Abstract

Introduction Maternal Group B Streptococcus (GBS) rectovaginal colonization is an important risk factor for invasive disease in neonates, yet availability of culture-based methods for detection is limited in low-resource settings. We evaluated the diagnostic performance of the HiberGene (HG) GBS loop-mediated isothermal amplification (LAMP) assay for the rapid detection of GBS in rectal/vaginal swabs collected from women in Uganda. This work forms a part of the PROGRESS GBS study. Methods In phase 1, 1294 rectal and vaginal swabs were collected from pregnant women and inoculated in enrichment (Lim) broth, which was then tested using the HG GBS LAMP assay (sip gene target) and culture on chromogenic agar. In phase 2, 166 swabs from nonpregnant women were tested directly (without the enrichment step). For samples with discordant results, an additional method of testing against multiplex real-time polymerase chain reaction assay was used. Results Overall, the HG GBS LAMP assay detected more GBS-positive samples (31.3%; 452/1445) than culture-based methods (13.3%; 192/1445). Multiplex polymerase chain reaction–tested results were concordant with LAMP results in 96.3% of cases. The sensitivity and specificity of the LAMP assay, after adjusting for the tiebreaker results of discordant samples, were 94.4% (95% confidence interval, 86.2–99.4) and 99.0% (95% confidence interval, 94.3–100), respectively. Conclusions The results of this study demonstrate high sensitivity and specificity of the HG GBS LAMP assay for the detection of GBS rectovaginal colonization in our setting. Given its rapid turnaround time, the HG GBS LAMP assay could appropriately be used to screen women for GBS rectovaginal colonization during labor to enable provision of intrapartum antibiotic prophylaxis.