Non-susceptibility trends among enterococci and non-pneumococcal streptococci from bacteraemias in the UK and Ireland,

Transcription

1 Journal of Antimicrobial Chemotherapy (2008) 62, Suppl. 2, ii75 ii85 doi: /jac/dkn354 Non-susceptibility trends among enterococci and non-pneumococcal streptococci from bacteraemias in the UK and Ireland, Derek F. J. Brown 1 *, Russell Hope 2, David M. Livermore 2, Geraldine Brick 2, Karen Broughton 3, Robert C. George 3 and Rosy Reynolds 4 on behalf of the BSAC Working Parties on Resistance Surveillance 1 Health Protection Agency, Clinical Microbiology and Public Health Laboratory, Addenbrooke s Hospital, Cambridge CB2 2QW, UK; 2 Antibiotic Resistance Monitoring and Reference Laboratory, Health Protection Agency, 61 Colindale Avenue, London NW9 5HT, UK; 3 Respiratory and Systemic Infection Laboratory, Health Protection Agency, 61 Colindale Avenue, London NW9 5HT, UK; 4 Department of Medical Microbiology, Southmead Hospital, Bristol BS10 5NB, UK Objectives: To describe the current patterns and trends in antimicrobial susceptibility in enterococci and streptococci (excepting pneumococci) from bacteraemia in the UK and Ireland from 2001 to Methods: In each year , blood culture isolates were collected by 25 laboratories distributed across the UK and Ireland. In total, there were 1408 isolates of enterococci, 1332 of b-haemolytic streptococci and 1012 of a- and non-haemolytic streptococci. A single central laboratory re-identified the isolates and measured MICs by the BSAC agar dilution method. Results: The prevalence of reduced susceptibility in streptococci and enterococci did not change significantly for most antibiotics, but trends were noted to increased ampicillin, imipenem and vancomycin resistance in Enterococcus faecium. The prevalence of reduced susceptibility to macrolides and tetracycline in streptococci, to tetracycline and gentamicin (high level) in enterococci and to b-lactams and glycopeptides in E. faecium were all high, with some differences in the prevalence among species or groups. Conclusions: Reduced susceptibility to some antimicrobial agents among streptococci and enterococci remains common and continued surveillance is warranted. Keywords: bacteraemia, antimicrobial agents, resistance, epidemiology Introduction The streptococci are a heterogeneous group of organisms including b-haemolytic, a- (viridans group) and non-haemolytic organisms, 1 together with the pneumococci, which are dealt with elsewhere in this supplement. 2 b-haemolytic streptococci in groups A, C and G cause a variety of deep or invasive soft tissue infections such as erysipelas, cellulitis and necrotizing fasciitis, whereas group B b-haemolytic streptococci are wellrecognized as a cause of invasive neonatal infections. Alphaand non-haemolytic streptococci are commensals of the oral cavity, gastrointestinal tract and female genital tract and are frequently found on the skin; they may cause transient bacteraemia in patients without symptoms and often represent contamination in blood cultures. However, a- and non-haemolytic streptococci are also among the most frequent causes of endocarditis 3,4 and may cause life-threatening infections in neutropenic cancer patients. 5,6 Around 3% of the bacteraemias reported as clinically significant are caused by a- and non-haemolytic streptococci other than pneumococci. 7 The enterococci are commensal organisms of the gut and are opportunistically pathogenic, particularly in immunocompromised patients who are hospitalized for long periods and in those with serious underlying disease They, too, are a frequent cause of endocarditis. 8,10 Most enterococcal infections are caused by either Enterococcus faecalis or Enterococcus faecium, the latter, in particular, being associated with outbreaks of infection caused by resistant strains in hospitals. 7 Although b-haemolytic streptococci remain susceptible to penicillin, they are suggested to be increasingly resistant to... *Corresponding author. Tel: þ ; Fax: þ ; dfjb2@cam.ac.uk... ii75 # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Brown et al. macrolides, 11,12 whereas a- and non-haemolytic streptococci 6,13 17 may be highly resistant to both penicillins and macrolides. Enterococci are commonly resistant to macrolides and tetracycline and often have high-level resistance to gentamicin, precluding synergy with b-lactams or glycopeptides. E. faecium is also generally resistant to penicillins and is the major enterococcal species manifesting resistance to glycopeptides. 8,9 This report reviews data on antimicrobial resistance prevalence in b-, a- and non-haemolytic streptococci (excluding pneumococci) and in enterococci collected as part of the BSAC Bacteraemia Resistance Surveillance Programme over the period Materials and methods Participating clinical laboratories were asked to collect up to 10 consecutive isolates per year of each of enterococci, b-haemolytic streptococci and a- and non-haemolytic streptococci (other than pneumococci). The procedures for collection, centralized identification and susceptibility testing of isolates, and for statistical analysis, are described in detail elsewhere in this Supplement. 