Antibiotics (ATB) are frequently used for lower respiratory tract infections treatment, with sometimes an antibiotic failure. We conducted a prospective cohort study in Amiens University Hospital. Patients treated by beta-lactams for a lower respiratory tract infection, who needed a flexible bronchoscopy, were included. The concentration of ATB was measured in serum and in bronchoalveolar lavage (BAL). The main objective was to evaluate the pulmonary penetration (PP), ratio between alveolar and serum concentration of ATB and to determine predictive factors for PP. Secondary objectives were to assess the role of viral infections explaining antibiotic failure and to identify viruses frequently implicated. Ninety  patients were included. Pulmonary penetration rate was 246.25 ± 1126 (27 [1-8410]). It ranged from 68 ± 82.87 (16 [1-236]) for CEFTRIAXONE to 463.47 ± 1924.47 (11[2-8420]) for PIPERACILLINE. None of the factors studied was predictive of PP. AFC/MIC ratio (AFC=alveolar film concentration; MIC=minimum inhibitory concentration) was used to study the efficacy of ATB. It was 4.85 ± 8.52 for all ATB. None of factors studied was predictive of antibiotic efficacy. No correlation was found between beta-lactam alveolar penetration and inhibitors alveolar penetration, when administrated simultaneously. At least one virus was found in 60% of BAL. Viruses were more significantly identified in patients with bronchiectasis. Fungal co-infections were more significantly found in patients with bronchiectasis or COPD. In conclusion, PP was highly variable. ATB concentration in serum does not predict alveolar concentration. Viruses were frequently found in bronchiectasis.