Monitoring of florfenicol susceptibility among bovine and porcine respiratory tract pathogens collected in Germany during the years 2002 and 2003

Sir, The fluorinated chloramphenicol derivative florfenicol is exclusively used since veterinary medicine. In Europe, it has been licensed since 1995 for the control of bacterial respiratory tract infections in cattle due to Pasteurella multocida, Mannheimia haemolytica, or Haemophilus somnus and since 2000 for the therapy of respiratory tract infections in pigs due to P. multocida or Actinobacillus pleuropneumoniae. The aim of this study was to investigate whether the most recently collected bovine and porcine respiratory tract pathogens from Germany have changed in their florfenicol susceptibility patterns since the introduction of florfenicol into clinical veterinary use. For this, respiratory tract pathogens were collected on the basis of one isolate per herd in 2002 and 2003, respectively. All isolates were obtained from nasal swabs or lung tissue samples of animals suffering from respiratory tract infections. As far as information on antimicrobial pre-treatment was available, the animals from which samples had been taken had not been pre-treated with any antimicrobial agents within the weeks before sample collection. A total of 595 isolates was included in this study: 95 bovine P. multocida, 98 M. haemolytica, 103 porcine P. multocida, 63 A. pleuropneumoniae, 131 Bordetella bronchiseptica, and 105 Streptococcus suis isolates. All isolates were biochemically identified to species level1 and comparatively tested for their in vitro susceptibilities to florfenicol by two different methods, agar disc diffusion and broth microdilution. Discs charged with 30 μg florfenicol (Becton Dickinson, Heidelberg, Germany) were used for disc diffusion, and microtitre plates (Sensititre, Westlake, OH, USA) containing florfenicol concentrations of 0.12–128 mg/L in serial two-fold dilutions for the broth microdilution method. Both susceptibility tests were carried out and evaluated according to the recommendations given in the National Committee for Clinical Laboratory Standards (NCCLS) document M31-A2.2 The four reference strains Escherichia coli ATCC 25922, Staphylococcus aureus ATCC 29213, S. aureus ATCC 25923 as well as A. pleuropneumoniae ATCC 27090 were used for quality control purposes. The florfenicol-specific breakpoints presented in the NCCLS document M31-A22 are valid for bovine P. multocida, M. haemolytica and H. somnus isolates. These breakpoints have recently also been accepted by the Antimicrobial Susceptibility Testing Subcommittee of the NCCLS not only for porcine P. multocida and A. pleuropneumoniae isolates, but also for B. bronchiseptica and S. suis isolates from porcine respiratory tract infections. For all bacteria included in this study, the approved florfenicol-specific breakpoints for susceptibility are zone diameters of ≥19 mm and MIC values of ≤2 mg/L. Those for classification as intermediate are 15–18 mm and 4 mg/L, and those for resistance are ≤14 mm and ≥8 mg/L.2 MIC50 and MIC90 values were calculated as the florfenicol concentrations that inhibited growth of 50% and 90% of the isolates, respectively. Table 1 shows the results of in vitro susceptibility testing of the isolates from 2002/2003 in comparison to the results from earlier studies conducted in Germany.3–6 All bovine P. multocida and M. haemolytica isolates collected during 2002/2003 were found to be susceptible to florfenicol by both methods. A comparison of the data from 2002/2003 with data from previous years revealed that the most recent bovine P. multocida and M. haemolytica isolates had the same MIC50 and MIC90 values as the isolates collected in the pre-approval phase in 1993/19944 (Table 1). A comparison between the data collected at the time of approval of florfenicol for use in cattle5 and those from 2002/2003 also revealed no increases in the MIC50 and MIC90 values of bovine P. multocida and M. haemolytica isolates (Table 1). This comparison showed that after 8 years of veterinary use, bovine P. multocida and M. haemolytica isolates from respiratory tract infections in Germany are still susceptible to florfenicol and that so far no resistance development has been detected in these target bacteria. A similar situation was seen among the porcine pathogens (Table 1). All porcine P. multocida, A. pleuropneumoniae, and S. suis isolates collected in 2002 and 2003 were susceptible to florfenicol. The S. suis isolates exhibited the same MIC50 and MIC90 values as those isolates collected at the time of introduction of florfenicol into use in pigs or shortly thereafter.3 Changes of not more than one dilution step in the MIC50 or MIC90 values were recorded when comparing the porcine P. multocida and A. pleuropneumoniae isolates collected in 2002 and 2003 with those collected either in the pre-approval period or at the time of florfenicol approval for pigs.3,5,6 A surprising development was seen among the B. bronchiseptica isolates. During the years 2000–2003, the