Identification of Streptococcus pneumoniae strains

British Journal of Biomedical Science, 2001 by Pease, Alan, Bamber, Andrw I

SIR, In response to the correspondence from Andrew Bamber,1 I would urge some re-assessment of his proposal to eliminate the use of Optochin sensitivity for the identification of Streptococcus pneumoniae strains.

Published work in 1986(2) described 10 biochemically and physiologically atypical S. pneumoniae strains isolated during a six-month period. All were isolated from eyes, in the absence of other recognised pathogens. They were sensitive to Optochin (minimum inhibitory concentration [MIC]: 1.5 mg/L) and all were non-capsulate, so bile solubility and coagglutination tests with Omni-Serum were inherently negative. Biochemical testing indicated (incorrectly) that the strains were either S. morbillorum or Gemella haemolysans. It should be noted that both of these profiles were significantly different to the expected profile of S. pneumoniae, but not dissimilar to those of S. mitis, as reported by Andrew Bamber. The true identity of the strains was confirmed by whole-cell protein analysis, using sodium dodecyl sulphate-polyacrylamide gel electrophoresis.

Similar organisms and findings have been reported in the USA3 and in Glasgow, UK.4 Similarly, all strains were isolated from eye and conjunctival specimens. Some sensitivity to Optochin has been observed in other strains of alpha-haemolytic streptococci; however, normally the MIC is higher. This characteristic frequently disappears following subculture.

From these findings and further work (unpublished), my recommendation is that reproducible sensitivity to Optochin is the only traditional method to presumptively identify S. pneumoniae, and avoids misidentification of non-capsulate strains.

Alan Pease

King's Mill Centre for Healthcare Services

Sutton in Ashfield

Nottinghamshire NG17 4JL, UK

E-mail: alan.pease@kmc-tr.nhs.uk

References

1 Bamber A. Presumptive identification and reporting of Streptococcus pneumoniae (Letter). Br J Biomed Sci 2001; 58: 130.

2 Pease AA, Douglas CWI, Spencer RC. Identifying non-capsulate strains of Streptococcus pneumoniae from eyes. J Clin Pathol 1986; 39: 871-5.

3 Shayegani M, Parsons LM, Gibbons Jnr WE, Campbell, D. Characterisation of non-typable Streptococcus pneumoniae-- like organisms isolated from outbreaks of conjunctivitis. J Clin Microbiol 1982; 16: 8-14.

4 Smart LE, Dougall AJ, Girdwood RWA. Identification of non-- capsulate strains of Streptococcus pneumoniae by coagglutination. J Clin Pathol 1987; 40: 243.

SIR, Alan Pease makes an informative point, and I acknowledge the atypical presentation of non-typable Streptococcus pneumoniae as quoted. However, it was not my intention to propose the elimination of Optochin sensitivity testing. Indeed, often it is used to aid recognition of pneumococci on primary isolation plates from respiratory samples, and is invaluable as a means of identification thereafter. I merely suggested that, like many identification tests, it is not 100% reliable when used in isolation.

Munoz et al.1 reported clinical isolates of S. pneumoniae showing Optochin-sensitive and Optochin-- resistant colonies around the Optochin disc. They suggested that equivocal Optochin disc test results should be confirmed either with bile solubility test or agglutination tests, or both. alpha-Haemolytic streptococci other than S. pneumoniae produce zones of inhibition to Optochin. Mundy et al.2 made this observation when examining 639 isolates, confirming the identity of S. pneumoniae with a commercial molecular probe, and they proposed a more detailed algorithm to facilitate routine clinical identification of S. pneumoniae.

Adjunction of a second test decreases the likelihood of error in presumptive identification, and bile solubility testing is a relatively simple way of doing this. Latex kits are available, but are more costly and they have their limitations.3 Wasilauskas et al. concluded that bile solubility and Optochin tests were more reliable than serological methods. In characterising atypical S. pneumoniae, Shayegani et al.,4 stated that their isolates were both bile-soluble and Optochin-sensitive.

Busy clinical laboratories are unable to use extensive biochemical methods to confirm identification of isolates for routine purposes, let alone use sodium dodecyl sulphate-polyacrylamide gel electrophoresis of wholecell proteins or molecular probes for confirmation; therefore, we must proceed with a presumptive identification. However, equivocal isolates may be examined more thoroughly, or referred to a reference laboratory.

All isolates tested in our study were respiratory isolates that required distinction from other alpha-haemolytic streptococci that colonise the oropharyngeal tract. Alan Pease acknowledges that other strains of alpha-- haemolytic streptococci show sensitivity to Optochin. Minimum inhibitory concentrations are not tested routinely here; we rely upon a zone of inhibition around an Optochin 5 (mu)g OPID disc (Mast Group Ltd., Merseyside UK). To qualify this and to clarify my original point, two tests are better than one; I was re-emphasising the need to use bile solubility testing as an adjunct to Optochin sensitivity.