Antimicrobial Susceptibility Profiles of Staphylococcus Intermedius Isolates from Clinical Cases of Canine Pyoderma in South Africa

'Antimicrobial susceptibility profiles of Staphylococcus intermedius isolates from clinical cases of canine pyoderma in South Africa', Successful treatment of canine pyoderma has become compromised owing to the development of antimicrobial resistance with accompanying recurrence of infection. Canine skin samples submitted to a veterinary diagnostic laboratory for microbiological culture and sensitivity between January 2007 and June 2010, from which Staphylococcus intermedius was isolated, were selected for this investigation. Antimicrobial resistance of S. intermedius was most prevalent with reference to ampicillin followed by resistance to tetracycline and then potentiated sulphonamides. In general, antimicrobial resistance was low and very few methicillin-resistant isolates were detected. Temporal trends were not noted, except for ampicillin, with isolates becoming more susceptible, and potentiated sulphonamides (co-trimoxazole), with isolates becoming more resistant. In general, both the Kirby–Bauer disc diffusion and broth dilution minimum inhibitory concentration tests yielded similar results for the antimicrobial agents tested. The main difference was evident in the over-estimation of resistance by the Kirby–Bauer test for ampicillin, co-trimoxazole, penicillin and doxycycline. Knowledge of trends in bacterial resistance is important for veterinarians when presented with canine pyoderma. Analysis of antimicrobial susceptibility profiles of S. intermedius isolated from canine pyodermas will guide veterinarians' use of the most appropriate agent and encourage prudent use of antimicrobials in companion animals.

• My supervisor, Dr Jackie Picard, for her patience, enthusiasm, motivation, advice and supervision in the planning and execution of this study.
• Prof Moritz van Vuuren for his assistance during the final stages of the project.
• My parents, for enabling me to pursue this degree and for their unconditional belief in and support of me.
• My husband, for his continual love and patience and, for always having faith in my abilities.
• The rest of my family, friends and colleagues for their constant assistance and encouragement.
intermedius was greatest to ampicillin followed by tetracycline and then potentiated sulphonamides.The results also showed that, in general, antimicrobial resistance was low.Very few methicillin resistant isolates were detected.Temporal trends were not noted, with the exception of ampicillin where isolates became more susceptible and potentiated sulphonamides (co-trimoxazole) where isolates were becoming more resistant.
Staphylococcus intermedius is significantly less resistant to erythromycin, clindamycin, cephalexin, oxacillin, amoxicillin-clavulanic acid, enrofloxacin, marbofloxacin and gentamicin, with most strains being susceptible to these drugs.Resistance to penicillin and tetracycline is frequently found in Staphylococcus intermedius and is on the increase.Resistance to most other antimicrobials, particularly newer generation antimicrobial agents such as the fluoroquinolones, is still comparatively low.In general, both the Kirby-Bauer and broth dilution MIC tests yielded similar results for the antimicrobial agents tested.The main difference between the two tests was evident in the over-estimation of resistance by the Kirby-Bauer test in the cases of ampicillin, co-trimoxazole, penicillin and doxycycline.This could be related to the instability of these particular drugs in vitro.Inoculum densities may also have played a role, with denser inocula producing smaller zone sizes for the drugs tested.Using the MIC method, all of the isolates tested were found to be completely sensitive to ticarcillin, oxacillin, amoxicillinclavulanic acid, imipenem, ceftiofur, chloramphenicol, doxycycline, gentamicin, amikacin and co-trimoxazole.Of the isolates tested using the MIC method, between 2-40% showed some level of resistance to the following antimicrobials: erythromycin, penicillin, ampicillin, enrofloxacin, clindamycin and marbofloxacin.The highest level of resistance observed was shown to erythromycin.The increase in resistance to the lincosamides, lincomycin, clindamycin and erythromycin may be attributed to the increased use of these drugs in the last decade.
Knowledge of trends in bacterial resistance is important for veterinarians when determining treatment for canine skin infections.The information obtained from the analysis of the antimicrobial susceptibility profiles of Staphylococcus intermedius isolated from canine pyoderma cases will provide veterinarians with valuable information on choosing the most appropriate drug to treat S. intermedius skin infections as well as re-enforcing the need for the prudent use of antimicrobial drugs in companion animals.Africa.The results from this study will be specific to South Africa and, since resistance is often based on usage patterns, which are largely regional in nature, these results cannot necessarily be extrapolated to other countries.

