Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

79

Resumo: O incremento de resistências a compostos antibac-terianos tem-se revelado, ao longo dos anos, alvo de crescente preocupação a nível Mundial, quer do ponto de vista da Saúde Pública, quer na perspetiva da Segurança Alimentar. A mastite bovina é a causa mais frequente de utilização destes fármacos em efectivos leiteiros, sendo o seu uso apontado como factor se-letivo na ecologia bacteriana do úbere bovino. A monitorização e análise dos padrões de resistências antibacterianas em agentes etiológicos de mastite tem fornecido informações de grande utilidade na escolha de antibióticos no tratamento de animais afectados, especialmente no que diz respeito a agentes etiológi-cos isolados em diferentes regiões geográficas.Este estudo retrospectivo teve como objetivo determinar os pa-drões de resistência a sete fármacos antibacterianos disponíveis no mercado Português e a sua evolução, em 47.413 isolados bac-terianos. Os isolados foram obtidos a partir de amostras de leite provenientes de explorações leiteiras das regiões Litoral Norte, Centro e Sul de Portugal Continental, entre 2004 e 2012. Testes de susceptibilidade foram executados pelo método de difusão em disco e procedeu-se à regressão logística de modo a verificar alterações nos níveis de resistência ao longo do tempo de estudo.Os resultados obtidos indicam uma evolução nos níveis de re-sistência, ao longo dos nove anos do estudo, entre certos agentes testados: no S. aureus para amoxicilina/ácido clavulânico (de 0,7% para 26,6%), cloxacilina (de 1,1% para 3,0%), cefquinoma (de 0% para 3.1%) e gentamicina (de 1,2% para 2,9%); na E. coli para amoxicilina/ácido clavulânico (de 22,6% para 55,0%), cefazolina (de 4,9% para 24,4%) e trimetoprim/sulfametoxazol (de 9,6% para 14,3%); na K. pneumoniae para amoxicilina/ácido clavulânico (de 57,1% para 71,4%); no S. uberis para gentamic-ina (de 41,5% para 88,2%); e no S. dysgalactiae (de 3,2% para 11,8%) e S. agalactiae (de 0% para 4,3%) para cloxacilina.

Palavras-Chave: Bovinos; Mastite; Antibióticos; Padrões; Resistências.

Summary: The worldwide development of resistance to anti-bacterial agents has been a growing concern over the years, from both public health and food safety perspectives. Bovine mastitis reports as the single most common pathology in dairy herds, and therefore, to which these drugs are mostly used. This use ac-counts as a selective factor in the bacterial ecology of the bovine

udder. The analysis and long-term monitoring of antibacterial susceptibility patterns has provided useful information in guid-ing the selection of antibacterial agents by local clinicians from different geographic regions, towards more efficient mastitis therapy protocols.This retrospective study aims to determine the evolution of re-sistance patterns of the major mastitis pathogens to seven an-tibacterial drugs available in the Portuguese market over time. Isolates were obtained from milk samples from dairy farms in the northwestern, central and southern regions of Portugal, between 2004 and 2012. Susceptibility testing was performed by the Kirby-Bauer disk diffusion method. Data from a total of 47,413 antibacterial susceptibility tests was used. Logistic re-gression was performed to verify whether the resistance propor-tions to the various antibacterials changed over the nine years of data.Results indicated a trend towards increased antibacterial resist-ance among certain mastitis pathogens: for S. aureus to amoxi-cillin/clavulanic acid (from 0.7% to 26.6%), cloxacillin (from 1.1% to 3.0%), cefquinome (from 0% to 3.1%) and gentamicin (from 1.2% to 2.9%); for E. coli to amoxicillin/clavulanic acid (from 22.6% to 55.0%), cefazolin (from 4,9% to 24.4%) and trimethoprim/sulfamethoxazole (from 9.6% to 14.3%); for K. pneumoniae to amoxicillin/clavulanic acid (from 57.1% to 71.4%); for S. uberis to gentamicin (from 41.5% to 88.2%); and for S dysgalactiae (from 3.2% to 11.8%) and S. agalactiae (from 0% to 4.3%) to cloxacillin.

Keywords: Dairy; Cattle; Bovine; Mastitis; Antibacterial agent;

Resistance; Patterns; Trends.

Introduction

The increase in resistance to antibacterial agents has raised serious concerns worldwide from both public health and food safety perspectives, putting their use in food-producing animals under constant scrutiny over the years (WHO, 1997, 1998; Agersø et al., 2011).

Bovine mastitis is the single most frequent cause for antibacterial use in dairy herds (White and McDermott, 2001; Hillerton and Berry, 2005; USDA, 2008; Agersø et al., 2011). Since this use has been suggested as a

Trends in Antibacterial Resistance of Major Bovine Mastitis Pathogens in Portugal

Evolução de Padrões de Resistência a Antibióticos em Agentes Etiológicos da Mastite Bovina em Portugal

Balbino Rocha1* , Denisa Mendonça2,3 , João Niza-Ribeiro2,3

1FAGRICOOP – Cooperativa Agrícola e dos Produtores de Leite de Vila Nova de Famalicão, Portugal2ICBAS – Instituto de Ciências Biomédicas de Abel Salazar, Universidade do Porto, Portugal – Departamento de Estudos População

3ISPUP – Instituto de Saúde Pública da Universidade do Porto, Portugal

*Correspondência: balbino.rocha@gmail.comTelef: 969 888 767

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

80

selective force in shaping the bacterial ecology of bo-vine mastitis (Myllys et al., 1994), antibacterial resist-ance of mastitis pathogens has received continuous particular attention (WHO, 1997, 1998; White and McDermott, 2001; Agersø et al., 2011).

