SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References

Aims: To evaluate the full test scheme of Facklam and Sahm (1995) for the identification of clinical enterococcal isolates to genus and species level.

Methods and Results: Fifty-nine clinical isolates, previously provisionally classed as enterococci on the basis of just four biochemical tests of Facklam and Sahm and one other test, were subjected to genus and species identification using the full identification scheme of Facklam and Sahm; 98% of these strains were confirmed to be enterococci and of these, 69% were identified as Enterococcus faecalis and 31% as Enterococcus faecium. Six tests in the scheme (out of 24) gave anomalous or unreliable results for some strains, and two gave unexpected results for the majority of strains presumptively identified as Ent. faecium.

Conclusions: Nine (out of 12) genus tests and nine (out of 12) species tests from the Facklam and Sahm scheme were reliable. Testing for the presence of the Lancefield antigen D was also useful. The majority of presumptive Ent. faecium strains gave different results for the sorbitol and raffinose tests from that expected.

Significance and Impact of the Study: This study indicates the level of reliability for each of the tests in a current enterococcal identification scheme for differentiating clinical isolates, and showed that two tests gave consistently different test results from those expected for Ent. faecium.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References

Enterococci form part of the bowel flora of humans and other animals, and are known to be responsible for several clinical infections, including urinary tract infections and infective endocarditis (Megran 1992; Moellering 1992). In addition, they are increasingly recognized as major nosocomial pathogens (Korten and Murray 1997; Marcus et al. 1997). Of particular concern is the rapid increase in strains isolated from the hospital setting that are resistant to vancomycin and other antibiotics (French 1998). Antibiotic-resistant strains have been shown to spread between, as well as within hospitals and consequently, there is a need for rapid and accurate identification and speciation of enterococci to improve the efficiency of patient treatment and control the spread of the organism in the hospital environment (Murray et al. 1991).

Several similar schemes for the identification of enterococci to species level have been published previously (Facklam and Collins 1989; Facklam and Sahm 1995; Manero and Blanch 1999). These schemes are time-consuming to perform, especially for the high through-put screening of large numbers of isolates required in a clinical setting. Over 85% of clinical enterococcal isolates are either Enterococcus faecalis or Enterococcus faecium, with few other species occurring (Facklam and Sahm 1995). As these are the strains of most clinical significance, the possibility is raised of devising a simplified identification scheme, i.e. one that comprises a simplified, less time-consuming set of tests based on the existing biochemical schemes, and removing those that contribute little to specific differentiation of these two species from each other and from all other enterococci. Therefore, the aim of this study was to use a full identification scheme (that of Facklam and Sahm 1995) to identify a collection of clinical isolates provisionally identified as enterococci, and to evaluate the reliability of each test for these clinical strains.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References

Strains

Type strains Ent. faecalis NCTC775 and Ent. faecium NCTC7171, and Leuconostoc mesenteroides NCTC10817, Pediococcus pentosaceus NCTC12274 and Gemellahaemolysans NCTC10243, were all obtained from the National Collection of Type Cultures. Lactococcuslactis and Streptococcus agalactiae were obtained from the J. Wolf Collection, University of Leeds. Fifty-nine clinical isolates were obtained for the present study; these were isolated from patients at the Leeds General Infirmary during 1997–98, and were provisionally identified as enterococci by the Gram reaction, cell arrangement, catalase, Lancefield group D antigen and bile-aesculin tests. Their role in infection is undetermined.

Identification tests

Tests were based on those of Facklam and Sahm (1995) for the differentiation of enterococci.

Genus identification.

Tests for reaction on bile-aesculin agar (containing 40% bile salts) and presence of antigen D were performed according to Facklam and Sahm (1995) and Krieg (1981). Gram staining was performed on 1-day-old cultures grown on GM17 agar (Terzaghi and Sandine 1975). The catalase test was performed by suspending single colonies in 3% (v/v) hydrogen peroxide; production of gas was checked for up to 5 min post-hydrogen peroxide addition (Smibert and Krieg 1981). Vancomycin susceptibility was tested by the BSAC standardized disc sensitivity testing method using 30 μg vancomycin discs; sensitivity or resistance was determined by comparison with the type strains (negative controls). The PYR (production of pyrrolidonyl arylamidase) and LAP (production of leucine aminopeptidase) tests were performed using the diagnostic test kits of A/S Rosco (Taastrup, Denmark) or Britania Laboratories S.A. (Buenos Aires, Argentina), according to the manufacturer’s instructions. Gas production from glucose was tested using Mann, Rogosa and Sharpe lactobacillus broth (De Man et al. 1960). Growth at 10 and 45°C was determined in GM17 broth. Growth in 6·5% sodium chloride was tested in GM17 broth containing 6·5% (w/v) sodium chloride. All media were incubated at 35°C for up to 7 days unless otherwise stated.

Species identification.