LIST OF ABBREVIATIONS
Knowledge of trends in bacterial resistance is important for veterinarians when determining treatment for canine skin infection.The information obtained from the analysis of the antimicrobial susceptibility profiles of Staphylococcus intermedius isolated from canine pyoderma cases will provide veterinarians with valuable information on choosing the most appropriate drug to treat S. intermedius skin infections as well as allowing for the prudent use of antimicrobial drugs in companion animals.

Aims
• To investigate the antimicrobial susceptibility profiles of Staphylococcus intermedius isolates from clinical cases of canine pyoderma in South Africa.
• To compare the Kirby-Bauer and broth microdilution methods of antimicrobial susceptibility testing.

Objectives
• To determine if there is a correlation between the antimicrobial profiles obtained, the age, sex, breed of dog as well as the geographical distribution of the isolates.
• To determine if there is any correlation between the results from the Kirby-Bauer and broth microdilution methods of antimicrobial susceptibility testing with particular reference to methicillin resistance testing.

Background
The staphylococci are facultatively anaerobic, Gram-positive, coccal-shaped bacteria (Figure 1) which belong to the family Micrococcaceae (Rich 2005).They are mostly harmless commensals of the skin and mucous membranes but are potentially pathogenic to man and many other animal species (Vanni et al. 2009).intermedius as part of their normal microflora.It is a transient inhabitant of the skin and hair coat.Reservoir sites include the oral and nasal cavities as well as the perineum and anus (Hartmann et al. 2005).Although Staphylococcus intermedius is not usually isolated from humans as it shows a host-specificity for canine corneocytes, transmission between humans and their pets has been shown to occur (Fitzgerald 2009).
Dogs affected by pyoderma usually have histories of multiple episodes of pyogenic and pruritic skin infections, usually associated with poor response to corticosteroids and partial response to antimicrobial drugs.Lesions may involve the superficial epidermis, dermis and hair follicles and may manifest as papules or pustules.Deeper infections are far more severe, with resulting destruction of the hair follicle and deep dermis and subcutaneous tissue invasion.The characteristics of deep pyoderma are cellulitis, furunculosis, fistulous tracts and skin ulceration (Morales et al. 1994).This skin condition typically results from an underlying skin disorder such as ectoparasitism, atopy, hormonal imbalances or immune-mediated dermatitis (Vanni et al. 2009).

Antimicrobial Resistance
Since the introduction of antimicrobials, the staphylococci have shown a rapid development and spread of resistance, particularly in nosocomial infections (Werckenthin et al. 2001).Antimicrobial resistance is of steadily increasing concern, in both veterinary and human medicine.Its development and spread has lead to treatment failures and hence increased morbidity, mortality and treatment costs (Pellerin et al. 1998).Increased attention has been devoted to small animal welfare which has resulted in increased expenditure on veterinary care.
Consequently, antimicrobial agents are frequently utilised in pet animals, including preparations licensed for human use and the treatment of human infections (Guardabassi 2004b).The treatment of deep pyoderma associated with S. intermedius is possibly the most common reason for antimicrobial treatment in dogs (Guardabassi et al. 2004a).This therapy is generally of an empirical nature.The following guidelines are widely accepted by small animal practitioners (DeBoer 2006): • The use of penicillin, ampicillin, amoxicillin and tetracycline is not indicated as resistance to these drugs is widespread; • Potentiated sulphonamides, erythromycin, lincomycin or clindamycin are reasonable choices as many isolates are still susceptible to these drugs; • Cephalosporins, amoxicillin-clavulanic acid and penicillinase-resistant penicillins such as oxacillin are good choices as nearly all strains are susceptible; • Aminoglycosides and fluoroquinolones should be reserved for more serious infections where other options are not available.Hoekstra and Paulton (2002) reported that resistance in S. intermedius was commonly observed for penicillin G, lincomycin, tetracycline and trimethoprim-sulphamethoxazole.They also found that adult male dogs were worse affected than juvenile female dogs.This could be due to the fact that they carry greater numbers of resistant S.
intermedius strains.Similarly, Hartmann et al (2005) recognised that resistance to penicillin and tetracycline was most common while most isolates tested were sensitive to erythromycin, clarithromycin and clindamycin.Pellerin et al (1998) established that more than 95% of the S. intermedius strains tested were susceptible to oxacillin, amoxicillin-clavulanic acid, cephalexin, gentamicin, fucidic acid, enrofloxacin and marbofloxacin.Vanni et al (2009) reported that resistance was observed mainly against macrolides, chloramphenicol and lincosamides.
The method of prescribing drugs without the use of microbiological culturing and sensitivity testing is most likely a major contributor to the emergence of resistant strains of staphylococci (Lilenbaum et al. 2000).Antimicrobial resistance is, however, a very complex problem involving various bacterial species, resistance mechanisms, transfer mechanisms and reservoirs (Guardabassi et al. 2004b).
Antimicrobial resistance occurs when an organism is resistant to one or more antimicrobials.It can also be acquired by spontaneous mutation or the attainment of resistance genes from another organism via conjugation, transduction or transformation (Cohn & Middleton 2010).The most well-known mechanism of resistance in staphylococci is through the production of various beta-lactamase enzymes.These are in turn classified as penicillinases and cephalosporinases.These enzymes may be encoded by chromosomes or plasmids and may be constitutive or inducible (Pellerin et al. 1998).One of the most problematic resistance mechanisms of recent times is methicillin resistance.Methicillin resistance refers to strains that are resistant to all beta-lactam antimicrobial agents, and can cause a spectrum of infection and disease, from subclinical carriage to lifethreatening focal infections and bacteraemia, It is mediated by vectors which are called staphylococcal cassette chromosome (SCC) elements containing the mecA gene which in turn encodes a penicillin-binding protein which has a broad affinity for beta-lactam antibiotics.These SCCs can be spread horizontally between staphylococci leading to a rapid spread of resistance (Epstein et al. 2009).
Awareness and monitoring of antimicrobial resistance in veterinary staphylococcal isolates is of great importance as the development of resistance in animal pathogens can result in treatment failure in individual patients which may pose a zoonotic risk to their owners.Moreover, increases in resistance to antimicrobial classes that are important in human medicine may result in the restriction of precious antibacterial agents from veterinary use (Loeffler et al. 2007).

