Search tips
Search criteria 


Logo of jcmPermissionsJournals.ASM.orgJournalJCM ArticleJournal InfoAuthorsReviewers
J Clin Microbiol. 2011 August; 49(8): 2884–2886.
PMCID: PMC3147760

Collaborative Evaluation of an Erythromycin-Clindamycin Combination Well for Detection of Inducible Clindamycin Resistance in Beta-Hemolytic Streptococci by Use of the CLSI Broth Microdilution Method [down-pointing small open triangle]


Constitutive or inducible clindamycin resistance can occur in beta-hemolytic streptococci due to the presence of an erm gene. The Clinical and Laboratory Standards Institute (CLSI) has recommended a disk approximation test (D-zone test) with erythromycin and clindamycin disks and a single-well broth test combining erythromycin and clindamycin for detection of inducible clindamycin resistance in staphylococci, but only a disk approximation test for the beta-hemolytic streptococci. This collaborative study assessed two different erythromycin and clindamycin concentration combinations in single wells (1 μg/ml + 0.25 μg/ml [erythromycin plus clindamycin] and 1 μg/ml + 0.5 μg/ml) with three different brands of Mueller-Hinton broth supplemented with 3% lysed horse blood for testing of frozen panels prepared for this study. All labs performed the D-zone test as described by the CLSI. A total of 155 nonduplicate streptococcal isolates (50 group A, 48 group B, 28 group C, and 29 group G isolates) were tested; 99 isolates showed inducible resistance by the D-zone test. There were some differences noted based upon the test medium. The sensitivity of the erythromycin plus clindamycin combination of 1 μg/ml + 0.25 μg/ml was 91 to 100%, while the sensitivity of the combination of 1 μg/ml + 0.5 μg/ml was 95 to 100%. Specificity overall was 98%. The slightly higher sensitivity of the combination of 1 μg/ml + 0.5 μg/ml is recommended. This study has demonstrated that a single-well microdilution test incorporating erythromycin and clindamycin in combination is a sensitive and specific indicator of inducible clindamycin resistance and could be included in routine test panels.


There are two principle mechanisms of macrolide resistance in the beta-hemolytic group of streptococci. One is an active efflux mechanism encoded by the mef genes that affects only macrolides (6). The second mechanism, known as the MLSB resistance phenotype, is methylation of a single adenine at the 50S ribosomal binding site used by the macrolides, lincosamides, and streptogramin B drugs, mediated by an erm gene (6). Expression of MLSB resistance in staphylococci and streptococci can be either constitutive or inducible (6, 9). Fourteen- and 15-member macrolides are good inducers of the ribosomal conformational change, but lincosamides such as clindamycin are poor inducers (6). Thus, inducible resistance is not detected by standard clindamycin MIC or disk testing. The CLSI has described a disk diffusion D-zone test for detection of inducible clindamycin resistance in staphylococci and beta-hemolytic streptococci (3). If an alternative test were available, most clinical laboratories would prefer not to have to perform a separate D-zone disk test and would rather have an inducible resistance test included in their routine broth dilution commercial test systems. A single-well erythromycin-clindamycin combination broth microdilution test for detection of inducible clindamycin resistance in staphylococci (but not streptococci) has been described by the CLSI (2, 3). The goal of this study was to assess a single-well broth microdilution test for detection of inducible clindamycin resistance in beta-hemolytic streptococci through a five-center collaborative study.


Test isolates.

Each of the five laboratories was asked to test at least 20 nonduplicate macrolide-resistant clinical isolates of beta-hemolytic streptococci of groups A, B, C, and G but not small-colony Streptococcus anginosus group isolates with group A, C, or G antigens (11).

Broth microdilution tests.

Frozen broth microdilution panels were prepared in one laboratory according to CLSI guidelines, using three different Mueller-Hinton broth preparations (BBL [BD Microbiology], Difco [BD Microbiology], and Oxoid), all supplemented with 3% lysed horse blood (2). The frozen panels were prepared to include erythromycin and clindamycin tested separately to define MICs and combinations of erythromycin and clindamycin of 1 μg/ml + 0.25 μg/ml and 1 μg/ml + 0.5 μg/ml in separate wells, based upon the results of a prior study (1). Each panel included a growth control well and a negative (medium only)-control well. Panels were shipped frozen to the four collaborating laboratories for testing. Panels were inoculated with the standard density of 5 × 105 CFU/ml and incubated for 20 to 24 h at 35°C prior to visual determination of MICs.

Disk diffusion D-zone tests.

Standard disk diffusion D-zone testing was performed using erythromycin (15 μg) and clindamycin (2 μg) disks placed 12 mm apart on Mueller-Hinton 5% sheep blood agar plates incubated at 35°C in 5% CO2 for 20 to 24 h (3). Each laboratory provided its own plates: three labs used the BBL brand (BD Microbiology Systems, Sparks, MD), and two labs used the Remel brand (Lenexa, KS). A positive D-zone test was noted by flattening of the clindamycin zone adjacent to the erythromycin disk with erythromycin-resistant isolates.

