Emergence of Macrolide Resistance in Throat Culture Isolates of Group A Streptococci in Ontario, Canada, in 2001

Abstract

Of 500 group A streptococci isolated from pharyngeal swabs, 72 (14.4%) were macrolide resistant, compared to 2.1% in 1997. Of these, 66 (92%) were of the M phenotype and 6 (8.3%) were of the MLS phenotype. Pulsed-field gel electrophoresis found that two clones, with patterns identical to those of serotypes M1 and M4, accounted for 19.4 and 68.1% of the macrolide-resistant isolates, respectively.

Increasing macrolide resistance of group A streptococci (GAS) has been observed worldwide during the last decade. Resistance rates in North America have been <3% until recently (4, 8).

In this study, we looked at the prevalence of macrolide resistance in community GAS throat isolates from southern Ontario and determined their mechanisms of resistance. The prevalence of macrolide resistance was compared to rates in GAS isolates obtained in 1997 from the same patient population and region (4). MDS Laboratories, Canada, a private laboratory serving community physicians throughout southern Ontario and representing a population base of 6 million people, collected 500 consecutive GAS throat isolates during the fall of 2001. Only one specimen per patient was included. Strains were identified with a Prolex Streptococcal Grouping Latex Kit (Pro-Lab Diagnostics, Austin, Tex.). Erythromycin-resistant strains were initially identified by disk diffusion on Mueller-Hinton agar supplemented with 5% sheep blood with 15-μg erythromycin disks (Becton Dickinson, Cockeysville, Md.) in accordance with NCCLS guidelines (12). Resistance identified by disk diffusion testing was verified by broth microdilution in accordance with NCCLS guidelines (11). Susceptibility of isolates to penicillin, erythromycin, clindamycin, tetracycline, linezolid, ciprofloxacin, levofloxacin, vancomycin, and telithromycin for erythromycin-resistant strains was determined by broth microdilution susceptibility testing.

Two principal mechanisms are responsible for acquired erythromycin resistance in GAS: target site modification and active efflux. Target site modification is most often due to N6 dimethylation of an adenine residue (A2058) in the peptidyl transferase circle of 23S rRNA domain V through the action of a family of enzymes encoded by erm class genes. Methylation results in reduced binding of and coresistance to macrolide, lincosamide, and streptogramin B (MLS) antibiotics. MLS resistance can be expressed either constitutively (cMLS phenotype) or inducibly (iMLS phenotype). In Streptococcus pyogenes, it is mediated by two classes of methylase genes, the conventional erm(B) genes and the recently described erm(TR) genes, hereafter designated erm(A) in accordance with recently accepted nomenclature (14).

Active efflux is the other major resistance mechanism for macrolide antibiotics (23). Only the 14- and 15-membered macrolides are affected (thus, M resistance), and resistance is usually low level. M resistance is mediated in S. pyogenes by the mef(A) gene.

MLS and M phenotypes of erythromycin-resistant strains were determined by double-disk diffusion testing (6). Erythromycin (15 μg) and clindamycin (2 μg) disks were placed on plates approximately 12 mm apart, and the plates were incubated overnight at 37°C in 5% CO₂. The absence of a significant zone of inhibition around the two disks was taken to indicate the cMLS phenotype, blunting of the clindamycin zone of inhibition proximal to the erythromycin disk was taken to indicate the iMLS phenotype, and susceptibility to clindamycin with no blunting of the zone of inhibition around the clindamycin disk was taken to indicate the M phenotype.

All isolates erythromycin resistant by MIC testing were characterized by PCR to detect the presence of mef(A), erm(B), and/or erm(A) genes with published primer sequences and in accordance with previously described methodology (4, 17).

Isolates were typed by pulsed-field gel electrophoresis (PFGE) after digestion with SfiI as described by Murray et al. (10). Strains were considered to be of the same PFGE type if the patterns had three or fewer different bands (18). M typing was performed at the National Centre for Streptococcus reference laboratory in Edmonton, Alberta, by standard methodology (5). IMS HEALTH, Canada, provided an estimate of the total number of antibiotic prescriptions dispensed in Canadian retail pharmacies on the basis of a representative sample of 2,000 pharmacies stratified by type, size, and province. Statistical comparisons of resistance rates were done with a Yates-corrected chi-square test with Epi Info 2000 (Centers for Disease Control and Prevention, Atlanta, Ga.).

Of the 500 isolates tested, 72 (14.4%) were found to be erythromycin resistant by disk diffusion testing. All of these were also found to be resistant by broth microdilution testing. A total of 91.5% of the resistant isolates consisted of two clones as determined by PFGE. A total of 59 (16%) of the patients <16 years of age carried a macrolide-resistant GAS, compared with 13 (10%) of those ≥16 years of age (P > 0.14). The prevalence of resistance in this study was significantly greater than in a similar study carried out in 1997 (4). That surveillance study had the same design and patient population and was carried out at the same laboratory as this study. In the 5-year period, macrolide resistance rates increased by more than sixfold (2.1% in 1997 to 14.4% in 2001 [P < 0.0001]) (Fig. 1).

FIG. 1.

Macrolide use and evolving GAS macrolide resistance. Macrolide usage trends are plotted along the left y axis in numbers of prescriptions per 100,000 people. Macrolide resistance rates (percent) are plotted along the right y axis. EryR, erythromycin resistance.

