Frequency of Isolation and Antimicrobial Susceptibility Patterns of Staphylococcus intermedius and Pseudomonas aeruginosa Isolates From Canine Skin and Ear Samples Over a 6-Year Period (1992–1997)

Introduction

Two of the most important bacterial pathogens of canine skin are Staphylococcus intermedius (S. intermedius) and Pseudomonas aeruginosa (P. aeruginosa). Staphylococcus intermedius is the primary bacterial pathogen of the skin of dogs and the most common isolate in canine pyoderma.¹ This organism is also a frequent isolate in otitis externa.²³ Because S. intermedius skin and ear infections are routinely treated with empirically selected antimicrobial agents, assessment of changes in antimicrobial susceptibility patterns over time is important to ensure that the organism remains susceptible to empirically selected antimicrobial agents. Pseudomonas aeruginosa can be isolated from dogs with chronic, deep pyoderma and chronic otitis externa/media.^2–4 Treatment of these infections is often challenging because P. aeruginosa isolates are frequently resistant to numerous antimicrobial agents, including those used empirically.⁴⁵ Because of the multidrug-resistant profile of P. aeruginosa isolates, in vitro antimicrobial susceptibility testing is advised to ensure appropriate antibiotic selection in addition to allowing for monitoring the development of resistant pathogens. The increase in the isolation of multidrug-resistant pathogens is of growing concern in human and veterinary medicine (e.g., oxacillin-resistant S. intermedius and methicillin-resistant Staphylococcus aureus).⁶⁷

The purposes of this retrospective study were to determine whether the frequency of isolation of S. intermedius and P. aeruginosa from bacterial culture samples collected from canine skin or ears changed over a 6-year period and whether the antimicrobial susceptibility patterns of S. intermedius and P. aeruginosa skin and ear isolates changed over the same time period. In addition, antimicrobial susceptibility patterns of skin and ear isolates of each pathogen were compared.

Materials and Methods

Results of 10,811 canine bacterial culture submissions to Michigan State University’s Animal Health Diagnostic Laboratory from 1992 through 1997 were reviewed. Samples yielding S. intermedius, P. aeruginosa, or both, were examined for changes in frequency of isolation from skin and ears and for changes in susceptibility of the isolates to selected antimicrobial agents. The site or skin lesions sampled for culture could not be determined from the computer records but likely included a variety of samples, including swabs of skin surface or pustules as well as macerated tissue. Samples were cultured by conventional methods by plating onto Colombia agar, supplemented with 5% defibrinated sheep-blood agar (EBA), Colistin and Nalidixic acid agar (CNA), MacConkey agar (MAC), and by inoculation of thioglycollate broth, supplemented with 1% hemin and 1% yeast extract^a (THIO). The EBA and CNA plates were incubated in 5% carbon dioxide for 24 and 48 hours. The MAC plate and the THIO plate were incubated in ambient air at 37°C for 24 hours. Staphylococcus-like colonies were identified by colonial morphology, Gram stain, catalase test, and coagulase test. The API STAPH IDENT strip^b was used to confirm identification as S. intermedius. Isolated Pseudomonas-like colonies were identified by Gram stain, colonial morphology, and oxidase test. Depending on the year of the study, one of three methods was used to confirm identification as P. aeruginosa: biochemical tests (i.e., triple sugar iron agar slant, Tech agar, growth on EBA at 42°C), the API 20 NE strip,^c or the Gram Negative Identification Plus card.^d Antimicrobial susceptibilities were reported as susceptible, intermediate, or resistant using the qualitative agar disk diffusion (Kirby-Bauer) test as described by the National Committee for Clinical Laboratory Standards (NCCLS).⁸ Quality control organisms included Escherichia coli, ATCC 25922, S. aureus, ATCC25923, and P. aeruginosa.

Frequencies of positive culture results for S. intermedius and P. aeruginosa from canine skin and ear cultures were compared over all years (1992 through 1997) by chi-square analysis, using a commercially available statistical software package.^e To compare susceptibility results for specific antimicrobial agents between skin and ear isolates of each organism, the proportions of susceptible isolates over the entire 6-year period were analyzed by a z-test. Significant differences were accepted when the value for alpha was <0.05 and the value for beta (power) was >0.8.