Antimicrobial Susceptibility Testing
The in vitro susceptibility of a pathogen to an antimicrobial agent can be performed by disk diffusion or by the measurement of the minimum inhibitory concentration (MIC), the lowest drug concentration capable of inhibiting the growth of the bacterium under investigation.Susceptibility testing is controlled for incubation in or on suitable media, atmosphere, temperature and incubation duration (Blondeau 2009).
The Kirby-Bauer disc diffusion method is a flexible and relatively inexpensive technique, which is commonly used in diagnostic laboratories.The standard procedure is used mainly to test rapidly growing aerobic bacteria.
Filter paper discs containing specified amounts of antimicrobial agents are placed on agar uniformly seeded with the test bacterium.The diameter of each zone of inhibition is measured in millimetres and the results compared with standards for interpretation of the zone size.Susceptibility to an antimicrobial drug indicates that the infection caused by the bacterium may respond to treatment if the drug reaches therapeutic levels in the affected tissues (Quinn et al. 2002).The disk diffusion technique has been standardised primarily for the testing of rapidly growing bacteria.The test has, however, been modified to allow testing of certain fastidious bacteria.The diameter of the zone of inhibition is influenced by the rate of diffusion of the antimicrobial agent through the agar, which may vary among different drugs.The zone size is however inversely proportional to the MIC.Criteria currently recommended for interpreting zone diameters and MIC results for commonly used antimicrobial agents are published by the Clinical and Laboratory Standards Institute (Murray et al. 1999, CLSI 2008).
In the broth microdilution test, drugs are added to a growth medium in a 96-well microtitre plate and serially diluted to the lowest drug concentration to be tested.After the addition of the test organism, the plate is incubated for 18-24 hours.The lowest drug concentration visibly preventing growth is recorded as the MIC.The antimicrobial susceptibility is subsequently determined by comparing the MIC value to established breakpoints that take the following into account: the drug's in vitro activity, achievable and sustainable drug concentrations within the host, distribution and elimination data and drug toxicity (Blondeau 2009).Dilution methods offer flexibility in the sense that the standard medium used to test frequently encountered organisms can be supplemented or replaced with another medium to allow for accurate testing of various fastidious bacteria not reliably tested by disk diffusion.Dilution methods are also adaptable to automated methods.The flexibility of dilution testing is also evident in the reporting formats that may be used.Quantitative results (in micrograms per millilitre) or category results (susceptible, intermediate and resistant) or both can be used (Murray et al. 1999).
Due to its greater stability, oxacillin has been widely used for the in vitro detection of staphylococcal resistance to the class of penicillinase-resistant penicillins (Bemis et al. 2009).Oxacillin susceptibility tests can be applied to other penicillinase-stable penicillins e.g.cloxacillin, dicloxacillin, flucloxacillin, methicillin and nafcillin.Oxacillin is more resistant to degradation and is therefore more likely to detect heteroresistant staphylococcal strains compared to methicillin or nafcillin (CLSI 2008).Most veterinary laboratories use the disk diffusion or broth microdilution methods for oxacillin susceptibility testing.Commercial microdilution test panels do, however, not always contain a sufficient range of oxacillin concentrations to detect low levels of oxacillin resistance (Bemis et al. 2009).Because of the heterogeneous nature of methicillin resistance, no single best method for susceptibility testing exists.The methods used in the clinical laboratory to detect methicillin resistant staphylococci are empirically derived and may be several steps removed from the detection of the genetic or biochemical determinants associated with methicillin resistant strains.This undoubtedly accounts for some of the confusion encountered in determining which strains are resistant and which are not (Hackbarth & Chambers 1989).
Tests for mecA e.g.PCR or the protein encoded by mecA (the penicillin-binding protein 2a) are the most accurate methods for prediction of resistance to oxacillin/methicillin.These tests could be used to confirm results of staphylococcal isolates from serious cases of infection (CLSI 2008).

