Journal of Antimicrobial Chemotherapy (1991) 27, 469-474

Activity of clindamycin against Staphylococcus aureus and Staphylococcus epidermidis from four UK centres D. S. Reeves", H. A. Holt", I. Phillips6, Anna King*, R. S. Miles', R. Paton', R. Wise' and J. M. Andrews''

MICs of penicillin, methicillin, clindamycin, erythromycin, sodium fusidate and gentamicin were determined by an agar dilution method for 300 current isolates of Staphylococcus aureus and 100 of S. epidermidis, collected from four centres, and 38 stock strains of methicillin-resistant S. aureus (MRSA). All but one of the 300 current isolates of S. aureus were sensitive to clindamycin (MIC < 0-5 mg/1), with an MIQo of 012 mg/1. Of a total of 39 MRSA strains, 11 (28-2%) were resistant to clindamycin (MIC > 32 mg/1); all of these strains were also resistant to erythromycin. Ten of the 100 strains of S. epidermidis were resistant to clindamycin; they came from a reasonably equal geographical distribution and were also resistant to erythromycin. The results suggest that clindamycin might still be useful as a secondline agent for infections caused by 5. aureus and S. epidermidis, although its activity against MRSA was limited to approximately two-thirds of the MRSA strains tested in this study.

Introduction Since its introduction into the UK in 1970, no systematic large scale study has been made of clindamycin activity against Staphylococcus aureus and S. epidermidis, although such studies have been performed in other centres (Evans, Patterson & Stratton, 1982; Van Landuyt, 1987; Fidalgo et al., 1988). This study was designed to reassess the therapeutic potential of clindamycin against current isolates of staphylococci from serious infection in hospital patients. Materials and methods Source of staphylococci In total, 300 current isolates of coagulase positive staphylococci and 100 S. epidermidis, derived from cases of serious infection, were collected in four centres. In addition 38 selected stock strains of methicillin-resistant 5. aureus(MRSA) were also tested. Serious infection included septicaemia, abscess, wound infection and osteomyelitis. The current clinical strains were isolated at four hospitals: Dudley Road, Birmingham; The Royal Infirmary, Edinburgh; Southmead Hospital, Bristol and St. Thomas's Hospital, 469 0305-7453/91/040469+06 $02.00/0

© 1991 The British Society for Antimicrobial Chemotherapy

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

"Department of Medical Microbiology, Southmead Hospital, Bristol;bDepartment of Microbiology, St. Thomas's Hospital, London; 'Clinical Bacteriology Services, University Medical School, Edinburgh, dDepartment of Medical Microbiology, Dudley Road Hospital, Birmingham, UK

470

D. S. Reeves et al.

London. Each centre collected 75 strains of S. aureus and 25 strains of S. epidermidis. The MRSA strains were supplied from the collection at St. Thomas's Hospital; these strains were epidemiologically distinct and split evenly between the following types: methicillin-resistant alone; gentamicin and methicillin-resistant; erythromycin and methicillin-resistant; fusidic acid and methicillin-resistant. Laboratory methods

Results The results obtained with a control strain (S. aureus NCTC 6571, Oxford strain) showed a good correlation between the centres, with all results lying within two doubling dilutions for clindamycin (mode 006mg/1), erythromycin (mode 012mg/1), fusidic acid (mode 006mg/1) and penicillin (mode 003mg/1), one dilution for gentamicin (mode 0-25 mg/1), and being identical for methicillin (mode 1 mg/1). Of the 300 S. aureus strains collected during the study, only one, from Edinburgh, was resistant to methicillin (MIC 8 mg/1). The distribution of MIC values of all Table I. Distribution of MICs for 300 contemporary isolates of 5. aureus from four UK centres No. of isolates with the indicated MIC MIC (mg/1) N.T. 32

