Journal of Antimicrobial Chemotherapy (996) 37, 37-375 In-vitro antimicrobial susceptibility of the 'Streptococcus millerv group {Streptococcus anginosus, Streptococcus constellatus and Streptococcus intermedius) Jan A. Jacobs* and Ellen E. Stobberingh University Hospital of Maastricht, Department of Medical Microbiology, P.O. Box 5800, NL-60, AZ Maastricht, The Netherlands A collection of 43 clinical 'Streptococcus milleri' strains was speciated and tested for susceptibilities to twelve antibiotics. Only.4% of the strains were of intermediate susceptibility to penicillin. None of the strains exhibited high-level resistance to gentamicin. Strains resistant to erythromycin, roxithromycin and clindamycin were found with a frequency of.6%,.4% and.4% respectively. resistance was found in 5.7% of the strains and occurred most frequently in Streptococcus anginosus. All the strains were susceptible to cefotaxime, vancomycin and teicoplanin. The results of this study do not indicate changing antibiotic resistance in strains of the 'S. milleri' group, but local differences in antibiotic susceptibilities may occur. Introduction Strains belonging to the 'Streptococcus milleri' group are distinct among the viridans streptococci because of their tendency to cause suppurative infections (Gossling, 988). At present, three newly denned species are recognised: Streptococcus anginosus, Streptococcus constellatus and Streptococcus intermedius (Whiley & Beighton, 99). Strains belonging to the 'S. miller? group are considered uniformly susceptible to the antibiotics usually employed for streptococcal infections (Gossling, 988). However, anecdotal reports (Quinn et al., 988) as well as studies on strain collections (Gomez-Garces, Alos & Cogollos, 994) have demonstrated the presence of resistant strains. Moreover, an increasing antimicrobial resistance among viridans streptococci has been noted (Potgieter et al., 99). Apart from one study (Gomez-Garces et al., 994), all studies on antibiotic susceptibility of the 'S. milleri' group were done before 990 and pre-dated current taxonomy. For these reasons, we were interested in the current susceptibility patterns of the '5. milleri' group in general and in the differences in susceptibilities between the three species in particular. We retrospectively identified a collection of clinical '5. milleri' isolates to the species level and determined susceptibilities to a range of antibiotics in current use. 'Phone: +3-43-3874644; Fax: +3-43-3876643 37 0305-7453/96/0037 + 05 $.00/0 / 996 The British Society for Antimicrobial Chemotherapy Downloaded from https://academic.oup.com/jac/article-abstract/37//37/753356 on 0 January 08
37 J. A. Jacobs and E. E. Stobberingh Methods The isolates were recovered over a three-year period (July 99-June 994) from clinical specimens received by the microbiology laboratory of the University Hospital of Maastricht. Four hundred and twenty three isolates were collected, 4 (53.0%) of which were infection-related and the remaining 99 (47.0%) were considered of questionable clinical significance. The isolates were recovered from the abdominal cavity (n = 8), the urogenital tract (05), head and neck area (56), the thoracic cavity including the respiratory tract (74), the skin and soft tissues (4), and the blood (8). They were stored at 70 C on porous beads in cryopreservative (Microbank, Pro-Lab Diagnostics, Richmond Hill, Ontario, Canada). Before inclusion in the study, their identification was confirmed with the API 0 Strep system supplied with the data base 4.0 (API System, biomerieux, Marcy-l'Etoile, France). Isolates were identified as S. anginosus, S. constellatus or S. intermedius using the scheme described by Whiley et al. (990). Antimicrobial susceptibility tests were carried out by the agar dilution method (NCCLS, 994) on Mueller-Hinton agar (Oxoid, Basingstoke, UK) supplemented with 5% (vol/vol) sheep blood and containing doubling dilutions of the different antibiotics. The antibiotics were obtained as standard reference powders of defined potency from their respective manufacturers in The Netherlands and included: penicillin (Yamanouchi, Leiderdorp), amoxycillin (Smith-Kline Beecham, Rijswijk), cefotaxime (Hoechst-Roussel, Hoevelaken), clindamycin (Upjohn, Ede), erythromycin (Abbott, Amsterdam), roxithromycin (Roussel, Hoevelaken), doxycycline (Pfizer, Rotterdam), gentamicin (Essex-Schering, Amstelveen), vancomycin (Eli Lilly, Nieuwegein), teicoplanin (Yamanouchi, Leiderdorp), ciprofloxacin (Bayer, Mijdrecht), and ofloxacin (Hoechst, Amsterdam). The plates were inoculated with lo^cfu/spot with a multipoint inoculator and