18,19 In analysis, when there was an intermediate category the results in the intermediate and resistant categories were combined as non-susceptible. Results Isolates and patients characteristics From 2001 to 2006, the BSAC Bacteraemia Resistance Surveillance Programme collected 1332 b-haemolytic streptococci, 1012 a- and non-haemolytic streptococci and 1408 enterococci (Table 1). Male patients contributed 51% of the b-haemolytic streptococcal isolates, 56% of the a- and nonhaemolytic streptococci and 58% of the enterococci. Some differences were seen between bacterial groups in relation to patients characteristics (Table 2) as follows: Age. There was no overall marked association of age with isolation of particular groups of bacteria. However, group B streptococci were most commonly isolated from younger patients, with 31.1% from patients aged,1 year, compared with 3.1% for group A, 0% for group C and 0.7% for group G. Care setting. In total 27.1% of the b-haemolytic streptococci, 46.3% of the a- and non-haemolytic streptococci and 68.7% of the enterococci were from patients in hospital for.48 h. Referring specialty. b-haemolytic streptococci were most commonly isolated from patients in accident and emergency, general medicine and paediatrics (mostly group B streptococci). Alphaand non-haemolytic streptococci were most commonly isolated from patients in haematology/oncology, as well as in general medicine and accident and emergency. Enterococci were also commonly from patients in general medicine, haematology/ oncology and surgery and from intensive care unit patients. Focus of infection. Where bacteraemia due to b-haemolytic streptococci could be associated with a focus of infection, skin and soft tissues were prominent, whereas bacteraemia due to a- and Table 1. Isolates collected in the BSAC Bacteraemia Resistance Surveillance Programme Group/organism number Isolates % group b-haemolytic streptococci group A group B group G group C a- and non-haemolytic streptococci S. oralis S. anginosus S. sanguinis S. bovis biotype II Streptococcus mitis Streptococcus parasanguinis Streptococcus vestibularis Streptococcus intermedius S. bovis I Streptococcus gordonii Streptococcus salivarius Streptococcus mutans Streptococcus constellatus Streptococcus sobrinus Enterococcus spp. E. faecalis E. faecium Enterococcus durans Enterococcus gallinarum Enterococcus casseliflavus Enterococcus avium Enterococcus hirae Other Enterococcus spp non-haemolytic streptococci was associated with subacute bacterial endocarditis or the gastrointestinal tract ( particularly for Streptococcus anginosus and Streptococcus bovis biotype II). In addition, Streptococcus oralis was particularly associated with line infections. Enterococcal bacteraemia was commonly associated with line infections, the gastrointestinal or genitourinary tracts. Non-susceptibility trends The prevalence of non-susceptibility varied considerably from year to year for some organism agent combinations. This is probably related to small sample sizes, variability in testing and the fact that some breakpoints are close to the MIC distributions of the susceptible wild-type populations. b-haemolytic streptococci Groups A, B and G composed most of the 1332 isolates of b-haemolytic streptococci, with 553, 414 and 303 isolates, respectively; the remaining 62 isolates were of group C. There was little evidence of significant trends in susceptibility over the ii76

3 Table 2. Characteristics of patients in the BSAC Bacteraemia Resistance Surveillance Programme Patient location at the time of bacteraemia Age (%) a Sex (%) a distribution a Group of organisms non-hospital b hospital,48 h hospital.48 h male female age (years) % Major specialties (%) c Major sources (%) c ii77 b-haemolytic streptococci accident and emergency (26.0) unknown (40.2) (n ¼ 1332) general medicine (22.2) skin and soft tissue (29.4) paediatrics (11.0) a- and non-haemolytic streptococci general medicine (25.4) unknown (37.1) (n ¼ 1015) accident and emergency (15.3) subacute bacterial endocarditis (20.0) haematology/oncology (12.9) gastrointestinal tract (16.0) þ 16.1 Enterococcus spp general medicine (18.8) unknown (33.8) (n ¼ 1047) surgery (17.2) line associated (25.1) intensive care (13.9) gastrointestinal tract (17.0) haematology/oncology (12.4) genitourinary tract (11.2) þ 18.1 Non-susceptibility in enterococci and streptococci a Percentages do not total 100 because data were missing for a few isolates. b Community or outpatients. c Sources and specialties are listed when they accounted for 10% of the cases. Downloaded from at Pennsylvania State University on March 3, 2014

4 Table 3. Non-susceptibility trends among b-haemolytic streptococci Brown et al. Breakpoint (mg/l) a Intermediate/resistant b isolates (%) in indicated year Agent S R P value c Cefotaxime NA Clindamycin Daptomycin 1 1 NT NT 0 NT 0 0 NA Ertapenem NT NA Erythromycin Imipenem d NA Linezolid /0 0/0 0/0 12.9/0 0/0 0/ Meropenem d 2 2 NT NT NT NT NT 0 NA Minocycline d NT 0/ / / /51.1 0/ Penicillin NA Teicoplanin d NA Tetracycline Tigecycline NT NT e NT e 1.3/0 2.2/0 0.9/ Vancomycin d NA NT, not tested; NA, not applicable. a Breakpoints BSAC unless otherwise indicated. b Where there is no intermediate category, only the percent resistant is given. c Test for trend (logistic regression with random effect for centre). d EUCAST breakpoints. 58,59 e Data not presented because raised MICs and consequent significant non-susceptible prevalence were artefactually high because the testing requirements for tigecycline were not understood at that time. period (Table 3). From 2001 to 2006, tetracycline resistance was frequent, with 48.6% resistant overall and was highest, at 82.4%, in group B isolates (Table 4). Other significant differences in non-susceptibility among the b-haemolytic streptococcal groups were seen with erythromycin (higher in groups B and G, with 15.2% and 16.8% resistant, respectively), clindamycin (higher in groups B and G, with 5.3% and 4.3% resistant, respectively) and tigecycline (higher in groups C and G, with 5.4% and 5.1% intermediate, 0% and 0% resistant, respectively) (Table 4). Table 4. Non-susceptible isolates (%) among different groups of b-haemolytic streptococci (breakpoints as in Table 3) Intermediate/resistant a (%) Agent N (total) total group A group B group C group G P value b Cefotaxime NA Clindamycin Daptomycin NA Ertapenem NA Erythromycin , Imipenem NA Linezolid /0 2.4/0 1.0/0 4.8/0 3.3/ Meropenem NA Minocycline / /18.0 0/81.6 0/ /35.4, Penicillin NA Teicoplanin NA Tetracycline , Tigecycline /0 0/0 5.4/0 5.1/0 NA Vancomycin NA NA, not applicable. a Where there is no intermediate category, only the percent resistant is given. b Test for difference among groups. ii78