Development of antibacterial resistance in mas-titis pathogens discloses three relevant aspects: 1) Reduction in treatment efficacy, disease control and productivity. In some instances, dairy cattle that devel-op mastitis caused by resistant bacteria may become chronically infected and be prematurely dispatched to slaughter; 2) Potential risk of transmission of resistant bacteria to humans via the food chain. This is, however, not likely to occur with milk from clinical mastitis cas-es, since this milk is banned from human consumption. Nonetheless, clinical cases may turn into subclinical cases or latent infections. Resistant bacteria from these infections are present in the bulk tank milk and may be transmitted to humans via raw milk products; and 3) Potential risk of transmission of resistance genes between mastitis and other environmental pathogens, which may affect humans through other routes (Owens et al., 1997; Khachatourinas, 1998; Sol et al., 2000).

Individual antibacterial susceptibility tests of mas-titis pathogens, based on the results of the agar disk diffusion method, have minimal clinical relevance in guiding individual cow therapy. However, since these results are replicable, large numbers of mastitis isolates may be processed to generate antibacterial resistance patterns of mastitis pathogens to selected antibacterial drugs, consequently providing useful information on the long-term trends of those traits in a certain popula-tion or region. Deliberations may therefore be made in the direction of guiding local clinicians to make more informed choices when selecting for the most suitable antibacterial drug to prevent and treat mastitis. This may contribute to a more successful therapeutic out-come, reducing the use of antibacterials and the selec-tion pressure among mastitis pathogens and their dis-semination in the environment (Constable and Morin, 2003).

The goal of this study was to establish, from a clinical perspective, antibacterial resistance patterns of major bovine mastitis pathogens isolated from milk samples from dairy herds in northwestern, central and southern Portugal in a nine-year period (2004 – 2012), and de-termine if those patterns have changed over time.

Materials and Methods

Criteria for selection of cases

This study consisted in reviewing records of all bac-teriological outcomes from clinical and subclinical mastitis milk samples from dairy cattle of Portuguese northwestern, central and southern herds. Milk samples were forwarded to an Animal Health and Food Safety Laboratory (Segalab, S.A. - Matosinhos, Portugal)

between January 2004 and September 2012. Results from antibacterial susceptibility testing were included for analysis.

Sample collection and microbiology

Over the years, mastitic milk sample submissions to the laboratory have been made as part of either oc-casional private individual initiatives by farmers or a herd’s assisting veterinarian, or by means of milk qual-ity programs established and performed by the labora-tory’s technical services to herds taking part in these programs. Milk samples were collected and submit-ted to the laboratory under National Mastitis Council (NMC) recommendations (NMC, 2004). In the labora-tory, milk samples were cultured and mastitis patho-gens identified using standard microbiologic methods. Contaminated samples were defined, by NMC guide-lines, as a combination of three or more isolated dis-similar colony types (NMC, 2004). Once identified, pure cultures of mastitis pathogens were tested for in vitro antibacterial susceptibility.

In vitro antibacterial susceptibility testing

In vitro antibacterial susceptibility testing was con-ducted by the Kirby-Bauer disk diffusion test meth-odology in accordance with Clinical and Laboratory Standards Institute (CLSI) standards (CLSI, 2008). Each isolate was added to sterile diluents to contain approximately 108 CFU/mL (0,5 on McFarland scale) and was plated on Mueller-Hinton agar (bioMérieux®), with or without supplementation of 5% defibrinated sheep blood, depending on the isolate’s genera. Disks impregnated with the tested antibacterial agents were placed over the agar and incubated at 37 ºC for 24 hours. Susceptibility data were determined by meas-urement of the inhibition zone around the antibacte-rial disks, according to the zone diameter interpretative CLSI standards (CLSI, 2008). E. coli - ATCC isolate 25922 and S. aureus - ATCC isolate 33862 were used as the quality control organisms. Only quality controlled results were reported. Isolates were classified as sus-ceptible, of intermediate susceptibility or resistant on the basis of CLSI standards (CLSI, 2008). Laboratory protocols, based on NMC guidelines, remained practi-cally unchanged during the study period. Antibacterial disks and/or manufacturers changed sporadically, de-pending on stock/market availability.

Tested antibacterials

The antibacterials considered for analysis were: Amoxicillin/clavulanic acid (AUG), 30 mg (20 mg + 10 mg); cloxacillin (OB/CX), 5 mg; penicillin G (P), 10 IU; cefazolin (KZ), 30 mg; cefquinome (CEQ), 10 mg; gentamicin (CN), 10 mg; and trimethoprim/sulfameth-oxazole (SXT), 25 mg (1,25 mg + 23,75 mg) (Table 1). The selection of the antibacterials was based on the fol-

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

81

lowing criteria: 1) License, availability and sales in the Portuguese market for mastitis intramammary (IMM) therapy in lactating cows (Apifarma, 2011; DGAV et al., 2012); 2) Use in human medicine (cefquinome as an exception); 3) Listed as “critically important” or “highly important” antibacterial in human medicine (cefquinome as an exception) (AGISAR, 2011).

Table 1 - Antibacterial agents used in 47,413 susceptibility tests of bacterial pathogen isolates obtained from dairy cow milk sam-ples and submitted for bacterial culture between January 2004 and September 2012.

Pathogens isolated Antibacterial agents tested

Staphylococcus aureus AUG, OB/CX, P, KZ, CEQ, CN and SXT

Streptococcus uberis AUG, OB/CX, P, KZ, CEQ, CN and SXT

Streptococcus agalactiae AUG, OB/CX, P, KZ, CEQ and SXT

Streptococcus dysgalactiae AUG, OB/CX, P, KZ and SXT

Enterococcus spp. * AUG, OB/CX, P, KZ and SXT

Escherichia coli AUG, OB/CX, P, KZ, CEQ, CN and SXT

Klebsiella pneumoniae AUG, OB/CX, P, KZ, CN and SXT

* includes only Enterococcus faecium and Enterococcus faecalis species.AUG - Amoxicillin/Clavulanic acid; OB/CX - Cloxacillin; P - Penicillin G; KZ - Cefazolin; CEQ - Cefquinome; CN - Gentamicin; SXT - Trimethoprim/Sulfamethoxazole.

