Background and Objectives: The bronchoscopic microsample (BMS) probe allows direct epithelial lining fluid (ELF) level measurement without saline lavage. We investigated whether cytokine levels in ELF from a BMS differed from those obtained by bronchoalveolar lavage (BAL) in stable and acute lung disease. Methods: In a single-centre, prospective observational cohort study of 45 patients, a sequential BMS probe procedure and BAL were performed on patients with stable chronic obstructive lung disease, interstitial lung disease, acute lung injury (ALI), burns-related inhalational injury or controls. ELF samples were assayed for IL-1β, IL-6, IL-8, TNF-α and G-CSF. Results: Both bronchoscopic microsampling and BAL showed significantly higher cytokine levels in the ELF from patients with ALI and burns-related inhalational injury than from those with chronic stable lung disease. The BMS method detected cytokine levels approximately 20- to 80-fold higher than the corresponding BAL (uncorrected for dilution). The ratio of BMS and BAL cytokine levels was as follows: the ratio for IL-1β [mean 55, 95% confidence interval (CI) 34-88] was higher than that for IL-6 (mean 16, 95% CI 10-23, p = 0.015) and IL-8 (mean 13, 95% CI -5 to 36, p = 0.03). The ratio for G-CSF (mean 43, 95% CI 24-75) was higher than that for IL-6 (mean 16, 95% CI 10-23, p = 0.008). Conclusions: The BMS probe safely collects ELF with higher equivalent inflammatory cytokine concentrations than via BAL from patients with both acute and chronic lung disease and can be an alternative to saline BAL. Variations in cytokine concentrations between BMS and BAL and sampling-site differences warrant further study.