Results

Staphylococcus intermedius and P. aeruginosa were isolated from 18.8% and 5.7%, respectively, of all canine samples submitted for bacterial culture during the years 1992 through 1997. Staphylococcus intermedius was isolated from 88.6% and 49.4% of all skin and ear samples, respectively, and frequency of isolation did not change during the 6-year period [Table 1]. Of 497 skin and 553 ear samples yielding S. intermedius, it was the only organism isolated (i.e., a pure culture) in 47.1% and 32.7% of cultures, respectively. Pseudomonas aeruginosa was isolated from 7.5% and 27.8% of all skin and ear samples, respectively, and frequency of isolation from skin samples increased (P<0.04) over the 6-year study period but remained unchanged for ear samples [Table 2]. Of 42 skin and 311 ear cultures yielding P. aeruginosa, it was the only organism isolated in 33.1% and 33.3% of cultures, respectively.

The antimicrobial susceptibility patterns for the S. intermedius isolates from canine skin are shown in Table 3. Only two of the 497 S. intermedius isolates from skin were resistant to both cephalothin and oxacillin, and two different isolates were resistant to ciprofloxacin. No significant differences were found in the percentages of skin isolates susceptible to any of the antibiotics over the 6-year period. The antimicrobial susceptibility patterns for the S. intermedius isolates from canine ears are shown in Table 4. Only three of the 553 ear isolates were resistant to cephalothin and oxacillin, and five ear isolates were resistant to ciprofloxacin. No significant differences were found in the percentages of ear isolates susceptible to cephalothin, erythromycin, gentamicin, oxacillin, trimethoprim/sulfadiazine, clindamycin, or ciprofloxacin over the 6-year period. However, the percentages of isolates susceptible to ampicillin, penicillin, and tetracycline were different (P<0.01 for all three antibiotics) over the 6-year period, with decreased resistance observed toward the end of the study period. When antimicrobial susceptibility of the S. intermedius isolates recovered from skin were compared to that of the ear isolates over the 6-year period, the percentages of skin isolates susceptible to ampicillin, penicillin, trimethoprim/sulfadiazine, tetracycline, and clindamycin were lower (P<0.01) for all.

The antimicrobial susceptibility patterns for the P. aeruginosa isolates from canine skin and ears are shown in Tables 5 and 6. All P. aeruginosa isolates, from both skin and ears, were resistant to ampicillin, cephalothin, trimethoprim/sulfadiazine, and tetracycline. All skin isolates were also resistant to cefotaxime but were susceptible to gentamicin, amikacin, and piperacillin. Only one skin isolate was resistant to ticarcillin, and two skin isolates were resistant to ciprofloxacin. Most ear isolates were susceptible to gentamicin, amikacin, ticarcillin, piperacillin, and ciprofloxacin; but when the aminoglycoside antibiotics were compared, more isolates were resistant to gentamicin than amikacin (P<0.01). Although there appeared to be trends toward increasing susceptibility of ear isolates to gentamicin (from 66% to 81%) and decreasing susceptibility to ticarcillin (from 100% to 82%) and cefotaxime (from 24% to 0%) over the 6-year period, these were not significant findings. Finally, when antimicrobial susceptibility of P. aeruginosa isolates recovered from skin was compared to that of the ear isolates over the 6-year period, the percentages of ear isolates susceptible to gentamicin (P<0.01) and amikacin (P<0.04) were lower.