Acid production from 1% (w/v) mannitol, sorbitol, sorbose, L-arabinose, D-ribose, sucrose and raffinose was tested in basal M17 broth (Terzaghi and Sandine 1975) (lacking lactose), using bromocresol purple as pH indicator. Deamination of arginine was tested in Thornley’s semi-solid arginine medium (Smibert and Krieg 1981). Motility was tested using phase contrast microscopy (Smibert and Krieg 1981). Pigment production was tested on tryptic soy agar. Tolerance to tellurite was tested on GM17 agar containing 0·04% (w/v) potassium tellurite. Utilization of pyruvate was tested using pyruvate broth (Facklam and Sahm 1995). All media were incubated at 35°C for up to 7 days unless otherwise stated.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References

The aim of this study was to subject a collection of clinical isolates of Gram-positive, catalase-negative cocci that had previously been provisionally identified as enterococci (on the basis of Gram reaction, cell arrangement, catalase test, the presence of antigen D and a positive reaction on bile-aesculin media only), to the full test scheme of Facklam and Sahm (1995), and then to evaluate the reliability of each test for these clinical isolates. The results of the genus identification tests, based on those of Facklam and Sahm (1995), are shown in Table 1. All the type genera and species, including the Ent. faecalis and Ent. faecium type strains, produced the expected result for each of the tests. Of the 59 clinical isolates, 58 were identified as enterococci. Isolate 11 was unable to grow at 10°C, 45°C, or in media containing 6·5% sodium chloride. Although it is not clear from the identification table of Facklam and Sahm which genus this isolate belongs to, these results would imply that it is not an Enterococcus.

Table 1.   Identification to genus level of a collection of 59 putative Enterococcus strains isolated from hospital patients with infections Thumbnail image of

Isolates 26, 27, 28, 31, 36 and 39 were unable to grow at 10°C but otherwise fitted well into the 12 test identification scheme for enterococci. This suggests that growth at 10°C may be an unreliable genus test. In addition, Manero and Blanch (1999) reported that most isolates of certain Enterococcus species are unable to grow at this temperature, again indicating that this test is unreliable for identifying enterococci to genus level. The results for the LAP test indicated that two of the 59 isolates were negative for production of leucine aminopeptidase, and other isolates were only weakly positive. The results of the other tests carried out indicate that these isolates are enterococci, so it would appear that the LAP test produces aberrant results for some clinical enterococcal strains. Another test which gave unreliable results was vancomycin sensitivity. With the emergence and spread of vancomycin resistance amongst enterococcal species (including Ent. faecalis and Ent. faecium) (Uttley et al. 1988) due to acquisition of the van genes (Korten and Murray 1997; Willey et al. 1999), care must clearly be taken in the interpretation of the vancomycin resistance test for differentiating genera. It is noteworthy that 10 of the 59 presumptive Ent. faecalis or Ent. faecium strains are more resistant to vancomycin than the control type strains.

The presumptive Enterococcus collection used for this study was originally derived from clinical isolates that gave appropriate reactions to just four of the Facklam and Sahm (1995) tests (catalase, Gram stain, cell arrangement and bile-aesculin reactions), plus an additional test for possession of the Lancefield Group D antigen. The present study indicates that approximately 98% of these strains were correctly identified as enterococci by these five tests alone. A further five Facklam and Sahm tests also appeared to be very reliable in the present study, notably gas production from glucose, PYR, motility, growth at 45°C and growth in the presence of 6·5% NaCl.

The results of the species identification tests of Facklam and Sahm (1995) are shown in Table 2. The control type strains of Ent. faecalis and Ent. faecium predominantly gave the expected result for each test, though Ent. faecalis NCTC775 gave inconclusive results for the acid production from mannitol test. Isolate 11 was not tested as it was identified as non-enterococci in the genus identification (Table 1). All 58 strains tested were identified as either Ent. faecalis (69%; 40 strains) or as Ent. faecium (31%; 18 strains), based on the full test scheme and allowing up to two anomalous results. There were two tests in which the majority of strains gave an unexpected result. Firstly, the species identification scheme of Facklam and Sahm (1995) indicates that more than 97% of Ent. faecium strains should test negative for the production of acid from sorbitol. Results from this study indicated that 16 out of 18 isolates identified as Ent. faecium by the complete test scheme were positive for acid production from sorbitol. This test was therefore ineffective for the differentiation of Ent. faecalis from Ent. faecium in the collection of enterococcal clinical isolates used. Secondly, the test for production of acid from raffinose also produced results conflicting with the test scheme of Facklam and Sahm (1995). The identification scheme of Facklam and Sahm indicates that more than 97% of Ent. faecium isolates can be expected to be negative for acid production from raffinose. Results from this study, however, indicated that 17 out of 18 isolates otherwise identified as Ent. faecium were positive for acid production from raffinose. The remaining tests (acid production from sorbose, arabinose, sucrose and ribose, deamination of arginine, pigment production, growth in the presence of tellurite, utilization of pyruvate and motility) all gave reliable results.

