Validation of serotyping of Streptococcus pneumoniae in Europe.

INTRODUCTION: Serotyping of pneumococci has become increasingly important as new pneumococcal vaccines are introduced and place emphasis on knowledge of national serotype-distributions and their development over time. AIM: The aim of this study was to evaluate the quality of serotyping of pneumococci in Europe, and to focus on possible problems with methods, procedures, etc. that may lead to wrong serotypings. METHODS: This study was part of a larger EU-project. Eleven reference laboratories in Europe participated in the validation of pneumococcal serotyping. The Streptococcus Unit at SSI functioned as the gold standard for use of the Neufeld test. Each laboratory was asked to type 70 blinded pneumococcal strains by use of their normal serotyping procedure and to answer to a questionnaire regarding their experience and serotyping procedure for pneumococci. The 70 strains were chosen to represent all the 23 pneumococcal types included in the 23-valent pneumococcal vaccine plus a number of other less common types. RESULTS: A total of 735 serotypings was performed. Five laboratories performed complete serotyping whereas the remaining six laboratories performed partial or variable serotyping into groups or types that did not belong to groups. In general, a high degree of consensus appeared between the 11 European reference laboratories. Of 735 serotypings, 39 erroneous serotypings were made (5% of all). Most serotyping errors included a wrong serotype within the correct serogroup, where especially types 9N, 18C and 19F were mistyped. Furthermore, misidentification of noncapsular pneumococci like S. mitis and S. oralis was also a frequent error. For 22 strains (30%) of pneumococci, serotyping mistakes were made. The erroneous serotypings were neither correlated to the use of other methods than the Neufeld test, nor to the serotyping routine of the laboratories. A number of errors may be due to a serotyping result based on a negative reaction with a specific factor serum, instead of a positive reaction with another factor serum. This may be chosen in order to simplify the serotyping procedure. Thus, all the necessary factor sera must be used in order to assure correct serotyping. CONCLUSIONS: Overall, the quality of serotyping of pneumococci was high, and a high degree of consensus was found between the eleven laboratories. It is important to use all the necessary factor sera for serotyping, to perform all the necessary tests and to base a serotyping result always on one or more positive reactions and not on a negative reaction alone. More focus on serotyping of serotypes within groups seems to be warranted.