European Journal of Clinical Microbiology and Infectious Diseases

, Volume 23, Issue 1, pp 61–62

Use of Direct Latex Agglutination Testing of Selective Broth in the Detection of Group B Strepptococcal Carriage in Pregnant Women


    • Department of Clinical MicrobiologyHospital “J. M. Morales Meseguer”
  • J. Martínez
    • Department of Clinical MicrobiologyHospital “J. M. Morales Meseguer”
  • A. Menasalvas
    • Department of Clinical MicrobiologyHospital “J. M. Morales Meseguer”
  • R. Blázquez
    • Department of Clinical MicrobiologyHospital “J. M. Morales Meseguer”
  • T. Rodríguez
    • Microbiogy and Genetic DepartmentMurcia University
  • M. Segovia
    • Microbiogy and Genetic DepartmentMurcia University
Concise Article

DOI: 10.1007/s10096-003-1052-x

Cite this article as:
Guerrero, C., Martínez, J., Menasalvas, A. et al. Eur J Clin Microbiol Infect Dis (2004) 23: 61. doi:10.1007/s10096-003-1052-x


Direct latex agglutination testing of selective broth medium for the detection of group B streptococci was evaluated. Results were compared with those obtained by the recommended subculture method. Among the 551 vaginal-rectal specimens tested, 101 (18.3%) were positive by the subculture method. Of these subcultures, latex agglutination testing detected 99 (98%) positive specimens. Agglutination testing of selective broth is a sensitive method which offers the advantage of saving 24 h in the turnaround time for detection of group B streptococci in pregnant women.


Group B streptococci (GBS) is a leading cause of neonatal infection worldwide. Detection of GBS carriage in pregnant women is recommended in order to prevent the transmission of GBS from mothers to neonates. The screening method recommended by the Centers for Disease Control and Prevention (CDC) requires the inoculation of vaginal and rectal swabs into a selective broth. After 24 h of incubation the broth is subcultured onto blood agar plates and suspected colonies are identified by agglutination tests [1]. Although this method is sensitive, it requires 48–72 h to complete. In order to decrease the turnaround time to detection of GBS, several strategies have been proposed. The combination of a direct plating method in a selective solid medium with a selective broth has proved to increase the sensitivity and to decrease the time to detection of GBS [2]. Differential media such as Granada agar and GBS medium [3, 4] have been evaluated and have demonstrated good sensitivities and the great advantage of identifying colonies of GBS easily by their orange pigment. Molecular methods have also been developed. Although the performance characteristics of these assays are equivalent to those of culture methods, the higher cost and the need for specific equipment makes them inaccessible to most conventional laboratories [5, 6]. Recently, Park et al. [7] have evaluated the efficacy of testing the selective broth after overnight incubation by using a latex agglutination test for detection of GBS. This method showed a better sensitivity than the subculture (98.8% versus 93.1%) and a reduction in turnaround time for results in 24 h.

In order to decrease the time to detection of GBS, we compared the use of both direct plating and latex agglutination testing of selective broth with the recommended subculture method.

Materials and Methods

Five hundred and fifty-one consecutive specimens (307 vaginal and 244 rectal swabs) from 309 pregnant women were submitted to our laboratory and included in the study. All specimens were inoculated onto blood agar plates containing 10 μg/ml of colistin and 10 μg/ml of nalidixic acid (CNA; bioMérieux, France) and the swabs were then immersed in Todd-Hewitt broth supplemented with 8 μg/ml of gentamicin and 15 μg/ml of nalidixic acid (SBM; Becton Dickinson Microbiology Systems, USA). All plated media were incubated for 24 h at 35°C with 5% CO2. SBM broths were incubated at 35°C in an ambient atmosphere. After incubation, CNA plates were examined by one investigator for colonies typical of GBS. Negative plates were reincubated for an additional 24 h and reexamined. After 24 h of incubation, the SBM broth was subcultured onto a blood agar plate and incubated for 24 h at 35°C with 5% CO2. As with CNA plates, subcultures were examined by the same investigator for colonies typical of GBS. Negative plates were reincubated and reexamined on the following day. In addition, latex testing for GBS (Slidex Strepto-Kit, BioMérieux) was performed by a second investigator on all SBM broth cultures after 24 h of incubation. Testing was performed according to the manufacturer’s protocol for cultures in broth specified in the package insert of the Slidex Strepto-Kit. The first investigator was blinded to the results of the latex testing and the second investigator was blinded to the results of the cultures until they were finished. Discrepant results were resolved with new subcultures of the broth tubes. Identification of GBS was performed by Gram stain, catalase reaction and an agglutination assay (Slidex Strepto-Kit).

