Bacterial respiratory infections often occur in swine populations and usually require antibiotic treatment. For an effective and responsible therapy, it is crucial to avoid the development of antimicrobial resistance in the causative pathogens. The results of susceptibility testing for Pasteurella multocida (P. mult.), Actinobacillus pleuropneumoniae(App), Streptococcus suis (S. suis), Bordetella bronchiseptica (B. bronch.) and Glaesserella parasuis (Gps) for relevant antimicrobials between 2018 and 2020 have been analysed in a retrospective study. Minimal inhibitory concentrations (MICs) for P mult., App, S. suis, B. bronch. and Gps (n=77, 71, 67, 63 and 33) were determined for currently licensed antibiotics using broth microdilution technique. Clinical breakpoints (CB) were applied if available from the Clinical and Laboratory Standards Institute (CLSI) for the relevant agents and antimicrobials. If no CB were available for the respective pathogens, MIC90 values were determined. P mult., App and S. suis were susceptible (s) (>99%) to ceftiofur. Moreover, cefquinome can be assumed as highly effective (MIC90 <= 1 mu g/ml) against these three pathogens and Gps. Both P mult. and App were susceptible to enrofloxacin (100% each) and florfenicol (95%). For tetracycline, the highest ratios of resistance could be found in P mult., App and S. suis (21%, 25%, 82%). P mult. and S. suis were susceptible to ampicillin (90%), whereas all B. bronch. isolates (100%) were resistant (r) here. Tulathromycin was effective against P mu/t. and App (s 96%), but not against B. bronch. (r 84%). P mult., B. bronch. and Gps showed high MIC90 values for trimethoprim/sul phonamide. The present data show that none of the antimicrobials can be recommended against all five respiratory pathogens in swine. Clinical therapy should therefore be based on prior identification and susceptibility testing of the causative bacterial pathogens.