Table 2.   Identification to species level of a collection of 58 putative Enterococcus strains isolated from hospital patients with infections Thumbnail image of

In conclusion, the results indicate that the collection of clinical enterococcal isolates used in this study were identifiable as Ent. faecalis or Ent. faecium using the Facklam and Sahm (1995) scheme. Six tests exhibited some unreliability or gave inconsistent results. Two of these gave anomalous results for the majority of the presumptive Ent. faecium strains. The most reliable tests found here comprised the following 10 genus tests (nine out of the original 12) for the differentiation of Enterococcus: growth and reaction on bile-aesculin, Gram reaction, cell arrangement, motility, catalase production, gas from glucose, PYR, growth in the presence of 6·5% sodium chloride and growth at 45°C, plus possession of Lancefield antigen D. The results of the LAP test and growth at 10°C indicated that for this collection of strains, results of these tests was less reliable and added only supporting information. Tests such as resistance to vancomycin as outlined in Facklam and Sahm (1995) were predictably unreliable (Facklam and Sahm 1995). Reliable differentiation of Ent. faecalis and Ent. faecium from other Enterococcus species was obtained using acid production from sorbose, sucrose, ribose and L-arabinose, utilization of pyruvate, deamination of arginine, motility, ability to produce pigment and growth on 0·04% tellurite (nine out of the 12 original tests). For the collection of strains used here, the tests for acid production from mannitol, sorbitol and raffinose proved less reliable.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References

AMD is supported by an Emma and Leslie Reed Scholarship from the University of Leeds.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. Acknowledgements
  7. References
  • 1
    De Man, J.C., Rogosa, M., Sharpe, M.E. (1960) A medium for the cultivation of lactobacilli. Journal of Applied Bacteriology 23 , 130135.
  • 2
    Facklam, R.R. & Collins, M.D. (1989) Identification of Enterococcus species isolated from human infections by a conventional test scheme. Journal of Clinical Microbiology 27 , 731734.
  • 3
    Facklam, R.R. & Sahm, D.F. (1995) Enterococcus. In Manual of Clinical Microbiology, 6th edn. eds Murray, P.R., Baron, E.J., Pfaller, M.A., Tenover, F.C. and Yolken, R.H. pp. 308–314. Washington DC: ASM Press.
  • 4
    French, G.L. (1998) Enterococci and vancomycin resistance. Clinical Infectious Disease 27 (Suppl.),S75S83.
  • 5
    Korten, V. & Murray, B.E. (1997) Enterococcus. In Principles and Practise of Clinical Bacteriology eds Emmerson, A.M., Hawkey, P.M. and Gillespie, S.H. pp. 93–107. John Wiley & Sons Ltd.
  • 6
    Krieg, N.R. (1981) Enrichment and isolation. In Manual of Methods of Bacteriology eds Gerhardt, P., Murray, R.G.E., Costilow, R.N., Nester, E.W., Wood, W.A., Krieg, N.R. and Briggs Phillips, G. pp. 112–142. Washington DC: ASM Press.
  • 7
    Manero, A. & Blanch, A.R. (1999) Identification of Enterococcus spp. with a biochemical key. Applied and Environmental Microbiology 65 , 44254430.
  • 8
    Marcus, N., Peled, N., Yagupsky, P. (1997) Rapid increase in the prevalence of antimicrobial drug resistance among enterococcal blood isolates in southern Israel. European Journal of Clinical Microbiology and Infectious Diseases 16 , 913915.
  • 9
    Megran, D.W. (1992) Enterococcal endocarditis. Clinical Infectious Disease 15 , 6371.
  • 10
    Moellering, R.C. (1992) Emergence of Enterococcus as a significant pathogen. Clinical Infectious Disease 14 , 11731178.
  • 11
    Murray, B.E., Evans Patterson, J., Zervos, M.J. et al. (1991) Evidence for clonal spread of a single strain of β-lactamase producing Enterococcus (Streptococcus) faecalis to six hospitals in five states. Journal of Infectious Disease 163 , 780785.
  • 12
    Smibert, R.M. & Krieg, N.R. (1981) General characterization. In Manual of Methods for General Bacteriology eds Gerhardt, P., Murray, R.G.E., Costilow, R.N., Nester, E.W., Wood, W.A., Krieg, N.R. and Briggs Phillips, G. pp. 409–443. Washington DC: ASM Press.
  • 13
    Terzaghi, B.E. & Sandine, W.E. (1975) Improved medium for lactic streptococci and their bacteriophages. Applied Microbiology 29 , 807813.
  • 14
    Uttley, A.H.C., Collins, C.H., Naidoo, J., George, R.C. (1988) Vancomycin resistant enterococci (letter). Lancet 1 , 5758.
  • 15
    Willey, B.M., Jones, R.N., McGeer, A. et al. (1999) Practical approach to the identification of clinically relevant Enterococcus species. Diagnostic Microbiology and Infectious Diseases 34 , 165171.