Respiratory tract infections are the leading cause of antimicrobial use in calves. Combining clinical examination and lung ultrasonography allows on-farm classification of calves as healthy or suffering from an upper respiratory tract infection (URTI), subclinical or clinical pneumonia. This might help to improve targeted antimicrobial therapy, restricting treatment to pneumonic cases. However, to what extent these diagnostic categories coincide with expected bacteriological and cytological bronchoalveolar lavage fluid (BALf) characteristics is currently unknown. The objective of this study was therefore to compare BALf bacteriology and cytology between healthy calves and calves with URTI, subclinical and clinical pneumonia. The hypothesis was that calves with subclinical and clinical pneumonia would have higher quantitative bacterial counts, bacterial isolation rates and neutrophil counts than URTIs or healthy animals. A cross-sectional study was performed on 305 indoor group-housed dairy and beef calves, from 62 farms. Calves were classified by combining clinical examination and lung ultrasonography. Clinical respiratory disease was defined using the Wisconsin score card and the Healthy Criterion (HC). The HC classified calves as clinically ill if at least one clinical sign was present. Ultrasonographic lung consolidation with a depth of ≥1 cm was considered indicative for pneumonia. Cytology and bacteriology were performed on BALf sampled by non-endoscopic bronchoalveolar lavage. Calves with clinical pneumonia were further subdivided based on culture result and presence of neutrophils phagocytosing bacteria. Combined lung ultrasonography and clinical examination (HC) classified 25.9 % (79/305) of the calves as healthy, 33.1 % (101/305) as URTI, 10.2 % (31/305) as subclinical and 30.8 % (94/305) as clinical pneumonia. Bacterial isolation rates and quantitative BALf culture results did not differ between groups. Calves with clinical pneumonia and neutrophil phagocytosis showed a significantly higher BALf neutrophil percentage compared to healthy calves (59.0 % vs. 37.7 % in healthy calves, P =.03). Inversely, lymphocyte percentage was lower in these calves (1.8 % vs. 5.3 % in healthy calves, P = .003). Classification of calves using lung ultrasonography and clinical scoring did not correspond with BALf bacteriology and cytology findings, as extrapolated from human and companion animal medicine. Under the current housing conditions of this study high rates of non-infectious airway inflammation or airway colonization by opportunistic pathogens, rather than infection might explain this. Isolation of respiratory pathogens from calves with various signs of respiratory disease or ultrasonographic lesions should be interpreted carefully. Of all cytological features, phagocytosis by neutrophils in BALf might be a useful criterion supporting the diagnosis of bacterial respiratory tract infection.