clindamycin erythromycin

fusidate

gentamicin

methicillin

penicillin

— 7 176 90 13

— — 6 117 99 76 2 — —

— — — — — 2 109 162 26 1 — — —

2 21 32 36 64 64 40 25 10 3 3 — —

. 3 173 111 12

116 171

1 — —

3 2

— —



8

1 2 2 2 3 2 2

— — —

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

The current isolates were collected between July and September 1988, and were stored at room temperature on agar slopes. MICs were determined in batches at each centre by means of an agar dilution technique in Oxoid DST agar. The inoculum size was 106cfu/spot for methicillin and 104cfu spot for all other antibiotics. Incubation was overnight in air at 30°C for tests with methicillin and at 37°C for all other antibiotics. The antibiotics tested (with their range of concentrations and BSAC recommended break-point in mg/1) were: penicillin (0015—32; 012); methicillin (0125-64; 4); sodium fusidate (0-03-32; 1); erythromycin (0-015-32; 0-5); clindamycin (0-03-32; 0-5); gentamicin (006-32; 4). The incidence of resistance was determined as defined by the BSAC break-points (Report by a Working Party of the British Society for Antimicrobial Chemotherapy, 1988).

471

Clindamycin activity against staphylococd

Table EL Distribution of MICs for 39 methicillin-resistant S. aureus (38 from the culture collection of St. Thomas's Hospital, London, and one from Edinburgh) No. of isolates with the indicated MIC MIC (mg/1)

clindamycin erythromycin

fusidate

gentamicin

— 3 11 16 1 — — 1 2 3 1 — 1

— — — 10 8 1 — — 2 5 3 6 4

penicillin

_

NT. 32

methicillin

2 18 7 — 1 — — —

4 1 — — 2 10

— — 11

2 20

— — — — 1 2 7 11 18

— — — — — — — 2 4 5 19 9 —

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

antibiotics tested against these strains is shown in Table I. Clindamycin showed good and consistent activity, with only one strain resistant (MIC 1 mg/1). The results were closely grouped, with 95% of strains having an MIC of 006 or 012mg/1. Thirteen of the 300 strains were resistant to erythromycin, of which seven came from Birmingham (9-3% resistance in this centre). There was no correlation between resistance to erythromycin and resistance to fusidic acid (14 in 300 strains). Ten of the fusidateresistant strains came from one centre, Edinburgh, making a resistance rate of 13-3% in this centre. In Birmingham and Edinburgh the MICs for strains sensitive to erythromycin and fusidate, respectively, were the same as those in Bristol and London, suggesting that the observed resistances were real and not due to technical factors. There was no gentamicin resistance amongst the methicillin-sensitive strains. The results for penicillin were clearly unreliable since 30% of strains were sensitive; this is a well known phenomenon with agar dilution MICs when an inoculum of < 106cfu is used. Testing for 0-lactamase production would have been more reliable. All but one of the MRSA strains (38/39) were collected at St. Thomas's Hospital, London, and had a methicillin MIC of > 4 mg/1. The remaining strain had an MIC of 4 mg/1 (Table II). Of these strains, 11 (28-2%) were resistant to clindamycin, and these 11 strains were also resistant to erythromycin. There was a high rate of resistance to erythromycin amongst the methicillin-resistant strains (34/39), and there seemed to be three populations of strains in terms of MIC values: sensitive (0-12 and 0-25 mg/1); moderately resistant (2 and 4 mg/1); and highly resistant ( > 32 mg/1). Of the clindamycin-resistant strains, only four were resistant also to fusidic acid. There were, however, six other strains that were resistant to fusidic acid, and therefore there was no apparent connection between the resistance to these two agents. Of the 39 strains, 20 were resistant also to gentamicin; this resistance did not correlate with that to clindamycin, although all but one of the gentamicin-resistant strains were also resistant to erythromycin. Six of the 17 methicillin and gentamicin-resistant strains were also

D. S. Reeves et al.

472

Table m . Distribution of MICs for 100 contemporary isolates of S. epidermidis from four UK centres No. of isolates with the indicated MIC MIC (mg/1)