incubated in 7% CO at 35 C for 8 h. Control organisms were Staphylococcus aureus ATCC 93 and Enterococcus faecalis ATCC 9. The susceptibility breakpoint concentrations were as recommended by the National Committee for Clinical Laboratory Standards (NCCLS, 994). Results The MICJO, MIG» and the MIC range of the three species for the different antibiotics are presented in the Table. For penicillin, 6/43 (.4%) isolates were moderately susceptible, and were randomly distributed among the three species. None of the isolates tested exhibited high-level resistance (MIC ^ 500 mg/l) to gentamicin. Overall resistance rates to macrolide antibiotics were limited to.6% ( isolates) for erythromycin and to.4% (ten isolates) for both clindamycin and roxithromycin. Resistance was found exclusively among the isolates of S. anginosus and S. constellatus. Resistance to all three aforementioned antibiotics was found in nine isolates, combined erythromycin and clindamycin resistance in another two and combined roxithromycin and clindamycin resistance in one. Overall, 35/43 (83.0%) of the organisms studied were susceptible to doxycycline, and 4 (5.7%) were resistant to this agent. Of the three species, S. anginosus was most frequently resistant. Downloaded from https://academic.oup.com/jac/article-abstract/37//37/753356 on 0 January 08
Antimicrobial susceptibility of "S. milleri" 373 Discussion Strains belonging to the 'S. milleri' group encompass a heterogeneous collection of viridans streptococci. Two classification schemes coexisted from 977 to 987: one British and one American. The American scheme distinguished between three species, that were named (in the final version), S. anginosus, S. constellatus and 5. intermedius. From 987 to 99, all strains were integrated into one single species, S. anginosus. Recently, molecular techniques delineated three species (Whiley & Beighton, 99). For reasons of nomenclature (the principle of priority of publication), they were given the same names as those used by the American scheme. The name 'S. milleri' was used by British taxonomists, who unified all strains into one species. The name is not legitimate but is still used because of its association with abscesses and other purulent infections. Although surgery constitutes the cornerstone of effective treatment of such infections, prolonged antibiotic therapy may be required (Jacobs et al., 994). Recent studies on antimicrobial susceptibilities of the '5. milleri' group are rare and only one report gives details of the susceptibilities of the three species described here (Gomez-Garces et al., 994). The literature on 'S. milleri' up to 988 has been reviewed by Gossling, who noted a wide variety in the results of susceptibility determinations, making comparisons difficult (Gossling, 988). In the present study, penicillin was shown to be intrinsically more active than amoxycillin, as reported by others (Gossling, 988; Gomez-Garces et al., 994). The number of isolates moderately susceptible to penicillin G was comparable with the data reviewed by Gossling (988) and to the results reported in recent studies (LePennec & Berardi-Grassias, 989; Gomez-Garces et al., 994). As many life-threatening infections due to '5. milleri' are empirically treated with an antibiotic regimen including a /Mactam antibiotic (Jacobs et al., 994), resistance to these agents may have serious consequences. Quinn et al. (988) report on a fatal case of post-neurosurgical meningitis due to a penicillin-resistant '5. milleri' strain (MIC 4 mg/l); resistance in this strain was shown to be mediated by alterations of penicillin-binding proteins. The MIC values for gentamicin in the present study are in line with those found in previous studies that used comparable methodologies (Gossling, 988). To our knowledge, high-level resistance to aminoglycosides has not yet been demonstrated in 'S. milleri'. The susceptibilities to clindamycin, erythromycin, and roxithromycin in the present study were similar to the data reviewed by Gossling (988). Resistance to the erythromycin-lincomycin-clindamycin group is transferable among the streptococci (Horaud, Le Bouguenec & Pepper, 985). One recent study from Spain reported resistance rates of 4.3% and.8% for erythromycin and clindamycin respectively (Gomez-Garces et al., 994). Although the authors did not specify the number of isolates that displayed combined resistance to these agents, these findings argue for periodic review of susceptibilities and invite a study of the mechanisms responsible for the resistance. The MIC concentrations of doxycycline for the present collection of isolates are in line with the data compiled for the tetracycline antibiotics (Gossling, 988). Local differences in susceptibility are again illustrated by the above-cited Spanish report (Gomez-Garces et al., 994), that quoted a resistance rate of 37.% for tetracycline. As in the present study, S. anginosus was noted to be more frequently resistant to tetracycline. Downloaded from https://academic.oup.com/jac/article-abstract/37//37/753356 on 0 January 08
