Comparison of selective and nonselective enrichment broth media for the detection of vaginal and anorectal colonization with group B streptococcus.

The use of a selective enrichment broth medium has been widely recommended to optimize the recovery of group B streptococci (GBS) from genital and anorectal samples. Because selective antibiotic-containing versions of broth media are significantly more expensive than their nonselective parent formulations, we sought to examine whether the use of the nonselective Todd-Hewitt broth (THB) could accomplish comparable recovery of GBS to the recommended, selective version of this medium (Lim broth). During the study, vaginal and anorectal swab samples submitted to our laboratory for GBS culture were all inoculated onto Columbia colistin-nalidixic acid agar (CNA) and into either THB or Lim broth (alternated on a weekly basis). During the 45-week study period, 1200 samples (600 in each study arm) were evaluated. GBS were recovered from 164 samples (27.3%) in the Lim arm of the study, of which 60 (36.7%) were positive only on subculture of broths. In the THB study arm, 161 samples (26.8%) yielded GBS, of which 58 (36.0%) were positive only in broths. This study conclusively demonstrates that a nonselective enrichment broth media provides comparable sensitivity to the recommended selective broth for detection of GBS colonization during pregnancy.