2 3 36 43 6 — — — — — — — 10

— 3 5 38 3 — 1 1 — 1 — 1 47

fusidate

gentamicin

methicillin

penicillin

2 10 55 19 — — — 1 — 6 5 2 —

1 4 38 6



6 4 2 8 16 10 23 18 5 3 1 3 1

2 — 2 12 5 20 4 6

1 3 20 23 28 11 8 4 2

resistant to clindamycin, and four of the 11 clindamycin-resistant strains were resistant also to gentamicin. Ten of 100 strains of S. spidermidis (Table III) were resistant to clindamycin with a reasonably equal geographical distribution; all were resistant also to erythromycin. Overall, 47 strains were resistant to erythromycin, with higher rates in London (68%) and Edinburgh (56%) than in Bristol and Birmingham (40%). The trimodal distribution of erythromycin resistance seen with MRSA was not apparent with S. epidermidis. In total, 14 strains were resistant to fusidate, with a wide geographical variation (Bristol 4%, Birmingham 12%, Edinburgh 16% and London 24%). In total, 35 strains were resistant to gentamicin, with a fairly even geographical distribution (Birmingham 24%, London 28%, Bristol 44%, Edinburgh 44%). All the clindamycin resistant strains were resistant to gentamicin. In total 25 strains were resistant to methicillin (three from Bristol, four from London, seven from Birmingham and 11 from Edinburgh). Although these differences in methicillin resistance seem significant, care must be taken in interpretation since the break-point is close to the main body of the MIC values and any small technical variations could, therefore, result in a large shift to the putatively resistant population. Only two methicillin-resistant strains (both from Edinburgh) were resistant also to clindamycin. There were no differences in susceptibility to clindamycin and the other antibiotics tested between the S. aureus strains from different specimen sites. This was also the case for the S. epidermidis strains. Discussion We are not aware of any previous survey of clindamycin activity against strains of S. aureus and S. epidermidis collected entirely from the UK. In a study of strains collected from 11 hospitals in Belgium and Luxembourg, 3% of 119 methicillin-

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

N.T. 32

clindamycin erythromycin

Clindamycin activity against staphylococd

473

Of the strains of S. epidermidis isolated, 10% were resistant to clindamycin, a figure which contrasts with 34% of the 106 strains described by Van Landuyt (1987). However, although there was a higher incidence of methicillin resistance in the study of Van Landuyt (1987), in contrast to the present study there was no correlation between clindamycin and methicillin resistance. Of the 25 methicillin-resistant strains, 12 (48%) were resistant also to gentamicin, which contrasts with 25% in the Belgian study. Amongst the methicillin-sensitive strains, 23 (31%) of 75 were gentamicin-resistant, as opposed to 59% of 70 strains in the Belgian study. In contrast to S. aureus, it appears, therefore, that there is no consistent correlation between methicillin and gentamicin resistance in 5. epidermidis. Fidalgo et al. (1988) found 57% of 79 strains of S. epidermidis from Spain to be resistant to clindamycin, as were 25% of the strains of coagulase-negative staphylococd (very largely S. epidermidis) considered to be pathogenic by Younger et al. (1987). In another study from North America, over 25% of 42 strains of S. epidermidis were resistant to clindamycin (Price & Flournay, 1982). It seems, therefore, that clindamycin resistance in S. epidermidis is more prevalent in countries other than the UK, which is perhaps a reflection on the relative usage of clindamycin or the 16-membered ring macrolides, which are known to select for constitutive resistance to the macrolide-lincosamide-streptogramin (MLS) group of antibiotics. The results in our study suggest that clindamycin may still be a useful second-line agent for infections caused by S. aureus and 5. epidermidis, although its usefulness in MRSA infection would be limited by the fact that nearly a third of strains are initially resistant. Whether the favourable position would be maintained if widespread usage occurred is debatable in view of the higher rates of resistance observed in both MSSA and 5. epidermidis from some other countries.

