Evaluation of procedures for typing of group B Streptococcus: a retrospective study

Abstract

Background This study evaluates two procedures for typing of Streptococcus agalactiae (group B streptococci; GBS) isolates, using retrospective typing data from the period 2010 to 2014 with a commercial latex agglutination test (latex test) and the Lancefield precipitation test (LP test). Furthermore, the genotype distribution of phenotypically non-typable (NT) GBS isolates is presented. We also raise the awareness, that the difference in typing results obtained by phenotypical methods and genotype based methods may have implications on vaccine surveillance in case a GBS vaccine is introduced. Methods A total of 616 clinical GBS isolates from 2010 to 2014 were tested with both a latex test and the LP test. Among these, 66 isolates were genotyped by PCR, including 41 isolates that were phenotypically NT. Results The latex test provided a serotype for 83.8% of the isolates (95% CI [80.7–86.6]) compared to 87.5% (95% CI [84.6–90.0]) obtained by the LP method. The two assays provided identical capsular identification for all sero-typeable isolates (excluding NT isolates). The PCR assay provided a genotype designation to the 41 isolates defined as phenotypically NT isolates. Discussion We found that the latex test showed a slightly lower identification percentage than the LP test. Our recommendation is to use the latex agglutination as the routine primary assay for GBS surveillance, and then use the more labour intensive precipitation test on the NT isolates to increase the serotyping rate. A genotype could be assigned to all the phenotypically NT isolates, however, as a consequence genotyping will overestimate the coverage from possible future capsular polysaccharide based GBS vaccines.