Conclusion
In order to detect early changes in Staphylococcus intermedius sensitivity patterns before a high prevalence of resistance is selected or developed, regular monitoring of antimicrobial resistance in companion animals is necessary.Prudent use of antimicrobial drugs to treat Staphylococcus intermedius infections should be employed in conjunction with laboratory culture and susceptibility testing, especially if a dog has received prior antimicrobial therapy.

Null (H O ) and Alternative (H A ) Hypotheses
• H O : There is no correlation between the antimicrobial profiles obtained, the age, sex, breed of dog as well as the geographical distribution of the isolates.
• H A : There is a correlation between the antimicrobial profiles obtained, the age, sex, breed of dog as well as the geographical distribution of the isolates.
• H O : There is no correlation between the results from the Kirby-Bauer and broth microdilution methods of antimicrobial susceptibility testing with particular reference to methicillin resistance testing.
• H A : There is a correlation between the results from the Kirby-Bauer and broth microdilution methods of antimicrobial susceptibility testing with particular reference to methicillin resistance testing.

Sampling
All canine skin samples submitted to Vetdiagnostix Veterinary Pathology Services for microbiological culture and sensitivity testing between January 2007 and June 2010, and from which Staphylococcus intermedius was isolated, were selected for this investigation.A total of 319 samples from dogs of various ages, sexes and breeds from various locations throughout South Africa yielded Staphylococcus intermedius and were therefore included in this study.Duplicates were not excluded.
These samples included skin swabs, skin biopsies, skin abscess and pustule swabs and fine needle aspirates.

Identification of Staphylococcus intermedius
Following growth on 5% sheep blood agar (Appendix A) after 24 hours incubation at 37˚C, Staphylococcus intermedius was identified on the basis of colony characteristics (Figure 2), catalase production, Gram's stain,

Antimicrobial Susceptibility Tests
The bacteria identified as Staphylococcus intermedius were tested for antimicrobial susceptibility by the disk diffusion method on Mueller Hinton agar (Appendix A).A 0.5 McFarland suspension of the test organism in sterile saline was evenly spread onto the Mueller Hinton agar plates, after which the plates were disked and incubated at 37˚C for 18-24 hours (CLSI, 2008).All media was quality controlled using Staphylococcus aureus ATCC 25923.
The isolates collected from June 2009 to June 2010 were tested against the following additional antimicrobial agents: amikacin (AK, 30 µg), ceftiofur (EFT, 30 µg), clindamycin (DA, 2 µg), doxycycline (DO, 30 µg), erythromycin (E, 15 µg), imipenem (IPM, 10 µg), marbofloxacin (MAR, 5 µg), penicillin (PEN, 10 IU), rifampicin (RD, 5 µg) and ticarcillin (TIC, 75 µg) (Appendix A).After measuring the zones of inhibition, the strains were classified as sensitive or resistant to the drugs tested according to the Clinical and Laboratory Standards Institute standards and criteria (Table 1) (CLSI 2008).The intermediate strains were interpreted as sensitive or resistant depending on whether the reading obtained fell closer to the sensitive or resistant cut-off.The MIC results for the above-mentioned antimicrobial drugs were determined for the Staphylococcus intermedius isolates collected from June 2009 to June 2010.The laboratory does not routinely make use of the MIC method and therefore retrospective data was not available.The broth microdilution method was utilised and commercial COMPAN1F Sensititre MIC plates were purchased for this purpose (Trek Diagnostics).Table 2 shows the dilution ranges used.A 0.5 McFarland suspension of the test organism was prepared in sterile saline and 10 µl of this suspension was added to 990 µl of cation-adjusted Mueller Hinton broth (Appendix A). 0.1ml of chilled calcium and magnesium ion stock solutions were added per litre for each desired increment of 1mg/l in the final concentration in the adjusted Mueller-Hinton broth (CLSI 2008).100µl of this inoculated broth suspension was then pipetted into each of the 96 wells on the commercial microtitre plate and the plate incubated at 37˚C for 18-24 hours.