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

sensitive S. aureus strains were resistant to clindamycin and 4% to erythromycin (Van Landuyt 1987). These values are in broad agreement with our own findings and are lower than those from North America. The corresponding values for MRSA (96 strains) were 60% and 89%, the former rate being rather higher than our finding in the UK (28%). Gentamicin resistance was much more prevalent in MRSA (95%) than in MSSA (2%) (50% and 0%, respectively, in our study). Similarly, of 209 strains of S. aureus collected from a hospital in Spain, 3-7% were resistant to clindamycin (Fidalgo et al, 1988), but the correlation with methicillin resistance was not stated. Clindamycin has received widespread use in North America as an agent active against anaerobes, and this may have resulted in the selection of resistant strains. In total, 28 of 140 strains of S. aureus from New Jersey were reported by Jenssen et al. (1987) to express constitutive resistance to clindamycin and erythromycin, with a further 13 strains resisant also to erythromycin. The susceptibility of the 140 strains to methicillin was not reported, a recent study of 106 strains of MRSA collected from 21 different countries world-wide (Maple, Hamilton-Miller & Brumfitt, 1989) showed clindamycin to be active against 34%, a figure similar to that found in this study. While 92% of these strains were resistant to erythromycin and gentamicin, only 12% were resistant to fusidic acid. It would be noted that the prevalence of gentamicin resistance was much higher than in the present study. A higher rate of clindamycin resistance in MRSA from the USA was reported by Evans etal. (1982), with at least 50% of 50 strains resistant to 32 mg/1. In the same study, at least 25% of the 50 methicillin-sensitive strains were resistant to clindamycin, whereas < 1 % were resistant in our own study.

474

D. S. Reeves et al.

References

(Received 12 August 1990; accepted 24 November 1990)

Downloaded from http://jac.oxfordjournals.org/ at University of Chicago on November 7, 2014

Evans, M. E., Patterson, L. S. & Stratum, C. W. (1982). In vitro comparison of clindamycin and pirlimycin (U-57930E) activity against Staphylocaccus aureus. Antimicrobial Agents and Chemotherapy 22, 334-5. Fidalgo, S., Mendez, F. J., Hardisson, C. & Salas, J. A. (1988). Epidemiology of macrolide and lincosamide resistance in species of staphylococci in a general hospital. Journal of Hospital Infection 11, 36-43. Jenssen, W. D., Thakker-Varia, S., Dubin, D. T. & Weinstein, M. P. (1987). Prevalence of macrolides-lincosamides-streptogramin B resistance and erm gene classes among clinical strains of staphylococci and streptococci. Antimicrobial Agents and Chemotherapy 31, 883-8. Maple, P. A. C , Hamilton-Miller, J. M. T. & Brumfitt, W. (1989). World-wide antibiotic resistance in methicillin-resistant Staphylococcus aureus. Lancet i, 537-40. Price, S. B. & Flournoy, D. J. (1982). Comparison of antimicrobial susceptibility patterns among coagulase-negative staphylococci. Antimicrobial Agents and Chemotherapy 21, 436-40. Report by a Working Party of the British Society for Antimicrobial Chemotherapy (1988). Breakpoints in in-vitro antibiotic sensitivity testing. Journal of Antimicrobial Chemotherapy 21, 701-10. Van Landuyt, H. W. (1987). Spectrum of clindamycin. In vitro effects and efficacy. Acta Therapeutica 13, 541-52. Younger, J. J., Christensen, G. D., Bartley, D. L., Simmons, J. C. H. & Barrett, F. F. (1987). Coagulase-negative staphylococci isolated from cerebrospinal fluid shunts: importance of slime production, species identification, and shunt removal to clinical outcome. Journal of Infectious Diseases 156, 548-54.

Activity of clindamycin against Staphylococcus aureus and Staphylococcus epidermidis from four UK centres.

MICs of penicillin, methicillin, clindamycin, erythromycin, sodium fusidate and gentamicin were determined by an agar dilution method for 300 current ...
323KB Sizes 0 Downloads 0 Views