Abstract
The nasal mucosa is a crucial filtering organ to prevent attachment and invasion of pathogens. To assess nasal health in relation to lung health, transverse cross sections of the nasal turbinates of 121 pigs suffering from respiratory disease and sent for diagnostic necropsy were scored visually and by an artificial intelligence (AI) medical diagnostic application (AI DIAGNOS), resulting in a high correlation of both scores (p < 0.001). Nasal samples of the diseased pigs were examined only for Bordetella (B.) bronchiseptica (PCR and bacteriological culture) and Pasteurella (P.) multocida (bacteriological culture). All pigs showed various degrees of inflammatory lung tissue alterations, and 35.5% of the pigs had atrophy of the nasal turbinates with no relation to detection rates of B. bronchiseptica (54.5%) and P. multocida (29.0%) in the nose. All P. multocida strains from nose samples were negative for the toxA gene so non-progressive atrophic rhinitis was diagnosed. Pigs positive for B. bronchiseptica in the nose were more often positive for B. bronchiseptica in the lung (p < 0.001) and for other bacterial species in the lower respiratory tract (p = 0.005). The new diagnostic application for scoring cross sections of nasal turbinates is a valuable tool for a fast and reproducible diagnostic.