Assessing the accuracy of a novel algorithm for group B Streptococcus detection in pregnant women and its impact on early-onset neonatal group B streptococcal infections

Objectives

The objective of this study was to assess the accuracy of a novel algorithm—designed in accordance with guidelines from the American College of Obstetricians and Gynecologists and the American Society for Microbiology—for detecting group B Streptococcus (GBS) colonization in pregnant women. Additionally, the study aimed to determine the potential impact of this algorithm on early-onset neonatal group B streptococcal disease (GBS-EOD) cases.

Methods

From March 2021 to March 2024, 5034 vaginal-rectal samples were collected from pregnant women at 35 weeks to 37 weeks gestation using ESwab™ medium. Samples were cultured in Todd Hewitt broth, subcultured on Granada medium, and tested with the DiaSorin Simplexa™ GBS Direct Assay (positive if Cycle threshold ≤ 35). Antimicrobial susceptibility testing was conducted on GBS isolates from penicillin-allergic women. GBS-positive women received intrapartum antibiotic prophylaxis. The study also reviewed cases of GBS-EOD during the same period.

Results

In total, 938 (18.6%) pregnant women were colonized with GBS. Among these, 625 samples (66.6%) were culture-positive, whereas 313 (33.4%) were culture-negative, but broth-enriched nucleic acid amplification test (NAAT)-positive. No samples were broth-enriched NAAT-negative and culturepositive. The broth-enriched NAAT demonstrated a sensitivity of 100 (95% CI: 99—100) and a specificity of 93 (95% CI: 92—94), along with a positive predictive value of 67 (95% CI: 67—70) and a negative predictive value of 100 (95% CI: 100—100). There was substantial agreement between methods (Cohen's kappa = 0.76 [95% CI: 0.74—0.79]). All 45 GBS isolates from penicillin-allergic women were susceptible to penicillin, ampicillin, and vancomycin, whereas 44.4% (20/45) were resistant to erythromycin, and 40% (18/45) were resistant to clindamycin. No cases of GBS-EOD were detected among the 5563 live newborns during the study period (95% CI: 0—0.054%).

Discussion

The novel algorithm was rapid, sensitive, and specific, effectively identifying candidates for intrapartum antibiotic prophylaxis, potentially reducing newborn infection and lowering GBS-EOD incidence and neonatal mortality rates. Integrating broth-enriched NAATs to verify negative culture samples could enhance detection and prevention efforts.