318] pg/L, p = 0.034). IL-3 did not differ between groups (blood p = 0.080; nasal p = 0.063). This group was also older (median 71 versus 63 years) and had more severe airflow limitation (GOLD 3–4: 97.2% versus 0%). Blood IL-1β correlated moderately with severity measures (rho 0.57–0.62) and these associations persisted after adjustment for age. Nasal IL-1β correlations with systemic severity measures were weak (rho 0.09–0.30) and were no longer significant after adjustment for age. In contrast, the association between nasal IL-1β and total SNOT-22 score persisted after adjustment for age and GOLD stage (partial rho = 0.491, p < 0.001; linear regression B = 4.6 SNOT-22 points per SD, p < 0.001). Blood IL-1β was independently associated with frequent hospitalisation after adjustment for age and SGRQ score (adjusted OR 3.62 per SD, 95% CI 1.41–9.31, p = 0.008) but only borderline after adjustment for age and GOLD stage (3.21, 0.96–10.78, p = 0.059). Nasal IL-1β was not an independent predictor (1.26, 0.67–2.36, p = 0.475) and discriminated poorly (area under the curve 0.651) compared with the SGRQ score (0.913) and GOLD stage (0.878). In COPD, systemic IL-1β tracks disease severity, whereas nasal IL-1β is associated specifically with upper airway symptom burden and adds no independent prognostic information beyond established clinical severity measures. IL-3 was not associated with disease severity in this unselected COPD population. Nasal IL-1β is best regarded as a marker of local upper airway inflammation rather than of COPD severity. ClinicalTrials.gov, NCT05980611.