Department of Infectious Diseases, University of Veterinary Medicine Institute for Microbiology, Hannover Foundation, Hannover Comparison of Three Common Methods of Lung Lavage in Healthy Pigs I. Hennig-Pauka 1,2,6 , S. Bremerich 3 , H. Nienhoff 4 , C. Schro ¨ der 2 , M. Ganter 2 , F. Blecha 5 , K.-H. Waldmann 2 and G. F. Gerlach 1 Addresses of authors: 1 Institute for Microbiology, Department of Infectious Diseases, University of Veterinary Medicine, Hannover, Foundation, Hannover; 2 Clinic for Swine and Small Ruminants, University of Veterinary Medicine, Hannover, Foundation, Hannover; 3 Veterinary Association of the BHZP (Bundeshybridzucht Programm), Ellringen, Germany; 4 Pig Health Service, Landwirtschaftskammer Hannover, Lower Saxony, Germany; 5 Department of Anatomy and Physiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS, USA; 6 Corresponding author: Tel.: 49 511 8567260; fax: 49 511 8567684; E-mail: isabel.hennig@tiho-hannover.de With 3 figures and 2 tables Received for publication November 18, 2006 Summary Bronchoscopic, endotracheal and transtracheal lung lavage were evaluated in 38 healthy pigs taken from a nucleus herd in a good state of health with respect to their applicability in practice and the traceability of bacteria, cellular parameters and the antimicrobial peptide PR-39 in the respective lavage fluid samples. The total cell count, qualitative morphological cellular characteristics as well as PR-39 could be determined in all lavage fluid samples, while quantitative cell differentiation was not possible in endotracheal lavage samples. The com- parison of the three methods resulted in a higher proportion of polymorphonuclear neutrophil granulocytes (PMNs) and higher concentrations of PR-39 in transtracheal samples. For this reason different valuation standards with respect to PMNs and PR-39 concentrations are presupposed for transtracheal lavage samples. The occurrence of pavement epithelial cells as well as the number of contaminating bacterial species per sample was the lowest in transtracheal lavage. Mycoplasma hyopneumoniae polymerase chain reaction appeared to have the highest diagnostic sensitivity in combination with bronchoscopic lavage. In conclusion, bronchoscopic and transtracheal lavage were considered to be more appropriate for bacteriological and cytological diagnostics than endo- tracheal lavage. Introduction Assessing respiratory health and thereby determining its influencing factors, are of great economic importance for the swine industry and provide the basis for control and preven- tion of disease. The collection and examination of lung lavage fluid by various methods is being performed with increasing frequency in practice to assess the alveolar and bronchial lung milieu in pigs, focusing on the diagnosis of infectious respir- atory diseases (Flaßhoff, 1996; Jolie et al., 2000). Intravital lung lavage in swine has to be performed in anaesthetized pigs regardless of the method used (Flaßhoff, 1996; Ganter and Hensel, 1997; Nienhoff et al., 2006). Different protocols for the lung lavage performance in swine exist and not all implement the technical recommendations made in human medicine which are essential for research work in veterinary medicine (Klech and Pohl, 1989; Ganter and Hensel, 1997). Until now routine use of fibreoptic bronchoscope has not been field- tested due to economy and biosecurity reasons while endotra- cheal and transtracheal lavage with disposable material are widely used in practice. In veterinary medicine lung lavage fluid is investigated using two different approaches. In one approach, classical micro- biological and virological methods or polymerase chain reaction (PCR) analyses are used to investigate the presence of potential pathogens associated with the porcine respiratory disease complex (Palzer et al., 2005). The bacteriological evaluation of lung lavage fluid was found to be more efficient than the bacteriological examination of lung tissue (Kipper, 1990). In a second approach inflammation markers are investi- gated. Here, a cut-off of 8% polymorphonuclear neutrophil granulocytes (PMNs) in lung lavage fluid has been suggested in order to differentiate between healthy pigs and pigs with respiratory disorders (Ganter and Hensel, 1997; Mombarg et al., 2002). The porcine antibacterial peptide PR-39, of which the amount is increased during experimental Actino- bacillus pleuropneumoniae infection and can be semiquantified in lung lavage fluid samples by ELISA, has also been discussed as a respiratory disease marker in pigs (Hennig- Pauka et al., 2006). Both approaches are difficult to interpret as the impact of the lavage method onto parameters in lung lavage fluid has not been investigated in pigs, yet. In the following comparison of bronchoscopic, transtracheal and endotracheal lung lavage fluid samples of healthy pigs the extent to which the lavage method itself influences the results is examined. Materials and Methods Animals and anaesthesia The pigs (German Yorkshire breed, approximately 15–30 kg body weight) originated from a closed nucleus herd with a basic stock of 500 sows routinely monitored and specified pathogen-free from epizootics, endo- and ectoparasites, toxi- genic Pasteurella multocida, Actinobacillus pleuropneumoniae and porcine respiratory and reproductive syndrome virus. Sows were vaccinated against porcine parvovirus and swine erysipelas. Piglets were vaccinated against M. hyopneumoniae. In Germany, no nucleus herds in better respiratory health are known. The pigs exhibited no clinical signs or history of www.blackwell-synergy.com J. Vet. Med. A 54, 428–433 (2007) Ó 2007 The Authors Journal compilation Ó 2007 Blackwell Verlag ISSN 0931–184X