Background: Antibiotic use in mild to severe acute exacerbations of COPD (AECOPD) remains controversial.

Aim: To analyze frequency and determinants of antibiotic use in severe AECOPD, and to investigate its association with hospital length of stay (LOS).

Methods: This retrospective, observational study in a Belgian tertiary hospital included 431 AECOPD patients (ICD-10 J44.0 and J44.1) discharged between 2016 and 2021. Determinants of in-hospital antibiotic use and its association with time to discharge alive were analyzed with logistic regression and Cox proportional hazards models, respectively.

Results: More than two-thirds (68%) of patients (mean age 70 years, 63% males) were treated with antibiotics, mainly amoxicillin-clavulanic acid. In multivariable analysis, patient-related variables (age, BMI, cancer), clinical variables (sputum volume, body temperature and C-reactive protein levels) and co-treatment (maintenance azithromycin and theophylline) were associated with in-hospital antibiotic use. The median hospital LOS was significantly longer in antibiotic-treated patients compared to non-antibiotic-treated patients (6 vs 4 days, p<0.001) indicated by a reduced probability of hospital discharge even after adjustment for age, sputum purulence, BMI, in-hospital systemic corticosteroid use and FEV₁ (aHR 0.60; 95% CI 0.43; 0.84).

Conclusions: In-hospital antibiotic use among AECOPD patients was common and determined by the exacerbation severity and patient characteristics. Moreover, in-hospital antibiotic use was associated with longer hospital stay.