Additional Data
In addition to the antimicrobial susceptibility data, data relating to dog signalment (age, gender, breed), case history and any other aspects of the history provided e.g.diagnosis, sample type and geographical location of the patient were also collected.

Statistical Analyses
Microsoft Excel was used to calculate, compare and graph the different parameters measured e.g.percentage susceptibility, age, breed, sex and geographical locations of the practices.The statistics used are descriptive using numerator/ denominator data.

Gender
The genders tended to be equally distributed throughout the years, with almost equal proportions of affected males and females present (Table 4 and Figure 5).

Breed
The Bull Terrier and Shepherd type dogs were grouped separately as they were over-represented and are known to be prone to pyoderma.As shown in Table 5 and Figure 6, the large short haired group was consistently worse affected throughout the years sampled, followed by Bull Terrier types.In 2010, small short haired breeds were worse affected compared to the numbers between 2007 and 2009.

Temporal Distribution
The monthly distribution, with the exception of 2009, tended to be consistent throughout the years (Table 6 and   Figure 7).

Geographical Distribution
The samples tended to be mainly from practices located in the KwaZulu-Natal province of South Africa (Figures [8][9][10][11].

Susceptibility Testing
As noted in Table 8 and Figure 12, antimicrobial resistance of S. intermedius was greatest to ampicillin followed by tetracycline and then potentiated sulphonamides.The results also showed that, in general, antimicrobial resistance was low.Very few methicillin resistant isolates were detected.Temporal trends were not noted, with the exception of ampicillin where isolates became more susceptible and potentiated sulphonamides (cotrimoxazole) where isolates were becoming more resistant.In general, both the Kirby-Bauer and broth dilution MIC tests yielded similar results for the antimicrobial agents tested (Table 8 and Figure 13).Table 9 depicts the percentage MIC distribution of the isolates for each dilution as well as the MIC 50 (median value) and MIC 90 .The percentage resistance represented in the table was based on published breakpoints (CLSI, 2008).Using the MIC method, all of the isolates tested were found to be completely sensitive to ticarcillin, oxacillin, amoxicillin-clavulanic acid, imipenem, ceftiofur, chloramphenicol, doxycycline, gentamicin, amikacin and co-trimoxazole.Of the isolates tested, between 2-40% showed some level of resistance to the following antimicrobials: erythromycin, penicillin, ampicillin, enrofloxacin, clindamycin and marbofloxacin.The highest level of resistance was shown to erythromycin.The shaded areas indicate the concentration range tested for each antimicrobial used.Blue shading indicates that the MIC was within the tested range while red shading indicates resistance to the particular antimicrobial, based on the various breakpoints used (CLSI, 2008).The concentration range for Amoxicillin-clavulanic acid was 4/2 -32/16(µg/ml).The concentration range for Co-Trimoxazole was 0.5/9.5 -2/38(µg/ml).

DISCUSSION, CONCLUSION AND RECOMMENDATIONS
The antimicrobial resistance predicament in human medicine has brought to light various aspects of the use of these substances in animals.There is, however, little useful data on antimicrobial resistance and usage in companion animals (Prescott et al., 2002).The purpose of this study was to investigate the antimicrobial resistance patterns in Staphylococcus intermedius isolates from canine pyoderma cases in South Africa.
Staphylococcus intermedius is a significant cause of juvenile and adult canine pyoderma.Chronic and recurrent pyoderma is a complex syndrome involving cell-mediated hypersensitivity, endocrine disorders and genetic predisposition.It responds poorly to antibiotic therapy alone.Staphylococcal pustular dermatitis occurs in neonates and adults under conditions of poor hygiene.It responds readily to antimicrobial therapy (Quinn et al., 1994).Pyoderma is one of the commonest causes of canine skin disease worldwide.The canine stratum corneum is thinner and more compact with intercellular spaces that are permeated with less protective emulsion when compared to other species.Canine hair follicle infundibula are also relatively unprotected (Siugzdaite et al., 2008).
Various aspects of the cases studied were considered such as age, gender, breed, temporal distribution, geographical distribution and the susceptibility profiles of the isolates.The Kirby-Bauer and broth microdilution methods were also compared.