Selection of pathogens

Only pure cultures of Staphylococcus aureus, Streptococcus agalactiae, Streptococcus uberis, Streptococcus dysgalactiae, Enterococcus spp. (E. faecium and E. faecalis only), Escherichia coli and Klebsiella pneumoniae were considered for analysis (Table 1). The selection of these pathogens was based on the following criteria: 1) classification as major mastitis pathogen; 2) importance for veterinary medi-cine and dairy industry and, 3) importance for public health.

Data analysis

For purposes of statistical analysis, isolates classified as being of intermediate susceptibility were not includ-ed in the study. Also, E. faecium and E. faecalis iso-lates were merged into the ‘Enterococcus spp.’ group. The proportion of tested pathogens that showed resist-ance to an individual antibacterial agent was calculat-ed for the nine years of data (Table 3). The resulting resistance levels were grouped in 7 categories, based on European Food Safety Authority (EFSA) standards: 1) Rare: <0.1%; 2) Very low: 0.1 to 1%; 3) Low: 1 to 10%; 4) Moderate: 10 to 20%; 5) High: 20 to 50%; 6) Very high: 50% to 70%; and 7) Extremely high: >70% (EFSA, 2010). Logistic regression was performed to determine the probability of antibacterial resistance by year (Table 4). The logistic regression model included ‘resistance’ as the response variable (“yes” vs. “no”) and ‘year’ as the explanatory variable (“2004” to “2012”). P-values of 0.05 were considered significant. Statistical analysis was performed using IBM SPSS Statistics, version 21.0.

Results

A total of 47,413 antibacterial susceptibility test results was included in the present study: 28,136 on S. aureus isolates; 5,916 on E. coli isolates; 5,799 on S. uberis isolates; 4,589 on S. agalactiae isolates; 1,231 on S. dysgalactiae isolates; 979 on Enterococcus spp. isolates; and 773 on K. pneumo-nia isolates (Table 2).

The proportion of tests that were found to be resist-ant to the selected antibacterial agents ranged from 0 to 100% (Table 3). Resistance proportions were, in majority, below 15%. S. agalactiae and S. dysga-lactiae were the pathogens with the lowest resistance proportions (0 to 11.4%, between both). Antibacterial resistance was prevalent at high levels among: S. au-

Table 2 - Number of antibacterial susceptibility tests throughout the study period (2004 - 2012).

Staphylococcus aureus

Escherichia coli

Streptococcus uberis

Streptococcus agalactiae

Streptococcus dysgalactiae

Enterococcusspp. *

Klebsiella pneumoniae

Total Total (%)

2004 1,491 127 604 159 137 199 56 2,773 5.9

2005 2,715 150 426 278 132 205 15 3,921 8.3

2006 3,766 364 298 620 87 84 131 5,350 11.3

2007 4,866 805 633 792 124 199 137 7,556 15.9

2008 5,070 935 925 586 323 119 76 8,034 16.9

2009 4,051 978 983 456 155 NT 138 6,761 14.3

2010 2,602 1,086 874 894 273 173 134 6,036 12.7

2011 2,138 846 580 376 NT NT 65 4,005 8.4

2012 1,427 625 476 428 NT NT 21 2,977 6.3

Total 28,126 5,916 5,799 4,589 1,231 979 773 47,413

Total (%) 59.3 12.2 9.7 2.6 2.1 12.5 1.6 100.0

* includes only Enterococcus faecium and Enterococcus faecalis species.NT - Not tested

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

82

reus for penicillin (44.7%); S. uberis for cloxacillin (22.1%) and gentamicin (71.0%); Enterococcus spp. for trimethoprim/sulfamethoxazole (23.9%), penicillin (70.0%), cefazolin (79.9%) and cloxacillin (96.2%); E. coli for amoxicillin/clavulanic acid (30.3%), penicillin (99.4%) and cloxacillin (99.5%); and K. pneumoniae for amoxicillin/clavulanic acid (37.4%), cloxacillin (100%) and penicillin (100%).

Logistic regression analysis (Table 4) determined that penicillin was the only antibacterial whose re-

sistance levels had no significant change over time. On the other hand, the other six tested antibacterial agents revealed an increasing trend in some of their resistance patterns.

The resistance levels of amoxicillin/clavulanic acid had a significant (p<0.001) increase over time to S. aureus (from 0.7% in 2004 to 26.6% in 2012), E. coli (from 22.6% in 2005 to 55.0% in 2012), and K. pneumoniae (from 57.1% in 2004 to 71.4% in 2012) (Table 4, Figure 1).

Table 3 - Resistance proportions of major mastitis bacterial pathogens, sorted by antibacterial agent.

S. aureus(n = 28,126)

S. uberis(n = 5,799)

S. agalactiae(n = 4,589)

S. dysgalactiae(n = 1,231)

Enterococcus spp. (n = 979)

E. coli(n = 5,916)

K. pneumoniae(n = 773)

N.º tested

R (%)N.º

testedR (%)

N.º tested

R (%)N.º

testedR (%)

N.º tested

R (%)N.º

testedR (%)

N.º tested

R (%)

AUG 5,387 12.3 1,309 0 1,108 0 307 0 257 3.9 1,292 30.3 198 37.4

OB/CX 5,348 2.9 1,257 22.1 1,093 11.4 307 5.5 263 96.2 214 99.5 58 100

P 4,834 44.7 517 0.4 404 0.2 130 0.8 50 70.0 333 99.4 73 100

KZ 3,844 0.5 1,225 1.0 1,029 0.1 307 0.7 254 79.9 1,224 9.6 189 8.5

CEQ 2,450 1.3 441 0.2 460 0.7 NT – NT – 707 8.2 NT –

CN 5,242 2.2 169 71.0 NT – NT – NT – 1,179 3.6 157 1.3

SXT 1,021 0.7 881 3.1 495 1.4 180 3.8 155 23.9 967 13.1 98 6.1

AUG - Amoxicillin/Clavulanic acid; OB/CX - Cloxacillin; P - Penicillin G; KZ - Cefazolin; CEQ - Cefquinome; CN - Gentamicin; SXT - Trimethoprim/Sulfamethoxazole; NT - Not tested; R - Resistance; Enterococcus spp. - includes only Enterococcus faecium and Enterococcus faecalis species.All proportions over 20% (‘High’) in bold.

