Antimicrobial susceptibility of Staphylococcus aureus and Staphylococcus pseudintermedius isolated from various animals

Abstract

This study characterized the antimicrobial susceptibility of 221 Staphylococcus aureus isolated from various species, and 60 canine Staphylococcus pseudintermedius isolated from 1986 through 2000 at the Western College of Veterinary Medicine (WCVM). Resistance of S. aureus was most common to penicillin (31%) and tetracycline (14%); resistance of S. pseudintermedius to penicillin was present in 8% and to tetracycline in 34% of isolates. Resistance to trimethoprim/sulfamethoxazole was only seen among S. pseudintermedius, and there was no resistance to amoxicillin/clavulanate, ampicillin/sulbactam, cephalothin, amikacin, gentamicin, enrofloxacin, chloramphenicol, or rifampin among any isolate. Inducible clindamycin resistance was found in both S. aureus and S. pseudintermedius, highlighting the need for careful interpretation of culture and susceptibility test results. There were significant differences in the minimum inhibitory concentrations of penicillin, ciprofloxacin, enrofloxacin, clindamycin, erythromycin, chloramphenicol, and tetracycline between avian, bovine, equine, and porcine isolates.

Introduction

Staphylococcus aureus and Staphylococcus pseudintermedius are coagulase positive staphylococci frequently implicated in opportunistic infections (1). Staphylococcus aureus affects humans, and companion and food animals causing both community acquired and nosocomial infections. One of the most common bacterial pathogens in humans, S. aureus is a major cause of morbidity and economic loss in production animals; it is increasingly recognized in companion animal infections. In poultry, S. aureus is a cause of foot and leg infections and septicemia; in dairy cattle, S. aureus mastitis is the single largest source of economic loss for the North American industry (2,3). Horses commonly carry S. aureus including methicillin resistant strains (MRSA) and a variety of infections including dermatitis, septic arthritis, and catheter site infections are reported (4–6). While the emergence of “livestock associated” MRSA has raised the profile of S. aureus in pigs, a wide variety of non-specific presentations including septicemia, abscesses, and osteomyelitis are associated with this organism (7,8).

In dogs, S. pseudintermedius (recently recognized as being distinct from S. intermedius) is more common than S. aureus both as a colonizer and as a cause of infections (9–11). It is a frequent cause of otitis, pyoderma, and urinary tract infections; although as an opportunist, any site with compromised host defenses is susceptible (12–14).

Resistance to commonly used antimicrobials is frequently encountered with both S. aureus and S. pseudintermedius. First reported among hospital acquired S. aureus strains from humans in the 1940s shortly after the introduction of penicillin, beta-lactamase (penicillinase) production is now widely disseminated among S. aureus and S. pseudintermedius in the community (15). Consequently, first generation cephalosporins, amoxicillin + clavulanate, clindamycin, trimethoprim + sulfamethoxazole, and enrofloxacin are routinely used in small animal practice for the treatment of staphylococcal infections (16). Emerging resistance to these drugs, particularly among methicillin resistant staphylococci, increases the likelihood of treatment failure and represents a risk to public health.

Inducible clindamycin resistance (iCR) is increasingly recognized, and complicates the interpretation of susceptibility reports. This phenomenon is frequently encountered among isolates which, without using specialized techniques, appear to be erythromycin-resistant and clindamycin-susceptible (E^RC^S) (17). In vitro, failure of clindamycin to induce resistance can result in misclassifying an isolate as susceptible. In human patients treatment failure following in vivo induction of resistance has been reported (18,19).

As the prevalence of resistance is geographically variable, local investigations are necessary. The objective of this study was to determine and compare the antimicrobial susceptibility profiles of an archived collection of S. aureus and S. pseudintermedius isolated from different animal species from clinical sample submissions to the WCVM bacteriology laboratory.