Control strains.

S. pneumoniae ATCC 49619 was used as the control strain on each day of testing, and Staphylococcus aureus ATCC BAA977 was used for quality assessment of the erythromycin-clindamycin combinations (3).

Data analysis.

Growth or no growth in the erythromycin-clindamycin combination wells was compared to the disk diffusion D-zone test results. A combination well was considered to have accurately detected inducible clindamycin resistance if growth was present for an isolate that was determined to have a positive D-zone test (1).


A total of 155 macrolide-resistant beta-hemolytic streptococcal isolates were tested in the five laboratories, including 50 group A, 48 group B, 28 group C, and 29 group G streptococci (Table 1). The number of isolates tested in the five laboratories varied from 20 to 67 (Table 1). From 69% (group C) to 90 to 93% (other groups) of isolates tested susceptible to clindamycin when tested alone (without clindamycin resistance induction [data not shown]). There was a slight difference in clindamycin susceptibility among the group G streptococci based upon one strain testing as resistant in only one of the three Mueller-Hinton broth bases. Among the test isolates, 99 demonstrated inducible clindamycin resistance by the disk diffusion D-zone test, and 9 showed constitutive clindamycin resistance. A total of 47 isolates were resistant to erythromycin but susceptible to clindamycin and D zone negative (data not shown).

Table 1.
Numbers of beta-hemolytic streptococcal isolates tested in participating laboratoriesa

In this study, combined broth concentrations of 1 μg/ml erythromycin and 0.25 μg/ml clindamycin detected 91 to 100% of the inducibly clindamycin-resistant strains, as evidenced by positive agar-based D-zone tests, and 1 μg/ml erythromycin plus 0.5 μg/ml clindamycin detected 95 to 100% of strains (Table 2). The ranges indicate some differences by the medium used, with lysed horse blood-supplemented Mueller-Hinton broth prepared using BBL Mueller-Hinton base providing the greatest sensitivity and Difco and Oxoid broth bases providing slightly lower sensitivities for detection of inducible resistance. Only one strain (a group B Streptococcus strain) of the 47 clindamycin-susceptible, D-zone-negative isolates grew in any of the drug combination wells (Table 2). Thus, the specificity of the combined-drug single-well tests was 98% with all media (Table 2). Table 3 details the small number of false-positive and false-negative errors noted in the study according to the four streptococcal groups, the three brands of media, and the two drug combinations tested. The combination of 1 μg/ml erythromycin plus 0.5 μg/ml clindamycin performed slightly better than the combination with the lower clindamycin concentration of 0.25 μg/ml, especially in the Difco and Oxoid media.

Table 2.
Overall sensitivities of single-well combined erythromycin-clindamycin broth tests with three media in five laboratories in comparison to agar D-zone testsa
Table 3.
Single-well broth dilution test errors compared to D-zone test resultsa

As noted in a prior single-center study (1), the CLSI recommends the use of the S. aureus control strain ATCC BAA977, with inducible clindamycin resistance, as a quality control/quality assessment strain. In this study, the S. aureus control strain performed well (8 of 8 samples were positive in the combined drug wells), despite the fact that the two drug combination concentrations tested with the streptococci contained a lower erythromycin concentration (1 μg/ml erythromycin) than the combination of 4 μg/ml erythromycin plus 0.5 μg/ml clindamycin recommended for staphylococci (1, 3).


This five-center collaborative study demonstrated that a single broth microdilution test well incorporating 1 μg/ml erythromycin and 0.5 μg/ml clindamycin accurately detected inducible clindamycin resistance in a collection of group A, B, C, and G streptococci weighted toward macrolide-resistant strains. The lower clindamycin concentration in the combination of 1 μg/ml erythromycin plus 0.25 μg/ml clindamycin provided a slightly lower rate of sensitivity. The single-well broth test was equivalent to performing a separate agar disk approximation test with erythromycin and clindamycin disks and provides a more convenient test format that could be incorporated into broth microdilution panels for routine testing of streptococci. There were some medium differences noted in the study that slightly affected the sensitivity of the single-well test.

