Blinded comparison of performance and cost-effectiveness of chromogenic and direct latex methods for detection of group B Streptococcus, with in-house PCR and LAMP as reference standards

Background: Group B Streptococcus (GBS) testing during pregnancy and application of intrapartum prophylaxis to carriers prevents early-onset disease due to GBS in neonates. New testing methods may be more accurate and cost-effective than conventional cultures. Objective: To compare performance and cost-effectiveness of three chromogenic agars and two direct latex agglutinations after carrot-broth enrichment, using in-house polymerase chain reaction (PCR) and loop-mediated isothermal amplification (LAMP) as reference standards. Methods: A total of 285 consecutive vaginal–rectal swabs were enriched with carrot broth and then tested using conventional Streptococcus selective agar (SSA), ChromAGAR Strep B (Colourex; Alere ULC, Ontario), ChromID Strepto B (bioMérieux Canada, Quebec), Brilliance GBS (Oxoid Company Inc, Ontario) and two latex agglutination kits: PathoDxtra Strep Grouping Reagent Kit (Thermofisher Scientific, Oxoid Company, Ontario) and MEDStrep (Alere ULC, Ontario). In-house PCR and LAMP reference methods were performed on frozen carrot broth. Results: Of the 285 samples received, 244 were analyzed by in-house PCR and 195 by LAMP. The conventional method (SSA) is less sensitive than PCR, but equally specific. Chromogenic agars were as sensitive as or more sensitive than SSA, and cost as much as or more than SSA. Direct latex antigens were less sensitive and specific than SSA, and cost less than SSA. In-house PCR was more sensitive than LAMP. Conclusion: Chromogenic media perform approximately as well as the conventional method (SSA) but are costlier. Direct latex agglutination methods do not achieve adequate performance. The ideal limit of detection for PCR is unknown, since the risk to the newborn of maternal carriage of a low inoculum of GBS is unknown.