Table 4 - Results of logistic regression analysis that determined, for the antibacterial susceptibility tests, whether resistance proportions to the various antibacterial agents changed with year.

AUG OB/CX P KZ CEQ CN SXT

OR(95% CI)

pvalue

OR(95% CI)

pvalue

OR(95% CI)

p value

OR(95% CI)

p value

OR(95% CI)

pvalue

OR(95% CI)

pvalue

OR(95% CI)

pvalue

S. aureus1.42

(1.36-1.48)1

< 0.0011.19

(1.10-1.28)1

< 0.0010.99

(0.97-1.03)1

0.7970.99

(0.76-1.32)2

0.9931.29

(1.09-1.54)1

0.0041.15

(1.06-1.26)1

0.0010.80

(0.45-1.43)3

0.453

S. agalactiae 1.00 1 1.0001.45

(1.22-1.73)21

< 0.0011.09

(0.46-2.58)4

0.8422.86

(0.45-18.22)1

0.2651.58

(0.71-3.54)1

0.264 NT NT0.93

(0.65-1.35)6

0.709

S uberis 1.00 1 1.0000.98

(0.90-1.06)20

0.5960.76

(0.35-1.66)7

0.4901.13

(0.85-1.50)1

0.416 0 8 0.9883.15

(2.09-4.73)9

< 0.0010.98

(0.82-1.17)1

0.825

S. dysgalactiae 1.00 10 1.0001.45

(1.04-2.00) 10

0.0280.62

(0.13-3.00)11

0.5480.83

(0.43-1.58)10

0.562 NT NT NT NT1.03

(0.65-1.63)10

0.906

Enterococcusspp. *

0.89(0.66-1.22)12

0.4751.04

(0.77-1.40)12

0.8231.08

(0.68-1.71)11

0.7411.10

(0.96-1.27)12

0.180 NT NT NT NT1.19

(0.95-1.49)12

0.126

E. coli1.50

(1.39-1.62)13

< 0.001 0 9 0.9940.32

(0.04-2.51)14

0.2811.18

(1.07-1.31)1

0.0010.83

(0.67-1.01)15

0.0640.99

(0.85-1.17)16

0.9391.13

(1.01-1.26)15

0.034

K. pneumonia1.35

(1.15-1.58)17

< 0.001 0 9 1.000 0 14 1.0001.14

(0.88-1.47)1

0.315 NT NT0.83

(0.46-1.51)18

0.5451.12

(0.57-2.19)19

0.749

AUG - Amoxicillin/Clavulanic acid; OB/CX - Cloxacillin; P - Penicillin G; KZ - Cefazolin; CEQ - Cefquinome; CN - Gentamicin; SXT - Trimethoprim/Sulfamethoxazole; NT - Not tested; OR - Odds ratio.Significant changes (p<0.05) in bold.1 All years included (2004-2012); 2 2004-2011; 3 All years included, except 2006, 2009, 2010 and 2012; 4 All years included, except 2008-2011; 5 All years included, except 2008; 6 All years included, except 2004, 2009 and 2011; 7 All years included, except 2009-2011; 8 All years included, except 2005, 2008 and 2009; 9 2004-2007; 10 2004-2010; 11 2004-2008; 12 All years included, except 2009, 2011 and 2012; 13 2005-2012; 14 2005-2008; 15 2007-2012; 16 All years included, except 2006; 17 All years included, except 2005; 18 All years included, except 2005 and 2006; 19 2006-2010; 20 All years included, except 2009; 21 All years included, except 2009 and 2011.

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

83

The resistance levels of cloxacillin had a significant (p<0.001) increase over time to both S. aureus (from 1.1% in 2004 to 3.0% in 2012) and S. agalactiae (from 0% in 2004 to 4.3% in 2011) (Table 4, Figure 2). A significant increase (p=0.028), from 3.2% in 2004 to 11.8% in 2012, was also verified in the resistance levels of this antibacterial to S. dysgalactiae (Table 4, Figure 2).

As for gentamicin, the resistance levels to S. aureus had an overall significant increase (p=0.001), from 1.2% in 2004 to 2.9% in 2012 (Table 4, Figure 3). A

significant increase (p<0.001) was also verified to S. uberis for this antibacterial, from 41.5% in 2004 to 88.2% in 2007 (Table 4, Figure 3).

E. coli resistance levels were the only to have a significant increase (p=0.001), from 4.9% in 2004 to 24.4% in 2012, for cefazolin (Table 4, Figure 4), as did S. aureus for cefquinome (p=0.004), from 0% in 2004 to 3.1% in 2012 (Table 4, Figure 5). The resistance lev-els of trimethoprim/sulfamethoxazole had a significant (p=0.034) increase over time to E. coli (from 9.6% in 2007 to 14.3% in 2012) (Table 4, Figure 6).

Figure 1 - Proportions of S. aureus (n = 28,126 tests), E. coli (n = 5,916 tests) and K. pneumoniae(n = 773 tests) resistant to amoxicillin/clavulanic acid (AUG) along the study period.

Figure 2 - Proportions of S. aureus (n = 28,126 tests), S. agalactiae (n = 4,589 tests) and S. dysgalactiae (n = 1,231 tests) resistant to cloxacillin (OB/CX) along the study period.

Figure 3 - Proportions of S. aureus (n = 28,126 tests) and S. uberis (n = 5,799 tests)resistant to gentamicin (CN) along the study period.