Some evidence exists, especially with Staphylococcus aureus, that there is a risk of spontaneous conversion from an inducible to a constitutive resistance phenotype during clindamycin therapy as a result of a single mutation in a promoter region that controls expression of the erm genes (7). Thus, patients could be at risk of clinical failure if inducible clindamycin resistance as well as constitutive resistance is not detected during routine antimicrobial susceptibility testing of individual patient isolates (4, 6, 7). Clindamycin monotherapy or combinations of penicillin plus clindamycin are sometimes used to treat severe streptococcal soft tissue infections, such as necrotizing fasciitis, where a therapeutic failure due to emergence of resistance could be limb- or life-threatening. Routine detection and reporting of inducible clindamycin resistance in beta-hemolytic streptococcal skin and soft tissue infections or bacteremia would provide an alert to clinicians that clinical failures due to emergence of resistance could occur during therapy. Both constitutive and inducible clindamycin resistance has increased in recent years, especially in group A and B streptococci (5, 8, 10). Indeed, the 2010 guidelines from the CDC on prevention of neonatal group B streptococcal infections now indicate that testing to detect inducible clindamycin resistance should be performed on all screening isolates from highly penicillin-allergic pregnant women (12). Broth microdilution testing with a single erythromycin-clindamycin combination well would provide a convenient option to clinical laboratories to facilitate routine detection of inducible clindamycin resistance.

Group F beta-hemolytic streptococci were not included in this study, and likewise, small-colony variants of group A, C, and G isolates of the S. anginosus group were not included (11). The CLSI recommends that all S. anginosus group isolates, whether alpha-, beta-, or nonhemolytic, should be considered members of the viridans group streptococci (3). The CLSI testing recommendations and certain drug interpretive breakpoints differ based upon isolates belonging to the beta-hemolytic group streptococci as opposed to the viridans group streptococci (3).

This multicenter study confirms the findings of an earlier single-center study (1) showing that a single-well broth screening test successfully detects inducible clindamycin resistance and is suitable for inclusion in broth panels used for routine testing of beta-hemolytic streptococci without the need for additional agar-based D-zone testing. Based upon a review of these data, the CLSI Subcommittee on Antimicrobial Susceptibility Testing has included a recommendation in document M100-S21 that the single-well test with 1 μg/ml erythromycin and 0.5 μg/ml clindamycin be used for detection of inducible clindamycin resistance in the beta-hemolytic streptococci (3).


No commercial support was received for the conduct of this study.


[down-pointing small open triangle]Published ahead of print on 22 June 2011.


1. Bowling J. E., et al. 2010. Detection of inducible clindamycin resistance in beta-hemolytic streptococci by using the CLSI broth microdilution test and erythromycin-clindamycin combinations. J. Clin. Microbiol. 48:2275–2277 [PMC free article] [PubMed]
2. Clinical and Laboratory Standards Institute 2009. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard, eighth edition, M07–A8. Clinical and Laboratory Standards Institute, Wayne, PA
3. Clinical and Laboratory Standards Institute 2011. Performance standards for antimicrobial susceptibility testing. Twenty-first informational supplement, M100–S21. Clinical and Laboratory Standards Institute, Wayne, PA
4. Desjardins M., Delgaty K. L., Ramotar K., Seetaram C., Toye B. 2004. Prevalence and mechanisms of erythromycin resistance in group A and group B Streptococcus: implications for reporting susceptibility results. J. Clin. Microbiol. 42:5620–5623 [PMC free article] [PubMed]
5. Diaz M., Torres Sanchez M. J., Aznar M. J. 2008. Prevalence and mechanisms of erythromycin and clindamycin resistance in clinical isolates of beta-hemolytic streptococci of Lancefield groups A, B, C, and G in Seville, Spain. Clin. Microbiol. Infect. 14:85–87 [PubMed]
6. Leclercq R. 2002. Mechanisms of resistance to macrolides and lincosamides: nature of the resistance elements and their clinical implications. Clin. Infect. Dis. 34:482–492 [PubMed]
7. Lewis J. S., II, Jorgensen J. H. 2005. Inducible clindamycin resistance in staphylococci: should clinicians and microbiologists be concerned? Clin. Infect. Dis. 40:280–285 [PubMed]
8. Michos A. G., et al. 2009. Macrolide resistance in Streptococcus pyogenes: prevalence, resistance determinants, and emm types. Diagn. Microbiol. Infect. Dis. 64:295–299 [PubMed]
9. Raney P. M., Tenover F. C., Carey R. B., McGowan J. E., Jr., Patel J. B. 2006. Investigation of inducible clindamycin and telithromycin resistance in isolates of beta-hemolytic streptococci. Diagn. Microbiol. Infect. Dis. 55:213–218 [PubMed]
10. Richter S. S., et al. 2005. Macrolide-resistant Streptococcus pyogenes in the United States, 2002–2003. Clin. Infect. Dis. 41:599–608 [PubMed]
11. Spellerberg B., Brandt C. 2007. Streptococcus, p. 412–429In Murray P. R., Baron E. J., Jorgensen J. H., Pfaller M. A., Yolken R. H., editors. (ed.), Manual of clinical microbiology, 9th ed. American Society for Microbiology, Washington, DC
12. Verani J. R., et al. 2010. Prevention of perinatal group B streptococcal disease; revised guidelines from CDC, 2010. MMWR Recomm. Rep. 59(RR-10):1–36 [PubMed]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)