5 Non-susceptibility in enterococci and streptococci All isolates were susceptible to b-lactams, glycopeptides and daptomycin, and the 2.3% of the isolates non-susceptible to linezolid were all categorized as intermediate (Table 4). The newer agents, doripenem (MICs 0.03 mg/l), ceftobiprole (MICs 0.25) and telavancin (MICs 0.25 mg/l), were highly potent against the b-haemolytic streptococci (data not shown). Alpha- and non-haemolytic streptococci Isolates of 14 different species of a- and non-haemolytic streptococci were collected; however, 65% of the 1012 isolates belonged to just four species, S. oralis (294 isolates), S. anginosus (151), Streptococcus sanguinis (124) and S. bovis biotype II (88) (Table 1). There was little evidence of significant trends in nonsusceptibility over the period (Table 5), but there were some significant differences in non-susceptibility among species (Table 6). The prevalence of penicillin resistance was the highest in S. sanguinis (34.7%) and lowest in S. anginosus (1.3%). Penicillin resistance in all species was at a low level, with most MICs mg/l and none.8 mg/l. Erythromycin resistance varied significantly from year to year, but the MIC distribution was broadly spread and the breakpoint does not clearly segregate resistant and susceptible subpopulations (Figure 1). The majority of the wild-type isolates were distributed around a mode erythromycin MIC of 0.12 mg/l. Among isolates with reduced susceptibility, there was a highly resistant population with erythromycin MICs 256 mg/l and a substantial proportion with MICs broadly spread, but mostly in the range 1 8 mg/l. The prevalence of erythromycin resistance was the highest in S. oralis (51.7%) and markedly lower in S. anginosus (17.9%). Most of the isolates with low-level Table 5. Non-susceptibility trends among a- and non-haemolytic streptococci Breakpoint (mg/l) a resistance were S. oralis or species other than S. anginosus, S. bovis biotype II and S. sanguinis. With tetracycline and minocycline, there were strikingly bimodal MIC distributions, with modes for the wild-type populations 0.5 and 0.12 mg/l, respectively, and substantial populations of resistant isolates with modal MICs of 64 and 16 mg/l, respectively. The prevalence of tetracycline resistance was the highest in S. bovis biotype II (76.1%). For tigecycline, there was a unimodal distribution with the modal MIC for all a- and non-haemolytic streptococci at 0.12 mg/l; of those falling in the non-susceptible group, 62 were intermediate (MIC 0.5 mg/l) and only 7 resistant (MIC 1 mg/l). There was no resistance to glycopeptides or linezolid, and only seven isolates (1.4%) were resistant to daptomycin (Table 6), all with MICs just above the breakpoint at 2 mg/l. For doripenem, MICs were distributed over a wide range, from to 1 mg/l, with a mode of mg/l (data not shown). MICs of ceftobiprole were similarly distributed over a wide range, from 0.06 to 4 mg/l, with a mode of 0.06 mg/l (data not shown). MICs of telavancin were mg/l, with a mode of 0.03 mg/l (data not shown). Enterococci Of the 1408 enterococci received, 875 (62%) were E. faecalis and 463 (33%) E. faecium (Table 1). For E. faecalis, there was little evidence of any significant trend in non-susceptibility over time (Table 7). For E. faecium, there was a significant increase in resistance to ampicillin from 86.8% in 2001 to 98.8% in 2006 and a parallel increase in non-susceptibility to imipenem from 89.7% in 2001 to 98.8% in 2006 (Table 7). There was also some indication of an increase in non-susceptibility to vancomycin Intermediate/resistant b isolates (%) in indicated year Agent S R P value c Amoxicillin Cefotaxime Clindamycin Daptomycin d 1 1 NT NT 0.6 NT Erythromycin Linezolid NA Minocycline e NT 1.2/ /23.7 0/ / / Penicillin Tetracycline Teicoplanin NA Tigecycline NT NT f NT f 3.7/0 1.1/0 2.4/ Vancomycin NA NT, not tested; NA, not applicable. a Breakpoints BSAC unless otherwise indicated. b Where there is no intermediate category, only the percent resistant is given. c Test for trend (logistic regression with random effect for centre). d EUCAST breakpoints. 60 e EUCAST breakpoints specified for other streptococci. 59 f Data not presented because raised MICs and consequent significant non-susceptible prevalence were artefactually high because the testing requirements for tigecycline were not understood at that time. ii79

6 Brown et al. Table 6. Non-susceptible isolates (%) among different groups of a- and non-haemolytic streptococci (breakpoints as in Table 5) Intermediate/resistant a (%) Agent N (total) total oralis anginosus sanguinis bovis II other P value b Amoxicillin NA Cefotaxime NA Clindamycin Daptomycin NA Erythromycin , Linezolid NA Minocycline / / / /27.4 0/ /22.0, Penicillin , Tetracycline , Teicoplanin NA Tigecycline c /0 0/0 0/0 4.7/0 0/0 5.7/0 NA Vancomycin NA NA, Not applicable. a Where there is no intermediate category, only the percent resistant is given. b Test for difference among species. c Excluding , see Table 5. Number of isolates Other spp. S. bovis biotype II S. anginosus S. sanguinis S. oralis MIC (mg/l) Figure 1. Distribution of erythromycin MICs (mg/l) for 1012 a- and non-haemolytic streptococci (dotted line indicates breakpoint, S 0.5, R. 0.5 mg/l). and teicoplanin, with higher prevalence in than in earlier years. For these glycopeptide agents, the MICs for most non-susceptible isolates are clearly in the resistant range (vancomycin MIC 32 mg/l for 100% of the non-susceptible isolates and teicoplanin MIC.8 mg/l for 80% of the non-susceptible isolates). There