Materials and methods

Bacterial isolates

Clinical isolates of S. aureus (n = 221) from various species (Table 1), and canine S. pseudintermedius (n = 60) collected from 1986 through 2000 at the WCVM bacteriology laboratory were tested. Only 1 isolate per animal was included. Bacteria were identified using standard methods: colony morphology, Gram stain, and biochemical tests including fermentation of mannitol and production of coagulase, DNase, acetoin, and hyaluronidase. Following identification, isolates were frozen in skim milk at −80ºC. As MRSA were not isolated at the WCVM until 2006, all isolates in this collection are methicillin susceptible. Methicillin susceptibility was confirmed by oxacillin disc diffusion testing according to CLSI guidelines, and by screening for the mecA gene (17).

Table 1.

Species of origin of Staphylococcus aureus strains

Species isolated from	Number of isolates
Avian	25
Bison, caribou, meerkat mouse, rat, sea otter	1 each
Bovine	123
Canine	5
Equine	21
Feline, caprine, iguana	2 each
Ovine	7
Porcine	22
Rabbit, wapiti	3 each
Total	221

Antimicrobial susceptibility testing

The minimum inhibitory concentrations (MIC) of 14 drugs were determined using the agar dilution technique according to CLSI guidelines (20). The following antimicrobials were used: amoxicillin/clavulanate, ampicillin/sulbactam, cephalothin, penicillin, ciprofloxacin, enrofloxacin, erythromycin, clindamycin, amikacin, gentamicin, chloramphenicol, rifampin, tetracycline, and trimethoprim/sulfamethoxazole. Ciprofloxacin, enrofloxacin, and sulbactam were purchased from the United States Pharmacopeia (USP, Rockville, Maryland, USA); all other drugs were purchased from Sigma-Aldrich (Sigma-Aldrich, St. Louis, Missouri, USA). The MICs were interpreted according to the Clinical and Laboratory Standards Institute (CLSI) guidelines (17,21). Escherichia coli ATCC 25922 and S. aureus ATCC 25923 were included for quality control. All E^RC^S resistant isolates were tested for iCR using the “D-test” according to CLSI guidelines (17). Commercially prepared erythromycin and clindamycin discs were used (Becton, Dickinson, Sparks, Maryland, USA).

Statistical analysis

Statistical analyses were conducted with a commercial software package (Stata/IC 10.1 for Windows, StataCorp LP, College Station, Texas, USA). Statistical analysis included only avian, bovine, equine and porcine isolates due to the small numbers of isolates from other species. Numbers of antimicrobials to which isolates were resistant were compared with the Kruskal-Wallis test, with alpha set to 0.05. Post-hoc analysis was performed using the Wilcoxon-Mann-Whitney test with Bonferroni’s correction. Analyses of the number of antimicrobials to which isolates were resistant included results reported for all antimicrobials. The MICs for bovine, avian, equine, and porcine S. aureus isolates were compared with the Kruskal-Wallis test. Post-hoc analysis was performed using the Wilcoxon-Mann-Whitney test with Bonferroni’s correction. For analysis of categorical resistance data, isolates of intermediate susceptibility were classified as “non-resistant.”

Results

Antimicrobial susceptibility profiles

Staphylococcus aureus isolates were resistant to between 0 and 4 antimicrobials, with a median of 1. The number of antimicrobials to which S. aureus isolates were resistant varied significantly between bovine, porcine, avian, and equine isolates (P = 0.0001). Porcine isolates were resistant to significantly more antimicrobials than were bovine, equine, and avian isolates (P < 0.0001). There were no significant differences in the number of antimicrobials to which isolates from bovine, equine, and avian hosts were resistant. No resistance to the aminogly-cosides, chloramphenicol, trimethoprim/sulfamethoxazole, rifampin, enrofloxacin, amoxicillin/clavulanate, ampicillin/sul-bactam, or cephalothin was observed in S. aureus isolated from any species (Table 2). Significant variations in MIC for a given drug, including chloramphenicol and enrofloxacin to which no isolates were resistant, were seen between species (Table 2).

Table 2.