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

84

Figure 4 - Proportions of E. coli (n = 5,916 tests) resistant to cefazolin (KZ) between January 2004 and September 2012.

Figure 5 - Proportions of S. aureus (n = 28,126 tests) resistant to cefquinome (CEQ) between January 2004 and September 2012.

Figure 6 - Proportions of E. coli (n = 5,916 tests) resistant to trimethoprim/sulfamethoxazole (SXT) between January 2007 and September 2012.

Discussion

Parallelisms may be established between this study and available relevant research worldwide as to antibac-terial resistance patterns of mastitis pathogens and the long-term trends of those traits (Erskine et al., 2002; Rossitto et al., 2002; San Martín et al., 2003; Rajala-Schultz et al., 2004; Roesch et al., 2006; Ebrahimi et al., 2007; Nunes et al., 2007b; Pol and Ruegg, 2007;

Bengtsson et al., 2009; Nam et al., 2009a; Nam et al., 2009b; Botrel et al., 2010; Suriyasathaporn, 2010; Ka-lmus et al., 2011; Persson et al., 2011; Sahebekhtiari et al., 2011). A critical approach should, however, be en-gaged when comparing and/or extrapolating any simi-larities or discrepancies between these due to differenc-es in the origin of isolates, laboratory procedures, inter-pretive guidelines, among other factors (De Oliveira et al., 2000; Gentilini et al., 2000; Erskine et al., 2002).

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

85

To the authors’ knowledge, this study is among those with the largest assembled data and the most extensive time-frame, compiling nine years of infor-mation (2004 – 2012). To date, the most widespread available data derived from a 7-year study by Erskine et al. (Erskine et al., 2002) and a 6-year study by Nam et al. (Nam et al., 2009a). The majority of studies has conducted their research over shorter time-frames (e.g., 6 months to 3 years) (San Martín et al., 2003; Rajala-Schultz et al., 2004; Roesch et al., 2006; Pol and Ruegg, 2007; Bengtsson et al., 2009; Botrel et al., 2010; Kalmus et al., 2011; Persson et al., 2011).

In Portugal, though some studies have established antibacterial susceptibility patterns for mastitis path-ogens, they all have used reduced sample sizes and none has determined the trends for those patterns for the study period (Nunes et al., 2007a; Nunes et al., 2007b; Rato et al., 2012).

Although laboratory protocols changed little dur-ing the study period, allowing for a consistent analy-sis in antibacterial resistance patterns, the results of this study were still limited due to the nature of the available data. For instance, the variable year was the only factor to be included in the logistic regression model to explain the changes in the resistance propor-tions of the tested antibacterials. It is quite obvious to the authors that factors other than year influenced those changes and would therefore need to be consid-ered in order to create a model that explains the data more accurately. Another limitation of the study was that several antibacterial agents were not tested in a consistent manner throughout the study period. This was because the antibacterial agents were employed depending on the type of pathogen isolated, clients’ request, and on laboratory/market availability of their respective diffusion disks.

The Kirby-Bauer disk diffusion methodology, inex-pensive and practical, was used in this study. The pri-mary disadvantage of using this method when moni-toring the development of resistance is the qualitative basis of outcomes (susceptible, intermediate or resist-ant). Additionally and similarly to what happens in all laboratories worldwide, human interpretive criteria were used to categorize the results, providing inap-propriate and potentially misleading conclusions. The validity of applying these breakpoints to the treatment of bovine mastitis has not been established and can be questioned because: 1) bovine milk pH, electrolyte, fat, protein, leukocyte concentrations, growth factor composition, and pharmacokinetic profiles are differ-ent than those for human plasma; and 2) human bac-terial pathogens are often different from bovine mas-titis pathogens (Constable and Morin, 2003). Ideally, accurate antibacterial susceptibility test breakpoints should result from: a) Minimum Inhibitory Concen-tration (MIC) data for mastitis pathogens; b) Pharma-cokinetic/pharmacodynamic (PK/PD) data for lactat-ing dairy cows; and c) the results of field studies that

measure the rates of clinical and bacteriologic cure. Clinical and bacteriologic cure rates may provide a clear breakpoint or, in other situations, these data can be used in conjunction with PK/PD data to suggest the most appropriate breakpoint (EUCAST, 2000).

The odds ratio (OR) from the logistic regression analysis can be used to determine the rate at which the prevalence of resistance is increasing or decreas-ing each year. Statistically significant (p<0.05) OR < 1.0 reflect a reduced odds of observing resistance during any given year, compared with the previous year. Statistically significant (p<0.05) OR > 1.0 re-flect an increased odds of observing resistance during any given year, compared with the previous year. For instance, the OR of 1.15 for S. aureus resistance to gentamicin can be interpreted to mean that the likeli-hood that S. aureus would be resistant to gentamicin was 1.15 that of a previous year.

With cloxacillin as an exception, when analysing the resistance patterns of the major mastitis patho-gens for the different tested β-lactam antibacteri-als, all three Streptococcus species (S. agalactiae, S. uberis and S. dysgalactiae) exhibited ‘very low’ re-sistance levels (<1.0%) (Table 3). This information is consistent with the Gram-positive spectrum of these compounds, especially in these species, and is also in agreement with what has been described in literature (Owens et al., 1990; Owens et al., 1997; Erskine et al., 2002; Rossitto et al., 2002; Roesch et al., 2006; Tenhagen et al., 2006; Nam et al., 2009b; Kalmus et al., 2011). Cloxacillin was, in fact, the only β-lactam to display evidence of a significant increasing trend for S. agalactiae and S. dysgalactiae (Table 4). Pos-sible explanations for this trend may be that, since: 1) this β-Lactam is among the most frequently used IMM antibacterials in lactating- and dry-cow therapy in Portugal; 2) these streptococci and Enterococcus spp. are environmental pathogens with similar bio-chemical and structural characteristics; and 3) Ente-rococcus spp. are known to be key reservoirs of anti-bacterial resistance genes – In response to the selec-tive pressure from the use of this antibacterial, genes expressing resistance to cloxacillin in Enterococcus spp. may have been exchanged and consequently dis-seminated amongst these streptococci over time.

