The data presented in this report originate from samples submitted to diagnostic laboratories in different countries. It was generally impossible to retrieve data on the specific age of the animals and additionally, the specific microbiological methods used to isolate and identify the organisms remained unreported. All participating laboratories are, however, appointed national reference laboratories for antimicrobial resistance.
Information on antimicrobial usage prior to collection of the samples could have indicated why resistance in some countries is higher than in others, and also, it could have indicated a reason for the difference seen between antimicrobials. As information on these key determinants is not available, comparisons and conclusions have to be done with care. Nevertheless, the study provides interesting data on the prevalence of antimicrobial resistance in bacterial pathogens in different European countries.
Although there was a prior agreement between the participating laboratories on a list of antimicrobials of relevance for each bacterial species, most laboratories provided data for different panels of antimicrobial agents. The same heterogeneity was observed in the breakpoints applied. This demonstrates an important problem in performing international monitoring based on data produced by routine diagnostic laboratories using different panels, methods, equipment etc. [6, 7]. We are aware that publishing data from multiple laboratories can be very problematic as there might be variations in methods used, interpretative criteria, etc. In the ideal world all laboratories would use the same methods and we could believe that data were directly comparable. However, since this unfortunately not is the case and there actually are major differences in the methods used we decided to ensure the comparability of the data as means of gaining reliable by conducting an external quality control system and only include data from laboratories and pathogens where the cut-off was met.
In a previous survey of isolates from nine European countries and USA, major variations in the occurrence of resistance between countries were reported . This emphasises that treatment strategies for bovine mastitis caused by S. aureus in the European countries have to be based on local knowledge and available resistance data. In the present study, it seems that the level of resistance in France has decreased between the years 2002 and 2003; for oxacillin (8.3% to 1%), for sulphonamide (59.2% to 2%) and for trimethoprim (11.1% to 0%). It was not possible to detect the same trend in the year 2004 due to the absence of submitted data. Similar trends were only observed in one other country (Switzerland), where a considerable decrease of gentamicin resistance from 30% to 2.5% was observed.
Of major concern is the level of resistance to oxacillin and 3rd generation cephalosporins (i.e. ceftiofur) in S. aureus. The prevalence of oxacillin resistance in Spain (3.7%) and France (8.3%) and the resistance towards cephalosporins in Spain (0.9% in 2004) and France (4.2% in 2002; 1% in 2003) indicate the presence of methicillin resistant S. aureus (MRSA) in these two countries. Most of the Spanish isolates were not multi resistant which usually a feature of MRSA isolated from humans is. Furthermore, the presence of the mecA- gene could not be confirmed since the strains were not stored.
For France the percentages of resistance observed by oxacillin resistant S. aureus isolates must be interpreted with a great deal of precaution. Data on bovine pathogens were collected through a multi centre study (see Table 1). The results could not be verified because some of the laboratories involved in the network did not store the strains. From 2003, all the strains collected in France were investigated specifically for methicillin resistance and a decrease was observed. MRSA are not commonly seen in bovine mastitis but this has recently been reported in Korea  and Hungary . MRSA also seems to be emerging within other livestock species [11, 12].
The decrease in resistance levels to antimicrobials in isolates from the Netherlands to ampicillin, fluoroquinolone, tetracycline, trimethoprim – sulphonamide may partly be explained by sampling bias (sampling of different age groups between years). The same tendency was observed in both England (and Wales) and France to some of the antimicrobials. In France the level of resistance to ampicillin and tetracycline decreased from 2002 to 2003 whereas a decrease in England (and Wales) was observed for ampicillin, amoxicillin – clavulanic acid, tetracycline and trimethoprim – sulphonamide in the period 2003 to 2004. Resistance to fluoroquinolones was detected in France with 5% (2002) and in the Netherlands (enrofloxacin) with 5.8% (2003). Similar results have previously been reported from France  and the Netherlands , whereas a similar or more frequent occurrence of resistance has been observed in USA [15, 16].
Antimicrobial resistance in P. multocida seems to be higher in the Netherlands compared to the other countries (e.g. 20% to tetracycline in 2002) though the amount of resistance shows a tendency to decrease over the years for ampicillin, tetracycline and trimethoprim – sulphonamide (Table 4). The same trend was observed in England (and Wales), France and Denmark to tetracycline and trimethoprim – sulphonamide. It should be noted, that in 2002 in France 1.2% of the isolates were resistant to ceftiofur and 4.7% to fluoroquinolones. Resistance to enrofloxacin in the Netherlands was 6.3% in 2004.
Resistance to trimethoprim – sulphonamide and penicillin seems to be low compared to the levels of resistance for other antimicrobials tested. The percentages of resistance to trimethoprim + sulphonamide (8.7%) and penicillin (3.9%) were the highest values observed in the Netherlands. A relatively high level of resistance could be detected in countries, which submitted data for tetracycline and erythromycin as well as gentamicin (Table 5). Similar data have been reported from USA [17, 18] and France . Thus, bovine mastitis caused by streptococci can be treated with penicillins, whereas the use of macrolides and tetracyclines probably should be avoided on the basis of the resistance data generated by this study.
Resistance to tetracycline and erythromycin was high compared to the other antimicrobials tested. In the Netherlands 67.8% of the isolates were resistant to tetracycline in 2004 and 17% of the isolates from France were resistant to erythromycin in 2003. The isolates from Sweden were susceptible to all antimicrobials tested, except for tetracycline, where the level was 6% (Table 5).
The percentage of resistant isolates increased in some countries, whereas it decreased in others. For instance in France and Italy the resistance levels show a tendency to decrease for most of antimicrobials investigated between the years 2002–2004. In Spain and Denmark the levels seem to increase between the years 2003 and 2004. In general, a relatively high frequency of resistance was observed in almost all countries over the years, with the exception of Sweden and the Netherlands the frequency was more moderate. Resistance to ceftiofur, representing the 3rd generation cephalosporins, was observed in France (2002, 1.2%; 2003, 0.7%), Spain (2002, 5.9%; 2003, 2.6%; 2004, 6%), Italy (2004, 7.1%) and arising in Belgium (2003, 2%; 2004, 6.3%). The frequent occurrence of resistance to more conventional antimicrobials in most of these countries makes cephalosporins a group of drugs that can be used without prior susceptibility testing. Greater than 10% fluoroquinolone resistance was observed in Sweden (enrofloxacin), France, Spain, Belgium, Germany and Italy. Higher rates of resistance towards newer antimicrobials (e.g. fluoroquinolones, aminoglycosides) may be due to the attitudes and practice among European countries in prescribing and administering drugs for therapy of enteric and respiratory tract diseases .
The seemingly emerging occurrence of resistance to the important antimicrobial agents; cephalosporins and fluoroquinolones in Belgium, France, Italy and Spain is worrying. These four countries had in general the highest frequency of resistance to most antimicrobial agents, potentially making treatment difficult.