Percent resistant, MIC₅₀ and MIC₉₀ of Staphylococcus aureus overall and isolated from avian, bovine, equine, and porcine sources (n = 221)

Class and/or Antimicrobial	Range tested (μg/mL)	Resistance breakpoint (μg/mL)	All (n = 221)	Avian (n = 25)	Bovine (n = 123)	Equine (n = 21)	Porcine (n = 22)
Amoxicillin/Clavulanate	2–16	8	0	0	0	0	0
MIC50a				2	2	2	2
Ampicillin/Sulbactam	4–64	32	0	0	0	0	0
MIC50a				4	4	4	4
Cephalothin	4–64	32	0	0	0	0	0
MIC50a				4	4	4	4
Penicillin	0.06–0.5	0.25	31	20	26	43	100b
MIC50				0.06	0.06	0.06	1
MIC90				1	0.5	1	1
Ciprofloxacin	0.25–16	4	< 1	0	0	5	0c
MIC50				0.25	0.25	0.25	0.5
MIC90				0.5	0.5	0.5	0.5
Enrofloxacin	0.25–16	4	< 1	0	0	0c	0
MIC50				0.25	0.25	0.25	0.25
MIC90				0.25	0.25	0.5	0.25
Clindamycin	0.25–16	4	7	0	0	5	64b
MIC50a				0.25	0.25	0.25	16
Erythromycin	0.25–16	8	8	4d	0	5	68b
MIC50a				0.25	0.25	0.25	32
Amikacin	8–64	64	0	0	0	0	0
MIC50a				8	8	8	8
Gentamicin	2–16	16	0	0	0	0	0
MIC50a				2	2	2	2
Chloramphenicol	4–32	32	0	0	0	0	0c
MIC50				4	4	8	8
MIC90				8	8	8	8
Rifampin	0.5–8	4	0	0	0	0	0
MIC50a				0.5	0.5	0.5	0.5
Tetracycline	2–32	16	12	24d	1	10	82b
MIC50				2	2	2	32
MIC90				64	2	2	64
Trimethoprim/Sulfamethoxazole	19–76	76	0	0	0	0	0
MIC50a				19	19	19	19

^aThe MIC₉₀ values were identical to the MIC₅₀.

^bSignificantly higher MICs (P < 0.008) than avian, bovine, and equine isolates.

^cSignificantly higher MICs (P < 0.008) than avian and bovine isolates.

^dSignificantly higher MICs (P < 0.008) than bovine isolates.

Fewer than 10 S. aureus isolates were archived from each of the following host species: bison, caribou, cat, dog, goat, iguana, meerkat, mouse, rabbit, rat, sea otter, sheep, and wapiti. Neither antimicrobial susceptibility nor statistical analyses are presented for these isolates due to the small number of isolates from each of these species.

Among S. pseudintermedius isolates, no resistance to the aminoglycosides, chloramphenicol, rifampin, amoxicillin/clavulanate, ampicillin/sulbactam, cephalothin, or the fluoroquinolones was found (Table 3). Tetracycline resistance was most common (34%), followed by erythromycin (13%), clindamycin (8%), penicillin (7%), and trimethoprim/sulfamethoxazole (5%).

Table 3.

Percent resistant, MIC₅₀ and MIC₉₀ of canine Staphylococcus pseudintermedius (n = 60)

Class and/or Antimicrobial	Range tested (μg/mL)	Resistance breakpoint (μg/mL)	Resistant strains (%)	MIC50	MIC90
Amoxicillin/Clavulanate	2–16	8	0	2	2
Ampicillin/Sulbactam	4–64	32	0	4	4
Cephalothin	4–64	32	0	4	4
Penicillin	0.06–0.5	0.25	7	0.06	0.12
Ciprofloxacin	0.25–16	4	0	0.25	0.25
Enrofloxacin	0.25–16	4	0	0.25	0.25
Clindamycin	0.25–16	4	8	0.25	1
Erythromycin	0.25–16	8	13	0.25	32
Amikacin	8–64	32	0	8	8
Gentamicin	2–16	8	0	2	2
Chloramphenicol	4–32	32	0	4	4
Rifampin	0.5–8	4	0	0.5	0.5
Tetracycline	2–32	16	34	2	64
Trimethoprim/Sulfamethoxazole	19–76	76	5	19	19

Inducible clindamycin resistance

Three S. aureus isolates were E^RC^S. Inducible clindamycin resistance was identified in 1 porcine and 1 avian isolate with this phenotype, while the third E^RC^S isolate was not inducibly resistant. The single E^RC^S S. pseudintermedius was inducibly resistant to clindamycin. Overall, iCR was found in 2/221 (0.9%) of S. aureus and 1/60 (1.7%) of S. pseudintermedius isolates.