Discussion

Staphylococcus intermedius was isolated from bacterial culture of 18.8% of 10,811 canine samples submitted from 1992 through 1997. This frequency of isolation was higher than a previously reported value of 12.9% from 1,868 canine samples.⁹ Frequencies of isolation of S. intermedius from skin (88.6%) and ear (49.4%) samples in this study were also higher than values of 53.9% and 29.3%, respectively, in the same report⁹ but were similar to another study in which 96 of 113 (85%) canine skin cultures yielded S. intermedius.¹⁰ Although case records from this study did not specify the precise location from which ear samples were collected, the isolation frequency (49.4%) fell between the values of 60.5% for samples collected from the horizontal canal and 36.8% for samples collected from the middle ear in 22 dogs with otitis externa/media.² In a recent report of 82 dogs with otitis media, S. intermedius was also the most common organism found and was isolated from 41.1% of affected ears.³ A pure culture of S. intermedius was recovered from 47.1% of the 497 skin cultures examined in this study. In contrast, a pure culture of S. intermedius was found in 73% of dogs with pyoderma in a previous study.¹¹ In that study the samples for culture were collected from pyogenic lesions only, making contamination with other skin organisms less likely. The results of this study in conjunction with the previous reports confirm that S. intermedius is an important skin pathogen in dogs.

Two of the 497 S. intermedius isolates from skin were resistant to both cephalothin and oxacillin, and two different isolates were resistant to ciprofloxacin. These low numbers of resistant isolates provided much less evidence of antimicrobial resistance than the 12% resistance to cephalothin, 11.5% resistance to methicillin, and 2.8% resistance to norfloxacin reported for all canine S. intermedius isolates in a previous study.⁹ However, these findings were similar to those of Lloyd, et al.,¹² who reported no resistance to oxacillin, methicillin, and enrofloxacin in 2,296 coagulase-positive staphylococci isolates collected from skin and mucous membranes. In fact, these authors found only one isolate resistant to cephalexin.¹² Similarly, resistance to cephalothin and methicillin was not found in 215 isolates of S. intermedius collected from the skin of dogs with pyoderma.¹¹

Three of the 553 ear isolates of S. intermedius were resistant to cephalothin and oxacillin, and five ear isolates were resistant to ciprofloxacin. In contrast, in a prior report of S. intermedius isolates from the horizontal ear canal, only 77.8% and 73.9% were susceptible to cephalothin and methicillin, respectively.² However, 92.9% and 92.3% of isolates from the middle ear were susceptible to these two antibiotic agents, respectively. With respect to fluoro-quinolones, the authors found a 99.1% overall susceptibility to ciprofloxacin. This value was greater than 96.3% and 85.7% enrofloxoxacin susceptibility of isolates collected from the horizontal canal and middle ear, respectively.² Although the results of this study are in general agreement with those of prior reports, subtle variation in resistance patterns underscores the importance of pursuing antimicrobial susceptibility testing of canine S. intermedius, especially in refractory cases of pyoderma or otitis externa/media.

One interesting finding in this study was that a greater number of ear isolates were observed to become increasingly susceptible to ampicillin, penicillin, and tetracycline over the 6-year period. This finding may simply be a reflection of the fact that these antibiotics are generally not present in topical otic preparations, and these antibiotics are rarely used systemically in the treatment of otitis externa/media.

Another interesting result was that fewer skin S. intermedius isolates were susceptible to ampicillin, penicillin, trimethoprim/sulfadiazine, tetracycline, and clindamycin in comparison to ear isolates. This finding may reflect greater use of these antibiotics (and thus higher selection pressure for less susceptible organisms) for treatment of skin infections in comparison to treatment of otitis externa/media. It may also represent a dose response, as skin pathogens may be exposed to lower concentrations of systemically administered antimicrobial agents than ear pathogens, which may be exposed to higher concentrations with locally administered topical medications. Although a curious finding, this difference has limited clinical significance, as these antibiotics (with the exception of clindamycin and trimethoprim/sulfadiazine) are not routinely selected for empirical treatment of canine pyoderma or otitis externa/media.

Pseudomonas aeruginosa was isolated from bacterial culture of 5.7% of 10,811 canine samples submitted from 1992 through 1997. Frequencies of isolation of P. aeruginosa from skin (7.5%) and ear samples (27.8%) were comparable to previous reports of isolation rates of 2.7% from skin samples⁸ and 26.3% and 36.8% from horizontal canal and middle ear samples, respectively.² Similar to the authors’ findings, P. aeruginosa was also the second most common organism isolated (from 35.5% of all ears) in a recent report of 82 dogs with otitis media.³ Frequency of isolation of P. aeruginosa from skin samples appeared to increase over the 6-year study period but remained unchanged for ear samples. Although the relatively small number of skin isolates (n=42) makes the apparent increase in isolation frequency difficult to interpret, it may suggest that P. aeruginosa is becoming a more significant pathogen in canine pyoderma.