Age
The number of skin samples yielding Staphylococcus intermedius was high in dogs up to the age of 6 years and, then decreased with only a few cases in dogs aged 11 years or older (Table 3 and Figure 4).The distribution of samples collected in 2010 was unusual in that there were two peaks, one in dogs two years old or less and one in 6 to 9 year-old dogs.This may just be related to the numbers of cases sampled in 2010 compared to the other years.With the exception of skin samples taken in 2008, a high percentage of dogs were under the age of one year old.2008 had a lower percentage of affected dogs less than one year of age compared to the other years.
In the literature, many authors have indicated that staphylococcal pyoderma is more common in dogs below the age of five years.Dogs above this age are less likely to contract this condition (Siugzdaite et al., 2008).This is largely in agreement with the results found in this study.

Gender
The genders tended to be equally distributed throughout the years, with almost equal proportions of affected males and females present (Table 4 and Figure 5).According to Holm et al. (2002), pyoderma is more common in males.Most authors, however, reported no statistically significant difference between the number of affected males and females (Siugzdaite et al., 2008).

Breed
The Bull Terriers and Shepherds were grouped separately as they were over-represented and are known to be prone to pyoderma.The large short haired group was consistently worst affected throughout the years sampled, followed by the Bull Terrier types.In 2010, small short haired breeds were worse affected compared to the numbers between 2007 and 2009 (Table 5 and Figure 6).This may just be related to the numbers of cases sampled in 2010 compared to the other years.Firm conclusions cannot, therefore, be made for this reason, The literature revealed some controversy with regards to breed disposition to this disease.Some authors stated that pyoderma is more common in long-haired breeds while others maintain that short-haired breeds were worse affected.The opinion of yet others was that hair length does not affect the occurrence of skin diseases (Siugzdaite et al., 2008).

Temporal Distribution
The monthly distribution, with the exception of 2009, tended to be consistent throughout the years (Table 6 and Figure 7).The unusual distribution in 2009 could be associated with a general distribution in the number of samples compared to the other years sampled.The literature presented limited data about the manner in which bacterial disease is affected by seasonal changes.Holm et al. (2002) reported that seasonal changes do not have any significant influence which is in line with the findings of this study.

Geographical Distribution
The samples tended to be mainly from practices located in the KwaZulu-Natal province of South Africa (Figures [8][9][10][11].This is most likely due to the fact that the laboratory is located in this province, with a courier network that arranges collection from these practices.Practices in other regions send their samples to the laboratory via private courier companies and the South African Post Office.Therefore, no conclusions can be drawn as to whether dogs in certain areas of South Africa are more prone to pyoderma than those in other areas.

Susceptibility Testing
Antimicrobial resistance of S. intermedius was greatest to ampicillin followed by tetracycline and then potentiated sulphonamides.The results also showed that, in general, antimicrobial resistance was low.Very few methicillin resistant isolates were detected.Temporal trends were not noted, with the exception of ampicillin where isolates became more susceptible and potentiated sulphonamides (co-trimoxazole) where isolates were becoming more resistant (Table 8 and Figure 12).Pellerin et al. (1998) and Hartmann et al. (2005) similarly showed that resistance was most commonly observed to penicillin, tetracycline and sulphamethoxazole-trimethoprim (cotrimoxazole).These drugs should not be used without prior susceptibility testing (Aarestrup, 2006).
Staphylococcus intermedius is less resistant to erythromycin, clindamycin, cephalexin, oxacillin, amoxicillinclavulanic acid, enrofloxacin, marbofloxacin and gentamicin, with most strains being susceptible to these drugs (Pellerin et al., 1998).Werckenthin et al. (2001) found that resistance to penicillin and tetracycline is frequently found in Staphylococcus intermedius and is on the increase.Resistance to most other antibiotics, particularly newer generation antimicrobial agents such as the fluoroquinolones, is still comparatively low.