Discussion

By testing this collection of isolates we aimed to define local, species specific resistance patterns. Resistance to tetracycline, erythromycin, clindamycin, and penicillin was significantly higher among porcine isolates than among isolates from other species. The high prevalence of macrolide resistance among porcine S. aureus may be partially attributable to tylosin usage in swine production (22). Tylosin fed to pigs has been experimentally shown to select for erythromycin-resistant staphylococci (23). In contrast, bovine and avian isolates were overwhelmingly susceptible. As previously reported, bovine mastitis isolates were susceptible to penicillin although penicillins are the most commonly used drugs for mastitis therapy; further research is required to determine the reason for this (24). Consistent with previous reports, low rates of resistance to chloramphenicol, the macrolides, aminoglycosides, and fluoroquinolones were found among avian isolates (3). A report documenting the susceptibilities of equine S. aureus at the WCVM up to 2003 found similar rates of resistance as are presented here (25). We report the first clinical isolation of S. aureus from a number of exotic species including caribou, iguana, meerkat, sea otter, and wapiti, consistent with this organism’s lack of host specificity.

To more sensitively detect differences between groups, MICs were compared, as opposed to comparing categorical resistance data. Measuring MICs allows detection of small changes in susceptibility that do not occur around defined resistance breakpoints. This approach identified significant differences in susceptibility for drugs where no isolates with MICs exceeding the CLSI resistance breakpoint were identified. Routine MIC determination has clinical and research benefits over disc diffusion testing and allows pharmacokinetic/pharmacodynamic principles to be applied when designing evidence based dosing regimens. For example, low level β-lactam resistance in organisms causing urinary tract infections can be overcome by high urine drug concentrations (26).

The S. pseudintermedius isolates were entirely susceptible to most commonly used antimicrobials in canine practice, including amoxicillin/clavulanate, cephalothin, enrofloxacin, and clindamycin. The relative susceptibility of S. pseudintermedius at the WCVM has previously been reported, and may reflect the large volume of primary cases seen at the WCVM which have not been previously treated with antimicrobials (14). Alternatively, this isolate collection may predate the emergence of more resistant isolates in this area.

Although inducible clindamycin resistance was only found in a single S. pseudintermedius and 2 isolates of S. aureus; 2 of the 3 S. aureus isolates with the E^RC^S phenotype were inducibly resistant. Despite recognized guidelines, testing for iCR is not yet standard procedure in many laboratories (27). As inducible resistance should be suspected when staphylococci are reported to be E^RC^S, its identification is confounded if both erythromycin and clindamycin susceptibilities are not tested. Practitioners should be aware of this increasingly recognized phenomenon as use of clindamycin to treat infections caused by iCR staphylococci can result in treatment failure (18,19). Previously, 17.7% to 37.5% of veterinary MRSA isolates were reported to be iCR (28,29). The low prevalence of iCR found here may be due to testing of bacteria that were methicillin susceptible or to geographic variation.

As this study comprised isolates collected in a diagnostic laboratory, the collection could be biased towards more resistant organisms as practitioners may be more likely to submit samples from difficult or refractory cases. However, as the WCVM has a largely primary caseload, this limitation may be of less importance and results can likely be generalized to western Canadian practices. A prospective study with a larger sample size would be better suited to assess changes in susceptibility between species and over time.

The highly resistant staphylococci often reported in the literature may be indicative of a publication and submission bias towards these organisms. This bias overemphasizes the relative frequency of multidrug resistant organisms and may influence practitioners to select unnecessarily broad spectrum drugs for empirical therapy. Identification of the offending organism followed by careful inspection of its susceptibility profile, including consideration of factors such as iCR, are all elements of rational antimicrobial selection.

The dissemination of antimicrobial resistant staphylococci is presenting challenges to both human and animal health professionals. Further studies are required to define the antimicrobial susceptibility profiles of veterinary staphylococci from western Canada to monitor the emergence and dissemination of resistance.