Results and Discussion

A total of 65 (21%) pregnant women were found to be colonized by GBS. One hundred and one (18.3%) vaginal or rectal swabs were found to be GBS positive. Table 1 shows the results of direct plating, SBM subculture and direct latex testing. Of the 551 specimens, GBS were recovered from 69 (68.3%) by both CNA direct plating and SBM broth. GBS were recovered from 32 (31.7%) additional specimens only on SBM broth subcultures. The direct latex testing was positive in 101 specimens. Of these, GBS were recovered upon subcultures of the broth in 99 (98%). Two of the specimens were positive by the latex method but negative for GBS upon repeated subcultures. Both samples revealed heavy growth of colonies resembling Enterococcus spp. Agglutination testing was positive with Streptococcus group B and D antibodies and definitive identification as Enterococcus faecalis was achieved using API 20 STREP (bioMérieux). Agglutination in more than one latex suspension may indicate a mixed culture of several streptococcal groups or the presence of an autoagglutinating strain. The possibility of a mixed culture was discarded by repeated subcultures of the broth, further isolation of suspected colonies and latex testing of isolated colonies on blood agar. The presence of polyagglutinations could also indicate an autoagglutination strain. This phenomenon is described as a limitation of the Slidex Strepto-Kit latex test, and in these cases biochemical identification is recommended by the manufacturer.
Table 1.

Recovery of GBS from direct CNA, SBM broth, and direct latex testing

Recovery results

No. of specimens

Direct CNA
















The use of a direct plating method in combination with a selective broth is based on the results of a previous study in which the authors found that direct plating increases test sensitivity by reducing the overgrowth of enterococci, which could mask the presence of GBS in the SBM subcultures [2]. In our work, direct plating did not increase the sensitivity of the subcultures. We have not found positive direct CNA cultures with negative subcultures of the selective broth. However, the use of direct plating provided an identification of GBS 24 h sooner than the broth method for more than half of the positive specimens.

Our results confirm those of Park et al. [7], who also found that direct latex testing detected 98% of positive subcultures. In their work they found seven latex-positive samples with negative subcultures from the broth. All these samples revealed heavy growth of Enterococcus faecalis, but PCR testing of the broth determined the presence of cfb genes specific for GBS. This finding led us to consider the possibility that in our study the two false latex-positive samples were in fact true latex-positive, although this could not be confirmed.

There were two false-negative specimens by latex testing. In both cases, the direct CNA culture was positive and subcultures from the SBM broth showed scant GBS growth. The insufficient growth of GBS in the broth could be the cause of these false-negative results.

In conclusion, the sensitivity and specificity of the direct latex agglutination method for detecting vaginal and rectal carriage of GBS was 98 and 99.5%, respectively. This sensitivity increases to 100% when used in combination with direct plating in a selective medium. In addition, if the direct plating is used and shows GBS after 24 h of incubation, there is no need to perform the latex agglutination from the broth. These results indicate that the combination of direct plating with the latex agglutination of the selective broth is comparable to the standard CDC method, and offers the advantage of providing an identification of GBS 24 h sooner than the subculture method.

Copyright information

© Springer-Verlag 2004