number of resistant isolates decreased from 25.6% in 2000/2001 to 0.8% in 2002/2003. Moreover, the percentage of B. bronchiseptica isolates classified as ‘intermediate’ dropped from 40.0% in 2000/2001 to 6.9% in 2002/2003. The reasons for this development towards increased susceptibility are unknown. However, it is likely that the temporal occurrence and disappearance of B. bronchiseptica isolates that exhibit varying levels of susceptibility/resistance is responsible for this development. Monitoring of porcine B. bronchiseptica isolates in the next few years will reveal whether this low level of resistance will remain stable in the future. There are currently very few studies available which allow the assessment of the actual susceptibility situation of bovine and porcine respiratory tract pathogens against antimicrobial agents in Germany. Therefore, continuous monitoring programmes for such veterinary pathogens are indispensable tools to determine the in vitro antimicrobial susceptibility status. The comparison of data from different years also permits the early recognition of trends in the development of resistance, especially against antimicrobial agents, such as florfenicol, which have been introduced more recently into clinical veterinary use. For such relatively new and highly potent antimicrobial agents, it is particularly important to see how the susceptibility of the target bacteria develops in the post-approval period. The data presented in this and other studies3–6 confirmed that the MIC values of the P. multocida, M. haemolytica, A. pleuropneumoniae and S. suis isolates have remained stable since the mid-1990s and—as a consequence—that no resistance development has occurred since then. Reliable data from monitoring programmes need to be based on a representative test population, standardized sample collection and susceptibility testing according to an internationally accepted methodology. In addition, monitoring programmes in the veterinary field should also contain data on overall antimicrobial use, e.g. classes of antimicrobial agents applied, quantities, duration of application, etc. Such data are at least in part already available from selected monitoring programmes, such as the DANMAP (www.dfvf.dk) or the SVARM (www.sva.se) programmes, and represent useful supplementary information that might help to explain trends in the development of bacterial susceptibility/resistance. Florfenicol susceptibility data for bovine and porcine respiratory tract pathogens ND, not determined. Susceptibility testing followed the recommendations of DIN 58940; approved breakpoints were not available at the time of testing, classification of the isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done in retrospect on the basis of the MIC data and the currently available NCCLS-approved breakpoints. Susceptibility testing followed the recommendations of NCCLS document M31-A2 and classification of the bovine isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done on the basis of the MIC data and the NCCLS-approved breakpoints. Susceptibility testing followed the recommendations of the NCCLS document M31-A2; approved breakpoints for porcine pathogens were not available at the time of testing, classification of the isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done in retrospect on the basis of the MIC data and the currently available NCCLS-approved breakpoints. Florfenicol susceptibility data for bovine and porcine respiratory tract pathogens ND, not determined. Susceptibility testing followed the recommendations of DIN 58940; approved breakpoints were not available at the time of testing, classification of the isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done in retrospect on the basis of the MIC data and the currently available NCCLS-approved breakpoints. Susceptibility testing followed the recommendations of NCCLS document M31-A2 and classification of the bovine isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done on the basis of the MIC data and the NCCLS-approved breakpoints. Susceptibility testing followed the recommendations of the NCCLS document M31-A2; approved breakpoints for porcine pathogens were not available at the time of testing, classification of the isolates as ‘resistant’, ‘intermediate’ or ‘susceptible’ was done in retrospect on the basis of the MIC data and the currently available NCCLS-approved breakpoints. Microtitre plates and discs were kindly provided by Schering-Plough. We thank Vivian Hensel for excellent technical assistance. 1Institut für Tierzucht, Bundesforschungsanstalt für Landwirtschaft (FAL), Höltystr. 10, 31535 Neustadt-Mariensee; 2Institut für Tiergesundheit, LUFA Nord-West, Hannover; 3Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL), Berlin; 4Institut für Mikrobiologie, Zentrum für Infektionsmedizin, Stiftung Tierärztliche Hochschule Hannover, Hannover; 5Auβenstelle für Epidemiologie, Stiftung Tierärztliche Hochschule Hannover, Bakum; 6Thüringer Landesamt für Lebensmittelsicherheit und Verbraucherschutz, Jena, Germany