The ‘extremely high’ resistance levels (>70%) of Enterococcus spp. (E. faecium and E. faecalis) for penicillin, cefazolin and cloxacillin (Table 3) is in agreement with available literature (Rossitto et al., 2002). This is explained because enterococci exhibit intrinsic resistance to penicillinase-susceptible peni-cillins (low level), penicillinase-resistant penicillins and cephalosporins (Murray, 2000). This is due to low affinity penicillin binding proteins (PBPs), which en-able these pathogens to synthesize cell wall compo-nents even in the presence of modest concentrations of most β-lactam antibacterials. In addition, entero-cocci are tolerant to the activity of β-lactams, that is,

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

86

they are inhibited but not killed by these agents. This property is an acquired characteristic (Murray, 2000). Moreover, strains of E. faecalis exclusively express-ing β-lactamase enzymes and exhibiting high resist-ance proportions to penicillin have been reported (Patterson et al., 1988; Rhinehart et al., 1990; Rice et al., 1991; Wells et al., 1992). These E. faecalis are not susceptible to anti-staphylococcal penicillins but are susceptible to ampicillin, amoxicillin and piperacillin combined with drugs that inhibit penicillinase such as clavulanic acid, sulbactam and tazobactam (Wells et al., 1992; Murray, 2000). This fact may explain the low resistance levels to amoxicillin/clavulanic acid, an outcome consistent with data from other reports (Tenhagen et al., 2006).

The wide variation observed in the resistance levels of S. aureus for the tested β-lactams (Table 3) can similarly be observed in other reports (Erskine et al., 2002; San Martín et al., 2003; Rajala-Schultz et al., 2004; Bengtsson et al., 2009; Kalmus et al., 2011; Persson et al., 2011; Sahebekhtiari et al., 2011). However, this in vitro susceptibility does not guar-antee nor reflect the in vivo treatment success rates. Several factors including the abilities of S. aureus to survive inside neutrophils (Yancey et al., 1991; Mullarky et al., 2001); to form small-colony vari-ants or L-forms (Brouillette et al., 2004); to induce fibrosis and microabscess formation (Ziv and Stor-per, 1985; Sordillo et al., 1989; Erskine et al., 2003); and to invade mammary epithelial cells (Lammers et al., 2000; Kerro Dego et al., 2002) are potential con-tributors to the poor response of chronic S. aureus to antibacterial therapy. When analysing the long-term trends of the resistance proportions of S. aureus to the different β-lactams, a significant increase was verified throughout the study period for amoxicillin/clavulanic acid, cloxacillin and cefquinome (Table 4). These results are not in agreement with other stud-ies, in which resistance of S. aureus to cloxacillin decreased, for example (Makovec and Ruegg, 2003). The authors consider this fact to be of paramount im-portance from both public health and epidemiologi-cal points-of-view and the reasons for these increases and respective points of origin need to be further in-vestigated and determined.

For both Gram-negative pathogens, the resistance levels verified towards the tested β-lactams (Table 3) relate to the fact that these compounds are considered to be more effective against Gram-positive bacteria. Even so, some molecules such as cefquinome (4th-generation cephalosporin) are broad spectrum agents with greater activity against Gram-negative bacteria and this is supported by the fact that this antibacterial presented the lowest resistance proportions. Actually, both of the tested cephalosporins showed the lowest resistance levels among all β-lactams. This was not only true for both Gram-negative pathogens, but for all pathogens in general (Table 3). Still in regard to

E. coli and K. pneumoniae, both pathogens also ex-hibited increases in resistance proportions that were significant for amoxicillin/clavulanic acid and ce-fazolin (E. coli only) (Table 4). Possible explanations for these facts may be the acquisition of plasmids containing genes that encode for extended-spectrum β-lactamases (ESBLs) in these species (Blanco et al., 1997; Davies, 1997; Paterson, 2006).

When analysing the resistance proportions for gen-tamicin and trimethoprim/sulfamethoxazole (Table 3), with the exception of S. uberis (for gentamicin) and Enterococcus spp. (for trimethoprim/sulfameth-oxazole), the resistance levels (<13.1%) were in con-formity with similar studies (Erskine et al., 2002; San Martín et al., 2003; Bengtsson et al., 2009; Botrel et al., 2010; Kalmus et al., 2011; Sahebekhtiari et al., 2011). Possible explanations for the significant in-creases in resistance levels to trimethoprim/sulfam-ethoxazole (for E. coli) and gentamicin (for S. uberis and S. aureus) (Table 4) may be the widespread use of these antibacterials in treatment protocols for gas-trointestinal disorders and other pathologies in cattle.

The observed high resistance levels to all tested an-tibacterials are of extreme importance since these sub-stances have been defined by the WHO as ‘critically important’ (i.e., amoxicillin/clavulanic acid, penicillin and gentamicin) or ‘highly important’ (cloxacillin, ce-fazolin and trimethoprim/sulfamethoxazole) antibacte-rials in human medicine (AGISAR, 2011).

Conclusions

The assessment of the evolution of the resistance patterns over the nine years of study, did indicate a trend towards increased resistance of the selected mastitis pathogens to the following antibacterials: Amoxicillin/clavulanic acid (S. aureus, E. coli and K. pneumoniae), Cloxacillin (S. aureus, S. agalac-tiae and S. dysgalactiae), Gentamicin (S. aureus and S. uberis), Cefazolin (E. coli), Cefquinome (S. au-reus), and trimethoprim/sulfamethoxazole (E. coli). Penicillin was the only antibacterial with no signifi-cant change over time. Enterococcus spp. did not ex-hibit any significant change over time for any of the tested antibacterials.