were some significant differences (P, ) in non-susceptibility among species (Table 8), with ampicillin, imipenem, meropenem, vancomycin and teicoplanin nonsusceptibility most prevalent in E. faecium. Conversely, tetracycline resistance was most prevalent in E. faecalis (82%). Vancomycin resistance in enterococci was significantly associated with resistance to ampicillin (P, ), imipenem (P, ) and meropenem (P ¼ ) and with susceptibility to tetracycline (P ¼ ). This was in line with the common pattern of susceptibility of E. faecium, which were mostly ampicillin-resistant and commonly vancomycinresistant. Most vancomycin-resistant enterococci had a VanA (resistant to vancomycin and teicoplanin) phenotype, with nonsusceptibility also to teicoplanin (of 167 vancomycin-non-susceptible isolates of enterococci, 15.0% were intermediate and 67.1% resistant to teicoplanin). Vancomycin non-susceptibility was clearly associated with haematology/ oncology (29.3%) and, to a lesser extent, nephrology (14.7%), compared with intensive care, surgery, general medicine and other sources (10.8, 8.7, 8.5 and 5.0% vancomycin-resistant, respectively). Vancomycin-non-susceptible enterococci were also more likely to be from patients in hospital for.48 h (13.9%) than otherwise (4.3%). There was clear evidence (P, ) that vancomycin and teicoplanin nonsusceptibility varied significantly among centres, with 6 of the 29 centres contributing 54.1% of the vancomycin-resistant isolates. This was not simply related to these six centres providing a high proportion of the isolates of E. faecium, the species where vancomycin resistance was most common, as they contributed only 28.3% of the E. faecium isolates. High-level gentamicin resistance (HLGR; MIC. 128 mg/l) was found in 50.6% of the E. faecalis and in 42.1% of the E. faecium isolates. In E. faecalis, HLGR was significantly more frequent (P, ) in hospital-acquired (57.5% of the 558 isolates) than other bacteraemias (36.9% of the 274 isolates). Differences between specialties in the prevalence of HLGR in E. faecalis (haematology/oncology 65.6%, ICU 59.4%, nephrology 50.4%, surgery 59.2%, general medicine 44.9% and other specialties 41.2%) were not significant when age distributions and rates of hospital acquisition of infections are taken into account. HLGR in E. faecalis varied significantly with age (P, ), being more common in the age groups years (57.3% HLGR) and years (54.1% HLGR) than in isolates from younger (20.3% HLGR) and older (45.3% HLGR) ii80

7 Table 7. Non-susceptible trends among enterococci Non-susceptibility in enterococci and streptococci Breakpoint (mg/l) a Intermediate/resistant b isolates (%) in indicated year Organism/agent S R P value c E. faecalis ampicillin NA gentamicin d imipenem 4 8 0/0 0/0 0/0 2.2/0 0/0 0/0.7 NA linezolid NA meropenem 4 4 NT NT NT NT NT 7.7 NA tetracycline 1 1 NT teicoplanin 4 8 0/2.0 0/ /1.3 0/3.6 0/1.4 0/ tigecycline NT NT e NT e 0/0 0/0 0/ vancomycin 4 8 0/2.0 0/3.4 0/3.8 0/3.6 0/1.4 0/ E. faecium ampicillin gentamicin d imipenem f /88.2 0/88.6 0/ / /95.6 0/ linezolid NA meropenem 4 4 NT NT NT NT NT 100 NA tetracycline 1 1 NT teicoplanin / / / / / / tigecycline NT NT e NT e 0/0 0/0 0/0 NA vancomycin 4 8 0/20.6 0/18.6 0/24.3 0/22.5 0/40.0 0/ NT, not tested; NA, not applicable. a Breakpoints BSAC except where indicated. b Where there is no intermediate category, only the percent resistant is given. c Test for trend (logistic regression with random effect for centre). d High-level gentamicin resistance. e Data not presented because raised MICs and consequent significant non-susceptible prevalence was artefactually high because the testing requirements for tigecycline were not understood at that time. f EUCAST breakpoints. 58 Table 8. Non-susceptible isolates among enterococci (breakpoints as in Table 7) Agent E. faecalis E. faecium Other enterococci N intermediate/resistant a (%) N intermediate/resistant a (%) N intermediate/resistant a (%) P value b Ampicillin , Gentamicin c , Imipenem / / /18.8, Linezolid NA Meropenem , Tetracycline , Teicoplanin / / /8.7, Tigecycline 419 0/ /0 46 0/2.2 NA Vancomycin 875 0/ / /8.7, NA, not applicable. a Where there is no intermediate category, only the percentage resistant is given. b Test for difference between species. c High-level gentamicin resistance. patients. Apparently similar relationships between age and resistance to vancomycin and teicoplanin in E. faecalis were not significant, there being few resistant isolates in any age group. There was very little non-susceptibility (overall,1%) to tigecycline and linezolid (Table 7). For doripenem, MICs were essentially divided into two populations with MICs for ii81

8 Brown et al. E. faecalis mostly mg/l (mode 2 mg/l) and for E. faecium mostly mg/l (mode 256 mg/l) (data not shown). With ceftobiprole there was a similar division between the species, with MICs for E. faecalis mostly mg/l (mode 2 mg/l) and for E. faecium mostly mg/l (mode 32 mg/l). Telavancin MICs were distributed over a wide range from mg/l, with a mode of 0.25 mg/l (data not shown). Telavancin MICs were 1 mg/l for all 383 E. faecalis and E. faecium susceptible to vancomycin, whereas telavancin MICs were.1 mg/l for 73.9% of 69 vancomycin-resistant enterococci (data not shown). Daptomycin MICs were in the range mg/l with.90% inhibited by 1 mg/l (data not shown). Discussion Among b-haemolytic streptococci, there was no evidence of any change in the prevalence of resistance over time for any of the agents studied. Penicillins are usually the antibiotics of choice for treatment of infections caused by b-haemolytic streptococci, and no penicillin resistance was seen in this study over the period Macrolides are mostly used to treat respiratory infections and superficial skin and soft tissue infections in patients with b-lactam hypersensitivity, and macrolide resistance in b-haemolytic streptococci is a problem in many parts of the world. 