Others in the United States have also documented increasing rates of resistance. York et al. (21) reported erythromycin resistance rates of 32% among isolates from patients with invasive disease and 9% among pharyngeal isolates obtained between 1994 and 1995 in the San Francisco, Calif., area. Martin et al. (9) described an outbreak of a macrolide-resistant clone of GAS that, within a few months, had spread among schoolchildren in Pittsburgh, Pa. Between October 1998 and May 2000, none of the GAS isolates studied were resistant to macrolide antibiotics. Between October 2000 and May 2001, however, no less than 48% of the isolates were resistant. This high rate was due to a single M6 clone. Shulman et al. (15) reported findings from a surveillance study in 2000 that included GAS isolates from children in geographically diverse sites. Data for the 2000 to 2001 isolates indicated that the rate of macrolide resistance was approximately 7.4%. Critchley et al. (3), in 1999, found that 6.2% of 2,742 GAS isolates collected from 324 clinical laboratories were resistant to macrolides.

In Quebec, Canada, Weiss et al. (20) characterized a total of 496 GAS strains isolated from patients with pharyngitis in five acute-care hospitals between September and December 1998. The overall resistance rate was 4.6% but varied from 0% in rural areas to 9.4% in Montreal. Of the 23 strains showing resistance to erythromycin, 15 (65%) had the same PFGE pattern. They were of serotype M28 and harbored the erm(A) gene, suggesting the spread of a single clone.

It is clear that macrolide resistance rates are climbing internationally. Our study and others, mentioned above, demonstrate that resistant clones enter communities and can lead to rapid and significant increases in resistance rates. Whether the clones are transiently present in the community or whether they remain for long periods of time remains unclear.

In Canada, there has been a reduction in the number of erythromycin prescriptions from 11.3/100 patients per year in 1995 to 2.9/100 patients per year in 2001 but an increase in the number of clarithromycin and azithromycin prescriptions from 3.9 to 12.2/100 patients per year over the same time period (Fig. 1). Data for the Province of Ontario are very similar to the national data. Several studies have shown an association with increasing macrolide use and resistance in pneumococci (7, 13). Baquero (1) applied pharmacodynamic concepts to suggest that bacterial exposure to low and prolonged concentrations of macrolides may have a role in the selection of resistance. Analysis of macrolide prescribing and resistance patterns indicated a correlation between increasing macrolide resistance and the increased use of newer, long-acting macrolides.

Of the 72 erythromycin-resistant isolates, 66 (91.5%) were of the M phenotype, all of which possessed the mef(A) gene. Those of the cMLS phenotype and one of the iMLS phenotype were found to possess the erm(B) gene. The remaining four of the iMLS phenotype were found to possess the erm(A) gene. Erythromycin MICs ranged from 0.06 to >32 μg/ml. No strains were resistant to penicillin, levofloxacin, vancomycin, or linezolid. Three isolates were resistant to tetracycline: one that harbored mef(A), one with erm(A), and one with erm(B). For one isolate, the telithromycin MIC was >1 μg/ml (i.e., 4 μg/ml) and the strain harbored erm(B). The MICs of representative antibiotics are shown in Table 1.

TABLE 1.

Distribution of in vitro susceptibility results for 72 erythromycin-resistant GAS isolates

Antimicrobial and gene	No. of isolates for which the MIC (μg/ml) was:
	≤0.06	0.12	0.25	0.5	1	2	4	8	16	≥32
Erythromycin
mef(A)							2	24	40
erm(B)										2
erm(A)						2	2
Clindamycin
mef(A)	63	3
erm(B)		1								1
erm(A)	3	1
Telithromycin
mef(A)			19	44	3
erm(B)		1					1
erm(A)	4

Telithromycin, a new ketolide, had reduced activity against isolates that were resistant to the macrolides because of efflux encoded by mef(A) and target site modification encoded by erm(B) compared to erythromycin-susceptible isolates and isolates that were macrolide resistant because of the presence of erm(A) (Table 1). Although the mechanism is unclear, similar findings have been noted by others.

We were able to characterize all 72 erythromycin-resistant isolates by PFGE (4, 22). Each of the six erm(A)- or erm(B)-containing strains constituted a unique PFGE type, and there were five different M types. Among the 66 mef(A)-containing strains, five different clones were identified by PFGE: 46 isolates with pattern A (M type 4), 14 with pattern B (M type 1), 4 with pattern C (M type 12), and 1 each with two other patterns (1 with pattern M2 and 1 with pattern M75). These data suggest that the emergence of macrolide resistance in southern Ontario is due primarily to the dissemination of a limited number of clones. Patterns A (M4) and B (M1) accounted for 65 and 19% of the macrolide-resistant isolates, respectively.

The outcome of the treatment of pharyngitis with a macrolide when the infecting GAS isolate is macrolide resistant is uncertain, although some studies have found higher than expected failure rates (19). The Infectious Diseases Society of America recommends laboratory confirmation of the clinical diagnosis by means of either throat culture or a rapid antigen detection test (2). A subcommittee of the American College of Physicians and the American Society of Internal Medicine, in collaboration with the Centers for Disease Control and Prevention, advocates the use of a clinical algorithm alone, in lieu of microbiologic testing, for confirmation of the diagnosis in adults for whom the suspicion of streptococcal infection is high (16). However, with increasing rates of resistance, pharyngeal cultures may be necessary to determine the optimal therapy.