Further research should be addressed in an effort to find the source of these increases and take actual steps in reversing these trends in these regions.

Acknowledgments

The authors wish to thank Segalab’s departments of technical services and microbiology for providing all data and rendering technical assistance, namely Helena Madeira, Paula Santos, Abigail Barbosa and Marta Barbosa.

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

87

Bibliography

Agersø Y, Hald T, Høg B, Jensen L, Jensen V, Korsgaard H, ... Sørum M (2011). Use of antimicrobial agents and oc-currence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. In: DANMAP 2010 - The Danish Integrated Antimicrobial Resistance Monitoring and Research Programme. Editors: Korsgaard H and Agersø Y (Copenhagen, Denmark), 23-40.

AGISAR (2011). Critically Important Antimicrobials for Human Medicine. In: AGISAR1. WHO Press, (Geneva, Switzerland), 18-26.

Apifarma (2011). Simposium Veterinário Apifarma. In: http://www.apifarma.pt/simposiumvet. Accessed 22-1-2013, 2013.

Bengtsson B, Unnerstad HE, Ekman T, Artursson K, Nilsson-Ost M and Waller KP (2009). Antimicrobial susceptibility of udder pathogens from cases of acute clinical mastitis in dairy cows. Veterinary Microbiology, 136 (1-2): 142-149.

Blanco JE, Blanco M, Mora A and Blanco J (1997). Prevalence of bacterial resistance to quinolones and other antimicrobials among avian Escherichia coli strains isolat-ed from septicemic and healthy chickens in Spain. Journal of Clinical Microbiology, 35 (8): 2184-2185.

Botrel MA, Haenni M, Morignat E, Sulpice P, Madec JY and Calavas D (2010). Distribution and antimicrobial resist-ance of clinical and subclinical mastitis pathogens in dairy cows in Rhone-Alpes, France. Foodborne Pathogens and Disease, 7 (5): 479-487.

Brouillette E, Grondin G, Lefebvre C, Talbot BG and Malouin F (2004). Mouse mastitis model of infection for antimicrobial compound efficacy studies against intracel-lular and extracellular forms of Staphylococcus aureus. Veterinary Microbiology, 101 (4): 253-262.

CLSI (2008). Performance standards for antimicrobial disk and difussion susceptibility tests for bacteria isolated from animals - Approved standard M31-A3. Wayne (PA), USA, Clinical and Laboratory Standards Institute, 28.

Constable PD and Morin DE (2003). Treatment of clini-cal mastitis. Using antimicrobial susceptibility profiles for treatment decisions. The Veterinary Clinics of North America - Food Animal Practice 19 (1): 139-155.

Davies JE (1997). Origins, acquisition and dissemina-tion of antibiotic resistance determinants. In: Antibiotic Resistance: Origins, Evolution, Selection and Spread. Editors: Chadwick DJ and Goode J. John Wiley and Sons Ltd. (Chichester, UK), 5-35.

De Oliveira AP, Watts JL, Salmon SA and Aarestrup FM (2000). Antimicrobial susceptibility of Staphylococcus aureus isolated from bovine mastitis in Europe and the United States. Journal of Dairy Science, 83 (4): 855-862.

DGAV, DRAV and DGAMPV (2012). Relatório Nacional de Monitorização do Consumo de Antimicrobianos - Ano 2011 Portugal. Ministério da Agricultura, Mar, Ambiente e Ordenamento Territorial, 1-12.

Ebrahimi A, Kheirabadi KH and Nikookhah F (2007). Antimicrobial susceptibility of environmental bovine mas-titis pathogens in west central Iran. Pakistan Journal of Biological Sciences, 10 (17): 3014-3016.

EFSA (2010). The Community Summary Report on antimi-crobial resistance in zoonotic and indicator bacteria from animals and food in the European Union in 2004-2007. European Food Safety Authority Journal, 8 (4): 1309.

Erskine RJ, Wagner S and DeGraves FJ (2003). Mastitis ther-apy and pharmacology. The Veterinary Clinics of North America - Food Animal Practice, 19 (1): 109-138, VI.

Erskine RJ, Walker RD, Bolin CA, Bartlett PC and White DG (2002). Trends in antibacterial susceptibility of masti-tis pathogens during a seven-year period. Journal of Dairy Science, 85 (5): 1111-1118.

EUCAST (2000). Determination of antimicrobial sus-ceptibility test breakpoints. Clinical Microbiology and Infection, 6 (10): 570-572.

Gentilini E, Denamiel G, Llorente P, Godaly S, Rebuelto M and DeGregorio O (2000). Antimicrobial susceptibility of Staphylococcus aureus isolated from bovine mastitis in Argentina. Journal of Dairy Science, 83 (6): 1224-1227.

Hillerton JE and Berry EA (2005). Treating mastitis in the cow: A tradition or an archaism. Journal of Applied Microbiology, 98 (6): 1250-1255.

Kalmus P, Aasmae B, Karssin A, Orro T and Kask K (2011). Udder pathogens and their resistance to antimi-

Rocha B. et al. RCPV (2014) 109 (591-592) 79-88

88

Nunes SF, Bexiga R, Cavaco LM and Vilela CL (2007a). Technical note: Antimicrobial susceptibility of Portuguese isolates of Staphylococcus aureus and Staphylococcus epi-dermidis in subclinical bovine mastitis. Journal of Dairy Science, 90 (7): 3242-3246.

Nunes SF, Cavaco LM, Vilela CL and Bexiga R (2007b). Antimicrobial susceptibility traits of subclinical bovine mastitis pathogens in Portugal. Revista Portuguesa de Ciências Veterinárias, 102: 275-280.

Owens WE, Ray CH, Watts JL and Yancey RJ (1997). Comparison of success of antibiotic therapy during lac-tation and results of antimicrobial susceptibility tests for bovine mastitis. Journal of Dairy Science, 80 (2): 313-317.