20 In this study, erythromycin resistance was seen in 5%, 15%, 7% and 17% of the group A, B, C and G isolates, respectively. A higher prevalence of erythromycin resistance in groups B and G than in groups A and C has also been reported in other studies. 15 The prevalence of tetracycline resistance among b-haemolytic streptococci was high at 19%, 82%, 37% and 59% in groups A, B, C and G, respectively. Resistance was most prevalent among group B isolates, which were often associated with younger patients and paediatric wards. However, it is not clear why tetracycline resistance should be more common in group B isolates. Tetracyclines should not be used to treat infections in infants, in whom most of these infections occurred, but group B streptococci in infants are usually acquired from their mothers, who may have been previously treated with tetracyclines, although tetracyclines are contraindicated in the second and third trimesters of pregnancy. Tetracyclines are most widely used in the community, and over 70% of the bacteraemias with b-haemolytic streptococci were community-associated (nonhospital or in hospital for,48 h). A high prevalence of tetracycline resistance was also reported in the UK Health Protection Agency LabBase bacteraemia surveillance programme, which is based on routine clinical laboratory reports. 7,21 Although 4.7% of the group A streptococci were resistant to erythromycin, only 0.4% were resistant to clindamycin, which may be used to treat severe group A streptococcal infections. However, the isolates were not tested for dissociated resistance to clindamycin. Linezolid non-susceptibility (12.9%) among b-haemolytic streptococci in 2004 is related to a marginally broader spread in the MIC distribution in that year combined with the close proximity of the wild-type distribution to the breakpoints (S 2and R. 4 mg/l). This resulted in 29 isolates from 2004 being reported as intermediate. A wide range of species of a- and non-haemolytic streptococci were collected (14 species or subgroups). There were no significant trends in non-susceptibility among this group, although resistance to some agents was common and there were some significant differences in non-susceptibility among species. Penicillin resistance was 16.1% overall and highest in S. sanguinis (33.9%). Penicillin resistance in a- and nonhaemolytic streptococci has been reported for many years, although resistance prevalence has varied widely in different 6,15,17,22 26 studies, partly because of application of different breakpoints, examination of different populations and differences in the nomenclature. Although 16.1% of the a- and nonhaemolytic streptococci were found to be penicillin-resistant in this study, MICs were no higher than 8 mg/l, and endocarditis with these bacteria would potentially be treatable with combinations of penicillin and an aminoglycoside. 27 Although there was no significant trend in erythromycin resistance among a- and non-haemolytic streptococci over time, resistance was common (overall 38.2%) and varied markedly between 23.8% and 58.5% from year to year. The distribution of MICs above those of the wild-type-susceptible population (MIC 0.5 mg/l) was broadly spread, with a highly resistant population (MIC 256 mg/l) and a substantial number of isolates with lower MICs, broadly spread down to the resistance breakpoint (.0.5 mg/l). As a consequence, the breakpoint cannot clearly distinguish a discrete resistant population, and this resulted in marked variation in resistance prevalence from year to year. Different levels of resistance to erythromycin probably relate to different mechanisms of resistance or combinations of mechanisms present in the organisms. 28 Erythromycin resistance in a- and non-haemolytic streptococci has been reported to be common worldwide. 6,15,17,22,26,28 It varied among species and was especially high in S. oralis (51.7%), as reported previously, 28,29 and low in S. anginosus (17.9%). The prevalence of tetracycline resistance was also high (33.9% overall), as reported by others. 22,25 Increasing resistance to antibiotics in enterococci has been reported, and glycopeptide resistance has been a particular concern. We found that vancomycin non-susceptibility increased in E. faecium, from20% in to.30% in , whereas vancomycin non-susceptibility in E. faecalis has remained relatively constant at 3%. Vancomycin resistance in enterococci increased markedly through the 1990s, with UK LabBase bacteraemia reports indicating 5% resistance in E. faecalis and 24% in E. faecium by The higher vancomycin resistance prevalence reported for E. faecalis in the early LabBase data may be due to the misidentification of E. faecium as E. faecalis in routine clinical laboratories, 7,33 (a view supported by a 10% ampicillin resistance rate among isolates reported as E. faecalis), and a more recent report on LabBase data indicates vancomycin resistance in E. faecalis at 3% over the period For E. faecium, the LabBase data show a reduction in the resistance prevalence during from 25% to 17%, followed by an increase to 24% in The lower prevalence of vancomycin resistance in E. faecium seen in the LabBase data compared with the data in this study may relate to the different sampling base, which includes all clinically significant bacteraemia isolates from laboratories in England and Wales in the LabBase data rather than a sample of isolates from 25 laboratories in this sentinel study. Major differences in the prevalence of vancomycin resistance in different hospitals lead to bias in any sentinel study according to how many centres with high vancomycin resistance rates are included. The detection methods used may also be ii82