All P. aeruginosa isolates, from both skin and ears, were resistant to ampicillin, cephalothin, trimethoprim/sulfadiazine, and tetracycline. This supports the recommendation that these antibiotics should not be used for suspected P. aeruginosa skin or ear infections. Fewer of the ear isolates, in comparison to skin isolates, were susceptible to gentamicin and amikacin. This finding may reflect the common use of topical aminoglycosides for treatment of otitis externa. Chronically inflamed ears often become colonized with P. aeruginosa, and this organism may become the most important pathogen over time as topical antibiotic treatment effectively eliminates other pathogens (i.e., S. intermedius). Use of topical aminoglycosides may be less efficacious against P. aeruginosa because the antimicrobial activity of these drugs is significantly reduced in the purulent debris that can accompany chronic otitis. This debris contains phagocytes, cellular remnants, and proteins (of both host and bacterial origin) that can bind or inactivate antibiotics.¹³ Long-term exposure to inadequate concentrations of aminoglycosides in this debris would lead to increased selection for resistant strains.

Two skin isolates and 22 ear isolates of P. aeruginosa were resistant to ciprofloxacin. This meant that 95.2% of the skin isolates and 92.9% of the ear isolates were susceptible to ciprofloxacin. This is in contrast to a previous report, which found that only 12.5% of Pseudomonas spp. isolated from the horizontal canal and 35.0% of Pseudomonas spp. isolated from the middle ear were susceptible to enrofloxaxin.² Unfortunately, because enrofloxacin was not considered to be an antipseudomonas drug by the laboratory used for this study, susceptibility of P. aeruginosa isolates to this fluoroquinolone was not determined. There are several possible explanations for the discrepancy in the percentages of P. aeruginosa isolates found to be susceptible to these two fluoroquinolones. First, ciprofloxacin is well documented to have greater activity against P. aeruginosa than enrofloxacin.^14–16 Second, the NCCLS-approved susceptibility breakpoint for enrofloxacin is lower (0.5 μg/mL) than the susceptibility breakpoint for ciprofloxacin (1.0 μg/mL). Third, different susceptibility results may also be reported when different methods are used to determine antimicrobial susceptibility. For example, in a recent study of 10 canine ear P. aeruginosa isolates, six were found to be susceptible to enrofloxacin when the dilution method was used to determine minimum inhibitory concentration of the antibiotic, while all 10 isolates were reported to be resistant to enrofloxacin by the semiquantitative Kirby-Bauer disk diffusion method.³ All in all, the data from this study and data of previous studies indicate that ciprofloxacin is preferred over enrofloxacin for systemic administration in the treatment of P. aeruginosa skin and ear infections.^14–16

Conclusion

The frequency of isolation of S. intermedius and P. aeruginosa from skin and ears of dogs and antimicrobial susceptibility patterns for these isolates did not change appreciably during the 6-year study period. Based on these findings, empirical treatment of canine superficial pyoderma caused by S. intermedius with cephalexin appears appropriate. However, bacterial culture and antimicrobial susceptibility testing should always be pursued in recurrent or refractory infections. Although oxacillin remains as a suitable choice, this antibiotic is no longer commercially available in the United States; however, as amoxicillin-clavulanic acid shares a similar pathogen profile, these antibiotics would be a suitable alternative. In contrast, empirical treatment of P. aeruginosa infections is not advisable because of the potential of encountering a multidrug-resistant isolate. Instead, bacterial culture and susceptibility testing should be pursued in all suspected P. aeruginosa infections. Finally, the authors did not find evidence for an increase in ciprofloxacin-resistant P. aeruginosa isolates in association with apparent increased use of this antimicrobial agent in canine practice.