Kirby-Bauer and MIC Data
In general, both the Kirby-Bauer and broth dilution MIC tests yielded similar results for the antimicrobial agents tested (Table 8 and Figure 13).The main difference between the two tests was evident in the over-estimation of resistance by the Kirby-Bauer test in the cases of ampicillin, co-trimoxazole, penicillin and doxycycline.This could be related to the instability of these particular drugs in vitro.Inoculum densities may also have played a role, with denser inocula producing smaller zone sizes for the drugs tested.
The Kirby-Bauer method remains as a convenient, low-cost means of conducting antimicrobial susceptibility tests and is widely used in veterinary laboratories.The test provides qualitative results that categorise isolates as susceptible, intermediate or resistant.Almost all veterinary-specific agents are available in the antimicrobialimpregnated disks.However, low-volume veterinary-specific agents may only be available from the pharmaceutical manufacturer.Smaller veterinary laboratories may have difficulties standardising the inoculum used in this method, however, commercial systems are available for this purpose (Aarestrup, 2006)., 2008).Using the MIC method, all of the isolates tested were found to be completely sensitive to ticarcillin, oxacillin, amoxicillin-clavulanic acid, imipenem, ceftiofur, chloramphenicol, doxycycline, gentamicin, amikacin and co-trimoxazole.Of the isolates tested, between 2-40% showed some level of resistance to the following antimicrobials: erythromycin, penicillin, ampicillin, enrofloxacin, clindamycin and marbofloxacin.The highest level of resistance was shown to erythromycin.The genes which confer erythromycin resistance in canine staphylococci are almost exclusively ermB genes.The increase in resistance to the lincosamides, lincomycin, clindamycin and erythromycin may be attributed to the increased use of these drugs in the last decade (Pellerin et al., 1998).
The MIC method may be performed in a variety of ways.This method provides a quantitative value as well as a categorisation of the organism as susceptible or resistant.Standardised methods for testing more fastidious organisms such as anaerobes and Campylobacter species have been developed.The MIC method is preferred for use in surveillance or epidemiological investigations as it allows for calculation of summary statistics.Of the various MIC formats used, the broth microdilution method is most widely used and is available in a variety of commercial systems as either dry or frozen panels.It permits testing of a wide range of antimicrobials on a small scale.The disadvantages of these MIC panels are that these panels are inflexible unless the laboratory is willing to bear the cost of custom panels.Not all veterinary-specific agents are available on all panels.Laboratories involved in surveillance programmes or epidemiological studies usually prefer to test a smaller number of antimicrobial agents for an extended number of dilutions.Many diagnostic laboratories choose to use a breakpoint panel.Breakpoint panels allow the laboratory to test a larger number of compounds with dilution ranges spanning the interpretive criteria or breakpoints for each agent (Aarestrup, 2006).

Conclusion and Recommendations
Antimicrobial resistance patterns largely reflect changing fashions in the use of antimicrobial drugs.Reports of resistance from diagnostic laboratories often represent treatment failures rather than treatment successes as animals that have been previously treated will be more likely to yield resistant bacteria than those that have not.
The problem of antimicrobial resistance in bacterial pathogens in dogs should be much less serious than in those isolated from humans.This is because dogs are less likely to be exposed to antimicrobials, other than for short, sporadic periods.They are also less likely to be hospitalised.Chronically ill dogs may be euthanized and immune-compromised animals are not normally treated with broad-spectrum and very potent drugs, like imipenem (Prescott et al., 2002).Staphylococcus intermedius isolates from dogs have increased in resistance to some drugs while decreasing in resistance to others.These changes reflect the changing patterns in the use of individual drugs over time.Antimicrobial resistance is not yet at a crisis stage in canine medicine but there have been warning signs.More information is needed on antimicrobial resistance and its molecular basis in canine medicine.
Bacteria that exhibit antimicrobial resistance will always be a part of our lives.The resistance dilemma in human medicine has brought to light how community-acquired resistance can arise in a relatively short time period.Data pertaining to antimicrobial resistance and the way in which drugs are used in companion animal practice is still relatively hard to come by.There needs to be a continual process of improving prudent use guidelines in companion animal practice so that resistance problems can be avoided.It would be extremely beneficial if an international agreement could be reached amongst veterinarians on a simple but effective approach to prudent companion animal drug use.There should be a concerted effort by veterinary hospitals and companion animal practices to develop formal guidelines for the usage of antimicrobials.Active and effective infection control programmes need to be implemented in veterinary hospitals in order to minimise the spread of resistant organisms or their resistance genes.Veterinary clinical microbiologists also need to agree on standards for monitoring and reporting resistance in companion animal bacteria.This data needs to be used in conjunction with data on antimicrobial drug usage.The limitations of the available data also need to be understood and improved on.
Ann Blunt declare that this dissertation is my own work, carried out originally under the supervision of Dr. Jackie Picard of the University of Pretoria and is in accordance with the requirements of the University for the degree Magister Scientiae (Veterinary Tropical Diseases).Prof. M. van Vuuren served as supervisor during the final stage of the project.
Abbreviation Meaning µg Micrograms µl Microlitres AK Amikacin AMC Amoxicillin-clavulanic acid AMP Ampicillin ATCC American type culture collection C Chloramphenicol CLSI Clinical Laboratory Standards Institute CN Gentamicin DA Clindamycin DNA Deoxyribose nucleic acid DO Doxycycline E Erythromycin EFT Ceftiofur ENR Enrofloxacin ermB Gene conferring erythromycin resistance IPM Imipenem KF Cephalothin MAR Marbofloxacin mecA Gene conferring methicillin resistance MIC Minimum inhibitory concentration MIC 50 Median MIC value MIC 90 MIC required to inhibit the growth of 90% of organisms ml Millilitres mm Millimetres N Neomycin OX Oxacillin PEN Penicillin