9 Non-susceptibility in enterococci and streptococci a contributory factor as a variety of unrecorded routine methods are used for the LabBase data compared with a defined MIC method in this study. Increasing vancomycin resistance, at a much higher level than in the UK, has been reported for North America in the SENTRY surveillance programme, which showed vancomycin non-susceptibility in E. faecium increasing from 40% in 1997 to 61% in 2002, whereas vancomycin non-susceptibility in E. faecalis remained at 1% to 4%. 34 Vancomycin resistance in E. faecium is much less common in Europe 34,35 than in North America, but there are marked differences among countries. The European Antimicrobial Resistance Surveillance System reported high vancomycin resistance prevalence in 2006 in Greece (42%), Ireland (36%) and Portugal (26%), although prevalence remained below 1% in several countries. 35 The changes in the prevalence of vancomycin nonsusceptibility in E. faecium in this study were mirrored by a similar but slightly lower prevalence of non-susceptibility to teicoplanin, indicating that most glycopeptide non-susceptibility is of the VanA phenotype rather than the VanB phenotype (vancomycin-resistant and teicoplanin-susceptible). This is in line with the predominance of the VanA phenotype in most countries. 30,36 39 As previously reported, vancomycin-nonsusceptible enterococci were particularly associated with haematology/oncology and nephrology wards. 31,40 42 The association of vancomycin non-susceptibility with particular hospitals in this study was not related to the specialties with higher risk for glycopeptide-resistant enterococci being at those centres, nor to E. faecium being isolated more frequently in those centres. Resistance to ampicillin in E. faecalis was not seen in this study. There are very few confirmed reports of resistance to ampicillin in clinical isolates of E. faecalis, and such isolates have been b-lactamase producers. 43 Reports of ampicillin resistance in E. faecalis are more likely to be attributable to the misidentification of E. faecium as E. faecalis. 8,33,44 Resistance to ampicillin and carbapenems in E. faecium is very common, with the prevalence of non-susceptibility to ampicillin and imipenem increasing here from 86.8% and 89.7%, respectively, to 98.8% for both agents between 2001 and These apparent changes should be viewed with caution as the UK LabBase data for the same period showed no trend in ampicillin resistance. 45 Combinations of aminoglycosides with b-lactam agents (or glycopeptides in the cases of b-lactam hypersensitivity or resistance to ampicillin) are the treatment of choice for enterococcal endocarditis. 27 The synergistic bactericidal activity of the combinations is, however, lost in isolates with HLGR. 8,46 HLGR prevalence increased in enterococci in the 1980s and 1990s. 8,9,47,48 In this study, there was no significant trend in HLGR prevalence over the period but resistance was very common, at 50.6% in E. faecalis and 42.1% in E. faecium. As reported by others, 8,47,48 HLGR, particularly in E. faecalis, was associated with acquisition in hospital and varied significantly among age groups. Resistance to tetracycline in enterococci was common, as in streptococci, but tigecycline retained activity against almost all enterococci, a- and b-haemolytic streptococci in this study, as reported by others, 49 and tigecycline non-susceptible isolates mostly had intermediate susceptibility. Non-susceptibility to the other newer agents, linezolid and daptomycin, was also uncommon, as reported by others. 49 For agents included more recently in the surveillance programme, there are insufficient data to examine trends, and clinical breakpoints were not available. However, MIC distributions indicate that activity was similar to that reported previously. As with other b-lactam agents, doripenem and ceftobiprole showed good activity against b-haemolytic streptococci, a range of activity against non- and a-haemolytic streptococci, moderate activity against E. faecalis and poor activity against E. faecium Telavancin showed good activity against b-haemolytic, non- and a-haemolytic streptococci and a range of activity against enterococci, with reduced activity against most vancomycin-resistant enterococci. 55,56 The results reported here from the first 6 years of the BSAC Bacteraemia Resistance Surveillance Programme show that the prevalence of non-susceptibility in streptococci (excluding pneumococci) and enterococci has not changed significantly for most of the antibiotics, but some possible trends were noted in increased ampicillin and vancomycin non-susceptibility in E. faecium. The prevalence of non-susceptibility to macrolides and tetracycline in streptococci, to tetracycline and gentamicin (high level) in enterococci and to b-lactams and glycopeptides in E. faecium remains high. Conversely, there is so far little evidence of increasing non-susceptibility to the newer agents such as linezolid, daptomycin and tigecycline. Statistical analysis has highlighted the risk factors for non-susceptibility, such as hospital acquisition of infection, therapeutic specialty and age. Continued surveillance is warranted and will increase the value of the dataset over time. Acknowledgements We are grateful to all who have contributed to the success of the BSAC Resistance Surveillance Project, in particular, the many laboratories that have collected isolates and all who have played a part in testing them [see page ii10 (Acknowledgements)]. Additional information on the isolates collected in the Project is available on the BSAC surveillance web site ( or through a link on the BSAC homepage See page ii12 (Publications) for a full list of previous publications from the Project, which may include parts of the information presented here. Funding The BSAC Bacteraemia Resistance Surveillance Programme has received financial support from AstraZeneca, Basilea, Cubist, Johnson & Johnson, Merck Sharp & Dohme, Novartis, Pfizer, Theravance and Wyeth, or their predecessors. The BSAC funds the work of the Resistance Surveillance Coordinator (R. R.) and Resistance Surveillance Working Party. Transparency declarations This article is part of a Supplement sponsored by the British Society for Antimicrobial Chemotherapy. D. M. L. has shareholdings in AstraZeneca, Pfizer, Schering Plough and GlaxoSmithKline and has accepted grants, speaking invitations and conference invitations from most major pharmaceutical companies. R. C. G. has received grant funding and/or speaking and conference invitations from Wyeth and GSK on ii83

