Rapid tests for detecting group B streptococcus colonization in pregnant women: A systematic review and meta-analysis of diagnostic accuracy.

BACKGROUND

Maternal colonization with group B Streptococcus is common and the leading cause of neonatal sepsis. Rapid tests can be an alternative for screening; however, their diagnostic accuracy is variable when used with non-enriched samples.

OBJECTIVES

To determine the diagnostic accuracy of rapid tests based on nucleic acid amplification techniques (NAAT) without prior enrichment for detecting group B Streptococcus colonization in pregnant women.

SEARCH STRATEGY: Systematic searches of databases (Medline, Embase, CENTRAL, LILACS) from inception until February 2025, ongoing studies and gray literature.

SELECTION CRITERIA: We included diagnostic accuracy studies in which participants were recruited prospectively, consecutively, or randomly; measurement and data collection had to be prospective. We excluded case-control studies.

DATA COLLECTION AND ANALYSIS: We extracted data in duplicate. We assessed the quality of the included studies using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool. For the meta-analysis, we used the bivariate model by estimating a summary sensitivity and specificity. We used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach to diagnostic test accuracy to compile the summary of findings tables.

MAIN

RESULTS

A total of 52 studies (24 653 participants) met the inclusion criteria; the median sample size was 299, and the median colonization prevalence was 21.4%. For qPCR-based rapid tests, the sensitivity was 90% (95% CI: 88%-92%), and the specificity was 96% (95% CI: 95%-97%). For the LAMP-based test, the sensitivity was 92% (95% CI: 84%-97%), and the specificity was 96% (95% CI: 94%-98%). When exploring study entry status, sample source, and test processing in the delivery room as sources of heterogeneity, we found no statistically significant differences between the subgroups analyzed (P = 0.19, P = 0.43, and P = 0.44). We found no evidence of publication bias when analyzing the asymmetry in the funnel plot.

CONCLUSIONS

Rapid qPCR-based tests may have acceptable sensitivity (low-certainty evidence), and they probably have acceptable specificity (moderate-certainty evidence). For the LAMP-based tests, the evidence is very uncertain regarding the acceptability of sensitivity (very low-certainty evidence), but specificity is probably acceptable (moderate-certainty evidence) for detecting GBS colonization in pregnant women. Applicability is limited in intrapartum settings due to sample processing and in low- and middle-income settings.

PROSPERO REGISTRATION: CRD42024536503.