Figure 1 :
Figure 1: Phagocytosed coccal-shaped staphylococci in a Diff-Quik® stained smear prepared from a canine skin sample (100X) (Picture: C. Blunt) Nearly all cases of pyoderma in dogs are caused by Staphylococcus intermedius (DeBoer 2006).The name Staphylococcus intermedius was proposed for those isolates which differed from Staphylococcus aureus in various biochemical reactions and in cell wall composition (Hajek 1976).The name Staphylococcus pseudintermedius is now given to the canine-specific strain of Staphylococcus intermedius.Staphylococcus intermedius is also an important cause of wound infections, otitis externa, cystitis, ocular and respiratory disease.Comparable to Staphylococcus aureus colonisation of humans, healthy dogs frequently carry S.
lack of pigment production, delayed acid production from D-mannitol, slow or weak maltose production and positive DNase reaction (Figure3) on DNase agar (Appendix A) as outlined in Clinical Veterinary Microbiology(Quinn, Carter, Markey & Carter 1994).All media used was quality controlled using Staphylococcus aureus ATCC 25923.The name Staphylococcus pseudintermedius is now given to the canine-specific strain of Staphylococcus intermedius.This can only be determined accurately by DNA sequencing.For the purposes of this study, the isolates will be referred to as Staphylococcus intermedius, as phenotypic typing cannot reliably distinguish the canine-specific species from others.

Figure 5 :
Figure 5: Gender distribution of dogs sampled between 2007 and 2010.

Figure 8 :
Figure 8: Overview of the geographical distribution of Staphylococcus intermedius strains isolated in 2007.

Figure 9 :
Figure 9: Geographical distribution of Staphylococcus intermedius strains isolated in 2008

Figure 12 :
Figure 12: Percentage susceptibility of Staphylococcus intermedius isolates from canine skin samples between 2007 and 2010.Breakpoint zone of inhibition diameters derived from CLSI, 2008.

Figure 13 :
Figure 13: Comparison of percentage susceptibility of Kirby-Bauer and broth dilution MIC data between 2007 and 2010

Table 8 :
Comparison of percentage susceptibility of

Table 2 :
Antimicrobial dilution ranges used on the COMPAN1F sensititre plates

Table 3 and
Figure 4show that the number of skin samples yielding Staphylococcus intermedius were high in dogs up to the age of 6 years and, then decreased with only a few cases in dogs aged 11 years or older.Of the 319 samples that yielded Staphylococcus intermedius, data related to age of the patients was only available for 289 of the cases.The distribution of samples collected in 2010 is unusual in that there are two peaks, one in dogs two years old or less and one in 6 to 9 year-old dogs.With the exception of skin samples taken in 2008, a high percentage of dogs were under the age of one year old.2008 had a lower percentage of affected dogs less than one year of age compared to the other years.

Table 3 :
Percentage age distribution of dogs from which skin samples yielded Staphylococcus intermedius isolates between 2007 and 2010 Figure 4: Age distribution of dogs from which skin samples yielded Staphylococcus intermedius isolates between 2007 and 2010.

Table 4 :
Gender distribution of dogs sampled between 2007 and2010.

Table 6 :
Monthly distribution of Staphylococcus intermedius isolates between 2007 and 2009.

Table 7 :
Percentage susceptibility of Staphylococcus intermedius isolates from canine skin samples between 2007 and 2010

Table 9 :
Percentage MIC distribution of Staphylococcus intermedius strains isolated between June 2009 and June 2010

Table 9
depicts the percentage MIC distribution of the isolates for each dilution as well as the MIC 50 (median value) and MIC 90 .The percentage resistance represented in the table was based on published breakpoints (CLSI