10 Brown et al. vaccine-related topics. D. M. L. and R. C. G. are both also employed within the UK public sector and are influenced by the HPA s views of antibiotic prescribing and usage. All other authors have none to declare. References 1. Spellerberg B, Brandt C. Streptococcus. In: Murray PR, Baron EJ, Jorgensen JH et al., eds. Manual of Clinical Microbiology, 9th edn. Washington, DC: ASM Press, 2007; Farrell DJ, Felmingham D, Shackcloth J et al. Non-susceptibility trends and serotype distributions among Streptococcus pneumoniae from community-acquired respiratory tract infections and from bacteraemias in the UK and Ireland, 1999 to J Antimicrob Chemother 2008; 62: Suppl 2: ii Wilson WR, Karchmer AW, Dajani AS et al. Antibiotic treatment of adults with infective endocarditis due to streptococci, enterococci, staphylococci, and HACEK microorganisms. JAMA 1995; 274: Mylonakis E, Calderwood SB. Infective endocarditis in adults. N Engl J Med 2001; 345: Tunkel AR, Sepkowitz KA. Infections caused by viridans streptococci in patients with neutropenia. Clin Infect Dis 2002; 34: Lyytikäinen O, Rautio M, Carlson P et al. Nosocomial bloodstream infections due to viridans streptococci in haematological and non-haematological patients: species distribution and antimicrobial resistance. J Antimicrob Chemother 2004; 53: Reacher MH, Shah A, Livermore DM et al. Bacteraemia and antibiotic resistance of its pathogens reported in England and Wales between 1990 and 1998: trend analysis. BMJ 2000; 320: Murray BE. The life and times of the enterococcus. Clin Microbiol Rev 1990; 3: Murray BE. Diversity among multidrug-resistant enterococci. Emerg Infect Dis 1997; 4: Teixeira LM, da Gloria Siqueira Carvalho M, Facklam RR. Enterococcus. In: Murray PR, Baron EJ, Jorgensen JH et al. eds. Manual of Clinical Microbiology, 9th edn. Washington, DC: ASM Press, 2007; Pfaller MA, Jones RN. In vitro evaluation of contemporary b-lactam drugs tested against viridans group and b-haemolytic streptococci. Diagn Microbiol Infect Dis 1997; 27: Renneberg J, Niemann LL, Gutschik E. Antimicrobial susceptibility of 278 streptococcal blood isolates to seven antimicrobial agents. J Antimicrob Chemother 1997; 39: Carratala J, Alcaide F, Fernandez-Sevilla A et al. Bacteremia due to viridans streptococci that are highly resistant to penicillin: increase among neutropenic patients with cancer. Clin Infect Dis 1995; 20: Doern GV, Ferraro MJ, Brueggemann AB et al. Emergence of high rates of antimicrobial resistance among viridans group streptococci in the United States. Antimicrob Agents Chemother 1966; 40: Pfaller MA, Jones RN, Marshall SA et al. Nosocomial streptococcal blood stream infections in the SCOPE program: species occurrence and antimicrobial resistance. Diagn Microbiol Infect Dis 1997; 29: Alcaide F, Benítez MA, Carratalà J et al. In vitro activities of the new ketolide HMR 3647 (telithromycin) in comparison with those of eight other antibiotics against viridans group streptococci isolated from blood of neutropenic patients with cancer. Antimicrob Agents Chemother 2001; 45: Reinert RR, von Eiff C, Kresken M et al. Nationwide German multicenter study on the prevalence of antibiotic resistance in streptococcal blood isolates from neutropenic patients and comparative in vitro activities of quinupristin dalfopristin and eight other antimicrobials. J Clin Microbiol 2001; 39: Reynolds R, Hope R, Williams L et al. Survey, laboratory, and statistical methods for the BSAC Resistance Surveillance Programmes. J Antimicrob Chemother 2008; 62 Suppl 2: ii Reynolds R, Lambert PC, Burton PR et al. Analysis, power and design of antimicrobial resistance surveillance studies, taking account of inter-centre variation and turnover. J Antimicrob Chemother 2008; 62 Suppl 2: ii Robinson DA, Sutcliffe JA, Tewodros W et al. Evolution and global dissemination of macrolide-resistant group A streptococci. Antimicrob Agents Chemother 2006; 50: Health Protection Agency. Antimicrobial Resistance in England, Wales and Northern Ireland, London: Health Protection Agency, (3 July 2008, date last accessed). 22. Fluit AC, Jones ME, Schmitz F et al. Antimicrobial susceptibility and frequency of occurrence of clinical blood isolates in Europe from the SENTRY antimicrobial surveillance program, Clin Infect Dis 2000; 30: Kennedy HF, Gemmell CG, Bagg J et al. Antimicrobial susceptibility of blood culture isolates of viridans streptococci: relationship to a change in empirical antibiotic therapy in febrile neutropenia. J Antimicrob Chemother 2001; 47: Wisplinghoff H, Bischoff T, Tallent SM et al. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004; 39: Han XY, Kamana M, Rolston KVI. Viridans streptococci isolated by culture from blood of cancer patients: clinical and microbiologic analysis of 50 cases. J Clin Microbiol 2006; 44: Westling K, Julander I, Ljungman P et al. Viridans streptococci in blood culture isolates in a Swedish university hospital: antibiotic susceptibility and identification of erythromycin resistance genes. Int J Antimicrob Agents 2006; 28: Elliott TS, Foweraker J, Gould FK et al. Guidelines for the antibiotic treatment of endocarditis in adults: report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 2004; 54: Ergin A, Ercis S, Hasçelik G. Macrolide resistance mechanisms and in vitro susceptibility patterns of viridans group streptococci isolated from blood cultures. J Antimicrob Chemother 2006; 57: Smith A, Jackson MS, Kennedy H. Antimicrobial susceptibility of viridans group streptococcal blood culture isolates to eight antimicrobial agents. Scand J Infect Dis 2004; 36: Woodford N, Johnson AP, Morrison D et al. Current perspectives on glycopeptide resistance. Clin Microbiol Rev 1995; 8: Cetinkaya Y, Falk P, Mayhall CG. Vancomycin-resistant enterococci. Clin Microbiol Rev 2000; 13: Chavers LS, Moser SA, Benjamin WH et al. Vancomycinresistant enterococci: 15 years and counting. J Hosp Infect 2003; 53: Brown DFJ, Brown N, Cookson B et al. National Glycopeptide- Resistant Enterococcal Bacteraemia Surveillance Working Group Report to the Department of Health, August J Hosp Infect 2006; 62: S Biedenbach DJ, Moet GJ, Jones RN. Occurrence and antimicrobial resistance pattern comparisons among bloodstream infection isolates from the SENTRY antimicrobial surveillance program ( ). Diagn Microbiol Infect Dis 2004; 50: ii84

11 Non-susceptibility in enterococci and streptococci 35. European Antimicrobial Resistance Surveillance System. EARSS Annual Report EARSS%202006%20Def_tcm pdf (1 May 2008, date last accessed). 36. Goossens H. Spread of vancomycin-resistant enterococci: differences between the United States and Europe. Infect Control Hosp Epidemiol 1998; 19: Kawalec M, Gniadkowski M, Hryniewicz W. Outbreak of vancomycin-resistant enterococci in a hospital in Gdańsk, Poland, due to horizontal transfer of different Tn1546-like transposon variants and clonal spread of several strains. J Clin Microbiol 2000; 38: Scagnelli M, Pellizer G, de Lalla F et al. Epidemiological analysis of vancomycin-resistant enterococci in a large tertiary-care hospital in Northern Italy. Eur J Clin Microbiol Infect Dis 2001; 20: Deshpande LM, Fritsche TR, Moet GJ et al. Antimicrobial resistance and molecular epidemiology of vancomycin-resistant enterococci from North America and Europe: a report from the SENTRY antimicrobial surveillance program. Diagn Microbiol Infect Dis 2007; 58: Lautenbach E, Bilker WB, Brennan PJ. Enterococcal bacteraemia: risk factors for vancomycin resistance and predictors of mortality. Infect Control Hosp Epidemiol 1999; 20: Byers KE, Anglim AM, Anneski CJ et al. A hospital epidemic of vancomycin-resistant Enterococcus: risk factors and control. Infect Control Hosp Epidemiol 2001; 22: Beltrami EM, Singer DA, Fish L et al. Risk factors for acquisition of vancomycin-resistant enterococci among patients on a renal ward during a community hospital outbreak. Am J Infect Control 2000; 28: Murray BE. b-lactamase-producing enterococci. Antimicrob Agents Chemother 1992; 36: EARSS Management Team. Quality assessment exercise EARSS EARSS Newslett 2003; 5: Health Protection Agency. Enterococcus spp. and group D Streptococcus spp. bacteraemia in England, Wales, and Northern Ireland: Health Protection Report 2007; 1: org.uk/hpr/archives/2007/hpr2407.pdf (3 July 2008, date last accessed). 46. Mederski-Samoraj BD, Murray BE. High-level resistance to gentamicin in clinical isolates of enterococci. JInfectDis1983; 147: Johnson AP, Warner M, Woodford N et al. Antibiotic resistance among enterococci causing endocarditis in the UK: analysis of isolates referred to a reference laboratory. BMJ 1988; 317: Schouten MA, Voss A, Hoogkamp-Korstanje JA. Antimicrobial susceptibility patterns of enterococci causing infections in Europe. The European VRE Study Group. Antimicrob Agents Chemother 1999; 43: Sader HS, Jones RN, Stilwell MG et al. Tigecycline activity tested against 26,474 bloodstream infection isolates: a collection from 6 continents. Diagn Microbiol Infect Dis 2005; 52: Ge Y, Wikler MA, Sahm DF et al. In vitro antimicrobial activity of doripenem, a new carbapenem. Antimicrob Agents Chemother 2004; 48: Fritsche TR, Stilwell MG, Jones RN. Antimicrobial activity of doripenem (S-4661): a global surveillance report (2003). Clin Microbiol Infect 2005; 11: Brown SD, Traczewski MM. Comparative in vitro antimicrobial activity of a new carbapenem, doripenem: tentative disc diffusion criteria and quality control. J Antimicrob Chemother 2005; 55: Jones ME. In-vitro profile of a new beta-lactam, ceftobiprole, with activity against methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect 2007; 13: Pillar CM, Aranza MK, Shah D et al. In vitro activity profile of ceftobiprole, an anti-mrsa cephalosporin, against recent Grampositive and Gram-negative isolates of European origin. J Antimicrob Chemother 2008; 61: King A, Phillips I, Kaniga K. Comparative in vitro activity of telavancin (TD-6424), a rapidly bactericidal, concentration-dependent anti-infective with multiple mechanisms of action against Gram-positive bacteria. J Antimicrob Chemother 2004; 53: Jansen WT, Verel A, Verhoef J et al. In vitro activity of telavancin against gram-positive clinical isolates recently obtained in Europe. Antimicrob Agents Chemother 2007; 51: Andrews JM, BSAC Working Party on Susceptibility Testing. BSAC standardized disc susceptibility testing method (version 6). J Antimicrob Chemother 2007; 60: European Committee on Antimicrobial Susceptibility Testing. Carbapenems EUCAST Clinical MIC Breakpoints, Version 1.1, (30 May 2008, date last accessed). 59. European Committee on Antimicrobial Susceptibility Testing. Tetracyclines EUCAST Clinical MIC Breakpoints, Version 2.0, (30 May 2008, date last accessed). 60. European Committee on Antimicrobial Susceptibility Testing. Miscellaneous Antimicrobials EUCAST Clinical MIC Breakpoints, Version 2.1, MICmiscellaneous.html (30 May 2008, date last accessed). ii85