374 J. A. Jacobs and E. E. Stobberingh As reported in other studies (LePennec & Berardi-Grassias, 989; Gomez-Garces et al., 994), the MIC values for ciprofloxacin and ofloxacin were close to the breakpoint susceptibility concentration. The glycopeptide antibiotics have been shown to possess excellent in-vitro activity (LePennec & Berardi-Grassias, 989; Gomez-Garces et al., 994) and the in-vitro Table. Activity of the antibiotics tested against the three species of the 'S. milleri' group Species (n)/ antibiotic 50% MIC (mg/l) 90% range Percentage of strains susceptible at breakpoint concentration" S. anginosus (/) = 5) 6 8 3 0.05-0.5-0.5 - ->56 ->56 ->56-3 4-3 - -0.5 0.5-^ 99. 77.3 97. 95.6 97. 76. 4.0 73.0 99.6 S. constellatus (n = 30) 6 0.5 4 3 0.05-0.5 - - ->56 ->56-3 -6 4-3 - -0.5 0.5-0.5-97.7 4.5 96.9 97.7 96.9 80. 6. 93.0 S. intermedius (n = 4) 6 0 5 0.5 3 0.05- - < -0.5 - - - -6 4-64 - - 97.6 85.4 90..0 8 9 "See text for details. Downloaded from https://academic.oup.com/jac/article-abstract/37//37/753356 on 0 January 08
Antimicrobial susceptibility of '5. milleri' 375 activity of teicoplanin has been reported to be superior to that of vancomycin (LePennec & Berardi-Grassias, 989). In conclusion, for the antibiotics studied, we found resistance amongst 'S. milleri' group rare. However, as uniform susceptibility cannot be assumed and as these organisms may be implicated in life-threatening infections, susceptibility studies should be done on isolates recovered from normally sterile body sites. Resistance to clindamycin and erythromycin was limited to S. anginosus and S. constellatus, and doxycycline resistance was more frequently found among S. anginosus. The high resistance rates to erythromycin, clindamycin and doxycycline reported by others (Gomez-Garces et al., 994) contrast with the data in the present study but warrant periodic review of local antimicrobial susceptibility of 'S. milleri'. References G6mez-Garces, J. L., Alos, J. I. & Cogollos, R. (994). Bacteriologic characteristics and antimicrobial susceptibility of 70 clinically significant isolates of Streptococcus milleri group. Diagnostic Microbiology and Infectious Diseases 9, 69-73. Gossling, J. (988). Occurrence and pathogenicity of the Streptococcus milleri group. Reviews of Infectious Diseases 0, 57-85. Horaud, T., Le Bouguenec, C. & Pepper, K. (985). Molecular genetics of resistance to macrolides, lincosamides and streptogamin B (MLS) in streptococci. Journal of Antimicrobial Chemotherapy 6, Suppl. A, -35. Jacobs, J. A., Pietersen, H. G., Stobberingh, E. E. & Soeters, P. B. (994). Bacteremia involving the "Streptococcus milleri" group: analysis of 9 cases. Clinical Infectious Diseases 9, 704-3. LePennec, M. P. & Berardi-Grassias, L. (989). In-vitro activity of 3 antibiotics against clinical isolates of Streptococcus milleri. Journal of Antimicrobial Chemotherapy 4, 68-9. National Committee for Clinical Laboratory Standards. (994). Methods for Dilution Antimicrobial Susceptibility Test for Bacteria that Grow Aerobically, 3rd edition; Approved Standard M7-A3. NCCLS, Villanova, PA. Potgieter, E., Carmichael, M., Koornhof, H. J. & Chalkley, L. J. (99). In vitro antimicrobial susceptibility of viridans streptococci isolated from blood cultures. European Journal of Clinical Microbiology and Infectious Diseases, 543-6. Quinn, J. P., DiVincenzo, C. A., Lucks, D. A., Luskin, R. L., Shatzer, K. L. & Lenner, S. A. (988). Serious infections due to penicillin-resistant strains of viridans streptococci with altered penicillin-binding proteins. Journal of Infectious Diseases 57, 764-9. Whiley, R. A. & Beighton, D. (99). Emended descriptions and recognition of Streptococcus anginosus, Streptococcus inlermedius and Streptococcus anginosus as distinct species. International Journal of Systematic Bacteriology 4, -5. Whiley, R. A., Fraser, H., Hardie, J. M. & Beighton, D. (990). Phenotypic differentiation of Streptococcus intermedius, Streptococcus constellatus,and Streptococcus anginosus strains within the "Streptococcus milleri" group. Journal of Clinical Microbiology 8, 497-50. (Received March 995; accepted September 995) Downloaded from https://academic.oup.com/jac/article-abstract/37//37/753356 on 0 January 08