Owens WE, Watts JL, Greene BB and Ray CH (1990). Minimum inhibitory concentrations and disk diffusion zone diameter for selected antibiotics against streptococci isolated from bovine intramammary infections. Journal of Dairy Science, 73 (5): 1225-1231.

Paterson DL (2006). Resistance in gram-negative bacteria: enterobacteriaceae. The American Journal of Medicine, 119 (6 Suppl 1): S20-28; discussion S62-70.

Patterson JE, Masecar BL and Zervos MJ (1988). Characterization and comparison of two penicillinase-pro-ducing strains of Streptococcus (Enterococcus) faecalis. Antimicrobial Agents and Chemotherapy, 32 (1): 122-124.

Persson Y, Nyman AK and Gronlund-Andersson U (2011). Etiology and antimicrobial susceptibility of udder patho-gens from cases of subclinical mastitis in dairy cows in Sweden. Acta Veterinaria Scandinavica, 53: 36.

Pol M and Ruegg PL (2007). Relationship between an-timicrobial drug usage and antimicrobial susceptibil-ity of gram-positive mastitis pathogens. Journal of Dairy Science, 90 (1): 262-273.

Rajala-Schultz PJ, Smith KL, Hogan JS and Love BC (2004). Antimicrobial susceptibility of mastitis pathogens from first lactation and older cows. Veterinary Microbiology, 102 (1-2): 33-42.

Rato MG, Bexiga R, Florindo C, Cavaco LM, Vilela CL and Santos-Sanches I (2012). Antimicrobial resistance and molecular epidemiology of streptococci from bovine mas-titis. Veterinary Microbiology, 161 (3-4): 286-94.

Rhinehart E, Smith NE, Wennersten C, Gorss E, Freeman J, Eliopoulos GM, ... Goldmann DA (1990). Rapid dissemi-nation of beta-lactamase-producing, aminoglycoside-re-sistant Enterococcus faecalis among patients and staff on an infant-toddler surgical ward. The New England Journal of Medicine, 323 (26): 1814-1818.

Rice LB, Eliopoulos GM, Wennersten C, Goldmann D, Jacoby GA and Moellering RC, Jr. (1991). Chromosomally mediated beta-lactamase production and gentamicin re-sistance in Enterococcus faecalis. Antimicrobial Agents and Chemotherapy 35 (2): 272-276.

Roesch M, Perreten V, Doherr MG, Schaeren W, Schallibaum M and Blum JW (2006). Comparison of antibiotic resist-ance of udder pathogens in dairy cows kept on organic and on conventional farms. Journal of Dairy Science, 89 (3): 989-997.

Rossitto PV, Ruiz L, Kikuchi Y, Glenn K, Luiz K, Watts JL and Cullor JS (2002). Antibiotic susceptibility patterns for environmental streptococci isolated from bovine mastitis in central California dairies. Journal of Dairy Science, 85 (1): 132-138.

Sahebekhtiari N, Nochi Z, Eslampour MA, Dabiri H, Bolfion M, Taherikalani M, ... Emaneini M (2011). Characterization of Staphylococcus aureus strains isolated from raw milk of bovine subclinical mastitis in Tehran and Mashhad. Acta Microbiologica et Immunologica Hungarica, 58 (2): 113-121.

San Martín B, Kruze J, Morales MA, Agüero H, Iragüen D, Espinoza S, ... Borie C (2003). Antimicrobial resistance in bacteria isolated from dairy herds in Chile. International Journal of Applied Research in Veterinary Medicine, 23-27.

Sol J, Sampimon OC, Barkema HW and Schukken YH (2000). Factors associated with cure after therapy of clini-cal mastitis caused by Staphylococcus aureus. Journal of Dairy Science, 83 (2): 278-284.

Sordillo LM, Nickerson SC and Akers RM (1989). Pathology of Staphylococcus aureus mastitis during lactogenesis: re-lationships with bovine mammary structure and function. Journal of Dairy Science, 72 (1): 228-240.

Suriyasathaporn W (2010). Milk quality and antimicrobial resistance against mastitis pathogens after changing from a conventional to an experimentally organic dairy farm. Asian-Australian Journal of Animal Science, 23 (5): 659-664.

Tenhagen BA, Koster G, Wallmann J and Heuwieser W (2006). Prevalence of mastitis pathogens and their re-sistance against antimicrobial agents in dairy cows in Brandenburg, Germany. Journal of Dairy Science, 89 (7): 2542-2551.

USDA (2008). Dairy 2007: Antibiotic use on U.S. dairy operations, 2002 and 2007. United States Department of Agriculture Animal Plant Health Inspection Service National Animal Health Monitoring System, 1-5.

Wells VD, Wong ES, Murray BE, Coudron PE, Williams DS and Markowitz SM (1992). Infections due to beta-lactamase-producing, high-level gentamicin-resistant Enterococcus faecalis. Annals of Internal Medicine, 116 (4): 285-292.

White DG and McDermott PF (2001). Emergence and Transfer of Antibacterial Resistance. Journal of Dairy Science, 84 (E. Suppl.): E151-E155.

WHO (1997). The Medical Impact of the Use of Antimicrobials in Food Animals. WHO Press (Berlin, Germany), 5-15.

WHO (1998). Use of Quinolones in Food Animals and Potential Impact on Human Health: Report and Proceedings of a WHO Meeting. WHO Press (Geneva, Switzerland), 5-12.

Yancey RJ, Sanchez MS and Ford CW (1991). Activity of antibiotics against Staphylococcus aureus within poly-morphonuclear neutrophils. European Journal Of Clinical Microbiology & Infectious Diseases, 10 (2): 107-113.

Ziv G and Storper M (1985). Intramuscular treatment of sub-clinical staphylococcal mastitis in lactating cows with pen-icillin G, methicillin and their esters. Journal of Veterinary Pharmacology and Therapeutics, 8 (3): 276-283.